ML23331A986: Difference between revisions

From kanterella
Jump to navigation Jump to search
(StriderTol Bot change)
(StriderTol Bot change)
 
Line 16: Line 16:


=Text=
=Text=
{{#Wiki_filter:EXHIBIT 6 Declaration of Robert E. Kopp I, ROBERT E. KOPP, declare as follow:
{{#Wiki_filter:}}
Personal Background and Experience
: 1. I am a climate and sea level scientist. My research focuses on past and future sea-level change, the interactions between physical climate change and the economy, and the use of climate risk information to inform decision-making.
: 2. Since 2011, I have served on the faculty of Rutgers University-New Brunswick. I am currently a Distinguished Professor in the Department of Earth & Planetary Sciences.
: 3. I have previously served at Rutgers University as an assistant professor from 2011-2014, as an associate professor from 2014-2017, as a professor from 2017-2023, as director of the Rutgers Institute of Earth, Ocean and Atmospheric Sciences from 2017-2021, and as founding co-director of the University Office of Climate Action from 2021-2023.
: 4. Prior to joining the Rutgers faculty, between 2009 and 2011, I was a Science &
Technology Policy Fellow at the U.S. Department of Energy Office of Climate Change Policy and Technology. Between 2007 and 2009, I was a postdoctoral fellow in geosciences and public policy at Princeton University.
: 5. I received a Bachelor of Science degree in Geophysical Sciences from the University of Chicago in 2002, a Master of Science degree in Geobiology from the California Institute of Technology in 2005, and a Ph.D. in Geobiology from the California Institute of Technology in 2007.
1
: 6. Since 2021, I have directed the Megalopolitan Coastal Transformation Hub, a National Science Foundation-funded consortium of 13 institutions that advances coastal climate adaptation and the scientific understanding of natural and human coastal climate dynamics.
: 7. I have authored over 145 scientific papers, the vast majority of which relate to climate change or sea-level rise.
: 8. I am a fellow of the American Geophysical Union and the American Association for the Advancement of Science.
: 9. I was one of the lead authors of the United Nations Intergovernmental Panel on Climate Change (IPCC)s Sixth Assessment Report (AR6), which was published between 2021 and 2023.1 In particular, I served as a lead author of the chapter addressing sea-level change in the 2021 Physical Science volume and also contributed to other portions of the report.2
: 10. I am a co-author on the reports that developed the sea-level rise projections used in the Fourth National Climate Assessment and the Fifth National Climate Assessment.3 I chaired New Jersey sea-level rise assessment panels in 2016 and 2019,4 and have also 1
THE SIXTH ASSESSMENT REPORT OF THE INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE, IPCC (2021-2023),
available at https://www.ipcc.ch/assessment-report/ar6/.
2 Fox-Kemper et al., OCEAN, CRYOSPHERE AND SEA LEVEL CHANGE. CLIMATE CHANGE 2021: THE PHYSICAL SCIENCE BASIS. CONTRIBUTION OF WORKING GROUP I TO THE SIXTH ASSESSMENT REPORT OF THE INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE (Cambridge U. 2021), available at https://doi.org/10.1017/9781009157896.011.
3 William V. Sweet et al., GLOBAL AND REGIONAL SEA LEVEL RISE SCENARIOS FOR THE UNITED STATES, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION (2017), available at https://goo.gl/YUehx6; William V. Sweet et al.,
GLOBAL AND REGIONAL SEA LEVEL RISE SCENARIOS FOR THE UNITED STATES: UPDATED MEAN PROJECTIONS AND EXTREME WATER LEVEL PROBABILITIES ALONG U.S. COASTLINES (2022), available at https://oceanservice.noaa.gov/hazards/sealevelrise/sealevelrise-tech-report.html.
4 R.E. Kopp et al., ASSESSING NEW JERSEYS EXPOSURE TO SEA-LEVEL RISE AND COASTAL STORMS: REPORT OF THE NEW JERSEY CLIMATE ADAPTATION ALLIANCE SCIENCE AND TECHNICAL ADVISORY PANEL (2016), available at https://doi.org/10.7282/T3ZP48CF; R.E. Kopp et al., NEW JERSEYS RISING SEAS AND CHANGING COASTAL STORMS: REPORT OF THE 2019 SCIENCE AND TECHNICAL ADVISORY PANEL (2019), available at https://doi.org/10.7282/t3-eeqr-mq48.
2
 
served on sea-level rise expert panels for the states of Maryland, California, and Massachusetts, as well as New York City and the city of Boston.
Key Messages
: 11. The United Nations created the IPCC in order to provide policymakers with scientific assessments on climate change and related risks. The 2021 Working Group 1 contribution to IPCC AR6 represents the most comprehensive, global assessment of the state of the physical science of climate change. This report includes local sea-level projections for global coastlines.5 The sea-level projections in the 2021 report represent a consensus assessment based on multiple lines of evidence. In my expert judgment, the 2021 IPCC report findings continue to provide a concise and accurate summary of the current state of the underlying science.
: 12. The sea-level projections in the 2021 IPCC report characterize the likely range of sea-level rise under different possible future emissions scenarios, and they also include projections considering the potential effect of faster-than-expected ice sheet mass loss on sea-level rise.
: 13. The 2022 Sea Level Rise Technical Report, Global and Regional Sea Level Rise Scenarios for the United States (also referred to as Sweet et al., 2022), was written on behalf of the U.S. Sea Level Rise and Coastal Flood Hazard Scenarios and Tools Interagency Task Force by twenty-four authors from federal agencies and academic institutions, of which I was one.6 The report is derived from the science presented in IPCC AR6. It is intended to help inform Federal agencies, state and local governments, and stakeholders in coastal communities about current and future sea level rise to help 5
Fox-Kemper et al., supra at 2.
6 Sweet et al. (2022), supra at 3.
3
 
contextualize its effects for decision-making purposes.7 It identified several key findings with respect to sea-level rise. In my expert judgment, these findings continue to provide a concise and accurate summary of the current state of the underlying science. Specifically, the 2022 Sea Level Rise Technical Report found that:
[i.] Relative sea level along the contiguous U.S. (CONUS) coastline is expected to rise on average as much over the next 30 years (0.25-0.30 m over 2020-2050) as it has over the last 100 years (1920-2020). Due to processes driving regional changes in sea level, there are similar regional differences in both the modeled scenarios and observation-based extrapolations, with higher RSL rise along the East (0-5 cm higher on average than CONUS) and Gulf Coasts (10-15 cm higher) as compared to the West (10-15 cm lower) and Hawaiian/Caribbean (5-10 cm lower) Coasts.8
[ii.] By 2050, the expected relative sea level (RSL) will cause tide and storm surge heights to increase and will lead to a shift in U.S. coastal flood regimes, with major and moderate high tide flood events occurring as frequently as moderate and minor high tide flood events occur today. Without additional risk-reduction measures, U.S. coastal infrastructure, communities, and ecosystems will face significant consequences.9
[iii.] Higher global temperatures increase the chances of higher sea level by the end of the century and beyond.10
: 14. The 2022 Sea Level Rise Technical Report includes sea level rise scenarios for US coastlines. These scenarios include a Low scenario, which reflects a continuation of late-20th-century/early-21st-century trends and a reversion of recent sea-level acceleration; Intermediate Low and Intermediate scenarios, which bracket the most likely range of future sea-level rise in the absence of rapid mass loss from the Antarctic and Greenland ice sheets; and Intermediate High and High scenarios, which reflect the potential for rapid 7
Id. at xii.
8 Id.
9 Id. at xiii.
10 Id.
4
 
ice-sheet mass loss. It also includes local extrapolations of current trends and accelerations through 2050.
: 15. Among the sites included in the 2022 Sea Level Rise Technical Report is the tide gauge at Key West.11 Scenarios for average sea-level rise at Key West over 2041-2059, relative to average sea level over 1995-2014, are as follows:
Scenario                              Central Value for 2050          Range for 2050 Low                                    0.8 ft                          0.7-1.0 ft Intermediate Low                      1.0 ft                          0.8-1.2 ft Intermediate                          1.1 ft                          0.9-1.4 ft Intermediate High                      1.3 ft                          1.0-1.8 ft High                                  1.6 ft                          1.1-2.1 ft Observation Extrapolation              1.2 ft                          1.0-1.4 ft Scenarios for average sea-level rise at Key West over 2051-2069, relative to average sea level over 1995-2014, are as follows:
Scenario                              Central Value for 2060          Range for 2060 Low                                    1.0 ft                          0.8-1.2 ft Intermediate Low                      1.2 ft                          1.0-1.4 ft Intermediate                          1.5 ft                          1.2-1.8 ft Intermediate High                      1.9 ft                          1.5-2.5 ft High                                  2.3 ft                          1.8-2.9 ft 11 Interagency Sea Level Rise Scenario Tool: Key West, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION (2022), available at https://sealevel.nasa.gov/task-force-scenario-tool?psmsl_id=188.
5
: 16. Based on the spatial scales of variability of sea level, conclusions drawn from this tide gauge can be reasonably construed to reflect the changes experienced at Turkey Point. As previously recognized by the Nuclear Regulatory Commission, The Miami Beach station was removed from service in 1981, but trends at Miami Beach are well correlated with trends at the Key West station.12 In my expert judgement, it would be appropriate to use the scenarios for Key West in the 2022 Sea Level Rise Technical Report to assess the range of potential future sea-level rise at Turkey Point.
: 17. The 2023 Federal Flood Risk Management Standard Climate-Informed Science Approach (CISA) State of the Science Report, written by the Federal Flood Risk Management Standard (FFRMS) Science Subgroup of the Flood Resilience Interagency Working Group of the National Climate Task Force,13 states:
[i.] Two new, major, authoritative global and regional SLR assessments have been published since 2015, which are the basis for mean SLR guidance under this 2023 CISA Report. Sweet et al. (2017) provided regional SLR projections for the United States for the first time on a 1-degree grid and at tide gauges. In addition, these projections were given exceedance probabilities associated with the Representative Concentration Pathways (RCPs) to assist with determining more likely scenarios. Sweet et al. (2022) updates the global mean SLR scenarios from Sweet et al. (2017) using output drawn directly from IPCC AR6. The underlying SLR science reported in AR6 is also used to calculate exceedance probabilities for each Sweet et al. (2022) global mean SLR scenario around the shared socioeconomic pathways (SSPs), global mean temperature targets, and the inclusion (or not) of lower-confidence ice sheet processes in the model-derived AR6 SLR projections (Working Group 1 2021; Working Group II 2022; Working Group III 2022).14 12 In the Matter of Florida Power & Light Co. (Turkey Point Nuclear Generating Units 6 and 7), CLI-18-01, at 25
: n. 110 (2018), available at https://www.nrc.gov/docs/ML1809/ML18095A117.pdf.
13 Maria Honeycutt et al., FEDERAL FLOOD RISK MANAGEMENT STANDARD CLIMATE-INFORMED SCIENCE APPROACH (CISA) STATE OF THE SCIENCE REPORT, NATIONAL CLIMATE TASK FORCE (Mar. 2023), available at https://www.whitehouse.gov/wp-content/uploads/2023/03/Federal-Flood-Risk-Management-Standard-Climate-Informed-Science-Approach-CISA-State-of-the-Science-Report.pdf.
14 Id. at 22.
6
 
[ii.] Federal agencies should apply this latest interagency Federal guidance for regionally-based SLR projections. Scenarios and time horizons should use a consistent national approach based on risk tolerance and criticality. The regional scenarios, based on the appropriate scenario at the closest tide gauge location or 1-degree grid, should be combined with the coastal hazard projection workflow using methods appropriate to policies, practices, criticality, and consequences.
Agencies should be aware that updates to the scenarios will continue to be made through the Interagency SLR Task Force process, in partnership with the NCA.
Each agency should factor projected regional/local sea level change into Federal investment decisions located as far inland as the extent of estimated tidal influence, now and in the future, using the most appropriate methods for the scale and consequence of the decision. Using the regional SLR scenarios will account for regional differences based on VLM, oceanographic processes, and ice sheet fingerprinting.15
[iii]. For short-term actions (~30 years to 2050), agencies should use the extrapolated trends in 2050, and then choose the SLR scenario curve immediately above the observational extrapolation to account for uncertainty, as follows:
                  - For a given location (tide gauge or grid cell), select the regional tide gauge extrapolation associated with that location (i.e., the region that the tide gauge/grid cell falls within).
                  - Identify the local (for the tide gauge/grid cell) model-derived scenario (e.g.,
Low, Intermediate Low, Intermediate, Intermediate High, High).
                  - That particular local model-derived scenario then becomes the planning curve or equivalent freeboard for the upcoming 30-year time horizon.16
: 18. For Turkey Point, application of the guidance laid out in the 2023 Federal Flood Risk Management Standard Climate-Informed Science Approach (CISA) State of the Science Report would imply use of either the Intermediate High (central value of 1.3 ft in 2050 and 1.9 ft in 2060, relative to a 1995-2014 baseline) or High (central value of 1.6 ft in 2050 and 2.3 ft in 2060, relative to a 1995-2014 baseline) scenarios to generate planning curves when considering the environmental impacts of a 30-year extension of the Turkey Point license. It is my expert opinion that it would be appropriate to follow this and other guidance laid out in the 2023 Federal Flood Risk Management Standard Climate-15 Id.
16 Id. at 23.
7
 
Informed Science Approach (CISA) State of the Science Report in relation to Turkey Point.
I declare under penalty of perjury under the laws of the United States that the foregoing is true and correct to the best of my knowledge.
Dated: November 7, 2023 Robert Kopp Attachments:
A. William V. Sweet et al., GLOBAL AND REGIONAL SEA LEVEL RISE SCENARIOS FOR THE UNITED STATES, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION (Feb. 2022),
available at https://oceanservice.noaa.gov/hazards/sealevelrise/noaa-nos-techrpt01-global-regional-SLR-scenarios-US.pdf.
B. Interagency Sea Level Rise Scenario Tool: Key West, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION (2022), available at https://sealevel.nasa.gov/task-force-scenario-tool?psmsl_id=188.
C. IPCC (AR6 WORKING GROUP 1 CONTRIBUTION): TECHNICAL
 
==SUMMARY==
INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE (2021), available at https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_TS.pdf.
D. IPCC (AR6 WORKING GROUP 1 CONTRIBUTION): CHAPTER 9, INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE (2021), available at https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_Chapter09.pdf.
E. Sea Level Projection Tool: Key West, INTERGOVERNMENTAL PANEL ON CLIMATE CHANGE (2021), available at https://sealevel.nasa.gov/ipcc-ar6-sea-level-projection-tool?psmsl_id=188.
F. Maria Honeycutt et al., FEDERAL FLOOD RISK MANAGEMENT STANDARD CLIMATE-INFORMED SCIENCE APPROACH (CISA) STATE OF THE SCIENCE REPORT, NATIONAL CLIMATE TASK FORCE (Mar. 2023), available at https://www.whitehouse.gov/wp-content/uploads/2023/03/Federal-Flood-Risk-Management-Standard-Climate-Informed-Science-Approach-CISA-State-of-the-Science-Report.pdf.
8
 
ATTACHMENT A Global and Regional Sea Level Rise Scenarios for the United States Credit ©GaryJ.Kohn
 
Cover Image: Flooding from 15-knot northerly winds on Smith Island, Maryland, on November 23, 2015.
Credit ©Gary J. Kohn National Oceanic and Atmospheric Administration U.S. Department of Commerce National Ocean Service Silver Spring, Maryland February, 2022 Recommended Citation:
Sweet, W.V., B.D. Hamlington, R.E. Kopp, C.P. Weaver, P.L. Barnard, D. Bekaert, W. Brooks, M. Craghan, G. Dusek, T. Frederikse, G. Garner, A.S. Genz, J.P. Krasting, E. Larour, D. Marcy, J.J. Marra, J. Obeysekera, M. Osler, M. Pendleton, D. Roman, L. Schmied, W. Veatch, K.D. White, and C. Zuzak, 2022: Global and Regional Sea Level Rise Scenarios for the United States: Up-dated Mean Projections and Extreme Water Level Probabilities Along U.S. Coastlines. NOAA Technical Report NOS 01. National Oceanic and Atmospheric Administration, National Ocean Service, Silver Spring, MD, 111 pp. https://oceanservice.noaa.gov/hazards/sealevelrise/noaa-nos-techrpt01-global-regional-SLR-scenarios-US.pdf Global and Regional Sea Level Rise Scenarios for the United States l ii
 
Global and Regional Sea Level Rise Scenarios for the United States:
Updated Mean Projections and Extreme Water Level Probabilities Along U.S. Coastlines Authors William V. Sweet                                                          John P. Krasting NOAA National Ocean Service                                              NOAA Geophysical Fluid Dynamics Laboratory Benjamin D. Hamlington                                                    Eric Larour NASA Jet Propulsion Laboratory,                                          NASA Jet Propulsion Laboratory, California Institute of Technology                                        California Institute of Technology Robert E. Kopp                                                            Doug Marcy Rutgers University                                                        NOAA National Ocean Service Christopher P. Weaver                                                    John J. Marra U.S. Environmental Protection Agency                                      NOAA National Centers for Environmental Patrick L. Barnard                                                        Information U.S. Geological Survey                                                    Jayantha Obeysekera David Bekaert                                                            Florida International University NASA Jet Propulsion Laboratory,                                          Mark Osler California Institute of Technology                                        NOAA National Ocean Service William Brooks                                                            Matthew Pendleton NOAA National Ocean Service                                              Lynker Michael Craghan                                                          Daniel Roman U.S. Environmental Protection Agency                                      NOAA National Ocean Service Gregory Dusek                                                            Lauren Schmied NOAA National Ocean Service                                              FEMA Risk Management Directorate Thomas Frederikse                                                        Will Veatch NASA Jet Propulsion Laboratory                                            U.S. Army Corps of Engineers California Institute of Technology                                        Kathleen D. White Gregory Garner                                                            U.S. Department of Defense Rutgers University                                                        Casey Zuzak Ayesha S. Genz                                                            FEMA Risk Management Directorate University of Hawaii at Mnoa, Cooperative Institute for Marine and Atmospheric Research Notice:
Mention of a commercial company or product does not constitute an endorsement by NOAA. Use of information from this publication for publicity or advertising purposes concerning proprietary products or the tests of such products is not authorized.
Global and Regional Sea Level Rise Scenarios for the United States l iii
 
Table of Contents List of Figures ............................................................................................................................................................ vi List of Tables ............................................................................................................................................................... x Executive Summary ................................................................................................................................................. xii Section 1: Introduction ...............................................................................................................................................1 Section 2: Future Mean Sea Level: Scenarios and Observation-Based Assessments .................................6 2.1. Overview of Regional and Global Sea Level Rise ........................................................................................6 2.2. Updates from Sweet et al. (2017) ....................................................................................................................9 2.2.1 Inclusion of Near-Term Time Period (2020-2050) ...................................................................................9 2.2.2 GMSL Scenario Divergence and Tracking .................................................................................................9 Box 2.1: Uncertainties .......................................................................................................................................... 10 2.2.3 Updates to the 2017 Sea Level Scenarios..............................................................................................11 2.2.4 Observation-Based Extrapolations .........................................................................................................12 2.3. Near-Term Sea Level Change (2020-2050) ............................................................................................. 13 2.4. Long-Term Sea Level Change (2050-2150) .............................................................................................20 2.5. Scenario Divergence and Tracking ............................................................................................................. 24 Section 3: Extreme Water Levels and Changing Coastal Flood Exposure .................................................. 28 3.1. Overview of Extreme Water Levels and Coastal Flooding ..................................................................... 28 3.2. Regional Frequency Analysis of Tide-Gauge Data................................................................................... 31 3.3. Average Event Frequencies of Extreme Water Levels ........................................................................... 32 3.4. Methods to Localize the Gridded Extreme Water Level Event Probabilities ................................... 35 3.5. The Changing Nature of Coastal Flood Exposure .................................................................................37 Box 3.1: Wave Contributions to Extreme Water Levels .................................................................................41 Section 4: Use Cases ............................................................................................................................................. 43 4.1. Mapping of NOAA High Tide Flood Thresholds and Flood Frequencies ........................................... 43 4.2. Application of Scenarios, Observation-Based Extrapolations, and Extreme Water Levels .......... 45 4.3. Growing Risk to Combined Storm and Wastewater Systems from Sea Level Rise ........................ 53 4.4. Use of InSAR Technology for Determining Regional Vertical Land Motion and Its Suitability for Computing Long-Term Sea Level Rise Projections .................................................................................. 55 Section 5: Conclusions .......................................................................................................................................... 60 Section 6: Acknowledgments .............................................................................................................................. 63 Section 7: References ............................................................................................................................................ 64 Global and Regional Sea Level Rise Scenarios for the United States l iv
 
Appendix ...................................................................................................................................................................74 Section A1: Tables and Figures .........................................................................................................................74 Section A2: Methods Appendix: Extreme Water Levels and Alaska Coastal Flood Height................86 A2.1: Data and Regional Frequency Analysis .................................................................................................... 86 A2.2: Gridded (Regional) Extreme Water Level Probabilities........................................................................ 88 A2.3: Localized Extreme Water Level Probabilities......................................................................................88 A2.3.1: Local Index Estimates from Short-Term Installations.......................................................................... 89 A2.3.2: Obtaining a Local Index from Tide Range Information ..................................................................... 91 A2.3.3: Uncertainties Using Alternative Methods to Estimate EWLlocal Probabilities ............................. 93 A2.3.4: Adjusting Local Extreme Water Level Probabilities to Time Periods............................................ 93 A2.4: Alaska Coastal Flood Heights ................................................................................................................93 Section 8: Acronyms .............................................................................................................................................. 95 Global and Regional Sea Level Rise Scenarios for the United States l v
 
List of Figures Figure 1.1: ................................................................................................................................................................................................... 1 Schematic (not to scale) showing physical factors affecting coastal flood exposure. Due to the clear and strong relative sea level rise signal (i.e., combination of sea level rise and sinking lands), the probability of flooding and impacts are increasing along most U.S. coastlines.
Figure 1.2: ................................................................................................................................................................................................. 2 a) Observed annual global mean sea level (GMSL) change from global tide gauges (blue line), along with the sum (orange line) of contributions from thermal expansion (thermosteric) and four distinct water-mass-driven increases in GMSL. b) GMSL change (blue line) as shown in a) with the annual average relative sea level change measured by tide gauges around the contiguous United States (black line; with a linear regression estimate of 28 cm of sea level rise from 1920 to 2020). (Adaptation of Frederikse et al., 2020).
Figure 1.3: ..................................................................................................................................................................................................3 a) Annual probability density and b) annual expected exceedances for daily highest water levels relative to the 1983-2001 mean higher high water (MHHW) tidal datum showing increases in NOAA minor, moderate, and major high tide flooding (HTF) probabilities/frequencies due to relative sea level (RSL) rise at the NOAA tide gauge in Charleston, South Carolina.
Figure 2.1................................................................................................................................................................................................... 8 Regional sea level linear rates of rise (mm/year) from satellite altimetry over three different time periods: (a) 1993-2006, (b) 2007-2020, and (c) 1993-2020. Linear rates of change of relative sea level (ocean and land height changes) from tide gauges over the same time period are also shown (circles).
Figure 2.2: ............................................................................................................................................................................................... 14 Observation-based extrapolations using tide-gauge data and five Scenarios, in meters, for a) global mean sea level and b) relative sea levels for the contiguous United States from 2020 to 2050 relative to a baseline of 2000.
Median values are shown by the solid lines, while the shaded regions represent the likely ranges for the observa-tion-based extrapolations and each scenario. Altimetry data (1993-2020) and tide-gauge data (1970-2020) are overlaid for reference.
Figure 2.3: ...............................................................................................................................................................................................18 Observation-based extrapolations and five regionalized global mean sea level scenario projections, in meters, of relative sea levels for eight coastal regions around the United States from 2020 to 2050 relative to a baseline of 2000. Median values are shown by the solid lines, while the shaded regions represent the likely ranges for the observation-based extrapolations and each scenario. Tide-gauge data (1970 to 2020) are overlaid for reference, along with satellite altimetry observations, which do not include contributions from vertical land motion.
Figure 2.4: ..............................................................................................................................................................................................20 Relative sea level rise, in meters, in 2050 for the a) Intermediate-Low and b) Intermediate-High scenarios relative to the year 2000.
Figure 2.5: ..............................................................................................................................................................................................24 Regional deviations of relative sea level from the global mean sea level (GMSL; in meters) value for each scenario in 2100. To obtain the regional projection in 2100 for each scenario, the mapped values must be added to the GMSL value for the associated scenario.
Figure 2.6: ..............................................................................................................................................................................................25 Divergence of global mean sea level (GMSL) trajectory and scenarios. The time series shows the observa-tion-based GMSL trajectory and the five GMSL scenarios from 2000 to 2100. The dots denote where each scenar-io significantly (2 sigma) deviates from the a) observation-based trajectory and from the b) Intermediate scenario.
Global and Regional Sea Level Rise Scenarios for the United States l vi
 
Figure 2.7: ............................................................................................................................................................................................... 27 Proportions of the contributions from different IPCC AR6 sea level trajectories to each of the five global mean sea level (GMSL) rise scenarios used in this report: Low, Intermediate-Low, Intermediate, Intermediate-High, and High. The IPCC AR6 trajectories are Low Emissions; Low Emissions, LC (where LC indicates inclusion of low-con-fidence ice-sheet processes); Intermediate Emissions; Intermediate Emissions, LC; High Emissions; and High Emissions LC. The emissions pathways associated with the IPCC AR6 trajectories are as follows: Low Emissions
    = Shared Socioeconomic Pathway (SSP) 1-1.9 or SSP1-2.6; Intermediate Emissions = SSP 2-4.5; High Emissions =
SSP3-7.0 or SSP5-8.5. Shifts between different GMSL rise scenarios approximately reflect the relative odds of be-ing close to a given scenario under different emissions pathways; e.g., the Low scenario is much more plausible under a low emissions pathway, while Intermediate and higher scenarios are much more likely to be associated with high emissions pathways, as well as with low-confidence ice-sheet processes.
Figure 3.1: ................................................................................................................................................................................................29 National median rate of minor high tide flooding and relative sea level, in meters, from 98 NOAA tide gauges along U.S. coastlines outside of Alaska used to monitor and track flood-frequency changes (from Sweet et al.,
2021). Relative sea levels reference the lowest annual (1925) level.
Figure 3.2: ..............................................................................................................................................................................................32 Regional Frequency Analysis 1-degree grids and local index values (u) relative to local mean higher high water tidal datum at the NOAA tide gauges used in this study.
Figure 3.3:............................................................................................................................................................................................... 33 a) Empirical probability densities of hourly water levels and their daily maxima measured by the NOAA tide gauge at The Battery (New York City), as well as the tidal datums of mean lower low water (MLLW), great diurnal tide range (GT), local high tide flood (HTF) heights, and the local index (u) used to localize the RFA-gridded EWL for this location (see Figure A2.2f). All values are referenced to the mean higher high water (MHHW) tidal datum and shown in b) as a return interval curve with the 95% confidence interval (2.5% and 97.5% levels) normalized to year 2020 RSLs.
Figure 3.4: .............................................................................................................................................................................................. 34 Current (circa 2020 relative sea levels) EWLlocal that a) occur annually on average and b) have a 0.01-year aver-age event frequency. Note: the scales in the two figures are not the same, and to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
Figure 3.5: ..............................................................................................................................................................................................35 Comparison between (a-c) this studys EWLlocal to those of NOAA (Zervas, 2013) based on a GEV fit of annual highest water levels and to (d-f) the stillwater (storm surge, tides, and wave set-up) components of FEMA used in their Flood Insurance Study at the 0.01-year, 0.1-year, and 0.5-year average event frequency levels.
Figure 3.6: .............................................................................................................................................................................................. 37 a) Map showing active NOAA tide gauges indicating Grand Bay, Mississippi, which has about 4-5 years of hourly data, b) tide range to local index (u) regression relative to the 1983-2001 tidal datum epoch with fit equation, goodness of fit (R2), and associated root mean square error (RMSE) for the surrounding region, c) RMSE for estimates of u based on 1-19 years of consecutive data over the 2001-2019 period based on the regional tide gauges for the surrounding region; and d) a 2020 EWLlocal return level curve for Grand Bay using a local index (u) from tide range regression. Note: to be useful for decision-making, a conversion to land-based heights (e.g.,
NAVD88) should be made.
Figure 3.7: ...............................................................................................................................................................................................38 NOAA minor (red layer: land between mean higher high water [MHHW] and minor high tide flood [HTF] height above MHHW), moderate (orange layer), and major (yellow layer) HTF maps showing a regional layered map with individual layer panes to the right for a) Charleston, South Carolina, and b) West Palm Beach, Florida. MHHW for 1983-2001 is the shoreline edge. Note: to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
Global and Regional Sea Level Rise Scenarios for the United States l vii
 
Figure 3.8: ..............................................................................................................................................................................................39 Average event frequencies in 2020 of a) minor high tide flooding (HTF); b) number of days (as compared to events) of HTF estimated in NOAAs annual outlook (Sweet et al., 2021) and regression between events and days; c) average event frequencies in 2020 of moderate HTF; and d) average event frequencies in 2020 of major HTF. Flood height-severity definitions are from NOAA (Sweet et al., 2018) and, specifically for Alaska locations, from Sweet et al. (2020b).
Figure 3.9: ..............................................................................................................................................................................................40 Coastal high tide flooding (HTF) frequencies projected at 2050 applying the sea level scenario that upper-bounds the regional observation-based extrapolations for NOAA a) minor, b) moderate, and c) major HTFs Figure Box 3.1: .......................................................................................................................................................................................42 Water level contribution due to a) wave set-up and b) wave swash; c) percent contribution of wave-driven water levels (i.e., wave run up = wave set-up and swash) relative to all components: tide, storm surge, and waves; and d) percent contribution of wave set-up relative to the sum of tide, storm surge, and wave set-up based on model reanalysis of Vitousek et al. (2017).
Figure 4.1:................................................................................................................................................................................................45 Maps of the NOAA minor, moderate, and major high tide flooding layers for a) Charleston, South Carolina, and b) West Palm Beach, Florida (as in Figure 3.7 but providing average event frequencies for each layer). Note: the shoreline on these maps is mean higher high water, but to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
Figure 4.2: ..............................................................................................................................................................................................46 Tide gauges selected for the application of sea level scenarios and extreme water level methods.
Figure 4.3: .............................................................................................................................................................................................. 47 a) RSL projections for the scenarios providing the upper bound to observation-based extrapolations to 2060 for the selected tide gauges. The corresponding scenario for each tide gauge is shown in parentheses in the leg-end. b) RFA-based EWL (see Section 3) return level curves relative to the 1983-2001 MHHW tidal datum. Notes:
(1) to be useful for decision-making, a conversion to land-based heights (e.g., geodetic datum such as NAVD88) should be made. (2) Average event frequency (x-axis label) is the reciprocal of average recurrence interval, which is also known as return period.
Figure 4.4: ..............................................................................................................................................................................................48 Recurrent flood frequency estimates for a) Sewells Point (Norfolk), Virginia, and b) Galveston Pier 21, Texas. For both, the relative sea level projection for the scenarios and the return level are the same as in Table 4.1. Note: to be useful for decision-making, a conversion of the return level to land-based heights (e.g., geodetic datum such as NAVD88) should be made.
Figure 4.5: ..............................................................................................................................................................................................50 Conceptual illustration of increasing exceedance probability (hence decreasing average recurrence interval) that assumes that the location parameter is a function of the magnitude of the relative sea level rise.
Figure 4.6: ...............................................................................................................................................................................................51 a) Average recurrence interval (due to rising RSL) curves (T versus T_0) at each tide gauge using the selected scenarios RSL projection (see Table 4.1). b) Risk curves as a function of design life: stationary (black curve), actual risk resulting from incorporating the sites RSL scenario projection (red curve), and risk curve for a specific risk (blue curve).
Figure 4.7: ...............................................................................................................................................................................................55 Location of combined stormwater and sewer system outfalls that are likely draining regions exposed to HTF within the Camden, New Jersey, region, with the minor (red: MHHW to 0.58 m [1.9 feet] above MHHW), moderate (orange: MHHW to 0.86 m [2.8 feet] above MHHW), and major (yellow: MHHW to 1.25 m [4.1 feet] above MHHW)
HTF layers stacked in the enlarged map and individual layers mapped to the right. Note: heights are relative to the 1983-2001 tidal epoch, and to be useful for decision-making, a conversion to land-based heights (e.g.,
NAVD88) should be made.
Global and Regional Sea Level Rise Scenarios for the United States l viii
 
Figure 4.8: .............................................................................................................................................................................................. 57 Comparison of vertical land motion (VLM) rate estimates (mm/year) from a) the scenario-based framework used in this report, and b) GPS-imaging estimates from Hammond et al. (2021). c) The difference between GPS-derived rates and scenario-derived rates and d) a comparison of the VLM estimates at the U.S. tide-gauge locations are also shown. Negative values of VLM reflect subsidence, while positive values reflect uplift.
Figure 4.9: ..............................................................................................................................................................................................58 Map showing VLM rates (mm/year) for the Hampton Roads region displayed on top of satellite imagery. Higher rates of subsidence are indicated by darker orange colors. Of particular interest is the range of rates in such a small region (e.g., on the order of up to 5 mm/year difference in places). Based on Buzzanga et al. (2020).
Figure A1.1: ............................................................................................................................................................................................. 74 Region definitions for observation-based extrapolations and scenarios in Section 2. These regions are used both to group tide gauges and also to generate regional averages for the gridded scenarios. A bathymetry mask is used to define the regions for the gridded scenarios.
Figure A1.2: ............................................................................................................................................................................................ 75 Shown for each tide gauge record with at least 30 years of record length between 1970 and 2020 are a) range, in meters, between median projection of Low and High Scenarios in 2050, and b) difference, in meters, between median observation-based extrapolation and Intermediate scenario in 2050.
Figure A2.1: .............................................................................................................................................................................................86 NOAA tide gauges used in the regional frequency analysis to generate extreme water level probabilities for U.S coastlines.
Figure A2.2: ........................................................................................................................................................................................... 87 Example of data from grid number 46415 showing exceedances above each local index (u) relative to the 1983-2001 mean higher high water (MHHW) tidal datum at a) Kings Point, New York; b) The Battery, New York; c) Ber-gen Point, New York; and d) Sandy Hook, New Jersey, which are e) aggregated into a single dataset and f) fit by a Generalized Pareto Distribution to form a return level interval curve for the grid.
Figure A2.3: ...........................................................................................................................................................................................89 Additional tide-gauge data available from NOAA that can be used to localize the 1-degree gridded set of region-al frequency analysis-based extreme water level probabilities.
Figure A2.4: ...........................................................................................................................................................................................90 Root mean square error for regional estimates of flood indices (u) based on 1-19 years of consecutive data over the 2001-2019 period, based on regional sets of tide gauges used in this study. Note: these regions are not the same as those shown in Figure A1.1 and used to describe results in Sections 2 and 3 of the report.
Figure A2.5: ...........................................................................................................................................................................................92 Tide range to local index (u) regressions with equations, goodness of fit (R2), and root mean squared error (RMSE) shown by regions. Note: all local indices (u) are relative to the 1983-2001 tidal datum epoch. In the equations, y represents the local index (u) and x represents tide range.
Figure A2.6: ...........................................................................................................................................................................................94 a) Quadratic regression of U.S. West Coast minor flood heights of NOAAs National Weather Service, following methods of Sweet et al. (2020b), to obtain a minor HTF definition for Alaskas coastline. The NOAA flood heights for b) minor, c) moderate, and d) major HTF are shown relative to mean higher high water.
Global and Regional Sea Level Rise Scenarios for the United States l ix
 
List of Tables Table 2.1:...................................................................................................................................................................................................15 Observation-based extrapolations and five scenarios, in meters, for global mean sea level and relative sea lev-el for the contiguous United States from 2020 to 2050 relative to a baseline of 2000. Median [likely ranges]
are shown.
Table 2.2: .................................................................................................................................................................................................19 Observation-based extrapolation and regionalized global mean sea level scenario-based estimates, in meters, of relative sea level in 2050 relative to a baseline of 2000 for eight coastal regions of the United States. Median
[likely ranges] are shown. The two scenarios that bound the median observation-based extrapolation are also provided for each region and indicated by red dividing lines. In regions where the observation-based extrapo-lation is the same as a particular scenario, the scenario is indicated in red text and the bounding scenarios can be assumed to be the next higher or lower scenario (e.g., the Intermediate bounds the Northeasts observa-tion-based extrapolation).
Table 2.3: ................................................................................................................................................................................................20 Global mean sea level and contiguous United States scenarios, in meters, relative to a 2000 baseline.
Table 2.4: ................................................................................................................................................................................................22 IPCC warming level-based global mean sea level projections. Global mean surface air temperature anomalies are projected for years 2081-2100 relative to the 1850-1900 climatology. Sea level anomalies are relative to a 2005 baseline (adapted from Fox-Kemper et al., 2021). The probabilities are imprecise probabilities, represent-ing a consensus among all projection methods applied. For imprecise probabilities >50%, all methods agree that the probability of the outcome stated is at least that value; for imprecise probabilities <50%, all methods agree that the probability of the outcome stated is less than or equal to the value stated.
Table 2.5: ................................................................................................................................................................................................23 Scenarios of relative sea level, in meters, for eight coastal regions of the United States in 2100 and 2150 relative to a baseline of 2000. Median values are shown.
Table 3.1: ..................................................................................................................................................................................................30 Physical processes affecting U.S. coastal water levels and their temporal and spatial scale properties (mod-ification of Sweet et al., 2017). Extreme water levels, which, as measured by tide gauges, generally exclude high-frequency wave effects, include processes between tsunami and ocean-basin variability and, to a lesser extent, low-frequency changes or trends associated with land ice melt/discharge, thermal expansion, and vertical land motion.
Table 3.2: ................................................................................................................................................................................................. 41 Annual average event frequencies for NOAA-defined minor, moderate, and major HTF heights by region that were typical (median values) in 1990, under current (circa 2020) sea levels and projected to occur considering the upper-bounding scenario of the observations-based extrapolations in 2050 (see Table 2.2).
Table 4.1:..................................................................................................................................................................................................46 Tide-gauge locations, scenarios bounding the observation-based extrapolations, and the extreme value dis-tribution Generalized Pareto Distribution (GPD) model parameters estimated using the regional frequency analysis (RFA).
Table 4.2: ................................................................................................................................................................................................49 Summary of design parameters to constrain the average event frequency, N, to 1 per year by 2060 (end-year of the design life). The 2005-2060 RSL projections are the local values associated with the scenarios providing the upper bound to the regional observation-based extrapolations shown in Table 2.2. Note: to be useful for decision-making, a conversion of the return level to land-based heights (e.g., geodetic datum such as NAVD88) should be made.
Global and Regional Sea Level Rise Scenarios for the United States l x
 
Table 4.3: .................................................................................................................................................................................................51 The parameters of generalized extreme value computed using the peaks-over-threshold Generalized Pareto Distribution model (Coles 2001).
Table 4.4: ................................................................................................................................................................................................53 Results of the risk-based design for all tide gauges shown in Figure 4.2. Average recurrence interval (ARI) is listed and is the reciprocal of average event frequency. Values in the last column have been rounded to the closest 5-year interval. Note: to be useful for decision-making, a conversion of the return level to land-based heights (e.g., geodetic datum such as NAVD88) should be made.
Table A1.1:................................................................................................................................................................................................ 76 Projections methods employed.
Table A1.2: .............................................................................................................................................................................................. 76 Offsets, in meters, for different time periods and for each region considered in Section 2. These offsets are as-sessed using the trajectory determined from the available tide-gauge data in each region.
Table A1.3: .............................................................................................................................................................................................. 77 Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Global and Regional Sea Level Rise Scenarios for the United States l xi
 
Executive Summary This report and accompanying datasets from the U.S. Sea Level Rise and Coastal Flood Hazard Scenarios and Tools Interagency Task Force provide 1) sea level rise scenarios to 2150 by decade that include esti-mates of vertical land motion and 2) a set of extreme water level probabilities for various heights along the U.S. coastline. These data are available at 1-degree grids along the U.S. coastline and downscaled specifical-ly at NOAA tide-gauge locations. Estimates of flood exposure are assessed using contemporary U.S. coastal flood-severity thresholds for current conditions (e.g., sea levels and infrastructure footprint) and for the next 30 years (out to year 2050), assuming no additional risk reduction measures are enacted.
This effort builds upon the 2017 Task Force report (Sweet et al., 2017). In particular, the set of global mean sea level rise scenarios from that report are updated and downscaled with output directly from the United Nations Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report (AR6; IPCC, 2021a),
through the efforts of the NASA Sea Level Change Team; updates include adjustments to the temporal tra-jectories and exceedance probabilities of these scenarios based upon end-of-century global temperatures.
As with the 2017 report, these global mean sea level rise scenarios are regionalized for the U.S. coastline.
In addition, methodology supporting the U.S. Department of Defense Regional Sea Level (DRSL) database1 (Hall et al., 2016) is adapted for the extreme water level dataset newly developed for this report.
This report will be a key technical input for the Fifth National Climate Assessment (NCA5). These data and information are being incorporated into current and planned agency tools and services, such as NOAAs Sea Level Rise Viewer and Inundation Dashboard,2 NASAs Sea Level Change Portal,3 and others. Although the intent of this report is not to provide authoritative guidance or design specifications for a specific project, it is intended to help inform Federal agencies, state and local governments, and stakeholders in coastal commu-nities about current and future sea level rise to help contextualize its effects for decision-making purposes.
Key Message #1:
Multiple lines of evidence provide increased confidence, regardless of the emissions pathway, in a narrower range of projected global, national, and regional sea level rise at 2050 than previously reported (Sweet et al., 2017).
* Both trajectories assessed by extrapolating rates and accelerations estimated from historical tide gauge observations, and model projections, fall within the same range in all cases, giving higher confidence in these relative sea level (RSL; land and ocean height changes) rise amounts by 2050.
* Relative sea level along the contiguous U.S. (CONUS) coastline is expected to rise on average as much over the next 30 years (0.25-0.30 m over 2020-2050) as it has over the last 100 years (1920-2020).
* Due to processes driving regional changes in sea level, there are similar regional differences in both the modeled scenarios and observation-based extrapolations, with higher RSL rise along the East (0-5 cm higher on average than CONUS) and Gulf Coasts (10-15 cm higher) as compared to the West (10-15 cm lower) and Hawaiian/Caribbean (5-10 cm lower) Coasts.
* The projections do not include natural year-to-year sea level variability that occurs along U.S.
coastlines in response to climatic modes such as the El Nino-Southern Oscillation.
1 https://drsl.serdp-estcp.org/
2 https://coast.noaa.gov/digitalcoast/tools/slr.html 3
https://sealevel.nasa.gov/
Global and Regional Sea Level Rise Scenarios for the United States l xii
 
Key Message #2 By 2050, the expected relative sea level (RSL) will cause tide and storm surge heights to increase and will lead to a shift in U.S. coastal flood regimes, with major and moderate high tide flood events occurring as frequently as moderate and minor high tide flood events occur today. Without additional risk-reduction mea-sures, U.S. coastal infrastructure, communities, and ecosystems will face significant consequences.
* Minor/disruptive high tide flooding (HTF; about 0.55 m above mean higher high water [MHHW]4) is projected to increase from a U.S average frequency of about 3 events/year in 2020 to >10 events/
year5 by 2050.
* Moderate/typically damaging HTF (about 0.85 m above MHHW) is projected to increase from a U.S. average frequency of 0.3 events/year in 2020 to about 4 events/year in 2050.
* Major/often destructive HTF (about 1.20 m above MHHW) is projected to increase from a U.S. aver-age frequency of 0.04 events/year in 2020 to 0.2 events/year by 2050.
* Across all severities (minor, moderate, major), HTF along the U.S. East and Gulf Coasts will largely continue to occur at or above the national average frequency.
Key Message #3:
Higher global temperatures increase the chances of higher sea level by the end of the century and beyond.
The scenario projections of relative sea level along the contiguous U.S. (CONUS) coastline are about 0.6-2.2 m in 2100 and 0.8-3.9 m in 2150 (relative to sea level in 2000); these ranges are driven by uncer-tainty in future emissions pathways and the response of the underlying physical processes.
* With an increase in average global temperature of 2&deg;C above preindustrial levels, and not consid-ering the potential contributions from ice-sheet processes with limited agreement (low confidence) among modeling approaches, the probability of exceeding 0.5 m rise globally (0.7 m along the CONUS coastline) by 2100 is about 50%. With 3&deg;-5&deg;C of warming under high emissions pathways, this probability rises to >80% to >99%. The probability of exceeding 1 m globally (1.2 m CONUS) by 2100 rises from <5% with 3&deg;C warming to almost 25% with 5&deg;C warming.
* Considering low-confidence ice-sheet processes and high emissions pathways with warming approaching 5&deg;C, probabilities rise to about 50%, 20%, and 10% of exceeding 1.0 m, 1.5 m, or 2.0 m of global rise by 2100, respectively. These processes are unlikely to make significant contributions with 2&deg;C of warming, but how much warming might be required to trigger them is currently un-known.
* As a result of improved understanding of the timing of possible large future contributions from ice-sheet loss, the Extreme scenario from the 2017 report (2.5 m global mean sea level rise by 2100) is now viewed as less plausible and has been removed. Nevertheless, the potential for in-creased acceleration in the late 21st century and beyond means that the other high-end scenarios provide pathways that could reach this threshold in the decades immediately following 2100 (and continue rising).
* Regionally, the projections are near or higher than the global average in 2100 and 2150 for almost all U.S. coastlines due to the effects from vertical land motion (VLM); gravitational, rotational, and deformational effects due to land ice loss; and ocean circulation changes. Largely due to VLM, RSL projections are lower than the global amounts along the southern Alaska coast and are higher along the Eastern and Western Gulf coastlines.
4 Mean higher high water (MHHW) level is estimated over the 1983-2001 tidal epoch period and, in this case, is considered a fixed elevation that does not change with sea level rise.
5 The extreme value statistical methods in this report do not directly resolve frequencies >10 events/year.
Global and Regional Sea Level Rise Scenarios for the United States l xiii
 
Key Message #4 Monitoring the sources of ongoing sea level rise and the processes driving changes in sea level is critical for assessing scenario divergence and tracking the trajectory of observed sea level rise, particularly during the time period when future emissions pathways lead to increased ranges in projected sea level rise.
* Efforts are under way to narrow the uncertainties in ice-sheet dynamics and future sea level rise amounts in response to increasing greenhouse gas forcing and associated global warming.
* Early indicators of changes in sea level rise trajectories can serve to trigger adaptive manage-ment plans and are identified through continuous monitoring and assessment of changes in sea level (on global and local scales) and of the key drivers of sea level change that most affect U.S.
coastlines, such as ocean heat content, ice-mass loss from Greenland and Antarctica, vertical land motion, and Gulf Stream system changes.
Global and Regional Sea Level Rise Scenarios for the United States l xiv
 
Section 1: Introduction Sea level rise driven by global climate change is a clear and present risk to the United States today and for the coming decades and centuries (USGCRP, 2018; Hall, Weaver et al., 2019). Sea levels will continue to rise due to the oceans sustained response to the warming that has already occurredeven if climate change mitigation succeeds in limiting surface air temperatures in the coming decades (Fox-Kemper et al., 2021).
Tens of millions of people in the United States already live in areas at risk of coastal flooding, with more moving to the coasts every year (NOAA NOS and U.S. Census Bureau, 2013). Rising sea levels and land sub-sidence are combining, and will continue to combine, with other coastal flood factors, such as storm surge, wave effects, rising coastal water tables, river flows, and rainfall (Figure 1.1), some of whose characteristics are also undergoing climate-related changes (USGCRP, 2017). The net result will be a dramatic increase in the exposure and vulnerability of this growing population, as well as the critical infrastructure related to transportation, water, energy, trade, military readiness, and coastal ecosystems and the supporting services they provide.
Figure 1.1: Schematic (not to scale) showing physical factors affecting coastal flood exposure. Due to the clear and strong relative sea level rise signal (i.e., combination of sea level rise and sinking lands), the probability of flooding and impacts are increasing along most U.S. coastlines.
Global mean sea level (GMSL) rise is a direct effect of climate change, resulting from a combination of ther-mal expansion of warming ocean waters and the addition of water mass into the ocean, largely associated with the loss of ice from glaciers and ice sheets. These processes are well understood for the recent past, and their contributions have been estimated for the 20th century (Figure 1.2a). With regard to increasing sea levels associated with climate change, the questions are when and how much, rather than if (USGCRP, 2017; Hall, Weaver et al., 2019). Increases in GMSL provide an important indicator of the changing climate, but it is the sea level rise on local and regional scalesmeasured by the global network of tide gauges and satel-litesthat is most relevant for coastal communities around the world. Regional and local sea level rise has not been and will not be uniform in time or space. Rather, sea levels change locally for a variety of reasons, such as vertical land motion (VLM), which can exacerbate the effects of the rising ocean. For context, where-as GMSL has risen by about 17 cm over the last 100 years (1920-2020), with noted acceleration since about 1970, relative sea level (RSL) averaged along the contiguous United States (CONUS) has risen about 28 cm over the same period with similar onset of acceleration (Figure 1.2b).
Global and Regional Sea Level Rise Scenarios for the United States l 1
 
Figure 1.2: a) Observed annual global mean sea level (GMSL) change from global tide gauges (blue line), along with the sum (orange line) of contributions from thermal expansion (thermosteric) and four distinct water-mass-driven increases in GMSL. b)
GMSL change (blue line) as shown in a) with the annual average relative sea level change measured by tide gauges around the contiguous United States (black line; with a linear regression estimate of 28 cm of sea level rise from 1920 to 2020). (Adaptation of Frederikse et al., 2020).
While this long-term and upward shift in mean RSL is the underlying driver of changes to the Nations coasts, extreme water levels (EWLs) occurring against the background of this shifting sea level baseline are respon-sible for many of the recurring and event-based impacts. In this report, EWLs are explicitly assumed to be ocean-related changes measured by tide gauges (e.g., high tides and storm surges), which typically do not measure other contributors such as direct rainfall or river flow unless they are positioned upstream of major river systems (Moftakhari et al., 2016). Specifically, EWLs are considered as those occurring with an average event frequency between 0.01 events/year (often referred to as the 1% annual chance event) and 10 events/
year. This range mostly spans the flood frequency of NOAA high tide flood (HTF) severity levels (minor, mod-erate, and major). HTF levels are nationally calibrated against NOAAs National Weather Service and local emergency managers depth-severity thresholds used in weather forecasting and impact communications (NOAA, 2020) to provide a consistent coastal-climate resilience standard (Sweet et al., 2018).
Higher sea levels amplify the impacts of storm surge, high tides, coastal erosion, and wetland loss, even absent any changes in storm frequency and intensity. Because of threshold effects related to changes mea-sured relative to a fixed elevation (Figure 1.3a), even the relatively small increases in sea level over the last several decades have led to greatly increased frequency of flooding6 at many places along the U.S. coast (Figure 1.3b). Much of the coastline is already close to a flood regime shift, with respect to flood frequency (and presumably damages). That is, only about a 0.3-0.7 m height difference currently separates infrequent, moderate/typically-damaging and major/often-destructive HTF from minor/disruptive nuisance HTF (Sweet et al., 2018), whose impacts are already remarkable throughout dozens of densely populated coastal cities (Moore and Obradovich, 2020). Decades ago, powerful storms were what typically caused coastal flooding, 6
The definition of a flood in this report is typically meant to refer to a water level associated with impacts rather than the occurrence of natural phenomena.
Global and Regional Sea Level Rise Scenarios for the United States l 2
 
but today, due to RSL rise, even common wind events and seasonal high tides regularly cause HTF with-in coastal communities, affecting homes and businesses, overloading stormwater and wastewater sys-tems, infiltrating coastal groundwater aquifers with saltwater, and stressing coastal wetlands and estuarine ecosystems.
At multiple locations along the U.S. coastline, the annual frequency of minor HTF is accelerating and has more than doubled over the past couple of decades, turning it from a rare event into a recurrent and disrup-tive problem (Sweet and Park, 2014; Sweet et al., 2018; USGCRP, 2018). For example, the trends in minor/
disruptive HTF have grown from about 5 days in 2000 to 10-15 days in New York City and Norfolk, Virginia, in 2020; in Miami, Florida, and Charleston, South Carolina, annual frequencies have grown from 0-2 days to about 5-10 days over the same period. These increases will continue, further accelerate, and spread to more locations over the next couple of decades (Sweet et al., 2021; Thompson et al., 2021). Thus, accurate projections of ongoing and future sea level rise and assessments that integrate across processes and tem-poral and spatial scales are key inputs to planning efforts and a key goal of this report.
Figure 1.3: a) Annual probability density and b) annual expected exceedances for daily highest water levels relative to the 1983-2001 mean higher high water (MHHW) tidal datum showing increases in NOAA minor, moderate, and major high tide flood-ing (HTF) probabilities/frequencies due to relative sea level (RSL) rise at the NOAA tide gauge in Charleston, South Carolina.
The Sea Level Rise and Coastal Flood Hazard Scenarios and Tools Interagency Task Force (hereafter Task Force) was jointly convened at the direction of the White House Resilience Council in 2015 under the U.S.
Global Change Research Program (USGCRP), the Subcommittee on Ocean Sciences and Technology (SOST),
and the National Ocean Council (NOC). This was in recognition of the strong need and demand for authori-tative, consistent, and accessible sea level rise and associated coastal hazard information for the entire U.S.
coastline, coordinated across the relevant Federal agencies, to serve as a starting point for on-the-ground coastal preparedness planning and risk management activities. The goal of the Task Force, since its incep-tion, has been to develop the necessary products through sustained and coordinated participation of key agencies, based on the best available science, including regional science and expertise when possible and appropriate. The goal has also been to incorporate those products into user-friendly mapping, visualization, and analysis tools made easily accessible through existing agency portals serving specific partners and stakeholders, as well as interagency venues such as the National Climate Assessment (NCA), the U.S. Cli-mate Resilience Toolkit, and others.
Global and Regional Sea Level Rise Scenarios for the United States l 3
 
The Task Force focused its initial efforts on the development of an interagency report (Sweet et al., 2017),
providing updated GMSL rise scenarios focused primarily on 2100 and integrating these GMSL rise scenarios with regional factors contributing to sea level change to produce, for the first time, a set of RSL scenarios for the entire U.S. coastline. These scenarios were also a major technical input to Volumes I and II of the Fourth NCA (NCA4; USGCRP 2017, 2018) and have been widely used in the development of state (e.g., Florida7 and Virginia [CCRM, 2019]) and local agency adaptation plans (e.g., Pensacola, Florida,8 and Portland, Maine [One Climate Future, 2019]), and processes for anticipating and managing future coastal risks.
The Task Forces first report (Sweet et al., 2017) built upon the most current scenarios at that time (e.g., Parris et al., 2012; Kopp et al., 2014; Hall et al., 2016) and estimated the full possible range for GMSL rise by 2100 as being bounded by 0.3 m on the low end, representing a simple linear extrapolation of the GMSL rate since the early 1990s, and by 2.5 m on the high end, representing an extreme ice-sheet melt/discharge scenario.
This 0.3-2.5 m range was discretized and aligned with emissions-based, conditional probabilistic storylines and global model projections into six GMSL rise scenarios: Low, Intermediate-Low, Intermediate, Intermedi-ate-High, High, and Extreme, corresponding to GMSL rise by 2100 of 0.3 m, 0.5 m, 1.0 m, 1.5 m, 2.0 m, and 2.5 m, respectively. These GMSL rise scenarios were then used to derive regional RSL responses on a 1-de-gree grid covering the coastlines of the U.S. mainland, Alaska, Hawaii, the Caribbean, and the Pacific Island territories, as well as at the precise locations of tide gauges along these coastlines.
This current report takes the Sweet et al. (2017) report as its starting point, updating the GMSL scenarios and the associated local and regional RSL projections to reflect recent advances in sea level science, as well as expanding the types of scenario information provided to better serve stakeholder needs for coastal risk man-agement and adaptation planning. As with the 2017 report, this iteration will also serve as a key technical input to the NCA, in this case NCA5. Specific updates in this report include the following:
* While this report still uses the same nomenclature as the NOAA 2017 GMSL scenarios, it draws upon new science of the Intergovernmental Panel on Climate Change (IPCC) Sixth Assessment Report (AR6; Fox-Kemper et al., 2021; Garner et al., 2021) to provide updated temporal trajectories and exceedance probabilities based on different levels of global warming. One effect is that the associated RSL projections for the U.S. coastline (gridded and at individual tide-gauge locations) differ in timing and magnitude as compared to the NOAA 2017 projections.
* In addition, in leveraging this updated science, including a longer observational record, improved understanding of ice-sheet dynamical processes, and better-constrained models, this report pro-vides a more comprehensive and detailed assessment of the distinct types and range of uncer-tainties associated with the GMSL rise scenarios, particularly at the high end.
* By utilizing 50-year regional sets of tide-gauge data, observation-based rates and accelerations are extrapolated to the year 2050 to identify the scenario projections aligning with current RSL trajectories.
* Lastly, gridded EWL probabilities are provided, along with methods to localize them along most U.S. coastlines, to contextualize each of the regionalized sea level scenarios across a range of flood frequencies under current standards, from recurrent tidal flooding to major storm-surge flooding, out to 2050.
To frame the remainder of this report, it is important to emphasize the distinction between describing scien-tific progress, in terms of current understanding and key uncertainties, and translating such advances in the scientific knowledge base into actionable science. The latter requires sustained engagement by groups such as NOAAs Office of Coast Management and the Sea Grant program with users, stakeholder groups, and as-sociated boundary organizations regarding their specific planning and decision contexts. Our development 7
https://floridadep.gov/rcp/florida-resilient-coastlines-program/documents/proposed-rule-development-draft-62s-7-sea-level 8
https://storymaps.arcgis.com/stories/e812723f69ad4a618c8f5f8b08cb208e Global and Regional Sea Level Rise Scenarios for the United States l 4
 
of scenarios in this report is grounded in the principles of risk-based framing for climate assessment (King et al., 2015; Weaver et al., 2017; Sutton, 2019; Kopp et al., 2019) and is consistent with adaptation pathways approaches for long-term planning. What we thus aim to provide are screening-level (suitable for first-order assessment) products appropriate for framing and bounding important problems in coastal risk assessment and management, along with contextualization of the underlying science and illustrative case studies. For example, consistent with this purpose, this report aims to provide the underlying scientific information to de-velop both planning- and bounding-type scenarios across the full spectrum of coastal risk; that is, 1) planning scenarios intended to frame near- to mid-term decision contexts and/or longer-term decisions with high-risk tolerance or ability to adjust plans, which address the question, What is most likely to happen? and 2) bound-ing scenarios designed to set the envelope of possible future outcomes, which can be used to stress-test long-term objectives, gauge the when, not if a given level of sea level rise might be reached, and address the question How bad could things get? What this report does NOT provide is official guidance nor design specifications for a specific project.
Section 2 describes advances in the understanding of the drivers of mean sea level since the 2017 report, discusses the use of observations for a near-term trajectory assessment, and provides the updated GMSL rise scenarios and their associated regional RSL projections. Section 3 focuses on high-frequency EWLs, including a regional frequency analysis of historical NOAA tide-gauge data to develop a set of EWL probabil-ities for assessing and projecting (to 2050) across a range of flood levels. Section 4 applies these scenarios and projections in illustrative use-case examples. Section 5 provides a summary of the report findings, as well as conclusions and next steps.
Global and Regional Sea Level Rise Scenarios for the United States l 5
 
Section 2: Future Mean Sea Level: Scenarios and Observation-Based Assessments Since Sweet et al. (2017), the observations and available data records of both sea level change and the asso-ciated processes have increased in number and length. In part due to these observations, our understanding of the drivers of sea level change has improved. There have also been significant advances in modeling how these processes will cause sea level to change in the future. This has led to an improved understand-ing of the possible trajectory of future sea level rise. In this report, these advances are reflected both in an update to the GMSL scenarios and a change in approach from Sweet et al. (2017). The primary change in approach is in separating this section into two different time periods: 1) near term (2020-2050) and 2) long term (2050-2150). There is also a section discussing divergence of the GMSL scenarios and tracking that is particularly relevant during the transition between the near- and long-term time periods. In the remainder of this section, a brief overview of the drivers of global and regional sea level rise is provided. Next, updates to Sweet et al. (2017) are discussed, and the motivation and scientific justification for these changes are given.
Finally, the updated information for the two time periods, along with the transition between these periods, is provided.
2.1. Overview of Regional and Global Sea Level Rise Over long, multidecadal to centennial timescales, the primary drivers of changes in GMSL are thermal ex-pansion due to the heating of the ocean and the addition of water mass associated with ice-mass loss from the ice sheets and glaciers. Other changes in the movement of water between ocean and land, including from groundwater depletion and water impoundment, have a secondary impact on GMSL, although they can increase in importance for certain time periods (see Frederikse et al., 2020). During the 20th century, GMSL estimated from tide-gauge records has been explained by the individual processes contributing to it (see Figure 1.2a; Frederikse et al., 2020). More recently, observed GMSL from satellite altimetry over the past 15 years has been explained using the in situ measurements of the Argo profiling floats and the observations of water-mass change from the GRACE and GRACE-FO satellites (WCRP, 2018). On shorter timescales, consid-erable interannual and decadal variability in GMSL is linked primarily to variations in terrestrial water storage and driven heavily by the El Nino-Southern Oscillation (ENSO; Boening et al. 2012; Fasullo et al., 2013; Pie-cuch and Quinn, 2016; Hamlington et al., 2020a, 2020b].
At the regional level, rates of sea level rise can deviate significantly from the globally averaged rate. Sea lev-el rise is not uniform across the globe; rather, it manifests as relative sea level (RSL) rise that also responds to several key factors important at regional and local scales (Kopp et al., 2014; Sweet et al., 2017; Hamlington et al., 2020a; Fox-Kemper et al., 2021). On short timescales and in short records, natural variations on inter-annual to decadal timescales can impact estimates of rates and accelerations. On long timescales, however, there are three primary causes of regional variations in estimated rates and accelerations: 1) sterodynamic sea level change; 2) gravitational, rotational, and deformational (GRD) changes due to contemporary ice-mass loss and the movement of water between land and ocean; and 3) vertical land movement (VLM; sub-sidence or uplift) due to glacial isostatic adjustment (GIA), tectonics, sediment compaction, groundwater and fossil fuel withdrawals, and other non-climatic factors. These three causes are discussed briefly below.
Sterodynamic sea level changes are those that arise from changes in the oceans circulation (currents) and its density (temperature and salinity). Sea level rise associated with sterodynamic sea level change is the combination of global mean thermosteric rise associated with global ocean warming and local deviations from the global mean due to ocean dynamic processes. It is these changes in ocean dynamics that lead to regional differences. Focusing on possible causes of long-term sterodynamic sea level changes for the U.S. coastlines, future changes in the Atlantic meridional overturning circulation (AMOC) are particular-ly relevant. The IPCC AR6 (IPCC, 2021a) determined that it is very likely that the AMOC will decline in the future, although there is still disagreement as to the extent of this decline. A weakening AMOC will lead to an increase in sea level along the coastal Northeast and Southeast regions (Yin et al., 2009; Krasting et al.,
Global and Regional Sea Level Rise Scenarios for the United States l 6
 
2016; see Figure A1.1 for region definitions). For the Northwest and Southwest coastal regions, ENSO plays a substantial role in interannual sea level change, although there is no clear evidence for a sustained shift in ENSO that will result in a long-term increase or decrease in sea level. Some models project future sea level changes associated with ocean dynamics to be large in magnitude in some locations, but these projections remain uncertain (Fox-Kemper et al., 2021).
The ice-mass loss from ice sheets and glaciers to the ocean has a strong influence on regional sea level.
Changes in Earths GRD responses dictate the spatial distribution of water across the global ocean (Farrell and Clark, 1976; Milne and Mitrovica, 1998; Mitrovica et al., 2001). These so-called sea level fingerprints are important to determining regional sea level rise. Mass loss causes a sea level fall in the near-field, a reduced sea level rise at intermediate distances, and a greater-than-global-average sea level rise at larger distances.
For U.S. coastlines, particularly in the Northeast, this means that a similar amount of ice-mass loss in Antarc-tica will have a larger impact than ice-mass loss in Greenland. Similarly, ice-mass loss in Greenland leads to bigger increases in sea level along the Northwest and Southwest coastal regions than along the Northeast coastal region. At any time horizon, the regional sea level rise associated with GRD will be driven both by the amount of ice that is being lost and the source of that ice. These regional fingerprints are tied to project-ed trajectories of mass loss from the associated source. Changes in terrestrial water storage (groundwater withdrawal and dam building) also have an associated fingerprint, but the regional contribution is generally smaller than that from the ice sheets and glaciers.
Lastly, the VLM considered in this report refers to either subsidence or uplift that occurs in coastal regions and can lead to the change in the height of sea level relative to land. VLM is not a singular phenomenon but instead results from various processes that display different patterns in space and time. These patterns have different impacts from place to place, especially in coastal settings where many of them operate at the same time. For much of the coastal United States, subsidence is driven on local scales by groundwater and fossil fuel withdrawal and on larger scales by GIA. However, in some regions such as southern Alaska, GIA leads to high rates of uplift in coastal regions. GIA is the ongoing response of the solid earth due to ice-mass changes in the past, particularly the deglaciation after the last glacial maximum. GIA induces VLM, in particu-lar subsidence along the U.S. East Coast, as well as changes in the gravity field, which cause local sea level changes. Accurate future projections of VLM require an understanding of the underlying processes and the time and space scales on which they vary. Currently, and in this report, VLM projections are based in part on analysis of past observations. If activities change in a particular location (e.g., reduction in groundwater pumping), an associated change in the rate of VLM will not necessarily be captured. Modeling of future VLM under a range of possible scenarios is not currently available over large scales. (See the vertical land motion use case in Section 4.4 for more information.)
Beyond these processes that impact long-term changes in sea level, there is also considerable natural (or unforced) climate variability that can lead to significant, albeit temporary, changes to sea level on the order of years or even decades. In many of the available observational records, it can be a challenge to distin-guish between these natural signals and those processes discussed above. As an example, in Figure 2.1, the regional rates of sea level rise along U.S. coastlines are shown for the first half (a, 1993-2006), second half (b, 2007-2020), and full (c, 1993-2020) satellite altimeter record (which do not measure VLM effects), along with overlaid tide-gauge rates (which measure VLM effects) measured over the same time period. A signifi-cant shift occurs from the first half of the record to the second half, with high sea level rise rates found along all coastlines of CONUS from 2007 to 2020. For the Northwest and Southwest coastal regions, in particular, the rate was near 0 for the first half of the record before shifting to almost 10 mm/year over the second half, driven by decadal variability linked to the Pacific Decadal Oscillation (PDO; e.g., Bromirski et al., 2011; Ham-lington et al., 2021). For the full record, there is considerably less spatial variability, with most regions ap-proaching the globally averaged rate of 3.1 mm/year.
Global and Regional Sea Level Rise Scenarios for the United States l 7
 
In this section of the report, the contribution of natural variability is not assessed directly, but its importance and contribution should be considered when looking at observed rates and assessing possible sea level at a specific time in the future. In other words, there is an envelope of naturally occurring sea level variability on top of the sea level rise discussed here that needs to be included to estimate sea level at a particular loca-tion at a specific time in the future. A depiction of the relationship between sea level rise and this envelope is provided in Figure 1.3. The median of the distribution increases over time as a result of the rising sea levels, while other sea level variability on a range of timescales contributes to the spread around this central value.
Figure 2.1: Regional sea level linear rates of rise (mm/year) from satellite altimetry over three different time periods:
(a) 1993-2006, (b) 2007-2020, and (c) 1993-2020. Linear rates of change of relative sea level (ocean and land height changes) from tide gauges over the same time period are also shown (circles).
Global and Regional Sea Level Rise Scenarios for the United States l 8
 
2.2. Updates from Sweet et al. (2017)
One of the main structural changes from the Sweet et al. (2017) report to this one is a specific emphasis on the near-term time period, 2020-2050. There is also a detailed discussion of GMSL scenario divergence and tracking that becomes particularly important in the transition from the near term to the long term. The motivation for the focus on these two topics is given below. Following this explanation, the primary advances in the sea level scenarios and assessments of future sea level are discussed in two subsections. The first provides an overview of the science and framework advancements that have led to an update of the scenar-ios first presented in Sweet et al. (2017). The second covers the inclusion of observation-based assessments of near-term sea level change for the first time.
2.2.1. Inclusion of Near-Term Time Period (2020-2050)
The dedicated focus on the near-term time period represents a new element in this report. Motivation for this change is provided briefly here. With increasing record lengths, the impact of natural sea level variability on estimated rates and accelerations diminishes, revealing more of the underlying climate change signal (see Figure 2.1c, for example). Tide gauges surrounding the U.S. coastlines provide records exceeding 100 years in some locations, and the satellite altimeter record is nearing three decades in length. Recent studies have assessed the degree to which rates and accelerations estimated from these records are reflective of the long-term increase in sea level (via satellite altimetry; e.g., Fasullo and Nerem, 2018; Richter et al., 2020) and RSL (via tide gauges; e.g., Wang et al., 2021). These studies suggest that with appropriate consideration of uncertainty, observation-based extrapolations can be informative in the near term. In this report, an assess-ment based solely on extrapolation of the observed rates and acceleration out to 2050 is used for trajectory tracking and a comparison to the GMSL and regional scenarios. These trajectories serve as an additional line of evidence for near-term sea level rise and provide a mostly independent (observational VLM information is shared in both) comparison to the model-based scenario. To maintain a distinction between estimates arising from observations and those coming from model-derived GMSL scenarios, the observation-based assess-ments are referred to in this report as extrapolations or trajectories and not as projections. These terms are also preceded by observation-based whenever used.
Beyond this renewed observational focus, the inclusion of this near-term time period is motivated by the fact that for certain decision types, short time horizons and nearer-term assessments are most relevant. For the typical lifetime of buildings and infrastructure in coastal areas, for example, a 30-year planning horizon has particular relevance (e.g., Fu, 2020; Hinkel et al., 2018). Additionally, flexible adaptation pathways and solu-tions typically require significant lead times on upgrades or replacements of coastal structures that necessi-tate assessments across a range of timescales. (Haasnoot et al., 2013, 2019; Bloeman et al., 2018; Werners et al., 2021; Hall, Harvey, and Manning, 2019). Knowing whether adaptation actions are required within the next 30 years or afterwards informs decisions about initial designs, the adaptations required, and the metrics that would trigger adaptation.
2.2.2. GMSL Scenario Divergence and Tracking After 2050, the assessments and comparisons made using the observation-based extrapolations of future sea level rise become less informative and should be made with caution. This is because uncertainty in the current estimates of rates and accelerations leads to large projected ranges and because current estimates may not be reflective of shifts or process changes that may occur in the future with additional emissions and global warming, resulting in increasing divergence between the future GMSL scenarios after 2050.
During the transition from near- to long-term assessments, an understanding of when the GMSL scenarios will diverge and what drives this divergence becomes increasingly important. Two types of uncertainty are important to consider in this context: uncertainty in physical processes and uncertainty in future emissions and ensuing warming. Although there are possible alternative definitions and framings, as used in this report, process uncertainty (Box 2.1) is associated with how well we currently understand why sea level has changed in the past and how it will change in the future. Stated another way, how well do we understand and model Global and Regional Sea Level Rise Scenarios for the United States l 9
 
the processes that will combine to impact sea level at a specific time and location in the future? This un-certainty is also reflected in the likely range of future sea level rise for a given GMSL scenario. The spread between the five GMSL rise scenarios is intended to reflect the range of potential future emissions pathways and associated warming levels that depends highly on global socioeconomic factors that have yet to unfold.
This unknown future pathway leads to what is referred to here as emissions uncertainty (Box 2.1).
At some point in the future, the separation between GMSL rise scenarios will overtake the process uncertain-ty associated with individual GMSL rise scenarios. In other words, scenario dependence will emerge, and it will be possible to distinguish between the observation-based trajectories associated with two neighboring GMSL rise scenarios. In general, these time periods are important for connecting the near-term similarities between scenarios to the time period where scenarios diverge rapidly. An effort is made here to understand when divergence of the GMSL rise scenarios might occur and to link them to possible future warming and emissions pathways. This analysis then serves as the foundation for process-based monitoring that could be useful in determining the trajectory of ongoing sea level rise and, by extension, the possible future sea level rise out to 2150.
Box 2.1: Uncertainties When assessing future changes in sea level, this report con-      Uncertainties in this Report siders two main sources of uncertainty.                          In this report, emissions uncertainty and process uncertainty are combined to generate five sea level scenarios with GMSL Process Uncertainty                                              target values in 2100: Low (0.3 m), Intermediate-Low (0.5 m),
An increase in emissions will cause ice-mass loss, ocean          Intermediate (1 m), Intermediate-High (1.5 m), and High (2 m).
thermal expansion, and local ocean dynamic changes, but the      These sea level scenarios are related to but distinct from the sensitivity of these processes to these forcing changes comes    emissions pathway scenarios in the IPCC AR6.
with uncertainty. For example, the sensitivity of the Antarctic ice sheet is not yet fully understood, leading to a substan-      Natural Variability tial uncertainty in how sea level reacts to forcing changes.      Next to sea level changes caused by changes in GHG forcing, Additionally, the future contributions from processes, such as    many physical processes cause natural variations (e.g., ENSO).
changes in ocean circulation and VLM, that impact RSL change      The scenarios and uncertainty ranges for each scenario and more locally have an associated uncertainty. This uncertainty    for the observation-based trajectories in this report do not in the contribution of these various processes to future RSL      include variations due to natural variability (the decadal sce-change is referred to in this report as process uncertainty.      nario values are 19-year averages that remove most variability effects). Natural variability is not directly considered a source Emissions Uncertainty                                            of uncertainty in the context of this report but does contribute Increasing the amount of greenhouse gases (GHGs) in the at-      to the uncertainty range in the observation-based extrapola-mosphere will trap more heat in the earth system. The amount      tions, as it can influence the estimated rates and accelerations of GHGs in the atmosphere determines the forcing of climate    in observational records. Natural, or non-forced, variations change and its effects, such as changes in temperature and        can also make significant contributions to sea level on a wide sea level rise. Various forcing scenarios describe possible      range of timescales. For example, along the U.S. West coast, GHG emissions pathways, which range from quick emissions          sea levels are higher during El Nino years. When assessing reduction to unmitigated future emissions. In the IPCC AR6        sea level at a specific location and time in the future, the sea (IPCC, 2021a), these possible future pathways are referred to    level contribution from natural variability must be combined as Shared Socioeconomic Pathways (SSPs). The uncertainty in      with the scenarios and trajectories provided here.
the future pathway is referred to as emissions uncertainty.
Global and Regional Sea Level Rise Scenarios for the United States l 10
 
2.2.3 Updates to the 2017 Sea Level Scenarios In order to support decision-making efforts related to future sea level risks, past interagency efforts (Par-ris et al., 2012; Hall et al., 2016; Sweet et al., 2017) have defined a set of GMSL rise scenarios spanning a range from a Low scenario, consistent with no additional GMSL acceleration, to a worst-case, or high-end, Extreme scenario, judged to be at the physically plausible limits based on the scientific literature. In Sweet et al. (2017), these scenarios were developed to span a range of 21st-century GMSL rise from 0.3 m to 2.5 m.
Sweet et al. (2017) built these scenarios upon the probabilistic emissions scenario-driven projections of Kopp et al. (2014). Kopp et al. (2014) combined a variety of different lines of evidenceglobal climate model (GCM) projections, the IPCC AR5 assessment of ice-sheet changes, and structured expert-judgment ice-sheet projections, among other sources of informationto generate distributions of future global and asso-ciated regional sea level changes consistent with low, medium, and high emissions scenarios. Sweet et al.
(2017) filtered the ensemble of different future projections generated by Kopp et al. (2014) to identify those subsets consistent with 0.3 m, 0.5 m, 1.0 m, 1.5 m, 2.0 m, and 2.5 m of 21st-century GMSL rise. These subsets constituted the six Sweet et al. (2017) GMSL scenarios. For most purposes, Sweet et al. (2017) focused on the median of each subset, although 17th and 83rd percentile levels were also reported.
This report retains the Sweet et al. (2017) scenarios (except the Extreme 2.5 m scenario, discussed below),
with the principal difference being updated temporal trajectories and exceedance probabilities now based on global warming levels rather than emissions scenarios. Linking to global warming levels provides a straightforward physical link for the GMSL scenarios and establishes a connection to global temperature monitoring efforts. The updates made in this report reflect the underlying ensemble of future projections based on methods used in the IPCC AR6 (Fox-Kemper et al., 2021; Garner et al., 2021) and listed in Table A1.1. As in Sweet et al. (2017), these projections are filtered based on 21st-century GMSL rise. In other words, projected pathways that intersect the GMSL scenario target values in 2100 are retained and then used to generate the GMSL scenarios from Low to High described here.
In addition to being updated based on the latest generation of GCMs and the IPCC AR6, this set of projec-tions incorporates multiple methods of projecting future ice-sheet changes, which are the major sources of future sea level rise and pose the biggest source of uncertainty in projecting the timing and magnitude of fu-ture possible rise amounts. For Antarctica, this includes emulators derived from two different ice-sheet mod-el intercomparison exercises (Edwards et al., 2021; Levermann et al., 2020), as well as from a single-model study focused on the potentially high-impact but uncertain-likelihood marine ice cliff instability (MICI) mecha-nism (DeConto et al. 2021) and a structured expert-judgment study (Bamber et al, 2019). For Greenland, this includes a single intercomparison-derived emulator (Edwards et al., 2021) and a structured expert-judgment study (Bamber et al., 2019). There is now a broader range of both Antarctic and Greenland potential con-tributions, compared to Sweet et al. (2017). Whereas the high-end scenarios of Sweet et al. (2017) were all dominated by Antarctic contributions, the potential for high Greenland contributions now also adds to these high-end scenarios, and due to its proximity, also drives larger differences along U.S. coastlines.
The use of multiple methods, including methods that consider mechanisms that could substantially increase ice-sheet sensitivity under high emissions scenarios, means that the time path of the higher GMSL scenarios is more realistic than in Sweet et al. (2017), which assumed (based on the underlying Kopp et al. [2014] pro-jections) that ice-sheet loss would accelerate at a constant rate over the remainder of the century. A result is that there is less acceleration in the higher scenarios until about 2050 and greater acceleration toward the end of this century. This has two primary implications. First, despite maintaining the same target values and having the same range between scenarios in 2100, the range covered by the scenarios is smaller in the near term than in Sweet et al. (2017). Second, the likely (17th-83rd percentile) ranges of projections consis-tent with each scenario before and after the 2100 time point used to define the scenarios tend to be broader than in Sweet et al. (2017).
Global and Regional Sea Level Rise Scenarios for the United States l 11
 
An important change from the Sweet et al. (2017) report is the exclusion of the Extreme (2.5 m) scenario in this report. Based on the most recent scientific understanding and as discussed in the IPCC AR6, the uncer-tain physical processes such as ice-sheet loss that could lead to much higher increases in sea level are now viewed as less plausible in the coming decades before potentially becoming a factor toward the end of the 21st century and beyond. A GMSL increase of 2.5 m by 2100 is thus viewed as less plausible, and the asso-ciated scenario has been removed from this report. Nevertheless, the increased acceleration in the late 21st century and beyond means that the other high-end scenarios provide pathways that potentially reach this threshold in the decades immediately following 2100 (and continue rising).
2.2.4. Observation-Based Extrapolations As discussed above, the pathways of the updated GMSL scenarios differ from those presented in Sweet et al. (2017), and the range between the scenarios in the near term is now reduced. This report, for the first time, includes observation-based extrapolations to serve as a near-term (2020-2050) comparison for the scenarios. They can also be viewed as trajectories of current sea level rise. When interpreting these ex-trapolations, they should be considered as an additional line of evidence for near-term sea level rise along-side the model-based GMSL scenarios. They are not intended to replace the GMSL scenarios. Additionally, such observation-based extrapolations, or trajectories, can be potentially misleading if not appropriately constrained. This report makes no detailed assessment of whether the long-term rate and acceleration have emerged from the influence of natural variability in the observational record, although recent studies suggest this could be the case in some regions (Lyu et al., 2014; Richter et al., 2020; Fasullo and Nerem, 2018; Wang et al., 2021). Instead, the observation-based extrapolations are presented as computed and without inter-pretation after several methodological choices were made to generate extrapolations that can be compared to the scenarios and identify those scenarios that bound the 2050 extrapolations. These methodological choices are described briefly below.
First, the rates and accelerations are estimated from the tide-gauge records starting in 1970. Recent studies have shown a consistent acceleration in GMSL since 1970 (Dangendorf et al., 2019; Frederikse et al., 2020),
and this is a primary motivator for the time period chosen. The impact of varying this start date on the region-al scales relied on here was assessed and found to be negligible within a few years of 1970 (more below).
This is not true, as a general statement, when using individual tide-gauge records. Second, the observa-tion-based extrapolations are made only to 2050. Beyond that date, it is assumed that processes not fully represented in the observations could become dominant. Third, the uncertainty in the rate and acceleration associated with the influence of natural variability is accounted for as fully as possible and included in the ex-trapolation. Finally, the extrapolations are made for GMSL, the coastlines of CONUS, and 10 separate coastal regions around the United States and outlying islands (see Figure A1.1 for region definitions). By grouping tide gauges regionally, the influence of localized variability is reduced, and challenges associated with individual tide gauges with incomplete or short records are overcome, thus yielding more useful and narrow-er extrapolated ranges. These regional comparisons also fulfill the intent of providing an additional line of evidence and comparison point to the GMSL scenarios.
For each individual region, the observation-based extrapolation is performed as follows:
: 1. The tide gauges in the region are grouped and combined following the virtual station method (see Frederikse et al., 2020) to generate a monthly time series of RSL from 1920 to present.
: 2. Natural variability is partially removed through regression analysis using climate indices represent-ing the El Nino-Southern Oscillation, Pacific Decadal Oscillation, and North Atlantic Oscillation (see Calafat et al., 2012; Hamlington et al., 2021).
: 3. The rate and acceleration from 1970 to present is computed, and the uncertainty on each term is assessed, accounting for the influence of remaining natural variability (see Hamlington et al., 2021) and serially correlated variability in the tide-gauge record (Bos et al., 2013, 2014).
Global and Regional Sea Level Rise Scenarios for the United States l 12
: 4. The rates, accelerations, and uncertainties are used to generate an ensemble of 5,000 extrap-olations with a baseline year of 2000 and extending to 2050. Median projections and a likely (17th-83rd) range are computed from this ensemble.
Following this procedure, observation-based extrapolations are obtained for GMSL, CONUS, and 8 coast-al regions (Figure A1.1)the Northeast (Maine to Virginia), the Southeast (North Carolina to the east coast of Florida), the Eastern Gulf (west coast of Florida to Mississippi), the Western Gulf (Louisiana to Texas), the Southwest (California), the Northwest (Oregon to Washington), the Hawaiian Islands, and the Caribbean.
Elsewhere in the report, projections are discussed for the Pacific Islands, but due to the availability of tide-gauge data and the geographic range covered by the region, the extrapolations are conducted using only those gauges on the Hawaiian Islands. Observation-based extrapolations are also made for the southern and northern coasts of Alaska and mentioned in the text but not included in the tables below. Differential VLM heavily impacts the tide-gauge records along the southern coastline of Alaska and makes the creation of a regionally representative time series challenging. The observation-based extrapolations for Alaska are thus caveated with increased uncertainty in the underlying regional processes that heavily limit their utility as a comparison to the GMSL scenarios.
2.3. Near-Term Sea Level Change (2020-2050)
In Sweet et al. (2017), the range between the median values of the Low and High GMSL scenarios in 2020, 2030, 2040, and 2050 was 0.05 m, 0.12 m, 0.23 m, and 0.38 m, respectively. As a result of improved science and the updated framework and procedure for generating the GMSL scenarios, the time path of the sce-nariosparticularly the higher scenariosis now more realistic and consistent with current process-based understanding. In this report, the range between the Low and High scenarios in 2020, 2030, 2040, and 2050 is now 0.02 m, 0.06 m, 0.15 m, and 0.28 m, respectively (Table 2.1). In other words, there is less di-vergence between the GMSL scenarios in this near-term time period, which reduces uncertainty in the projected amount of GMSL rise up to the year 2050. The Low scenario remains largely the same between this report and Sweet et al. (2017); this range reduction reflects a downward shift in the higher scenarios in 2050 and times prior, as discussed above. As an example, the projected value in 2050 for the High scenario in this report (~0.4 m) is the same as that for the Intermediate-High projected value in 2050 in Sweet et al.
(2017). In short, while the scenarios continue to be defined by projected values of GMSL increase in 2100, it is important to note that the paths to get to these target values have changed in this report compared to the previous one.
Following the procedure outlined in Section 2.2.4, an observation-based extrapolation of GMSL is comput-ed using the global tide-gauge reconstruction from Frederikse et al. (2020; Figure 2.2a; also see top row of Table 2.1). The extrapolated value of GMSL increase in 2050 relative to a baseline of 2000 is 0.24 m, with a likely (17th-83rd percentile) range between 0.19 m and 0.29 m. A similar extrapolation was made using GMSL data measured by satellite altimeters over 1993-2021, resulting in an estimate of 0.23 m of rise from 2000 to 2050 and in agreement with the results of the tide-gauge extrapolation. Based on the updated GMSL sce-narios, the median of the 2050 observation-based extrapolation is bounded by (i.e., it falls between) the In-termediate-Low and Intermediate scenarios. The likely ranges for the Low and High scenarios do not overlap with the likely range of observation-based extrapolation in 2050, although the very likely ranges (5th-95th percentiles) do overlap. The likely range of the Intermediate-High scenario does overlap with the likely range of the observation-based extrapolation. A similar observation-based extrapolation is completed using only the tide gauges located around CONUS (Figure 2.2b), resulting in a projected increase of 0.38 m in 2050, with a likely range of 0.32 m to 0.45 m. This range for CONUS is again narrower than in Sweet et al. (2017).
Similar to GMSL, this observation-based assessment is bounded by the Intermediate-Low and Intermediate scenarios in 2050.
Global and Regional Sea Level Rise Scenarios for the United States l 13
 
Figure 2.2: Observation-based extrapolations using tide-gauge data and five Scenarios, in meters, for a) global mean sea level and b) relative sea levels for the contiguous United States from 2020 to 2050 relative to a baseline of 2000. Median values are shown by the solid lines, while the shaded regions represent the likely ranges for the observation-based extrapolations and each scenario. Altimetry data (1993-2020) and tide-gauge data (1970-2020) are overlaid for reference.
As a result of the smaller region used and the increased influence of natural variability and VLM, the likely ranges in 2050 for CONUS in both the scenario projections and observation-based extrapolations are larger than those associated with the GMSL scenarios themselves. The likely range from the observation-based extrapolation does overlap with the likely ranges from both the Low and High scenarios. This is both a reflec-tion of the larger range in the extrapolation for CONUS and the narrower range between the High and Low scenarios in this report. A key takeaway from this assessment is that on global and national scales, two lines of evidence (observations and GMSL scenarios) are consistent out to 2050 and support a narrower range in possible near-term sea level change than provided in Sweet et al. (2017). As discussed previously, this is consistent with and a result of the improved process-based understanding and projection approach that has been incorporated in this report.
The observation-based extrapolations are also computed for 10 coastal regions of the United States. Only 8 of these regions are shown in the tables and figures below, with the coastlines of Alaska covered sepa-rately in the text. As in the global and national cases, the observation-based extrapolations are extended out to 2050. Following the procedure outlined in section 2.2.4, tide gauges within each of these regions are combined into a single time series prior to extrapolating estimated rates and accelerations. Building on the discussion in section 2.2.4, the motivation for doing these assessments regionally as opposed to at each individual tide gauge location is two-fold. First, the observation-based extrapolations are intended to serve as a comparison to the model-based GMSL scenarios. Outside the possibility of very localized VLM, the processes included in the regionalized GMSL scenarios are generally spatially coherent over the re-gions considered. Indeed, the selection of specific regions is driven by process-based similarities mostly associated with ocean dynamics and large-scale VLM. Grouping the tide gauges and generating regional comparisons yields a closer analog to the information contained in the scenarios. The regional averages also reduce the influence of local signalsincluding VLM and other natural ocean variabilitythat can influence extrapolations and associated ranges. Second, some of the individual tide gauges around the U.S. coastlines have records that either do not span the full time period from 1970 to 2020 or contain data gaps. Generating Global and Regional Sea Level Rise Scenarios for the United States l 14
 
Table 2.1: Observation-based extrapolations and five scenarios, in meters, for global mean sea level and relative sea level for the contiguous United States from 2020 to 2050 relative to a baseline of 2000. Median [likely ranges] are shown.
Global Mean Sea Level 2020                    2030                      2040                    2050 Obs. Extrapolation        0.07 [0.06, 0.08]        0.12 [0.11, 0.13]        0.18 [0.16, 0.19]      0.24 [0.19, 0.29]
Low              0.06 [0.05, 0.07]        0.09 [0.08, 0.10]          0.12 [0.11, 0.13]        0.15 [0.14, 0.17]
Intermediate-Low          0.07 [0.06, 0.07]        0.11 [0.09, 0.12]          0.15 [0.13, 0.17]      0.20 [0.18, 0.23]
Intermediate          0.07 [0.07, 0.09]        0.13 [0.11, 0.15]        0.19 [0.16, 0.23]      0.28 [0.22, 0.32]
Intermediate-High        0.08 [0.07, 0.10]        0.14 [0.11, 0.20]        0.23 [0.18, 0.32]        0.37 [0.27, 0.46]
High              0.08 [0.07, 0.10]        0.15 [0.11, 0.22]        0.27 [0.18, 0.39]        0.43 [0.31, 0.57]
Contiguous United States 2020                    2030                      2040                    2050 Obs. Extrapolation        0.11 [0.09, 0.13]      0.19 [0.16, 0.21]        0.28 [0.23, 0.32]        0.38 [0.32, 0.45]
Low              0.12 [0.09, 0.15]        0.18 [0.14, 0.23]        0.25 [0.19, 0.31]        0.31 [0.24, 0.39]
Intermediate-Low          0.13 [0.10, 0.16]      0.20 [0.15, 0.25]        0.28 [0.22, 0.34]        0.36 [0.28, 0.44]
Intermediate            0.13 [0.10, 0.16]      0.21 [0.16, 0.26]        0.30 [0.23, 0.37]        0.40 [0.31, 0.49]
Intermediate-High          0.13 [0.10, 0.16]      0.22 [0.16, 0.28]        0.33 [0.24, 0.43]        0.46 [0.35, 0.61]
High                0.13 [0.10, 0.16]      0.22 [0.17, 0.29]        0.35 [0.26, 0.47]      0.52 [0.39, 0.68]
regional time series alleviates these challenges and allows us to provide generalized comparisons and assessments about the match between observations and model-based scenarios along the U.S. coastlines.
These regional comparisons then provide an additional line of evidence for the possible overall trajectory of sea level in the near term. The result is shown in Figure 2.3, with corresponding values in Table 2.2 for each of the eight regions and compared to the scenarios in each region.
The regional differences in the observation-based extrapolations and scenarios in Figure 2.3 are consistent with the current process-based understanding of sea level rise. Processes such as ocean dynamics, the GRD response to contemporary ice-mass loss (i.e., fingerprints), and coastal VLM lead to differences between the eight regions. Additionally, uncertainty ranges on the extrapolations can be bigger or smaller depending on the number of tide gauges in a particular region and the influence of natural variability on the rate and acceleration estimates. To demonstrate this regionalization, Figure 2.4 shows these regional variations of sea level in 2050 for the Intermediate-Low and Intermediate-High scenarios. In 2050, the regional variation in future sea levels does not change significantly between scenarios. Although the values increase from the Intermediate-Low scenario to the Intermediate-High scenario, the east-west difference in sea level rise is similar. Higher values for both scenarios are found along the entire East and Gulf Coasts. Subsidence leads to the highest rates along the Gulf Coast, driven by regional and local factors, such as river sediment com-paction and withdrawal of subsurface fluids (Dokka, 2011; NGS, 2001; Rydlund and Densmore, 2012). Along the East Coast, subsidence is generally associated with the large-scale process of GIA, with fluid extraction being an issue in some areas (Frederikse et al., 2017; Karegar et al., 2016). Beyond VLM, many of the re-gional differences are driven by differences in the ocean dynamic variability. For example, the sterodynamic contribution from 2000 to 2050 in the Northeast is more than double the sterodynamic contribution in the Southwest. This regional difference is similarly reflected in the observation-based extrapolations in 2050. It should be noted that this difference arises from higher-than-global-average projections for the Northeast as opposed to lower-than-global-average projections for the Southwest, which tracks very closely to the GMSL values shown in Table 2.1.
Global and Regional Sea Level Rise Scenarios for the United States l 15
 
For the observation-based extrapolations, the largest estimates of sea level rise in 2050 are found along the entire Gulf Coast (Table 2.2). The Western Gulf has the highest extrapolated values in 2050, driven by high rates of coastal subsidence in the region and consistent with the scenarios discussed above. The Northwest and Southwest coastal regions have the lowest observation-based extrapolations to 2050. For the purposes of offering a comparison to the scenarios, the scenarios that either bound or track the median of the obser-vation-based extrapolations are provided (denoted by red text or markers in Table 2.2). Two regions track the Intermediate-Low scenario (Northeast and Hawaiian Islands), and two regions track the Intermediate scenar-io (Southwest and Caribbean). The Intermediate-Low to Intermediate scenarios bound the Northwest, and the Intermediate to Intermediate-High scenarios bound the Southeast and Western Gulf regions. Finally, the Intermediate-High to High scenarios bound the Eastern Gulf region. With only the exceptions of the low-end scenarios in the Southwest and Eastern Gulf, the likely ranges from the observation-based extrapolations have at least some overlap with the likely ranges of all the scenarios within a given region. This is due to a combination of the larger uncertainty on the observation-based assessments at these regional levels for an individual scenario and the narrower ranges between the median values of each GMSL scenario found in this report compared to Sweet et al. (2017). While not shown in Table 2.2, the observation-based extrapola-tion for the northern coast of Alaska in 2050 (median value of 0.27 cm) is bracketed by the Intermediate and Intermediate-High scenarios. The extrapolation of the southern coast of Alaska leads to a large RSL de-crease in 2050 and is inconsistent with the scenario median values. As mentioned above, this is a result of challenges in generating a representative tide-gauge time series to use in the extrapolation.
As a note on the interpretation of the results provided in this near-term section, the regional comparisons between the observation-based extrapolations and scenarios need to be considered in the context of the global comparison in Figure 2.2. The regional scenarios are intrinsically linked to their associated GMSL tar-get values in 2100. In an ideal framework that perfectly represented the regionalization of these GMSL sce-narios and the relevant regional processes, separate comparisons on a regional level would be unnecessary.
In other words, all regions and locations would track the same GMSL scenario. Since this is not the case, if a particular region deviates from the others, it would be an indication that either the observation-based extrap-olation for that region is biased high or low or that the framework used to generate the regionalization of the GMSL scenarios is not adequately representing the contribution of a regional process. Since the observed GMSL trajectory is near the Intermediate-Low scenario, as shown here, based on the current understanding of the processes driving regional RSL, it is not expected that a particular region would track a much higher scenario. These regional comparisons during the near-term time period then serve two potential purposes:
: 1) they provide an additional line of evidence along with the GMSL and CONUS comparisons for the near-term trajectory of sea level rise, and 2) they can serve to identify cases when the contributions of regional processes may be tracking differently than represented by the regionalization of the GMSL scenarios.
As a general assessment of these two purposes, the likely ranges of all but one of the regions are either bounded on one side by the Intermediate scenario or tracks a scenario neighboring the Intermediate sce-nario, showing some level of consistency with the GMSL and CONUS comparisons. This provides additional confidence in the narrower range (when compared to Sweet et al., 2017) of sea level rise at the regional level out to 2050 presented in this report. The Eastern Gulf is the only region bounded by the High scenario.
The high observation-based extrapolation for the Eastern Gulf should be interpreted with caution, as it does not necessarily mean a higher scenario is applicable compared to other regions. As a possible explanation, unresolved natural ocean variability in the observational record could lead to an observation-based extrap-olation that is biased high. Such variability would need to be low-frequencyor long periodto significantly impact a rate and acceleration estimated in a 50-year record. For all regions considered here, it is likely that natural variability still contributes to the median observation-based extrapolation, and as seen in Figure 2.1, this variability has a substantial impact on the coastlines of the United States. This influence of natural variability on rates and accelerations is captured to the extent possible in the likely ranges of the observa-tion-based extrapolations, and these likely ranges should be considered in tandem with the median values Global and Regional Sea Level Rise Scenarios for the United States l 16
 
when assessing near-term trajectories. Beyond the possible influence of natural variability, there may also be a mismatch in the process representation between the observations and regionalized, model-based GMSL scenarios that leads to a projection that is too low in the latter. One possibility is non-linear or unresolved VLM in the region. The regionalized GMSL scenarios consider only long-term linear rates of VLM, while the observation-based extrapolations could represent a shift in the rate of VLM in the estimated acceleration.
An explanation of regional differences between observation-based extrapolations and model-based scenar-ios requires additional investigation, likely on a tide gauge-by-tide gauge basis. As a first step in this direc-tion, the range between Low and High scenarios at each individual tide gauge (considering only the tide gauges with at least 30 years of data102 of the full set of 121) is provided in Figure A1.2a, and the depar-ture between the observation-based extrapolation and Intermediate scenario at each individual tide gauge is shown in Figure A1.2b. These figures show that the range between Low and High scenarios is generally lower than 20 cm in 2050 at the local level and that most observation-based extrapolations are within 15 cm of the Intermediate scenario in 2050. Of the 102 tide gauges used in this report, 65 have observation-based extrapolations that fall within the narrower Low to High ranges in 2050, and 80 of these 102 are within 15 cm of the Intermediate scenario. The majority of those falling below the Low scenario are found in the Northwest and southern Alaska regions, and the majority of those exceeding the High scenario are found in the two Gulf regions. This supports the regional comparisons shown in Figure 2.3 and Table 2.2 while also conveying that there is general agreement and consistency between the ranges of the observation-based extrapola-tions and regionalized GMSL scenarios even on a local, tide gauge-by-tide gauge level. A more definitive assessment of why some regions track higher (e.g., Eastern Gulf) or lower relative to others requires further analysis that should be done with consideration of the associated uncertainty and ranges.
As a general concluding statement on this near-term section, the link between the regional and global scenarios needs to be considered when drawing conclusions at the regional level based on the observa-tion-based extrapolations. In practice, regionally identifying the scenario that upper-bounds the observa-tion-based extrapolation at year 2050 (Table 2.2) may help compensate for potential interannual variability when projecting sea level for a particular location. The associated uncertainties in the approaches adopted here do emphasize the importance of ongoing monitoring using the observations and the need to update trajectories. As records continue to lengthen, likely ranges on near-term assessments will narrow. Additional-ly, satellite altimeter records are reaching sufficient length to be important in such monitoring. As a final note, the same framework used for extrapolating the observations forward can also be used to assess the increas-esor offsetsobserved over different recent time periods. These offsets are useful for adjusting baselines of the scenarios and are provided for each region in Table A1.2.
Global and Regional Sea Level Rise Scenarios for the United States l 17
 
Figure 2.3: Observation-based extrapolations and five regionalized global mean sea level scenario projections, in meters, of relative sea levels for eight coastal regions around the United States from 2020 to 2050 relative to a baseline of 2000. Median values are shown by the solid lines, while the shaded regions represent the likely ranges for the observation-based extrapola-tions and each scenario. Tide-gauge data (1970 to 2020) are overlaid for reference, along with satellite altimetry observations, which do not include contributions from vertical land motion.
Global and Regional Sea Level Rise Scenarios for the United States l 18
 
Table 2.2: Observation-based extrapolation and regionalized global mean sea level scenario-based estimates, in meters, of relative sea level in 2050 relative to a baseline of 2000 for eight coastal regions of the United States. Median [likely ranges]
are shown. The two scenarios that bound the median observation-based extrapolation are also provided for each region and indicated by red dividing lines. In regions where the observation-based extrapolation is the same as a particular scenario, the scenario is indicated in red text and the bounding scenarios can be assumed to be the next higher or lower scenario (e.g., the Intermediate bounds the Northeasts observation-based extrapolation).
Median Observation                            Intermediate-                      Intermediate-Low                            Intermediate                              High            Bounding Extrapolations                                Low                                High Scenarios Northeast 0.40              0.36                0.40            0.43              0.49              0.54 Int-Low
[0.30, 0.47]      [0.27, 0.45]          [0.31, 0.49]    [0.34, 0.54]      [0.38, 0.64]      [0.40, 0.69]
Southeast 0.41              0.28                0.32            0.36              0.43              0.49 Int-Int-High
[0.32, 0.50]      [0.20, 0.35]        [0.25, 0.40]    [0.28, 0.46]      [0.32, 0.58]      [0.35, 0.64]
Eastern Gulf 0.48              0.30                0.34            0.38              0.45              0.51 Int-High-High
[0.43, 0.54]      [0.22, 0.38]        [0.26, 0.42]    [0.30, 0.48]      [0.34, 0.60]      [0.38, 0.68]
Western Gulf 0.59              0.49                0.53            0.57              0.63              0.69 Int-Int-High
[0.51,0.67]        [0.41, 0.57]        [0.44, 0.62]    [0.47, 0.67]      [0.51, 0.79]    [0.56, 0.87]
Southwest 0.24                0.15                0.20            0.24              0.31              0.38 Intermediate
[0.20,0.29]      [0.10, 0.20]          [0.14, 0.26]    [0.18, 0.32]      [0.22, 0.45]      [0.26, 0.54]
Northwest 0.16              0.10                0.15            0.18              0.25              0.31 Int-Low-Int
[0.08, 0.24]        [0.05, 0.15]        [0.09, 0.20]      [0.12, 0.26]      [0.15, 0.39]      [0.19, 0.47]
Hawaiian Islands 0.24                0.19                0.24            0.29              0.38              0.46 Int-Low
[0.20, 0.28]        [0.13, 0.24]        [0.18, 0.31]    [0.22, 0.39]      [0.27, 0.53]      [0.31, 0.64]
Caribbean 0.28                0.19                0.24            0.28              0.35              0.42 Intermediate
[0.24, 0.31]      [0.10, 0.29]        [0.14, 0.33]    [0.18, 0.39]      [0.22, 0.51]      [0.27, 0.59]
Global and Regional Sea Level Rise Scenarios for the United States l 19
 
Figure 2.4: Relative sea level rise, in meters, in 2050 for the a) Intermediate-Low and b) Intermediate-High scenarios relative to the year 2000.
2.4. Long-Term Sea Level Change (2050-2150)
The updated GMSL values in 2050, 2100, and 2150 relative to a 2000 baseline are shown for each of the five scenarios in Table 2.3. Note that the current National Tidal Datum Epoch (NTDE) has a baseline of 1992 (midpoint of the 1983-2001 epoch). Comparisons between the projections here and calculations tied to the NTDE will require an adjustment between 1992 and 2000 (see Table A1.2 for offsets). Beyond the middle of this century, the differences between sea level scenarios become increasingly large, and the differences between sea level scenarios become more closely associated with differences in potential future GHG emis-sions pathways and associated global warming. Although the GMSL scenarios (names and their values) are the same at 2100 for this report and for Sweet et al. (2017), there is a narrowing in the range covered by the scenarios in both 2050 and 2150, driven primarily by a reduction in the values at those two target dates as-sociated with the Intermediate-High and High scenarios in this report. As previously discussed, in 2050, the updated median value for the High scenario is similar to the median value for the Intermediate-High scenario from Sweet et al. (2017). This is not the case in 2150, however, where the separation between the scenarios remains similar to Sweet et al. (2017). Because of this, and because the scenarios are defined by the 2100 values, the same scenario naming is used in this report as in Sweet et al. (2017), with the notable exception of the omission of the Extreme (2.5 m) scenario.
In the very long term (over millennia), the magnitude of global mean sea level rise closely relates to the mag-nitude of global warming; however, over the timescales of decades and centuries, the magnitude of global warming more closely relates to the rate of GMSL rise. It is thus not possible to tie specific levels of warm-ing in general to amounts of sea level rise, but it is possible to relate specific levels of warming at specific points in time (e.g., at the end of the century) to different levels of sea level rise. Thus, based on the IPCC AR6 (&sect;9.6.3.4 in Fox-Kemper et al., 2021), it is possible to connect the GMSL rise scenarios to different levels Table 2.3: Global mean sea level and contiguous United States scenarios, in meters, relative to a 2000 baseline.
Global Mean Sea Level                                            Contiguous United States 2050            2100          2150                                2050        2100      2150 Low            0.15            0.3            0.4                Low              0.31        0.6        0.8 Intermediate-Low      0.20            0.5            0.8        Intermediate-Low        0.36          0.7        1.2 Intermediate        0.28            1.0            1.9            Intermediate        0.40          1.2        2.2 Intermediate-High      0.37            1.5            2.7        Intermediate-High      0.46          1.7      2.8 High            0.43            2.0            3.7                High            0.52          2.2        3.9 Global and Regional Sea Level Rise Scenarios for the United States l 20
 
of future global mean surface air temperature occurring at the end of the century. The median GMSL pro-jection for 2100 for a world with global mean surface air temperature in 2081-2100 averaging 2.0&deg;C above 1850-1900 levels is about 0.5 m (likely range of 0.4-0.7 m; Table 2.4), consistent with the Intermediate-Low scenario. The median GMSL projection for a world with global mean surface air temperature in 2081-2100 averaging 4.0&deg;C higher is about 0.7 m (likely range of 0.6-0.9 m), between the Intermediate-Low and Inter-mediate scenarios, with the upper end of the likely range approaching the Intermediate scenario. These two scenarios are also consistent with the current observed acceleration, which, if extrapolated, would yield about 0.24 m of GMSL rise by 2050 and 0.69 m by 2100.
However, these projections include only physical processes in which there is at least medium confidence in the current scientific understanding. As described in the IPCC AR6 (Box 9.4 in Fox-Kemper et al., 2021), the largest potential contributions to long-term GMSL rise come from ice-sheet processes in which there is cur-rently low confidence. Projections that include the magnitudes, rates, and thresholds associated with these ice-sheet processes, particularly under higher emissions futures, could give rise to GMSL rise values well above the likely range. Pathways to such unknown-likelihood, high-impact outcomespotential surprises in the words of NCA4 (Kopp et al., 2017)include
* earlier-than-projected ice-shelf disintegration in Antarctica,
* abrupt, widespread onset of marine ice-sheet instability and/or marine ice-cliff instability in Antarctica, and
* faster-than-projected changes in surface-mass balance on Greenland, potentially associated with changes in atmospheric circulation, cloud processes, or albedo changes.
These outcomes are represented in the IPCC projections (&sect;9.6.3 in Fox-Kemper et al., 2021) through the in-clusion of an illustrative very high emissions (SSP5-8.5), low-confidence projection range, the 83rd percentile of which for 2100 extends to 1.6 m (modestly above the Intermediate-High scenario) and the 95th percentile of which extends to 2.3 m (above the High scenario). In 2150, the 83rd and 95th percentiles of this low-con-fidence scenario are 4.8 and 5.4 m, respectively. Because these outcomes are based on processes poorly represented in climate and ice-sheet models, the IPCC assessment of these processes incorporates informa-tion from a structured expert-judgement study (Bamber et al., 2019) and a single Antarctic ice-sheet model-ing study that explicitly incorporates ice-shelf hydrofracturing and ice-cliff collapse mechanisms (DeConto et al., 2021). (See &sect;9.6.3.2, &sect;9.6.3.3, and Box 9.4 of Fox-Kemper et al., 2021, for further discussion.)
To connect this to the scenarios provided here, the Intermediate-High and High scenarios represent poten-tial futures in which these deeply uncertain ice-sheet processes play important roles in the late 21st century and beyond. After 2100, these processes may also play important roles in the Intermediate scenario. These trajectories are highly emissions-dependent. For example, in an illustrative low emissions (SSP1-2.6) future, in which the world achieves net-zero carbon dioxide emissions by the 2070s and net-negative emissions thereafter, the corresponding AR6 low-confidence ranges in 2100 extend to 0.8 m at the 83rd percentile (between the Intermediate-Low and Intermediate scenarios) and 1.1 m at the 95th percentile (modestly above the Intermediate scenario), reaching 1.3 m (between the Intermediate-Low and Intermediate scenarios) and 1.9 m (consistent with the Intermediate scenario), respectively, in 2150. Thus, in a low emissions future, there is little evidence to support the plausibility of GMSL projections substantially higher than the median Interme-diate scenario.
These warming levels are further compared to the five scenarios in this report by assessing the probability that the given GMSL value in 2100 will be exceeded for a particular warming level (Table 2.4). At all warming levels, there is at least a 92% chance of exceeding the Low scenario in 2100. The probability for exceeding the Intermediate-Low (0.5 m) scenario drops for all warming levels when compared to the probability for ex-ceeding the Low scenario. For the Intermediate, Intermediate-High, and High scenarios, the probability drops Global and Regional Sea Level Rise Scenarios for the United States l 21
 
off at each warming level. Consistent with the framing of the five scenarios in this report, greater warming and higher emissions are generally needed to arrive at the Intermediate through High scenarios in 2100.
Table 2.4: IPCC warming level-based global mean sea level projections. Global mean surface air temperature anomalies are projected for years 2081-2100 relative to the 1850-1900 climatology. Sea level anomalies are relative to a 2005 baseline (adapted from Fox-Kemper et al., 2021). The probabilities are imprecise probabilities, representing a consensus among all projection methods applied. For imprecise probabilities >50%, all methods agree that the probability of the outcome stated is at least that value; for imprecise probabilities <50%, all methods agree that the probability of the outcome stated is less than or equal to the value stated.
Unknown              Unknown Global Mean Surface Likelihood, High    Likelihood, High Air Temperature          1.5&deg;C            2.0&deg;C        3.0&deg;C          4.0&deg;C        5.0&deg;C Impact - Low    Impact - Very High 2081-2100 Emissions          Emissions Low      Intermediate Closest Emissions                                                                                  Low (SSP1-2.6),  Very High (SSP5-8.5),
Low        (SSP1-2.6) to  (SSP2-4.5) to      High        Very High Scenario-Based GMSL                                                                              Low Confidence      Low Confidence (SSP1-2.6)    Intermediate        High        (SSP3-7.0)    (SSP5-8.5)
Projection                                                                                            processes            processes (SSP2-4.5)    (SSP3-7.0) 0.18            0.20      0.21 (0.18-        0.22          0.25            0.20                0.24 Total (2050)
(0.16-0.24)      (0.17-0.26)      0.27)      (0.19-0.28)  (0.22-0.31)    (0.16-0.31)        (0.20-0.40) 0.44            0.51      0.61 (0.50-    0.70 (0.58-        0.81          0.45                0.88 Total (2100)
(0.34-0.59)    (0.40-0.69)        0.81)          0.92)    (0.69-1.05)    (0.32-0.79)          (0.63-1.60)
Low to    Intermediate-  Intermediate-  Intermediate- Intermediate-Bounding Median                                                                                        Low to      Intermediate-Low to Intermediate-        Low to        Low to        Low to        Low to Scenarios in 2100                                                                                Intermediate-Low      Intermediate Low        Intermediate  Intermediate  Intermediate  Intermediate Probability > Low 92%              98%          >99%            >99%          >99%            89%                >99%
(0.3 m) in 2100 Probability > Int.-Low 37%              50%          82%            97%          >99%            49%                  96%
(0.5 m) in 2100 Probability > Int.
                              <1%              2%            5%            10%          23%            7%                  49%
(1.0 m) in 2100 Probability > Int.-High
                              <1%              <1%          <1%              1%          2%              1%                  20%
(1.5 m) in 2100 Probability > High
                              <1%              <1%          <1%            <1%          <%            <1%                  8%
(2.0 m) in 2100 The median regional scenario values in 2100 and 2150 for the eight coastal regions discussed in Section 2.3 are provided in Table 2.5. The values in 2100 for each region differ from the GMSL value used to define a given scenario due to the combination of regionally relevant factors that are discussed in Section 2.1. Similar to the near term, the highest values across all scenarios are found in the Western Gulf region, followed by the Eastern Gulf. These high values are heavily driven by the high rates of subsidence in the region. For all but two regions (Southwest and Northwest), the projected values exceed the GMSL values associated with a particular scenario. The values for each scenario in the Southwest region correspond closely to the GMSL values, which is consistent with the agreement seen between the observation-based extrapolations in 2050 for the global and regional case discussed in Section 2.3. To further understand the regional variability for a given scenario, Figure 2.5 shows the regional departure from the GMSL value for each scenario in 2100.
In other words, the provided maps display the amount that needs to be added to the global value to get the associated regional value for a given scenario. The regional pattern is similar in each case. The Eastern Gulf and Western Gulf regions are consistently much higher than the global value, and the southern coast of Alaska is much lower across all scenarios. In the highest scenarios, the Northeast, Southeast, Northwest, and Southwest regions are near the global values, although there is a larger east-west separation in the lower scenarios. In these lower scenarios, the higher projections for the Northeast, when compared to the Southwest, are a result of both VLM and ocean circulation changes along the U.S. East Coast. In the higher Global and Regional Sea Level Rise Scenarios for the United States l 22
 
scenarios, the contributions from the ice sheets dominate and lead to less separation between the U.S. East and West Coasts.
Table 2.5: Scenarios of relative sea level, in meters, for eight coastal regions of the United States in 2100 and 2150 relative to a baseline of 2000. Median values are shown.
Intermediate-                      Intermediate-Region          Low                      Intermediate                        High Low                                High Northeast 2100            0.6          0.8              1.3              1.6          2.1 2150            0.9          1.3              2.3              2.7          3.7 Southeast 2100            0.5          0.7              1.1              1.6          2.1 2150            0.7          1.1              2.1              2.7          3.7 Eastern Gulf 2100            0.6          0.8              1.2              1.7          2.2 2150            0.8          1.2              2.2              2.8          3.9 Western Gulf 2100            0.9          1.1              1.6              2.1          2.6 2150            1.3          1.7              2.8              3.4          4.5 Southwest 2100            0.3          0.5              1.0              1.5          2.0 2150            0.4          0.8              1.9              2.6          3.7 Northwest 2100            0.2          0.4              0.8              1.3          1.8 2150            0.3          0.7              1.6              2.3          3.3 Pacific Islands 2100            0.4          0.6              1.1              1.7          2.3 2150            0.6          1.0              2.2              2.9          4.2 Caribbean 2100            0.4          0.6              1.0              1.5          2.1 2150            0.5          0.9              2.0              2.6          3.7 Global and Regional Sea Level Rise Scenarios for the United States l 23
 
Figure 2.5: Regional deviations of relative sea level from the global mean sea level (GMSL; in meters) value for each scenario in 2100. To obtain the regional projection in 2100 for each scenario, the mapped values must be added to the GMSL value for the associated scenario.
2.5. Scenario Divergence and Tracking In this report, for the first time, a specific focus is given to the near-term time period (2020-2050). During this window, observations can provide useful information on the trajectory of sea level rise on global and regional scales and serve as a comparison to the model-based GMSL scenarios. Prior to 2050, there is rela-tively small process uncertainty and little sensitivity to different emissions trajectories, and there is reduced spread between the scenarios in this report compared to Sweet et al. (2017). Connected to this reduced spread, the likely ranges of the revised GMSL scenarios presented here remain overlapping after 2050, whereas the Sweet et al. (2017) scenarios do not overlap after about 2040. In other words, in this report, the process uncertainty continues to exceed the GMSL scenario divergence past the near-term time period. Until the divergence exceeds the range for a given scenario, it will not be possible to determine when higher-end GMSL scenarios will unambiguously emerge from the potential range of the lower-end GMSL scenarios for decades to come. In this report, the time periods (or gates) when the scenarios become separable are estimated. Different considerations for determining these gates must be made before and after the near-term time period, when the observations are most useful. It should be noted that the gates presented here are based solely on the GMSL differences between scenarios. Regionally, the timing of these gates may be different due to uncertainty in the contributing regional processes. Additionally, other lines of evidence including monitoring of individual processes or emissions trajectories could allow for distinguishing between the scenarios earlier than the gates provided here.
In Figure 2.6, the time pathways of the five GMSL scenarios from 2020 to 2100 are shown, and the gates at which the likely ranges diverge from a particular trajectory or scenario are determined. In Figure 2.6a, the divergence relative to the observation-based GMSL extrapolation is assessed. Note: the GMSL observa-tion-based extrapolation is extended only to 2100 here for the purposes of this divergence assessment. For Global and Regional Sea Level Rise Scenarios for the United States l 24
 
the Low and High scenarios, the likely ranges separate prior to 2060, with the Intermediate-High scenario separating after 2060. On the other hand, the Intermediate-Low and Intermediate scenarios do not diverge from the extrapolated observation-based trajectory until after 2080. Consistent with the discussion in Sec-tion 2.3, if the processes driving sea level rise are assumed to remain similar for the next three decades, the Intermediate-Low and Intermediate scenarios provide useful bounds on GMSL rise for the near-term time period.
In the decades beyond 2050, however, the more uncertain processes described in Section 2.4 could be-come a factor and the observation-based trajectory becomes less informative. Instead of assessing the divergence relative to this trajectory, the separation gates relative to the Intermediate scenario are shown in Figure 2.6b. In this case, the Intermediate-High and High scenarios will not diverge from the Intermedi-ate scenario until after 2070 and 2060, respectively. Only the Low scenario diverges from the Intermediate scenario prior to 2050. Although not depicted in Figure 2.6, the higher scenarios also start to overlap again after 2100; for example, GMSL rise consistent with the Intermediate scenario in 2100 (1.0 m) does not rule out GMSL rise consistent with the Intermediate-High scenario by 2150. In tying the two different gate assess-ments together, even though the Intermediate scenario tracks near the current observation-based trajectory, it will not be possible to statistically distinguish between the Intermediate scenario and the two higher sce-narios for decades to come. This also provides important context and caution if attempting to use the obser-vations directly to infer future sea level rise beyond the near-term time period.
Figure 2.6: Divergence of global mean sea level (GMSL) trajectory and scenarios. The time series shows the observation-based GMSL trajectory and the five GMSL scenarios from 2000 to 2100. The dots denote where each scenario significantly (2 sigma) deviates from the a) observation-based trajectory and from the b) Intermediate scenario.
To explore this further, the proportions of the IPCC AR6 sea level projections contributing to each GMSL rise scenario are shown in Figure 2.7, with contributing emissions pathways specified. As an example interpreta-tion of this figure, the Low scenario generally requires a low emissions pathway, while the Intermediate-Low scenario arises from low, intermediate, and high emissions pathways. Pathways consistent with the Interme-diate scenario include low emissions trajectories but are mostly related to high emissions scenarios. In fact, the Intermediate, Intermediate-High, and High scenarios are all heavily driven by high emissions scenarios, and differences between these scenarios are associated predominantly with the possible role and contribu-tions of the low-confidence ice-sheet processes described in section 2.4. The other processes that cause Global and Regional Sea Level Rise Scenarios for the United States l 25
 
future sea level change have similar contributions across these scenarios. In other words, sterodynamic sea level change is similar for the Intermediate, Intermediate-High, and High scenarios.
These estimates provide a link between the emissions trajectories in the near term and the possible sce-nario for GMSL rise in the long term. When coupled with the gating assessment in Figure 2.6, these esti-mates hold particular relevance for assessing the pathway of sea level rise and determining which long-term scenarios are then possible or even likely. As a way of connecting the elements of the report, the time period where the GMSL scenarios begin to diverge can be put in the context of the analysis done in both the near-term and long-term sections. The likely ranges of the Low and Intermediate-Low versus Intermediate scenarios separate at about 2040 and 2065, respectively. The observation-based extrapolations of global GMSL rise have a relatively narrow range out to this time horizon and can therefore play a role in determin-ing whether a particular low-end trajectory or scenario is more or less likely to be exceeded in the coming decades. As shown in Figure 2.7, the Low scenario depends very heavily on a low emissions pathway on any time horizon. Monitoring using observations of both sea level and emissions can be useful for evaluating the likelihood of the Low scenario, both in the near term and long term.
On the other hand, the separations of the likely ranges for the Intermediate to Intermediate-High and Inter-mediate to High scenarios do not occur until after 2060 and 2070, respectively. The values at the end of the 21st century and beyond for these scenarios can arise under a variety of different emissions pathways, although higher scenarios are predominantly linked to higher emissions, as expected. To state it another way, the near-term trajectories discussed in Section 2.3 do not currently inform the likelihood of a given sce-nario occurring in 2100 or 2150. However, the observations can provide useful monitoring as the windows of separation (gates) for a different scenario approach in the future. On these global scales, process-based monitoring of the ice sheets, for example, can play an important role, as the higher scenarios (Intermediate to High) are closely linked to the potential for ice-sheet changes. Additionally, a link between the scenarios in 2100-2150, emissions pathways, and warming levels has been established here. Ongoing and continuous monitoring of both global temperatures9 and emissions10 will aid in determining the possible trajectory of future GMSL rise. It should be noted that while the windows provided in Figure 2.6 would be different on the national or regional level, the scenarios for a given location are still closely linked to emissions and warming, and the monitoring discussion above is still relevant.
Finally, regardless of future emissions pathways, GMSL rise will continue past 2150. The amount of commit-ted rise can be assessed based on historical comparisons, modeling, and the current process-based un-derstanding of GMSL rise. This committed rise is the amount of total sea level rise that will likely occur for a given warming level. For higher warming levels, the ranges of committed sea level are wide, but the possible values are large in magnitude. Even for a relatively low warming level of 1.5&deg;C, the committed sea level over the next 2000 years still ranges between about 2 m and 3 m. For 2&deg;C, the upper range increases to 6 m (IPCC, 2021a). Although the focus of this report is on the time period between 2020 and 2150, it does rein-force the when, not if framing provided in Section 1.
9 https://climate.nasa.gov/vital-signs/global-temperature/
10 https://gml.noaa.gov/ccgg/trends/
Global and Regional Sea Level Rise Scenarios for the United States l 26
 
Figure 2.7: Proportions of the contributions from different IPCC AR6 sea level trajectories to each of the five global mean sea level (GMSL) rise scenarios used in this report: Low, Intermediate-Low, Intermediate, Intermediate-High, and High. The IPCC AR6 trajectories are Low Emissions; Low Emissions, LC (where LC indicates inclusion of low-confidence ice-sheet processes);
Intermediate Emissions; Intermediate Emissions, LC; High Emissions; and High Emissions LC. The emissions pathways associated with the IPCC AR6 trajectories are as follows: Low Emissions = Shared Socioeconomic Pathway (SSP) 1-1.9 or SSP1-2.6; Interme-diate Emissions = SSP 2-4.5; High Emissions = SSP3-7.0 or SSP5-8.5. Shifts between different GMSL rise scenarios approximately reflect the relative odds of being close to a given scenario under different emissions pathways; e.g., the Low scenario is much more plausible under a low emissions pathway, while Intermediate and higher scenarios are much more likely to be associated with high emissions pathways, as well as with low-confidence ice-sheet processes.
Global and Regional Sea Level Rise Scenarios for the United States l 27
 
Section 3: Extreme Water Levels and Changing Coastal Flood Exposure Since Sweet et al. (2017), some objectives of the Task Force have been to define and develop for the U.S.
coastline 1) a set of coastal-climate flood-resilience standards and 2) a gridded set of extreme water level (EWL) probabilities that span flood frequencies with associated impacts to assess these standards. Together, these sets of information are used to describe how flood exposure within coastal floodplains are slated to change from rising sea levels (i.e., without mitigative action). Specifically for 1), we use a nationally calibrated set of the coastal water-level-impact-severity thresholds from the NOAA National Weather Service (Sweet et al., 2018), which are used in public communications. For 2), a regional frequency analysis (RFA) of tide-gauge observations is developed by adapting methods for exposure assessments within the Pacific Basin (Sweet et al., 2020b) and for the U.S. Department of Defense coastal installations worldwide11 (Hall et al., 2016).
Regional frequency analysis can provide many types of geospatial information based on limited sets of local observations, such as rainfall characteristics published by NOAA12 (Perica et al., 2018), which are widely used in stormwater design and management within the United States. Both the RFA-based extremes and NOAA flood-threshold information are discussed below.
There are a few important notes about terminology for this section (and the report as a whole). First, aver-age event frequency terminology is used throughout (except in Section 4.2 to build off of relevant papers/
concepts) to describe extreme water level probabilities instead of the more traditional return period termi-nology. This is done primarily to address best practices (or avoid bad practices), which have been reviewed by the United States Corps of Engineers (USACE; USACE, 1994). Although frequency and period are relat-ed (they are reciprocals), the use of periods can be misconstrued; e.g., the so-called 100-year event can be easily confused or communicated (e.g., IPCC, 2021b) as an event that occurs once per century. Such an interpretation could be assumed to imply a static and permanent water level that happens, on average, 100 years from the last event. In reality, such coastal water levels have and will continue to change with sea level rise, among other potential factors, and can occur (albeit with low probability) several times over the span of a few years. Second, although annual exceedance probability terminology is often used to describe average event frequencies (e.g., 0.1 events/year frequency expressed as the 10% annual chance event), we again stick to events/year frequency terminology, partly due to underlying method but also because events occur-ring more often than once a year are also being quantified and communicated (a 5 events/year frequency is poorly conveyed as a 500% annual chance event). Finally, the use of the word occurrence in this section means has the probability of equaling or exceeding, as it applies to a particular water level or flood height.
3.1. Overview of Extreme Water Levels and Coastal Flooding As sea levels continue to rise, coastal water levelsfrom the mean to the extremeare growing deeper and reaching farther inland along most U.S. coastlines. Where local relative sea level (RSL) is rising, the wet-dry land delineation (i.e., mean higher high water [MHHW] tidal datum; NOAA, 2003) is encroaching landward, causing more permanent inundation and land loss (e.g., in Louisiana); affecting groundwater levels (Befus et al., 2020), stormwater systems effectiveness (Habel et al., 2020), and water quality (McKenzie et al., 2021);
and altering the intertidal zone and its ecosystems (Kirwan and Gedan, 2019). Where local RSL is falling relative to the land surface, other problems can occur, such as changes in coastal erosion processes, inci-sion of tributaries, decreased draft for waterborne transport, decreased sedimentation in saltwater marshes, and alterations in intertidal zones and estuaries (Larsen et al., 2004; Sweeny and Becker, 2020). Especially problematic for societys coastal footprint is that the entire spectrum of flood exposure is also growing where RSL is rising, from minor high tide flooding (HTF) to more severe major flooding during storms (Sweet and Park, 2014; Fox-Kemper et al., 2021). For example, the national rate of minor HTF is accelerating and is now (circa 2020) more than double what it was in 2000 due to RSL rise (Figure 3.1), with projections suggesting 11 https://drsl.serdp-estcp.org/
12 https://www.weather.gov/owp/hdsc Global and Regional Sea Level Rise Scenarios for the United States l 28
 
Figure 3.1: National median rate of minor high tide flooding and relative sea level, in meters, from 98 NOAA tide gauges along U.S. coastlines outside of Alaska used to monitor and track flood-frequency changes (from Sweet et al., 2021). Relative sea levels reference the lowest annual (1925) level.
a doubling of its current rate by 2030 (Sweet et al., 2018, 2021; The State of High Tide Flooding and Annual Outlook13; Thompson et al., 2021; Flooding Days Projection Tool14).
Assessments of current and future changes in minor to major HTF using RSL projections require probabilistic information about local water level variability. Specifically, they require the envelope of variability encapsulat-ing EWLs that define the magnitude and frequency of events capable of causing a range of known or as-sumed impacts (Tebaldi et al., 2012; Church et al., 2013; Hall et al., 2016; USGCRP, 2017; Oppenheimer et al.,
2019; Fox-Kemper et al., 2021). The basis for quantifying EWLs along U.S. coastlines originates with NOAAs tide-gauge network, which measures water level responses from multiple processes operating over a range of frequencies (Table 3.1). However, due to their general placement (e.g., in harbors), protective housings that dampen wave effects, and their multi-minute sampling rates, tide gauges typically do not measure or report values that include higher-frequency wave effects (Sweet et al., 2015; see Box 3.1). Other sources of useful tide level information for the U.S. and globally include USACE inventories (e.g., USACE MRG&P, 2017), the University of Hawaii Sea Level Center,15 and the Global Extreme Sea Level Analysis database.16 Extreme water levels are often used as a proxy for impacts, such as the 0.01 events/year frequency level, better known as the once per century event (Oppenheimer et al., 2019), with connotations of the flood of the century. However, such a probabilistically defined event can be both misleading about its true frequency (USACE, 1994) or might go mostly unnoticed in some locations (Sweet et al., 2020b). High tide flood heights, on the other hand, are absolute heights that are calibrated to the depth-severity impact thresholds of the NOAA National Weather Service and local emergency managers to trigger public notification of impending flood risks (NOAA, 2020). NOAA minor, moderate, and major HTF is defined as a water level reaching or exceeding about (national median values) 0.55 m, 0.85 m, and 1.20 m above current MHHW, respectively (Sweet et al., 2018). Put another way, an EWL is only a flood if it actually impacts the public in some manner and is not necessarily a description of a meteorological event.
13 https://tidesandcurrents.noaa.gov/HighTideFlooding_AnnualOutlook.html 14 https://sealevel.nasa.gov/data_tools/15 15 https://uhslc.soest.hawaii.edu/
16 https://www.gesla.org/
Global and Regional Sea Level Rise Scenarios for the United States l 29
 
But the NOAA tide-gauge network is relatively sparse compared to the density of coastal communities, and the tide gauges have varying record lengths. From the perspective of a particular coastal community, this may result in either 1) a lack of local data (often data that are simply extrapolated from the closest NOAA tide gauge) or 2) a data record that is biased by lack of or overexposure to regionally significant rare events such as storm surges from landfalling tropical cyclones. Probabilistic assessments using atmospheric/ocean circu-lation models can increase spatial coverage (Vousdoukas et al., 2018), but they often perform poorly in areas with high tropical storm activity or with complex bathymetries (Muis et al., 2016). Targeted deployments of in situ sensors by communities to monitor changes in sea level, tide heights, and flood exposure (McCallum et al., 2013) can be informative but still lack the necessary longer-term regional perspective.
Table 3.1: Physical processes affecting U.S. coastal water levels and their temporal and spatial scale properties (modification of Sweet et al., 2017). Extreme water levels, which, as measured by tide gauges, generally exclude high-frequency wave effects, include processes between tsunami and ocean-basin variability and, to a lesser extent, low-frequency changes or trends associated with land ice melt/discharge, thermal expansion, and vertical land motion.
Spatial Scale                                    Potential Magnitude Physical Process                                                  Temporal Scale Global    Regional    Local                                    (yearly)
Wind Waves Effects                                  X        seconds to minutes            <10 m Tsunami                              X        X          minutes to hours          <10s of ms Storm Surge (e.g., tropical and X        X          minutes to days            <10 m extra-tropical storms)
Tides                                X        X            hours to years              <15 m Ocean/Atmospheric Variability X        X            days to years              <0.5 m (e.g., ENSO response)
Ocean Gyre and Over-turning X        X          years to decades            <0.5 m Variability Land Ice Melt/Discharge            X            X        X          years to centuries        mms to cms Thermal Expansion              X            X        X          years to centuries        mms to cms Vertical Land Motion                        X        X        minutes to centuries        mms to ms For the U.S., there are two primary sources of federally provided EWL probabilities. The first comes from the Federal Emergency Management Agency (FEMA, 2016b), which provides sets of regional solutions using a combination of NOAA storm-tide observations, historical high-water marks,17 synthetic storm simulations (e.g.,
Nadal-Caraballo et al., 2020; ERDC Coastal Hazards System18), and wave effects to estimate the regulatory floodplain and its exposure to the rarest of events (e.g., 1% and 0.2% annual chance events). FEMA provides this information for national flood insurance purposes19 but does not consider future sea levels. Another set of EWL probabilities is from NOAAs Center for Operational Oceanographic Products and Services (Zervas, 2013), which currently uses a generalized extreme value (GEV) distribution fit to annual highest water levels for tide-gauge records of >30 years).20 The U.S. Army Corps of Engineers and their Sea Level Change Cal-culator21 provide the NOAA EWL probabilities (Zervas, 2013) with several projections of future RSL to help in project planning but only for specific long-term tide-gauge locations.
A primary goal of the following subsections is to introduce a new set of EWL probabilities to support sea lev-el rise and flood-exposure assessments and planning. The EWL set is applicable for most of the U.S. coast-line and further resolves (both in physical and probability space) the EWL information currently available from 17 https://stn.wim.usgs.gov/FEV/
18 https://chs.erdc.dren.mil/
19 https://www.fema.gov/flood-maps/national-flood-hazard-layer 20 https://tidesandcurrents.noaa.gov/est/
21 https://cwbi-app.sec.usace.army.mil/rccslc/slcc_calc.html Global and Regional Sea Level Rise Scenarios for the United States l 30
 
FEMA and NOAA; although again, the EWL data here, which are derived from tide-gauge data, generally do not include wave effects (see Table 3.1 and Figure 1.1). Section 3.2 briefly describes the RFA of NOAA tide-gauge data with pointers to the Appendix for a fuller description. In Section 3.3, data for all NOAA tide gaug-es with >10 years of record are used to compute EWL probabilities, and these results are compared to NOAA and FEMA datasets. Section 3.4 discusses methods on how local EWL probabilities can be 1) computed us-ing other records, such as those of shorter duration (<10 years) from NOAA or other (user supplied) sources, and 2) estimated approximately every 500 m along the U.S. coastline based on local tide range information from NOAA models (e.g., VDatum22). Lastly, Section 3.5 assesses current and future flood exposure within the coastal floodplain using NOAAs height-severity categories of minor, moderate, and major HTF (Sweet et al., 2018), which broadly define water levels where U.S. infrastructure becomes impacted and are used in weather forecasting to trigger emergency responses (NOAA, 2020). Estimates of how flood exposure is projected to change by 2050 (assuming no additional adaptation or risk-deduction measures) are provided using the upper-bounding scenarios of the regional observation-based extrapolations along U.S. coastlines (see Table 2.2).
3.2. Regional Frequency Analysis of Tide-Gauge Data Extreme water level probabilities and their 95% confidence intervals are provided at a 1-degree spacing along nearly the entire U.S. coastline (Figure 3.2). The EWL information is based on an RFA (Hosking and Wallis, 1997) of NOAA tide gauges within a 400-km radius of the center of each individual 1-degree grid and fit with a Generalized Pareto Distribution (GPD) of threshold exceedances (Coles, 2001). The RFA process not only better assesses EWL exceedance probabilities from a regional perspective as compared to a sin-gle-gauge assessment but also can supply information where no tide gauges exist. Furthermore, a GPD fit to exceedances above a high threshold as compared to a GEV fit to annual maxima uses more of the data re-cord (e.g., two or more significant events within a particular year), not just those maxima within a certain (e.g.,
annual) time block. This approach, using RFA-based GPD fits, better resolves both the low- and high-fre-quency spectrum with output in this report ranging from 0.01 events/year to 10 events/year frequencies.
Combining an RFA with GPD fits to obtain EWL probabilities is unique for U.S. coastlines, although there are other statistical methods such as the joint probability method (Baranes et al., 2020) and Bayesian hierarchi-cal modeling (Calafat and Marcos, 2020), which may also prove useful in assessing rare event probabilities or providing information where no tide gauges exist.
22 https://vdatum.noaa.gov/
Global and Regional Sea Level Rise Scenarios for the United States l 31
 
Figure 3.2: Regional Frequency Analysis 1-degree grids and local index values (u) relative to local mean higher high water tidal datum at the NOAA tide gauges used in this study.
To be useful for local decision-making, the gridded EWLs (EWLgridded) derived by RFA need to be further localized (EWLlocal), which is achieved via a local index (u) estimated at a particular tide gauge (u values are shown in Figure 3.2) or for a particular location and converted to the vertical control datum on the land surface, normally the North American Vertical Datum of 1988 (NAVD88). The following equation is used to estimate EWLlocal probabilities (median and 95% confidence intervals):
1) where EWLgridded is the gridded EWL composed on normalized (unitless) sets of tide-gauge data, and ulocal, referred to simply as u, are the same value and represent the height of the 98th percentile of daily highest water levels with a 4-day filter applied and are relative to the 1983-2001 (or 5-year modified epoch; Gill et al.,
2014) MHHW tidal datum. For statistical independence when quantifying the EWL probabilities, the filtering process is needed to isolate and only include the peak water level value from a particular storm or event, rather than including multiple consecutive daily peak levels resulting from the same event (e.g., a multiday storm surge). See Section A2 for more details.
3.3. Average Event Frequencies of Extreme Water Levels The focus of this analysis is on EWL events and their probabilities that span the frequency space associated with coastal flooding under current sea levels (Sweet et al., 2018). An example for the NOAA tide gauge at The Battery in New York City (NYC) in Figure 3.3a shows the NOAA HTF heights and probability distributions for hourly water levels and also for their daily maxima.23 Also shown is the local index (u = 0.55 m above MHHW) computed for this tide gauge, which is used to estimate EWLlocal from the EWLgridded probabilities for this location (Figure 3.3b). See Figure A2.2f for the gridded probabilities applicable for NYC. At higher frequencies, such as those associated with the height of the minor HTF level (0.56 m above MHHW), the EWLlocal probabilities for events (about 4-5 events/year) are close but slightly underestimate flood fre-quency estimates for days (about 11 days/year; not shown), which are based on a multidecadal distribution 23 https://tidesandcurrents.noaa.gov/stationhome.html?id=8518750 Global and Regional Sea Level Rise Scenarios for the United States l 32
 
of daily highest water levels (shown in Figure 3.3a) used by NOAA when making projections of minor HTF (Sweet et al., 2018). This difference reflects the 4-day event filter in estimates of the EWLlocal probabilities discussed above. A similar ratio (about 2 days per event) exists in NOAAs HTF Outlook (about 11 days/year for 2020 at NYC, which is based on an extrapolation of quadratic or linear fits to annual counts of minor HTF days (Sweet et al., 2020a). The ratio of minor HTF events to days estimated at NOAA tide gauges as a whole is further discussed later in this section. The main point is that, typically, the duration of a minor HTF event, as in NYC and along U.S. coastlines, spans about 2 days and multiple tide cycles on average.
Figure 3.3: a) Empirical probability densities of hourly water levels and their daily maxima measured by the NOAA tide gauge at The Battery (New York City), as well as the tidal datums of mean lower low water (MLLW), great diurnal tide range (GT), local high tide flood (HTF) heights, and the local index (u) used to localize the RFA-gridded EWL for this location (see Figure A2.2f).
All values are referenced to the mean higher high water (MHHW) tidal datum and shown in b) as a return interval curve with the 95% confidence interval (2.5% and 97.5% levels) normalized to year 2020 RSLs.
Some general patterns emerge in regional EWLslocal with 1 event/year (Figure 3.4a) and 0.01 events/year frequencies (Figure 3.4b). Locations with higher 0.01 events/year EWLlocal are found adjacent to wide, shal-low continental coasts that are exposed to frequent tropical or extratropical storm surges, such as occur along the Eastern and Western Gulf coastal regions at 2.5 +/- 1.1 m and 2.8 +/- 0.8 m (median +/- 1 standard deviation), respectively. In contrast, the U.S. Pacific/Hawaiian Islands and Southwest Pacific coastal regions have lower 0.01 events/year EWLslocal due to deep, narrow continental shelves and generally calmer condi-tions (0.8 +/- 0.1 m and 1.0 +/- 0.1 m, respectively), although wave effects not inherent to the EWL probabilities are often the primary factor causing flooding, overwash, and erosion along natural landscapes in these locations (Barnard et al., 2019; see Box 3.1). In terms of the 1 event/year heights, tide ranges become influ-ential (correlation of ~0.7 between great diurnal tide range [GT] and u across all locations), as is the case in the Northwest Pacific coastal region and the southern Alaska coasts, where the highest 1-year EWLs occur (0.8 +/- 0.1 m and 1.0 +/- 0.3 m, respectively) and larger tide ranges are found.
Global and Regional Sea Level Rise Scenarios for the United States l 33
 
Figure 3.4: Current (circa 2020 relative sea levels) EWLlocal that a) occur annually on average and b) have a 0.01-year average event frequency. Note: the scales in the two figures are not the same, and to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
There are differences when comparing the RFA-based EWLslocal from this study to current FEMA and NOAA governmental datasets. Comparisons to NOAA EWLs (Zervas, 2013) in Figure 3.5a-c show that the RFA-based 0.01, 0.1, and 0.5 events/year levels are about 6%, 9%, and 13% higher across the board based on linear regression, respectively. The bias between datasets is not unexpected, as an RFA typically results in higher EWL probabilities with narrowed confidence intervals due to the regionalization process as compared to a single-gauge analysis (Sweet et al., 2020b). Overall, there is strong correlation between datasets, al-though less so at the 0.01 events/year EWLlocal (R2 = 0.49) due in part to the large differences occurring along the Gulf coastlines of Alabama, Mississippi, and Louisiana, where the RFA-based 0.01 events/year EWLlocal
(~4 m above MHHW) values are substantially higher (>1 m) than the NOAA GEV estimates in a few locations.
The RFA-based EWLlocal probabilities are also compared to the tide-gauge-equivalent stillwater compo-nent (tides, storm surge, and limited wave set-up, but not wave swash; see Figure 1.1) generated by FEMA and used within their regional Flood Insurance Studies24 (Figure 3.5d-f). The FEMA EWLs vary in their con-struction by region, using a combination of singular and RFA tide-gauge analyses, storm-surge modeling, and synthetic tropical storm modeling (for the Northeast, Southeast, and Eastern and Western Gulf coastal regions) via a joint probability method-optimal sampling (JPM-OS) procedure (FEMA, 2016a, 2016b). The 0.01 and 0.1 events/year EWLlocal are slightly lower (7% and 4%, respectively), with differences again noted along the Eastern and Western Gulf and Caribbean coastal regions. At the 0.5 events/year levels, both sets of EWLs are nearly the same based on linear regression. The goodness-of-fit (R2) values are about the same as with the NOAA (2013) GEV results, although a little less at the 0.01 events/year levelslikely due to the inclusion of synthetic storm-surge modeling in the FEMA estimates, compared to the NOAA (2013) values, which are based on tide-gauge observations. Thus, it is concluded that the RFA-based EWL provides higher estimates than a single-gauge analysis (Zervas, 2013) but less than those of FEMA stillwater values at lower probabilities, since FEMAs data also include storm-surge modeling, synthetic storms, and high-water marks in addition to tide-gauge data.
24 https://www.fema.gov/glossary/flood-insurance-study-fis Global and Regional Sea Level Rise Scenarios for the United States l 34
 
Figure 3.5: Comparison between (a-c) this studys EWLlocal to those of NOAA (Zervas, 2013) based on a GEV fit of annual highest water levels and to (d-f) the stillwater (storm surge, tides, and wave set-up) components of FEMA used in their Flood Insurance Study at the 0.01-year, 0.1-year, and 0.5-year average event frequency levels.
3.4. Methods to Localize the Gridded Extreme Water Level Event Probabilities There are several ways to obtain EWLlocal from the EWLgridded. All require a local index (u), which can be ob-tained from 1) a NOAA tide gauge used in this study (Figure 3.2; Table A1.3); 2) alternative sources of water level/tide-gauge data not used in this study (e.g., see Figure A2.3); or 3) tide range knowledge from mea-surements or models. When using short-term water level measurements (Figure A2.4), additional uncertainty, dependent on record length, is factored into the 95% confidence interval of the EWLlocal estimate (see Equa-tion 4 in the Appendix). This additional uncertainty relates to the fact that the local index (u) will vary from year to year akin to how RSL varies through time.25 On a national scale (and for most regions as well; see Figure A2.4), the root mean square error (RMSE) in local index estimates is about 6-7 cm after 5 years and falls to less than 3 cm at 10 years, which is close to the standard error in tidal datum calculations themselves (see datum errors in Bodnar, 1981).
Where local water level measurements are not available, another option is to estimate a local index (u) and EWLlocal probabilities based on an underlying relationship between local index values and tide range along U.S. coastlines. Additional uncertainty using this method will need to be factored into the results as well.
25 https://tidesandcurrents.noaa.gov/sltrends/sltrends.html Global and Regional Sea Level Rise Scenarios for the United States l 35
 
This relationship (Figure A2.5) builds off of the findings of Sweet et al. (2020b) within the Pacific Ocean and of Merrifield et al. (2013) globally, who found a strong global correlation between the range of water level variability and average annual highest water level across the globe. Nationally, there exists a strong positive relationship (R2 = 0.72 in Figure A2.5), although with fairly large uncertainty (RMSE of 0.11 m). But when tide range and local index values are regressed regionally, all the fits RMSEs are less (see Figure A2.5). Across all U.S. regions, it takes about 6 years of data for the RMSE (see Figure A2.4) in local index (u) estimates to match the RMSE values based on measured tide range (see Figure A2.5). Tide range information can be ob-tained from NOAA Vertical Datum Transformation (VDatum).26 Comparison of RMSEs based on multiple years of record versus tide range estimates of a local index (u) will vary by region (see Figures A2.4 and A2.5), and the lesser of the two is considered the better option in estimating an EWLlocal for any specific location not associated with a tide-gauge location used in the study.
Here we provide an example of how to obtain EWLlocal probabilities for a location not used in this study. The location for this example is the NOAA National Estuarine Research Reserve in Grand Bay, Mississippi (Figure 3.6a), which has a NOAA tide gauge, but the hourly record is only about 4 years long.27
: 1. The first step is to identify the specific EWL grid where the location resides, which in this case is grid number 42811 (Figure 3.6a), and obtain the EWLgridded probabilities.
: 2. Next, a local index needs to be estimated for an EWLlocal to be computed, either by the tide-range-based method (Figure 3.6b) or using the existing short data record (Figure 3.6c) for the specific region, depending on the smaller RMSE of the two methods. The RMSE based on the tide range regression is 0.078 m (Figure 3.6b) and is less than the 0.099 m RMSE based on a 4-year water level record for this region (solving the equation shown in Figure 3.6c).
: 3. Using the published NOAA tide range value at this location (0.49 m) leads to an estimated local index value of 0.47 m through the regional regression (solving the equation shown in Figure 3.6b).
: 4. An EWLlocal return level curve (Figure 3.6d) relative to the 1983-2001 tidal epoch is generated by substituting a local index value of 0.47 m and an RMSE of 0.078 m (with a variance of 0.0782) into Appendix Equations 1 and 4 (see Section A2), respectively.
: 5. Finally, to update the curve to current conditions (circa 2020) from the midpoint of the 1983-2001 epoch (1992), 0.12 m is added to the return level curve values. The 0.12 m value represents the re-gional-median trend in u of 4.3 mm/year multiplied by 28 years (see Table A1.3 and Section A2.3.4 for more information). Alternatively, 0.15 m could be added instead by applying the RSL offsets from the regional observation-based extrapolations for this region (Table A1.2).
The resultant EWLlocal probabilities estimated for Grand Bay are similar to others at nearby tide gauges that share the same 1-degree EWLgridded (see Figure 3.4). Less noticeable is that the 95th confidence intervals are more inflated (i.e., 0.5 m vs. 0.1 m at the 1 event/year EWL) because of the additional uncertainty from using the tide-range-based method to obtain a local index. Nationally, the spread of the 95% confidence interval at the 1 event/year EWLlocal using a local index (u) estimated by tide range (Figure 3.6b and Figure A2.5) is 0.32 m as compared to 0.03 m when assessed across all NOAA tide gauges.
26 https://vdatum.noaa.gov/
27 https://tidesandcurrents.noaa.gov/stationhome.html?id=8740166 Global and Regional Sea Level Rise Scenarios for the United States l 36
 
Figure 3.6: a) Map showing active NOAA tide gauges indicating Grand Bay, Mississippi, which has about 4-5 years of hourly data, b) tide range to local index (u) regression relative to the 1983-2001 tidal datum epoch with fit equation, goodness of fit (R2), and associated root mean square error (RMSE) for the surrounding region, c) RMSE for estimates of u based on 1-19 years of consecutive data over the 2001-2019 period based on the regional tide gauges for the surrounding region; and d) a 2020 EWLlocal return level curve for Grand Bay using a local index (u) from tide range regression. Note: to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
3.5. The Changing Nature of Coastal Flood Exposure To assess U.S. coastal flood exposure using the EWLlocal probabilities, we use the nationally calibrated coastal HTF heights of NOAA (Sweet et al., 2018) and a modification of Sweet et al. (2020b) for Alaska coastlines (see Section A2.4). The NOAA HTF heights include three categories: minor, moderate, and major (national median) starting at about 0.55 m, 0.85 m, and 1.20 m, respectively (Figure A2.6), whose impacts are disrup-tive, typically damaging, and often destructive, respectively, under current flood defenses. NOAA provides data (e.g., Flood Frequency [MapServer]28) and maps (Figure 3.7) in its SLR Viewer of exposure to HTF to help communities recognize potential flood exposure associated with weather-water level forecasts and for vulnerability assessments associated with sea level rise.
Currently (with EWLlocal relative to year 2020 trend levels), minor HTF events occur (median value) about 3 times per year along U.S. coastlines and are most frequent along the Northeast, Western Gulf, and Northwest coastlines (about 4 events/year) and along the Southeast and Eastern Gulf coastlines (about 2 events/year; Figure 3.8a). A similar pattern emerges when comparing the 2020 NOAA minor HTF outlook (Sweet et al.,
2020a) for the number of flood days at about 100 of the tide gauges (Figure 3.8b). The NOAA outlook for 28 https://coast.noaa.gov/arcgis/rest/services/dc_slr/Flood_Frequency/MapServer Global and Regional Sea Level Rise Scenarios for the United States l 37
 
Figure 3.7: NOAA minor (red layer: land between mean higher high water [MHHW] and minor high tide flood [HTF] height above MHHW), moderate (orange layer), and major (yellow layer) HTF maps showing a regional layered map with individual layer panes to the right for a) Charleston, South Carolina, and b) West Palm Beach, Florida. MHHW for 1983-2001 is the shoreline edge. Note:
to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
minor HTF days uses extrapolations of linear and/or quadratic fits to days per year with a water level at or above the flood height. As a whole, there are about twice the number of days of minor HTF than the number of discrete events (Figure 3.8b inset), which is largely reflective of typical synoptic-scale (temporal) variability and the 4-day event filtering used in the RFA process and GPD fitting. The national (median) outlook for mi-nor HTF in 2020 was 4-5 days, with about 8-9 days each along the Northeast and Western Gulf coastlines and 3-5 days each along the Southeast and Eastern Gulf coastlines (Sweet et al., 2020a).
Currently, moderate HTF in 2020 (Figure 3.8c) has about a 0.3 events/year frequency (median value) nation-ally and a similar 0.2-0.4 events/year frequency along the Southeast, Eastern Gulf, and Northwest coast-lines. Moderate HTF is most likely along the Western Gulf coastlines (0.6-0.7 events/year). Major HTF (Figure 3.8d) nationally and along the Southeast coastline has about a 0.04 events/year frequency. Major HTF is most likely along the Western Gulf coastline (0.15 events/year) and along the Northeast and Eastern Gulf coastlines (0.08-0.09 events/year). For a more local perspective (see Figure 3.7), 2020 annual frequencies of minor, moderate, and major HTF in Charleston, South Carolina, and West Palm Beach, Florida, were about 2-3 events/year, 0.15-0.25 events/year, and about 0.02-0.04 events/year, respectively, based on the near-est tide gauge (see Table A1.2).
Changes in flood exposure are projected to 2050 considering no additional flood risk reduction or adap-tation (e.g., via improved stormwater system functionalities) at NOAA tide gauges (Figure 3.9). The EWLlocal probabilities are brought to 2050 levels by adding the local RSL projections initiating in year 2005 associat-ed with the upper-bounding sea level scenario identified by the regional observation-based extrapolations (Table 2.2). Other scenarios could be used, but we opted for this particular set because it uses observational evidenceextrapolation of fits over the last 50-years (i.e., 1970-2020) to provide some level of prediction for the next 30 years. For instances where the extrapolations are the same as a particular scenario (e.g., North-east), the adjacent (higher) scenario is used (e.g., the Intermediate is considered the upper-bounding scenar-io for the Northeast), which also serves to partially compensate for natural variability that is not reflected in the extrapolations.
Global and Regional Sea Level Rise Scenarios for the United States l 38
 
Figure 3.8: Average event frequencies in 2020 of a) minor high tide flooding (HTF); b) number of days (as compared to events) of HTF estimated in NOAAs annual outlook (Sweet et al., 2021) and regression between events and days; c) average event frequencies in 2020 of moderate HTF; and d) average event frequencies in 2020 of major HTF. Flood height-severity definitions are from NOAA (Sweet et al., 2018) and, specifically for Alaska locations, from Sweet et al. (2020b).
Nationally and along all regions except the Hawaiian/Pacific Islands (about 9 events/year), the Caribbean (about 6 events/year), and Alaska (0.7 events/year) coastlines, the median event frequency in minor HTF is projected to increase to >10 events/year (Figure 3.9a). Moderate HTF (median) frequencies (Figure 3.9b) are projected by 2050 to increase nationally to about 4 events/year; >10 events/year along the Western Gulf coastline; 3-6 events/year along the Northeast, Southeast, and Eastern Gulf coastlines; about 1 event/year along the Northwest coastline; and 0.7 events/year along the Southwest coastline. Major HTF frequencies (Figure 3.9c) are projected to increase to about 0.2 events/year nationwide (median), with 1 event/year along the Western Gulf coastline, 0.5 events/year along the Northeast coastline, and 0.2-0.3 events/year along the Southeast Atlantic and Eastern Gulf coastlines. For a local perspective, the 2050 projections of annual frequencies of minor HTF in Charleston and West Palm Beach are >10 events/year, with 4-5 of those events reaching or exceeding moderate HTF and the possibility (0.1-0.2 events/year) of major HTF.
For perspective and a summary assessment by region, Table 3.2 quantifies how minor, moderate, and major HTF frequencies have changed and are projected to change considering the local RSL scenari-os associated with the upper-bounding scenario of the regional observation-based extrapolations (Table 2.2) using 1990, 2020, and 2050 time slices. Nationally, minor HTF frequencies nearly tripled between 1990 and 2020, growing from about 1 to 3 events/year. They are projected to more than triple by 2050 to Global and Regional Sea Level Rise Scenarios for the United States l 39
 
Figure 3.9: Coastal high tide flooding (HTF) frequencies projected at 2050 applying the sea level scenario that upper-bounds the regional observation-based extrapolations for NOAA a) minor, b) moderate, and c) major HTFs.
>10 events/year. Moderate HTF frequencies nationally experienced about a 50% increase (0.2 events/year growing to 0.3 events/year) from 1990 to 2020, which is slightly higher than the frequency increase in major HTF frequencies. By 2050, moderate HTF frequencies nationally are projected to increase by more than a factor of 10, with about a factor of 5 increase in major HTF frequencies. In short, assuming continuation of current trends and summarized at the national level, a flood regime shift is projected by 2050, with moder-ate HTF occurring a bit more frequently than minor HTF events occur today and major HTF events occurring about as frequently as moderate HTF frequencies occur today.
Global and Regional Sea Level Rise Scenarios for the United States l 40
 
Table 3.2: Annual average event frequencies for NOAA-defined minor, moderate, and major HTF heights by region that were typical (median values) in 1990, under current (circa 2020) sea levels and projected to occur considering the upper-bounding scenario of the observations-based extrapolations in 2050 (see Table 2.2).
1990                                      2020                                  2050 U.S. Region              Minor    Moderate        Major          Minor      Moderate      Major        Minor      Moderate        Major Flood        Flood        Flood          Flood        Flood        Flood        Flood        Flood        Flood National                1          0.2          0.03            3            0.3        0.04        >10            4            0.2
          *Hawaii/Pac Is            0.06        <0.02        <0.02          0.2          <0.02        <0.02          9          0.1        <0.02 NE Atlantic                2          0.3        0.06            4            0.6        0.09        >10            6            0.4 SE Atlantic              0.9          0.1          0.03            2            0.2        0.04        >10            4            0.2 E Gulf                0.7          0.2        0.06            2            0.3        0.08        >10            3            0.3 W Gulf                  1          0.3          0.1            4            0.7          0.2        >10          >10            1 SW Pacific              0.8        0.02        <0.02            1          0.04        <0.02        >10          0.7        <0.02 NW Pacific                3          0.3        <0.02            4            0.4        <0.02        >10            1          0.03
              **Alaska                0.7        <0.02        <0.02          0.2          <0.02        <0.02        0.7        0.03        <0.02 US Carib              0.02        <0.02        <0.02          0.04          <0.02        <0.02          6          0.04        <0.02
    *The Pacific Island locations use the same scenario assigned to the Hawaiian Islands (see Table 2.2); **Alaska locations, which as a whole could not be regionalized due to large differences in VLM, use the lower-bounding scenario per CONUS, which is the Intermediate-Low scenario (see Table 2.1). The lower-bounding scenario for Alaska is used to reflect the significant deviations below the Intermediate scenario (Figure A1.2b).
Box 3.1: Wave Contributions to Extreme Water Levels Water level heights are a common proxy for coastal flooding                      Leveraging the global total water level assessment of Vitousek (e.g., Sweet et al., 2018) and consist of a variety of compo-                    et al. (2017), which combines reanalysis models for waves, nents (see Figure 1.1). This report focuses primarily on projec-                  surge, and tides (total water level implying that all relevant tions of relative sea level (RSL) rise together with tides and                    components in Table 3.1 are included), we demonstrate the storm surge contributions to extreme water levels (EWLs).                        relative influence of waves on coastal water levels during However, along exposed coasts, wave-driven water levels can                      extreme events (Figure Box 3.1). Even though the coarse play a significant role in EWLs during storm events and during                    resolution of this study (1&deg; x 1&deg; grid cells) cannot fully resolve lesser storm conditions as exacerbated by sea level rise. Here                    tropical cyclones, which play a significant role in EWL events we illustrate the relative influence of wave-driven water levels,                for the Southeast, Eastern and Western Gulf, Caribbean, and broken down into the components of set-up and swash during                        Hawaiian/Pacific Islands regions, this analysis demonstrates extreme events across the United States, compared to tide                        the relevance of waves in contributing to EWLs. Across the and surge contributions.                                                          United States and its territories, using the 0.1 events/year EWL Wave set-up is the quasi-static rise in water level at the shore-                event as an example, this study estimates that wave set-up line due to breaking waves (Longuet-Higgins and Stewart,                          ranges from about 20-75 cm (Figure Box 3.1a) and swash 1963). Swash is the time-varying elevation of the leading edge                    from 35-125 cm (Figure Box 3.1b), together accounting for of wave uprush, which varies in frequency from seconds (due                      25%-90% of EWLs (Figure Box 3.1c and based on Vistousek to incident waves) to minutes (e.g., surf beat; Guza and Thorn-                  et al., 2017not this studys RFA-based EWLs) for open-coast ton, 1982). Wave set-up and swash components, collectively                        beaches (i.e., not for embayments protected from ocean known as wave run-up, are dependent on wave height, peri-                        waves). Wave-driven water levels (i.e., wave run-up) represent od, and beach slope (Stockdon et al., 2006) and are therefore                    ~50% or more of the EWL contributions (again, not from this controlled by local beach morphology and transient ocean                          study) in areas with narrow continental shelves (reduces surge conditions. To perform regional assessments of present-day or                    potential) and/or small tidal ranges, in particular the Hawaiian future wave-driven water level contributions, wave conditions                    and Pacific Islands, the Caribbean, the Outer Banks (North Car-are typically determined via global wave models forced by                        olina), most of Florida, the entire U.S. West Coast, and portions wind-reanalysis studies (e.g., Reguero et al., 2012) or histori-                  of Louisiana, Texas, and Alaska. But swash oscillations only cal/future wind fields produced by global climate models (e.g.,                  amplify coastal EWLs over short periods (i.e., seconds to min-Hemer et al., 2013).                                                              utes), whereas wave set-up represents a relatively sustained Global and Regional Sea Level Rise Scenarios for the United States l 41
 
Box 3.1 (cont.): Wave Contributions to Extreme Water Levels contribution during storm events with about a 10% to 80%            underestimated for open-coast beaches, especially along U.S.
contribution to EWLs, with the highest values in the tropics        island coastlines. Including wave-driven processes will be a (Figure Box 3.1d). As these examples indicate, when omitting        focus of subsequent Task Force attention leading up to the wave-driven processes, coastal flood risk can be significantly      Sixth National Climate Assessment (NCA6).
Figure Box 3.1. Water level contribution due to a) wave set-up and b) wave swash; c) percent contribution of wave-driven water levels (i.e., wave run up = wave set-up and swash) relative to all components: tide, storm surge, and waves; and d) percent contribution of wave set-up relative to the sum of tide, storm surge, and wave set-up based on model reanalysis of Vitousek et al. (2017).
Global and Regional Sea Level Rise Scenarios for the United States l 42
 
Section 4: Use Cases Below are four use cases, which use
* the (regional frequency analysis) RFA-based extreme water levels (EWLs) to map (at city scales) the annual probabilities/frequencies for the NOAA minor (disruptive), moderate (typically damag-ing), and major (often destructive) high tide flooding (HTF) layer classifications that are nationally calibrated to those used in weather-warning forecasting by NOAA;
* the relative sea level (RSL) projections and the RFA-based EWLs to incorporate trends (e.g., sea level rise projections) into design engineering criteria for risk management and adaptive planning;
* the RSL projections and RFA-based EWL probabilities with maps of NOAA minor, moderate, and major HTF layers to assess current and future vulnerabilities to combined storm and wastewater systems; and
* vertical land motion (VLM) rates inherent to the RSL projections are compared to rates from new satellite technologies at very high spatial resolution to showcase possibilities to monitor current rates from space and further localize the RSL projections.
The goal is to contextualize how the emerging science and this reports datasets can assist in developing products suitable for approaching (mapping, designing, or bounding) important problems in coastal risk assessment and management.
4.1. Mapping of NOAA High Tide Flood Thresholds and Flood Frequencies High tide flooding29 is increasingly common due to years of RSL rise. NOAA has been 1) documenting chang-es in minor HTF patterns since 2015, with about 100 NOAA tide gauges along the U.S. coastlines, and 2) providing a yearly coastal HTF outlook for these locations for the coming year,30 as well as projections for the next several decades based on RSL projections from NCA4/Sweet et al., 2017. NOAA has also mapped the three HTF depth-severity (minor, moderate, and major) categories based on the relationship with tide range (Sweet et al., 2018) to show the spatial extent of associated impacts (see Figure 3.7). The minor HTF maps are provided in the NOAA SLR Viewer,31 and all three map layers are accessible through NOAA map services.32 In an effort to provide better flood exposure information, NOAA is developing a product with input from partners (e.g., the Federal Emergency Management Agency [FEMA]) to assign exceedance probabilities us-ing the RFA-based EWLs to the minor, moderate, and major HTF categories as shown for Charleston, South Carolina, and West Palm Beach, Florida (Figure 4.1). The annual event frequency shown for each NOAA HTF zone is assigned to the particular flood height. For example, the moderate HTF zone in Charleston is shown as the orange-brown layer in Figure 4.1a, which includes all land elevations between the minor HTF height threshold (0.570 m above mean higher high water [MHHW]; see Table A1.2) and the moderate HTF threshold (0.853 m above MHHW). This moderate HTF zone is expected to be completely (up to 0.853 m above MHHW) at risk of flooding, with an average event frequency between about 1 event/year and 0.2 events/year. A frequency range is provided to partially address the 95% confidence intervals in both the EWL statistics and the mapping data. In the case of local maps, like Charleston and West Palm Beach, the average event frequency for each NOAA HTF layer is a constant across the area shown.
These types of products can help inform the probability of higher-frequency, lower-impact events. As agen-cies (e.g., FEMA) start to develop products that provide more comprehensive hazard and risk information 29 https://oceanservice.noaa.gov/facts/high-tide-flooding.html 30 https://tidesandcurrents.noaa.gov/HighTideFlooding_AnnualOutlook.html 31 https://coast.noaa.gov/slr/
32 https://coast.noaa.gov/arcgis/rest/services/dc_slr/Flood_Frequency/MapServer Global and Regional Sea Level Rise Scenarios for the United States l 43
 
(e.g., graduated flood risk; see The Future of Flood Risk Data33), there is a need to better define and resolve the probabilities of these more frequent flood conditions. In addition, considering todays height-severity flood thresholds in the face of sea level rise (see Figure 1.3), understanding the event probabilities in this more frequent space is critical. Such information would help graduate the flood probabilities more compre-hensively than FEMAs binary 1% annual chance floodplain definition and allow for a more comprehensive picture of structure-level risk.
How Can This Be Done?
The process to spatially assign probabilities again relies on a relationship to tide range (see Figure A2.5),
with tide range values obtained by subtracting VDatums MHHW and mean lower low water [MLLW] modeled tidal surfaces.34 Using VDatums tide range and the regional regression equations (Figure A2.5) to obtain a local index (u), the EWL return level (or rather, average event frequency) curves for the associated grid are downscaled to individual VDatum grid cells (~100 m) using Equation 1 in Section 3.2. With these downscaled curves, the HTF levels at each VDatum cellalso based on VDatums tide range (i.e., great diurnal tide range [GT] tide datum) relationships (Sweet et al., 2018)are intersected with the localized frequency curve (expected values) for the cell in order to determine event frequencies on a cell-by-cell basis. The average event frequencies are then associated with their respective mapped inundation footprints (3-5 m horizontal resolution). To refine the data, they were clipped to the coastal HUC (hydrologic unit code) 12 watersheds35 that overlapped VDatum model data. This was done in order to provide a probability in watersheds that con-tained source VDatum data only.
The value of these data is that we can now provide not only the mapped inundation extent of each of the three HTF levels (see Figure 3.7) but also the probability, or event frequencies, for each level on high-res-olution inundation data (Figure 4.1). By leveraging the relationship between the local indices (u) to GT on a regional basis, the EWL statistics can provide event frequencies for 1) most water levels or flood heights of interest and 2) most locations, even if there is not a local tide gauge nearby to assist coastal managers when planning for potential impacts to their communities. In terms of the mapped product and inherent uncer-tainties, it should be recognized that the VDatum models standard error is on the order of 15 cm,36 which is similar to that of the LIDAR elevation data.37 The associated 95% confidence intervals from both VDatum and the LIDAR used in the mapping is then (standard error x 1.96) about 30 cm and similar to that of the EWL at the 1 event/year frequency (0.3 m median) using tide range to spatially derive EWLlocal (Figure A2.5), although it increases to about 0.9 m at the 0.01 events/year frequency. Thus, it is recommended that these maps be used cautiously in any type of application.
Both NOAA and FEMA are currently exploring methods to further localize the EWLgridded probabilities, such as using NOAA short-term gauges (e.g., Section 3.4) and multidecadal hindcast modeling to develop a higher resolution set of local indices (u). FEMA is working to merge the higher-frequency portion of the EWL distri-butions (e.g., > 0.05 events/year) with the FEMA EWL stillwater datasets (some of which are shown in Figure 3.5). These efforts will serve, in general, to refine coastal exposure by todays standards and, specifically, mi-nor to major HTF probabilities to better understand and communicate the Nations coastal flood risk through products such as FEMAs National Risk Index.38 33 https://www.fema.gov/fact-sheet/future-flood-risk-data-ffrd 34 https://vdatum.noaa.gov/
35 https://www.usgs.gov/core-science-systems/ngp/national-hydrography/watershed-boundary-dataset?qt-science_support_page_related
_con=4 - qt-science_support_page_related_con 36 https://vdatum.noaa.gov/docs/est_uncertainties.html 37 https://www.usgs.gov/ngp-standards-and-specifications/lidar-base-specification-online 38 https://hazards.fema.gov/nri/
Global and Regional Sea Level Rise Scenarios for the United States l 44
 
Figure 4.1: Maps of the NOAA minor, moderate, and major high tide flooding layers for a) Charleston, South Carolina, and b)
West Palm Beach, Florida (as in Figure 3.7 but providing average event frequencies for each layer). Note: the shoreline on these maps is mean higher high water, but to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
4.2. Application of Scenarios, Observation-Based Extrapolations, and Extreme Water Levels Because future sea level rise amounts are inherently uncertain, planners and engineers who engage in ad-dressing adaptation to future sea level rise in coastal communities often adopt a scenario approach. Based on several national and regional sea level projections (Hall, Weaver et al., 2019; Parris et al., 2012; USACE, 2014; Hall et al., 2016; Sweet et al., 2017), many communities have developed their own specific scenario sets and guidelines for how to use them. In this section, the application of the regional sea level scenarios (see Section 2) that leverage the newly developed observation-based extrapolations (see Section 2.3) and the EWL probabilities produced using the RFA (see Section 3) are illustrated for representative locations around the United States.
This use case is not meant to provide standardized planning guidance for using information on sea level rise projections; rather, it is provided as an example of applying concepts of time-varying extreme value proba-bilities due to sea level rise, risk reduction, and adaptive planning that may be used in practice (Salas and Obeysekera, 2014; Salas et al., 2018). One of the primary tasks in coastal infrastructure projects is to deter-mine the design elevation (also known as the return level) of a particular structure (e.g., seawall or building) for a desired level of risk or probability. Such design problems typically require the knowledge of advanced statistical methods associated with extreme values such as those illustrated in the commonly referenced textbook by Coles (2001).
The use case is illustrated for 10 tide gauges around the United States (Figure 4.2). For reference, the up-per-bounding scenarios of the observation-based extrapolations for 2050 (see Table 2.2) and the RFA-based EWL distribution parameters (Section 3) are provided in Table 4.1. The EWL probability parameters are neces-sary to replicate this use case, and they are specifically from a Generalized Pareto Distribution (GPD) peaks-Global and Regional Sea Level Rise Scenarios for the United States l 45
 
over-threshold approach (Coles 2001): a) the local Index, u; b) rate of exceedances above the local index, ;
c) scale, RFA; and d) shape,  (see Section A2 for more details). In the examples below, the upper-bounding scenario is used (Figure 4.3a) with the corresponding return level curves for the selected tide-gauge loca-tions (Figure 4.3b).
Figure 4.2: Tide gauges selected for the application of sea level scenarios and extreme water level methods.
Table 4.1: Tide-gauge locations, scenarios bounding the observation-based extrapolations, and the extreme value distribution Generalized Pareto Distribution (GPD) model parameters estimated using the regional frequency analysis (RFA).
Upper-bounding scenarios circa 2050 Tide-gauge location details                                      RFA-based GPD parameters of the observation-based extrapolations Local NOAA ID          Location      Region      Upper Bound        Index                  RFA u
1612340      Honolulu, HI      Haw.              Int          0.248      3.19      0.218    0.066 8518750      The Battery, NY      NE              Int          0.546      2.98      0.261    0.179 8638610    Sewells Point, VA      NE              Int          0.502      2.95      0.332      0.067 8723214      Virginia Key, FL      SE            Int-High        0.284      3.00      0.152    0.251 8726520    St. Petersburg, FL  E. Gulf          High          0.337      2.99      0.266      0.354 8729840      Pensacola, FL      E. Gulf          High          0.345      2.85      0.212    0.456 Galveston Pier 8771450                        W. Gulf          Int-High        0.366      2.75      0.289    0.340 21, TX 9410660      Los Angeles, CA      SW            Int-High        0.472      3.21      0.150    0.063 9414290    San Francisco, CA      SW            Int-High        0.375      3.15      0.211    0.038 9447130        Seattle, WA        NW              Int          0.541      3.07      0.233    0.110 Global and Regional Sea Level Rise Scenarios for the United States l 46
 
Figure 4.3: a) RSL projections for the scenarios providing the upper bound to observation-based extrapolations to 2060 for the selected tide gauges. The corresponding scenario for each tide gauge is shown in parentheses in the legend. b) RFA-based EWL (see Section 3) return level curves relative to the 1983-2001 MHHW tidal datum. Notes: (1) to be useful for decision-making, a conversion to land-based heights (e.g., geodetic datum such as NAVD88) should be made. (2) Average event frequency (x-axis label) is the reciprocal of average recurrence interval, which is also known as return period.
As shown in Figure 4.3a, 2005 is the reference year for the projection scenarios. However, the return level curves shown in Figure 4.3b are referenced to the year 2000. The return level curves are first adjusted to the year 2005 by raising the curves by an amount equivalent to the local trend in the flood index (u) from 2000 to 2005 (see Table A1.3). Alternatively, the RSL offsets (see Table A1.2) could be applied, with differ-ences between the two insignificant to the results here.
Accounting for Time-Varying Relative Sea Level Rise A particular scenario depicts the changes in RSL at a selected location. A common assumption is that as RSL rises, the EWLs also increase, and that must be accounted for in the changing behavior of the probability distribution of the EWLs. One approach for developing a time-varying extreme value distribution is to as-sume that one or more parameters (location, scale, and shape) are functions of time or some other covariate (e.g., El Nino-Southern Oscillation index; Coles, 2001; Menendez and Woodworth, 2010). When two or more parameters evolve with time (i.e., strong nonstationarity), the paradigm shifts from a stationary approach, typically used for planning infrastructure until recently, to one reflecting significant temporal change in the probability distribution. A common practice is to remove the trend in the extreme dataset and then to as-sume the distribution of the detrended extremes to be stationary. This approach is similar to the case when only the location parameter is varying with time and the other parameters are constant.
In the ensuing sections, it is assumed that only the location parameter (i.e., local index, u, in GPD) changes as a function of RSL (i.e., per the specified sea level scenario). This may be expressed as where u is the RFA/GPD local index that is a function of RSL, and  and  are scale and shape parameters, respectively, which are assumed to be constant over time. However, this assumption does not preclude the analysis of using a higher degree of temporal variability (e.g., both u and  are functions of RSL or some other covariate). As a consequence of the above assumption, the local index u is adjusted by a magnitude (i.e., the regional mean sea level change from the reference year) obtained from a selected scenario.
Global and Regional Sea Level Rise Scenarios for the United States l 47
 
For planning infrastructure using the scenarios RSL projections and the EWL probabilities, two approaches are illustrated: 1) recurrent flood frequency and 2) time-varying average recurrence interval (ARI; which is the reciprocal of average event frequency [AEF]) and risk.39 While the infrastructure designs are based on a vari-ety of factors, one or both of these approaches may be used to support that process (e.g., height of a sea-wall or base-flood elevation). In this use case, the term flood could pertain to a particular NOAA HTF level or an arbitrary probabilistic EWL level, although not necessarily to imply a meteorological (e.g., storm) event.
Designs Based on Recurrent Flood Frequency In many U.S. coastal locations, the frequency of flooding is increasing, mostly due to rising sea levels (Sweet et al., 2021). A community may tolerate infrequent flooding initially, but at some point, when the sea level rise is significant, the flooding frequency will increase, which in turn may exceed that communitys risk tolerance for flooding. Using the extreme value distributions and the sea level scenarios, it is possible to predict the time-varying change in frequency (e.g., as in Figure 3.9). In case of the GPD, the recurrent flood frequency (number of exceedances above a return level [z]) may be computed as (Buchanan et al., 2017) where  is the change in RSL (relative to the project construction year) obtained from Figure 4.3a.
In the example used here, the planning problem may be stated as follows: What should the initial return level (used for the design) be to ensure that the recurrent flood frequency is limited to a specified number of events at the end of the design life? It is now possible to lay this out graphically, as shown in Figure 4.4 for two tide gauges (Sewells Points, Virginia, and Galveston Pier 21, Texas).
Figure 4.4: Recurrent flood frequency estimates for a) Sewells Point (Norfolk), Virginia, and b) Galveston Pier 21, Texas. For both, the relative sea level projection for the scenarios and the return level are the same as in Table 4.1. Note: to be useful for deci-sion-making, a conversion of the return level to land-based heights (e.g., geodetic datum such as NAVD88) should be made.
In Figure 4.4, the number to the right of each point along the curve shows the recurrent flood frequency, N, corresponding to the year indicated on the left. For this example, it was assumed that by 2060, the desired value of N = 1, and the design AEF necessary for this criterion, is indicated in Figure 4.4 (AEF = 0.06 events/
year for Sewells Point and AEF = 0.05 events/year for Galveston Pier 21). The corresponding design re-turn levels are 1.31 m and 1.35 m, respectively, relative to MHHW datum. A summary of results for all 10 tide 39 In the context of Section 4.2, risk is defined as the probability of one or more events exceeding a given height threshold over the life of a project.
Global and Regional Sea Level Rise Scenarios for the United States l 48
 
gauges is shown in Table 4.2. The design average event frequency required in 2005 to meet the flood fre-quency criteria shows significant variability across the sites. The design return level depends on two factors:
: 1) the magnitude of the sea level rise from 2005 to 2060 (end of the design life); and 2) the slope (a function of the scale and shape parameters) of the return level curve (Figure 4.3b).
Table 4.2: Summary of design parameters to constrain the average event frequency, N, to 1 per year by 2060 (end-year of the design life). The 2005-2060 RSL projections are the local values associated with the scenarios providing the upper bound to the regional observation-based extrapolations shown in Table 2.2. Note: to be useful for decision-making, a conversion of the return level to land-based heights (e.g., geodetic datum such as NAVD88) should be made.
Design average event Return level (m above Return level (m above Relative Sea level                                              frequency (events/
1983-2001 MHHW)        1983-2001 MHHW)
NOAA ID              Location        rise (in meters from                                              year) required in corresponding to      required in 2005 to 2005 to 2060)                                                2005 to achieve N = 1 AEF = 1 year      ensure N = 1 by 2060 by 2060 1612340            Honolulu, HI              0.39                0.33                  0.72                <0.01 8518750          The Battery, NY            0.50                0.76                  1.26                0.10 8638610          Sewells Point, VA            0.56                0.75                  1.31                0.06 8723214          Virginia Key, FL            0.55                0.44                  0.99                  0.01 8726520        St. Petersburg, FL            0.70                0.49                  1.19                0.05 8729840            Pensacola, FL              0.66                0.47                  1.13                0.06 8771450        Galveston Pier 21, TX          0.77                0.58                  1.35                0.05 9410660          Los Angeles, CA              0.41                0.57                  0.98                <0.01 9414290        San Francisco, CA            0.46                0.49                  0.95                <0.01 9447130            Seattle, WA              0.29                0.70                  0.99                  0.05 Design Based on Time-Varying Exceedance Probabilities Average recurrence interval is used to describe EWL probabilities in the following examples to directly relate to and build off of a couple of recent, relevant focused studies on the topic. Interpretation of the results should follow guidelines of the U.S Army Corps of Engineers (USACE, 1994).
In current practice, the projects with a longer design life (> 25 years) typically use a low average event fre-quency (<0.1 events/year) or, equivalently, a high/long ARI (> 10 years or more). At high recurrence intervals, the peaks-over-threshold and the annual maxima recurrence intervals converge (Langbein, 1949), although not necessarily where tropical storm surges are present (Wahl et al., 2017). Revisiting the concepts of tra-ditional ARI and risk concepts for annual maxima in time-varying frameworks has been addressed recently (e.g., Salas and Obeysekera, 2014). The application of time-varying ARI and risk concepts is illustrated by converting the GPD model to an equivalent annual maxima model, which in this case is the GEV distribution.
The equivalent annual-maxima modeling approach, as used here, will also facilitate the direct application of emerging risk and recurrent interval concepts already developed for situations of time-varying extreme prob-abilities (Salas and Obeysekera, 2014; Salas et al., 2018; Obeysekera and Salas, 2020).
The cumulative distribution function (CDF) of the GEV model of annual maxima is expressed as where , ,  are the location, scale, and shape parameters of the GEV (Coles 2001).
Global and Regional Sea Level Rise Scenarios for the United States l 49
 
For computing u, the local index is further adjusted to reflect the translation of the return level curve from 2000 to the reference year (i.e., 2005). The GEV scale parameter,  =  , where the at-site scale parameter
, is computed as  = RFA*u. For this use case, the adjusted local index is computed as uadj = u
* s (2005-2000), where s is the trend of the local index u at the site (see Table A1.3). If desirable, other adjustment procedures may be used. Finally, the time-varying GEV model assumes that only the location parameter, ,
changes with sea level change,  and the time varying annual extreme value distribution is given by The exceedance probability, pt, which corresponds to an initial return level (zq0, initial design), changes with time because of the rising RSL,  (Figure 4.5). Consequently, the ARI is not a fixed measure but decreases with increasing sea level.
Figure 4.5: Conceptual illustration of increasing exceedance probability (hence decreasing average recurrence interval) that assumes that the location parameter is a function of the magnitude of the relative sea level rise.
The traditional concept of the ARI is the average waiting time for between two successive exceedances of the return level. Using the same definition but in a time-varying exceedance probability framework (Figure 4.5), an equivalent measure of ARI (T) may be derived as (Cooley, 2013; Salas and Obeysekera, 2014) where pt = 1  Ft (z, ) i is the time-varying exceedance probability. If a project is designed for a return period, T0[t = t0], then T < T0 implies that the actual recurrence interval due to rising RSL will be less.
Global and Regional Sea Level Rise Scenarios for the United States l 50
 
The methods described in the preceding paragraphs are applied to the 10 tide-gauge locations shown in Figure 4.2. For illustration, it was assumed that the projection scenario for each tide gauge would continue beyond 2060. However, the methodology described above can be used with any other scenario. The de-rived GEV parameters for each gauge are shown in Table 4.3.
Table 4.3: The parameters of generalized extreme value computed using the peaks-over-threshold Generalized Pareto Distribu-tion model (Coles 2001).
Local index At-site scale                            GEV location        GEV scale          GEV shape NOAA ID            Location                            adjustment from parameter                                parameter          parameter            parameter 2000-2005 (m) 1612340        Honolulu, HI            0.054              0.007                0.330              0.058                0.066 8518750      The Battery, NY          0.142              0.016                0.757              0.173                0.179 8638610      Sewells Point, VA          0.167              0.023                0.748              0.179                0.067 8723214      Virginia Key, FL          0.048              0.026                0.444              0.063                0.251 8726520      St. Petersburg, FL        0.090              0.014                0.494              0.132                0.354 8729840        Pensacola, FL            0.073              0.012                0.474              0.118                0.456 8771450    Galveston Pier 21, TX        0.106              0.033                0.579              0.149              0.340 9410660      Los Angeles, CA          0.071              0.005                0.565              0.066                0.063 9414290      San Francisco, CA          0.079              0.010                0.492              0.083                0.038 9447130          Seattle, WA            0.126              0.010                0.701              0.111              0.110 The ARI curves,T, as a function of T0, for all 10 tide gauge locations are shown in Figure 4.6a. This figure demonstrates that, in all cases, the actual ARI is less than the design recurrence interval. For instance, for a location near Pensacola, Florida, if a project is designed for T0 = 100 years, the actual ARI, due to future RSL rise (Table 4.1, Upper Bound column), is only about 50 years. As another example, for a location near The Battery, New York City, a project may need to be designed for T0 = 90 years if the desired ARI under its asso-ciated (Table 4.1, Upper Bound column) RSL rise scenario is 40 years.
Figure 4.6: a) Average recurrence interval (due to rising RSL) curves ( T versus T0 ) at each tide gauge using the selected scenar-ios RSL projection (see Table 4.1). b) Risk curves as a function of design life: stationary (black curve), actual risk resulting from incorporating the sites RSL scenario projection (red curve), and risk curve for a specific risk (blue curve).
Global and Regional Sea Level Rise Scenarios for the United States l 51
 
Risk-Based Design Under stationary conditions, the risk (defined as the probability of one or more exceedances above the design elevation) is a function of the life of the project, n. The risk formula under stationarity is given by R = 1  (1  1/T0)n. For example, there is about a (R = 0.26) 26% chance of experiencing an event with an ARI of (T0) 100 years over the course of (n) 30 years under a non-changing (stationary statistical) environment. As the length of the design life increases, risk also increases. Under conditions of time-varying exceedance probability, pt, the risk (R) formula is (Salas and Obeysekera, 2014)
With rising relative sea levels, pt increases, and the risk is higher than that under stationarity. This increase in risk is illustrated for the San Francisco, California, tide gauge in Figure 4.6b when the initial design, T0 = 50 years (the event level with a 50-year ARI). The black curve in Figure 4.6b shows the increasing risk as the design life becomes longer even under stationarity. For instance, if the design life, equals 25 years, this risk is about 0.4 (40%). However, when the local sea level rise scenario is incorporated, the risk over a given life of the project increases more rapidly, exceeding the corresponding risk under stationarity (see red curve in Figure 4.6b). In the above example, when n = 25 years, the risk will increase to about 60% due to the RSL scenario projection. Moreover, the RSL rise causes the risk to approach 100% (R = 1) when the design life is about 50 years or more. In the risk-based design approach, one can specify the tolerable risk and determine the initial design period (or return level).
One option is to design a project in such a way that the resulting increasing risk profile due to application of the scenarios RSL projection is at or below that under stationarity. While the risk-reduction approach described below is illustrated for a selected RSL scenario for the future, it can be implemented for multiple scenarios, leading to a variety of risk-reduction options depending on the future RSL scenarios. In such a broader application, a risk-based framing founded on risk tolerance may be adopted.
Considering uncertainty in the sea level rise projections, one may wish to approach the problem using con-cepts of dynamically adaptive planning. In the example shown in Figure 4.6b (blue curve), two parameters are specified to illustrate this concept. First, it is assumed that the project will be constructed in, for example, two or more phases. Considering such a planning assumption, phase I is 25 years long (i.e., n = 25 years),
and the maximum tolerable risk during this phase is 0.3 (30%), as opposed to the 60% risk mentioned above.
The blue curve shows the risk profile for such a design. This curve was computed by constraining R = 0.3 when n = 25, as shown by the green dot in Figure 4.6b. The implication of this adaptive approach is that the initial return level will need to increase from 0.84 m MHHW to 0.93 m MHHW (Table 4.4), and the corre-sponding initial ARI has to increase from 50 years to 125 years. In this approach, one must also assume that the project will be expanded after that initial period, and measures must be adopted to prevent locking in the design and preempting the planners from expanding it into a bigger project after the initial 25-year peri-od. For example, the foundation design of the project may need to assume the eventual capacity expansion and allow for it in the initial design. This approach of dynamically adaptive planning is becoming increasingly popular as a way to deal with deep uncertainties associated with sea level rise.
Global and Regional Sea Level Rise Scenarios for the United States l 52
 
Table 4.4 shows that with a relatively small increase in initial design elevation, the risk can be managed to a desirable level. In this example, however, the ultimate design (at the end of the full design life; e.g., 50 or 100 years) needs to be assessed to ensure that resources (e.g., land) that may be needed for the build-out are considered.
Table 4.4: Results of the risk-based design for all tide gauges shown in Figure 4.2. Average recurrence interval (ARI) is listed and is the reciprocal of average event frequency. Values in the last column have been rounded to the closest 5-year interval. Note: to be useful for decision-making, a conversion of the return level to land-based heights (e.g., geodetic datum such as NAVD88) should be made.
Design return                                      Average recurrence interval Design return level to level for T0 = 50                                  (ARI) of the design to constrain NOAA ID              Location                          constrain risk to 30% over a years (m above                                    probability (risk) to 30% over a 25-year period (m MHHW)
MHHW)                                                  25-year period 1612340          Honolulu, HI            0.59                    0.69                              >100 8518750        The Battery, NY            1.74                    1.95                              90 8638610        Sewells Point, VA          1.55                    1.75                              >100 8723214        Virginia Key, FL          0.78                    1.00                              >100 8726520        St. Petersburg, FL          1.61                    1.88                              80 8729840          Pensacola, FL            1.75                    2.09                              75 8771450      Galveston Pier 21, TX        1.79                    2.13                              85 9410660        Los Angeles, CA            0.79                    0.86                              >100 9414290        San Francisco, CA          0.84                    0.93                              >100 9447130            Seattle, WA            1.05                      1.13                            >100 4.3. Growing Risk to Combined Storm and Wastewater Systems from Sea Level Rise Sea level rise is causing HTF to become more severemore frequent, deeper, and more widespreadin terms of its impacts (Sweet et al., 2021). Coastal areas that are not exposed to HTF now may become so in the coming decades. As the footprint of flooding expands, water from adjacent estuaries and bays will flood into communities and encounter previously unaffected urban infrastructure.
Many places already see backflow from tidal waters through stormwater pipes that spill out of catch basins into neighborhood streets. Cities with combined sewer systems often have backflow preventers on their vulnerable outfall pipes (EPA, 1995a, 1995b). However, combined sewers will be open to inflow from surface flooding. If floodwater in the streets encounters a catch basin that connects to a combined sewer system, then high tide waters will enter the sewer. At best, the tide waters will be on their way to the sewage treat-ment plant; at worst, a combined sewer outflow would be triggered if the sewer pipes cannot handle the volume of water.
While Camden, New Jersey, has taken action to prevent runoff from entering its system,40 tidal inflow is a novel problem. Identification of risks like this can provide lead time to take adaptation actions. Still, in some combined sewer communities, such as Camden, the onset of risk can arrive well before midcentury. Map-ping shows that minor HTF at a height of 0.58 m above current MHHW tidal datum (Table A1.3) begins to have a footprint in Camden neighborhoods served by combined sewers (red shade in Figure 4.7, spanning from MHHW to 0.58 m [1.9 feet] above MHHW; locations are provided by the New Jersey Department of Environmental Protection41). By the time the tide reaches the moderate (0.86 m above MHHW) and major 40 https://www.epa.gov/arc-x/camden-new-jersey-uses-green-infrastructure-manage-stormwater 41 https://njdep.maps.arcgis.com/apps/Viewer/index.html?appid=70dd49de342949ca933e840d0c530fc7 Global and Regional Sea Level Rise Scenarios for the United States l 53
 
(1.25 m above MHHW) HTF levels, the extent of flooding increases dramatically, and many intersections will be flooded.
The Camden region currently (circa 2020) experiences
* about 2 events/year (or about 4 days/year per Figure 3.8b) of minor HTF;
* 0.2 events/year of moderate HTF; and
* 0.03 events/year of major HTF, based on the EWLlocal directly across the Delaware River at the NOAA tide gauge in Philadelphia. The EWL-based probabilities support actual observations in 2020, when the Camden/Philadelphia region experienced 4 days of minor HTF, with 4-8 days projected to occur in 2021 (Sweet et al., 2021).
Considering the Intermediate scenario, which is the upper-bounding scenario for this regions RSL obser-vation-based extrapolations (see Table 2.2), a rise of 0.19 m by 2030 (measured since 2005) is projected to result in
* 5-10 events/year (on the order of 10-20 days/year) of minor HTF,
* 0.6 events/year of moderate HTF, and
* 0.07 events/year of major HTF.
By 2050, a 0.38 m RSL rise is projected (above 2005 levels) for this area, resulting in
    *  >10 events/year (perhaps >20 days/year) of minor HTF,
* about 3 events/year (6 days/year) of moderate HTF, and
* 0.3 events/year of major HTF.
So, within about the next 30 years (by 2050), a surface flood regime shift with subsurface impacts is project-ed to occur in Camden, considering current RSL rise trajectories. By then, moderate and major HTF (flooding upwards of 0.9 m and 1.2 m above MHHW, respectively) is projected to occur with similar frequencies/prob-abilities as minor (about 0.6 m above MHHW) and moderate HTF occur today. With nearly 4 high tides per event (1 event lasts about 2 days; 2 high tides occur almost every day), this implies that by 2050, upwards of 80 tides per year or more at the minor HTF level are projected, with about 12 of those tides per year exceed-ing the moderate HTF level and a 0.3 events/year frequency of major HTF flooding. Any time street inter-sections are underwater, tidal waters could flow down catch basins into the combined system (Figure 4.7).
Beyond 2050, HTF frequency, depth, and extent will continue to grow. It is unclear how this increased flood frequency will affect the combined sewer systems functionality and surrounding water quality.
Global and Regional Sea Level Rise Scenarios for the United States l 54
 
Figure 4.7: Location of combined stormwater and sewer system outfalls that are likely draining regions exposed to HTF within the Camden, New Jersey, region, with the minor (red: MHHW to 0.58 m [1.9 feet] above MHHW), moderate (orange: MHHW to 0.86 m [2.8 feet] above MHHW), and major (yellow: MHHW to 1.25 m [4.1 feet] above MHHW) HTF layers stacked in the enlarged map and individual layers mapped to the right. Note: heights are relative to the 1983-2001 tidal epoch, and to be useful for decision-making, a conversion to land-based heights (e.g., NAVD88) should be made.
4.4: Use of InSAR Technology for Determining Regional Vertical Land Motion and Its Suitability for Computing Long-Term Sea Level Rise Projections Vertical land motion is an important component of RSL rise, leading to changes in the height of the ocean relative to land. Vertical land motion is not a singular phenomenon but instead results from various process-es that display different patterns in space and time. These patterns have different impacts from place to place, especially in coastal settings where many of them operate at the same time and can serve to either increase RSL (subsidence) or decrease RSL (uplift). For much of the coastal United States, subsidence is driven on local scales by both natural processes, such as compaction of river sediments, and unnatural, hu-man-caused reasons, such as groundwater and fossil fuel withdrawal; on larger scales, subsidence is driven by glacial isostatic adjustment (GIA). On the other hand, in some regions, such as southern Alaska, GIA leads to high rates of uplift in coastal regions. For example, Grand Isle, Louisiana, has experienced more than 0.9 m (3 feet) of RSL rise, whereas Juneau, Alaska, has experienced more than 1.2 m (4 feet) of RSL fall based on a 100-year historical linear rate value,42 in large part due to VLM. For perspective, the national median RSL rise along U.S. coastlines during this 100-year period was about 0.25-0.30 m (see Figure 1.2b).
42 https://tidesandcurrents.noaa.gov/sltrends/
Global and Regional Sea Level Rise Scenarios for the United States l 55
 
Accurate future projections of VLM require an understanding of and accounting for the underlying process-es and the time and space scales on which they vary. In this report, VLM projections are based in part on analysis of past observations. Vertical land motion rates are estimated at tide-gauge locations as well as at 1-degree grids using a statistical model of tide-gauge observations (Kopp et al., 2014; Sweet et al., 2017; Fox-Kemper et al., 2021; Garner et al., 2021). The model assesses RSL change across the global tide-gauge network43 with data through about 2019 and separates the tide-gauge observations into 3 modes: 1) a global rise signal (Dangendorf et al., 2019), 2) a long-term linearbut regionally varyingrate, and 3) local effects that vary in time and by region. It is the second mode that defines this reports linear VLM rates, which have been incorporated into the RSL projections for each GMSL rise scenario. These rates are assumed to be linear over the past record and to persist linearly into the future over the length of the projected record.
Assumed persistence may not necessarily be valid over the long term (e.g., if groundwater pumping ceases) but may be necessary due to a lack of data. As shown in Figure 4.8a, high rates of subsidence are estimated along the entire Gulf Coast, and moderate rates of subsidence are assessed along the entire East Coast. On the other hand, high rates of uplift are estimated for the southern coast of Alaska.
Over the past couple of decades, GPS stations have provided estimates of VLM in coastal areas across the United States. These GPS-based VLM estimates provide a comparison to the VLM rates in this report, al-beit with a couple of caveats. First, the record lengths over which the GPS-based estimates are computed are significantly shorter than the tide-gauge data records used to infer the VLM rates in this report. Second, many tide-gauge locations do not have a co-located GPS station. While it is not possible to extend the record lengths of the available GPS measurements, the second challenge has been addressed using the GPS-im-aging technique discussed in Hammond et al. (2021), which leverages the GPS network in coastal areas of the United States to generate VLM estimates at all tide-gauge locations (Figure 4.8b). Note that negative values of VLM reflect subsidence while positive values reflect uplift. To determine the VLM contribution to RSL at the coast, the negative and positive direction would be reversed. Broadly, the GPS-based estimates are consistent with the VLM estimates contained in this report. However, when subtracting the VLM rates in this report from the GPS-derived rates, differences become apparent (Figure 4.8c). The largest differences are found along the Southern Alaska coastlines, where rates of uplift are very large, and along the entire Gulf Coast, where subsidence rates are large. The rates are further compared in Figure 4.8d, which again reflects general agreement between the two sets of estimates, although at roughly 75% of the gauges, the tide-gauge-based VLM estimate in this report is greater (less negative in the case of subsidence) than that from GPS. In other words, there are generally higher rates of subsidence indicated in the GPS rates when compared to the VLM estimates in this report.
This comparison with the GPS is not intended to be an assessment of the accuracy of VLM rates and asso-ciated projections included in this report. Instead, it highlights some of the challenges associated with both estimating VLM rates at the coast and then projecting these into the future, particularly away from the tide-gauge and GPS stations. The spatial variability and local drivers of VLM are clear in Figure 4.8, and extending the tide-gauge-centered estimates to fill in spatial gaps either through the projection framework in this report or with GPS imaging is challenging to validate, particularly as these methods are not intended to capture VLM varying on small spatial scales. An opportunity is provided, however, by new technologies using sat-ellite-based advanced Interferometric Synthetic Aperture Radar (InSAR) analysis, which can provide higher spatial resolution measurements of VLM rates. Calibrated to land GPS station estimates, measurements of land elevations over time by InSAR are producing VLM rates for large swaths or the U.S. coastal plain (e.g.,
Bekaert et al., 2017; Buzzanga et al., 2020; Bekaert et al., 2019; all InSAR VLM estimates are publicly avail-able through references). Having a higher-resolution assessment of VLM rates can in turn help communities understand where VLM is now occurring at very fine scales (e.g., street block level) and help make informed decisions of how continued VLM will contribute to future RSL projections. Furthermore, InSAR provides an 43 https://www.psmsl.org/data/
Global and Regional Sea Level Rise Scenarios for the United States l 56
 
additional component to the coastal VLM observing network. Integrated assessments across tide gauges, GPS, and InSAR are likely to be most useful for inferring VLM rates and projecting these rates forward at the spatial scales key to coastal communities. Following is a case study of how the InSAR VLM connects to this VLM-observing network. In general, as there is the possibility of using a user-defined VLM rate within the RSL projections, we examine other sources of VLM that may offer options.
Figure 4.8: Comparison of vertical land motion (VLM) rate estimates (mm/year) from a) the scenario-based framework used in this report, and b) GPS-imaging estimates from Hammond et al. (2021). c) The difference between GPS-derived rates and scenar-io-derived rates and d) a comparison of the VLM estimates at the U.S. tide-gauge locations are also shown. Negative values of VLM reflect subsidence, while positive values reflect uplift.
Hampton Roads, Virginia The historical long-term linear RSL rise rate at the Sewells Point, Virginia, tide gauge44 is about 4.7 mm/
year. More than half of this rate is estimated to be from downward VLM or subsidence with a rate of about 2.9 mm/year, which is close to previous estimates (Zervas, 2013; Kopp et al., 2014; Sweet et al., 2017). This subsidence is driven by both GIA and more localized groundwater withdrawal. If assumed to be linear and persistent into the future, VLM will contribute about 0.29 m to projections of RSL over the next 100 years.
For example, by 2050 under the Intermediate-Low and Intermediate scenarios, the amount of RSL rise is projected to be between about 0.4 m and 0.45 m, respectively, with about 35% and 30% of that rise amount, respectively, from VLM.
However, VLM rates across the Hampton Roads region are not uniform. A past study (Eggleston and Pope, 2013) leveraged a variety of in situ observations to find a spatially varying pattern of subsidence ranging from 1.8 to 4.4 mm/year in the region from 1940 to 1971. The variations were connected to groundwater withdrawal in the region, which was captured via this assessment even with an effective spatial resolution on the order of tens of kilometers. More recently, InSAR rate maps have shown a range of subsidence from 44 https://tidesandcurrents.noaa.gov/stationhome.html?id=8638610 Global and Regional Sea Level Rise Scenarios for the United States l 57
 
about 1 mm to 5 mm/year in the region over the time period from 2014 to present, with locally higher rates (Figure 4.9; Buzzanga et al., 2020). Importantly, the satellite-based assessment revealed spatial variations on sub-kilometer scales, with some of the most prominent features in the spatial map connected to specific construction projects and land-use changes. With an average rate of subsidence around 3 mm/year over the course of the 21st century, VLM could contribute about 0.3 m to projected RSL, with locally higher amounts elsewhere in the region. Furthermore, comparing the InSAR-derived spatial pattern of VLM to that in either Eggleston and Pope (2013) or the gridded rates in this report provides important information about the linearity of VLM and the timescales on which VLM varies. There are considerable differences between the different assessments, indicating a shift in rates over the time periods considered. While it is necessary to consider the uncertainty in the VLM rate estimates and differences in measurement type, users of VLM infor-mation should assess land-use changes over the time periods considered along with the relevant processes driving VLM in the region. InSAR-derived VLM maps will play an increasingly key role in this assessment due to the spatial coverage and resolution provided by the satellites.
Figure 4.9: Map showing VLM rates (mm/year) for the Hampton Roads region displayed on top of satellite imagery. Higher rates of subsidence are indicated by darker orange colors. Of particular interest is the range of rates in such a small region (e.g., on the order of up to 5 mm/year difference in places). Based on Buzzanga et al. (2020).
Global and Regional Sea Level Rise Scenarios for the United States l 58
 
Observing and Projecting Coastal Vertical Land Motion While InSAR-measured VLM provides advantages over other measurement platforms in terms of spatial cov-erage and resolution, it should be considered in the context of the larger observing network when assessing VLM at the coast. In particular, InSAR serves two potential roles. First, InSAR can be used to provide ongoing monitoring of VLM at high spatial resolutions. InSAR has the potential to generate time series of VLM on a fine spatial scale. Subsidence hotspots can be identified along with abrupt shifts in VLM, which can assist in planning and executing adaptation efforts. For coastal communities attempting to alleviate subsidence in their region through efforts such as groundwater reinjection, InSAR provides a potentially better alternative to in situ monitoring to assess the effectiveness of these efforts. Second, InSAR can serve to assess spatial variability in VLM, filling in the gaps between tide gauges and GPS stations in coastal regions. The obser-vations can then be combined in a statistical framework to provide more accurate projections of VLM with better estimates of uncertainty.
Assessing VLM with InSAR is not without challenges, however, although many of these are being addressed in ongoing and planned efforts. First, to be useful for assessing long-term VLM rates with the still relatively short satellite records, the shorter-term VLM rates can be calibrated and tied into the existing National Spa-tial Reference System (NSRS)45 to improve accuracy and representativeness of long-term changes. Second, the availability and coverage of GPS in coastal regions impact the accuracy of VLM by InSAR. To provide a measurement of absolute VLM, InSAR needs to be tied to available GPS measurements. In areas with large gaps between GPS stations, this can lead to reduced accuracy of the InSAR estimates. Ideally, analysis would be conducted to determine optimal GPS station spacing for maintaining integrity of the InSAR-derived velocity field in various environments, including, but not limited to, regions of coastal subsidence, landslide/
earthquake/volcanic activity, high plains aquifer depletion, and aquifer depletion in a tectonic area. Finally, In-SAR VLM estimates are computationally expensive to perform over large regions, making national coverage a challenge. Efforts are underway, however, to generate a consistent surface displacement product (a pre-liminary step to estimating VLM) for the United States. A generalized approach for generating absolute VLM estimates from this product could then be created, paving the way for ongoing monitoring of VLM along the U.S. coastlines at high spatial resolutions.
To improve projections of VLM, InSAR alone is not sufficient. Instead, InSAR should be analyzed in tan-dem with available tide-gauge, GPS, and any other available in situ observations to assess both the spatial variability of VLM rates and potential non-linearities in the VLM rates estimated over these records. These non-linearities are critical for determining the future contribution of VLM to RSL. For example, the long-term rate assessed at a tide gauge as done in this report could differ significantly from the rate of VLM over the past decade because of a sustained land-use change. The comparison between the two types of VLM esti-mates in Figure 4.9 indicate that these shifts may be present at some locations along the U.S. coastlines and need to be assessed to improve projections of VLM.
45 https://oceanservice.noaa.gov/education/tutorial_geodesy/geo08_spatref.html Global and Regional Sea Level Rise Scenarios for the United States l 59
 
Section 5: Conclusions Sea level rise driven by global climate change is a clear and present risk to the United States, now and for the foreseeable future. It is the goal of the Sea Level Rise and Coastal Flood Hazard Scenarios and Tools Interagency Task Force to continue to provide projections and future scenarios to assist decision-makers for both planning and risk-bounding purposes. This report builds upon the progress made in Sweet et al. (2017),
updating the GMSL scenarios and the associated local and regional RSL projections to reflect recent ad-vances in sea level science, as well as expanding the types of scenario information provided to better serve stakeholder needs for coastal risk management and adaptation planning.
The major findings of this report are as follows:
Multiple lines of evidence provide increased confidence, regardless of the emissions pathway, in a narrower range of projected global, national, and regional sea level rise at 2050 than previously reported (Sweet et al., 2017).
Both trajectories assessed by extrapolating rates and accelerations estimated from historical tide-gauge ob-servations, and model projections, fall within the same range in all cases, giving higher confidence in these relative sea level (RSL; land and ocean height changes) rise amounts by 2050. Specifically, RSL along the contiguous U.S. (CONUS) coastline is expected to rise, on average, as much over the next 30 years (0.25-0.30 m over 2020-2050) as it has over the last 100 years (1920-2020). Due to processes driving regional changes in sea level, the report found regional differences in both the modeled scenarios and observa-tion-based extrapolations, with higher RSL rise along the East (0-5 cm higher on average than CONUS) and Gulf Coasts (10-15 cm higher) as compared to the West (10-15 cm lower) and Hawaiian/Caribbean (5-10 cm lower) Coasts.
For coastlines outside CONUS, and for individual regions and locations within CONUS, the projections can differ from the aforementioned mean values. In addition, it is important to note that the projections do not include natural year-to-year sea level variability that occurs along U.S. coastlines in response to climatic modes such as the El Nino-Southern Oscillation. Nevertheless, if we assume that regional sea level will keep following its present trajectory for the coming three decades, most U.S. regions are mostly tracking between the Intermediate-Low and Intermediate-High scenarios. Although the near-term observation-based extrap-olations will continue to evolve over time with new observations and analyses, this updated information should help inform both near-term decisions and projects that may require decades worth of planning prior to actual implementation.
By 2050, the expected relative sea level (RSL) will cause tide and storm surge heights to increase and will lead to a shift in U.S. coastal flood regimes, with major and moderate high tide flood events occurring as frequently as moderate and minor high tide flood events occur today. Without additional risk-reduction mea-sures, U.S. coastal infrastructure, communities, and ecosystems will face significant consequences.
Minor/disruptive high tide flooding (HTF; about 0.55 m above mean higher high water [MHHW]) is projected to increase from a U.S average frequency of about 3 events/year in 2020 to >10 events/year by 2050. The projected increases for moderate/typically damaging (about 0.85 m above MHHW) and major/often de-structive (about 1.20 m above MHHW) HTF are 0.3 events/year in 2020 to about 4 events/year in 2050 and 0.04 events/year in 2020 to 0.2 events/year by 2050, respectively. Across all severities (minor, moderate, major), HTF along the U.S. East and Gulf Coasts will largely continue to occur at or above the national aver-age frequency.
Global and Regional Sea Level Rise Scenarios for the United States l 60
 
In other words, much of the coastline is already close to a flood regime shift with respect to flood frequen-cy and, consequently, damages. Only a small height difference (0.3-0.7 m) currently separates infrequent, damaging, or destructive HTF from the current regime of more frequent, so-called nuisance, flooding (whose impacts are in fact already remarkable throughout dozens of densely populated coastal cities). Decades ago, powerful storms were what typically caused coastal flooding, but due to RSL rise, even todays common wind events and seasonal high tides are already regularly flooding communities, and they will do so to an ever greater extent in the next few decades, affecting homes and businesses, overloading stormwater and wastewater systems, infiltrating coastal groundwater aquifers with saltwater, and stressing coastal wetlands and estuarine ecosystems.
Higher global temperatures increase the chances of higher sea level by the end of the century and beyond.
The scenario projections of relative sea level (RSL) along the contiguous U.S. (CONUS) coastline are about 0.6-2.2 m in 2100 and 0.8-3.9 m in 2150 (relative to sea level in 2000); these ranges are driven by uncer-tainty in future emissions pathways and the response of the underlying physical processes.
With an increase in average global temperature of 2&deg;C above preindustrial levels, and not considering the potential contributions from ice-sheet processes with limited agreement (low confidence) among modeling approaches, the probability of exceeding 0.5 m rise globally (0.7 m along the CONUS coastline) by 2100 is about 50%. With 3&deg;-5&deg;C of warming under high emissions pathways, this probability rises to >80% to >99%.
The probability of exceeding 1 m globally (1.2 m CONUS) by 2100 rises from <5% with 3&deg;C warming to almost 25% with 5&deg;C warming. Considering low-confidence ice-sheet processes and high emissions pathways with warming approaching 5&deg;C, these probabilities rise to about 50%, 20%, and 10% of exceeding 1.0 m, 1.5 m, or 2.0 m of global rise by 2100, respectively. While these low-confidence ice-sheet processes are unlikely to make significant contributions with 2&deg;C of warming, how much warming might be required to trigger them is currently unknown.
In addition, as a result of improved understanding of the timing of possible large future contributions from ice-sheet loss, the Extreme scenario from the 2017 report (2.5 m GMSL rise by 2100) is now viewed as less plausible and has been removed from consideration. Nevertheless, the increased acceleration in the late 21st century and beyond means that the other high-end scenarios provide pathways that potentially reach this threshold in the decades immediately following 2100 (and continue rising). Regionally, the projections are near or higher than the global average in 2100 and 2150 for almost all U.S. coastlines due to vertical land motion (VLM); gravitational, rotational, and deformational effects due to land ice loss; and ocean circulation changes. Largely due to VLM, RSL projections are lower than the global amounts along the southern Alaska coast and are higher along the Eastern and Western Gulf coastlines.
Monitoring the sources of ongoing sea level rise and the processes driving changes in sea level is critical for assessing scenario divergence and tracking the trajectory of observed sea level rise, particularly during the time period when future emissions pathways lead to increased ranges in projected sea level rise.
Efforts are currently under way to narrow the uncertainties in ice-sheet dynamics and future sea level rise amounts in response to increasing greenhouse gas forcing and associated global warming. Early indicators of changes in sea level rise trajectories can serve to trigger adaptive management plans and are identified through continuous monitoring and assessment of changes in sea level (on global and local scales) and of the key drivers of sea level change that most affect U.S. coastlines, such as ocean heat content, ice-mass loss from Greenland and Antarctica, vertical land motion, and Gulf Stream system changes.
As emphasized in the summary findings above, beyond 2050 the amount of sea level rise is strongly af-fected by future global warming. By reducing greenhouse gas (GHG) emissions, severe and transformative Global and Regional Sea Level Rise Scenarios for the United States l 61
 
impacts occurring later this century or early next century along U.S. coastlines are more likely to be avoided.
As GHG emissions and global temperatures continue to rise, the likelihood of very high U.S. sea level rise does too. If global warming reaches 2&deg;C (warming levels are already >1&deg;C), corresponding to a 50% chance that U.S. sea level as a whole will rise at least 0.7 m by 2100 and 1.2 m by 2150 (measured since 2000), major HTF by 2100 would occur more often than minor HTF occurs today in many coastal communities if risk-re-duction action is not taken. If global mean temperatures were to rise as high as about 3&deg;-5&deg;C, much larger amounts of sea level rise would become increasingly possible, as instabilities in ice-sheet dynamics would potentially come into play. Constant monitoring of global to local sea levels and their source contributions by Federal agencies, such as NOAA and NASA, will be key to help assess potential trajectory divergence for triggering adaptive management plans.
The updated sea level scenarios and the EWL probability datasets in this study are being delivered or planned via numerous agency data servers, tools, and associated guidance products. Additionally, this report is a key technical input to the Fifth National Climate Assessment (NCA5 currently under way), and the datasets and derived information are being delivered to the NCA5 author teams. In terms of next steps, the Task Force will continue to refine these sea level projections and extreme (e.g., high tides, storms) water level probabilities while working to improve understanding of the implications of these projections for coast-al hazards (e.g., flooding, erosion, and rising water tables), societal exposure and risk, infrastructure vulnera-bility, ecosystem health (including habitat transformation/loss), and cascading societal impacts. In order to do so, additional and improved observations and more sophisticated modeling approaches that incorporate the relevant physical processes (e.g., waves; see Box 3.1) will be needed at the regional scale, with local granu-larity to assess the impacts of these coastal hazards. Such information is expected to ultimately feed into the next generation of interagency reports and assessments to enable informed climate adaptation planning.
Global and Regional Sea Level Rise Scenarios for the United States l 62
 
Section 6: Acknowledgments The authors appreciate the review and constructive comments from the following external reviewers*: Dr.
Mark Merrifield (Scripps Institution of Oceanography), Dr. Gary Mitchum (University of South Florida), Dr.
Claudia Tebaldi (Lawrence Berkeley National Laboratory), Dr. Thomas Wahl (University of Central Florida),
Dr. Steve Nerem (University of Colorado), Abby Sullivan (Philadelphia Water Department), and David Behar (San Francisco Public Utilities Commission).
We also thank the following agencies and/or their personnel for the reviews provided: Dr. Davina Pas-seri (U.S. Geological Survey [USGS]), Dr. Erika Lentz (USGS), Dr. Rebecca Beavers (National Park Ser-vice), Heidi Stiller (NOAA), Jamie Carter (National Oceanic and Atmospheric Administration [NOAA]),
Lisa Auermuller (Rutgers University), Dr. Renee Collini (Mississippi State University), Laura Engeman (Scripps Institution of Oceanography), Dr. Ian Miller, (University of Washington), Katy Hintzen (University of Hawaii),
Jill Gambill (University of Georgia), Carey Schafer (EcoAdapt), and Rachel Johnson (NOAA). We would like to thank Sean Vitousek (model output) and Amy Foxgrover (figure illustration) of the USGS for Wave Call-out box support.
The contributions of Robert Kopp and Greg Garner were supported by the National Science Foundation (ICER-1663807, ICER-2103754) and the National Aeronautics and Space Administration (award 80NSS-C20K1724 and JPL task 105393.509496.02.08.13.31). Contributions of John Marra, William Sweet, Jayantha Obeysekera, and Ayesha Genz were supported by the U.S. Department of Defense Strategic Environmental Research and Development Program through work carried out under Project RC-2644. The contributions of Benjamin Hamlington, Thomas Frederikse, Eric Larour, and David Bekaert were carried out in part at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration (80NM0018D0004). The contributions of Patrick Barnard were supported by the USGS Coastal and Marine Hazards and Resources Program.T he authors thank NOAAs Ashely Miller for her geo-graphic information support.
Production Team Brooke C. Stewart, Science Editor, NC State University Andrea L. McCarrick, Technical Editor, NC State University Jessicca Allen, Lead Graphic Designer, NC State University S. Elizabeth Love-Brotak, Graphic Designer, NOAA Deborah Misch, Lead Graphic Designer, Innovative Consulting & Management Services, LLC Sara W. Veasey, Visual Communications Team Lead, NOAA Jacquelyn Crossman, Librarian, ZAI-MPF
*The identification of the external reviewers does not imply that they agreed with all of the content of this report.
Global and Regional Sea Level Rise Scenarios for the United States l 63
 
Section 7: References Bamber, J.L. and W.P. Aspinall, 2013: An expert judgement assessment of future sea level rise from the ice sheets. Nature Climate Change, 3 (4), 424-427. https://doi.org/10.1038/nclimate1778 Bamber, J.L., M. Oppenheimer, R.E. Kopp, W.P. Aspinall, and R.M. Cooke, 2019: Ice sheet contributions to fu-ture sea-level rise from structured expert judgment. Proceedings of the National Academy of Sciences of the United States of America, 116 (23), 11195-11200. https://doi.org/10.1073/pnas.1817205116 Baranes, H.E., J.D. Woodruff, S.A. Talke, R.E. Kopp, R.D. Ray, and R.M. DeConto, 2020: Tidally driven interan-nual variation in extreme sea level frequencies in the Gulf of Maine. Journal of Geophysical Research:
Oceans, 125 (10), e2020JC016291.
http://dx.doi.org/10.1029/2020JC016291 Barnard, P.L., L.H. Erikson, A.C. Foxgrover, J.A. Finzi Hart, P. Limber, A.C. ONeill, M. van Ormondt, S. Vitousek, N. Wood, M.K. Hayden, and J.M. Jones, 2019: Dynamic flood modeling essential to assess the coastal impacts of climate change. Scientific reports, 9 (1), 1-13. https://doi.org/10.1038/s41598-019-40742-z Befus, K.M., P.L. Barnard, D.J. Hoover, J.A. Finzi Hart, and C.I. Voss, 2020: Increasing threat of coastal ground-water hazards from sea-level rise in California. Nature Climate Change, 10 (10), 946-952.
https://doi.org/10.1038/s41558-020-0874-1 Bekaert, D.P.S., B.D. Hamlington, B. Buzzanga, and C.E. Jones, 2017: Spaceborne synthetic aperture radar survey of subsidence in Hampton Roads, Virginia (USA). Scientific Reports, 7 (1), 1-9.
https://doi.org/10.1038/s41598-017-15309-5 Bekaert, D.P.S., C.E. Jones, K. An, and M.-H. Huang, 2019: Exploiting UAVSAR for a comprehensive analysis of subsidence in the Sacramento Delta. Remote Sensing of Environment, 220, 124-134.
https://doi.org/10.1016/j.rse.2018.10.023 Bloemen, P., T. Reeder, C. Zevenbergen, J. Rijke, and A. Kingsborough, 2018: Lessons learned from apply-ing adaptation pathways in flood risk management and challenges for the further development of this approach. Mitigation and Adaptation Strategies for Global Change, 23 (7), 1083-1108.
https://doi.org/10.1007/s11027-017-9773-9 Bodnar, A.N., 1981: Estimating Accuracies of Tidal Datums from Short Term Observations. Technical Report NOS CO-OPS 077. National Oceanic and Atmospheric Administration, National Ocean Service, Center for Operational Oceanographic Products and Services, Silver Spring, MD, 40 pp.
https://tidesandcurrents.noaa.gov/publications/NOAA_Technical_Report_NOS_COOPS_077.pdf Boening, C., J.K. Willis, F.W. Landerer, R.S. Nerem, and J. Fasullo, 2012: The 2011 La Nina: So strong, the oceans fell. Geophysical Research Letters, 39 (19). https://doi.org/10.1029/2012GL053055 Bos, M.S., R.M.S. Fernandes, S.D.P. Williams, and L. Bastos, 2013: Fast error analysis of continuous GNSS observations with missing data. Journal of Geodesy, 87 (4), 351-360.
http://dx.doi.org/10.1007/s00190-012-0605-0 Bos, M.S., S.D.P. Williams, I.B. Ara&#xfa;jo, and L. Bastos, 2014: The effect of temporal correlated noise on the sea level rate and acceleration uncertainty. Geophysical Journal International, 196 (3), 1423-1430.
http://dx.doi.org/10.1093/gji/ggt481 Bromirski, P.D., A.J. Miller, R.E. Flick, and G. Auad, 2011: Dynamical suppression of sea level rise along the Pa-cific coast of North America: Indications for imminent acceleration. Journal of Geophysical Research, 116, C07005. https://doi.org/10.1029/2010jc006759 Global and Regional Sea Level Rise Scenarios for the United States l 64
 
Buchanan, M.K., M. Oppenheimer, and R.E. Kopp, 2017: Amplification of flood frequencies with local sea level rise and emerging flood regimes. Environmental Research Letters, 12 (6), 064009.
https://doi.org/10.1088/1748-9326/aa6cb3 Buzzanga, B., D.P.S. Bekaert, B.D. Hamlington, and S.S. Sangha, 2020: Toward sustained monitoring of sub-sidence at the coast using InSAR and GPS: An application in Hampton Roads, Virginia. Geophysical Research Letters, 47 (18), e2020GL090013. https://doi.org/10.1029/2020GL090013 Calafat, F.M., D.P. Chambers, and M.N. Tsimplis, 2012: Mechanisms of decadal sea level variability in the east-ern North Atlantic and the Mediterranean Sea. Journal of Geophysical Research: Oceans, 117, C09022.
https://doi.org/10.1029/2012JC008285 Calafat, F.M. and M. Marcos, 2020: Probabilistic reanalysis of storm surge extremes in Europe. Proceedings of the National Academy of Sciences, 117 (4), 1877-1883.
http://dx.doi.org/10.1073/pnas.1913049117 CCRM, 2019: Recommendations for Sea Level Rise Projections: A Report for the Governors Coastal Cli-mate Resiliency Plan. Center for Coastal Resources Management, Virginia institute of Marine Science, Gloucester Point, VA, 6 pp.
https://www.naturalresources.virginia.gov/media/governorvirginiagov/secretary-of-natural-resources/
images/1c.-Sea-level-rise-projections-for-Virginia-planning-purposes-(2)-FINAL-10_31.pdf Chao, B.F., Y.-H. Wu, and Y.S. Li, 2008: Impact of artificial reservoir water impoundment on global sea level.
Science, 320 (5873), 212-214. https://doi.org/10.1126/science.1154580 Church, J.A., P.U. Clark, A. Cazenave, J.M. Gregory, S. Jevrejeva, A. Levermann, M.A. Merrifield, G.A. Milne, R.S. Nerem, P.D. Nunn, A.J. Payne, W.T. Pfeffer, D. Stammer and A.S. Unnikrishnan, 2013: Sea Level Change. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cam-bridge University Press, Cambridge, United Kingdom and New York, NY, USA https://www.ipcc.ch/site/assets/uploads/2018/02/WG1AR5_Chapter13_FINAL.pdf Coles, S., 2001: An Introduction to Statistical Modeling of Extreme Values. Springer-Verlag, London, 224 pp.
Coles, S.G., and M.J. Dixon, 1999: Likelihood-based inference for extreme value models. Extremes, 2, 5-23.
https://doi.org/10.1023/A:1009905222644 Cooley, D., 2013: Return periods and return levels under climate change. In: Extremes in a Changing Climate.
AghaKouchak, A., D. Easterling, K. Hsu, S. Schubert, and S. Sorooshian, Eds. Springer, Dordrecht, Neth-erlands, 97-114.
Dalrymple, T. (1960). Flood Frequency Analyses, Manual of Hydrology: Part 3. U.S. Geological Survey Water-Supply Paper 1543-A. U.S. Geological Survey, Reston, VA, 86 pp. https://doi.org/10.3133/ws-p1543A Dangendorf, S., C. Hay, F.M. Calafat, M. Marcos, C.G. Piecuch, K. Berk, and J. Jensen, 2019: Persistent accel-eration in global sea-level rise since the 1960s. Nature Climate Change, 9 (9), 705-710.
https://doi.org/10.1038/s41558-019-0531-8 DeConto, R.M., D. Pollard, R.B. Alley, I. Velicogna, E. Gasson, N. Gomez, S. Sadai, A. Condron, D.M. Gilford, E.L. Ashe, R.E. Kopp, D. Li, and A. Dutton, 2021: The Paris Climate Agreement and future sea-level rise from Antarctica. Nature, 593 (7857), 83-89. https://doi.org/10.1038/s41586-021-03427-0 Dokka, R.K., 2011: The role of deep processes in late 20th century subsidence of New Orleans and coastal areas of southern Louisiana and Mississippi. Journal of Geophysical Research, 116, B06403.
https://doi.org/10.1029/2010JB008008 Global and Regional Sea Level Rise Scenarios for the United States l 65
 
Edwards, T.L., S. Nowicki, B. Marzeion, R. Hock, H. Goelzer, H. Seroussi, N.C. Jourdain,D.A. Slater, F.E. Turner, C.J. Smith, C.M. McKenna, E. Simon, A. Abe-Ouchi, J.M. Gregory, E. Larour, W.H. Lip-scomb, A.J. Payne, A. Shepherd, C. Agosta, P. Alexander, T. Albrecht, B. Anderson, X. Asay-Davis, A. Aschwanden, A. Barthel, A. Bliss, R. Calov, C. Chambers, N. Champollion, Y. Choi, R. Cullath-er, J. Cuzzone, C. Dumas, D. Felikson, X. Fettweis, K. Fujita, B.K. Galton-Fenzi, R. Gladstone, N.R.
Golledge, R. Greve, T. Hattermann, M.J. Hoffman, A. Humbert, M. Huss, P. Huybrechts, W. Immer-zeel, T. Kleiner, P. Kraaijenbrink, S. Le clech, V. Lee, G.R. Leguy, C.M. Little, D.P. Lowry, J.-H. Malles, D.F. Martin, F. Maussion, M. Morlighem, J.F. ONeill, I. Nias, F. Pattyn, T. Pelle, S.F. Price, A. Quiquet, V. Radi, R. Reese, D.R. Rounce, M. R&#xfc;ckamp, A. Sakai, C. Shafer, N.-J. Schlegel, S. Shannon, R.S.
Smith, F. Straneo, S. Sun, L. Tarasov, L.D. Trusel, J. Van Breedam, R. van de Wal, M. van den Broeke, R. Winkelmann, H. Zekollari, C. Zhao, T. Zhang, and T. Zwinger, 2021: Projected land ice contributions to twenty-first-century sea level rise. Nature, 593 (7857), 74-82. https://doi.org/10.1038/s41586-021-03302-y Eggleston, J. and J. Pope, 2013: Land Subsidence and Relative Sea-Level Rise in the Southern Chesapeake Bay Region. Circular 1392. U.S. Geological Survey, Reston, VA, 30 pp., http://dx.doi.org/10.3133/cir1392 EPA, 1995a: Combined Sewer Overflows: Guidance for Long-Term Control Plan. EPA/832-B-95-002. U.S. En-vironmental Protection Agency, Office of Wastewater Management, Washington, DC, 201 pp.
https://www.epa.gov/sites/default/files/2015-10/documents/owm0272_0.pdf EPA, 1995b: Combined Sewer Overflows: Guidance for Nine Minimum Controls. EPA 832-B-95-003. U.S. En-vironmental Protection Agency, Office of Wastewater Management, Washington, DC, 69 pp.
https://www.epa.gov/sites/default/files/2015-10/documents/owm0030_2.pdf Farrell, W.E. and J.A. Clark, 1976: On postglacial sea level. Geophysical Journal International, 46 (3),
647-667. https://doi.org/10.1111/j.1365-246X.1976.tb01252.x.
Fasullo, J. T., Boening, C., Landerer, F. W., & Nerem, R. S. (2013). Australias unique influence on global sea level in 2010-2011. Geophysical Research Letters, 40 (16), 4368-4373.
https://doi.org/10.1002/grl.50834 Fasullo, J.T. and R.S. Nerem, 2018: Altimeter-era emergence of the patterns of forced sea-level rise in climate models and implications for the future. Proceedings of the National Academy of Sciences, 115 (51),
12944-12949. https://doi.org/10.1073/pnas.1813233115 FEMA, 2016a: Guidance for Flood Risk Analysis Mapping: Coastal Flood Frequency and Extreme Value Anal-ysis. Guidance Document 76. Federal Emergency Management Agency, Hyattsville, MD, 22 pp.
https://www.fema.gov/sites/default/files/2020-02/Coastal_Flood_Frequency_and_Extreme_Value_
Analysis_Guidance_Nov_2016.pdf FEMA, 2016b: Guidance for Flood Risk Analysis Mapping: Statistical Simulation Methods. Guidance Docu-ment 77. Federal Emergency Management Agency, Hyattsville, MD, 37 pp. https://www.fema.gov/sites/
default/files/documents/fema_coastal-statistical-simulation-methods_nov-2016.pdf Fox-Kemper, B., H.T. Hewitt, C. Xiao, G. Aalgeirsd&#xf3;ttir, S.S. Drijfhout, T.L. Edwards, N.R. Golledge, M. Hem-er, R.E. Kopp, G. Krinner, A. Mix, D. Notz, S. Nowicki, I.S. Nurhati, L. Ruiz, J.-B. Sall&#xe9;e, A.B.A. Slangen, and Y. Yu, 2021: Ocean, Cryosphere and Sea Level Change. In: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovern-mental Panel on Climate Change [MassonDelmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. P&#xe9;an, S.
Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelek&#xe7;i, R. Yu, and B. Zhou (eds.)]. Cambridge University Press. In Press.
https://www.ipcc.ch/report/ar6/wg1/
Frau, R., M. Andreewsky, and P. Bernardara, 2018: The use of historical information for regional frequency analysis of extreme skew surge. Natural Hazards and Earth System Sciences, 18 (3), 949-962.
https://doi.org/10.5194/nhess-18-949-2018 Global and Regional Sea Level Rise Scenarios for the United States l 66
 
Frederikse, T., K. Simon, C.A. Katsman, and R. Riva, 2017: The sea-level budget along the Northwest Atlantic coast: GIA, mass changes, and large-scale ocean dynamics. Journal of Geophysical Research: Oceans, 122, 5486-5501. https://doi.org/10.1002/2017JC012699 Frederikse, T., F. Landerer, L. Caron, S. Adhikari, D. Parkes, V.W. Humphrey, S. Dangendorf, P. Hogarth, L. Zanna, L. Cheng, and Y.-H. Wu, 2020: The causes of sea-level rise since 1900. Nature, 584 (7821),
393-397. https://doi.org/10.1038/s41586-020-2591-3 Fu, X., 2020: Measuring local sea-level rise adaptation and adaptive capacity: A national survey in the United States. Cities, 102, 102717. https://doi.org/10.1016/j.cities.2020.102717.
Garner, G.G., T. Hermans, R.E. Kopp, A.B.A. Slangen, T.L. Edwards, A. Levermann, S. Nowikci, M.D. Palmer, C. Smith, B. Fox-Kemper, H.T. Hewitt, C. Xiao, G. Aalgeirsd&#xf3;ttir, S.S. Drijfhout, T.L. Edwards, N.R. Golledge, M. Hemer, R.E. Kopp, G. Krinner, A. Mix, D. Notz, S. Nowicki, I. S. Nurhati, L. Ruiz, J-B. Sall&#xe9;e, Y. Yu, L. Hua, T. Palmer, and B. Pearson, 2021. IPCC AR6 Sea-Level Rise Projections. Ver-sion 20210809. PO.DAAC, CA, USA. Dataset accessed [2021-09-01].
https://podaac.jpl.nasa.gov/announcements/2021-08-09-Sea-level-projections-from-the-IPCC-6th Gill, S., G. Hovis, K. Kriner, and M. Michalski, 2014: Implementation of Procedures for Computation of Tidal Datums in Areas with Anomalous Trends of Relative Mean Sea Level. NOAA Technical Report NOS CO-OPS 068. National Oceanic and Atmospheric Administration, National Ocean Service, Silver Spring, MD, 58 pp.
https://tidesandcurrents.noaa.gov/publications/NOAA_Technical_Report_NOS_COOPS_68.pdf Guza, R.T. and E.B. Thornton, 1982: Swash oscillations on a natural beach. Journal of Geophysical Research, 87 (C1), 483-491. https://doi.org/10.1029/JC087iC01p00483 Haasnoot, M., J.H. Kwakkel, W.E. Walker, and J. ter Maat, 2013: Dynamic adaptive policy pathways: A meth-od for crafting robust decisions for a deeply uncertain world. Global Environmental Change, 23 (2),
485-498. https://doi.org/10.1016/j.gloenvcha.2012.12.006 Haasnoot, M., S. Brown, P. Scussolini, J.A. Jimenez, A.T. Vafeidis, and R.J. Nicholls, 2019: Generic adap-tation pathways for coastal archetypes under uncertain sea-level rise. Environmental Research Communications, 1 (7), 071006. https://doi.org/10.1088/2515-7620/ab1871 Habel, S., C.H. Fletcher, T.R. Anderson, and P.R. Thompson, 2020: Sea-level rise induced multi-mechanism flooding and contribution to urban infrastructure failure. Scientific Reports, 10 (1), 3796.
http://dx.doi.org/10.1038/s41598-020-60762-4 Hall, J.A., S. Gill, J. Obeysekera, W. Sweet, K. Knuuti, and J. Marburger, 2016: Regional Sea Level Scenarios for Coastal Risk Management: Managing the Uncertainty of Future Sea Level Change and Extreme Wa-ter Levels for Department of Defense Coastal Sites Worldwide. U.S. Department of Defense, Strategic Environmental Research and Development Program, Alexandria, VA, 224 pp.
https://climateandsecurity.files.wordpress.com/2014/01/regional-sea-level-scenarios-for-coast-al-risk-management_managing-uncertainty-of-future-sea-level-change-and-extreme-water-levels-for-department-of-defense.pdf Hall, J.W., H. Harvey, and L.J. Manning, 2019: Adaptation thresholds and pathways for tidal flood risk manage-ment in London. Climate Risk Management, 24, 42-58. https://doi.org/10.1016/j.crm.2019.04.001 Hall, J.A., C.P. Weaver, J. Obeysekera, M. Crowell, R.M. Horton, R.E. Kopp, J. Marburger, D.C. Marcy, A. Parris, W.V. Sweet, W.C. Veatch, and K.D. White, 2019: Rising sea levels: Helping decision-makers confront the inevitable. Coastal Management, 47 (2), 127-150. http://dx.doi.org/10.1080/08920753.2019.1551012 Global and Regional Sea Level Rise Scenarios for the United States l 67
 
Hamlington, B.D., A.S. Gardner, E. Ivins, J.T.M. Lenaerts, J.T. Reager, D.S. Trossman, E.D. Zaron, S. Adhikari, A. Arendt, A. Aschwanden, B.D. Beckley, D.P.S. Bekaert, G. Blewitt, L. Caron, D.P. Chambers, H.A. Chandanpurkar, K. Christianson, B. Csatho, R.I. Cullather, R.M. DeConto, J.T. Fasullo, T. Frederikse, J.T. Freymueller, D.M. Gilford, M. Girotto, W.C. Hammond, R. Hock, N. Holschuh, R.E. Kopp, F. Landerer, E. Larour, D. Menemenlis, M. Merrifield, J.X. Mitrovica, R.S. Nerem, I.J. Nias, V. Nieves, S. Nowicki, K. Pangaluru, C.G. Piecuch, R.D. Ray, D.R. Rounce, N.-J. Schlegel, H. Seroussi, M. Shirzaei, W.V. Sweet, I. Velicogna, N. Vinogradova, T. Wahl, D.N. Wiese, and M.J. Willis, 2020a: Understanding of contem-porary regional sea-level change and the implications for the future. Reviews of Geophysics, 58 (3),
e2019RG000672. https://doi.org/10.1029/2019RG000672 Hamlington, B.D., C.G. Piecuch, J.T. Reager, H. Chandanpurkar, T. Frederikse, R.S. Nerem, J.T. Fasullo, and S.-H. Cheon, 2020b: Origin of interannual variability in global mean sea level. Proceedings of the National Academy of Sciences, 117 (25), 13983. http://dx.doi.org/10.1073/pnas.1922190117 Hamlington, B.D., T. Frederikse, P.R. Thompson, J.K. Willis, R.S. Nerem, and J.T. Fasullo, 2021: Past, present, and future Pacific sea-level change. Earths Future, 9 (4), e2020EF001839.
https://doi.org/10.1029/2020EF001839 Hammond, W.C., G. Blewitt, C. Kreemer, and R.S. Nerem, 2021: GPS imaging of global vertical land motion for studies of sea level rise. Journal of Geophysical Research: Solid Earth, 126 (7), e2021JB022355.
https://doi.org/10.1029/2021JB022355 Hawley, W.B., C.C. Hay, J.X. Mitrovica, and R.E. Kopp., 2020: A spatially variable time series of sea level change due to artificial water impoundment. Earths Future, 8 (7), e2020EF001497.
https://doi.org/10.1029/2020EF001497 Hemer, Mark A., Y. Fan, N. Mori, A. Semedo, and X.L. Wang, 2013: Projected changes in wave climate from a multi-model ensemble. Nature Climate Change, 3 (5), 471-476. https://doi.org/10.1038/nclimate1791 Hinkel, J., J.C.J.H. Aerts, S. Brown, J.A. Jim&#xe9;nez, D. Lincke, R.J. Nicholls, P. Scussolini, A. Sanchez-Arcilla, A. Vafeidis, and K.A. Addo, 2018: The ability of societies to adapt to twenty-first-century sea-level rise.
Nature Climate Change, 8 (7), 570-578. https://doi.org/10.1038/s41558-018-0176-z Hosking, J.R.M. and J. R. Wallis, 1997: Regional Frequency Analysis: An Approach Based on L-Moments.
Cambridge University Press, Cambridge, U.K, 224 pp.
IPCC, 2021a: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. P&#xe9;an, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelek&#xe7;i, R. Yu, and B.
Zhou (eds.)]. Cambridge University Press. In Press. https://www.ipcc.ch/report/ar6/wg1/
IPCC, 2021b: Summary for Policymakers. In: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change
[MassonDelmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. P&#xe9;an, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelek&#xe7;i, R. Yu, and B. Zhou (eds.)]. Cambridge University Press. In Press. https://www.ipcc.ch/report/ar6/wg1/
Karegar, M.A., T.H. Dixon, and S.E. Engelhart, 2016: Subsidence along the Atlantic Coast of North America:
Insights from GPS and late Holocene relative sea level data. Geophysical Research Letters, 43, 3126-3133. https://doi.org/10.1002/2016GL068015 King, D., D. Schrag, Z. Dadi, Q. Ye, and A. Gosh, 2015: Climate Change: A Risk Assessment. Hynard, J. and T. Rodger, Eds. Centre for Science and Policy, Cambridge, UK, 154 pp.
www.csap.cam.ac.uk/projects/climate-change-risk-assessment/
Global and Regional Sea Level Rise Scenarios for the United States l 68
 
Kirwan, M.L. and K.B. Gedan, 2019: Sea-level driven land conversion and the formation of ghost forests.
Nature Climate Change, 9 (6), 450-457. https://doi.org/10.1038/s41558-019-0488-7 Konikow, L.F., 2011: Contribution of global groundwater depletion since 1900 to sea-level rise. Geophysical Research Letters, 38 (17). https://doi.org/10.1029/2011GL048604 Kopp, R.E., R.M. Horton, C.M. Little, J.X. Mitrovica, M. Oppenheimer, D.J. Rasmussen, B.H. Strauss, and C. Tebaldi, 2014: Probabilistic 21st and 22nd century sea-level projections at a global network of tide-gauge sites. Earths Future, 2 (8), 383-406. https://doi.org/10.1002/2014EF000239 Kopp, R.E., R.M. DeConto, D.A. Bader, C.C. Hay, R.M. Horton, S. Kulp, M. Oppenheimer, D. Pollard, and B.H. Strauss, 2017: Evolving understanding of Antarctic ice-sheet physics and ambiguity in probabilistic sea-level projections. Earths Future, 5 (12), 1217-1233. https://doi.org/10.1002/2017EF000663 Kopp, R.E., E.A. Gilmore, C.M. Little, J. Lorenzo-Trueba, V.C. Ramenzoni, and W.V. Sweet, 2019: Usable science for managing the risks of sea-level rise. Earths Future, 7, 1235-1269.
https://doi.org/10.1029/2018EF001145 Krasting, J.P., J.P. Dunne, R.J. Stouffer, and R.W. Hallberg, 2016: Enhanced Atlantic sea-level rise relative to the Pacific under high carbon emission rates. Nature Geoscience, 9 (3), 210-214.
http://dx.doi.org/10.1038/ngeo2641 Langbein W.B., 1949: Annual floods and the partial-duration flood series. Eos, Transactions American Geophysical Union, 30 (6), 879-881. https://doi.org/10.1029/TR030i006p00879 Larsen, C.F., R.J. Motyka, J.T. Freymueller, K.A. Echelmeyer, and E.R. Ivins, 2004: Rapid uplift of southern Alaska caused by recent ice loss. Geophysical Journal International, 158 (3), 1118-1133.
http://dx.doi.org/10.1111/j.1365-246X.2004.02356.
Levermann, A., R. Winkelmann, T. Albrecht, H. Goelzer, N.R. Golledge, R. Greve, P. Huybrechts J. Jordan, G. Leguy, D. Martin, M. Morlighem, F. Pattyn, D. Pollard, A. Quiquet, C. Rodehacke, H. Seroussi, J. Sutter, T. Zhang, J. Van Breedam, R. Calov, R. DeConto, C. Dumas, J. Garbe, G.H. Gudmunds-son, M.J. Hoffman, A. Humbert, T. Kleiner, W.H. Lipscomb, M. Meinshausen, E. Ng, S.M.J. Nowicki, M. Perego, S.F. Price, F. Saito, N.J. Schlegel, S. Sun, and R.S.W. van de Wal, 2020: Projecting Antarcti-cas contribution to future sea level rise from basal ice shelf melt using linear response functions of 16 ice sheet models (LARMIP-2). Earth System Dynamics, 11 (1), 35-76. https://doi.org/10.5194/esd-11      2020 Longuet-Higgins, M.S. and R.W. Stewart, 1963: A note on wave set-up. Journal of Marine Research, 21 (4),
4-10. https://images.peabody.yale.edu/publications/jmr/jmr21-01-01.pdf Lyu, K., X. Zhang, J.A. Church, A.B.A. Slangen, and J. Hu. 2014: Time of emergence for regional sea-level change. Nature Climate Change, 4 (11), 1006-1010. https://doi.org/10.1038/nclimate2397 Marzeion, B., A.H. Jarosch, and M. Hofer, 2012: Past and future sea-level change from the surface mass bal-ance of glaciers. The Cryosphere, 6 (6), 1295-1322. https://doi.org/10.5194/tc-6-1295-2012 Marzeion, B., R. Hock, B. Anderson, A. Bliss, N. Champollion, K. Fujita, M. Huss, W.W. Immerzeel, P. Kraaijenbrink, J.-H. Malles, F. Maussion, V. Radi, D.R. Rounce, A. Sakai, S. Shannon, R. van de Wal, and H. Zekollari, 2020: Partitioning the uncertainty of ensemble projections of global glacier mass change. Earths Future, 8 (7), e2019EF001470. https://doi.org/10.1029/2019EF001470 McCallum, B.E., S.M. Wicklein, R.G. Reiser, R. Busciolano, J. Morrison, R.J. Verdi, J.A. Painter, E.R. Frantz, and A.J. Gotvald, 2013: Monitoring Storm Tide and Flooding from Hurricane Sandy along the Atlantic Coast of the United States, October 2012. Open File Report 2013-1043. U.S. Geological Survey, Reston, VA, 42 pp. http://dx.doi.org/10.3133/ofr20131043 Global and Regional Sea Level Rise Scenarios for the United States l 69
 
McKenzie, T., S. Habel, and H. Dulai, 2021: Sea-level rise drives wastewater leakage to coastal waters and storm drains. Limnology and Oceanography Letters, 6, 154-163. https://doi.org/10.1002/lol2.10186 Men&#xe9;ndez, M., and P.L. Woodworth, 2010: Changes in extreme high water levels based on a quasi-global tide-gauge data set. Journal of Geophysical Research: Oceans, 115, C10011.
https://doi.org/10.1029/2009JC005997 Merrifield, M.A., A.S. Genz, C.P. Kontoes, and J.J. Marra, 2013: Annual maximum water levels from tide gaug-es: Contributing factors and geographic patterns. Journal of Geophysical Research: Oceans, 118 (5),
2535-2546. https://doi.org/10.1002/jgrc.20173 Milne, G.A. and J.X. Mitrovica, 1998: Postglacial sea-level change on a rotating Earth. Geophysical Journal International, 133 (1), 1-19. https://doi.org/10.1046/j.1365-246X.1998.1331455.x Mitrovica, J.X., M.E. Tamisiea, J.L. Davis, and G.A. Milne, 2001: Recent mass balance of polar ice sheets in-ferred from patterns of global sea-level change. Nature, 409 (6823), 1026-1029.
https://doi.org/10.1038/35059054 Moftakhari, H. R., Jay, D. A., & Talke, S. A. (2016). Estimating river discharge using multiple-tide gauges distrib-uted along a channel. Journal of Geophysical Research: Oceans, 121 (4), 2078-2097.
https://doi.org/10.1002/2015JC010983 Moore, F.C. and N. Obradovich, 2020: Using remarkability to define coastal flooding thresholds. Nature Communications, 11 (1), 1-8. https://doi.org/10.1038/s41467-019-13935-3 Muis, S., M. Verlaan, H.C. Winsemius, J.C.J.H. Aerts, and P.J. Ward, 2016: A global reanalysis of storm surges and extreme sea levels. Nature Communications, 7 (1), 11969.
http://dx.doi.org/10.1038/ncomms11969 Nadal-Caraballo, N.C., M.O. Campbell, V.M. Gonzalez, M.J. Torres, J.A. Melby, and A.A. Taflanidis, 2020:
Coastal Hazards System: A probabilistic coastal hazard analysis framework. Journal of Coastal Research, 95 (SI), 1211-1216. http://dx.doi.org/10.2112/SI95-235.1 NGS, 2001: Geodetic Program Needs of Louisiana and Wisconsin: Report to Congress. National Oceanic and Atmospheric Administration, National Ocean Service, National Geodetic Survey, Silver Spring, MD, 38 pp. https://www.ngs.noaa.gov/PUBS_LIB/NGSreport_823.pdf NOAA, 2003: Computational Techniques for Tidal Datums Handbook. NOAA Special Publication NOS CO-OPS 2. National Oceanic and Atmospheric Administration, National Ocean Service, Center for Opera-tional Oceanographic Products and Services, Silver Spring, MD, 113 pp. https://tidesandcurrents.noaa.
gov/publications/Computational_Techniques_for_Tidal_Datums_handbook.pdf NOAA, 2020: Surf Zone Forecast and Coastal/Lakeshore Hazard Services. National Weather Service Instruc-tion 10-320. National Oceanic and Atmospheric Administration, National Weather Service, 38 pp.
http://www.nws.noaa.gov/directives/sym/pd01003020curr.pdf NOAA NOS and U.S. Census Bureau, 2013: National Coastal Population Report: Population Trends from 1970 to 2020. National Oceanic and Atmospheric Administration, National Ocean Service, and U.S. Census Bureau, Silver Spring, MD, 22 pp. https://aambpublicoceanservice.blob.core.windows.net/oceanser-viceprod/facts/coastal-population-report.pdf Obeysekera, J. and J. Salas, 2020: Hydrologic designs for extreme events under nonstationarity. In:
Engineering Methods for Precipitation under a Changing Climate. Olsen, J.R. and K.T. Adamec, Eds.
American Society of Civil Engineers, 63-82. http://dx.doi.org/10.1061/9780784415528.ch04 One Climate Future, 2019: One Climate Future Climate Change Vulnerability Assessment. One Climate Future, Cities of Portland and South Portland, 181 pp. https://www.oneclimatefuture.org/wp-content/up-loads/2020/12/OneClimateFuture_VulnerabilityAssessment_Final.pdf Global and Regional Sea Level Rise Scenarios for the United States l 70
 
Oppenheimer, M., B.C. Glavovic , J. Hinkel, R. van de Wal, A.K. Magnan, A. Abd-Elgawad, R.
Cai, M. Cifuentes-Jara, R.M. DeConto, T. Ghosh, J. Hay, F. Isla, B. Marzeion, B. Meyssignac, and Z.
Sebesvari, 2019: Sea Level Rise and Implications for Low-Lying Islands, Coasts and Communities. In:
IPCC Special Report on the Ocean and Cryosphere in a Changing Climate [H.-O. Prtner, D.C. Rob-erts, V. Masson-Delmotte, P. Zhai, M. Tignor, E. Poloczanska, K. Mintenbeck, A. Alegr&#xed;a, M. Nicolai, A. Okem, J. Petzold, B. Rama, N.M. Weyer (eds.)]. In Press. https://www.ipcc.ch/site/assets/uploads/
sites/3/2019/11/08_SROCC_Ch04_FINAL.pdf Parris, A., P. Bromirski, V. Burkett, D. Cayan, M. Culver, J. Hall, R. Horton, K. Knuuti, R. Moss, J. Obeysekera, A. Sallenger, and J. Weiss, 2012: Global Sea Level Rise Scenarios for the US National Climate Assess-ment. NOAA Tech Report OAR CPO-1. National Oceanic and Atmospheric Administration, Climate Program Office, 37 pp. https://repository.library.noaa.gov/view/noaa/11124 Perica, S., S. Pavlovic, M. St. Laurent, C. Trypaluk, D. Unruh, and O. Wilhite, 2018: Precipitation-Frequency At-las of the United States. Volume 11, Version 2.0 Texas. National Oceanic and Atmospheric Administra-tion, National Weather Service, Silver Spring, MD. https://repository.library.noaa.gov/view/noaa/22619 Piecuch, C. G. and, & Quinn, K. J. 2016: El Nino, La Nina, and the global sea level budget. Ocean Science, 12 (6), 1165-1177. https://doi.org/10.5194/os-12-1165-2016 Reguero, B.G., M. Men&#xe9;ndez, F.J. M&#xe9;ndez, R. M&#xed;nguez, and I.J. Losada. 2012: A Global Ocean Wave (GOW) calibrated reanalysis from 1948 onwards. Coastal Engineering. 65, 38-55.
https://doi.org/10.1016/j.coastaleng.2012.03.003 Richter, K., B. Meyssignac, A.B.A. Slangen, A. Melet, J.A. Church, X. Fettweis, B. Marzeion, C. Agosta, S.R.M.
Ligtenberg, G. Spada, M.D. Palmer, C.D. Roberts, and N. Champollion, 2020: Detecting a forced signal in satellite-era sea-level change. Environmental Research Letters, 15 (9), 094079.
https://doi.org/10.1088/1748-9326/ab986e Rydlund, P. H. Jr., B. J. Densmore (2012). Methods of practice and guidelines for using survey-grade Global Navigation Satellite Systems (GNSS) to establish vertical datum in the United States Geological Survey.
Chapter 1 of Section D, Field Survey Methods, Book 11, Collection and Delineation of Spatial Data. U.S.
Geological Survey, Reston, Virginia, 102 pp. with appendixes. http://pubs.usgs.gov/tm/11d1/
Salas, J.D., and J. Obeysekera, 2014: Revisiting the concepts of return period and risk for nonstationary hy-drologic extreme events. Journal of Hydrologic Engineering, 19 (3) 554-568.
https://doi.org/10.1061/(ASCE)HE.1943-5584.0000820 Salas, J.D., J. Obeysekera, and R. Vogel, 2018: Techniques for assessing water infrastructure for nonstation-ary extreme events: A review. Hydrologic Science Journal, 63 (3), 325-352.
https://doi.org/10.1080/02626667.2018.1426858 Salas, J.D. and J. Obeysekera, 2019: Probability distribution and risk of the first occurrence of k extreme hy-drologic events. Journal of Hydrologic Engineering, 24 (10), 04019032.
https://doi.org/10.1061/(ASCE)HE.1943-5584.0001809 Slangen, A.B.A., M. Carson, C.A. Katsman, R.S.W. van de Wal, A. Khl, L.L.A. Vermeersen, and D. Stammer, 2014: Projecting twenty-first century regional sea-level changes. Climatic Change, 124 (1), 317-332.
http://dx.doi.org/10.1007/s10584-014-1080-9 Stockdon H.F., R.A. Holman, P.A. Howd, and A.H. Sallenger, Jr., 2006: Empirical parameterization of setup, swash, and runup. Coastal Engineering, 53 (7), 573-588.
https://doi.org/10.1016/j.coastaleng.2005.12.005 Sutton, R.T., 2019: Climate science needs to take risk assessment much more seriously. Bulletin of the American Meteorological Society, 100 (9), 1637-1642. https://doi.org/10.1175/BAMS-D-18-0280.1 Global and Regional Sea Level Rise Scenarios for the United States l 71
 
Sweeney, B. and A. Becker, 2020: Considering future sea level change in maritime infrastructure design:
A survey of US Engineers. Journal of Waterway, Port, Coastal, and Ocean Engineering, 146 (4),
04020019. http://dx.doi.org/10.1061/(ASCE)WW.1943-5460.0000583 Sweet, W.V. and J. Park, 2014: From the extreme to the mean: Acceleration and tipping points of coastal inun-dation from sea level rise. Earths Future, 2 (12), 579-600. https://doi.org/10.1002/2014EF000272 Sweet, W.V., J. Park, S. Gill, and J. Marra, 2015: New ways to measure waves and their effects at NOAA tide gauges: A Hawaiian-network perspective. Geophysical Research Letters, 42 (21), 9355-9361.
https://doi.org/10.1002/2015GL066030 Sweet, W.V., R.E. Kopp, C.P. Weaver, J. Obeysekera, R.M. Horton, E.R. Thieler, and C. Zervas, 2017: Global and Regional Sea Level Rise Scenarios for the United States. NOAA Technical Report NOS CO-OPS 083.
National Oceanic and Atmospheric Administration, National Ocean Service, Center for Operational Oceanographic Products and Services, Silver Spring, MD, 75 pp. https://tidesandcurrents.noaa.gov/
publications/techrpt83_Global_and_Regional_SLR_Scenarios_for_the_US_final.pdf Sweet, W.V., G. Dusek, J. Obeysekera, and J.J. Marra, 2018: Patterns and Projections of High Tide Flooding Along the U.S. Coastline Using a Common Impact Threshold. NOAA Technical Report NOS CO-OPS 086. National Oceanic and Atmospheric Administration, National Ocean Service, Silver Spring, MD, 44 pp. https://tidesandcurrents.noaa.gov/publications/techrpt86_PaP_of_HTFlooding.pdf Sweet, W.V., G. Dusek, G. Carbin, J.J. Marra, D. Marcy, and S. Simon, 2020a: 2019 State of U.S. High Tide Flooding with a 2020 Outlook. NOAA Technical Report NOS CO-OPS 092. National Oceanic and At-mospheric Administration, National Ocean Service, Silver Spring, MD, 24 pp.
https://tidesandcurrents.noaa.gov/publications/Techrpt_092_2019_State_of_US_High_Tide_Flood-ing_with_a_2020_Outlook_30June2020.pdf Sweet, W.V., A.S. Genz, J. Obeysekera, and J.J. Marra. 2020b: A regional frequency analysis of tide gauges to assess Pacific coast flood risk. Frontiers in Marine Science, 7, 883.
https://doi.org/10.3389/fmars.2020.581769 Sweet, W.V., S. Simon, G. Dusek, D. Marcy, W. Brooks, M. Pendleton, and J. Marra, 2021: 2021 State of High Tide Flooding and Annual Outlook. NOAA High Tide Flooding Report. National Oceanic and Atmo-spheric Administration, National Ocean Service, Silver Spring, MD, 28 pp.
https://tidesandcurrents.noaa.gov/publications/2021_State_of_High_Tide_Flooding_and_Annual_Out-look_Final.pdf Tebaldi, C., B.H. Strauss, and C.E. Zervas, 2012: Modelling sea level rise impacts on storm surges along US coasts. Environmental Research Letters, 7 (1), 014032. https://doi.org/10.1088/1748-9326/7/1/014032 Thompson, P.R., M.J. Widlansky, B.D. Hamlington, M.A. Merrifield, J.J. Marra, G.T. Mitchum, and W. Sweet, 2021: Rapid increases and extreme months in projections of United States high-tide flooding. Nature Climate Change, 1-7. https://doi.org/10.1038/s41558-021-01077-8 USACE, 1994: Engineering and Design: Hydrologic Frequency Estimates. ER 1110-2-1450. U.S. Army Corps of Engineers, Washington, DC, 8 pp. https://www.publications.usace.army.mil/Portals/76/Publications/En-gineerRegulations/ER_1110-2-1450.pdf USACE. 2014. Procedures to Evaluate Sea Level Change: Impacts, Responses, and Adaptation. Technical Let-ter 1100-2-1. U.S. Army Corps of Engineers, Washington, DC, 254 pp. https://www.publications.usace.
army.mil/portals/76/publications/engineertechnicalletters/etl_1100-2-1.pdf USACE MRG&P, 2017: 2015 Updated Atlas of U.S. Army Corps of Engineers Historic Daily Tide Data in Coastal Louisiana. MRG&P Report No. 14. U.S. Army Corps of Engineers, Mississippi River Geomor-phology and Potamology Program, New Orleans, LA, 41 pp. https://erdc-library.erdc.dren.mil/jspui/bit-stream/11681/25484/1/MRG%26P%20Report%20No%2014.pdf Global and Regional Sea Level Rise Scenarios for the United States l 72
 
USGCRP, 2017: Climate Science Special Report: Fourth National Climate Assessment, Volume I
[Wuebbles, D.J., D.W. Fahey, K.A. Hibbard, D.J. Dokken, B.C. Stewart, and T.K. Maycock (eds.)]. U.S.
Global Change Research Program, Washington, DC, USA, 470 pp. doi:10.7930/J0J964J6 USGCRP, 2018: Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment, Volume II [Reidmiller, D.R., C.W. Avery, D.R. Easterling, K.E. Kunkel, K.L.M. Lewis, T.K. Maycock, and B.C. Stewart (eds.)]. U.S. Global Change Research Program, Washington, DC, USA, 1515 pp.
doi:10.7930/NCA4.2018 Vitousek, S., P.L. Barnard, C.H. Fletcher, N. Frazer, L. Erikson, and C.D. Storlazzi, 2017: Doubling of coastal flooding frequency within decades due to sea-level rise. Scientific Reports, 7 (1), 1-9.
http://dx.doi.org/10.1038/s41598-017-01362-7 Vousdoukas, M.I., L. Mentaschi, E. Voukouvalas, M. Verlaan, S. Jevrejeva, L.P. Jackson, and L. Feyen, 2018:
Global probabilistic projections of extreme sea levels show intensification of coastal flood hazard.
Nature Communications, 9 (1), 2360. http://dx.doi.org/10.1038/s41467-018-04692-w Wada, Y., L.P.H. Van Beek, F.C. Sperna Weiland, B.F. Chao, Y.-H. Wu, and M.F.P. Bierkens, 2012: Past and future contribution of global groundwater depletion to sea-level rise. Geophysical Research Letters, 39 (9). https://doi.org/10.1029/2012GL051230 Wada, Y., M.-H. Lo, P.J.F. Yeh, J.T. Reager, J.S. Famiglietti, R.-J. Wu, and Y.-H. Tseng, 2016: Fate of water pumped from underground and contributions to sea-level rise. Nature Climate Change, 6 (8), 777-780.
http://dx.doi.org/10.1038/nclimate3001 Wahl, T., I.D. Haigh, R.J. Nicholls, A. Arns, S. Dangendorf, J. Hinkel, and A.B. Slangen, 2017: Understanding extreme sea levels for broad-scale coastal impact and adaptation analysis. Nature Communications, 8 (1), 16075. http://dx.doi.org/10.1038/ncomms16075 Wang, J., Church, J. A., Zhang, X., & Chen, X. (2021). Reconciling global mean and regional sea level change in projections and observations. Nature Communications, 12 (1), 1-12.
https://doi.org/10.1038/s41467-021-21265-6 WCRP Global Sea Level Budget Group, 2018: Global sea-level budget 1993-present. Earth System Science Data, 10, 1551-1590. http://dx.doi.org/10.5194/essd-10-1551-2018 Weaver, C.P., R.H. Moss, K.L. Ebi, P.H. Gleick, P.C. Stern, C. Tebaldi, R.S. Wilson, and J.L. Arvai, 2017: Refram-ing climate change assessments around risk: Recommendations for the US National Climate Assess-ment. Environmental Research Letters, 12 (8), 080201. https://doi.org/10.1088/1748-9326/aa7494 Werners, S.E., R.M. Wise, J.R.A. Butler, E. Totin, and K. Vincent, 2021: Adaptation pathways: A review of ap-proaches and a learning framework. Environmental Science & Policy, 116, 266-275.
https://doi.org/10.1016/j.envsci.2020.11.003 Yin, J., M.E. Schlesinger, and R.J. Stouffer, 2009: Model projections of rapid sea-level rise on the northeast coast of the United States. Nature Geoscience, 2 (4), 262-266. https://doi.org/10.1038/ngeo462 Zervas, C., 2013: Extreme Water Levels of the United States 1893-2010. NOAA Technical Report NOS CO-OPS 067. National Oceanic and Atmospheric Administration, National Ocean Service, Silver Spring, MD, 212 pp.
https://tidesandcurrents.noaa.gov/publications/NOAA_Technical_Report_NOS_COOPS_067a.pdf Global and Regional Sea Level Rise Scenarios for the United States l 73
 
Section A1: Tables and Figures Figure A1.1: Region definitions for observation-based extrapolations and scenarios in Section 2. These regions are used both to group tide gauges and also to generate regional averages for the gridded scenarios. A bathymetry mask is used to define the regions for the gridded scenarios.
Global and Regional Sea Level Rise Scenarios for the United States l 74
 
Figure A1.2. Shown for each tide gauge record with at least 30 years of record length between 1970 and 2020 are a) range, in meters, between median projection of Low and High Scenarios in 2050, and b) difference, in meters, between median observa-tion-based extrapolation and Intermediate scenario in 2050.
Global and Regional Sea Level Rise Scenarios for the United States l 75
 
Table A1.1: Projections methods employed.
Driver of GMSL or        Kopp et al. (2014) projection method          AR6 (Fox-Kemper et al., 2021) projection methods RSL change                (used in Sweet et al., 2017)                                      (used here)
Two-layer model with climate sensitivity calibrated to the Thermal expansion      CMIP5 ensemble drift-corrected zostoga      IPCC assessment and expansion coefficients calibrated to emulate CMIP6 models
: 1. Emulated ISMIP6 simulations through 2100 (Edwards et al.,
Likely range from IPCC AR5, with shape of 2021), extended after 2100 based on constant post-2100 Greenland ice sheet    tails based on structured expert judgment rates (Bamber and Aspinall, 2013)
: 2. Structured expert judgment (Bamber et al., 2019)
: 1. Emulated ISMIP6 simulations through 2100 (Edwards et al.,
2021), extended after 2100 with constant rates based on the IPCC AR5 parametric Antarctic Ice Sheet model (Church et al., 2013)
Likely range from IPCC AR5, with shape of
: 2. LARMIP-2 simulations (Levermann et al., 2020) augmented Antarctic ice sheet    tails based on structured expert judgment by AR5 surface mass balance model (Church et al., 2013),
(Bamber and Aspinall, 2013) extended past 2100 based on constant rates
: 3. Single ice-sheet model incorporated marine ice cliff insta-bility (DeConto et al., 2021)
: 4. Structured expert judgment (Bamber et al., 2019)
Emulated GlacierMIP (Marzeion et al., 2020; Edwards et al.,
Distribution based on Marzeion et al. (2012)
Glaciers                                                      2021) extended after 2100 with IPCC AR5 parametric model surface mass balance model refit to GlacierMIP (Marzeion et al., 2020)
Groundwater depletion: Population/          Groundwater depletion: Updated population/groundwater groundwater depletion relationship          depletion relationship calibrated based on Konikow (2011) calibrated based on Konikow (2011) and      and Wada et al. (2012, 2016)
Land water storage    Wada et al. (2012)                          Water impoundment: Population/dam impoundment Water impoundment: Population/dam            relationship calibrated based on Chao et al. (2008), adjusted impoundment relationship calibrated          for new construction, following Hawley et al. (2020) for 2020 based on Chao et al. (2008)                  to 2040 Ocean dynamic sea      Distribution derived from CMIP5 ensemble    Distribution derived from CMIP6 ensemble zos field after level          zos field                                    linear drift removal Gravitational, Sea-level equation solver (Mitrovica et al., Sea-level equation solver (Slangen et al., 2014) driven by rotational, and 2011) driven by projections of ice-sheet and projections of ice-sheet, glacier, and land water storage deformational glacier changes                              changes effects GIA and other      Spatiotemporal statistical model of tide-    Spatiotemporal statistical model of tide-gauge data drivers of VLM      gauge data                                  (updated from Kopp et al., 2014)
Table A1.2: Offsets, in meters, for different time periods and for each region considered in Section 2. These offsets are assessed using the trajectory determined from the available tide-gauge data in each region.
1992-2000              2000-2005                  2005-2020 Contiguous U.S.                    0.02                      0.03                    0.08 Northeast                        0.03                      0.02                    0.09 Southeast                        0.03                      0.02                    0.09 Eastern Gulf                      0.03                      0.02                      0.1 Western Gulf                      0.05                      0.04                    0.14 Southwest                        0.01                      0.01                    0.05 Northwest                        0.01                      0.01                    0.04 Hawaiian Islands                    0.02                      0.02                    0.06 Caribbean                        0.02                      0.01                    0.06 Global and Regional Sea Level Rise Scenarios for the United States l 76
 
Table A1.3: Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                    Tide                                  Minor            Major US                                                                        Index u Trend Epoch                  Moderate Grid  NOAA ID        Location    Latitude Longitude  Range                                Flood (m,          Flood Region                                                                      u (m, (mm/yr) of u                  Flood (m)
No.                                                      (m)                                  MHHW)                (m)
MHHW)
Pacific                                                                                      1983-39509    1611400    Nawiliwili, HI  21.95  159.36    0.558    0.244      1.7              0.522      0.817  1.192 2001 1983-39511    1612340    Honolulu, HI    21.31    157.87  0.580    0.248      1.3              0.523      0.817  1.193 2001 1983-39511    1612480    Mokuoloe, HI      21.43    157.79  0.646    0.265      2.0              0.526      0.819  1.196 2001 1983-39153    1615680      Kahului, HI    20.90    156.48    0.686    0.252      2.1              0.527      0.821  1.197 2001 1983-39154    1617433    Kawaihae, HI    20.04    155.83    0.659      0.237    7.9              0.526      0.820  1.196 2001 1983-38795    1617760        Hilo, HI      19.73  155.06    0.731    0.272      3.1              0.529      0.822  1.199 2001 1983-37704    1619000    Johnston Atoll    16.74  169.53    0.674    0.295      2.2              0.527      0.820  1.197 2001 1983-42004    1619910  Midway Islands    28.21    177.36  0.381    0.303      1.9              0.515      0.811  1.185 2001 Apra Harbor,                                                    1983-36941  1630000                      13.44    144.65    0.715    0.249      4.2              0.529      0.821  1.199 Guam                                                        2001 Pago Bay,                                                    1983-36941    1631428                      13.43    144.80    0.525    0.287      4.2              0.521      0.816  1.191 Guam                                                        2001 American                                                      1983-26574  1770000                      14.28    189.32    0.848    0.338      3.8              0.497      0.788  1.167 Samoa                                                        2001 1983-35169  1820000        Kwajalein      8.73    167.74  1.194    0.446      3.1              0.548      0.836  1.218 2001 1983-39117  1890000      Wake Island      19.29    166.62    0.718    0.295      2.1              0.529      0.822  1.199 2001 NE                                                                                          1983-47859    8410140    Eastport, ME    44.90    66.98    5.874    0.930      2.1              0.735      0.976  1.405 2001 Cutler Naval                                                    1983-47858    8411250                    44.64    67.30    4.133      0.716    2.4              0.665      0.924  1.335 Base, ME                                                      2001 1983-47857    8413320  Bar Harbor, ME    44.39    68.21    3.465      0.657    2.1              0.639      0.904  1.309 2001 1983-47496    8418150    Portland, ME    43.66    70.25    3.019    0.605      1.9              0.621      0.891  1.291 2001 1983-47496    8419317      Wells, ME    43.32    70.56    2.914    0.667      3.5                0.617    0.887  1.287 2001 1983-47496  8423898    Fort Point, NH  43.07      70.71  2.864    0.662      3.5              0.615      0.886  1.285 2001 1983-47136  8443970      Boston, MA    42.35    71.05    3.131    0.634      2.8              0.625      0.894  1.295 2001 1983-46777  8447386    Fall River, MA    41.70    71.16  1.456    0.566      3.5              0.558      0.844  1.228 2001 Woods Hole,                                                    1983-46778  8447930                      41.52    70.67    0.672    0.446      3.2              0.527      0.820  1.197 MA                                                          2001 Nantucket                                                      1983-46778  8449130                      41.29    70.10  1.089      0.418    3.8              0.544      0.833  1.214 Island, MA                                                      2001 1983-46777  8452660      Newport, RI      41.51    71.33    1.174    0.478    2.8              0.547      0.835  1.217 2001 Global and Regional Sea Level Rise Scenarios for the United States l 77
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                        Index u Trend Epoch                  Moderate Grid    NOAA ID      Location      Latitude Longitude  Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
NE                            Conimicut                                                      1983-46777    8452944                      41.72    71.34    1.398  0.560      3.5              0.556      0.842  1.226 (cont.)                          Light, RI                                                    2001 1983-46777    8454000    Providence, RI    41.81    71.40    1.476  0.549      2.3              0.559      0.844  1.229 2001 Quonset Point,                                                    1983-46777    8454049                      41.59    71.41  1.249    0.547    3.5              0.550      0.837  1.220 RI                                                        2001 New London,                                                      1983-46776    8461490                      41.36    72.09  0.930    0.468      2.6              0.537      0.828  1.207 CT                                                        2001 1983-46776    8465705  New Haven, CT      41.28    72.91    2.045  0.603      3.5              0.582      0.861  1.252 2001 1983-46775    8467150  Bridgeport, CT      41.17    73.18    2.231  0.555      3.0              0.589      0.867  1.259 2001 1983-46777    8510560    Montauk, NY      41.05    71.96    0.771    0.487    3.4              0.531      0.823    1.201 2001 Port Jefferson,                                                  1983-46416    8514560                      40.95    73.08    2.181    0.527    2.5              0.587      0.865  1.257 NY                                                        2001 1983-46416    8516945  Kings Point, NY    40.81    73.76    2.378  0.638      2.5              0.597      0.873  1.267 2001 1983-46415    8518750  The Battery, NY    40.70    74.01    1.542  0.546      3.1              0.562      0.846  1.232 2001 Bergen Point,                                                    1983-46415    8519483                      40.64    74.14    1.681  0.549      4.4              0.567      0.850  1.237 NY                                                        2001 1983-46415    8531680  Sandy Hook, NJ      40.47    74.01    1.593  0.552      2.7              0.564      0.848  1.234 2001 1983-46056    8534720  Atlantic City, NJ  39.36    74.42    1.403  0.534      4.1              0.556      0.842  1.226 2001 1983-45697    8536110    Cape May, NJ      38.97    74.96    1.659  0.486      4.7              0.566      0.850  1.236 2001 Ship John                                                      1983-46055    8537121                      39.31    75.38    1.894  0.578      3.5              0.576      0.857  1.246 Shoal, NJ                                                      2001 Marcus Hook,                                                    1983-46055    8540433                      39.81    75.41    1.871  0.563      3.5              0.575      0.856  1.245 PA                                                        2001 Philadelphia,                                                  1983-46055    8545240                      39.93    75.14    2.039  0.462      3.1              0.582      0.861  1.252 PA                                                        2001 Delaware City,                                                  1983-46055    8551762                      39.58    75.59    1.830  0.540      3.5              0.573      0.855  1.243 DE                                                        2001 1983-46055    8551910  Reedy Point, DE    39.56    75.57    1.779  0.423      4.1              0.571      0.853    1.241 2001 Brandywine                                                      1983-45696    8555889                      38.99      75.11  1.676    0.616    3.5              0.567      0.850  1.237 Shoal, DE                                                      2001 1983-45696    8557380      Lewes, DE      38.78    75.12    1.418  0.530      3.5              0.557      0.843  1.227 2001 1983-45696    8570280  Ocean City, MD    38.33    75.08    1.187    0.413    3.5              0.547      0.836    1.217 2001 Ocean City                                                    1983-45696    8570283                      38.33    75.09    0.751  0.360      3.5              0.530      0.823  1.200 Inlet, MD                                                      2001 Bishops Head,                                                    1983-45695    8571421                      38.22    76.04    0.624  0.503      3.5              0.525      0.819    1.195 MD                                                        2001 Global and Regional Sea Level Rise Scenarios for the United States l 78
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                        Index u Trend Epoch                  Moderate Grid    NOAA ID      Location    Latitude Longitude    Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
NE                                                                                          1983-45695    8571892  Cambridge, MD    38.57    76.07    0.622    0.414    4.9              0.525      0.819    1.195 (cont.)                                                                                        2001 Tolchester                                                    1983-46054    8573364                    39.21    76.25      0.527  0.484      2.5              0.519      0.814    1.189 Beach, MD                                                      2001 Chesapeake                                                      1983-46055    8573927                    39.53    75.81    0.980    0.470    3.8              0.539      0.829  1.209 City, MD                                                      2001 Havre De                                                      1983-46054    8574070                    39.54    76.09      0.746  0.482      3.5              0.530      0.822  1.200 Grace, MD                                                      2001 1983-46054    8574680  Baltimore, MD    39.27    76.58    0.506    0.443      3.2              0.520      0.815    1.190 2001 1983-45695    8575512  Annapolis, MD    38.98    76.48      0.438  0.430      3.7              0.518      0.813    1.188 2001 Solomons                                                      1983-45695    8577330                    38.32    76.45      0.449  0.398      6.0              0.518      0.813    1.188 Island, MD                                                      2001 1983-45694    8594900  Washington, DC  38.87    77.02    0.965    0.461    3.3              0.539      0.829  1.209 2001 Wachapreague,                                                    1983-45337    8631044                    37.61    75.69      1.376  0.508      5.4              0.564      0.850  1.234 VA                                                        2001 1983-45337    8632200    Kiptopeke, VA    37.17    75.99    0.896    0.435      4.7              0.536      0.827  1.206 2001 Colonial Beach,                                                  1983-45695    8635150                    38.25    76.96      0.591  0.406      4.7              0.524      0.818    1.194 VA                                                        2001 1983-45336    8635750    Lewisetta, VA  38.00    76.46      0.458  0.420      5.6              0.518      0.814    1.188 2001 Windmill Point,                                                  1983-45336    8636580                    37.62    76.29      0.424    0.419    5.2              0.532      0.828  1.202 VA                                                        2001 1983-45336    8637689    Yorktown, VA    37.23    76.48      0.786  0.567      3.5              0.531      0.824    1.201 2001 Sewells Point,                                                  1983-44977    8638610                    36.95    76.33      0.841  0.502      4.6              0.534      0.825  1.204 VA                                                        2001 1983-44977    8638863      CBBT, VA    36.97      76.11    0.885    0.503      6.0              0.535      0.827  1.205 2001 Money Point,                                                    1983-44977    8639348                    36.78    76.30      0.977  0.528      5.6              0.539      0.829  1.209 VA                                                        2001 Global and Regional Sea Level Rise Scenarios for the United States l 79
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                        Index u Trend Epoch                  Moderate Grid    NOAA ID        Location    Latitude Longitude  Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
SE                                                                                          1983-44978    8651370      Duck, NC      36.18    75.75    1.124  0.494      4.6              0.545      0.834    1.215 2001 Oregon Inlet,                                                  1983-44619    8652587                    35.80    75.55    0.360    0.384      4.6              0.514      0.811  1.184 NC                                                          2001 Cape Hatteras,                                                    1983-44619    8654400                    35.22    75.64    1.056    0.412    3.2              0.542      0.832    1.212 NC                                                          2001 USCG Hatteras,                                                    1983-44619    8654467                      35.21    75.70    0.186  0.598      3.2              0.507      0.806    1.177 NC                                                          2001 1983-44259    8656483    Beaufort, NC    34.72    76.67    1.079  0.362      3.8              0.543      0.832    1.213 2001 1983-44258    8658120  Wilmington, NC    34.23    77.95    1.427    0.327    2.3              0.557      0.843  1.227 2001 Wrightsville                                                  1983-44258    8658163                      34.21    77.79    1.366  0.564      3.2              0.555      0.841  1.225 Beach, NC                                                      2001 Springmaid                                                    1983-43898    8661070                    33.66    78.92    1.707  0.493      2.9              0.568      0.851  1.238 Pier, SC                                                      2001 Oyster Landing,                                                  1983-43897    8662245                    33.35    79.19    1.561  0.496      3.2              0.562      0.847  1.232 SC                                                        2001 1983-43538    8665530  Charleston, SC    32.78    79.93    1.757  0.453      3.3              0.570      0.853  1.240 2001 1983-43537    8670870  Fort Pulaski, GA  32.03    80.90    2.287  0.500      3.3              0.591      0.869    1.261 2001 Fernandina                                                    1983-42818    8720030                    30.67    81.47    1.999    0.473    2.3              0.580      0.860  1.250 Beach, FL                                                      2001 1983-42818    8720218      Mayport, FL    30.40    81.43    1.508  0.378      2.6              0.557      0.842  1.227 2001 St Johns River,                                                  1983-42818    8720357                      30.19    81.69    0.312  0.333      3.2              0.512      0.809    1.182 FL                                                        2001 St. Augustine                                                  1983-42459    8720587                    29.86    81.26    1.569    0.531    3.2              0.563      0.847  1.233 Beach, FL                                                      2001 1983-42101    8721604  Trident Pier, FL  28.42    80.59    1.193    0.407    5.1              0.537      0.825  1.207 2001 1983-41024    8723214  Virginia Key, FL  25.73    80.16    0.667    0.317    5.1              0.518      0.811  1.188 2001 1983-40664    8723970    Vaca Key, FL    24.71    81.11    0.297  0.249      4.2              0.512      0.809    1.182 2001 1983-40664    8724580    Key West, FL    24.56    81.81    0.551  0.262      2.5              0.522      0.817    1.192 2001 Global and Regional Sea Level Rise Scenarios for the United States l 80
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                        Index u Trend Epoch                  Moderate Grid    NOAA ID      Location    Latitude Longitude    Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
E. Gulf                                                                                        1983-41382    8725110      Naples, FL    26.13    81.81    0.875  0.323      2.9              0.535      0.826  1.205 2001 1983-41382    8725520  Fort Myers, FL  26.65    81.87      0.401  0.325      3.1              0.516      0.812    1.186 2001 Port Manatee,                                                    1983-41740    8726384                    27.64    82.56    0.669    0.260      6.6              0.527      0.820    1.197 FL                                                          2001 St Petersburg,                                                    1983-41740    8726520                    27.76    82.63    0.688    0.337      2.8              0.528      0.821    1.198 FL                                                          2001 Old Port                                                      1983-41740    8726607                    27.86    82.55      0.749  0.304      3.2              0.530      0.822  1.200 Tampa, FL                                                      2001 Mckay Bay                                                      1983-41740    8726667                    27.91    82.43      0.814  0.320      3.2              0.533      0.824  1.203 Entrance, FL                                                    2001 Clearwater                                                    1983-41740    8726724                    27.98    82.83      0.841  0.294      7.1              0.540      0.831  1.210 Beach, FL                                                      2001 1983-42457    8727520    Cedar Key, FL    29.14    83.03      1.157    0.415    2.2              0.546      0.835  1.216 2001 Apalachicola,                                                    1983-42456    8728690                    29.73    84.98      0.492  0.390      3.0              0.520      0.815    1.190 FL                                                          2001 Panama City,                                                    1983-42814    8729108                    30.15    85.67    0.409    0.368      2.5              0.516      0.812    1.186 FL                                                          2001 Panama City                                                    1983-42814    8729210                    30.21    85.88      0.420  0.348      4.3              0.517      0.813    1.187 Beach, FL                                                      2001 1983-42812    8729840    Pensacola, FL  30.40    87.21    0.383    0.345      2.4              0.515      0.811  1.185 2001 1983-42812    8732828  Mobile Bay, AL  30.42    87.83    0.490    0.519    4.3              0.520      0.815    1.190 2001 Dauphin Island,                                                  1983-42811    8735180                    30.25    88.08      0.367  0.354      4.3              0.512      0.808    1.182 AL                                                          2001 1983-42811    8736897      Mobile, AL    30.65    88.06      0.517  0.535      4.3              0.521      0.816    1.191 2001 Mobile State                                                    1983-42811    8737048                    30.71    88.04      0.501  0.439      4.3              0.520      0.815    1.190 Docks, AL                                                      2001 Pascagoula                                                      1983-42811    8741533                    30.37    88.56    0.466    0.494      4.3              0.519      0.814    1.189 NOAA Lab, MS                                                      2001 Bay Waveland                                                    1983-42810    8747437                    30.33    89.33    0.529    0.498      4.6              0.522      0.816    1.192 Yacht Club, MS                                                    2001 Global and Regional Sea Level Rise Scenarios for the United States l 81
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                        Index u Trend Epoch                  Moderate Grid    NOAA ID        Location    Latitude Longitude    Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
W. Gulf                      Pilots Station 2012-42092    8760922  East, SW Pass,  28.93    89.41    0.356    0.399      4.3              0.514      0.811  1.184 2016 LA 2012-42451    8761724    Grand Isle, LA  29.26    89.96    0.323    0.309      7.8              0.428      0.725  1.098 2016 New Canal                                                      1983-42809    8761927                    30.03      90.11    0.164  0.485      5.6              0.507      0.805    1.177 Station, LA                                                    2001 Port Fourchon,                                                    2012-42450    8762075                    29.11    90.20    0.368    0.298      4.3              0.515      0.811  1.185 LA                                                          2016 Amerada Pass,                                                    1983-42449    8764227                    29.45    91.34      0.487  0.535      4.3              0.519      0.815    1.189 LA                                                          2001 Cypremort                                                      1983-42449    8765251                    29.71    91.88      0.518  0.458      4.3              0.521      0.816    1.191 Point, LA                                                      2001 Freshwater                                                    1983-42448    8766072                    29.56    92.31      0.657  0.696      4.3              0.526      0.820    1.196 Canal Locks, LA                                                    2001 Lake Charles,                                                    1983-42806    8767816                    30.22    93.22      0.427  0.494      4.3              0.517      0.813    1.187 LA                                                          2001 Calcasieu Pass,                                                  1983-42447    8768094                    29.77    93.34    0.589    0.465      6.1              0.524      0.818    1.194 LA                                                          2001 Sabine Pass                                                    1983-42447    8770570                    29.73    93.87    0.488    0.368      6.1              0.520      0.815    1.190 North, TX                                                      2001 Morgans Point,                                                    1983-42446    8770613                    29.68    94.99    0.398    0.488      3.1              0.535      0.831  1.205 TX                                                          2001 1983-42446    8771013  Eagle Point, TX  29.48    94.92    0.338    0.331    13.8              0.494      0.790    1.164 2001 Galveston Bay                                                    1983-42446    8771341                    29.36    94.72      0.510  0.499      6.1              0.520      0.815    1.190 Entrance, TX                                                    2001 Galveston Pier                                                    1983-42446    8771450                    29.31    94.79      0.429  0.366      6.5              0.517      0.813    1.187 21, TX                                                        2001 Galveston 1983-42446    8771510    Pleasure Pier,  29.29    94.79      0.622  0.425      6.5              0.525      0.819    1.195 2001 TX 1983-42086    8772440    Freeport, TX  28.95    95.31    0.536    0.391    9.0              0.521      0.816    1.191 2001 USCG Freeport,                                                    1983-42086    8772447                    28.94    95.30    0.549    0.460      6.1              0.522      0.816    1.192 TX                                                          2001 2002-42084    8774770    Rockport, TX    28.02    97.05      0.111  0.336      5.7              0.504      0.803    1.174 2006 Corpus Christi,                                                  1983-41725    8775870                    27.58    97.22      0.497    0.391      4.8              0.529      0.824    1.199 TX                                                          2001 1983-41366    8779770  Port Isabel, TX  26.06    97.22      0.418  0.337      4.0              0.517      0.813    1.187 2001 Global and Regional Sea Level Rise Scenarios for the United States l 82
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                        Index u Trend Epoch                  Moderate Grid    NOAA ID        Location    Latitude Longitude    Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
SW                                                                                          1983-43500    9410170    San Diego, CA    32.71    117.17    1.745  0.490      2.2              0.570      0.852  1.240 2001 1983-43500    9410230    La Jolla, CA  32.87    117.26    1.624  0.468      2.1              0.565      0.849  1.235 2001 Los Angeles,                                                    1983-43858    9410660                    33.72    118.27    1.674  0.472      1.0              0.567      0.850  1.237 CA                                                          2001 Santa Monica,                                                    1983-44217    9410840                    34.01    118.50    1.654  0.489      1.8              0.566      0.850  1.236 CA                                                          2001 Santa Barbara,                                                    1983-44216    9411340                    34.41    119.69    1.645  0.485      0.6              0.566      0.849  1.236 CA                                                          2001 Port San Luis,                                                  1983-44574    9412110                    35.18    120.76    1.623  0.449      1.0              0.565      0.849  1.235 CA                                                          2001 1983-44932    9413450    Monterey, CA    36.61    121.89    1.627    0.431      1.6              0.565      0.849  1.235 2001 San Francisco,                                                    1983-45290    9414290                    37.81    122.47    1.780  0.375      1.9              0.571      0.853    1.241 CA                                                          2001 Redwood City,                                                    1983-45290    9414523                    37.51    122.21    2.501  0.400      2.7              0.600      0.875  1.270 CA                                                          2001 1983-45290    9414750    Alameda, CA    37.77    122.30    2.010    0.411      0.4              0.580      0.860  1.250 2001 1983-45290    9414863    Richmond, CA    37.93    122.40    1.846  0.359      3.1              0.574      0.855  1.244 2001 1983-45290    9415020  Point Reyes, CA  38.00    122.98    1.758  0.447      2.1              0.570      0.853  1.240 2001 Port Chicago,                                                    1983-45649    9415144                    38.06    122.04    1.498  0.388      1.4              0.560      0.845  1.230 CA                                                          2001 1983-45648    9416841  Arena Cove, CA  38.91    123.71    1.787  0.500      0.6              0.573      0.856  1.243 2001 1983-46365    9418767    North Spit, CA  40.77    124.22    2.090    0.491      4.8              0.584      0.863  1.254 2001 Crescent City,                                                  1983-46724    9419750                    41.75    124.18    2.095    0.548      -0.8              0.584      0.863  1.254 CA                                                          2001 1983-47083    9431647  Port Orford, OR  42.74    124.50    2.220    0.594      0.2              0.572      0.850  1.242 2001 1983-47442    9432780  Charleston, OR  43.35    124.32    2.323    0.586      1.1              0.593      0.870  1.263 2001 South Beach,                                                    1983-47801    9435380                    44.63    124.04    2.543    0.579      1.7              0.602      0.876  1.272 OR                                                          2001 1983-48161    9437540    Garibaldi, OR  45.55    123.92    2.536    0.597      2.4              0.601      0.876    1.271 2001 1983-48520    9439040      Astoria, OR  46.21    123.77    2.624  0.629    0.2              0.605      0.879  1.275 2001 Global and Regional Sea Level Rise Scenarios for the United States l 83
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                        Index u Trend Epoch                  Moderate Grid    NOAA ID        Location    Latitude Longitude    Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
NW                                                                                            1983-48520    9440910  Toke Point, WA    46.71    123.97    2.720  0.807      0.6              0.609      0.882  1.279 2001 1983-48519    9441102    Westport, WA    46.90    124.11    2.786  0.670      1.9              0.611    0.884  1.281 2001 1983-48878    9442396    La Push, WA    47.91    124.64    2.577  0.766      1.9              0.603      0.877  1.273 2001 1983-49237    9443090    Neah Bay, WA    48.37    124.61    2.425  0.688      1.7              0.597      0.873  1.267 2001 Port Angeles,                                                    1983-49238    9444090                    48.13    123.44    2.153  0.562      0.2              0.586      0.865  1.256 WA                                                          2001 Port Townsend,                                                    1983-49239    9444900                    48.11    122.76    2.597  0.538      1.7              0.604      0.878  1.274 WA                                                          2001 1983-48880    9446484    Tacoma, WA      47.27    122.41    3.595    0.517      3.4              0.644      0.908    1.314 2001 1983-48880    9447130      Seattle, WA  47.60    122.34    3.462    0.541      2.1              0.639      0.904  1.309 2001 Cherry Point,                                                    1983-49239    9449424                    48.86    122.76    2.788  0.585      0.4              0.612      0.884  1.282 WA                                                          2001 Friday Harbor,                                                  1983-49238    9449880                    48.55    123.01    2.364    0.554      1.2              0.595      0.871  1.265 WA                                                          2001 Alaska                                                                                        1983-51743    9450460    Ketchikan, AK  55.33    131.63    4.708    1.086    0.4              2.059      2.359  2.759 2001 Port Alexander,                                                  1983-52099    9451054                    56.25    134.65    3.329  0.738      5.8              1.031      1.331  1.731 AK                                                          2001 1983-52457    9451600      Sitka, AK    57.05    135.34    3.029  0.768      2.4              0.883      1.183  1.583 2001 2012-52817    9452210      Juneau, AK    58.30    134.41    4.970    1.152    15.1              2.319      2.619  3.019 2016 2012-53175    9452400    Skagway, AK    59.45    135.33    5.100    1.218    19.9              2.456      2.756  3.156 2016 1983-52815    9452634    Elfin Cove, AK  58.19    136.35    3.360    1.149    5.8              1.048      1.348  1.748 2001 Yakutat,                                                    2012-53171    9453220                    59.55    139.73    3.070    0.891    10.7              0.902      1.202  1.602 Yakutat Bay, AK                                                    2016 1983-53524    9454050    Cordova, AK    60.56    145.75    3.838    0.937      0.8              1.344      1.644  2.044 2001 1983-53882    9454240      Valdez, AK    61.13    146.36    3.702  0.878      5.8              1.253      1.553  1.953 2001 1983-53520    9455090      Seward, AK    60.12    149.43    3.238    0.884      4.0              0.983      1.283  1.683 2001 2012-53159    9455500    Seldovia, AK  59.44    151.72    5.499    1.350    9.8              2.906      3.206  3.606 2016 2012-53518    9455760      Nikiski, AK  60.68    151.40    6.262    1.254    9.9              NaN        NaN    NaN 2016 1983-53879    9455920  Anchorage, AK    61.24    149.89    8.889    1.269    2.7              NaN        NaN    NaN 2001 Kodiak Island,                                                  2012-52440    9457292                    57.73    152.51    2.675    0.715    9.2              0.743      1.043  1.443 AK                                                          2016 Global and Regional Sea Level Rise Scenarios for the United States l 84
 
Table A1.3 (cont.): Regional designation, tide gauge information, extreme water level metadata, and high tide flood heights.
Flood EWL                                                      Tide                                Minor              Major US                                                                      Index u Trend Epoch                  Moderate Grid    NOAA ID      Location    Latitude Longitude    Range                              Flood (m,            Flood Region                                                                      u (m, (mm/yr) of u                    Flood (m)
No.                                                        (m)                                MHHW)                (m)
MHHW)
Alaska                                                                                        2012-52079    9457804      Alitak, AK    56.90    154.25    3.578  0.908      5.8              1.174      1.474  1.874 (cont.)                                                                                        2016 1983-51714    9459450  Sand Point, AK  55.34    160.50    2.204    0.737      1.4              0.615      0.915  1.315 2001 1983-51712    9459881    King Cove, AK  55.06    162.33    2.082    0.753      5.8              0.592      0.892  1.292 2001 50262    9461380  Adak Island, AK  51.86    176.63      1.131  NaN      NaN      NaN      0.572      0.872  1.272 1983-50623    9461710      Atka, AK    52.23      174.17    1.041  0.424      5.8              0.584      0.884  1.284 2001 1983-50629    9462450    Nikolski, AK  52.94    168.87    1.213  0.537      5.8              0.563      0.863  1.263 2001 50990    9462620    Unalaska, AK    53.88    166.54    1.098    NaN      NaN      NaN      0.576      0.876  1.276 1983-51714    9463502  Port Moller, AK  55.99    160.57    3.175  0.697      5.8              0.952      1.252  1.652 2001 Village Cove, 52422    9464212                    57.13    170.29    1.005    NaN      NaN      NaN      0.589      0.889  1.289 AK 54940    9468756      Nome, AK      64.50    165.43    0.464    NaN      NaN      NaN      0.719      1.019  1.419 Red Dog Dock, 56018    9491094                    67.58    164.07    0.269    NaN      NaN      NaN      0.787      1.087  1.487 AK Prudhoe Bay, 57111    9497645                    70.40    148.53    0.214    NaN      NaN      NaN      0.808      1.108  1.508 AK Carib                                                                                        1983-38168    9751364    St. Croix, VI  17.75    64.71    0.226    0.205      2.4              0.509      0.807    1.179 2001 1983-38527    9751381    St. John, VI    18.32    64.72    0.252    0.210      2.4              0.510      0.808    1.180 2001 Lime Tree Bay,                                                    1983-38168    9751401                    17.69    64.75    0.216    0.154      3.0              0.509      0.806    1.179 VI                                                          2001 Charlotte                                                      1983-38527    9751639                    18.34    64.92    0.240    0.172    2.3              0.510      0.807    1.180 Amalie, VI                                                      2001 Vieques Island,                                                  1983-38526    9752695                    18.09      65.47    0.225    0.190      2.4              0.509      0.807    1.179 PR                                                          2001 1983-38525    9755371    San Juan, PR    18.46    66.12    0.481    0.191      2.4              0.519      0.814    1.189 2001 Magueyes                                                      1983-38165    9759110                    17.97    67.05    0.204    0.157      1.9              0.508      0.806    1.178 Island, PR                                                      2001 Mona Island,                                                    1983-38524    9759938                    18.09      67.94    0.247  0.257      2.4              0.510      0.807    1.180 PR                                                          2001 Global and Regional Sea Level Rise Scenarios for the United States l 85
 
Section A2: Methods Appendix: Extreme Water Levels and Alaska Coastal Flood Height A2.1: Data and Regional Frequency Analysis A regional frequency analysis (RFA) of NOAA tide gauges is used to estimate extreme water levels (EWLs) along U.S. coastlines at and away from tide gauges. The RFA method (Hosking and Wallis, 1997) is based on the assumption that similar physical forcing across a region will produce a similar frequency of events and a probability density up to a local index (u), which is a local scaling factor that captures response pe-culiarities (Dalrymple, 1960). An RFA uses regional sets of data that have been locally normalized by their respective local index with a statistical heterogeneity test (H value) to assess the extent that the data are sufficiently similar. Using statistical L-moments, heterogeneity is a measure of the variation between sites of a locations summary distribution statistics and the amount of dispersion expected if the locations were indeed a homogeneous region (Hosking and Wallis, 1997). If H < 1, the region is considered acceptably homogeneous. If 1  H < 2, the region is considered possibly heterogeneous but acceptable for our study. If H  2, then the tide-gauge group is definitely heterogeneous and not suitable for analysis. Once the region-al bounds are established whose data are acceptably homogeneous, the aggregated data are fit with an extreme value distribution.
This study uses hourly and top ten data from all NOAA tide gauges46 with at least 10 years of record (Fig-ure A2.1). Water levels are put onto the mean higher high water (MHHW) tidal datum and detrended (the trend value is retained and shown in Table A1.3) relative to the midpoint of the current national datum tid-al epoch (1983-2001), which is similar for NOAA EWL procedures using a single-gauge analysis (Zervas, 2013; Extreme Water Levels47). From the datasets, daily highest water levels are picked and declustered at each tide gauge using a 4-day storm window to ensure event independence. The 98th percentile of the Figure A2.1: NOAA tide gauges used in the regional frequency analysis to generate extreme water level probabilities for U.S. coastlines.
46 https://tidesandcurrents.noaa.gov/
47 https://tidesandcurrents.noaa.gov/est/
Global and Regional Sea Level Rise Scenarios for the United States l 86
 
declustered daily highest levels at each tide gauge is used as the local index (u) to normalize the data for the RFA process.
To form regions, the tide-gauge data is aggregated across a 400 km radius, similar to methods of Hall et al.
(2016) but from the midpoint of a continuous set of coastline-intersecting 1-degree grids instead of site-spe-cific installations. A maximum of 10 and a minimum of 3 tide gauges are included for each grid. Next, the regional data are spatially declustered with an additional 4-day event (i.e., storm) window to ensure that only the maximum water level within a region is retained (keep only the highest peak water levels for a particu-lar event). Then, the statistical heterogeneity measure is estimated to ensure that the grouped tide-gauge data are sufficiently homogeneous (H < 2). In some instances, when a region surrounding a grid centroid Figure A2.2: Example of data from grid number 46415 showing exceedances above each local index (u) relative to the 1983-2001 mean higher high water (MHHW) tidal datum at a) Kings Point, New York; b) The Battery, New York; c) Bergen Point, New York; and d) Sandy Hook, New Jersey, which are e) aggregated into a single dataset and f) fit by a Generalized Pareto Distribution to form a return level interval curve for the grid.
Global and Regional Sea Level Rise Scenarios for the United States l 87
 
has H  2, tide gauges farthest away are sequentially dropped until homogeneity is achieved. In the end, all 1-degree grids along the contiguous United States (CONUS) had H < 2 (considered acceptably homoge-neous) except a grid (number 48519) along the Northwest Pacific coastline, which, along with the Hawaiian and other U.S. Pacific Islands, uses the much larger physical-process regions identified and quantified in Sweet et al. (2020b). Grids along the Alaska coastline are fairly well resolved by the RFA except along the western and northern coasts.
An example is shown for grid number 46415, which is where the NOAA tide gauge at The Battery in New York City (NYC) is located (Figure A2.2). Four tide gauges are included in this grid (Kings Point, New York; The Battery, New York; Bergen Point, New York; and Sandy Hook, New Jersey [Figure A2.2a-d]), and their data are considered homogeneous (H value of 0.32). After the 4-day spatial filtering for events, each of the tide-gauge datasets is normalized by (divided by) its respective local index (u) value and aggregated as shown in Figure A2e.
A2.2: Gridded (Regional) Extreme Water Level Probabilities With the tide gauges identified for each 1-degree grid, the aggregated and normalized datasets are fit with a Generalized Pareto Distribution (GPD; Coles, 2001). Using the penalized maximum likelihood method (Coles and Dixon, 1999; Frau et al., 2018; Sweet et al., 2020b), expected and 95% confidence interval (2.5th% and 97.5th% levels) values are estimated for the gridded EWL probabilities and defined as:
1) where G is the exceedance probability (P[Z > z]),  is the probability of an individual (normalized) observation exceeding the local index (u),  is the scale parameter, and  is the shape parameter. It is assumed that the distribution of the number of exceedances per year follows a Poisson distribution and that the return level for an EWL of height (Z) is given by:
2) where N is the average recurrence interval (referred to in this study as the average event frequency, which is the reciprocal value), ny is number of days per year (365.25), and  is the average number of event exceed-ances per year (about 3 on average across all tide gauges in the study). To estimate EWLs with return levels with a 10 events/year frequency, we extrapolate the gridded GPD model with a logarithmic fit for return levels between the 0.5-3 events/year frequencies. A return level interval curve fit to the aggregated data (Figure A2.2e) for the grid where NYC is located is shown in Figure A2.2f.
A2.3: Localized Extreme Water Level Probabilities When fitting a GPD to the RFA of aggregated tide-gauge data, the local EWL (EWLlocal) probabilities including the model of expected values and their 95% confidence interval at a particular location are given as 3) where EWLgridded is the gridded return level for a particular coastal 1-degree grid and ulocal is the local index used in both the RFA and GPD processes. The value of u is a height (98th percentile of 4-day event filtered daily highest water level) above the local MHHW tidal datum for the current (1983-2001) national tidal da-tum epoch (NTDE) or for a modified 5-year epoch. The associated uncertainty of the EWLgridded estimated during the RFA is expressed as gridded. When localized at a tide gauge used in the formulation of the grids (see Figure A1), u is assumed to have no uncertainty. However, just as the location parameters in generalized extreme value (GEV) have time-dependent characteristics (Men&#xe9;ndez and Woodworth, 2010), it is recognized that u would experience similar behavior, but that is not quantified in this study.
Global and Regional Sea Level Rise Scenarios for the United States l 88
 
In this RFA framework, it is possible to estimate EWLlocal from the EWLgridded probabilities (expected values and 95% confidence interval) through the use of other sources of data. Specifically, the local indices needed to localize the EWLgridded values can either be 1) obtained from short-term tide-gauge data (or by targeted de-ployments) within a particular grid that is not included in the RFA formulation (<10 years; Figure A2.3) or 2) based on an underlying relationship between regional sets of local index (u) values and tide range available from, for example, NOAA VDatum.48 In both cases, we establish large U.S. coastal regions (note: these are slightly different than the regions discussed in Sections 2 and 3 of the report and shown in Figure A1.1) that encompass several 1-degree grids to quantify information needed to obtain local indices and/or estimate variance/uncertainties (e.g., RMSE). These alternative methods, which are discussed below, may be of inter-est to coastal communities that are not co-located to a tide gauge used in this study but have predictions of tide range or have access to or are planning temporary tide-gauge installations to establish tidal datums and/or EWLs.
Figure A2.3: Additional tide-gauge data available from NOAA that can be used to localize the 1-degree gridded set of regional frequency analysis-based extreme water level probabilities. See https://tidesandcurrents.noaa.gov/.
A2.3.1: Local Index Estimates from Short-Term Installations When other sets of tide/water level data are available, a local index can be directly estimated to obtain EWLlocal probabilities from the EWLgridded probabilities. The first step for using data that are not from NOAA would be to estimate a local MHHW tidal datum using, for example, NOAAs online datum tool.49 Following Equation 3 above, there will be some uncertainty in the local index value that is dependent on record length (e.g., 1-10 years). To account for short-record uncertainty in the local indices (u), RMSE (1 standard error) is es-timated for regional estimates of u for the tide gauges used in the RFA (see Figure A2.1). Root mean square error is estimated using a logarithmic fit over a 19-year record length (Figure A4). To compute the RMSE, the maximum absolute differences are computed between u derived over the entire record and for progres-sively longer consecutive record lengths between 2001 and 2019 at each tide gauge (e.g., 19 discrete 1-year 48 https://vdatum.noaa.gov/
49 https://access.co-ops.nos.noaa.gov/datumcalc/
Global and Regional Sea Level Rise Scenarios for the United States l 89
 
records; 18 consecutive 2-year records). The maximum (absolute) difference is used to account for interannu-al variability that can be significant (e.g., during phases of El Nino-Southern Oscillation [ENSO]). This differ-ence is considered the error in estimating u for shorter records, and the average of the absolute differences across the regional set of tide gauges is considered the bias. The standard deviation of the absolute differ-ences is also computed across all tide gauges, and an estimate of the RMSE is then computed as the square Figure A2.4: Root mean square error for regional estimates of flood indices (u) based on 1-19 years of consecutive data over the 2001-2019 period, based on regional sets of tide gauges used in this study. Note: these regions are not the same as those shown in Figure A1.1 and used to describe results in Sections 2 and 3 of the report.
Global and Regional Sea Level Rise Scenarios for the United States l 90
 
root of the sum of the square of the bias and the standard deviation (variance). The estimates for Hawaiian and U.S. Pacific Islands follow estimates of Sweet et al. (2020b).
A2.3.2: Obtaining a Local Index from Tide Range Information Another method to obtain an estimate of a local index (u) and its uncertainty is based on a dependency (correlation) that exists with tide range (great diurnal [GT]) along most coastal regions similar to findings of Merrifield et al. (2013). In essence, tide range (GT), which represents the spread between MHHW and mean lower low water (MLLW), partially quantifies the variance of the daily highest water level distribution and the height of the local index u. Figure A2.5 illustrates the regression-based relationships between tide range and u along U.S. coastal regions (these are the same regions used in Figure A2.4). All regressions are significant above the 90% significance level (p values < 0.1) and applicable for the 1983-2001 tidal epoch. For the Ha-waiian and U.S. Pacific Islands, the Pacific-wide regression of Sweet et al. (2020b) is used.
Global and Regional Sea Level Rise Scenarios for the United States l 91
 
Figure A2.5: Tide range to local index (u) regressions with equations, goodness of fit (R2), and root mean squared error (RMSE) shown by regions. Note: all local indices (u) are relative to the 1983-2001 tidal datum epoch. In the equations, y represents the local index (u) and x represents tide range.
Global and Regional Sea Level Rise Scenarios for the United States l 92
 
A2.3.3: Uncertainties Using Alternative Methods to Estimate EWLlocal Probabilities When using either alternative method (tide range or short-record estimates) to obtain a local index (u), the uncertainty estimates of EWLlocal probabilities will include additional uncertainty in u (u). Following methods of Sweet et al. (2020b), it can be shown that 4) where EWLgridded and u are the expected values of the gridded return levels and the expected value of u, for example, estimated by the tide-range and u dependency (see Figure A2.5), respectively, u2 is the uncertain-ty inherent to any u-prediction relationship (e.g., RMSE). Thus, there is an additive uncertainty in u as estimat-ed from this relationship, which would introduce additional uncertainty in estimates of EWLlocal.
A2.3.4: Adjusting Local Extreme Water Level Probabilities to Time Periods To adjust the EWLlocal probabilities to a different sea level other than the current tidal epoch (e.g., from 1992 to 2000 or 2005 so as to apply the sea level rise scenarios), RSL estimates using the trends inherent to the hourly data used to compute the local index (u) should be applied (Table A1.3) to the epoch-specific EWLlocal probabilities themselves. For tide gauges used in the RFA analysis and with more than 20 years of data, the local u trend can be used; otherwise, a median regional trend as defined in Figures A2.4 and A2.5 can be used. Alternatively, the RSL offsets derived from the regional observational RSL data (Table A1.2) could be used with differences between methods considered insignificant. For example, to estimate probabilities for the year 2000, the EWLlocal probabilities values would be increased by an amount equal to the trend in u (or the median u trend value for the region) multiplied by 8 years (since 1992, which is the midpoint of the 1983-2001 epoch). The same procedure should be followed to adjust EWLlocal probabilities for a given loca-tion estimated via the tide range regression (see Figure A5). In the case of a short-term estimate of u, similar procedures should be followed if local tidal datums have been computed and adjusted to the national tidal datum epoch (e.g., using the CO-OPS Tidal Analysis Datum Calculator50); in the case where no epoch can be established (see the CO-OPS Tidal Analysis Datum Calculator for guidance), then the measurements will be assumed to be referenced to the period of collection, and trend adjustment may be less straightforward.
A2.4: Alaska Coastal Flood Heights To assess flood exposure, the coastal high tide flooding (HTF) heights of Sweet et al. (2018) are used for all U.S. coastlines outside of Alaska. Used in NOAA annual outlooks (e.g., Sweet et al., 2021; The State of High Tide Flooding and Annual Outlook51), these heights are a best-fit solution (regression) to the dozens of National Weather Service (NWS) emergency response warning thresholds established at many (but not all)
NOAA tide gauges along the countrys coastline. The NWS thresholds are used to communicate expected or ongoing coastal flood hazards to the public (NOAA, 2020), but often their depth-severity thresholds vary according to specific features near the tide gauge that affect both the associated flood frequency and the degree of broader vulnerabilities. Along the Alaska coastline, we follow the methodologies of Sweet et al.
(2020b), who used a slight modification to assess damaging flood heights for the Pacific Basin coastlines.
Here, the Alaska flood heights are based on a quadratic regression model using only Pacific Coast NWS minor flood heights and considered for only tide ranges below 6 meters (Figure A2.6a). To obtain moderate and major flood heights for Alaska, 0.3 m and 0.7 m are added to the regression, which is approximately the median difference between these heights and those for minor flooding along CONUS (Sweet et al., 2018).
With flood heights defined nationally, minor, moderate, and major HTF are defined as occurring when water levels reach or exceed heights of about (median values) 0.55 m, 0.85 m, and 1.2 m above MHHW, respective-ly, and linearly vary with tide range (Figures A2.6b-d).
50 https://access.co-ops.nos.noaa.gov/datumcalc/index.jsp 51 https://tidesandcurrents.noaa.gov/HighTideFlooding_AnnualOutlook.html Global and Regional Sea Level Rise Scenarios for the United States l 93
 
Figure A2.6: a) Quadratic regression of U.S. West Coast minor flood heights of NOAAs National Weather Service, following methods of Sweet et al. (2020b), to obtain a minor HTF definition for Alaskas coastline. The NOAA flood heights for b) minor, c) moderate, and d) major HTF are shown relative to mean higher high water.
Global and Regional Sea Level Rise Scenarios for the United States l 94
: 8. Acronyms Note: state abbreviations have been omitted AIS: Antarctic ice sheet AEF: average event frequency AMOC: Atlantic meridional overturning circulation AR5: [IPCC] Fifth Assessment Report AR6: [IPCC] Sixth Assessment Report ARI: average return interval C: celsius CDF: cumulative distribution function cm: centimeter CMIP5: Coupled Model Intercomparison Project Phase 5 CMIP6: Coupled Model Intercomparison Project Phase 6 CONUS: contiguous United States CO-OPS: Center for Operational Oceanographic Products and Services CoSMoS: Coastal Storm Modeling System DRSL: Department of Defense Regional Sea Level [database]
ENSO: El Nino-Southern Oscillation EPA: Environmental Protection Agency EWL: extreme water level FEMA: Federal Emergency Management Agency FFRD: Future of Flood Risk Data GCM: global climate model GEV: generalized extreme value GHG: greenhouse gas GIA: glacial isostatic adjustment GlacierMIP: Glacier Model Intercomparison Project GMSL: global mean sea level GPD: Generalized Pareto Distribution GPS: Global Positioning System GRD: gravitational, rotational, and deformational GT: great diurnal tide range HTF: high tide flood, flooding HUC: hydrologic unit code InSAR: Interferometric Synthetic Aperture Radar IPCC: Intergovernmental Panel on Climate Change ISMIP6: Ice Sheet Model Intercomparison Project for CMIP6 JPM-OS: joint probability method-optimal sampling [procedure]
LARMIP-2: Linear Antarctic Response Model Intercomparison Project [version 2]
m: meter MHHW: mean higher high water MICI: marine ice cliff instability MLLW: mean lower low water mm: millimeter NASA: National Aeronautics and Space Administration NAVD88: North American Vertical Datum of 1988 NCA: National Climate Assessment NCA5: Fifth National Climate Assessment NCA4: Fourth National Climate Assessment NOAA: National Oceanic and Atmospheric Administration NOC: National Ocean Council Global and Regional Sea Level Rise Scenarios for the United States l 95
 
NSRS: National Spatial Reference System NTDE: national tidal datum epoch NWS: National Weather Service NYC: New York City PDO: Pacific Decadal Oscillation R2: goodness of fit RFA: regional frequency analysis RMSE: root mean square error RSL: relative sea level SOST: Subcommittee on Ocean Sciences and Technology SSP: Shared Socioeconomic Pathway USACE: U.S. Army Corps of Engineers USGCRP: U.S. Global Change Research Program USGS: U.S. Geological Survey VDatum: Vertical Datum Transformation VLM: vertical land motion
 
Global and Regional Sea Level Rise Scenarios for the United States l 97 ATTACHMENT B 18 16 14 Sea Level Change (ft) 12 10 8
6 4
2 2020  2040      2060  2080          2100  2120 Year RESET
 
0.55 0.50 0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 Low Intermediate Low Intermediate Intermediate High High
 
ATTACHMENT C Foreword Technical and Preface Summary 33
 
34 TS Technical Summary Coordinating Authors:
Paola A. Arias (Colombia), Nicolas Bellouin (United Kingdom/France), Erika Coppola (Italy),
Richard G. Jones (United Kingdom), Gerhard Krinner (France/Germany, France), Jochem Marotzke (Germany), Vaishali Naik (United States of America), Matthew D. Palmer (United Kingdom),
Gian-Kasper Plattner (Switzerland), Joeri Rogelj (United Kingdom/Belgium), Maisa Rojas (Chile),
Jana Sillmann (Norway/Germany), Trude Storelvmo (Norway), Peter W. Thorne (Ireland/United Kingdom), Blair Trewin (Australia)
Authors:
Krishna Achuta Rao (India), Bhupesh Adhikary (Nepal), Richard P. Allan (United Kingdom),
Kyle Armour (United States of America), Govindasamy Bala (India/United States of America),
Rondrotiana Barimalala (South Africa/Madagascar), Sophie Berger (France/Belgium),
Josep G. Canadell (Australia), Christophe Cassou (France), Annalisa Cherchi (Italy), William Collins (United Kingdom), William D. Collins (United States of America), Sarah L. Connors (France/United Kingdom), Susanna Corti (Italy), Faye Cruz (Philippines), Frank J. Dentener (EU/The Netherlands),
Claudine Dereczynski (Brazil), Alejandro Di Luca (Australia, Canada/Argentina), Aida Diongue Niang (Senegal), Francisco J. Doblas-Reyes (Spain), Alessandro Dosio (Italy), Herv&#xe9; Douville (France),
Fran&#xe7;ois Engelbrecht (South Africa), Veronika Eyring (Germany), Erich Fischer (Switzerland), Piers Forster (United Kingdom), Baylor Fox-Kemper (United States of America), Jan S. Fuglestvedt (Norway), John C. Fyfe (Canada), Nathan P. Gillett (Canada), Leah Goldfarb (France/United States of America), Irina Gorodetskaya (Portugal/Russian Federation, Belgium), Jose Manuel Gutierrez (Spain), Rafiq Hamdi (Belgium), Ed Hawkins (United Kingdom), Helene T. Hewitt (United Kingdom),
Pandora Hope (Australia), AKM Saiful Islam (Bangladesh), Christopher Jones (United Kingdom),
Darrell S. Kaufman (United States of America), Robert E. Kopp (United States of America),
Yu Kosaka (Japan), James Kossin (United States of America), Svitlana Krakovska (Ukraine),
June-Yi Lee (Republic of Korea), Jian Li (China), Thorsten Mauritsen (Sweden, Denmark), Thomas K. Maycock (United States of America), Malte Meinshausen (Australia/Germany), Seung-Ki Min (Republic of Korea), Pedro M. S. Monteiro (South Africa), Thanh Ngo-Duc (Vietnam), Friederike Otto (United Kingdom/Germany), Izidine Pinto (South Africa/Mozambique), Anna Pirani (Italy),
Krishnan Raghavan (India), Roshanka Ranasinghe (The Netherlands/Sri Lanka, Australia), Alex C. Ruane (United States of America), Lucas Ruiz (Argentina), Jean-Baptiste Sall&#xe9;e (France), Bj&#xf8;rn H. Samset (Norway), Shubha Sathyendranath (UK/Canada, United Kingdom, Overseas Citizen of India), Sonia I. Seneviratne (Switzerland), Anna A. Srensson (Argentina), Sophie Szopa (France),
Izuru Takayabu (Japan), Anne-Marie Treguier (France), Bart van den Hurk (The Netherlands),
35
 
Technical Summary Robert Vautard (France), Karina von Schuckmann (France/Germany), Snke Zaehle (Germany),
Xuebin Zhang (Canada), Kirsten Zickfeld (Canada/Germany)
Contributing Authors:
Gufinna Aalgeirsd&#xf3;ttir (Iceland), Lincoln M. Alves (Brazil), Terje Berntsen (Norway),
Sara M. Blichner (Norway), Lisa Bock (Germany), Gregory G. Garner (United States of America),
Joelle Gergis (Australia), Sergey K. Gulev (Russian Federation), Mathias Hauser (Switzerland),
Flavio Lehner (United States of America/Switzerland), Chao Li (China), Marianne T. Lund (Norway), Daniel J. Lunt (United Kingdom), Sebastian Milinski (Germany), Gemma Teresa Narisma (Philippines), Zebedee R. J. Nicholls (Australia), Dirk Notz (Germany), Sophie Nowicki (United States of America/France, United States of America), Bette Otto-Bliesner (United States of America), Brodie Pearson (United States of America/United Kingdom), Adam S. Phillips (United States of America), James Renwick (New Zealand), St&#xe9;phane S&#xe9;n&#xe9;si (France), Lucas Silva (Portugal/Switzerland), Aimee B. A. Slangen (The Netherlands), Thomas F. Stocker (Switzerland),
TS                  Claudia Tebaldi (United States of America), Laurent Terray (France), Sabin Thazhe Purayil (India),
Andrew Turner (United Kingdom), Steven Turnock (United Kingdom), Carolina Vera (Argentina),
Cunde Xiao (China), Panmao Zhai (China)
Review Editors:
Val&#xe9;rie Masson-Delmotte (France), Gregory M. Flato (Canada), Noureddine Yassa (Algeria)
This Technical Summary should be cited as:
Arias, P.A., N. Bellouin, E. Coppola, R.G. Jones, G. Krinner, J. Marotzke, V. Naik, M.D. Palmer, G.-K. Plattner, J. Rogelj, M. Rojas, J. Sillmann, T. Storelvmo, P.W. Thorne, B. Trewin, K. Achuta Rao, B. Adhikary, R.P. Allan, K. Armour, G. Bala, R. Barimalala, S. Berger, J.G. Canadell, C. Cassou, A. Cherchi, W. Collins, W.D. Collins, S.L. Connors, S. Corti, F. Cruz, F.J. Dentener, C. Dereczynski, A. Di Luca, A. Diongue Niang, F.J. Doblas-Reyes, A. Dosio, H. Douville, F. Engelbrecht, V. Eyring, E. Fischer, P. Forster, B. Fox-Kemper, J.S. Fuglestvedt, J.C. Fyfe, N.P. Gillett, L. Goldfarb, I. Gorodetskaya, J.M. Gutierrez, R. Hamdi, E. Hawkins, H.T. Hewitt, P. Hope, A.S. Islam, C. Jones, D.S. Kaufman, R.E. Kopp, Y. Kosaka, J. Kossin, S. Krakovska, J.-Y. Lee, J. Li, T. Mauritsen, T.K. Maycock, M. Meinshausen, S.-K. Min, P.M.S. Monteiro, T. Ngo-Duc, F. Otto, I. Pinto, A. Pirani, K. Raghavan, R. Ranasinghe, A.C. Ruane, L. Ruiz, J.-B. Sall&#xe9;e, B.H. Samset, S. Sathyendranath, S.I. Seneviratne, A.A. Srensson, S. Szopa, I. Takayabu, A.-M. Tr&#xe9;guier, B. van den Hurk, R. Vautard, K. von Schuckmann, S. Zaehle, X. Zhang, and K. Zickfeld, 2021: Technical Summary. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. P&#xe9;an, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelek&#xe7;i, R. Yu, and B. Zhou (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 33144.
doi:10.1017/9781009157896.002.
36
 
Technical Summary Table of Contents Introduction      38 TS.3    Understanding the Climate System Response and Implications for Limiting Global Warming                                                               90 Box TS.1 l Core Concepts Central to This Report                                                39    TS.3.1  Radiative Forcing and Energy Budget                                                 90 TS.3.2  Climate Sensitivity and TS.1    A Changing Climate                              43            Earth System Feedbacks                               93 TS1.1    Context of a Changing Climate                                              43    TS.3.3  Temperature Stabilization, Net Zero Emissions and Mitigation                                             97 Box TS.2 l Paleoclimate            45    Box TS.7 l Climate and Air Quality Responses                                                                                      TS TS.1.2  Progress in Climate Science                                        47    to Short-lived Climate Forcers TS.1.3  Assessing Future Climate Change                                                  52    in Shared Socio-economic Pathways                                              103 TS.1.4  From Global to Regional Climate Information                                                                          Box TS.8 l for Impact and Risk Assessment  57                                                Earth System Response to Solar Radiation Modification                                    104 Cross-Section Box TS.1l Box TS.9 l Global Surface Temperature Change                                            59 Irreversibility, Tipping Points and Abrupt Changes  106 TS.2    Large-scale Climate Change: Mean Climate, Variability and Extremes  63 TS.4    Regional Climate Change                                    107 TS.2.1  Changes Across the Global Climate System                                                              63 TS.4.1  Generation and Communication of Regional TS.2.2  Changes in the Drivers of the Climate System                                                               67            Climate Change Information  107 TS.2.3  Upper Air Temperatures                                                                                                Box TS.10 l and Atmospheric Circulation                                          70    Event Attribution              108 Box TS.3 l                                                                                                                    Box TS.11 l Low-likelihood, High-warming Storylines                                                      72    Climate Services              111 TS.2.4  The Ocean                  74 Box TS.12 l TS.2.5  The Cryosphere                        76    Multiple Lines of Evidence for Assessing Box TS.4 l                                                                                                                    Regional Climate Change and the Interactive Atlas  111 Sea Level      77    TS.4.2  Drivers of Regional Climate Variability Box TS.5 l                                                                                                                              and Change  113 The Carbon Cycle                  79    Box TS.13 l TS.2.6  Land Climate, Including Biosphere                                                                                    Monsoons  118 and Extremes  82                          TS.4.3  Regional Climate Change Box TS.6 l                                                                                                                              and Implications for Climate Extremes Water Cycle          85            and Climatic Impact-Drivers  120 Box TS.14 l Infographic TS.1 l                                                                                                                Urban Areas        144 Climate Futures  88 37
 
Technical Summary Introduction                                                                                              TS.3 summarizes knowledge and understanding of climate forcings, feedbacks and responses. Infographic TS.1 uses a storyline approach The Working Group I (WGI) contribution to the Intergovernmental                                            to integrate findings on possible climate futures. Finally, Section TS.4 Panel on Climate Change (IPCC) Sixth Assessment Report (AR6)                                              provides a synthesis of climate information at regional scales.3 The assesses the physical science basis of climate change. As part of                                          list of acronyms used in the WGI Report is in Annex VIII.
that contribution, this Technical Summary (TS) is designed to bridge between the comprehensive assessment of the WGI Chapters                                                      Text at the beginning of a section presented in dark blue and its Summary for Policymakers (SPM). It is primarily built from                                            with a blue vertical bar at the left, as shown here, provides the Executive Summaries of the individual chapters and Atlas and                                              a summary of the findings discussed in that section.
provides a synthesis of key findings based on multiple lines of evidence (e.g., analyses of observations, models, paleoclimate                                            The AR6 WGI Report promotes best practices in traceability information and understanding of physical, chemical and biological                                        and reproducibility, including through adoption of the Findable, processes and components of the climate system). All the findings                                          Accessible, Interoperable, and Reusable (FAIR) principles for scientific and figures here are supported by and traceable to the underlying                                          data. Each chapter has a data table (in its Supplementary Material)
TS chapters, with relevant chapter sections indicated in curly brackets.                                      documenting the input data and code used to generate its figures and tables. In addition, a collection of data and code from the report Throughout this Technical Summary, key assessment findings are                                            has been made freely-available online via long-term archives.4 reported using the IPCC calibrated uncertainty language (Chapter 1, Box 1.1). Two calibrated approaches are used to communicate the                                            These FAIR principles are central to the WGI Interactive Atlas5, an degree of certainty in key findings, which are based on author teams                                      online tool that complements the WGI Report by providing flexible evaluations of underlying scientific understanding:                                                        spatial and temporal analyses of past, observed and projected climate change information. It comprises a regional information
: 1) Confidence1 is a qualitative measure of the validity of a finding,                                      component that supports many of the chapters of the Report and based on the type, amount, quality and consistency of evidence                                        a regional synthesis component that supports the Technical Summary (e.g., data, mechanistic understanding, theory, models, expert                                        and Summary for Policymakers.
judgment) and the degree of agreement.
: 2) Likelihood2 provides a quantified measure of confidence in                                              Regarding the representation of robustness and uncertainty in maps, a finding expressed probabilistically (e.g., based on statistical                                      the method chosen for the AR66 differs from the method used in the analysis of observations or model results, or both, and expert                                        Fifth Assessment Report (AR5). This choice is based on new research judgement by the author team or from a formal quantitative                                            on the visualization of uncertainty and on user surveys.
survey of expert views, or both).
Where there is sufficient scientific confidence, findings can also be formulated as statements of fact without uncertainty qualifiers.
Throughout IPCC reports, the calibrated language is clearly identified by being typeset in italics.
The context and progress in climate science (Section TS.1) is followed by a Cross-Section Box TS.1 on global surface temperature change.
Section TS.2 provides information about past and future large-scale changes in all components of the climate system. Section 1  In this Technical Summary, the following summary terms are used to describe the available evidence: limited, medium, or robust; and for the degree of agreement: low, medium, or high. A level of confidence is expressed using five qualifiers: very low, low, medium, high, and very high, and typeset in italics, e.g., medium confidence. For a given evidence and agreement statement, different confidence levels can be assigned, but increasing levels of evidence and degrees of agreement are correlated with increasing confidence (see Chapter 1, Box 1.1 for more details).
2  In this Technical Summary, the following terms are used to indicate the assessed likelihood of an outcome or a result: virtually certain 99-100% probability, very likely 90-100%, likely 66-100%,
about as likely as not 33-66%, unlikely 0-33%, very unlikely 0-10%, exceptionally unlikely 0-1%. Additional terms (extremely likely: 95-100%, more likely than not >50-100%, and extremely unlikely 0-5%) may also be used when appropriate. Assessed likelihood is typeset in italics, e.g., very likely (see Chapter 1, Box 1.1 for more details). Throughout the WGI report and unless stated otherwise, uncertainty is quantified using 90% uncertainty intervals. The 90% uncertainty interval, reported in square brackets [x to y], is estimated to have a 90% likelihood of covering the value that is being estimated. The range encompasses the median value, and there is an estimated 10% combined likelihood of the value being below the lower end of the range (x) and above its upper end (y). Often, the distribution will be considered symmetric about the corresponding best estimate, but this is not always the case. In this Report, an assessed 90% uncertainty interval is referred to as a very likely range. Similarly, an assessed 66% uncertainty interval is referred to as a likely range.
3  The regional traceback matrices that provide the location of the assessment findings synthesized in Section TS.4 are in the Supplementary Material (SM) of Chapter 10.
4  Data archive is available at https://catalogue.ceda.ac.uk/uuid/3234e9111d4f4354af00c3aaecd879b7.
5  https://interactive-atlas.ipcc.ch/
6  The AR6 figures use one of the following approaches. For observations, the absence of x symbols shows areas with statistical significance, while the presence of x indicates non-significance.
For model projections, the method offers two approaches with varying complexity. In the simple approach, high agreement (80%) is indicated with no overlay, and diagonal lines (///) show low agreement (<80%); In the advanced approach, areas with no overlay display robust signal (66% of models show change greater than the variability threshold and 80% of all models agree on the sign of change), reverse diagonal lines (\\\) show no robust signal, and crossed lines show conflicting signals (i.e., significant change but low agreement). Cross-Chapter Box Atlas.1 provides more information on the AR6 method for visualizing robustness and uncertainty on maps.
38
 
Technical Summary Box TS.1 l Core Concepts Central to This Report This box provides short descriptions of key concepts that are relevant to the AR6 WGI assessment, with a focus on their use in the Technical Summary and the Summary for Policymakers. The Glossary (Annex VII) includes more information on these concepts along with definitions of many other important terms and concepts used in this Report.
Characteristics of Climate Change Assessment Global warming: Global warming refers to the change of global surface temperature relative to a baseline depending upon the application. Specific global warming levels, such as 1.5&deg;C, 2&deg;C, 3&deg;C or 4&deg;C, are defined as changes in global surface temperature relative to the years 1850-1900 as the baseline (the earliest period of reliable observations with sufficient geographic coverage). They are used to assess and communicate information about global and regional changes, linking to scenarios and used as a common basis for Working Group II (WGII) and Working Group III (WGIII) assessments. (Section TS.1.3, Cross-Section Box TS.1) {1.4.1, 1.6.2, 4.6.1, Cross-Chapter Boxes 1.5, 2.3, 11.1, and 12.1, Atlas Sections 3-11, Glossary}                                                                TS Emergence: Emergence refers to the experience or appearance of novel conditions of a particular climate variable in a given region.
This concept is often expressed as the ratio of the change in a climate variable relative to the amplitude of natural variations of that variable (often termed a signal-to-noise ratio, with emergence occurring at a defined threshold of this ratio). Emergence can be expressed in terms of a time or a global warming level at which the novel conditions appear and can be estimated using observations or model simulations. (Sections TS.1.2.3 and TS.4.2) {1.4.2, FAQ 1.2, 7.5.5, 10.3, 10.4, 12.5.2, Cross-Chapter Box Atlas.1, Glossary}
Cumulative carbon dioxide (CO2) emissions: The total net amount of CO2 emitted into the atmosphere as a result of human activities. Given the nearly linear relationship between cumulative CO2 emissions and increases in global surface temperature, cumulative CO2 emissions are relevant for understanding how past and future CO2 emissions affect global surface temperature.
A related term - remaining carbon budget - is used to describe the total net amount of CO2 that could be released in the future by human activities while keeping global warming to a specific global warming level, such as 1.5&deg;C, taking into account the warming contribution from non-CO2 forcers as well. The remaining carbon budget is expressed from a recent specified date, while the total carbon budget is expressed starting from the pre-industrial period. (Sections TS.1.3 and TS.3.3) {1.6.3, 5.5, Glossary}
Net zero CO2 emissions: A condition that occurs when the amount of CO2 emitted into the atmosphere by human activities equals the amount of CO2 removed from the atmosphere by human activities over a specified period of time. Net negative CO2 emissions occur when anthropogenic removals exceed anthropogenic emissions. (Section TS.3.3) {Box 1.4, Glossary}
Human Influence on the Climate System Earths energy imbalance: In a stable climate, the amount of energy that Earth receives from the Sun is approximately in balance with the amount of energy that is lost to space in the form of reflected sunlight and thermal radiation. Climate drivers, such as an increase in greenhouse gases or aerosols, interfere with this balance, causing the system to either gain or lose energy. The strength of a climate driver is quantified by its effective radiative forcing (ERF), measured in W m-2. Positive ERF leads to warming, and negative ERF leads to cooling. That warming or cooling in turn can change the energy imbalance through many positive (amplifying) or negative (dampening) climate feedbacks. (Sections TS.2.2, TS.3.1 and TS.3.2) {2.2.8, 7.2, 7.3, 7.4, Box 7.1, Box 7.2, Glossary}
Attribution: Attribution is the process of evaluating the relative contributions of multiple causal factors to an observed change in climate variables (e.g., global surface temperature, global mean sea level), or to the occurrence of extreme weather or climate-related events. Attributed causal factors include human activities (such as increases in greenhouse gas concentration and aerosols, or land-use change) or natural external drivers (solar and volcanic influences), and in some cases internal variability. (Sections TS.1.2.4 and TS.2, Box TS.10) {Cross-Working Group Box: Attribution in Chapter 1; 3.5; 3.8; 10.4; 11.2.4; Glossary}
Committed change, long-term commitment: Changes in the climate system, resulting from past, present and future human activities, which will continue long into the future (centuries to millennia) even with strong reductions in greenhouse gas emissions.
Some aspects of the climate system, including the terrestrial biosphere, the deep ocean and the cryosphere, respond much more slowly than surface temperatures to changes in greenhouse gas concentrations. As a result, there are already substantial committed changes associated with past greenhouse gas emissions. For example, global mean sea level will continue to rise for thousands of years, even if future CO2 emissions are reduced to net zero and global warming halted, as excess energy due to past emissions continues to propagate into the deep ocean and as glaciers and ice sheets continue to melt. (Section TS.2.1, Box TS.4, Box TS.9) {1.2.1, 1.3, Box 1.2, Cross-Chapter Box 5.3}
39
 
Technical Summary Box TS.1 (continued)
Climate Information for Regional Climate Change and Risk Assessment Distillation: The process of synthesizing information about climate change from multiple lines of evidence obtained from a variety of sources, taking into account user context and values. It leads to an increase in the usability, usefulness and relevance of climate information, enhances stakeholder trust, and expands the foundation of evidence used in climate services. It is particularly relevant in the context of co-producing regional-scale climate information to support decision-making. (Section TS.4.1, Box TS.11) {10.1, 10.5, 12.6}
(Climate change) risk: The concept of risk is a key aspect of how the IPCC assesses and communicates to decision-makers about the potential for adverse consequences for human or ecological systems, recognizing the diversity of values and objectives associated with such systems. In the context of climate change, risks can arise from potential impacts of climate change as well as human responses to climate change. WGI contributes to the common IPCC risk framing through the assessment of relevant climate information, including TS      climatic impact-drivers and low-likelihood, high-impact outcomes. (Sections TS.1.4 and TS.4.1, Box TS.4) {Cross-Chapter Boxes 1.3 and 12.1, Glossary}
Climatic impact-drivers: Physical climate system conditions (e.g., means, events, extremes) that can be directly connected with having impacts on human or ecological systems are described as climatic impact-drivers (CIDs) without anticipating whether their impacts are detrimental (i.e., as for hazards in the context of climate change risks) or provide potential opportunities. A range of indices may capture the sector- or application-relevant characteristics of a climatic impact-driver and can reflect exceedances of identified tolerance thresholds. (Sections TS.1.4 and TS.4.3) {12.1-12.3, FAQ 12.1, Glossary}
Storylines: The term storyline is used both in connection to scenarios (related to a future trajectory of emissions or socio-economic developments) or to describe plausible trajectories of weather and climate conditions or events, especially those related to high levels of risk. Physical climate storylines are introduced in AR6 to explore uncertainties in climate change and natural climate variability, to develop and communicate integrated and context-relevant regional climate information, and to address issues with deep uncertainty7, including low-likelihood, high-impact outcomes. (Section TS.1.4, Box TS.3, Infographic TS.1) {1.4.4, Box 10.2, Glossary}
Low-likelihood, high impact outcomes: Outcomes/events whose probability of occurrence is low or not well known (as in the context of deep uncertainty) but whose potential impacts on society and ecosystems could be high. To better inform risk assessment and decision-making, such low-likelihood outcomes are considered if they are associated with very large consequences and may therefore constitute material risks, even though those consequences do not necessarily represent the most likely outcome. (Section TS.1.4, Box TS.3, Figure TS.6) {1.4.4, 4.8, Cross Chapter Box 1.3, Glossary}
As part of the AR6 cycle, the IPCC produced three Special Reports                                  The Report has been peer-reviewed by the scientific community in 2018 and 2019: the Special Report on Global Warming of                                          and governments (Annex X provides the Expert Reviewer list). The 1.5&deg;C (SR1.5), the Special Report on the Ocean and Cryosphere in                                    substantive introduction provided by Chapter 1 is followed by a first a Changing Climate (SROCC), and the Special Report on Climate                                      set of chapters dedicated to large-scale climate knowledge (Chapters Change and Land (SRCCL).                                                                            2-4), which encompasses observations and paleoclimate evidence, causes of observed changes, and projections; these are complemented The AR6 WGI Report provides a full and comprehensive assessment                                    by Chapter 11 for large-scale changes in extremes. The second set of of the physical science basis of climate change that builds on the                                  chapters (Chapters 5-9) is orientated around the understanding of previous assessments and these Special Reports and considers new                                    key climate system components and processes, including the global information and knowledge from the recent scientific literature8,                                  cycles of carbon, energy and water; short-lived climate forcers and their including longer observational datasets and new scenarios and                                      link to air quality; and the ocean, cryosphere and sea level change.
model results.                                                                                      The last set of chapters (Chapters 10-12 and the Atlas) is dedicated to the assessment and distillation of regional climate information from The structure of the AR6 WGI Report is designed to enhance the                                      multiple lines of evidence at sub-continental to local scales (including visibility of knowledge developments and to facilitate the integration                              urban climate), with a focus on recent and projected regional changes of multiple lines of evidence, thereby improving confidence in findings.                            in mean climate, extremes, and climatic impact-drivers. The new online 7    Although not a core concept of the WGI Report, deep uncertainty is used in the Technical Summary in the following sense: A situation of deep uncertainty exists when experts or stakeholders do not know or cannot agree on: (1) appropriate conceptual models that describe relationships among key driving forces in a system; (2) the probability distributions used to represent uncertainty about key variables and parameters; and/or (3) how to weigh and value desirable alternative outcomes (Lempert et al., 2003). Lempert, R. J., Popper, S. W., and Bankes, S. C. (2003). Shaping the next one hundred years: New methods for quantitative long-term strategy analysis (MR-1626-RPC). Santa Monica, CA: The RAND Pardee Center.
8    The assessment covers scientific literature accepted for publication by 31 January 2021.
40
 
Technical Summary Interactive Atlas allows users to interact in a flexible manner through
* Updated assessment of recent warming: The AR5 reported maps, time series and summary statistics with climate information for                                      a smaller rate of increase in global mean surface temperature over a set of updated WGI reference regions. The Report also includes 34                                        the period 1998-2012 than the rate calculated since 1951. Based Frequently Asked Questions and answers for the general public (https://                                    on updated observational datasets showing a larger trend over www.ipcc.ch/report/ar6/wg1/faqs).                                                                          1998-2012 than earlier estimates, there is now high confidence that the observed 1998-2012 global surface temperature trend Together, this Technical Summary and the underlying chapters aim at                                        is consistent with ensembles of climate model simulations, providing a comprehensive picture of knowledge progress since the                                          and there is now very high confidence that the slower rate of WGI contribution to AR5. Multiple lines of scientific evidence confirm                                      global surface temperature increase observed over this period that the climate is changing due to human influence. Important                                              was a temporary event induced by internal and naturally forced advances in the ability to understand past, present and possible                                            variability that partly offset the anthropogenic surface warming future changes should result in better-informed decision-making.                                            trend over this period, while heat uptake continued to increase in the ocean. Since 2012, strong warming has been observed, Some of the new results and main updates to key findings in this                                            with the past five years (2016-2020) being the hottest five-Report compared to AR5, SR1.5, SRCCL, and SROCC are summarized                                              year period in the instrumental record since at least 1850 (high                            TS below. Relevant Technical Summary sections with further details are                                        confidence). (Section TS.1.2, Cross-Section Box TS.1) shown in parentheses after each bullet point.
* Magnitude of climate system response: In this Report, it has been possible to reduce the long-standing uncertainty ranges for metrics that quantify the response of the climate system to Selected Updates and/or New Results since AR5                                                              radiative forcing, such as the equilibrium climate sensitivity (ECS) and the transient climate response (TCR), due to substantial
* Human influence9 on the climate system is now an                                                          advances (e.g., a 50% reduction in the uncertainty range of cloud established fact: The Fourth Assessment Report (AR4) stated                                            feedbacks) and improved integration of multiple lines of evidence, in 2007 that warming of the climate system is unequivocal, and                                        including paleoclimate information. Improved quantification AR5 stated in 2013 that human influence on the climate system                                          of ERF, the climate system radiative response, and the observed is clear. Combined evidence from across the climate system                                            energy increase in the Earth system over the past five decades strengthens this finding. It is unequivocal that the increase of                                        demonstrate improved consistency between independent CO2, methane (CH4) and nitrous oxide (N2O) in the atmosphere                                            estimates of climate drivers, the combined climate feedbacks, and over the industrial era is the result of human activities and that                                      the observed energy increase relative to AR5. (Section TS.3.2) human influence is the main driver10 of many changes observed
* Improved constraints on projections of future climate across the atmosphere, ocean, cryosphere and biosphere.                                                change: For the first time in an IPCC report, the assessed future (Sections TS.1.2, TS.2.1 and TS.3.1)                                                                    change in global surface temperature is consistently constructed
* Observed global warming to date: A combination of                                                        by combining scenario-based projections (which AR5 focused improved observational records and a series of very warm years                                          on) with observational constraints based on past simulations since AR5 have resulted in a substantial increase in the estimated                                      of warming as well as the updated assessment of ECS and TCR.
level of global warming to date. The contribution of changes in                                        In addition, initialized forecasts have been used for the period observational understanding alone between AR5 and AR6 leads                                            2019-2028. The inclusion of these lines of evidence reduces to an increase of about 0.1&deg;C in the estimated warming since                                            the assessed uncertainty for each scenario. (Section TS.1.3, 1850-1900. For the decade 2011-2020, the increase in global                                            Cross-Section Box TS.1) surface temperature since 1850-1900 is assessed to be 1.09
* Air quality: The AR5 assessed that projections of air quality
[0.95 to 1.20] &deg;C.11 Estimates of crossing times of global warming                                      are driven primarily by precursor emissions, including CH4. New levels and estimates of remaining carbon budgets are updated                                            scenarios explore a diversity of future options in air pollution accordingly. (Section TS.1.2, Cross-Section Box TS.1)                                                  management. The AR6 reports rapid recent shifts in the
* Paleoclimate evidence: The AR5 assessed that many of the                                                  geographical distribution of some of these precursor emissions, changes observed since the 1950s are unprecedented over                                                confirms the AR5 finding, and shows higher warming effects decades to millennia. Updated paleoclimate evidence strengthens                                        of short-lived climate forcers in scenarios with the highest air this assessment; over the past several decades, key indicators of                                      pollution. (Sections TS.1.3 and TS.2.2, Box TS.7) the climate system are increasingly at levels unseen in centuries
* Effects of short-lived climate forcers on global warming:
to millennia and are changing at rates unprecedented in at least                                        The AR5 assessed the radiative forcing for emitted compounds.
the last 2000 years. (Box TS.2, Section TS.2)                                                          The AR6 has extended this by assessing the emissions-based ERFs 9  Human influence on the climate system refers to human-driven activities that lead to changes in the climate system due to perturbations of Earths energy budget (also called anthropogenic forcing). Human influence results from emissions of greenhouse gases, aerosols and tropospheric ozone precursors, ozone-depleting substances, and land-use change.
10  Throughout this Technical Summary, main driver means responsible for more than 50% of the change.
11  Throughout the WGI report and unless stated otherwise, uncertainty is quantified using 90% uncertainty intervals. The 90% uncertainty interval, reported in square brackets [x to y], is estimated to have a 90% likelihood of covering the value that is being estimated. The range encompasses the median value and there is an estimated 10% combined likelihood of the value being below the lower end of the range (x) and above its upper end (y). Often the distribution will be considered symmetric about the corresponding best estimate, but this is not always the case. In this Report, an assessed 90% uncertainty interval is referred to as a very likely range. Similarly, an assessed 66% uncertainty interval is referred to as a likely range.
41
 
Technical Summary also accounting for aerosol-cloud interactions. The best estimates
* Effect of short-lived climate forcers on global warming of ERF attributed to sulphur dioxide (SO2) and CH4 emissions are                                        in coming decades: The SR1.5 stated that reductions in substantially greater than in AR5, while that of black carbon is                                        emissions of cooling aerosols partially offset greenhouse gas substantially reduced. The magnitude of uncertainty in the ERF                                          mitigation effects for two to three decades in pathways limiting due to black carbon emissions has also been reduced relative to                                          global warming to 1.5&deg;C. The AR6 assessment updates the AR5 AR5. (Section TS.3.1)                                                                                    assessment of the net cooling effect of aerosols and confirms
* Global water cycle: The AR5 assessed that anthropogenic                                                  that changes in short-lived climate forcers will very likely cause influences have likely affected the global water cycle since                                            further warming in the next two decades across all scenarios.
1960. The dedicated chapter in AR6 (Chapter 8) concludes with                                            (Section TS.1.3, Box TS.7) high confidence that human-caused climate change has driven
* COVID-19: Temporary emissions reductions in 2020 associated detectable changes in the global water cycle since the mid-20th                                          with COVID-19 containment led to small and positive net radiative century, with a better understanding of the response to aerosol                                          effect (warming influence). However, global and regional climate and greenhouse gas changes. The AR6 further projects with high                                          responses to this forcing are undetectable above internal climate confidence an increase in the variability of the water cycle in most                                    variability due to the temporary nature of emissions reductions.
TS    regions of the world and under all emissions scenarios. (Box TS.6)                                      (Section TS.3.3)
* Extreme events: The AR5 assessed that human influence had been detected in changes in some climate extremes.
A dedicated chapter in AR6 (Chapter 11) concludes that it is                                      Selected Updates and/or New Results Since AR5, SRCCL now an established fact that human-induced greenhouse gas                                          and SROCC emissions have led to an increased frequency and/or intensity of some weather and climate extremes since 1850, in particular
* Atmospheric concentration of methane: The SRCCL reported for temperature extremes. Evidence of observed changes and                                              a resumption of atmospheric CH4 concentration growth since attribution to human influence has strengthened for several types                                        2007. The AR6 reports a faster growth over 2014-2019 and of extremes since AR5, in particular for extreme precipitation,                                          assesses growth since 2007 to be largely driven by emissions droughts, tropical cyclones and compound extremes (including                                            from the fossil fuels and agriculture (dominated by livestock) fire weather). (Sections TS.1.2 and TS.2.1, Box TS.10)                                                  sectors. (Section TS.2.2)
* Land and ocean carbon sinks: The SRCCL assessed that the persistence of the land carbon sink is uncertain due to climate Selected Updates and/or New Results Since AR5 and SR1.5                                                    change. The AR6 finds that land and ocean carbon sinks are projected to continue to grow until 2100 with increasing
* Timing of crossing 1.5&deg;C global warming: Slightly different                                              atmospheric concentrations of CO2, but the fraction of emissions approaches are used in SR1.5 and in this Report. SR1.5 assessed                                          taken up by land and ocean is expected to decline as the CO2 a likely range of 2030 to 2052 for reaching a global warming                                            concentration increases, with a much larger uncertainty range for level of 1.5&deg;C (for a 30-year period), assuming a continued,                                            the land sink. The AR5, SR1.5 and SRCCL assessed carbon dioxide constant rate of warming. In AR6, combining the larger estimate                                          removal options and scenarios. The AR6 finds that the carbon of global warming to date and the assessed climate response to                                          cycle response is asymmetric for pulse emissions or removals, all considered scenarios, the central estimate of crossing 1.5&deg;C of                                      which means that CO2 emissions would be more effective at global warming (for a 20-year period) occurs in the early 2030s, in                                      raising atmospheric CO2 than CO2 removals are at lowering the early part of the likely range assessed in SR1.5, assuming no                                        atmospheric CO2. (Section TS.3.3, Box TS.5) major volcanic eruption. (Section TS.1.3, Cross-Section Box TS.1)
* Ocean stratification increase12 : Refined analyses of available
* Remaining carbon budgets: The AR5 had assessed the                                                        observations in the AR6 lead to a reassessment of the rate of transient climate response to cumulative emissions of CO2 to be                                          increase of the global stratification in the upper 200 m to be double likely in the range of 0.8&deg;C to 2.5&deg;C per 1000 GtC (1 Gigatonne                                          that estimated in SROCC from 1970 to 2018. (Section TS.2.4) of carbon, GtC, = 1 Petagram of carbon, PgC, = 3.664 Gigatonnes
* Projected ocean oxygen loss: Future subsurface oxygen of carbon dioxide, GtCO2), and this was also used in SR1.5. The                                          decline in new projections assessed in WGI AR6 is substantially assessment in AR6, based on multiple lines of evidence, leads to                                        greater in 2080-2099 than assessed in SROCC. (Section TS.2.4) a narrower likely range of 1.0&deg;C-2.3&deg;C per 1000 GtC. This has
* Ice loss from glaciers and ice sheets: Since SROCC, globally been incorporated in updated estimates of remaining carbon                                              resolved glacier changes have improved estimates of glacier mass budgets (see Section TS.3.3.1), together with methodological                                            loss over the past 20 years, and estimates of the Greenland and improvements and recent observations. (Sections TS.1.3                                                  Antarctic Ice Sheet loss have been extended to 2020. (Section TS.2.5) and TS.3.3)
* Observed global mean sea level change: new observation-based estimates published since SROCC lead to an assessed sea level rise estimate from 1901 to 2018 that is now consistent with the sum of individual components and consistent with closure of the global energy budget. (Box TS.4) 12 Increased stratification reduces the vertical exchange of heat, salinity, oxygen, carbon and nutrients. Stratification is an important indicator for ocean circulation.
42
 
Technical Summary
* Projected global mean sea level change: The AR6 projections                                      TS.1            A Changing Climate of global mean sea level are based on projections from ocean thermal expansion and land ice contribution estimates, which are                                This section introduces the assessment of the physical science basis of consistent with the assessed ECS and assessed changes in global                                  climate change in the AR6 and presents the climate context in which surface temperature. They are underpinned by new land ice model                                  this assessment takes place, recent progress in climate science and intercomparisons and consideration of processes associated with                                  the relevance of global and regional climate information for impact low confidence to characterize the deep uncertainty in future ice                                and risk assessments. The future emissions scenarios and global loss from Antarctica. The AR6 projections based on new models                                    warming levels, used to integrate assessments across this Report, and methods are broadly consistent with SROCC findings.                                          are introduced and their applications for future climate projections (Box TS.4)                                                                                      are briefly addressed. Paleoclimate science provides a long-term context for observed climate change of the past 150 years and the projected changes in the 21st century and beyond (Box TS.2). The assessment of past, current and future global surface temperature changes relative to the standard baselines and reference periods13 used throughout this Report is summarized in Cross-Section Box TS.1.                              TS TS1.1            Context of a Changing Climate This Report assesses new scientific evidence relevant for a world whose climate system is rapidly changing, overwhelmingly due to human influence. The five IPCC assessment cycles since 1990 have comprehensively and consistently laid out the rapidly accumulating evidence of a changing climate system, with the Fourth Assessment Report in 2007 being the first to conclude that warming of the climate system is unequivocal. Sustained changes have been documented in all major elements of the climate system: the atmosphere, land, cryosphere, biosphere and ocean (Section TS.2). Multiple lines of evidence indicate the recent large-scale climatic changes are unprecedented in a multi-millennial context and that they represent a millennial-scale commitment for the slow-responding elements of the climate system, resulting in continued worldwide loss of ice, increase in ocean heat content, sea level rise and deep ocean acidification (Box TS.2; Section TS.2). {1.2.1, 1.3, Box 1.2, 2.2, 2.3, Figure 2.34, 5.1, 5.3, 9.2, 9.4-9.6, Appendix 1.A}
Earths climate system has evolved over many millions of years, and evidence from natural archives provides a long-term perspective on observed changes and projected changes over the coming centuries.
These reconstructions of past climate also show that atmospheric CO2 concentrations and global surface temperature are strongly coupled (Figure TS.1), based on evidence from a variety of proxy records over multiple time scales (Box TS.2, Section TS.2). Levels of global warming (see Core Concepts Box) that have not been seen in millions of years could be reached by 2300, depending on the emissions pathway that is followed (Section TS.1.3). For example, there is medium confidence that, by 2300, an intermediate scenario14 used in this Report leads to global surface temperatures of [2.3&deg;C to 4.6&deg;C] higher than 1850-1900, similar to the mid-Pliocene Warm 13 Several baselines or reference periods are used consistently throughout this Report. Baseline refers to a period against which anomalies (i.e., differences from the average value for the baseline period) are calculated. Examples include the 1750 baseline (used for anthropogenic radiative forcings), the 1850-1900 baseline (an approximation for pre-industrial global surface temperature from which global warming levels are calculated) and the 1995-2014 baseline (used for many climate model projections). A reference period indicates a time period over which various statistics are calculated (e.g., the near-term reference period, 2021-2040). Paleo reference periods are listed in Box TS.2. {1.4.1, Cross-Chapter Boxes 1.2 and 2.1}
14 Please refer to Section TS.1.3.1 for an overview of the climate change scenarios used in this Report.
43
 
Technical Summary Atmospheric CO2 concentration and global surface temperature change Atmospheric CO2 concentration (ppm) during the last 60 million years and projections for the next 300 years          SSP5-8.5 2000                                                                                                                              SSP5-8.5 1000 SSP2-4.5                                2300 400 SSP1-2.6 200 2100 early Eocene                mid-Pliocene                    2020 TS                                                                                                                                                                              SSP1-2.6 Global surface temperature change (&deg;C) 15 SSP5-8.5                                2300 10 (relative to 1850-1900) 5                                                                                            SSP2-4.5 0                                                                                            SSP1-2.6                                2100
                                                              -5 0 2 4 6 8 10 12 14 16 18 60 50 40 30 20 10      9    7  5  3    1 800 600 400 200        0 1850      2000  2150    2300 Temperature (&deg;C)
Millions of years                Thousands of years                Year CE                  relative to 1850-1900 Figure TS.1 l Changes in atmospheric CO2 and global surface temperature (relative to 1850-1900) from the deep past to the next 300 years. The intent of this figure is to show that CO2 and temperature covary, both in the past and into the future, and that projected CO2 and temperatures are similar to those only from many millions of years ago. CO2 concentrations from millions of years ago are reconstructed from multiple proxy records (grey dots are data from Section 2.2.3.1, Figure 2.3 shown with cubic-spline fit). CO2 levels for the last 800,000 years through the mid-20th century are from air trapped in polar ice; recent values are from direct air measurements. Global surface temperature prior to 1850 is estimated from marine oxygen isotopes, one of multiple sources of evidence used to assess paleo temperatures in this Report. Temperature of the past 170 years is the AR6 assessed mean. CO2 levels and global surface temperature change for the future are shown for three Shared Socio-economic Pathway (SSP) scenarios through 2300 CE, using Earth system model emulators calibrated to the assessed global surface temperatures. Their smooth trajectories do not account for inter-annual to inter-decadal variability, including transient response to potential volcanic eruptions. Global maps for two paleo reference periods are based on Coupled Model Intercomparison Project Phase 6 (CMIP6) and pre-CMIP6 multi-model means, with site-level proxy data for comparison (squares and circles are marine and terrestrial, respectively). The map for 2020 is an estimate of the total observed warming since 1850-1900. Global maps at right show two SSP scenarios at 2100 (2081-2100) and at 2300 (2281-2300; map from CMIP6 models; temperature assessed in 4.7.1). A brief account of the major climate forcings associated with past global temperature changes is in Cross-Chapter Box 2.1.
(Section TS.1.3, Figure TS.9, Cross-Section Box TS.1, Box TS.2) {1.2.1.2; Figures 1.14 and 1.5; 2.2.3; 2.3.1.1; 2.3.1.1.1; Figures 2.4 and 2.5; Cross-Chapter Box 2.1, Figure 1; 4.5.1; 4.7.1; Cross-Chapter Box 4.1; Cross-Chapter Box 7.1; Figure 7.13}
Period [2.5&deg;C to 4&deg;C], about 3.2 million years ago, whereas the high                                                  been studied systematically since the early 20th century. Other CO2 emissions scenario SSP5-8.5 leads to temperatures of [6.6&deg;C                                                      major anthropogenic drivers, such as atmospheric aerosols (fine to 14.1&deg;C] by 2300, which overlaps with the Early Eocene Climate                                                      solid particles or liquid droplets), land-use change and non-CO2 Optimum [10&deg;C to 18&deg;C], about 50 million years ago. {Cross-Chapter                                                    greenhouse gases, were identified by the 1970s. Since systematic Boxes 2.1 and 2.4, 2.3.1, 4.3.1.1, 4.7.1.2, 7.4.4.1}                                                                  scientific assessments began in the 1970s, the influence of human activities on the warming of the climate system has evolved from Understanding of the climate systems fundamental elements is                                                        theory to established fact (see also Section TS.2). The evidence for robust and well established. Scientists in the 19th century identified                                                human influence on recent climate change strengthened from the the major natural factors influencing the climate system. They                                                        IPCC First Assessment Report in 1990 to the IPCC Fifth Assessment also hypothesized the potential for anthropogenic climate change                                                      Report in 2013/14, and is now even stronger in this assessment due to CO2 emitted by combustion of fossil fuels (petroleum,                                                          (Sections TS.1.2.4 and TS.2). Changes across a greater number of coal, natural gas). The principal natural drivers of climate change,                                                  climate system components, including changes in regional climate including changes in incoming solar radiation, volcanic activity,                                                    and extremes can now be attributed to human influence (see orbital cycles and changes in global biogeochemical cycles, have                                                      Sections TS.2 and TS.4). {1.3.1-1.3.5, 3.1, 11.2, 11.9}
44
 
Technical Summary Box TS.2 l Paleoclimate Paleoclimate evidence is integrated within multiple lines of evidence across the WGI Report to more fully understand the climate system. Paleo evidence extends instrument-based observations of climate variables and climate drivers back in time, providing the long-term context needed to gauge the extent to which recent and potential future changes are unusual (Section TS.2, Figure TS.1). Pre-industrial climate states complement evidence from climate model projections by providing real-world examples of climate characteristics for past global warming levels, with empirical evidence for how the slow-responding components of the climate system operate over centuries to millennia - the time scale for committed climate change (Core Concepts Box, Box TS.4, Box TS.9). Information about the state of the climate system during well-described paleoclimate reference periods helps narrow the uncertainty range in the overall assessment of Earths sensitivity to climate forcing (Section TS.3.2.1). {Cross-Chapter Box 2.1, FAQ 1.3, FAQ 2.1}
Paleoclimate reference periods. Over the long evolution of Earths climate, several periods have received extensive research attention as examples of distinct climate states and rapid climate transitions (Box TS.2, Figure 1). These paleoclimate reference periods                                                            TS represent the present geological era (Cenozoic; past 65 million years) and are used across chapters to help structure the assessment of climate changes prior to industrialization. Cross-Chapter Box 2.1 describes the reference periods, along with a brief account of their climate forcings, and lists where each is discussed in other chapters. Cross-Chapter Box 2.4 summarizes information on one of the reference periods, the mid-Pliocene Warm Period. The Interactive Atlas includes model output from the World Climate Research Programme Coupled Model Intercomparison Project Phase 6 (CMIP6) for four of the paleoclimate reference periods.
Three selected global climate indicators covary across multiple paleoclimate reference periods (a)                                                                                      (b)
Reference period                                                      CO2        Temperature Sea level
(*See Interactive Atlas for climate model output)        Age        (ppm)          (&deg;C)      (m) 2000 Recent past                                        1995-2014 CE  360    397    0.66 to 1.00    0.15 to 0.25                                                      Early Eocene Atmospheric CO2 (ppm)
Approximate pre-industrial                          1850-1900 CE  286    296    -0.15 to +0.11  -0.03 to 0.00 1000 Last Millennium                                    850-1850 CE    278 to 285    -0.14 ~ 0.24    -0.05 to 0.03 Mid-Holocene*                                      6.5-5.5 ka    260 to 268      0.2 to 1.0      -3.5 to +0.5 Last Deglacial Transition                                                                                                                  500 18-11 ka      193    271    not assessed    -120    -50                                    Recent past        Mid-Pliocene Last Glacial Maximum*                              23-19 ka      188 to 194      -5 to -7      -134 to -125                                    1850-1900 Last Interglacial Last Interglacial*                                  129-116 ka    266 to 282      0.5 to 1.5        5 to 10 200              Mid-Holocene Mid-Pliocene Warm Period*                          3.3-3.0 ka    360 to 420      2.5 to 4.0        5 to 25 Last Glacial Maximum Early Eocene                                        53-49 Ma      1150 to 2500    10 to 18        70 to 76 Paleocene-Eocene Thermal Maximum                    55.9-55.7 Ma  900    2000    10 to 25        not assessed                            100
                                                                                                                                                -10      -5      0      5    10    15    20 X to Y: very likely range (caveats in Figure 2.34)                      lower      1850-1900      higher X Y: start to end of period, with no stated uncertainty                                                                                                Global surface temperature (&deg;C) relative to 1850-1900 X ~ Y: lowest and highest values, with not stated uncertainty Box TS.2, Figure 1 l Paleoclimate and recent reference periods, with selected key indicators. The intent of this figure is to list the paleoclimate reference periods used in this Report, to summarize three key global climate indicators, and compare CO2 with global temperature over multiple periods. (a) Three large-scale climate indicators (atmospheric CO2, global surface temperature relative to 1850-1900, and global mean sea level relative to 1900), based on assessments in Chapter 2, with confidence levels ranging from low to very high. (b) Comparison between global surface temperature (relative to 1850-1900) and atmospheric CO2 concentration (shown on a log scale) for multiple reference periods (mid-points with 5-95% ranges). {2.2.3, 2.3.1.1, 2.3.3.3, Figure 2.34}
Paleoclimate models and reconstructions. Climate models that target paleoclimate reference periods have been featured by the IPCC since the First Assessment Report. Under the framework of CMIP6-PMIP4 (Paleoclimate Modelling Intercomparison Project),
new protocols for model intercomparisons have been developed for multiple paleoclimate reference periods. These modelling efforts have led to improved understanding of the climate response to different external forcings, including changes in Earths orbital and plate movements, solar irradiance, volcanism, ice-sheet size and atmospheric greenhouse gases. Likewise, quantitative reconstructions of climate variables from proxy records that are compared with paleoclimate simulations have improved as the number of study sites and variety of proxy types have expanded, and as records have been compiled into new regional and global datasets. {1.3.2, 1.5.1, Cross-Chapter Boxes 2.1 and 2.4}
Global surface temperature. Since AR5, updated climate forcings, improved models, new understanding of the strengths and weaknesses of a growing array of proxy records, better chronologies and more robust proxy data products have led to better agreement between models and reconstructions. For global surface temperature, the mid-point of the AR6-assessed range and the median of the model-simulated temperatures differ by an average of 0.5&deg;C across five reference periods; they overlap within their 45
 
Technical Summary Box TS.2 (continued) 90% ranges in four of five cases, which together span from about 6 [5 to 7]&deg;C colder during the Last Glacial Maximum to about 14 [10 to 18] &deg;C warmer during the Early Eocene, relative to 1850-1900 (Box TS.2, Figure 2a). Changes in temperature by latitude in response to multiple forcings show that polar amplification (stronger warming at high latitudes than the global average) is a prominent feature of the climate system across multiple climate states, and the ability of models to simulate this polar amplification in past warm climates has improved since AR5 (high confidence). Over the past millennium, and especially since about 1300 CE, simulated global surface temperature anomalies are well within the uncertainty of reconstructions (medium confidence), except for some short periods immediately following large volcanic eruptions, for which different forcing datasets disagree (Box TS.2, Figure 2b). {2.3.1.1, 3.3.3.1, 3.8.2.1, 7.4.4.1.2}
Proxy-based and model-simulated estimates of global surface temperature agree across multiple reference periods TS                                                (a)                                                                                    (b) 20 Global surface temperature relative to                                            1.0        Observed          Reconstructed        Simulated 1850-1900                                                                          0.8 15 Early Eocene Simulated temperature (&deg;C) 0.6 Global surface temperature 10 0.4 5                                                                                    0.2 mid-Pliocene relative to 1850-1900 (&deg;C)
Recent past                                                                    0 0  1850-1900            Last Interglacial mid-Holocene                                              -0.2
                                                                                                                                  -0.4
                                              -5          Last Glacial Maximum
                                                                                                                                  -0.6
                                            -10
                                                -10      -5        0      5      10      15  20                                            1000      1200          1400        1600          1800  2000 Reconstructed temperature (&deg;C)                                                                                  Year (CE)
Box TS.2, Figure 2 l Global surface temperature as estimated from proxy records (reconstructed) and climate models (simulated). The intent of this figure is to show the agreement between observations and models of global temperatures during paleo reference periods. (a) For individual paleoclimate reference periods. (b) For the last millennium, with instrumental temperature (AR6 assessed mean, 10-year smoothed). Model uncertainties in (a) and (b) are 5-95% ranges of multi-model ensemble means; reconstructed uncertainties are 5-95% ranges (medium confidence) of (a) midpoints and (b) multi-method ensemble median. {2.3.1.1, Figure 2.34, Figure 3.2c, Figure 3.44}
Equilibrium climate sensitivity. Paleoclimate data provide evidence to estimate equilibrium climate sensitivity (ECS15) (Section TS.3.2.1). In AR6, refinements in paleo data for paleoclimate reference periods indicate that ECS is very likely greater than 1.5&deg;C and likely less than 4.5&deg;C, which is largely consistent with other lines of evidence and helps narrow the uncertainty range of the overall assessment of ECS. Some of the CMIP6 climate models that have either high (>5&deg;C) or low (<2&deg;C) ECS also simulate past global surface temperature changes outside the range of proxy-based reconstructions for the coldest and warmest reference periods. Since AR5, independent lines of evidence, including proxy records from past warm periods and glacial-interglacial cycles, indicate that sensitivity to forcing increases as temperature increases (Section TS.3.2.2). {7.4.3.2, 7.5.3, 7.5.6, Table 7.11}
Water cycle. New hydroclimate reconstructions and model-data comparisons have improved the understanding of the causes and effects of long-term changes in atmospheric and ocean circulation, including monsoon variability and modes of variability (Box TS.13, Section TS.4.2). Climate models are able to reproduce decadal drought variability on large regional scales, including the severity, persistence and spatial extent of past megadroughts known from proxy records (medium confidence). Some long-standing discrepancies remain, however, such as the magnitude of African monsoon precipitation during the early Holocene (the past 11,700 years), suggesting continuing knowledge gaps. Paleoclimate evidence shows that, in relatively high CO2 climates such as the Pliocene, Walker circulation over the equatorial Pacific Ocean weakens, supporting the high confidence model projections of weakened Walker cells by the end of the 21st century. {3.3.2, 8.3.1.6, 8.4.1.6, 8.5.2.1, 9.2}
15  In this Report, equilibrium climate sensitivity is defined as the equilibrium (steady state) change in the surface temperature following a doubling of the atmospheric carbon dioxide (CO2) concentration from pre-industrial conditions.
46
 
Technical Summary Box TS.2 (continued)
Sea level and ice sheets. Although past and future global warming differ in their forcings, evidence from paleoclimate records and modelling show that ice-sheet mass and global mean sea level (GMSL) responded dynamically over multiple millennia (high confidence). This evidence helps to constrain estimates of the committed GMSL response to global warming (Box TS.4). For example, under a past global warming levels of around [2.5&deg;C to 4&deg;C] relative to 1850-1900, like during the mid-Pliocene Warm Period, sea level was [5 to 25 m] higher than 1900 (medium confidence); under past global warming levels of [10&deg;C to 18&deg;C], like during the Early Eocene, the planet was essentially ice free (high confidence). Constraints from these past warm periods, combined with physical understanding, glaciology and modelling, indicate a committed long-term GMSL rise over 10,000 years, reaching about 8 to 13 m for sustained peak global warming of 2&deg;C and up to 28 to 37 m for 5&deg;C, which exceeds the AR5 estimate. {2.3.3.3, 9.4.1.4, 9.4.2.6, 9.6.2, 9.6.3.5}
Ocean. Since AR5, better integration of paleo-oceanographic data with modelling along with higher-resolution analyses of transient changes have improved understanding of long-term ocean processes. Low-latitude sea surface temperatures at the Last                TS Glacial Maximum cooled more than previously inferred, resolving some inconsistencies noted in AR5. This paleo context supports the assessment that ongoing increase in ocean heat content (OHC) represents a long-term commitment (see Core Concepts Box),
essentially irreversible on human time scales (high confidence). Estimates of past global OHC variations generally track those of sea surface temperatures around Antarctica, underscoring the importance of Southern Ocean processes in regulating deep-ocean temperatures. Paleoclimate data, along with other evidence of glacial-interglacial changes, show that Antarctic Circumpolar flow strengthened and that ventilation of Antarctic Bottom Water accelerated during warming intervals, facilitating release of CO2 stored in the deep ocean to the atmosphere. Paleo evidence suggests significant reduction of deep-ocean ventilation associated with meltwater input during times of peak warmth. {2.3.1.1, 2.3.3.1, 9.2.2, 9.2.3.2}
Carbon cycle. Past climate states were associated with substantial differences in the inventories of the various carbon reservoirs, including the atmosphere (Section TS.2.2). Since AR5, the quantification of carbon stocks has improved due to the development of novel sedimentary proxies and stable-isotope analyses of air trapped in polar ice. Terrestrial carbon storage decreased markedly during the Last Glacial Maximum by 300-600 PgC, possibly by 850 PgC when accounting for interactions with the lithosphere and ocean sediments, a larger reduction than previously estimated, owing to a colder and drier climate. At the same time, the storage of remineralized carbon in the ocean interior increased by as much as 750-950 PgC, sufficient to balance the removal of carbon from the atmosphere (200 PgC) and terrestrial biosphere reservoirs combined (high confidence). {5.1.2.2}
TS.1.2      Progress in Climate Science                                  satellite measurement techniques are now long enough to be relevant for climate assessments. For example, globally distributed, TS.1.2.1 Observation-based Products and their Assessments                high-vertical-resolution profiles of temperature and humidity in the upper troposphere and stratosphere can be obtained from the early Observational capabilities have continued to improve and              2000s using global navigation satellite systems, leading to updated expand overall since AR5, enabling improved consistency                estimates of recent atmospheric warming. Improved measurements between independent estimates of climate drivers, the                  of ocean heat content, warming of the land surface, ice-sheet mass combined climate feedbacks, and the observed energy                    loss and sea level changes allow a better closure of the global and sea level increase. Satellite climate records and                  energy and sea level budgets relative to AR5. For surface and improved reanalyses are used as an additional line of                  balloon-based networks, apparent regional data reductions result evidence for assessing changes at the global and regional              from a combination of data policy issues, data curation/provision scales. However, there have also been reductions in some              challenges, and real cessation of observations, and are to an extent observational data coverage or continuity and limited                  counter-balanced by improvements elsewhere. Limited observational access to data resulting from data policy issues. Natural              records of extreme events and spatial data gaps currently limit the archives of past climate, such as tropical glaciers, have also        assessment of some observed regional climate change. {1.5.1, 2.3.2, been subject to losses (in part due to anthropogenic climate          7.2.2, Box 7.2, Cross-Chapter Box 9.1, 9.6.1, 10.2.2, 10.6, 11.2, 12.4}
change). {1.5.1, 1.5.2, 10.2.2}
New paleoclimate reconstructions from natural archives have enabled Earth system observations are an essential driver of progress in our      more robust reconstructions of the spatial and temporal patterns of understanding of climate change. Overall, capabilities to observe        past climate changes over multiple time scales (Box TS.2). However, the physical climate system have continued to improve and expand.        paleoclimate archives, such as tropical glaciers and modern natural Improvements are particularly evident in ocean observing networks        archives used for calibration (e.g., corals and trees), are rapidly and remote-sensing systems. Records from several recently instigated      disappearing owing to a host of pressures, including increasing 47
 
Technical Summary temperatures (high confidence). Substantial quantities of past            Section Box TS.1). Increasing horizontal resolution in global instrumental observations of weather and other climate variables,          climate models improves the representation of small-scale over both land and ocean, which could fill gaps in existing datasets,      features and the statistics of daily precipitation (high remain un-digitized or inaccessible. These include measurements of        confidence). Earth system models, which include additional temperature (air and sea surface), rainfall, surface pressure, wind        biogeochemical feedbacks, often perform as well as their strength and direction, sunshine amount and many other variables          lower-complexity global climate model counterparts, which dating back into the 19th century. {1.5.1}                                do not account for these additional feedbacks (medium confidence). {1.3.6, 1.5.3, 3.1, 3.5.1, 3.8.2, 4.3.1, 4.3.4, 7.5, Reanalyses combine observations and models (e.g., a numerical              8.5.1, 9.6.3.1}
weather prediction model) using data assimilation techniques to provide a spatially complete, dynamically consistent estimate of        Climate model simulations coordinated and collected as part of the multiple variables describing the evolving climate state. Since AR5,    World Climate Research Programmes Coupled Model Intercomparison new reanalyses have been developed for the atmosphere and the          Project Phase 6 (CMIP6), complemented by a range of results from ocean with various combinations of increased resolution, extended      the previous phase (CMIP5), constitute a key line of evidence TS records, more consistent data assimilation and larger availability      supporting this Report. The latest generation of CMIP6 models have of uncertainty estimates. Limitations remain, for example, in how      an improved representation of physical processes relative to previous reanalyses represent global-scale changes to the water cycle. Regional  generations, and a wider range of Earth system models now represent reanalyses use high-resolution, limited-area models constrained by      biogeochemical cycles. Higher-resolution models that better capture regional observations and with boundary conditions from global          smaller-scale processes are also increasingly becoming available reanalyses. There is high confidence that regional reanalyses better    for climate change research (Figure TS.2, Panels a and b). Results represent the frequencies of extremes and variability in precipitation, from coordinated regional climate modelling initiatives, such as the surface air temperature and surface wind than global reanalyses        Coordinated Regional Climate Downscaling Experiment (CORDEX) and provide estimates that are more consistent with independent        complement and add value to the CMIP global models, particularly in observations than dynamical downscaling approaches. {1.5.2,            complex topography zones, coastal areas and small islands, as well 10.2.1.2, Annex I}                                                      as for extremes. {1.5.3, 1.5.4, 2.8.2, FAQ 3.3, 6.2.2, 6.4, 6.4.5, 8.5.1, 10.3.3, Atlas.1.4}
TS.1.2.2 Climate Model Performance Projections of the increase in global surface temperature and the This report assesses results from climate models                    pattern of warming from previous IPCC Assessment Reports and other participating in the Coupled Model Intercomparison Project          studies are broadly consistent with subsequent observations (limited Phase 6 (CMIP6) of the World Climate Research Programme.            evidence, high agreement), especially when accounting for the These models include new and better representation of                difference in radiative forcing scenarios used for making projections physical, chemical and biological processes, as well as              and the radiative forcings that actually occurred (Figure TS.3). The higher resolution, compared to climate models considered            AR5 and SROCC projections of GMSL for the 2007-2018 period have in previous IPCC Assessment Reports. This has improved              been shown to be consistent with observed trends in GMSL and the simulation of the recent mean state of most large-scale          regional weighted mean tide gauges. {1.3.6, 9.6.3.1}
indicators of climate change and many other aspects across the climate system. Some differences from observations              For most large-scale indicators of climate change, the simulated remain, for example in regional precipitation patterns.              recent mean climate from CMIP6 models underpinning this Projections of the increase in global surface temperature,          assessment have improved compared to the CMIP5 models used in the pattern of warming, and global mean sea level rise              AR5 (high confidence). This is evident from the performance of 18 from previous IPCC Assessment Reports and other studies              simulated atmospheric and land large-scale indicators of climate are broadly consistent with subsequent observations,                change between the three generations of models (CMIP3, CMIP5, especially when accounting for the difference in radiative          and CMIP6) when benchmarked against reanalysis and observational forcing scenarios used for making projections and the                data (Figure TS.2, Panel c). Earth system models, characterized by radiative forcings that actually occurred.                          additional biogeochemical feedbacks, often perform at least as well as related, more constrained, lower-complexity models lacking these The CMIP6 historical simulations assessed in this report            feedbacks (medium confidence). {3.8.2, 10.3.3.3}
have an ensemble mean global surface temperature change within 0.2&deg;C of the observations over most of the            The CMIP6 multi-model mean global surface temperature change historical period, and observed warming is within the very          from 1850-1900 to 2010-2019 is close to the best estimate of likely range of the CMIP6 ensemble. However, some CMIP6              the observed warming. However, some CMIP6 models simulate models simulate a warming that is either above or below              a warming that is below or above the assessed very likely range. The the assessed very likely range of observed warming. The              CMIP6 models also reproduce surface temperature variations over information about how well models simulate past warming,            the past millennium, including the cooling that follows periods of as well as other insights from observations and theory, are          intense volcanism (medium confidence). For upper air temperature, used to assess projections of global warming (see Cross-            an overestimation of the upper tropical troposphere warming by 48
 
Technical Summary about 0.1&deg;C per decade between 1979 and 2014 persists in most                                      evolution of the satellite-observed Arctic sea ice loss (high confidence).
CMIP5 and CMIP6 models (medium confidence), whereas the                                            The ability to model ice-sheet processes has improved substantially differences between simulated and improved satellite-derived                                        since AR5. As a consequence, there is medium confidence in the estimates of change in global mean temperature through the depth                                    representation of key processes related to surface-mass balance and of the stratosphere have decreased. {3.3.1}                                                        retreat of the grounding-line (the junction between a grounded ice sheet and an ice shelf, where the ice starts to float) in the absence Some CMIP6 models demonstrate an improvement in how clouds                                          of instabilities. However, there remains low confidence in simulations are represented. CMIP5 models commonly displayed a negative                                        of ice-sheet instabilities, ice-shelf disintegration and basal melting shortwave cloud radiative effect that was too weak in the present                                  owing to their high sensitivity to both uncertain oceanic forcing and climate. These errors have been reduced, especially over the                                        uncertain boundary conditions and parameters. {1.5.3, 2.3.2, 3.4.1, Southern Ocean, due to a more realistic simulation of supercooled                                  3.4.2, 3.8.2, 9.3.1, 9.3.2, 9.4.1, 9.4.2}
liquid droplets with sufficient numbers and an associated increase in the cloud optical depth. Because a negative cloud optical depth                                  CMIP6 models are able to reproduce most aspects of the spatial feedback in response to surface warming results from brightening                                  structure and variance of the El Nino-Southern Oscillation (ENSO) of clouds via active phase change from ice to liquid cloud particles                                and Indian Ocean Basin and Dipole modes of variability (medium                                TS (increasing their shortwave cloud radiative effect), the extratropical                              confidence). However, despite a slight improvement in CMIP6, some cloud shortwave feedback in CMIP6 models tends to be less negative,                                underlying processes are still poorly represented. Models reproduce leading to a better agreement with observational estimates (medium                                  observed spatial features and variance of the Southern Annular confidence). CMIP6 models generally represent more processes                                        Mode (SAM) and Northern Annular Mode (NAM) very well (high that drive aerosol-cloud interactions than the previous generation                                  confidence). The summertime SAM trend is well captured, with of climate models, but there is only medium confidence that those                                  CMIP6 models outperforming CMIP5 models (medium confidence).
enhancements improve their fitness-for-purpose of simulating                                        By contrast, the cause of the NAM trend towards its positive phase radiative forcing of aerosol-cloud interactions. {6.4, 7.4.2, FAQ 7.2}                              is not well understood. In the Tropical Atlantic basin, which contains the Atlantic Zonal and Meridional modes, major biases in modelled CMIP6 models still have deficiencies in simulating precipitation                                    mean state and variability remain. Model performance is limited in patterns, particularly in the tropical ocean. Increasing horizontal                                reproducing sea surface temperature anomalies for decadal modes resolution in global climate models improves the representation of                                  of variability, despite improvements from CMIP5 to CMIP6 (medium small-scale features and the statistics of daily precipitation (high                                confidence) (see also Section TS.1.4.2.2, Table TS.4). {3.7.3-3.7.7}
confidence). There is high confidence that high-resolution global, regional and hydrological models provide a better representation of                                Earth system models (ESMs) simulate globally averaged land land surfaces, including topography, vegetation and land-use change,                                carbon sinks within the range of observation-based estimates which can improve the accuracy of simulations of regional changes in                                (high confidence), but global-scale agreement masks large regional the terrestrial water cycle. {3.3.2, 8.5.1, 10.3.3, 11.2.3}                                        disagreements. There is also high confidence that the ESMs simulate the weakening of the global net flux of CO2 into the ocean during There is high confidence that climate models can reproduce the recent                              the 1990s, as well as the strengthening of the flux from 2000. {3.6}
observed mean state and overall warming of temperature extremes globally and in most regions, although the magnitude of the trends                                  Two important quantities used to estimate how the climate system may differ. There is high confidence in the ability of models to capture                            responds to changes in greenhouse gas (GHG) concentrations are the the large-scale spatial distribution of precipitation extremes over land.                          equilibrium climate sensitivity (ECS) and transient climate response The overall performance of CMIP6 models in simulating the intensity                                (TCR16). The CMIP6 ensemble has broader ranges of ECS and TCR and frequency of extreme precipitation is similar to that of CMIP5                                  values than CMIP5 (see Section TS.3.2 for the assessed range). These models (high confidence). {Cross-Chapter Box 3.2, 11.3.3, 11.4.3}                                  higher sensitivity values can, in some models, be traced to changes in extratropical cloud feedbacks (medium confidence). To combine The structure and magnitude of multi-model mean ocean temperature                                  evidence from CMIP6 models and independent assessments of ECS biases have not changed substantially between CMIP5 and CMIP6                                      and TCR, various emulators are used throughout the report. Emulators (medium confidence). Since AR5, there is improved consistency                                      are a broad class of simple climate models or statistical methods between recent observed estimates and model simulations of changes                                  that reproduce the behaviour of complex ESMs to represent key in upper (<700 m) ocean heat content. The mean zonal and overturning                                characteristics of the climate system, such as global surface temperature circulations of the Southern Ocean and the mean overturning                                        and sea level projections. The main application of emulators in AR6 circulation of the North Atlantic (AMOC) are broadly reproduced by                                  is to extrapolate insights from ESMs and observational constraints to CMIP5 and CMIP6 models. {3.5.1, 3.5.4, 9.2.3, 9.3.2, 9.4.2}                                        produce projections from a larger set of emissions scenarios, which is achieved due to their computational efficiency. These emulated CMIP6 models better simulate the sensitivity of Arctic sea ice area                                projections are also used for scenario classification in WGIII. {Box 4.1, to anthropogenic CO2 emissions, and thus better capture the time                                    4.3.4, 7.4.2, 7.5.6, Cross-Chapter Box 7.1, FAQ 7.2}
16  In this Report, transient climate response is defined as the surface temperature response for the hypothetical scenario in which atmospheric carbon dioxide (CO2) increases at 1% yr -1 from pre-industrial to the time of a doubling of atmospheric CO2 concentration.
49
 
Technical Summary (a) Model resolution                                                                    (b) Model complexity 0                                                    110                                60 HighRes    100                                                                                                                                        CMIP6 50                                        CMIP6                                        50                                                                                                      CMIP5 90 CMIP5                                                                                                                                                  CMIP3 100                                                  80 CMIP3                                        40 Number of models 150                                                  70 60                                30 km            200 50 Level 250                                                  40                                20 300                                                  30 20                                10 350 10 0
400                                                  0 Atm Res Oce Res Atm lev Ocn lev                                                          # Mod      Aerosols        Atm Chem      Land carbon      N cycle          vegetation      Ocean BGC cycle                          prognostic (c) Pattern correlation with observational reference TS                1.0 0.8 Correlation  0.4 CMIP6 0.2    CMIP5 CMIP3 Additional observations 0
Near-Surface Precipitation TOA        TOA    TOA SW  TOA LW Sea Level Temperature Temperature Eastward              Eastward  Northward Northward Geopotential Specific    Soil      Leaf Area Gross Primary Air Temperature            Outgoing Outgoing Cloud Rad Cloud Rad Pressure  850 hPa    200 hPa    Wind                  Wind        Wind      Wind    Height    Humidity    Moisture      Index    Productivity Shortwave Longwave  Effect  Effect                                  850 hPa              200 hPa    850 hPa  200 hPa  500 hPa    400 hPa Radiation Radiation Figure TS.2 l Progress in climate models. The intent of this figure is to show present improvements in climate models in resolution, complexity and representation of key variables. (a) Evolution of model horizontal resolution and vertical levels (based on Figure 1.19). (b) Evolution of inclusion of processes and resolution from Coupled Model Intercomparison Project Phase 3 (CMIP3), Phase 5 (CMIP5) and Phase 6 (CMIP6; Annex II). (c) Centred pattern correlations between models and observations for the annual mean climatology over the period 1980-1999. Results are shown for individual CMIP3 (cyan), CMIP5 (blue) and CMIP6 (red) models (one ensemble member is used) as short lines, along with the corresponding ensemble averages (long lines). The correlations are shown between the models and the primary reference observational data set (from left to right: ERA5, GPCP-SG, CERES-EBAF, CERES-EBAF, CERES-EBAF, CERES-EBAF, JR-55, ERA5, ERA5, ERA5, ERA5, ERA5, ERA5, AIRS, ERA5, ESACCI-Soilmoisture, LAI3g, MTE). In addition, the correlation between the primary reference and additional observational data sets (from left to right: NCEP, GHCN, -, -, -, -, ERA5, HadISST, NCEP, NCEP, NCEP, NCEP, NCEP, NCEP, ERA5, NCEP, -, -, FLUXCOM) are shown (solid grey circles) if available. To ensure a fair comparison across a range of model resolutions, the pattern correlations are computed after regridding all datasets to a resolution of 4&#xba; in longitude and 5&#xba; in latitude. (Expanded from Figure 3.43; produced with ESMValTool version 2). {Figure 3.43}
TS.1.2.3 Understanding Climate Variability and                                                                          Observational datasets have been extended and improved since Emerging Changes                                                                                                AR5, providing stronger evidence that the climate is changing and allowing better estimates of natural climate variability on decadal Observed changes in climate are unequivocal at the global                                                      time scales. There is very high confidence that the slower rate of global scale and are increasingly apparent on regional and local                                                      surface temperature change observed over 1998-2012 compared to spatial scales. Both the rate of long-term change and                                                          1951-2012 was temporary, and was, with high confidence, induced the amplitude of year-to-year variations differ between                                                        by internal variability (particularly Pacific Decadal Variability) and regions and across climate variables, thus influencing when                                                    variations in solar irradiance and volcanic forcing that partly offset the changes emerge or become apparent compared to natural                                                          anthropogenic warming over this period. Global ocean heat content variations (see Emergence in Core Concepts Box). The signal                                                    continued to increase throughout this period, indicating continuous of temperature change has emerged more clearly in tropical                                                      warming of the entire climate system (very high confidence). Hot regions, where year-to-year variations tend to be small over                                                    extremes also continued to increase during this period over land land, than in regions with greater warming but larger year-                                                    (high confidence). Even in a continually warming climate, periods to-year variations (high confidence) (Figure TS.3). Long-                                                      of reduced and increased trends in global surface temperature at term changes in other variables have emerged in many                                                            decadal time scales will continue to occur in the 21st century (very regions, such as for some weather and climate extremes                                                          high confidence). {Cross-Chapter Box 3.1, 3.3.1, 3.5.1, 4.6.2, 11.3.2}
and Arctic sea ice area. {1.4.2, Cross-Chapter Box 3.1, 9.3.1, 11.3.2, 12.5.2}                                                                                                Since AR5, the increased use of large ensembles, or multiple simulations with the same climate model but using different initial conditions, supports improved understanding of the relative roles 50
 
Technical Summary of internal variability and forced change in the climate system.                          occurred, and there is medium confidence that many of these changes Simulations and understanding of modes of climate variability,                            are attributable to human activities. Several impact-relevant changes including teleconnections, have improved since AR5 (medium                                have not yet emerged from natural variability but will emerge sooner confidence), and larger ensembles allow a better quantification of                        or later in this century depending on the emissions scenario (high uncertainty in projections due to internal climate variability. {1.4.2,                    confidence). Ocean acidification and deoxygenation have already 1.5.3, 1.5.4, 4.2, 4.4.1, Box 4.1, 8.5.2, 10.3.4, 10.4}                                    emerged over most of the global open ocean, as has a reduction in Arctic sea ice (high confidence). {9.3.1, 9.6.4, 11.2, 11.3, 12.4, 12.5, Changes in regional climate can be detected even though natural                            Atlas.3-Atlas.11}
climate variations can temporarily increase or obscure anthropogenic climate change on decadal time scales. While anthropogenic forcing                        TS.1.2.4 Understanding of Human Influence has contributed to multi-decadal mean precipitation changes in several regions, internal variability can delay emergence of the                              The evidence for human influence on recent climate change anthropogenic signal in long-term precipitation changes in many                              has strengthened progressively from the IPCC Second land regions (high confidence). {10.4}                                                        Assessment Report to AR5 and is even stronger in this assessment, including for regional scales and for extremes.                        TS Mean temperatures and heat extremes have emerged above natural                                Human influence in the IPCC context refers to the human variability in almost all land regions with high confidence. Changes                          activities that lead to or contribute to a climate response, in temperature-related variables, such as regional temperatures,                              such as the human-induced emissions of greenhouse growing season length, extreme heat and frost, have already                                  gases that subsequently alter the atmospheres radiative Emergence of changes in surface temperature Annual mean temperature change and the change relative to year-to-year variations (a) Change in temperature at a global warming level of 1&deg;C Zonal mean 90&deg;N Observation (Berkeley Earth) 60&deg;N                              Other observation datasets CMIP6 multi-model mean (45) 30&deg;N                              5-95% model range 0
30&deg;S 60&deg;S Missing 90&deg;S data                                                                                0.5 1  2  3  4    5
                          -2.4 -2.1 -1.8 -1.5 -2.1 -0.9 -0.6 -0.3    0  0.3 0.6 0.9 1.2 1.5  1.8 2.1 2.4
                                                                      &deg;C (b) Change in temperature at a global warming level of 1&deg;C relative to the size of year-to-year variations Zonal mean 90&deg;N 60&deg;N 30&deg;N 0
30&deg;S 60&deg;S Missing                                                                      90&deg;S data                                                                                0.5 1  2  3  4    5 3.5 2.5 1.5 0.5          0    0.5  1    1.5 2 2.5 3  3.5  4 Signal to noise ratio Figure TS.3 l Emergence of changes in temperature over the historical period. The intent of this figure is to show how observed changes in temperature have emerged and that the emergence pattern agrees with model simulations. The observed change in temperature at a global warming level of 1&deg;C (a), and the signal-to-noise ratio (the change in temperature at a global warming level of 1&deg;C, divided by the size of year-to-year variations, (b)) using data from Berkeley Earth. The right panels show the zonal means of the maps and include data from different observational datasets (red) and the Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations (black, including the 5-95% range) processed in the same way as the observations. {1.4.2, 10.4.3}
51
 
Technical Summary properties, resulting in warming of the atmosphere, ocean          of scientific literature combining different lines of evidence, and and land components of the climate system. Other human            improved accessibility to different types of climate models (high activities influencing climate include the emission of            confidence) (see Sections TS.2 and TS.4). {Cross-Working Group Box:
aerosols and other short-lived climate forcers, and land-use      Attribution in Chapter 1, 1.5, 3.2, 3.5, 5.2, 6.4.3, 8.3, 9.6, 10.1, 10.2, change such as urbanization. Progress in our understanding        10.3.3, 10.4.1, 10.4.2, 10.4.3, 10.5, 10.6, Cross-Chapter Box 10.3, of human influence is gained from longer observational            Box 10.3, 11.1.6, 11.2-11.9, 12.4}
datasets, improved paleoclimate information, a stronger warming signal since AR5, and improvements in climate models, physical understanding and attribution techniques          TS.1.3      Assessing Future Climate Change (see Core Concepts Box). Since AR5, the attribution to human influence has become possible across a wider range          Various frameworks can be used to assess future climatic changes and of climate variables and climatic impact-drivers (CIDs,            to synthesize knowledge across climate change assessment in WGI, see Core Concepts Box). New techniques and analyses                WGII and WGIII. These frameworks include: (i) scenarios, (ii) global drawing on several lines of evidence have provided greater        warming levels and (iii) cumulative CO2 emissions (see Core Concepts TS    confidence in attributing changes in regional weather and          Box). The latter two offer scenario- and path-independent approaches climate extremes to human influence (high confidence).            to assess future projections. Additional choices, for instance with regard
{1.3, 1.5.1, Appendix 1.A, 3.1-3.8, 5.2, 6.4.2, 7.3.5, 7.4.4,      to common reference periods and time windows for which changes 8.3.1, 10.4, Cross-Chapter Box 10.3, 11.2-11.9, 12.4}              are assessed, can further help to facilitate integration across the WGI report and across the whole AR6 (see Section TS.1.1). {1.4.1, 1.6, Cross-Combining the evidence from across the climate system increases      Chapter Box 1.4, 4.2.2, 4.2.4, Cross-Chapter Box 11.1}
the level of confidence in the attribution of observed climate change to human influence and reduces the uncertainties associated with      TS.1.3.1 Climate Change Scenarios assessments based on single variables. {Cross-Chapter Box 10.3}
A core set of five illustrative scenarios based on the Shared Since AR5, the accumulation of energy in the Earth system has            Socio-economic Pathways (SSPs) are used consistently become established as a robust measure of the rate of global            across this Report: SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0, climate change on interannual-to-decadal time scales. The rate of        and SSP5-8.5. These scenarios cover a broader range of accumulation of energy is equivalent to Earths energy imbalance        greenhouse gas and air pollutant futures than assessed in and can be quantified by changes in the global energy inventory for      earlier WGI reports, and they include high-CO2 emissions all components of the climate system, including global ocean heat        pathways without climate change mitigation as well as new uptake, warming of the atmosphere, warming of the land and melting      low-CO2 emissions pathways (Figure TS.4). In these scenarios, of ice. Compared to changes in global surface temperature, Earths      differences in air pollution control and variations in climate energy imbalance (see Core Concepts Box) exhibits less variability,      change mitigation stringency strongly affect anthropogenic enabling more accurate identification and estimation of trends. {Box    emissions trajectories of SLCFs. Modelling studies relying 7.2 and Section 7.2}                                                    on the Representative Concentration Pathways (RCPs) used in AR5 complement the assessment based on SSP scenarios, Identifying the human-induced components contributing to the            for example at the regional scale.
energy budget provides an implicit estimate of the human influence on global climate change (Sections TS.2 and TS.3.1). {Cross-Working      A comparison of simulations from CMIP5 using the RCPs Group Box: Attribution in Chapter 1, 3.8, 7.2.2, Box 7.2,                with SSP-based simulations from CMIP6 shows that about Cross-Chapter Box 9.1}                                                  half of the increase in simulated warming in CMIP6 versus CMIP5 arises because higher climate sensitivity is more Regional climate changes can be moderated or amplified by regional      prevalent in CMIP6 model versions; the other half arises forcing from land-use and land-cover changes or from aerosol            from higher radiative forcing in nominally corresponding concentrations and other short-lived climate forcers (SLCFs). For        scenarios (e.g., RCP8.5 and SSP5-8.5; medium confidence).
example, the difference in observed warming trends between cities        The feasibility or likelihood of individual scenarios is not and their surroundings can partly be attributed to urbanization          part of this assessment, which focuses on the climate (very high confidence). While established attribution techniques        response to a large range of emissions scenarios. {1.5.4, 1.6, provide confidence in our assessment of human influence on              Cross-Chapter Box 1.4, 4.2, 4.3, 4.6, 6.6, 6.7, Cross-Chapter large-scale climate changes (as described in Section TS.2), new          Box 7.1, Atlas.2.1}
techniques developed since AR5, including attribution of individual events, have provided greater confidence in attributing changes in    Climate change projections with climate models require information climate extremes to climate change (Box TS.10). Multiple attribution  about future emissions or concentrations of greenhouse gases, approaches support the contribution of human influence to several    aerosols, ozone-depleting substances, and land use over time regional multi-decadal mean precipitation changes (high confidence).  (Figure TS.4). This information can be provided by scenarios, which Understanding about past and future changes in weather and climate    are internally consistent projections of these quantities based on extremes has increased due to better observation-based datasets,      assumptions of how socio-economic systems could evolve over the physical understanding of processes, an increasing proportion        21st century. Emissions from natural sources, such as the ocean and 52
 
Technical Summary Human activities Other CO2                                                                                                                      (e.g., SLCF, land use albedo)
Non-CO2 Greenhouse gases Emissions                                                                                          N2O HFCs etc.
200 NH3 Land use albedo etc.
900 175 NOx                                                                                  TS 125 CO2                                                            800          CH4 100                                                                      700                                                                          150 MtNOx / yr 600                                                                          125 GtCO2 / yr                                                          MtCH4 / yr 75                                                                  500                                                                          100 50                                                                  400 75 25                                                                  300 200                                                                          50 0                                                                  100                                                                          25
                                                    -25                                                                    0                                                                            0 1950        2000      2050          2100                          1950                                2000    2050      2100                  1950  2000                          2050      2100 Concentrations                                                                  4000 N2O HFCs etc.
140 120 1000 CO2                                                  3500              CH4                                                                    100 MtSO2 / yr 3000                                                                                          80 800 2500 ppm                                                                ppb                                                                                                    60 600                                                                2000                                                                                          40 400                                                                1500                                                                                          20      SO2 1000                                                                                          0 200                                                                                                                                                              1950          2000          2050      2100 1950            2000      2050          2100                      1950                                    2000    2050      2100 12 Radiative Effective radiative forcing 10 Forcing                                                              8 6
Total anthropogenic (W/m-2) 4 Carbon-                                                                                                          2 Cycle and                                                                                                        0 non-CO2                                                                                                                        Natural
                                                                                                                                                          -2 biogeo-                                                                                                            1950                    2000                                      2050                      2100 chemical feedbacks 7
Global Change in global surface 6
Warming                                                                  5                                                            Projections temperature 4
3 Legend:
(&deg;C rel. to 1850-1900) 2 Historical                                                                                                                                    Observations 1
SSP5-8.5 SSP3-7.0                                                                                                                        0 1950                2000                                    2050                      2100 SSP2-4.5                      2100 SSP1-2.6                      RCP SSP1-1.9                      range                                                Regional Climate Change Temperature                                                                                                                                                    Precipitation GWL 2&deg;C                                                                                                                                                            GWL 2&deg;C Climatic Impact-Robust significant change No or no robust                                4 2.5 1.5 0 1.5 2 2.5 3 4 5 Drivers                                              -40      -20    0    +20 +40 Legend see left significant change                                                                                                                                                                                Change in annual mean Change in annual mean Conflicting signal                                surface temperature (&deg;C)                                                                                                                          precipitation (%)
Figure TS.4 l The climate change cause-effect chain: The intent of this figure is to illustrate the process chain starting from anthropogenic emissions, to changes in atmospheric concentration, to changes in Earths energy balance (forcing), to changes in global climate and ultimately regional climate and climatic impact-drivers. Shown is the core set of five Shared Socio-economic Pathway (SSP) scenarios as well as emissions and concentration ranges for the previous Representative Concentration Pathway (RCP) scenarios in year 2100; carbon dioxide (CO2) emissions (GtCO2 yr-1), panel top left; methane (CH4) emissions (middle) and sulphur dioxide (SO2), nitrogen oxide (NOx) emissions (all in Mt yr-1), top right; concentrations of atmospheric CO2 (ppm) and CH4 (ppb), second row left and right; effective radiative forcing for both anthropogenic and natural forcings (W m-2), third row; changes in global surface air temperature (&deg;C) relative to 1850-1900, fourth row; maps of projected temperature change (&deg;C) (left) and changes in annual-mean precipitation (%) (right) at a global warming level (GWL) of 2&deg;C relative to 1850-1900 (see also Figure TS.5), bottom row. Carbon cycle and non-CO2 biogeochemical feedbacks will also influence the ultimate response to anthropogenic emissions (arrows on the left). {1.6.1, Cross-Chapter Box 1.4, 4.2.2, 4.3.1, 4.6.1, 4.6.2}
53
 
Technical Summary the land biosphere, are usually assumed to be constant, or to evolve                                    at the regional scale (Section TS.4). Scenario extensions are based in response to changes in anthropogenic forcings or to projected                                        on assumptions about the post-2100 evolution of emissions or of climate change. Natural forcings, such as past changes in solar                                          radiative forcing that are independent from the modelling of socio-irradiance and historical volcanic eruptions, are represented in model                                  economic dynamics, which does not extend beyond 2100. To explore simulations covering the historical era. Future simulations assessed                                    specific dimensions, such as air pollution or temporary overshoot of in this Report account for projected changes in solar irradiance and                                    a given warming level, scenario variants are used in addition to the for the long-term mean background forcing from volcanoes, but not                                        core set. {1.6.1, Cross-Chapter Box 1.4, 4.2.2, 4.2.6, 4.7.1, Cross-for individual volcanic eruptions. Scenarios have a long history in                                      Chapter Box 7.1}
IPCC as a method for systematically examining possible futures and following the cause-effect chain: from anthropogenic emissions, to                                      SSP1-1.9 represents the low end of future emissions pathways, changes in atmospheric concentrations, to changes in Earths energy                                      leading to warming below 1.5&deg;C in 2100 and limited temperature balance (forcing), to changes in global climate and ultimately                                        overshoot of 1.5&deg;C over the course of the 21st century (see regional climate and climatic impact-drivers (Figure TS.4, Section                                      Figure TS.6). At the opposite end of the range, SSP5-8.5 represents TS.2, Infographic TS.1). {1.5.4, 1.6.1, 4.2.2, 4.4.4, Cross-Chapter                                      the very high warming end of future emissions pathways from the TS Box 4.1, 11.1}                                                                                          literature. SSP3-7.0 has overall lower GHG emissions than SSP5-8.5 but, for example, CO2 emissions still almost double by 2100 compared The uncertainty in climate change projections that results from                                          to todays levels. SSP2-4.5 and SSP1-2.6 represent scenarios with assessing alternative socio-economic futures, the so-called scenario                                    stronger climate change mitigation and thus lower GHG emissions.
uncertainty, is explored through the use of scenario sets. Designed                                      SSP1-2.6 was designed to limit warming to below 2&deg;C. Infographic to span a wide range of possible future conditions, these scenarios                                      TS.1 presents a narrative depiction of SSP-related climate futures.
do not intend to match how events actually unfold in the future,                                        No likelihood is attached to the scenarios assessed in this Report, and they do not account for impacts of climate change on the socio-                                      and the feasibility of specific scenarios in relation to current trends economic pathways. Besides scenario uncertainty, climate change                                          is best informed by the WGIII contribution to AR6. In the scenario projections are also subject to climate response uncertainty (i.e., the                                  literature, the plausibility of some scenarios with high CO2 emissions, uncertainty related to our understanding of the key physical processes                                  such as RCP8.5 or SSP5-8.5, has been debated in light of recent and structural uncertainties in climate models) and irreducible and                                      developments in the energy sector. However, climate projections from intrinsic uncertainties related to internal variability. Depending on                                    these scenarios can still be valuable because the concentration levels the spatial and temporal scales of the projection, and on the variable                                  reached in RCP8.5 or SSP5-8.5 and corresponding simulated climate of interest, the relative importance of these different uncertainties                                    futures cannot be ruled out. That is because of uncertainty in carbon-may vary substantially. {1.4.3, 1.6, 4.2.5, Box 4.1, 8.5.1}                                              cycle feedbacks which, in nominally lower emissions trajectories, can result in projected concentrations that are higher than the central Scenarios in AR6 cover a broader range of emissions futures than                                        concentration levels typically used to drive model projections. {1.6.1; considered in AR5, including high CO2 emissions scenarios without                                        Cross-Chapter Box 1.4; 4.2.2, 5.4; SROCC; Chapter 3 in WGIII}
climate change mitigation as well as a low CO2 emissions scenario reaching net zero CO2 emissions (see Core Concepts Box) around mid-                                      The socio-economic narratives underlying SSP-based scenarios century. In this Report, a core set of five illustrative scenarios is used                              differ in their assumed level of air pollution control. Together with to explore climate change over the 21st century and beyond (Section                                      variations in climate change mitigation stringency, this difference TS.2). They are labelled SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0, and                                    strongly affects anthropogenic emissions trajectories of SLCFs, SSP5-8.517 and span a wide range of radiative forcing levels in 2100.                                    some of which are also air pollutants. SSP1 and SSP5 assume strong They start in 2015 and include scenarios with high and very high                                        pollution control, projecting a decline of global emissions of ozone GHG emissions and CO2 emissions that roughly double from current                                        precursors (except methane; CH4) and of aerosols and most of levels by 2100 and 2050, respectively (SSP3-7.0 and SSP5-8.5);                                          their precursors in the mid- to long term. The reductions due to air scenarios with intermediate GHG emissions and CO2 emissions                                              pollution controls are further strengthened in scenarios that assume remaining around current levels until the middle of the century                                          a marked decarbonization, such as SSP1-1.9 or SSP1-2.6. SSP2-4.5 (SSP2-4.5); and scenarios with very low and low GHG emissions and                                        is a medium pollution-control scenario with air pollutant emissions CO2 emissions declining to net zero around or after 2050, followed by                                    following current trends, and SSP3-7.0 is a weak pollution-control varying levels of net negative CO2 emissions (SSP1-1.9 and SSP1-2.6).                                    scenario with strong increases in emissions of air pollutants over the These SSP scenarios offer unprecedented detail of input data for ESM                                    21st century. Methane emissions in SSP-based scenarios vary with simulations and allow for a more comprehensive assessment of                                            the overall climate change mitigation stringency, declining rapidly climate drivers and responses, in particular because some aspects,                                      in SSP1-1.9 and SSP1-2.6 but declining only after 2070 in SSP5-8.5.
such as the temporal evolution of pollutants, emissions or changes in                                    SSP trajectories span a wider range of air pollutant emissions than land use and land cover, span a broader range in the SSP scenarios                                      considered in the RCP scenarios (see Figure TS.4), reflecting the than in the RCPs used in AR5. Modelling studies utilizing the RCPs                                      potential for large regional differences in their assumed pollution complement the assessment based on SSP scenarios, for example, 17  Throughout this Report, scenarios are referred to as SSPx-y, where SSPx refers to the Shared Socio-economic Pathway or SSP describing the socio-economic trends underlying the scenario, and y refers to the approximate target level of radiative forcing (in W m--2) resulting from the scenario in the year 2100.
54
 
Technical Summary policies. Their effects on climate and air pollution are assessed in Box                          TS.1.3.2 Global Warming Levels and Cumulative CO2 Emissions TS.7. {4.4.4, 6.6.1, Figure 6.4, 6.7.1, Figure 6.19}
Quantifying geographical response patterns of climate Since the RCPs are also labelled by the level of radiative forcing                                  change at various global warming levels (GWLs), such as they reach in 2100, they can in principle be related to the core set                                1.5&deg;C or 2&deg;C above the 1850-1900 period, is useful for of AR6 scenarios (Figure TS.4). However, the RCPs and SSP-based                                      characterizing changes in mean climate, extremes and scenarios are not directly comparable. First, the gas-to-gas                                        climatic impact-drivers. Global warming levels are used compositions differ; for example, the SSP5-8.5 scenario has higher                                  in this Report as a dimension of integration independent CO2 but lower CH4 concentrations compared to RCP8.5. Second, the                                    of the timing when the warming level is reached and of projected 21st-century trajectories may differ, even if they result                                  the emissions scenario that led to the warming. For many in the same radiative forcing by 2100. Third, the overall effective                                  climate variables the response pattern for a given GWL is radiative forcing (see Core Concepts Box) may differ, and tends                                      consistent across different scenarios. However, this is not to be higher for the SSPs compared to RCPs that share the same                                      the case for slowly responding processes, such as ice-sheet nominal stratospheric-temperature-adjusted radiative forcing label.                                  and glacier mass loss, deep ocean warming, and the related Comparing the differences between CMIP5 and CMIP6 projections                                        sea level rise. The response of these variables depends on        TS (Cross-Section Box TS.1) that were driven by RCPs and SSP-based                                      the time it takes to reach the GWL, differs if the warming is scenarios, respectively, indicates that about half of the difference in                              reached in a transient warming state or after a temporary simulated warming arises because of higher climate sensitivity being                                overshoot of the warming level, and will continue to evolve, more prevalent in CMIP6 model versions; the remainder arises from                                    over centuries to millennia, even after global warming higher ERF in nominally corresponding scenarios (e.g., RCP8.5 and                                    has stabilized. Different GWLs correspond closely to SSP5-8.5; medium confidence) (see Section TS.1.2.2). In SSP1-2.6 and                                specific cumulative CO2 emissions due to their near-linear SSP2-4.5, changes in ERF also explain about half of the changes in the                              relationship with global surface temperature. This Report range of warming (medium confidence). For SSP5-8.5, higher climate                                  uses 1.0&deg;C, 1.5&deg;C, 2.0&deg;C, 3.0&deg;C and 4.0&deg;C above 1850-1900 sensitivity is the primary reason behind the upper end of the CMIP6-                                conditions as a primary set of GWLs. {1.6.2, 4.2.4, 4.6.1, 5.5, projected warming being higher than for RCP8.5 in CMIP5 (medium                                      Cross-Chapter Box 11.1, Cross-chapter Box 12.1}
confidence). Note that AR6 uses multiple lines of evidence beyond CMIP6 results to assess global surface temperature under various                                  For many indicators of climate change, such as seasonal and annual scenarios (see Cross-Section Box TS.1 for the detailed assessment).                              mean and extreme surface air temperatures and precipitation, the
{1.6, 4.2.2, 4.6.2.2, Cross-Chapter Box 7.1}                                                      geographical patterns of changes are well estimated by the level of global surface warming, independently of the details of the emissions Earth system models can be driven by anthropogenic CO2                                            pathways that caused the warming, or the time at which the level of emissions (emissions-driven runs), in which case atmospheric                                    warming is attained. GWLs, defined as a global surface temperature CO2 concentration is a projected variable; or by prescribed time-                                increase of, for example, 1.5&deg;C or 2&deg;C relative to the mean of 1850-varying atmospheric concentrations (concentration-driven runs). In                              1900, are therefore a useful way to integrate climate information emissions-driven runs, changes in climate feed back on the carbon                                independently of specific scenarios or time periods. {1.6.2, 4.2.4, cycle and interactively modify the projected CO2 concentration                                    4.6.1, 11.2.4, Cross-Chapter Box 11.1}
in each ESM, thus adding the uncertainty in the carbon cycle response to climate change to the projections. Concentration-                                    The use of GWLs allows disentangling the contribution of changes driven simulations are based on a central estimate of carbon cycle                                in global warming from regional aspects of the climate response, feedbacks, while emissions-driven simulations help quantify the role                              as scenario differences in response patterns at a given GWL are of feedback uncertainty. The differences in the few ESMs for which                                often smaller than model uncertainty and internal variability. The both emissions and concentration-driven runs were available for the                              relationship between the GWL and response patterns is often linear, same scenario are small and do not affect the assessment of global                                but integration of information can also be done for non-linear surface temperature projections discussed in Cross-Section Box TS.1                              changes, like the frequency of heat extremes. The requirement is that and Section TS.2 (high confidence). By the end of the 21st century,                              the relationship to the GWL is broadly independent of the scenario emissions-driven simulations are on average around 0.1&deg;C cooler                                  and relative contribution of radiative forcing agents. {1.6, 11.2.4, than concentration-driven runs, reflecting the generally lower CO2                                Cross-Chapter Box 11.1}
concentrations simulated by the emissions-driven ESMs, and have a spread about 0.1&deg;C greater, reflecting the range of simulated CO2                                The GWL approach to integration of climate information also has concentrations. However, these carbon cycle-climate feedbacks do                                  some limitations. Variables that are quick to respond to warming, affect the transient climate response to cumulative CO2 emissions                                like temperature and precipitation, including extremes, sea ice area, (TCRE18), and their quantification is crucial for the assessment of                              permafrost and snow cover, show little scenario dependence for a remaining carbon budgets consistent with global warming levels                                    given GWL, whereas slow-responding variables such as glacier and simulated by ESMs (see Section TS.3). {1.6.1, Cross-Chapter Box 1.4,                              ice-sheet mass, warming of the deep ocean and their contributions 4.2, 4.3.1, 5.4.5, Cross-Chapter Box 7.1}                                                        to sea level rise, have substantial dependency on the trajectory of 18  The transient surface temperature change per unit of cumulative CO2 emissions, usually 1000 GtC.
55
 
Technical Summary warming taken to reach the GWL. A given GWL can also be reached                                SR1.5 concluded that climate models project robust differences in for different balances between anthropogenic forcing agents, such                              regional climate characteristics between present-day and global as long-lived greenhouse gas and SLCF emissions, and the response                              warming of 1.5&deg;C, and between 1.5&deg;C and 2&deg;C. This Report adopts patterns may depend on this balance. Finally, there is a difference                            a set of common GWLs across which climate projections, impacts, in the response even for temperature-related variables if a GWL is                            adaptation challenges and climate change mitigation challenges can reached in a rapidly warming transient state or in an equilibrium                              be integrated, within and across the three Working Groups, relative state when the land-sea warming contrast is less pronounced. In this                          to 1850-1900. The core set of GWLs in this Report are 1.0&deg;C (close Report, the climate responses at different GWLs are calculated based                          to present day conditions), 1.5&deg;C, 2.0&deg;C, 3.0&deg;C and 4.0&deg;C. {1.4, 1.6.2, on climate model projections for the 21st century (see Figure TS.5),                          Cross-Chapter Box 1.2, Table 1.5, Cross-Chapter Box 11.1}
which are mostly not in equilibrium. The SSP1-1.9 scenario allows assessing the response to a GWL of about 1.5&deg;C after a (relatively)                            Connecting Scenarios and Global Warming Levels short-term stabilization by the end of the 21st century. {4.6.2, 9.3.1.1, 9.5.2.3, 9.5.3.3, 11.2.4, Cross-Chapter Box 11.1, Cross-Chapter                                In this Report, scenario-based climate projections are translated Box 12.1}                                                                                      into GWLs by aggregating the ESM model response at specific TS                                                                                                  GWLs across scenarios (see Figure TS.5 and Figure TS.6). The climate Global warming levels are highly relevant as a dimension of                                    response pattern for the 20-year period around when individual integration across scientific disciplines and socio-economic actors                            simulations reach a given GWL are averaged across all models and and are motivated by the long-term goal in the Paris Agreement                                scenarios that reach that GWL. The best estimate and likely range of holding the increase in the global average temperature to well                            of the timing of when a certain GWL is reached under a particular below 2&deg;C above pre-industrial levels and to pursue efforts to limit                          scenario (or GWL-crossing time), however, is based not only on the temperature increase to 1.5&deg;C above pre-industrial levels.                                CMIP6 output, but on a combined assessment taking into account The evolution of aggregated impacts with temperature levels has                                the observed warming to date, CMIP6 output and additional lines also been widely used and embedded in the WGII assessment.                                    of evidence (see Cross-Section Box TS.1). {4.3.4, Cross-Chapter This includes the Reasons for Concern (RFC) and other burning                              Box 11.1, Atlas.2, Interactive Atlas}
ember diagrams in IPCC WGII. The RFC framework has been further expanded in SR1.5, SROCC and SRCCL by explicitly looking at the                                Global warming levels are closely related to cumulative CO2 (and differential impacts between half-degree GWLs and the evolution of                            in some cases CO2-equivalent) emissions. This Report confirms the risk for different socio-economic assumptions. {1.4.4, 1.6.2, 11.2.4,                          assessment of the WGI contribution to AR5 and SR1.5 that a near-12.5.2, Cross-Chapter Box 11.1, Cross-Chapter Box 12.1}                                        linear relationship exists between cumulative CO2 emissions and the (a) Global mean temperature in CMIP6                      (b) Patterns of change in near-surface air temperature, precipitation and soil moisture 5  SSP3-7.0 (20-yr GSAT means)                                      Temperature change                          Precipitation change                  Soil moisture change SSP1-2.6 (20-yr GSAT means)                                                                          48                                    47                                            43
                                                        +4&deg;C      +4&deg;C 4
137                                  132                                            119 3
    &deg;C                                                            +2&deg;C
                                                        +2&deg;C 2
                                                        +1.5&deg;C                                                    153                                  147                                            132 1                                                        +1.5&deg;C 2000    2020    2040    2060    2080    2100              4 2.5-2 -1.5 0 1.5 2 2.5 3 4 5
                                                                                            &deg;C 30 10 0 10 20 30 40
                                                                                                                                                                -2.5 1.5 0.5 0 0.5 1 1.5 2 2.5 Figure TS.5 l Scenarios, global warming levels, and patterns of change. The intent of this figure is to show how scenarios are linked to global warming levels (GWLs) and to provide examples of the evolution of patterns of change with global warming levels. (a) Illustrative example of GWLs defined as global surface temperature response to anthropogenic emissions in unconstrained Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations, for two illustrative scenarios (SSP1-2.6 and SSP3-7.0). The time when a given simulation reaches a GWL, for example, +2&deg;C, relative to 1850-1900 is taken as the time when the central year of a 20-year running mean first reaches that level of warming. See the dots for +2&deg;C, and how not all simulations reach all levels of warming. The assessment of the timing when a GWL is reached takes into account additional lines of evidence and is discussed in Cross-Section Box TS.1. (b) Multi-model, multi-simulation average response patterns of change in near-surface air temperature, precipitation (expressed as percentage change) and soil moisture (expressed in standard deviations of interannual variability) for three GWLs. The number to the top right of the panels shows the number of model simulations averaged across including all models that reach the corresponding GWL in any of the five Shared Socio-economic Pathways (SSPs). See Section TS.2 for discussion. {Cross-Chapter Box 11.1}
56
 
Technical Summary resulting increase in global surface temperature (Section TS.3.2). This as sub-continents and oceanic regions, or to typological regions, such implies that continued CO2 emissions will cause further warming        as monsoon regions, coastlines, mountain ranges or cities, as used in and associated changes in all components of the climate system. For    Section TS.4. A new set of standard AR6 WGI reference regions has declining cumulative CO2 emissions (i.e., if negative net emissions    also been included in this Report (Figure TS.6, bottom panels). {1.4.5, are achieved), the relationship is less strong for some components,    10.1, 11.9, 12.1-12.4, Atlas.1.3.3-1.3.4}
such as the hydrological cycle. The WGI report uses cumulative CO2 emissions to compare climate response across scenarios and provides    Global and regional climate models are important sources of climate a link to the emissions pathways assessment in WGIII. The advantage    information at the regional scale. Since AR5, a more comprehensive of using cumulative CO2 emissions is that it is an inherent emissions  assessment of past and future evolution of a range of climate variables scenario characteristic rather than an outcome of the scenario-based    on a regional scale has been enabled by the increased availability projections, where uncertainties in the cause-effect chain from        of coordinated ensemble regional climate model projections and emissions to temperature change are important (Figure TS.4), for        improvements in the level of sophistication and resolution of example, the uncertainty in ERF and TCR. Cumulative CO2 emissions      global and regional climate models. This has been complemented can also provide a link to the assessments of mitigation options.      by observational, attribution and sectoral-vulnerability studies Cumulative CO2 emissions do not carry information about non-CO2        informing, for instance, about impact-relevant tolerance thresholds. TS emissions, although these can be included with specific emissions      {10.3.3, 11.9, 12.1, 12.3, 12.6, Atlas.3-Atlas.11}
metrics to estimate CO2-equivalent emissions. (Section TS.3.3) {1.3.2, 1.6, 4.6.2, 5.5, 7.6}                                                  Multiple lines of evidence derived from observations, model simulations and other approaches can be used to construct climate information on a regional scale as described in detail in Sections TS.4.1.1 and TS.1.4      From Global to Regional Climate Information for            TS.4.1.2. Depending on the phenomena and specific context, these Impact and Risk Assessment                                sources and methodologies include theoretical understanding of the relevant processes, drivers and feedbacks of climate at regional scale; The AR6 WGI Report has an expanded focus on regional                trends in observed data from multiple datasets; and the attribution information supported by the increased availability of              of these trends to specific drivers. Furthermore, simulations from coordinated regional climate model ensemble projections            different model types (including global and regional climate models, and improvements in the sophistication and resolution              emulators, statistical downscaling methods, etc.) and experiments of global and regional climate models (high confidence).            (e.g., CMIP, CORDEX, and large ensembles of single-model simulations Multiple lines of evidence can be used to construct climate        with different initial conditions), attribution methodologies and other information on a global to regional scale and can be further        relevant local knowledge (e.g., indigenous knowledge) are utilized distilled in a co-production process to meet user needs (high      (see Box TS.11). {1.5.3, 1.5.4, Cross-Chapter Box 7.1, 10.2-10.6, 11.2, confidence). To better support risk assessment, a common            Atlas.1.4, Cross-Chapter Box 10.3}
risk framework across all three Working Groups has been implemented in AR6, and low-likelihood but high-impact              From the multiple lines of evidence, climate information can be outcomes are explicitly addressed in WGI by using physical          distilled in a co-production process that involves users, related climate storylines (see Core Concepts Box).                        stakeholders and producers of climate information, considering the specific context of the question at stake, the underlying values and Climatic impact-drivers are physical climate system                the challenge of communicating across different communities. The conditions (e.g., means, events, extremes) that affect              co-production process is an essential part of climate services, which an element of society or ecosystems. They are the WGI              are discussed in Section TS.4.1.2. {10.5, 12.6, Cross-Chapter Box 12.2}
contribution to the risk framing without anticipating whether their impact provides potential opportunities              With the aim of informing decision-making at local or regional or is detrimental (i.e., as for hazards). Many global and          scales, a common risk framework has been implemented in AR6.
regional climatic impact-drivers have a direct relation to          Methodologies have been developed to construct more impact- and global warming levels (high confidence). {1.4.4, 1.5.2-1.5.4,      risk-relevant climate change information tailored to regions and Cross-Chapter Box 1.3, 4.8, 10.1, 10.5.1, Box 10.2, Cross-          stakeholders. Physical storyline approaches are used in order to Chapter Box 10.3, 11.2.4, 11.9, Box 11.2, Cross-Chapter            build climate information based on multiple lines of evidence, and Box 11.1, 12.1-12.3, 12.6, Cross-Chapter Boxes 12.1 and            which can explicitly address physically plausible, but low-likelihood, 12.2, Atlas.1.3.3-1.3.4, Atlas.1.4, Atlas.1.4.4}                    high-impact outcomes and uncertainties related to climate variability for consideration in risk assessments (Figure TS.6). {Cross-Chapter Climate change is a global phenomenon, but manifests differently        Box 1.3, 4.8, Box 9.4, 10.5, Box 10.2, Box 11.2, 12.1-12.3, 12.6, in different regions. The impacts of climate change are generally      Glossary}
experienced at local, national and regional scales, and these are also the scales at which decisions are typically made. Robust climate  The climatic impact-driver framework developed in AR6 supports change information is increasingly available at regional scales for    an assessment of changing climate conditions that are relevant impact and risk assessments. Depending on the climate information      for sectoral impacts and risk assessment. Climatic impact-drivers context, geographical regions in AR6 may refer to larger areas, such    (CIDs) are physical climate system conditions (e.g., means, extremes, 57
 
Technical Summary Schematic ECS likelihood                                Assessed changes in global surface temperature Observations                              Projections SSP5-8.5 6              Low-likelihood                                                                                                              SSP3-7.0 High ECS high warming                                                                                                              SSP2-4.5 SSP1-2.6 Equilibrium climate sensitivity (ECS, &deg;c) 5 SSP1-1.9 1980              2000            2020        2040          2060        2080 4                                                                                                                                          SSP5-8.5 Mid-range ECS SSP3-7.0 Likely range                                                                                                                      SSP2-4.5 3
SSP1-2.6 TS                                                                                                                                                                                          SSP1-1.9 2                                                        1980              2000            2020        2040          2060        2080 SSP5-8.5 Low-likelihood                                                                                                              SSP3-7.0 1
Low ECS low warming                                                                                                                SSP2-4.5 SSP1-2.6 0                                                                                                                                          SSP1-1.9 Relative likelihood Global Warming Level (GWL, &deg;C) 0.5      1    1.5      2            3        4 GWL 1.5&deg;C                                          GWL 2.0&deg;C                                    GWL 4.0&deg;C Heat warning index 1 day/year  3 days/year            2 weeks/year          1 month/year  3 months/year  6 months/year GWL 1.5&deg;C                                          GWL 2.0&deg;C                                    GWL 4.0&deg;C Changes in extreme rainfall 0                5                10                15                  20            25                30
                                                                                                          % more rain on wettest day of the year Figure TS.6 l A graphical abstract for key aspects of the Technical Summary. The intent of this figure is to summarize many different aspects of the Technical Summary related to observed and projected changes in global temperature and associated regional changes in climatic impact-drivers relevant for impact and risk assessment. Top left:
a schematic representation of the likelihood for equilibrium climate sensitivity (ECS), consistent with the AR6 assessment (see Chapter 7 and Section TS.3). ECS values above 5&deg;C and below 2&deg;C are termed low-likelihood, high warming (LLHW) and low-likelihood, low warming, respectively (Box TS.3). Top right: Observed (see Cross-Section Box TS.1) and projected global surface temperature changes, shown as global warming levels (GWLs) relative to 1850-1900, using the assessed 95% (top), 50% (middle) and 5%
(bottom) likelihood time series (see Chapter 4 and Section TS.2). Bottom panels show maps of Coupled Model Intercomparison Project Phase 6 (CMIP6) median projections of two climatic impact-drivers (CIDs, see Section TS.1.4) at three different GWLs (columns for 1.5, 2 and 4&deg;C) for the AR6 land regions (see Chapters 1, 10, and Atlas and Section TS.4). The heat warning index is the number of days per year averaged across each region at which a heat warning for human health at level danger would be issued according to the U.S. National Oceanic and Atmospheric Administration (NOAA) (NOAA HI41, see Chapter 12 and Annex VI). The maps of extreme rainfall changes show the percentage change in the amount of rain falling on the wettest day of a year (Rx1day, relative to 1995-2014, see Chapter 11) averaged across each region when the respective GWL is reached. Additional CIDs are discussed in Section TS.4. {1.4.4, Box 4.1, 7.5, 11.4.3, 12.4}
58
 
Technical Summary events) that affect an element of society or ecosystems and are thus      Many global- and regional-scale CIDs, including extremes, have a potential priority for providing climate information. For instance,    a direct relation to global warming levels (GWLs) and can thus inform the heat index used by the U.S. National Oceanic and Atmospheric          the hazard component of Representative Key Risks and Reasons for Administration (NOAA HI) for issuing heat warnings is a CID index        Concern assessed by AR6 WGII. These include heat, cold, wet and dry that can be associated with adverse human health impacts due to          hazards, both mean and extremes; cryospheric hazards (snow cover, heat stress (see Figure TS.6). Depending on system tolerance, CIDs and    ice extent, permafrost) and oceanic hazards (marine heatwaves) their changes can be detrimental (i.e., hazards in the risk framing),    (high confidence) (Figure TS.6). Establishing links between specific beneficial, neutral, or a mixture of each across interacting system      GWLs with tipping points and irreversible behaviour is challenging elements, regions and sectors (aligning with WGII Sectoral Chapters      due to model uncertainties and lack of observations, but their 2-8). Each sector is affected by multiple CIDs, and each CID affects      occurrence cannot be excluded, and their likelihood of occurrence multiple sectors. Climate change has already altered CID profiles and    generally increases at greater warming levels (Box TS.1, Section TS.9).
resulted in shifting magnitude, frequency, duration, seasonality and      {11.2.4, Box 11.2, Cross-Chapter Boxes 11.1 and 12.1}
spatial extent of associated indices (high confidence) (see regional details in Section TS.4.3). {12.1-12.4, Table 12.1, Table 12.2, Annex VI}
TS Cross-Section Box TS.1: Global Surface Temperature Change This box synthesizes the outcomes of the assessment of past, current and future global surface temperature. Global mean surface temperature (GMST) and global surface air temperature (GSAT) are the two primary metrics of global surface temperature used to estimate global warming in IPCC reports. GMST merges sea surface temperature (SST) over the ocean and 2 m air temperature over land and sea ice areas and is used in most paleo, historical and present-day observational estimates. The GSAT metric is 2 m air temperature over all surfaces and is the diagnostic generally used from climate models. Changes in GMST and GSAT over time differ by at most 10% in either direction (high confidence), but conflicting lines of evidence from models and direct observations, combined with limitations in theoretical understanding, lead to low confidence in the sign of any difference in long-term trend. Therefore, long-term changes in GMST/GSAT are presently assessed to be identical, with expanded uncertainty in GSAT estimates. Hence the term global surface temperature is used in reference to both quantities in the text of the TS and SPM. {Cross-Chapter Box 2.3}
Global surface temperature has increased by 0.99 [0.84 to 1.10] &deg;C from 1850-1900 to the first two decades of the 21st century (2001-2020) and by 1.09 [0.95 to 1.20] &deg;C from 1850-1900 to 2011-2020. Temperatures as high as during the most recent decade (2011-2020) exceed the warmest centennial-scale range reconstructed for the present interglacial, around 6500 years ago [0.2&deg;C to 1&deg;C] (medium confidence). The next most recent warm period was about 125,000 years ago during the last interglacial when the multi-centennial temperature range [0.5&deg;C to 1.5&deg;C]
encompasses the 2011-2020 values (medium confidence). The likely range of human-induced change in global surface temperature in 2010-2019 relative to 1850-1900 is 0.8&deg;C to 1.3&deg;C, with a central estimate of 1.07&deg;C, encompassing the best estimate of observed warming for that period, which is 1.06&deg;C with a very likely range of [0.88&deg;C to 1.21&deg;C],
while the likely range of the change attributable to natural forcing is only -0.1&deg;C to +0.1&deg;C.
Compared to 1850-1900, average global surface temperature over the period 2081-2100 is very likely to be higher by
[1.0&deg;C to 1.8&deg;C] in the low CO2 emissions scenario SSP1-1.9 and by [3.3&deg;C to 5.7&deg;C] in the high CO2 emissions scenario SSP5-8.5. In all scenarios assessed here except SSP5-8.5, the central estimate of 20-year averaged global surface warming crossing the 1.5&deg;C level lies in the early 2030s, which is in the early part of the likely range (2030-2052) assessed in SR1.5. It is more likely than not that under SSP1-1.9, global surface temperature relative to 1850-1900 will remain below 1.6&deg;C throughout the 21st century, implying a potential temporary overshoot of 1.5&deg;C global warming of no more than 0.1&deg;C. Global surface temperature in any individual year could exceed 1.5&deg;C relative to 1850-1900 by 2030 with a likelihood between 40% and 60% across the scenarios considered here (medium confidence). A 2&deg;C increase in global surface temperature relative to 1850-1900 will be crossed under SSP5-8.5 but is extremely unlikely to be crossed under SSP1-1.9. Periods of reduced and increased global surface temperature trends at decadal time scales will continue to occur in the 21st century (very high confidence). The effect of strong mitigation on 20-year global surface temperature trends would be likely to emerge during the near term (2021-2040), assuming no major volcanic eruptions occur. (Figure TS.8, Cross-Section Box TS.1, Figure 1) {2.3, 3.3, 4.3, 4.4, 4.5, 4.6, 7.3}
Surface Temperature History Dataset innovations, particularly more comprehensive representation of polar regions, and the availability of new datasets have led to an assessment of increased global surface temperature change relative to the directly equivalent estimates reported in AR5. The contribution of changes in observational understanding alone between AR5 and AR6 in assessing temperature changes from 1850-1900 59
 
Technical Summary Cross-Section Box TS.1 (continued) to 1986-2005 is estimated at 0.08 [-0.01 to 0.12] &deg;C. Global surface temperature increased from 1850-1900 to 1995-2014 by 0.85
[0.69 to 0.95] &deg;C, between 1850-1900 and the first two decades of the 21st century (2001-2020) by 0.99 [0.84 to 1.20] &deg;C, and to the most recent decade (2011-2020) by 1.09 [0.95 to 1.20] &deg;C. Each of the last four decades has in turn been warmer than any decade that preceded it since 1850. Temperatures have increased faster over land than over the ocean since 1850-1900, with warming to 2011-2020 of 1.59 [1.34 to 1.83] &deg;C over land and 0.88 [0.68 to 1.01] &deg;C over the ocean. {2.3.1, Cross-Chapter Box 2.3}
Global surface temperature during the period 1850-1900 is used as an approximation for pre-industrial conditions for consistency with AR5 and AR6 Special Reports, whilst recognizing that radiative forcings have a baseline of 1750 for the start of anthropogenic influences. It is likely that there was a net anthropogenic forcing of 0.0-0.3 Wm-2 in 1850-1900 relative to 1750 (medium confidence),
and from the period around 1750 to 1850-1900, there was a change in global surface temperature of around 0.1&deg;C (likely range
      -0.1 to +0.3&deg;C, medium confidence), with an anthropogenic component of 0.0&deg;C to 0.2&deg;C (likely range, medium confidence). {Cross-TS    Chapter Box 1.2, 7.3.5}
Global surface temperature has evolved over geological time (Figure TS.1, Box TS.2). Beginning approximately 6500 years ago, global surface temperature generally decreased, culminating in the coldest multi-century interval of the post-glacial period (since roughly 7000 years ago), which occurred between around 1450 and 1850 (high confidence). Over the last 50 years, global surface temperature has increased at an observed rate unprecedented in at least the last two thousand years (high confidence). Temperatures as high as during the most recent decade (2011-2020) exceed the warmest centennial-scale range reconstructed for the present interglacial, around 6500 years ago [0.2&deg;C to 1&deg;C] (medium confidence). The next most recent warm period was about 125,000 years ago during the Last Interglacial when the multi-centennial temperature range [0.5&deg;C to 1.5&deg;C] encompasses the 2011-2020 values (medium confidence) (Cross-Section Box TS.1, Figure 1). During the mid-Pliocene Warm Period, around 3.3-3.0 million years ago, global surface temperature was 2.5&deg;C to 4&deg;C warmer (medium confidence). {2.3.1, Cross-Chapter Box 2.1 and 2.4}
Current Warming There is very high confidence that the CMIP6 model ensemble reproduces observed global surface temperature trends and variability since 1850 with errors small enough to allow for detection and attribution of human-induced warming. The CMIP6 multi-model mean global surface warming between 1850-1900 and 2010-2019 is close to the best estimate of observed warming, though some CMIP6 models simulate a warming that is outside the assessed very likely observed range. {3.3.1}
The likely range of human-induced change in global surface temperature in 2010-2019 relative to 1850-1900 is 0.8&deg;C to 1.3&deg;C, with a central estimate of 1.07&deg;C (Figure Cross-Section Box TS.1, Figure 1), encompassing the best estimate of observed warming for that period, which is 1.06&deg;C with a very likely range of [0.88&deg;C to 1.21&deg;C], while the likely range of the change attributable to natural forcing is only -0.1&deg;C to +0.1&deg;C. This assessment is consistent with an estimate of the human-induced global surface temperature rise based on assessed ranges of perturbations to the top of the atmosphere (effective radiative forcing) and with metrics of feedbacks of the climate response (equilibrium climate sensitivity and the transient climate response). Over the same period, well-mixed greenhouse gas forcing likely warmed global surface temperature by 1.0&deg;C to 2.0&deg;C, while aerosols and other anthropogenic forcings likely cooled global surface temperature by 0.0&deg;C to 0.8&deg;C. {2.3.1, 3.3.1, 7.3.5, Cross-Chapter Box 7.1}
The observed slower increase in global surface temperature (relative to preceding and following periods) in the 1998-2012 period, sometimes referred to as the hiatus, was temporary (very high confidence). The increase in global surface temperature during the 1998-2012 period is also greater in the data sets used in the AR6 assessment than in those available at the time of AR5. Using these updated observational data sets and a like-for-like consistent comparison of simulated and observed global surface temperature, all observed estimates of the 1998-2012 trend lie within the very likely range of CMIP6 trends. Furthermore, the heating of the climate system continued during this period, as reflected in the continued warming of the global ocean (very high confidence) and in the continued rise of hot extremes over land (medium confidence). Since 2012, global surface temperature has risen strongly, with the past five years (2016-2020) being the hottest five-year period between 1850 and 2020 (high confidence). {2.3.1, 3.3.1, 3.5.1, Cross-Chapter Box 3.1}
Future Changes in Global Surface Temperature The AR6 assessment of future change in global surface temperature is, for the first time in an IPCC report, explicitly constructed by combining new projections for the SSP scenarios with observational constraints based on past simulated warming as well as the AR6-updated assessment of equilibrium climate sensitivity and transient climate response. In addition, climate forecasts initialized from the observed climate state have been used for the period 2019-2028. The inclusion of additional lines of evidence has reduced the assessed uncertainty ranges for each scenario (Cross-Section Box TS.1, Figure 1). {4.3.1, 4.3.4, Box 4.1, 7.5}
60
 
Technical Summary Changes in surface temperature (a) Global surface temperatures are more likely than not unprecedented in the past 125,000 years Kaufman et al.                                    The latest decade was warmer than any multi-century period 1.5    Pages 2k                                                after the Last Interglacial, around 125,000 years ago AR6 mean Over the last 50 years, global temperature Global surface temperature 1.0                                                              has increased at a rate unprecedented in at least the last 2000 years 0.5 relative to 1850-1900 (&deg;C)                                                                                                                                                                                                                                Latest decade 0.0 Mid-Holocene 0.5                                                                                                                                                                                                                                                                                      TS 10,000        6000          2000          1000              1400                                                                                                        1800 1900                            2000    Last Interglacial Year (BCE)                                                                                                                                                                                                        Year (CE)
Age:                                        Agricultural                                    Historical                                                                                                                          Industrial Resolution:                                  Centuries                                      Decades                                                                                                                              Annual (b) Observed and projected warming are stronger over                                                                                                        (c) Global surface temperature has risen more than 1&deg;C land than oceans, and strongest in the Arctic                                                                                                                from 1850-1900 1.0                                          HadCRUT .5.0 NOAAGlobalTemp Kadow et al. Berkeley Earth 1981-2020                                                                                                    0.5 Global surface temperature relative to 1850-1900 (&deg;C) 0.0 0.5 1850                                                1900              1950                2000 (d) Internal variability will influence near-term warming rates 2.5                                              CMIP6 historical HadCRUT5 assessed SSP1-2.6 2.0                                              assessed SSP2-4.5 assessed SSP3-7.0 1.5                                              models 1-4 0.6                    0.4  0.2    0.1      0.0    0.1      0.2  0.4      0.6                                                                1.0 Colour Significant                            Trend (&deg;C per decade) 0.5 Non significant 0.0 1980                                          1990    2000      2010    2020      2030      2040            2050 SSP3-7.0 (2081-2100)                                                                                          (e) Warming to 2100 depends on the scenario 5
GSAT 2081-2100 relative to 1995-2014 (&deg;C)
SSP1-1.9 SSP1-2.6 SSP2-4.5                                                        5 4      SSP3-7.0 Relative to 1850-1900 (&deg;C)
SSP5-8.5 4
3 3
2 1                                                                      2 6 4                        3 2        1 0.5 0        0.5    1    2    3    4    6                                                                                                                  1.5 &deg;C Total change (&deg;C) 1 0
2000-2019                          2020-2039 2040-2059          2060-2079          2081-2100 Cross-Section Box TS.1, Figure 1 l Earths surface temperature history and future with key findings annotated within each panel.
61
 
Technical Summary Cross-Section Box TS.1 (continued)
Cross-Section Box TS.1, Figure 1 (continued): The intent of this figure is to show global surface temperature observed changes from the Holocene to now, and projected changes. (a) Global surface temperature over the Holocene divided into three time scales: (i) 12,000 to 1000 years ago (10,000 BCE to 1000 CE) in 100-year time steps, (ii) 1000 to 1900 CE, 10-year smooth, and (iii) 1900 to 2020 CE (mean of four datasets in panel c). Bold lines show the median of the multi-method reconstruction, with 5% and 95% percentiles of the ensemble members (thin lines). Vertical bars are 5-95th percentile ranges of estimated global surface temperature for the Last Interglacial and mid-Holocene (medium confidence) (Section 2.3.1.1). All temperatures are relative to 1850-1900. (b) Spatially resolved trends (&deg;C per decade) for (upper map) HadCRUTv5 over 1981-2020, and (lower map, total change) multi-model mean projected changes from 1995-2014 to 2081-2010 in the SST3-7.0 scenario. Observed trends have been calculated where data are present in both the first and last decade and for at least 70% of all years within the period using ordinary least squares. Significance is assessed with autoregressive AR(1) model correction and denoted by stippling. Hatched areas in the lower map show areas of conflicting model evidence on significance of changes. (c) Temperature from instrumental data for 1850-2020, including annually resolved averages for the four global surface temperature datasets assessed in Section 2.3.1.1.3 (see text for references). The grey shading shows the uncertainty associated with the HadCRUTv5 estimate. All temperatures are relative to the 1850-1900 reference period. (d) Recent past and 2015-2050 evolution of annual mean global surface temperature change relative to 1850-1900, from HadCRUTv5 (black), Coupled Model Intercomparison Project Phase 6 (CMIP6) historical simulations (up to 2014, in grey, ensemble mean solid, 5% and 95% percentiles dashed, individual models thin), and CMIP6 projections under scenario SSP2-4.5, from four models that have an equilibrium climate sensitivity near the assessed central value (thick yellow). Solid thin coloured lines show the assessed central estimate of 20-year change in global surface temperature for 2015-2050 under three scenarios, and dashed thin coloured lines the corresponding 5% and 95% quantiles. (e)
TS    Assessed projected change in 20-year running mean global surface temperature for five scenarios (central estimate solid, very likely range shaded for SSP1-2.6 and SSP3-7.0), relative to 1995-2014 (left y-axis) and 1850-1900 (right y-axis). The y-axis on the right-hand side is shifted upward by 0.85&deg;C, the central estimate of the observed warming for 1995-2014, relative to 1850-1900. The right y-axis in (e) is the same as the y-axis in (d). {2.3, 4.3, 4.4}
During the near term (2021-2040), a 1.5&deg;C increase in global surface temperature, relative to 1850-1900, is very likely to occur in scenario SSP5-8.5, likely to occur in scenarios SSP2-4.5 and SSP3-7.0, and more likely than not to occur in scenarios SSP1-1.9 and SSP1-2.6. The time of crossing a warming level is defined here as the midpoint of the first 20-year period during which the average global surface temperature exceeds the level. In all scenarios assessed here except SSP5-8.5, the central estimate of crossing the 1.5&deg;C level lies in the early 2030s. This is in the early part of the likely range (2030-2052) assessed in SR1.5, which assumed continuation of the then-current warming rate; this rate has been confirmed in the AR6. Roughly half of this difference arises from a larger historical warming diagnosed in AR6. The other half arises because for central estimates of climate sensitivity, most scenarios show stronger warming over the near term than was estimated as current in SR1.5 (medium confidence). When considering scenarios similar to SSP1-1.9 instead of linear extrapolation, the SR1.5 estimate of when 1.5&deg;C global warming is crossed is close to the central estimate reported here. (Cross-Section Box TS.1, Table 1) {2.3.1, Cross-Chapter Box 2.3, 3.3.1, 4.3.4, Box 4.1}
It is more likely than not that under SSP1-1.9, global surface temperature relative to 1850-1900 will remain below 1.6&deg;C throughout the 21st century, implying a potential temporary overshoot of 1.5&deg;C global warming of no more than 0.1&deg;C. If climate sensitivity lies near the lower end of the assessed very likely range, crossing the 1.5&deg;C warming level is avoided in scenarios SSP1-1.9 and SSP1-2.6 (medium confidence). Global surface temperature in any individual year, in contrast to the 20-year average, could by 2030 exceed 1.5&deg;C relative to 1850-1900 with a likelihood between 40% and 60%, across the scenarios considered here (medium confidence).
(Cross-Section Box TS.1, Table 1) {4.3.4, 4.4.1, Box 4.1, 7.5}
During the 21st century, a 2&deg;C increase in global surface temperature relative to 1850-1900 will be crossed under SSP5-8.5 and SSP3-7.0, is extremely likely to be crossed under SSP2-4.5, but is unlikely to be crossed under SSP1-2.6 and extremely unlikely to be crossed under SSP1-1.9. For the mid-term period 2041-2060, this 2&deg;C global warming level is very likely to be crossed under SSP5-8.5, likely to be crossed under SSP3-7.0, and more likely than not to be crossed under SSP2-4.5. (Cross-Section Box TS.1, Table 1) {4.3.4}
Events of reduced and increased global surface temperature trends at decadal time scales will continue to occur in the 21st century but will not affect the centennial-scale warming (very high confidence). If strong mitigation is applied from 2020 onward as reflected in SSP1-1.9, its effect on 20-year trends in global surface temperature would likely emerge during the near term (2021-2040),
measured against an assumed non-mitigation scenario such as SSP3-7.0 or SSP5-8.5. All statements about crossing the 1.5&deg;C level assume that no major volcanic eruption occurs during the near term (Cross-Section Box TS.1, Table 1). {2.3.1, Cross-Chapter Box 2.3, 4.3.4, 4.4.1, 4.6.3, Box 4.1}
Compared to 1850-1900, average global surface temperature over the period 2081-2100 is very likely to be higher by [1.0&deg;C to 1.8&deg;C] in the low CO2 emissions scenario SSP1-1.9 and by [3.3&deg;C to 5.7&deg;C] in the high CO2 emissions scenario SSP5-8.5. For the scenarios SSP1-2.6, SSP2-4.5, and SSP3-7.0, the corresponding very likely ranges are [1.3&deg;C to 2.4&deg;C], [2.1&deg;C to 3.5&deg;C], and [2.8&deg;C to 4.6&deg;C], respectively. The uncertainty ranges for the period 2081-2100 continue to be dominated by the uncertainty in equilibrium climate sensitivity and transient climate response (very high confidence) (Cross-Section Box TS.1, Table 1). {4.3.1, 4.3.4, 4.4.1, 7.5}
The CMIP6 models project a wider range of global surface temperature change than the assessed range (high confidence); furthermore, the CMIP6 global surface temperature increase tends to be larger than that in CMIP5 (very high confidence). {4.3.1, 4.3.4, 4.6.2, 7.5.6}
62
 
Technical Summary Cross-Section Box TS.1 (continued)
Cross-Section Box TS.1, Table 1 l Assessment results for 20-year averaged change in global surface temperature based on multiple lines of evidence. The change is displayed in &deg;C relative to the 1850-1900 reference period for selected time periods (first three rows), and as the first 20-year period during which the average global surface temperature change exceeds the specified level relative to the period 1850-1900 (last four rows). The entries give both the central estimate and, in parentheses, the very likely (5-95%) range. An entry n.c. means that the global warming level is not crossed during the period 2021-2100.
SSP1-1.9                  SSP1-2.6                    SSP2-4.5                  SSP3-7.0                      SSP5-8.5 Near term, 1.5 [1.2 to 1.7]            1.5 [1.2 to 1.8]            1.5 [1.2 to 1.8]          1.5 [1.2 to 1.8]              1.6 [1.3 to 1.9]
2021-2040 Mid-term, 1.6 [1.2 to 2.0]            1.7 [1.3 to 2.2]            2.0 [1.6 to 2.5]          2.1 [1.7 to 2.6]              2.4 [1.9 to 3.0]
2041-2060 Long term, 1.4 [1.0 to 1.8]            1.8 [1.3 to 2.4]            2.7 [2.1 to 3.5]          3.6 [2.8 to 4.6]              4.4 [3.3 to 5.7]
2081-2100 1.5&deg;C 2025-2044                  2023-2042                    2021-2040                  2021-2040                    2018-2037            TS
[2013-2032 to n.c.]        [2012-2031 to n.c.]      [2012-2031 to 2037-2056]    [2013-2032 to 2033-2052]      [2011-2030 to 2029-2048]
n.c.                      n.c.                    2043-2062                  2037-2056                    2032-2051 2&deg;C
[n.c. to n.c.]        [2031-2050 to n.c.]      [2028-2047 to 2075-2094]    [2026-2045 to 2053-2072]      [2023-2042 to 2044-2063]
n.c.                      n.c.                        n.c.                  2066-2085                    2055-2074 3&deg;C
[n.c. to n.c.]            [n.c. to n.c.]          [2061-2080 to n.c.]        [2050-2069 to n.c.]      [2042-2061 to 2074-2093]
n.c.                      n.c.                        n.c.                      n.c.                      2075-2094 4&deg;C
[n.c. to n.c.]            [n.c. to n.c.]              [n.c. to n.c.]          [2070-2089 to n.c.]          [2058-2077 to n.c.]
TS.2        Large-scale Climate Change: Mean                                                substantial reductions in global GHG emissions. Continued Climate, Variability and Extremes                                                GHG emissions greatly increase the likelihood of potentially irreversible changes in the global climate system (Box TS.9),
This section summarizes knowledge about observed and projected                                in particular with respect to the contribution of ice sheets large-scale climate change (including variability and extremes),                              to global sea level change (high confidence). {2.3, 3.8, 4.3, drivers and attribution of observed changes to human activities. It                          4.6, 4.7, 7.2-7.4, Cross-Chapter Box 7.1, 9.2-9.6}
describes observed and projected large-scale changes associated with major components of the climate system: atmosphere, ocean                          Earth system model simulations of the historical period since 1850 (including sea level change), land, biosphere and cryosphere, and the                    are only able to reproduce the observed changes in key climate carbon, energy and water cycles. In each subsection, reconstructed                      indicators when anthropogenic forcings are included (Figure TS.7).
past changes, observed and attributed recent changes, and projected                      Taken together with numerous formal attribution studies across an near- and long-term changes to mean climate, variability and                            even broader range of indicators and theoretical understanding, extremes are presented, where possible, in an integrated way. See                        this underpins the unequivocal attribution of observed warming of Section TS.1.3.1 for information on the scenarios used for projections.                  the atmosphere, ocean, and land to human influence (Table TS.1).
{2.3, 3.8}
TS.2.1      Changes Across the Global Climate System In addition to global surface temperature (Cross-Section Box TS.1), a wide range of indicators across all components of the climate system are changing rapidly (Figure TS.7), with many at levels unseen in millennia. The observed changes provide a coherent picture of a warming world, many aspects of which have now been formally attributed to human influence, and human influence on the atmosphere, ocean, and land components of the climate system, taken together, is assessed as unequivocal for the first time in an IPCC assessment report (Table TS.1, Figure TS.7).
It is virtually certain that global surface temperature rise and associated changes can be limited through rapid and 63
 
Technical Summary Global Near-surface air temperature over land                                  Near-surface air temperature                                Ocean heat content Near-surface air temperature                                                                        Sea ice North America                                                  Asia                                                Arctic TS Central and South America                                          Australasia                                            Antarctic Europe and North Africa                                          Antarctic                                        Precipitation Global Africa                                                  Arctic 60&deg;N-90&deg;N anthropogenic + natural              natural        observations Figure TS.7 l Simulated and observed changes compared to the 1995-2014 average in key large-scale indicators of climate change across the climate system, for continents, ocean basins and globally up to 2014. The intent of this figure is to compare the observed and simulated changes over the historical period for a range of variables and regions, with and without anthropogenic forcings, for attribution. Black lines show observations, orange lines and shading show the multi-model mean and 5-95th percentile ranges for Coupled Model Intercomparison Project Phase 6 (CMIP6) historical simulations including anthropogenic and natural forcing, and green lines and shading show corresponding ensemble means and 5-95th percentile ranges for CMIP6 natural-only simulations. Observations after 2014 (including, for example, a strong subsequent decrease of Antarctic sea ice area that leads to no significant overall trend since 1979) are not shown because the CMIP6 historical simulations end in 2014.
A 3-year running mean smoothing has been applied to all observational time series. {3.8, Figure 3.41}
64
 
Technical Summary Table TS.1 l Assessment of observed changes in large-scale indicators of mean climate across climate system components and their attribution to human influence. The colour coding indicates the assessed confidence in/likelihood of the human contribution as a driver or main driver19 (main driver is specified in that case) where available (see colour key). Otherwise, explanatory text is provided in cells with white background. The relevant chapter section with more detailed information is listed in each table cell.
Change in Indicator                                        Observed Change Assessment                            Human Contribution Assessment Atmosphere and Water Cycle Warming of global mean surface air temperature since                                                                      Likely range of human contribution (0.8&deg;C-1.3&deg;C) encom-
{2.3.1, Cross-Chapter Box 2.3}
1850-1900                                                                                                                passes observed warming (0.9&deg;C-1.2&deg;C) {3.3.1}
Warming of the troposphere since 1979                          {2.3.1}                                                  Main driver {3.3.1}
Cooling of the lower stratosphere                              Since mid-20th century {2.3.1}                            Main driver 1979-mid-1990s {3.3.1}
Large-scale precipitation and upper troposphere humidity
{2.3.1}                                                  {3.3.2, 3.3.3}
changes since 1979 Expansion of the zonal mean Hadley Circulation since
{2.3.1}                                                  Southern Hemisphere {3.3.3}
the 1980s Ocean                                                                                                                                                                                TS Ocean heat content increase since the 1970s                    {2.3.3, 2.3.4, 9.2.1, Cross-Chapter Box 9.1}              Main driver {3.5.1}
Salinity changes since the mid-20th century                    {2.3.3, 2.3.4, 9.2.2}                                    {3.5.2}
Global mean sea level rise since 1971                          {2.3.3, 9.6.1}                                            Main driver {3.5.3}
Cryosphere Arctic sea ice loss since 1979                                  {2.3.2, 9.3.1}                                            Main driver {3.4.1}
Reduction in Northern Hemisphere spring snow cover since
{2.3.2, 9.5.3}                                            {3.4.2}
1950 Greenland Ice Sheet mass loss since 1990s                      {2.3.2, 9.4.1}                                            {3.4.3}
Antarctic Ice Sheet mass loss since 1990s                      {2.3.2, 9.4.2}                                            Limited evidence and medium agreement {3.4.3}
Retreat of glaciers                                            {2.3.2, 9.5.1}                                            Main driver {3.4.3}
Carbon Cycle Increased amplitude of the seasonal cycle of atmospheric
{2.3.4}                                                  Main driver {3.6.1}
CO2 since the early 1960s Acidification of the global surface ocean                      {SROCC, 5.3.2, Cross-Chapter Box 5.3}                    Main driver {3.6.2}
Land Climate (Extremes, see Table TS.12)
Mean 2 m land warming since 1850-1900 (about 40%
{2.3.1}                                                  Main driver {3.3.1}
larger than global mean warming)
Synthesis Warming of the global climate system since
{2.3.5}                                                  {3.8.1}
pre-industrial times See text description          medium confidence              likely/high confidence            very likely    extremely likely          virtually certain        fact Future climate change across a range of atmospheric, cryospheric,                                  centuries to millennia. Furthermore, it is likely that at least one large oceanic and biospheric indicators depends upon future emissions                                    volcanic eruption will occur during the 21st century. Such an eruption pathways. Outcomes for a broad range of indicators increasingly                                    would reduce global surface temperature for several years, decrease diverge through the 21st century across the different SSPs (Section                                land precipitation, alter monsoon circulation and modify extreme TS.1.3.1, Figure TS.8). Due to the slow response of the deep ocean                                  precipitation, at both global and regional scales. {4.3, 4.7, 9.4, 9.6, and ice sheets, this divergence continues long after 2100, and 21st                                Cross-Chapter Box 4.1}
century emissions choices will have implications for GMSL rise for 19    Throughout this Technical Summary, main driver means responsible for more than 50% of the change.
65
 
Technical Summary Recent and future change of four key indicators of the climate system Atmospheric temperature, ocean heat content, Arctic summer sea ice, and land precipitation (a) Global surface air temperature                                                          (b) Global ocean heat content and thermosteric sea level 4
CMIP6    Emulator Future (assessed)                              3 3                                                                                                                                        Future (assessed)          0.3 2                                                                                          2                                                                      0.2 m
    &deg;C                                                                                                    YJ  1 1                                                                                                          Past (observed)                                          0.1 Past (observed) 0  Past (simulated)                                                                        0 Past (simulated)                                                      0
                  -1                                                                                          -1                                                                      -0.1 1950                    2000      [2000/        [2040/          [2080/ Change          1950                  2000                  2050                    2100        Change 2019]          2059]            2100] in 2100                                                                                            in 2100 (c) Arctic September sea ice area                                                          (d) Global land precipitation 10                                                                                        20 TS                  8 Past (simulated)                                                                                                                              Future (CMIP6)
Past (observed) 10 6
million km2 Past (observed)
Future (CMIP6)                  %
4                                                                                          0 2
Practically sea ice free                                                                  Past (simulated) 0                                                                                      -10 1950                      2000                2050                  2100 Change          1950                      2000                2050                2100 Change in 2100 in 2100 Past (simulated); 5-95% range        Future (SSP1-1.9) mean Past (observed)                      Future (SSP1-2.6) mean; 5-95% range Future (SSP2-4.5) mean Future (SSP3-7.0) mean; 5-95% range Future (SSP5-8.5) mean Figure TS.8 l Observed, simulated and projected changes compared to the 1995-2014 average in four key indicators of the climate system through to 2100 differentiated by Shared Socio-economic Pathway (SSP) scenario. The intent of this figure is to show how future emissions choices impact key, iconic large-scale indicators and to highlight that our collective choices matter. Past simulations are based on the Coupled Model Intercomparison Project Phase 6 (CMIP6) multi-model ensemble.
Future projections are based on the assessed ranges based upon multiple lines of evidence for (a) global surface temperature (Cross-Section Box TS.1) and (b) global ocean heat content and the associated thermosteric sea level contribution to global mean sea level change (right-hand axis) using a climate model emulator (Cross-Chapter Box 7.1), and CMIP6 simulations for (c) Arctic September sea ice and (d) global land precipitation. Projections for SSP1-1.9 and SSP1-2.6 show that reduced greenhouse gas emissions lead to a stabilization of global surface temperature, Arctic sea ice area and global land precipitation over the 21st century. Projections for SSP1-2.6 show that emissions reductions have the potential to substantially reduce the increase in ocean heat content and thermosteric sea level rise over the 21st century but that some increase is unavoidable. The brackets in the x axis in panel (a) indicate assessed 20-year-mean periods. {4.3, Figure 4.2, 9.3, 9.6, Figure 9.6}
Observational records show changes in a wide range of climate                                          human influence is the main contributor to the observed increase extremes that have been linked to human influence on the climate                                      (decrease) in the likelihood and severity of hot (cold) extremes (Table system (Table TS.2). In many cases, the frequency and intensity of                                    TS.2). The frequency of extreme temperature and precipitation events future changes in extremes can be directly linked to the magnitude                                    in the current climate will change with warming, with warm extremes of future projected warming. Changes in extremes have been                                            becoming more frequent (virtually certain), cold extremes becoming widespread over land since the 1950s, including a virtually certain                                    less frequent (extremely likely) and precipitation extremes becoming global increase in extreme air temperatures and a likely intensification                              more frequent in most locations (very likely). {9.6.4, 11.2, 11.3, 11.4, in global-scale extreme precipitation. It is extremely likely that                                    11.6, 11.7, 11.8, 11.9, Box 9.2}
66
 
Technical Summary Table TS.2 l Summary table on observed changes in extremes, their attribution since 1950 (except where stated otherwise), and projected changes at
+1.5&deg;C, +2&deg;C and +4&deg;C of global warming, on global and continental scales. An increase in warm/hot extremes refers to warmer and/or more frequent hot days and nights and warm spells/heatwaves, over most land areas. A decrease in cold extremes refers to warmer and/or fewer cold days and nights and cold spells/cold waves, over most land areas. Drought events are relative to a predominant fraction of land area. For tropical cyclones, observed changes and attribution refer to Categories 3-5, while projected changes refer to Categories 4-5. Tables 11.1 and 11.2 are more detailed versions of this table, containing, in particular, information on regional scales. In general, higher warming levels also imply stronger projected changes for indicators where the confidence level does not depend on the warming level and the table does not explicitly quantify the global sensitivity. See also Box TS.10. {9.6, Box 9.2, 11.3, 11.7}
Observed                      Attributed                                          Projected at GWL (&deg;C)
Change in Indicator (since 1950)                  (since 1950)                      +1.5                            +2                            +4 Warm/hot extremes: Frequency or intensity                                                                      Main driver Cold extremes: Frequency or intensity                                            Main driver Main driver of the TS Heavy precipitation events: Frequency,        Over majority of land observed intensification intensity and/or amount                        regions with good                                                          in most land regions                          in most land regions of heavy precipitation in observational coverage land regions Agricultural and ecological droughts:                                                                        in more regions              in more regions                in more regions Intensity and/or frequency                      in some regions                in some regions            compared to observed          compared to 1.5&deg;C              compared to 2&deg;C of changes                of global warming                global warming Precipitation associated with tropical cyclones                                                                                                Rate +11%                      Rate +14%                      Rate +28%
Tropical cyclones: Proportion of intense cyclones                                                                                                +10%                          +13%                          +20%
Compound events: Co-occurrent heatwaves and droughts                              (Frequency)                    (Frequency)                            (Frequency and intensity increases with warming)
Marine heatwaves: Intensity & frequency (since 1900)                  (since 2006)                                    Strongest in tropical and Arctic Ocean Extreme sea levels: Frequency (since 1960)                                                              (Scenario-based assessment for 21st century) medium confidence            likely/high confidence              very likely            extremely likely              virtually certain TS.2.2          Changes in the Drivers of the Climate System                                          The total anthropogenic effective radiative forcing (ERF) in 2019, relative to 1750, was 2.72 [1.96 to 3.48] W m-2 Since 1750, changes in the drivers of the climate system                                          (medium confidence) and has likely been growing at an are dominated by the warming influence of increases in                                            increasing rate since the 1970s. {2.2, 6.4, 7.2, 7.3}
atmospheric GHG concentrations and a cooling influence from aerosols, both resulting from human activities. In                                        Solar activity since 1900 was high but not exceptional compared to comparison there has been negligible long-term influence                                      the past 9000 years (high confidence). The average magnitude and from solar activity and volcanoes. Concentrations of CO2,                                      variability of volcanic aerosols since 1900 has not been unusual methane (CH4), and nitrous oxide (N2O) have increased to                                      compared to at least the past 2500 years (medium confidence).
levels unprecedented in at least 800,000 years, and there                                      However, sporadic strong volcanic eruptions can lead to temporary is high confidence that current CO2 concentrations have not                                    drops in global surface temperature lasting 2-5 years. {2.2.1, 2.2.2, been experienced for at least 2 million years. Global mean                                    2.2.8, Cross-Chapter Box 4.1}
concentrations of anthropogenic aerosols peaked in the late 20th century and have slowly declined since in northern                                  Atmospheric CO2 concentrations have changed substantially mid-latitudes, although they continue to increase in South                                    over millions of years (Figure TS.1). Current levels of atmospheric Asia and East Africa (high confidence).                                                        CO2 have not been experienced for at least 2 million years (high 67
 
Technical Summary (a) Last time CO2 levels were as high as present was at least 2 million years ago 11B-foraminifera 13C-alkenone 450 Ant. ice core 409.9 350 CO2 (ppm) 250 150 3.5          3.0          2.5            2.0            1.5                          1.0      0.5          0.0 Age (millions of years ago)
TS                                                                                                                  (c) Since 1960-1980 several high-accuracy global networks measure (b) Information from multiple ice cores depicts a strong                                          surface concentrations of CO2, CH4, and N2O. Current concentrations increase of CO2, CH4, and N2O since the 19th century                                              are higher than measured in ice cores during the last 800,000 years 1960-2019                                                                    1960-2019 420 409.9                                        420 400      WAIS Divide 400 409.9 380      Law Dome CO2 (ppm) 380          NOAA CO2 (ppm) 360      EDML 360          CSIRO 340                                                                                                                        SIO 340 320 300                                                              1866.3                                        320 1800                                      300                                                                        1866.3 280                                                                                                                                                                                      1800 1600                                      280 1600 CH4 (ppb) 1400 CH4 (ppb)
GISP2                                                                                                                                1400 1200 WAIS Divide                                                                                                                              1200 1000 Law Dome                                                                                                                          NOAA 1000 800                                                                                                          CSIRO AGAGE 800 600                                                                                                          UCI 340                                                                                                                                                                                      600 332.1                                        340 320                                                                                                                                                                                        332.1 320 N2O (ppb)
GISP2 N2O (ppb) 300      NEEM                                                                                                              NOAA H15                                                    300          CSIRO Law Dome Styx                      EURO                                                                          AGAGE 280 280 260 260 0            400            800    1200      1600      2000                                                  1960    1970  1980    1990    2000              2010            2020 Year (CE)
(d) The increase in effective radiative forcing (ERF) since the late 19th century is driven predominantly by warming GHGs and cooling aerosol. ERF is changing at a faster rate since the 1970s Carbon dioxide (CO2)        Halogenated gases                Volcanic 2          Methane (CH4)              Tropospheric Aerosol              Solar Nitrous oxide (N2O)        Other anthropogenic              Total Ozone (O3) 0 W m2 2
(W m2 per decade1) 0.5 0.4 Rate of change anthropogenic ERF                                  0.3 4                                                                                                                                            0.2 0.1 0.0 1750                      1800              1850                        1900                            1950            2000 Figure TS.9 l Changes in well-mixed greenhouse gas (WMGHG) concentrations and effective radiative forcing (EFR). The intent of this figure is to show that the changes of the main drivers of climate system over the industrial period are exceptional in a long-term context. (a) Changes in carbon dioxide (CO2) from proxy records over the past 3.5 million years. (b) Changes in all three WMGHGs from ice core records over the Common Era. (c) Directly observed WMGHG changes since the mid-20th century. (d)
Evolution of ERF and components since 1750. Further details on data sources and processing are available in the associated FAIR data table. {2.2, Figures 2.3, 2.4 and 2.10}
68
 
Technical Summary confidence, Figure TS.9a). Over 1750-2019, CO2 increased by 131.6      the late 20th century, with low confidence in the magnitude of
+/- 2.9 ppm (47.3%). The centennial rate of change of CO2 since 1850      post-2014 changes due to conflicting evidence (Section TS.3.1).
has no precedent in at least the past 800,000 years (Figure TS.9),      {2.2.6, 6.2.1, 6.3.5, 6.4.1, 7.3.3}
and the fastest rates of change over the last 56 million years were at least a factor of four lower (low confidence) than over 1900-        There is high confidence that tropospheric ozone has been increasing 2019. Several networks of high-accuracy surface observations            from 1750 in response to anthropogenic changes in ozone precursor show that concentrations of CO2 have exceeded 400 ppm, reaching        emissions (nitrogen oxides, carbon monoxide, non-methane volatile 409.9 (+/- 0.3) ppm in 2019 (Figure TS.9c). The ERF from CO2 in 2019      organic compounds, and methane), but with medium confidence in (relative to 1750) was 2.16 Wm-2. {2.2.3, 5.1.2, 5.2.1, 7.3}            the magnitude of this change, due to limited observational evidence and knowledge gaps. Since the mid-20th century, tropospheric By 2019, concentrations of CH4 reached 1866.3 (+/- 3.3) ppb              ozone surface concentrations have increased by 30-70% across the (Figure TS.9c). The increase since 1750 of 1137 +/- 10 ppb (157.8%) far  Northern Hemisphere (medium confidence); since the mid-1990s, exceeds the range over multiple glacial-interglacial transitions of the free tropospheric ozone has increased by 2-7% per decade in most past 800,000 years (high confidence). In the 1990s, CH4 concentrations  northern mid-latitude regions and 2-12% per decade in sampled plateaued, but started to increase again around 2007 at an average      tropical regions. Future changes in surface ozone concentrations will  TS rate of 7.6 +/- 2.7 ppb yr -1 (2010-2019; high confidence). There is high be primarily driven by changes in precursor emissions rather than confidence that this recent growth is largely driven by emissions      climate change (high confidence). Stratospheric ozone has declined from fossil fuel exploitation, livestock, and waste, with ENSO driving  between 60&deg;S-60&deg;N by 2.2% from 1964-1980 to 2014-2017 multi-annual variability of wetland and biomass burning emissions.      (high confidence), with the largest declines during 1980-1995.
In 2019, ERF from CH4 was 0.54 Wm-2. {2.2.3, 5.2.2, 7.3}                The strongest loss of stratospheric ozone continues to occur in austral spring over Antarctica (ozone hole), with emergent signs of Since 1750, N2O increased by 62.0 +/- 6.0 ppb, reaching a level of        recovery after 2000. The 1750-2019 ERF for total (stratospheric and 332.1 (+/- 0.4) ppb in 2019. The increase since 1750 is of comparable    tropospheric) ozone is 0.47 [0.24 to 0.71] W m2, which is dominated magnitude to glacial-interglacial fluctuations of the past              by tropospheric ozone changes. {2.2.5, 6.3.2, 7.3.2, 7.3.5}
800,000 years (Figure TS.9c). N2O concentration trends since 1980 are largely driven by a 30% increase in emissions from the expansion    The global mean abundance of hydroxyl (OH) radical, or oxidizing and intensification of global agriculture (high confidence). By 2019    capacity, chemically regulates the lifetimes of many SLCFs, and its ERF was 0.21 W m-2. {2.2.3, 5.2.3}                                  therefore the radiative forcing of CH4, ozone, secondary aerosols and many halogenated species. Model estimates suggest no Halogenated gases consist of chlorofluorocarbons (CFCs),                significant change in oxidizing capacity from 1850 to 1980 (low hydrochlorofluorocarbons (HCFCs), hydrofluorocarbons (HFCs) and        confidence). Increases of about 9% over 1980-2014 computed by other gases, many of which can deplete stratospheric ozone and          ESMs and carbon cycle models are not confirmed by observationally warm the atmosphere. In response to controls on production and          constrained inverse models, rendering an overall medium confidence consumption mandated by the Montreal Protocol on Substances            in stable OH or positive trends since the 1980s, and implying that OH that Deplete the Ozone Layer and its amendments, the atmospheric        is not the primary driver of recent observed growth in CH4. {6.3.6, abundances of most CFCs have continued to decline since AR5.            Cross-Chapter Box 5.2}
Abundances of HFCs, which are replacements for CFCs and HCFCs, are increasing (high confidence), though increases of the major HCFCs  Land use and land-cover change exert biophysical and biogeochemical have slowed in recent years. The ERF from halogenated components        effects. There is medium confidence that the biophysical effects in 2019 was 0.41 Wm-2. {2.2.4, 6.3.4, 7.3.2}                            of land-use change since 1750, most notably the increase in global albedo, have had an overall cooling on climate, whereas Tropospheric aerosols mainly act to cool the climate system,            biogeochemical effects (i.e., changes in GHG and volatile organic directly by reflecting solar radiation, and indirectly by enhancing    compound emissions or sinks) led to net warming. Overall land-use cloud reflectance. Ice cores show increases in aerosols across the      and land-cover ERF is estimated at -0.2 [-0.3 to -0.1] W m2. {2.2.7, Northern Hemisphere mid-latitudes since 1700 and reductions            7.3.4, SRCCL Section 2.5}
since the late 20th century (high confidence). Aerosol optical depth (AOD), derived from satellite- and ground-based radiometers,            The total anthropogenic ERF in 2019 relative to 1750 was 2.72 [1.96 has decreased since 2000 over the mid-latitude continents of            to 3.48] W m2 (Figure TS.9), dominated by GHGs (positive ERF) and both hemispheres, but increased over South Asia and East Africa        partially offset by aerosols (negative ERF). The rate of change of ERF (high confidence). Trends in AOD are more pronounced from sub-          likely has increased since the 1970s, mainly due to growing CO2 micrometre aerosols for which the anthropogenic contribution is        concentrations and less negative aerosol ERF (Section TS.3.1). {2.2.8, particularly large. Global carbonaceous aerosol budgets and trends      7.3}
remain poorly characterized due to limited observations, but black carbon (BC), a warming aerosol component, is declining in several regions of the Northern Hemisphere (low confidence). Total aerosol ERF in 2019, relative to 1750, is 1.1 [1.7 to 0.4] W m2 (medium confidence) and more likely than not became less negative since 69
 
Technical Summary TS.2.3            Upper Air Temperatures and Atmospheric                                                  track in the Southern Hemisphere by 2100 under the high Circulation                                                                            CO2 emissions scenarios. It is likely that the proportion of intense tropical cyclones has increased over the last four The effects of human-induced climate change have                                                  decades and that this cannot be explained entirely by been clearly identified in observations of atmospheric                                            natural variability. There is low confidence in observed temperature and some aspects of atmospheric circulation,                                          recent changes in the total number of extratropical cyclones and these effects are likely to intensify in the future.                                          over both hemispheres. The proportion of tropical cyclones Tropospheric warming and stratospheric cooling are                                                that are intense is expected to increase (high confidence),
virtually certain to continue with continued net emissions                                        but the total global number of tropical cyclones is expected of greenhouse gases. Several aspects of the atmospheric                                            to decrease or remain unchanged (medium confidence).
circulation have likely changed since the mid-20th century,                                        {2.3, 3.3, 4.3, 4.4, 4.5, 8.3, 8.4, 11.7}
and human influence has likely contributed to the observed poleward expansion of the Southern Hemisphere Hadley Cell                                      The troposphere has warmed since at least the 1950s, and it is virtually and very likely contributed to the observed poleward shift                                      certain that the stratosphere has cooled. It is very likely that human-TS          of the Southern Hemisphere extratropical jet in summer. It                                      induced increases in GHGs were the main driver of tropospheric is likely that the mid-latitude jet will shift poleward and                                    warming since 1979. It is extremely likely that anthropogenic strengthen, accompanied by a strengthening of the storm                                        forcing, both from increases in GHG concentrations and depletion of Observed trends                                                                  Projected long-term change (2081-2100)
GNSS-RO 2002-2019                                                Multi-model projections (SSP1-2.6)                                  Multi-model projections (SSP3-7.0) 36                                                                    32 Pressure (hPa)                                                          Pressure (hPa)                                                Pressure (hPa)
Colour High model agreement
                      -0.5-0.4-0.3-0.2-0.1 0 0.1 0.2 0.3 0.4 0.5                                                          4 2 -1 0 1                      2  3  4  5          Low model agreement
                                      &deg;C per decade                                                                                                          &deg;C ERA5 1979-2019 (DJF)                                          Multi-model projections (DJF, SSP1-2.6)                              Multi-model projections (DJF, SSP3-7.0) 34                                                                  31 Pressure (hPa)                                                        Pressure (hPa)                                                Pressure (hPa)
Non-signficant trend Contour from -50 to 40 by 10 Contour from -50 to 40 by 10
                      -1.5  -1    -0.5    0      0.5      1    1.5                                                    5 3 1 0 1 2 3 4 5 6                          Colour High model agreement m/s per decade                                                                                                          m/s                            Low model agreement Figure TS.10 l Observed and projected upper air temperature and circulation changes. The intent of this figure is to visualize upper air temperature and circulation changes and the similarity between observed and projected changes. Upper panels: (Left) Zonal cross-section of temperature trends for 2002-2019 in the upper troposphere region for the ROM SAF radio-occultation dataset. (Middle) Change in the annual and zonal mean atmospheric temperature (&deg;C) in 2081-2100 in SSP1-2.6 relative to 1995-2014 for 36 Coupled Model Intercomparison Project Phase 6 (CMIP6) models. (Right) the same in SSP3-7.0 for 32 models. Lower panels: (Left) Long-term mean (thin black colour) and linear trend (colour) of zonal mean December-January-February (DJF) zonal winds for ERA5. (Middle) multi-model mean change in annual and zonal mean wind (m s-1) in 2081-2100 in SSP1-2.6 relative to 1995-2014 based on 34 CMIP6 models. The 1995-2014 climatology is shown in contours with spacing of 10 m s-1. (Right) the same for SSP3-7.0 for 31 models. {2.3.1; Figures 2.12 and 2.18; 4.5.1; Figure 4.2.6}
70
 
Technical Summary stratospheric ozone due to ozone-depleting substances, was the main      the Southern Hemisphere mid-latitude jet is likely to shift poleward driver of upper stratospheric cooling since 1979. It is very likely that and strengthen under the SSP5-8.5 scenario relative to 1995-2014, global mean stratospheric cooling will be larger for scenarios with      accompanied by an increase in the SAM (Section TS.4.2.2). It is higher atmospheric CO2 concentrations. In the tropics, since at least    likely that wind speeds associated with extratropical cyclones will 2001 (when new techniques permit more robust quantification), the        strengthen in the Southern Hemisphere storm track for SSP5-8.5.
upper troposphere has warmed faster than the near-surface (medium        There is low confidence in the potential role of Arctic warming and confidence) (Figure TS.10). There is medium confidence that most        sea ice loss on historical or projected mid-latitude atmospheric CMIP5 and CMIP6 models overestimate the observed warming in              variability. {2.3.1, 3.3.3, 3.7.2, 4.3.3, 4.4.3, 4.5.1, 4.5.3, 8.2.2, 8.3.2, the upper tropical troposphere over the period 1979-2014, in part        Cross-Chapter Box 10.1}
because they overestimate tropical SST warming. It is likely that future tropical upper tropospheric warming will be larger than at the    It is likely that the proportion of major (Category 3-5) tropical tropical surface. {2.3.1, 3.3.1, 4.5.1}                                  cyclones (TCs) and the frequency of rapid TC intensification events have increased over the past four decades. The average location The Hadley Circulation has likely widened since at least the 1980s,      of peak TC wind-intensity has very likely migrated poleward in predominantly in the Northern Hemisphere, although there is only        the western North Pacific Ocean since the 1940s, and TC forward              TS medium confidence in the extent of the changes. This has been            translation speed has likely slowed over the contiguous USA since accompanied by a strengthening of the Hadley Circulation in the          1900. It is likely that the poleward migration of TCs in the western Northern Hemisphere (medium confidence). It is likely that human        North Pacific and the global increase in TC intensity rates cannot be influence has contributed to the poleward expansion of the zonal        explained entirely by natural variability. There is high confidence mean Hadley cell in the Southern Hemisphere since the 1980s, which      that average peak TC wind speeds and the proportion of Category is projected to further expand with global warming (high confidence). 4-5 TCs will increase with warming and that peak winds of the There is medium confidence that the observed poleward expansion          most intense TCs will increase. There is medium confidence that the in the Northern Hemisphere is within the range of internal variability. average location where TCs reach their maximum wind-intensity will
{2.3.1, 3.3.3, 8.4.3}                                                    migrate poleward in the western North Pacific Ocean, while the total global frequency of TC formation will decrease or remain unchanged Since the 1970s, near-surface average winds have likely weakened        with increasing global warming. {11.7.1}
over land. Over the ocean, near-surface average winds likely strengthened over 1980-2000, but divergent estimates lead to            There is low confidence in observed recent changes in the total low confidence thereafter. Extratropical storm tracks have likely        number of extratropical cyclones over both hemispheres. There is also shifted poleward since the 1980s. There is low confidence in            low confidence in past-century trends in the number and intensity of projected poleward shifts of the Northern Hemisphere mid-latitude        the strongest extratropical cyclones over the Northern Hemisphere jet and storm tracks due to large internal variability and structural    due to the large interannual-to-decadal variability and temporal and uncertainty in model simulations. There is medium confidence in          spatial heterogeneities in the volume and type of assimilated data a projected decrease in the frequency of atmospheric blocking over      in atmospheric reanalyses, particularly before the satellite era. Over Greenland and the North Pacific in boreal winter in 2081-2100 under      the Southern Hemisphere, it is likely that the number of extratropical the SSP3-7.0 and SSP5-8.5 scenarios. There is high confidence that      cyclones with low central pressures (<980 hPa) has increased since Southern Hemisphere storm tracks and associated precipitation have      1979. The frequency of intense extratropical cyclones is projected to migrated polewards over recent decades, especially in the austral        decrease (medium confidence). Projected changes in the intensity summer and autumn, associated with a trend towards more positive        depend on the resolution of climate models (medium confidence).
phases of the Southern Annular Mode (SAM) (Section TS.4.2.2) and        There is medium confidence that wind speeds associated with the strengthening and southward shift of the Southern Hemisphere        extratropical cyclones will change following changes in the storm extratropical jet in austral summer. In the long term (2081-2100),      tracks. {2.3.1, 3.3.3, 4.5.1, 4.5.3, 8.3.2, 8.4.2, 11.7.2}
71
 
Technical Summary Box TS.3 l Low-likelihood, High-warming Storylines Future global warming exceeding the assessed very likely range cannot be ruled out and is potentially associated with the highest risks for society and ecosystems. Such low-likelihood, high-warming storylines tend to exhibit substantially greater changes in the intensity of regional drying and wetting than the multi-model mean. Even at levels of warming within the very likely range, global and regional low-likelihood outcomes might occur, such as large precipitation changes, additional sea level rise associated with collapsing ice sheets (see Box TS.4), or abrupt ocean circulation changes. While there is medium confidence that the Atlantic Meridional Overturning Circulation (AMOC) will not experience an abrupt collapse before 2100, if it were to occur, it would very likely cause abrupt shifts in regional weather patterns and water cycle. The probability of these low-likelihood outcomes increases with higher global warming levels. If the real-world climate sensitivity lies at the high end of the assessed range, then global and regional changes substantially outside the very likely range projections occur for a given emissions scenario.
With increasing global warming, some very rare extremes and some compound events (multivariate or concurrent TS      extremes) with low likelihood in past and current climate will become more frequent, and there is a higher chance that events unprecedented in the observational record occur (high confidence). Finally, low-likelihood, high-impact outcomes may also arise from a series of very large volcanic eruptions that could substantially alter the 21st century climate trajectory compared to SSP-based Earth system model (ESM) projections. {Cross-Chapter Box 4.1, 4.3, 4.4, 4.8, 7.3, 7.4, 7.5, 8.6, 9.2, 9.6, Box 9.4, Box 11.2, Cross-Chapter Box 12.1}
Previous IPCC reports largely focused their assessment on the projected very likely range of future surface warming and associated climate change. However, a comprehensive risk assessment also requires considering the potentially larger changes in the physical climate system that are unlikely or very unlikely but possible and potentially associated with the highest risks for society and ecosystems (Figure TS.6). Since AR5, the development of physical climate storylines of high warming has emerged as a useful approach for exploring the future risk space that lies outside of the IPCC very likely range projections. {4.8}
Uncertainty in the true values of equilibrium climate sensitivity (ECS) and transient climate response (TCR) dominate uncertainty in projections of future warming under moderate to strong emissions scenarios (Section TS.3.2). A real-world ECS higher than the assessed very likely range (2&deg;C-5&deg;C) would require a strong historical aerosol cooling and/or a trend towards stronger warming from positive feedbacks linked to changes in SST patterns (pattern effects), combined with a strong positive cloud feedback and substantial biases in paleoclimate reconstructions - each of which is assessed as either unlikely or very unlikely, but not ruled out. Since CMIP6 contains several ESMs that exceed the upper bound of the assessed very likely range in future surface warming, these models can be used to develop low-likelihood, high warming storylines to explore risks and vulnerabilities, even in the absence of a quantitative assessment of likelihood. {4.3.4, 4.8, 7.3.2, 7.4.4, 7.5.2, 7.5.5, 7.5.7}
CMIP6 models with surface warming outside, or close to, the upper bound of the very likely range exhibit patterns of large widespread temperature and precipitation changes that differ substantially from the multi-model mean in all scenarios. For SSP5-8.5, the high-warming models exhibit widespread warming of more than 6&deg;C over most extratropical land regions and parts of the Amazon.
In the Arctic, annual mean temperatures increase by more than 10&deg;C relative to present-day, corresponding to about 30% more than the best estimate of warming. Even for SSP1-2.6, high-warming models show on average 2&deg;C-3&deg;C warming relative to present-day conditions over much of Eurasia and North America (about 40% more than the best estimate of warming) and more than 4&deg;C warming relative to the present over the Arctic in 2081-2100 (Box TS.3, Figure 1). Such a high global warming storyline would imply that the remaining carbon budget consistent with a 2&deg;C warming is smaller than the assessed very likely range. Put another way, even if a carbon budget that likely limits warming to 2&deg;C is met, a low-likelihood, high-warming storyline would result in warming of 2.5&deg;C or more. {4.8}
CMIP6 models with global warming close to the upper bound of the assessed very likely warming range tend to exhibit greater changes in the intensity of regional drying and wetting than the multi-model mean. Furthermore, these model projections show a larger area of drying and tend to show a larger fraction of strong precipitation increases than the multi-model mean. However, regional precipitation changes arise from both thermodynamic and dynamic processes so that the most pronounced global warming levels are not necessarily associated with the strongest precipitation response. Abrupt human-caused changes to the water cycle cannot be ruled out. Positive land surface feedbacks, involving vegetation and dust, can contribute to abrupt changes in aridity, but there is only low confidence that such changes will occur during the 21st century. Continued Amazon deforestation, combined with a warming climate, raises the probability that this ecosystem will cross a tipping point into a dry state during the 21st century (low confidence). (See also Box TS.9). {4.8, 8.6.2}
72
 
Technical Summary Box TS.3 (continued)
While there is medium confidence that the projected decline in the AMOC (Section TS.2.4) will not involve an abrupt collapse before 2100, such a collapse might be triggered by an unexpected meltwater influx from the Greenland Ice Sheet. If an AMOC collapse were to occur, it would very likely cause abrupt shifts in the regional weather patterns and water cycle, such as a southward shift in the tropical rain belt, and could result in weakening of the African and Asian monsoons, strengthening of Southern Hemisphere monsoons, and drying in Europe. (See also Boxes TS.9 and TS.13). {4.7.2, 8.6.1, 9.2.3}
Very rare extremes and compound or concurrent events, such as the 2018 concurrent heatwaves across the Northern Hemisphere, are often associated with large impacts. The changing climate state is already altering the likelihood of extreme events, such as decadal droughts and extreme sea levels, and will continue to do so under future warming. Compound events and concurrent extremes contribute to increasing probability of low-likelihood, high-impact outcomes and will become more frequent with increasing global warming (high confidence). Higher warming levels increase the likelihood of events unprecedented in the observational record. {9.6.4, Box 11.2}                                                                                                                                                            TS Finally, low likelihood storylines need not necessarily relate solely to the human-induced changes in climate. A low-likelihood, high-impact outcome, consistent with historical precedent in the past 2500 years, would be to see several large volcanic eruptions that could greatly alter the 21st century climate trajectory compared to SSP-based Earth system model projections. {Cross-Chapter Box 4.1}
SSP1-2.6 (2081-2100)
(a) Best estimate (scaled)                            (b) High-warming models                      (c) Very-high-warming models
                              -4    -3      -2.5      -2    -1.5      -1        0        1      1.5  2      2.5      3      4
                                                                                &deg;C SSP5-8.5 (2081-2100)
(d) Best estimate (scaled)                            (e) High-warming models                      (f) Very-high-warming models
                              -8      -6      -5      -4      -3      -2        0        2      3  4      5      6      8
                                                                                &deg;C Box TS.3, Figure 1 l High-warming storylines. The intent of this figure is to illustrate high warming storylines compared to the CMIP6 multi-model-mean.
(a) Coupled Model Intercomparison Project Phase 6 (CMIP6) multi-model mean linearly scaled to the assessed best global surface temperature estimate for SSP1-2.6 in 2081-2100 relative to 1995-2014, (b) mean across five high-warming models with global surface temperature changes nearest to the upper bound of the assessed very likely range, and (c) mean across five very high-warming models with global surface temperature changes higher than the assessed very likely. (d-f)
Same as (a-c) but for SSP5-8.5. Note the different colour bars in (a-c) and (d-f). {4.7, Figure 4.41}
73
 
Technical Summary TS.2.4        The Ocean                                                    Global mean SST has increased since the beginning of the 20th century by 0.88 [0.68 to 1.01] &deg;C, and it is virtually certain it will Observations, models and paleo-evidence indicate that                  continue to increase throughout the 21st century, with increasing recently observed changes in the ocean are unprecedented              hazards to marine ecosystems (medium confidence). Marine for centuries to millennia (high confidence). Over the past            heatwaves have become more frequent over the 20th century (high four to six decades, it is virtually certain that the global          confidence), approximately doubling in frequency (high confidence) ocean has warmed, with human influence extremely                      and becoming more intense and longer since the 1980s (medium likely the main driver since the 1970s, making climate                confidence). Most of the marine heatwaves over 2006-2015 have change irreversible over centuries to millennia (medium                been attributed to anthropogenic warming (very likely). Marine confidence). It is virtually certain that upper ocean salinity        heatwaves will continue to increase in frequency, with a likely global contrasts have increased since the 1950s and extremely                increase of 2-9 times in 2081-2100 compared to 1995-2014 under likely that human influence has contributed. It is virtually          SSP1-2.6, and 3-15 times under SSP5-8.5 (Figure TS.11a), with certain that upper ocean stratification has increased since            the largest changes in the tropical and Arctic ocean. {2.3.1, Cross-1970 and that sea water pH has declined globally over the              Chapter Box 2.3, 9.2.1, Box 9.2, 12.4.8}
TS    last 40 years, with human influence being the main driver of the observed surface open ocean acidification (virtually            Observed upper-ocean stratification (0-200 m) has increased certain). A long-term increase in surface open ocean pH                globally since at least 1970 (virtually certain). Based on recent refined occurred over the past 50 million years (high confidence),            analyses of the available observations, there is high confidence and surface ocean pH as low as recent times is uncommon                that it increased by 4.9 +/- 1.5% from 1970-2018, which is about in the last 2 million years (medium confidence). There is              twice as much as assessed in SROCC, and will continue to increase high confidence that marine heatwaves have become more                throughout the 21st century at a rate depending on the emissions frequent in the 20th century, and most of those since 2006            scenario (virtually certain). {2.3.3, 9.2.1}
have been attributed to anthropogenic warming (very likely). There is high confidence that oxygen levels have              It is virtually certain that since 1950 near-surface high-salinity dropped in many regions since the mid 20th century and                regions have become more saline, while low-salinity regions have that the geographic range of many marine organisms has                become fresher, with medium confidence that this is linked to an changed over the last two decades.                                    intensification of the hydrological cycle (Box TS.6). It is extremely likely that human influence has contributed to this salinity change The amount of ocean warming observed since 1971                        and that the large-scale pattern will grow in amplitude over the 21st will likely at least double by 2100 under a low warming                century (medium confidence). {2.3.3, 3.5.2, 9.2.2, 12.4.8}
scenario (SSP1-2.6) and will increase by 4-8 times under a high warming scenario (SSP5-8.5). Stratification (virtually          The AMOC was relatively stable during the past 8000 years (medium certain), acidification (virtually certain), deoxygenation            confidence). There is low confidence in the quantification of AMOC (high confidence) and marine heatwave frequency (high                  changes in the 20th century because of low agreement in quantitative confidence) will continue to increase in the 21st century.            reconstructed and simulated trends, missing key processes in both While there is low confidence in 20th century AMOC change,            models and measurements used for formulating proxies, and new it is very likely that AMOC will decline over the 21st century        model evaluations. Direct observational records since the mid-2000s (Figure TS.11). {2.3, 3.5, 3.6, 4.3.2, 5.3, 7.2, 9.2, Box 9.2, 12.4}  are too short to determine the relative contributions of internal variability, natural forcing and anthropogenic forcing to AMOC It is virtually certain that the global ocean has warmed since at least    change (high confidence). An AMOC decline over the 21st century 1971, representing about 90% of the increase in the global energy          is very likely for all SSP scenarios (Figure TS.11b); a possible abrupt inventory (Section TS.3.1). The ocean is currently warming faster than    decline is assessed further in Box TS.3. {2.3.3, 3.5.4, 4.3.2, 8.6.1, at any other time since at least the last deglacial transition (medium    9.2.3, Cross-Chapter Box 12.3}
confidence), with warming extending to depths well below 2000 m (very high confidence). It is extremely likely that human influence        There is high confidence that many ocean currents will change in was the main driver of this recent ocean warming. Ocean warming            the 21st century in response to changes in wind stress. There is low will continue over the 21st century (virtually certain), and will likely  confidence in 21st century change of Southern Ocean circulation, continue until at least to 2300 even for low CO2 emissions scenarios.      despite high confidence that it is sensitive to changes in wind Ocean warming is irreversible over centuries to millennia (medium          patterns and increased ice-shelf melt. Western boundary currents confidence), but the magnitude of warming is scenario-dependent from      and subtropical gyres have shifted poleward since 1993 (medium about the mid-21st century (medium confidence). The warming will          confidence). Subtropical gyres, the East Australian Current Extension, not be globally uniform, with heat primarily stored in Southern Ocean      the Agulhas Current, and the Brazil Current are projected to intensify water-masses and weaker warming in the subpolar North Atlantic (high      in the 21st century in response to changes in wind stress, while the confidence). Limitations in the understanding of feedback mechanisms      Gulf Stream and the Indonesian Throughflow are projected to weaken limit our confidence in future ocean warming close to Antarctica and      (medium confidence). All of the four main eastern boundary upwelling how this will affect sea ice and ice shelves. {2.3.3, 3.5.1, 4.7.2, 7.2.2, systems are projected to weaken at low latitudes and intensify at high 9.2.2, 9.2.3, 9.2.4, 9.3.2, 9.6.1, Cross-Chapter Box 9.1}                  latitudes in the 21st century (high confidence). {2.3.3, 9.2.3}
74
 
Technical Summary Recent and Future change in the ocean Marine heatwaves, Atlantic Meridional Overturning Circulation (AMOC), Dissolved oxygen, and pH (a) Marine heatwaves                                                                                                        (c) Dissolved oxygen (100-600 m) 15                                                                                                                        5 Future (CMIP6)                                                        Past (simulated)                Past (observed) 0 multiplication factor 10
                                                                                                                                            % change    -5 5                        Past (observed)                                                                                                                                            Future (CMIP6)
                                                                                                                                                        -10 Past (simulated)
                                                                                                                                                        -15 Change                                                                                                              Change in 2100                                                                                                              in 2100 (b) Atlantic Meridional Overturning Circulation (AMOC)                                                                      (d) Ocean acidification 5                                                                                                                      8.2 Past (observed)
Past (simulated)                                                                                                          Past (simulated)                              Past (observed) 0                                                                                                                        8                                                                                                              TS low acidity Sv                                                                                                                                    pH
                              -5 7.8 high acidity                                    Future (CMIP6)
                              -10                                                              Future (CMIP6) 7.6 Change                                                                                                              Change 1850          1900      1950                          2000          2050      2100 in 2100                            1850        1900        1950                2000              2050      2100 in 2100 Past (simulated)        Future (SSP1-2.6) mean; 5-95% range Past (observed)          Future (SSP2-4.5) mean Future (SSP3-7.0) mean; 5-95% range Future (SSP5-8.5) mean; 5-95% range Recent and future change in ice sheets Greenland and Antarctic ice sheets (e) Greenland ice sheet                                                                                                  (f) Antarctic ice sheet 4                                                                                      -0.1                              4            Historical                                      Future (RCP/SSP) -0.1 Historical                                                                                                      (observation-based)
Box  (observation-based)                                  Future (RCP/SSP)                                        0                                                                                  0 0                                                                                      0 Observations ISMIP6                                                                                                        Observations  ISMIP6 (2010-2017) (2093-2100)                                                                                            -4                                                                                0.1 104 Gt m                                    (1978-2017) (2061-2100)                                                          m 1                                                    0.1                  104 Gt      -8                                1                                              0.2
                              -4 Elevation change                                                                                                          Elevation change 0.5                                                                                                                        0.5 0                                                                                                                          0                                              0.3
                                                                                                                                                        -12 (m/yr)                                                                                                                    (m/yr)
                                                                -0.5                                                  0.2                                                                  -0.5
                              -8                                -1                                                                                                                        -1 0.4 ISMIP6 Emulator -16 1980      2000    2020              2040                2060    2080      2100          Change                  1980      2000        2020    2040                    2060    2080      2100 in 2100 ISMIP6    LARMIP Observation-based:                                                                                                                                              Emulator Emulator median (SSP1-2.6); 17-83% & 5-95% ranges Bamber                                                                                                                                                    Change Emulator median (SSP5-8.5); 17-83% & 5-95% ranges                                                                            in 2100 Mouginot/Rignot              ISMIP6 models (SSP1-2.6/RCP2.6)
IMBIE                        ISMIP6 models (SSP5-8.5/RCP8.5)
Figure TS.11 l Past and future ocean and ice-sheet changes. The intent of this figure is to show that observed and projected time series of many ocean and cryosphere indicators are consistent. Observed and simulated historical changes and projected future changes under varying greenhouse gas emissions scenarios. Simulated and projected ocean changes are shown as Coupled Model Intercomparison Project Phase 6 (CMIP6) ensemble mean, and 5-95% range (shading) is provided for scenarios SSP1-2.6 and SSP3-7.0 (except in panel a where the range is provided for scenario SSP1-2.6 and SSP5-8.5). Mean and 5-95% range in 2100 are shown as vertical bars on the right-hand side of each panel. (a) Change in multiplication factor in surface ocean marine heatwave days relative to 1995-2014 (defined as days exceeding the 99th percentile in sea surface temperature (SST) from 1995-2014 distribution). Assessed observational change span 1982-2019 from AVHRR satellite SST. (b) Atlantic Meridional Overturning Circulation (AMOC) transport relative to 1995-2014 (defined as maximum transport at 26&deg;N). Assessed observational change spans 2004-2018 from the RAPID array smoothed with a 12-month running mean (shading around the mean shows the 12-month running standard deviation around the mean). (c) Global mean percent change in ocean oxygen (100-600 m depth), relative to 1995-2014. Assessed observational trends and very likely range are from the SROCC assessment, and span 1970-2010 centred on 2005. (d)
Global mean surface pH. Assessed observational change spans 1985-2019, from the CMEMS SOCAT-based reconstruction (shading around the global mean shows the 90%
confidence interval). (e), (f): Ice sheet mass changes. Projected ice-sheet changes are shown as median, 5-95% range (light shading), and 17-83% range (dark shading) of cumulative mass loss and sea level equivalent from ISMIP6 emulation under SSP1-2.6 and SSP5-8.5 (shading and bold line), with individual emulated projections as thin lines.
Median (dot), 17-83% range (thick vertical bar), and 5-95% range (thin vertical bar) in 2100 are shown as vertical bars on the right-hand side of each panel, from ISMIP6, ISMIP6 emulation, and LARMIP-2. Observation-based estimates: For Greenland (e), for 1972-2018 (Mouginot), for 1992-2016 (Bamber), for 1992-2020 (IMBIE) and total estimated mass loss range for 1840-1972 (Box). For Antarctica (f), estimates based on satellite data combined with simulated surface mass balance and glacial isostatic adjustment for 1992-2020 (IMBIE), 1992-2016 (Bamber), and 1979-2017 (Rignot). Left inset maps: mean Greenland elevation changes 2010-2017 derived from CryoSat-2 radar altimetry (e) and mean Antarctica elevation changes 1978-2017 derived from restored analogue radar records (f). Right inset maps: ISMIP6 model mean (2093-2100) projected changes under the MIROC5 climate model for the RCP8.5 scenario. {2.3.3; 2.3.4; 3.5.4; 4.3.2; 5.3.2; 5.3.3; 5.6.3; 9.2.3; 9.4.1; 9.4.2; Box 9.2; Box 9.2, Figure 1; Figures 9.10, 9.17 and 9.18}
75
 
Technical Summary It is virtually certain that surface pH has declined globally over the    become practically sea ice-free in late summer under high last 40 years and that the main driver is uptake of anthropogenic        CO2 emissions scenarios by the end of the 21st century (high CO2. Ocean acidification and associated reductions in the saturation      confidence). It is virtually certain that further warming will state of calcium carbonate - a constituent of skeletons or shells of      lead to further reductions of Northern Hemisphere snow a variety of marine organisms - is expected to increase in the 21st      cover, and there is high confidence that this is also the case century under all emissions scenarios (high confidence). A long-term      for near-surface permafrost volume.
increase in surface open ocean pH occurred over the past 50 million years (high confidence), and surface ocean pH as low as recent            Glaciers will continue to lose mass at least for several times is uncommon in the last 2 million years (medium confidence).        decades even if global temperature is stabilized (very There is very high confidence that present-day surface pH values          high confidence), and mass loss over the 21st century are unprecedented for at least 26,000 years and current rates of pH      is virtually certain for the Greenland Ice Sheet and likely change are unprecedented since at least that time. Over the past 2-3      for the Antarctic Ice Sheet. Deep uncertainty persists with decades, a pH decline in the ocean interior has been observed in all      respect to the possible evolution of the Antarctic Ice Sheet ocean basins (high confidence) (Figure TS.11d). {2.3.3, 2.3.4, 3.6.2,    within the 21st century and beyond, in particular due to the TS 4.3.2, 5.3.2, 5.3.3, 5.6.3, 12.4.8}                                      potential instability of the West Antarctic Ice Sheet. {2.3, 3.4, 4.3, 8.3, 9.3-9.6, Box 9.4, 12.4}
Open-ocean deoxygenation and expansion of oxygen minimum zones have been observed in many areas of the global ocean since the        Current Arctic sea ice coverage levels (both annual and late summer) mid 20th century (high confidence), in part due to human influence    are at their lowest since at least 1850 (high confidence), and for (medium confidence). Deoxygenation is projected to continue to        late summer for the past 1000 years (medium confidence). Since the increase with ocean warming (high confidence) (Figure TS.11c).        late 1970s, Arctic sea ice area and thickness have decreased in both Higher climate sensitivity and reduced ocean ventilation in CMIP6      summer and winter, with sea ice becoming younger, thinner and more compared to CMIP5 results in substantially greater projections of      dynamic (very high confidence). It is very likely that anthropogenic subsurface (100-600 m) oxygen decline than reported in SROCC for      forcing, mainly due to greenhouse gas increases, was the main driver the period 2080-2099. {2.3.3, 2.3.4, Cross-Chapter Box 2.4, 3.6.2,    of this loss, although new evidence suggests that anthropogenic 5.3.3, 12.4.8}                                                        aerosol forcing has offset part of the greenhouse gas-induced losses since the 1950s (medium confidence). The annual Arctic sea ice area Over at least the last two decades, the geographic range of many      minimum will likely fall below 1 million km2 at least once before marine organisms has shifted towards the poles and towards greater    2050 under all assessed SSP scenarios. This practically sea ice-free depths (high confidence), indicative of shifts towards cooler waters. state will become the norm for late summer by the end of the 21st The range of a smaller subset of organisms has shifted equatorward    century in high CO2 emissions scenarios (high confidence). Arctic and to shallower depths (high confidence). Phenological metrics        summer sea ice varies approximately linearly with global surface associated with the life cycles of many organisms have also            temperature, implying that there is no tipping point and observed/
changed over the last two decades or longer (high confidence).        projected losses are potentially reversible (high confidence). {2.3.2, Since the changes in the geographical range of organisms and their    3.4.1, 4.3.2, 9.3.1, 12.4.9}
phenological metrics have been observed to differ with species and location, there is the possibility of disruption to major marine  For Antarctic sea ice, there is no significant trend in satellite-observed ecosystems. {2.3.4}                                                    sea ice area from 1979 to 2020 in both winter and summer, due to regionally opposing trends and large internal variability. Due to mismatches between model simulations and observations, combined TS.2.5      The Cryosphere                                            with a lack of understanding of reasons for substantial inter-model spread, there is low confidence in model projections of future Antarctic Over recent decades, widespread loss of snow and ice has          sea ice changes, particularly at the regional level. {2.3.2, 3.4.1, 9.3.2}
been observed, and several elements of the cryosphere are now in states unseen in centuries (high confidence).          In permafrost regions, increases in ground temperatures in the upper Human influence was very likely the main driver of                30 m over the past three to four decades have been widespread (high observed reductions in Arctic sea ice since the late 1970s        confidence). For each additional 1&deg;C of warming (up to 4&deg;C above the (with late-summer sea ice loss likely unprecedented for at        1850-1900 level), the global volume of perennially frozen ground to least 1000 years) and the widespread retreat of glaciers          3 m below the surface is projected to decrease by about 25% relative (unprecedented in at least the last 2,000 years, medium            to the present volume (medium confidence). However, these decreases confidence). Furthermore, human influence very likely              may be underestimated due to an incomplete representation of contributed to the observed Northern Hemisphere spring            relevant physical processes in ESMs (low confidence). Seasonal snow snow cover decrease since 1950.                                    cover is treated in Section TS.2.6. {2.3.2, 9.5.2, 12.4.9}
By contrast, Antarctic sea ice area experienced no significant    There is very high confidence that, with few exceptions, glaciers net change since 1979, and there is only low confidence            have retreated since the second half of the 19th century; this in its projected changes. The Arctic Ocean is projected to        behaviour is unprecedented in at least the last 2000 years (medium 76
 
Technical Summary confidence). Mountain glaciers very likely contributed 67.2 [41.8 to      of the Antarctic Ice Sheet, is assessed in Box TS.9. {2.3.2, 3.4.3, 9.4.1, 92.6] mm to the observed GMSL change between 1901 and 2018.              9.4.2, 9.6.3, Atlas.11.2}
This retreat has occurred at increased rates since the 1990s, with human influence very likely being the main driver. Under RCP2.6 and      It is likely that the Antarctic Ice Sheet has lost 2670 +/- 530 Gt, RCP8.5, respectively, glaciers are projected to lose 18% +/- 13% and        contributing 7.4 +/- 1.5 mm to GMSL rise over 1992-2020. The total 36% +/- 20% of their current mass over the 21st century (medium            Antarctic ice mass losses were dominated by the West Antarctic Ice confidence). {2.3.2, 3.4.3, 9.5.1, 9.6.1}                                Sheet, with combined West Antarctic and Peninsula annual loss rates increasing since about 2000 (very high confidence). Furthermore, it The Greenland Ice Sheet was smaller than at present during the            is very likely that parts of the East Antarctic Ice Sheet have lost mass Last Interglacial period (roughly 125,000 years ago) and the mid-        since 1979. Since the 1970s, snowfall has likely increased over the Holocene (roughly 6,000 years ago) (high confidence). After reaching      western Antarctic Peninsula and eastern West Antarctica, with large a recent maximum ice mass at some point between 1450 and 1850,            spatial and interannual variability over the rest of Antarctica. Mass the ice sheet retreated overall, with some decades likely close to        losses from West Antarctic outlet glaciers, mainly induced by ice shelf equilibrium (i.e., mass loss approximately equalling mass gained). It    basal melt (high confidence), outpace mass gain from increased is virtually certain that the Greenland Ice Sheet has lost mass since    snow accumulation on the continent (very high confidence).                TS the 1990s, with human influence a contributing factor (medium            However, there is only limited evidence, with medium agreement, confidence). There is high confidence that annual mass changes have      of anthropogenic forcing of the observed Antarctic mass loss since been consistently negative since the early 2000s. Over the period        1992 (with low confidence in process attribution). Increasing mass 1992-2020, Greenland likely lost 4890 +/- 460 Gt of ice, contributing      loss from ice shelves and inland discharge will likely continue to 13.5 +/- 1.3 mm to GMSL rise. There is high confidence that Greenland      outpace increasing snowfall over the 21st century (Figure TS.11f).
ice mass losses are increasingly dominated by surface melting and        Deep uncertainty persists with respect to the possible evolution of runoff, with large interannual variability arising from changes in        the Antarctic Ice Sheet along high-end mass-loss storylines within surface mass balance. Projections of future Greenland ice-mass loss      the 21st century and beyond, primarily related to the abrupt and (Box TS.4, Table 1; Figure TS.11e) are dominated by increased surface    widespread onset of marine ice sheet instability and marine ice cliff melt under all emissions scenarios (high confidence). Potential          instability. (See also Boxes TS.3 and TS.4). {2.3.2, 3.4.3, 9.4.2, 9.6.3, irreversible long-term loss of the Greenland Ice Sheet, and of parts      Box 9.4, Atlas.11.1}
Box TS.4 l Sea Level Global mean sea level (GMSL) increased by 0.20 [0.15 to 0.25] m over the period 1901 to 2018, with a rate of rise that has accelerated since the 1960s to 3.7 [3.2 to 4.2] mm yr -1 for the period 2006-2018 (high confidence). Human activities were very likely the main driver of observed GMSL rise since 1971, and new observational evidence leads to an assessed sea level rise over the period 1901 to 2018 that is consistent with the sum of individual components contributing to sea level rise, including expansion due to ocean warming and melting of glaciers and ice sheets (high confidence). It is virtually certain that GMSL will continue to rise over the 21st century in response to continued warming of the climate system (Box TS.4, Figure 1). Sea level responds to greenhouse gas (GHG) emissions more slowly than global surface temperature, leading to weaker scenario dependence over the 21st century than for global surface temperature (high confidence). This slow response also leads to long-term committed sea level rise, associated with ongoing ocean heat uptake and the slow adjustment of the ice sheets, that will continue over the centuries and millennia following cessation of emissions (high confidence) (Box TS.9). By 2100, GMSL is projected to rise by 0.28-0.55 m (likely range) under SSP1-1.9 and 0.63-1.01 m (likely range) under SSP5-8.5 relative to the 1995-2014 average (medium confidence). Under the higher CO2 emissions scenarios, there is deep uncertainty in sea level projections for 2100 and beyond associated with the ice-sheet responses to warming. In a low-likelihood, high-impact storyline and a high CO2 emissions scenario, ice-sheet processes characterized by deep uncertainty could drive GMSL rise up to about 5 m by 2150. Given the long-term commitment, uncertainty in the timing of reaching different GMSL rise levels is an important consideration for adaptation planning. {2.3, 3.4, 3.5, 9.6, Box 9.4, Cross-Chapter Box 9.1, Table 9.5}
GMSL change is driven by warming or cooling of the ocean (and the associated expansion/contraction) and changes in the amount of ice and water stored on land. Paleo-evidence shows that GMSL has been about 70 m higher and 130 m lower than present within the past 55 million years and was likely 5 to 10 m higher during the Last Interglacial (Box TS.2, Figure 1). Sea level observations show that GMSL rose by 0.20 [0.15 to 0.25] m over the period 1901-2018 at an average rate of 1.7 [1.3 to 2.2] mm yr -1. New analyses and paleo-evidence since AR5 show this rate is very likely faster than during any century over at least the last three millennia (high confidence). Since AR5, there is strengthened evidence for an increase in the rate of GMSL rise since the mid-20th century, with an average rate of 2.3 [1.6-3.1] mm yr -1 over the period 1971-2018 increasing to 3.7 [3.2-4.2] mm yr -1 for the period 2006-2018 (high confidence). {2.3.3, 9.6.1, 9.6.2}
77
 
Technical Summary Box TS.4 (continued)
(a) Global mean sea level rise from 1900-2150 2.5 SSP1-1.9 SSP1-2.6 SSP2-4.5 SSP3-7.0 SSP5-8.5 Median (medium confidence) 2                Likely range (medium confidence)
SSP3-7.0 SSP5-8.5 Low confidence 83rd percentile GMSL rise (m) 1.5                    SSP5-8.5 Low confidence 95th percentile 1
0.5 Observations                                                              SSP1-2.6        2150 medium & low TS                              0                                                                                                                            confidence projections (see caption) 1900                        1950                      2000                  2050              2100                    2150 (b) Committed sea level rise by warming level and time scale                    (c) Projected timing of sea level rise milestones Paleo ranges 35 Year by which a rise of 2.0 m above 1995-2014 is expected 30                                                                10,000-yr Mid-Pliocene Warm Period 25 GMSL rise (m) 1.5 m 20 15                                                                2,000-yr Last Interglacial                                                                    1.0 m 10 Lowconfidence processes included      SSP58.5 SSP37.0 5                                                                                                                                            SSP24.5 SSP12.6 SSP11.9 100-yr                  0.5 m 0
1        1.5  2            3            4                5 2000              2100              2200                2300+
Peak global surface temperature (&deg;C)
Box TS.4, Figure 1 l Global mean sea level (GMSL) change on different time scales and under different scenarios. The intent of this figure is to (i) show the century-scale GMSL projections in the context of the 20th century observations, (ii) illustrate deep uncertainty in projections by considering the timing of GMSL rise milestones, and (iii) show the long-term commitment associated with different warming levels, including the paleo evidence to support this. (a) GMSL change from 1900 to 2150, observed (1900-2018) and projected under the SSP scenarios (2000-2150), relative to a 1995-2014 baseline. Solid lines show median projections. Shaded regions show likely ranges for SSP1-2.6 and SSP3-7.0. Dotted and dashed lines show respectively the 83rd and 95th percentile low confidence projections for SSP5-8.5. Bars at right show likely ranges for SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5 in 2150. Lightly shaded thick/thin bars show 17th-83rd/5th-95th percentile low-confidence ranges in 2150 for SSP1-2.6 and SSP5-8.5, based upon projection methods incorporating structured expert judgement and marine ice cliff instability. Low confidence range for SSP5-8.5 in 2150 extends to 4.8/5.4 m at the 83rd/95th percentile. (b) GMSL change on 100- (blue), 2000-(green) and 10,000-year (magenta) time scales as a function of global surface temperature, relative to 1850-1900. For 100-year projections, GMSL is projected for the year 2100, relative to a 1995-2014 baseline, and temperature anomalies are average values over 2081-2100. For longer-term commitments, warming is indexed by peak warming above 1850-1900 reached after cessation of emissions. Shaded regions show paleo-constraints on global surface temperature and GMSL for the Last Interglacial and mid-Pliocene Warm Period. Lightly shaded thick/thin blue bars show 17th-83rd/5th-95th percentile low confidence ranges for SSP1-2.6 and SSP5-8.5 in 2100, plotted at 2&deg;C and 5&deg;C. (c) Timing of exceedance of GMSL thresholds of 0.5, 1.0, 1.5 and 2.0 m, under different SSPs. Lightly shaded thick/thin bars show 17th-83rd/5th-95th percentile low-confidence ranges for SSP1-2.6 and SSP5-8.5. {4.3.2, 9.6.1, 9.6.2, 9.6.3, Box 9.4}
78
 
Technical Summary Box TS.4 (continued)
GMSL will continue to rise throughout the 21st century (Box TS.4, Figure 1a). Considering only those processes in whose projections we have at least medium confidence, relative to the period 1995-2014, GMSL is projected to rise between 0.18 m (0.15-0.23 m, likely range; SSP1-1.9) and 0.23 m (0.20-0.30 m, likely range; SSP5-8.5) by 2050. By 2100, the projected rise is between 0.38 m (0.28-0.55 m, likely range; SSP1-1.9) and 0.77 m (0.63-1.01 m, likely range; SSP5-8.5) {Table 9.9}. The methods, models and scenarios used for sea level projections in the AR6 are updated from those employed by SROCC, with contributions informed by the latest model projections described in the ocean and cryosphere Sections (Sections TS.2.4 and TS.2.5). Despite these differences, the sea level projections are broadly consistent with those of SROCC. {4.3.2, 9.6.3}
Importantly, likely range projections do not include those ice-sheet-related processes whose quantification is highly uncertain or that are characterized by deep uncertainty. Higher amounts of GMSL rise before 2100 could be caused by earlier-than-projected disintegration of marine ice shelves, the abrupt, widespread onset of marine ice sheet instability (MISI) and marine ice cliff instability (MICI) around Antarctica, and faster-than-projected changes in the surface mass balance and dynamical ice loss from Greenland                TS (Box TS.4, Figure 1). In a low-likelihood, high-impact storyline and a high CO2 emissions scenario, such processes could in combination contribute more than one additional meter of sea level rise by 2100 (Box TS.3). {4.3.2, 9.6.3, Box 9.4}
Beyond 2100, GMSL will continue to rise for centuries to millennia due to continuing deep ocean heat uptake and mass loss from ice sheets, and will remain elevated for thousands of years (high confidence). By 2150, considering only those processes in whose projections we have at least medium confidence and assuming no acceleration in ice-mass flux after 2100, GMSL is projected to rise between 0.6 m (0.4-0.9 m, likely range, SSP1-1.9) and 1.3 m (1.0-1.9 m, likely range) (SSP5-8.5), relative to the period 1995-2014 based on the SSP scenario extensions. Under high CO2 emissions, processes in which there is low confidence, such as MICI, could drive GMSL rise up to about 5 m by 2150 (Box TS.4, Figure 1a). By 2300, GMSL will rise 0.3-3.1 m under low CO2 emissions (SSP1-2.6) (low confidence). Under high CO2 emissions (SSP5-8.5), projected GMSL rise is between 1.7 and 6.8 m by 2300 in the absence of MICI and by up to 16 m considering MICI (low confidence). Over 2000 years, there is medium agreement and limited evidence that committed GMSL rise is projected to be about 2-3 m with 1.5&deg;C peak warming, 2-6 m with 2&deg;C of peak warming, 4-10 m with 3&deg;C of peak warming, 12-16 m with 4&deg;C of peak warming, and 19-22 m with 5&deg;C of peak warming. {9.6.3}
Looking at uncertainty in time provides an alternative perspective on uncertainty in future sea level rise (Box TS.4, Figure 1c). For example, considering only medium confidence processes, GMSL rise is likely to exceed 0.5 m between about 2080 and 2170 under SSP1-2.6 and between about 2070 and 2090 under SSP5-8.5. Given the long-term commitment, uncertainty in the timing of reaching different levels of GMSL rise is an important consideration for adaptation planning. {9.6.3}
At regional scales, additional processes come into play that modify the local sea level change relative to GMSL, including vertical land motion, ocean circulation and density changes, and gravitational, rotational, and deformational effects arising from the redistribution of water and ice mass between land and the ocean. These processes give rise to a spatial pattern that tends to increase sea level rise at the low latitudes and reduce sea level rise at high latitudes. However, over the 21st century, the majority of coastal locations have a median projected regional sea level rise within +/-20% of the projected GMSL change (medium confidence). Further details on regional sea level change and extremes are provided in Section TS.4. {9.6.3}
Box TS.5 l The Carbon Cycle The continued growth of atmospheric CO2 concentrations over the industrial era is unequivocally due to emissions from human activities. Ocean and land carbon sinks slow the rise of CO2 in the atmosphere. Projections show that while land and ocean sinks absorb more CO2 under high emissions scenarios than low emissions scenarios, the fraction of emissions removed from the atmosphere by natural sinks decreases with higher concentrations (high confidence).
Projected ocean and land sinks show similar responses for a given scenario, but the land sink has a much higher interannual variability and wider model spread. The slowed growth rates of the carbon sinks projected for the second half of this century are linked to strengthening carbon-climate feedbacks and stabilization of atmospheric CO2 under medium-to-no-mitigation and high-mitigation scenarios, respectively (see FAQ 5.1). {5.2, 5.4}
79
 
Technical Summary Box TS.5 (continued)
Carbon sinks for anthropogenic CO2 are associated with mainly physical ocean and biospheric land processes that drive the exchange of carbon between multiple land, ocean and atmospheric reservoirs. These exchanges are driven by increasing atmospheric CO2, but are modulated by changes in climate (Box TS.5, Figure 1c,d). The Northern and Southern Hemispheres dominate the land and ocean sinks, respectively (Box TS.5, Figure 1). Ocean circulation and thermodynamic processes also play a critical role in coupling the global carbon and energy (heat) cycles. There is high confidence that this ocean carbon-heat nexus is an important basis for one of the most important carbon-climate metrics, the transient climate response to cumulative CO2 emissions (TCRE; Section TS.3.2.1) used to determine the remaining carbon budget. {5.1, 5.2, 5.5, 9.2, Cross-Chapter Box 5.3}
Based on multiple lines of evidence using interhemispheric gradients of CO2 concentrations, isotopes, and inventory data, it is unequivocal that the growth in CO2 in the atmosphere since 1750 (see Section TS.2.2) is due to the direct emissions from human activities. The combustion of fossil fuels and land-use change for the period 1750-2019 resulted in the release of 700 +/- 75 PgC (likely TS    range, 1 PgC = 1015 g of carbon) to the atmosphere, of which about 41% +/- 11% remains in the atmosphere today (high confidence).
Of the total anthropogenic CO2 emissions, the combustion of fossil fuels was responsible for about 64% +/- 15%, growing to an 86% +/-
14% contribution over the past 10 years. The remainder resulted from land-use change. During the last decade (2010-2019), average annual anthropogenic CO2 emissions reached the highest levels in human history at 10.9 +/- 0.9 PgC yr -1 (high confidence). Of these emissions, 46% accumulated in the atmosphere (5.1 +/- 0.02 PgC yr -1), 23% (2.5 +/- 0.6 PgC yr -1) was taken up by the ocean and 31%
(3.4 +/- 0.9 PgC yr -1) was removed by terrestrial ecosystems (high confidence). {5.2.1, 5.2.2, 5.2.3}
The ocean (high confidence) and land (medium confidence) sinks of CO2 have increased with anthropogenic emissions over the past six decades (Box TS.5, Figure 1). This coherence between emissions and the growth in ocean and land sinks has resulted in the airborne fraction of anthropogenic CO2 remaining at 44 +/- 10% over the past 60 years (high confidence). Interannual and decadal variability of the ocean and land sinks indicate that they are sensitive to changes in the growth rate of emissions as well as climate variability and are therefore also sensitive to climate change (high confidence). {5.2.1}
The land CO2 sink is driven by carbon uptake by vegetation, with large interannual variability, for example, linked to the El Nino-Southern Oscillation (ENSO). Since the 1980s, carbon fertilization from rising atmospheric CO2 has increased the strength of the net land CO2 sink (medium confidence). During the historical period, the growth of the ocean sink has been primarily determined by the growth rate of atmospheric CO2. However, there is medium confidence that changes to physical and chemical processes in the ocean and in the land biosphere, which govern carbon feedbacks, are already modifying the characteristics of variability, particularly the seasonal cycle of CO2, in both the ocean and land. However, changes to the multi-decadal trends in the sinks have not yet been observed. {2.3.4, 3.6.1, 5.2.1}
In AR6, ESM projections are assessed with CO2 concentrations by 2100 from about 400 ppm (SSP1-1.9) to above 1100 ppm (SSP5-8.5).
Most simulations are performed with prescribed atmospheric CO2 concentrations, which already account for a central estimate of climate-carbon feedback effects. Carbon dioxide emissions-driven simulations account for uncertainty in these feedbacks, but do not significantly change the projected global surface temperature changes (high confidence). Although land and ocean sinks absorb more CO2 under high emissions than low emissions scenarios, the fraction of emissions removed from the atmosphere decreases (high confidence). This means that the more CO2 that is emitted, the less efficient the ocean and land sinks become (high confidence),
an effect which compensates for the logarithmic relationship between CO2 and its radiative forcing, which means that for each unit increase in additional atmospheric CO2 the effect on global temperature decreases. (Box TS.5, Figure 1f,g). {4.3.1, 5.4.5, 5.5.1.2}
Ocean and land sinks show similar responses for a given scenario, but the land sink has a much higher interannual variability and wider model spread. Under SSP3-7.0 and SSP5-8.5, the initial growth of both sinks in response to increasing atmospheric concentrations of CO2 is subsequently limited by emerging carbon-climate feedbacks (high confidence) (Box TS.5, Figure 1f).
Projections show that the ocean and land sinks will stop growing from the second part of the 21st century under all emissions scenarios, but with different drivers for different emissions scenarios. Under SSP3-7.0 and SSP5-8.5, the weakening growth rate of the ocean CO2 sink in the second half of the century is primarily linked to the strengthening positive feedback from reduced carbonate buffering capacity, ocean warming and altered ocean circulation (e.g., AMOC changes). In contrast, for SSP1-1.9, SSP1-2.6 and SSP2-4.5, the weakening growth rate of the ocean carbon sink is a response to the stabilizing or declining atmospheric CO2 concentrations. Under SSP1-1.9, models project that combined land and ocean sinks will turn into a weak source by 2100 (medium confidence). Under high CO2 emissions scenarios, it is very likely that the land carbon sink will grow more slowly due to warming and drying from the mid-21st century, but it is very unlikely that it will switch from being a sink to a source before 2100.
80
 
Technical Summary Box TS.5 (continued)
                                          
Latitute Colour High model agreement      Low model agreement TS
            
number of model simulations shades: uncertainty range
             )
               
Box TS.5, Figure 1 l Carbon cycle processes and projections.
81
 
Technical Summary Box TS.5 (continued)
Box TS.5, Figure 1 (continued): The intent of this figure is to show the response of the carbon cycle to carbon dioxide (CO2) emissions and climate and its role in determining future CO2 levels through projected changes to sinks and sink fractions. The figure shows changes in carbon storage in response to elevated CO2 (a, b) and the response to climate warming (c, d). Maps show spatial patterns of changes in carbon uptake during simulations with 1% per year increase in CO2 (Section 5.4.5.5), and zonal mean plots show distribution of carbon changes is dominated by the land (green lines) in the tropics and Northern Hemisphere and ocean (blue lines) in the Southern Hemisphere. Hatching indicates regions where fewer than 80% of models agree on the sign of response. (e) Future CO2 projections:
projected CO2 concentrations in the Shared Socio-economic Pathway (SSP) scenarios in response to anthropogenic emissions, results from coupled Earth system models for SSP5-8.5 and from the MAGICC7 emulator for other scenarios (Section 4.3.1). (f) Future carbon fluxes: projected combined land and ocean fluxes (positive downward) up to 2100 for the SSP scenarios, and extended to 2300 for available scenarios, 5-95% uncertainty plumes shown for SSP1-2.6 and SSP3-7.0 (Sections 4.3.2.4, 5.4.5.4 and 5.4.10). The numbers near the top show the number of model simulations used. (g) Sink fraction: the fraction of cumulative emissions of CO2 removed by land and ocean sinks. The sink fraction is smaller under conditions of higher emissions. {Figure 4.3; 5.4.5; Figures 5.25, 5.27 and 5.30}
Climate change alone is expected to increase land carbon accumulation in the high latitudes (not including permafrost, which is assessed in Sections TS.2.5 and TS.3.2.2), but also to lead to a counteracting loss of land carbon in the tropics (medium confidence).
Earth system model projections show that the overall uncertainty of atmospheric CO2 by 2100 is still dominated by the emissions TS      pathway, but carbon-climate feedbacks (see Section TS.3.3.2) are important, with increasing uncertainties in high emissions pathways (Box TS.5, Figure 1e). {4.3.2, 5.4.1, 5.4.2, 5.4.4, 5.4.5, 11.6, 11.9, Cross-Chapter Box 5.1, Cross-Chapter Box 5.3}
Under three SSP scenarios with long-term extensions until 2300 (SSP5-8.5, SSP5-3.4-OS, SSP1-2.6), ESMs project a change of the land from a sink to a source (medium confidence). The scenarios make simplified assumptions about emissions reductions, with SSP1-2.6 and SSP5-3.4-OS reaching about 400 ppm by 2300, while SSP5-8.5 exceeds 2000 ppm. Under high emissions, the transition is warming-driven, whereas it is linked to the decline in atmospheric CO2 under net negative CO2 emissions. The ocean remains a sink throughout the period to 2300 except under very large net negative emissions. The response of the natural aspects of the carbon cycle to carbon dioxide removal is further developed in Section TS.3.3.2. {5.4.9}
TS.2.6        Land Climate, Including Biosphere and Extremes                              and about 80% larger than warming of the ocean surface. Warming of the land surface during the period 1971-2018 contributed about 5%
Land surface air temperatures have risen faster than the                              of the increase in the global energy inventory (Section TS.3.1), nearly global surface temperature since the 1850s, and it is virtually                        twice the estimate in AR5 (high confidence). It is virtually certain that certain that this differential warming will persist into the                          the average surface warming over land will continue to be higher than future. It is virtually certain that the frequency and intensity                      over the ocean throughout the 21st century. The warming pattern of hot extremes and the intensity and duration of heatwaves                            will likely vary seasonally, with northern high latitudes warming more have increased since 1950 and will further increase in the                            during winter than summer (medium confidence). {2.3.1, 4.3.1, 4.5.1, future even if global warming is stabilized at 1.5&deg;C. The                              7.2.2, Box 7.2, Cross-Chapter Box 9.1, 11.3, Atlas 11.2}
frequency and intensity of heavy precipitation events have increased over a majority of those land regions with good                              The frequency and intensity of hot extremes (warm days and nights) observational coverage (high confidence) and will extremely                            and the intensity and duration of heatwaves have increased globally likely increase over most land regions with additional global                          and in most regions since 1950, while the frequency and intensity warming.                                                                              of cold extremes have decreased (virtually certain). There is high confidence that the increases in frequency and severity of hot Over the past half century, key aspects of the biosphere                              extremes are due to human-induced climate change. Some recent have changed in ways that are consistent with large-                                  extreme events would have been extremely unlikely to occur without scale warming: climate zones have shifted poleward, and                                human influence on the climate system. It is virtually certain that the growing season length in the Northern Hemisphere                                  further changes in hot and cold extremes will occur throughout the extratropics has increased (high confidence). The amplitude                            21st century in nearly all inhabited regions, even if global warming of the seasonal cycle of atmospheric CO2 poleward of 45&deg;N                              is stabilized at 1.5&deg;C (Table TS.2, Figure TS.12a). {1.3, Cross-Chapter has increased since the 1960s (very high confidence), with                            Box 3.2, 11.1.4, 11.3.2, 11.3.4, 11.3.5, 11.9, 12.4}
increasing productivity of the land biosphere due to the increasing atmospheric CO2 concentration as the main driver                            Greater warming over land alters key water cycle characteristics (medium confidence). Global-scale vegetation greenness has                            (Box TS.6). The rates of change in mean precipitation and runoff, increased since the 1980s (high confidence). {2.3, 3.6, 4.3,                          and their variability, increase with global warming (Figure TS.12e,f).
4.5, 5.2, 11.3, 11.4, 11.9, 12.4}                                                      Human-induced climate change has contributed to increases in agricultural and ecological droughts in some regions due to increases Observed temperatures over land have increased by 1.59 [1.34-                            in evapotranspiration (medium confidence). More regions are 1.83] &deg;C between the period 1850-1900 and 2011-2020. Warming                              affected by increases in agricultural and ecological droughts with of the land is about 45% larger than for global surface temperature                      increasing global warming (high confidence; see also Figure TS.12c).
82
 
Technical Summary (a) Hot extreme events                                                                                                                                                      (b) Heavy precipitation events 40 10-year event, frequency 6                                                                                                      6              10-year event, frequency 40 Relative frequency (pre-industrial = 1)                                                                                                                                    Relative frequency (pre-industrial = 1) 50-year event, frequency                                                                                                                                                      50-year event, frequency                                    35 35          10-year event, intensity                                5                                                                                                                      10-year event, intensity 50-year event, intensity                                                                                                                                        5              50-year event, intensity                                    30 Intensity change (&deg;C)                                                                                                                                                              Intensity change (%)
30 4                                                                                                                                                                                  25 25                                                                                                                                                                          4 20                                                                                                                                                                                                                                                      20 3
3 15                                                                                                                                                                                                                                                      15 2
10 2                                                                          10 5                                                                  1 5
0                                                                                                                                                                          1 1.0        1.5    2.0              3.0          4.0                                                                                                                        1.0        1.5      2.0              3.0                  4.0 Global surface temperature change since 1850-1900 (&deg;C)                                                                                                                      Global surface temperature change since 1850-1900 (&deg;C)
TS (c) Droughts in drought-prone regions                                                                                                                                        (d) Northern Hemisphere March-May snow cover 8
Amplitude of annual soil moisture decrease (-)
20                                                        SSP5-8.5 10-year event, frequency SSP3-7.0 Relative frequency (pre-industrial = 1) 7          10-year event, intensity                                2                                                                                                                                                                SSP2-4.5 Snow cover extent change (%)
0                                                        SSP1-2.6 6                                                                                                                                                                                                                                    SSP1-1.9 1.5 5                                                                                                                                                                        - 20 4                                                                  1
                                                                                                                                                                                                                                - 40 3
0.5
                                                                                                                                                                                                                                - 60 2
0 1                                                                                                                                                                        - 80 1.0        1.5    2.0              3.0          4.0                                                                                                                          0        1      2        3    4        5          6      7 Global surface temperature change since 1850-1900 (&deg;C)                                                                                                                      Global surface temperature change since 1850-1900 (&deg;C)
(e) Hydrological change over tropical land                                                                                                                                  (f) Hydrological change over extratropical land 50 Precipitable water annual mean                                                                                                                                                Precipitable water annual mean Precipitation                                                                                                                                                        60 Precipitation
      % change - 15 models mean - SSP5-8.5                                                                                                                                        % change - 15 models mean - SSP5-8.5 annual mean                                                                                                                                                                annual mean 40 interannual variability                                                                                                                                          50        interannual variability Runoff                                                                                                                                                                      Runoff 30      annual mean                                                                                                                                                                annual mean interannual variability                                                                                                                                          40        interannual variability 20                                                                                                                                                                        30 10                                                                                                                                                                        20 10 0
17-83% range                                                                                                          0                                                            17-83% range
                                                -10 2.0          3.0        4.0        5.0                                                                                                                                    2.0          3.0          4.0              5.0 Global surface temperature change since 1850-1900 (&deg;C)                                                                                                                      Global surface temperature change since 1850-1900 (&deg;C)
Figure TS.12 l Land-related changes relative to the 1850-1900 as a function of global warming levels. The intent of this figure is to show that extremes and mean land variables change consistently with warming levels and to show the changes with global warming levels of water cycle indicators (i.e., precipitation and runoff) over tropical and extratropical land in terms of mean and interannual variability (interannual variability increases at a faster rate than the mean). (a) Changes in the frequency (left scale) and intensity (in &deg;C, right scale) of daily hot extremes occurring every 10 and 50 years. (b) as (a), but for daily heavy precipitation extremes, with intensity change in %.
(c) Changes in 10-year droughts aggregated over drought-prone regions (WNA, CNA, NCA, SCA, NSA, NES, SAM, SWS, SSA, WCE, MED, WSAF, ESAF, MDG, SAU, and EAU; for definitions of these regions, see Figure Atlas.2), with drought intensity (right scale) represented by the change of annual mean soil moisture, normalized with respect to interannual variability. Limits of the 5%95% confidence interval are shown in panels (a-c). (d) Changes in Northern Hemisphere spring (March-April-May) snow cover extent relative to 1850-1900; (e,f) Relative change (%) in annual mean of total precipitable water (grey line), precipitation (red solid lines), runoff (blue solid lines) and in standard deviation (i.e., variability) of precipitation (red dashed lines) and runoff (blue dashed lines) averaged over (e) tropical and (f) extratropical land as function of global warming levels. Coupled Model Intercomparison Project Phase 6 (CMIP6) models that reached a 5&deg;C warming level above the 1850-1900 average in the 21st century in SSP5-8.5 have been used. Precipitation and runoff variability are estimated by respective standard deviation after removing linear trends. Error bars show the 17-83% confidence interval for the warmest +5&deg;C global warming level. {Figures 8.16, 9.24, 11.6, 11.7, 11.12, 11.15, 11.18 and Atlas.2}
83
 
Technical Summary There is low confidence that the increase of plant water-use efficiency Hemisphere extratropics since the mid-20th century (high due to higher atmospheric CO2 concentration alleviates extreme          confidence), are consistent with large-scale warming. At the agricultural and ecological droughts in conditions characterized        same time an increase in the amplitude of the seasonal cycle of by limited soil moisture and increased atmospheric evaporative          atmospheric CO2 poleward of 45&deg;N since the early 1960s (high demand. {2.3.1, Cross-Chapter Box 5.1, 8.2.3, 8.4.1, 11.2.4, 11.4,      confidence) and a global-scale increase in vegetation greenness 11.6, Box 11.1}                                                        of the terrestrial surface since the early 1980s (high confidence) have been observed. Increasing atmospheric CO2, warming at high Northern Hemisphere spring snow cover has decreased since at            latitudes, and land management interventions have contributed least 1978 (very high confidence), and there is high confidence        to the observed greening trend, but there is low confidence in that trends in snow cover loss extend back to 1950. It is very likely  their relative roles. There is medium confidence that increased that human influence contributed to these reductions. Earlier onset    plant growth associated with CO2 fertilization is the main driver of snowmelt has contributed to seasonally dependent changes            of the observed increase in amplitude of the seasonal cycle of in streamflow (high confidence). A further decrease of Northern        atmospheric CO2 in the Northern Hemisphere. Reactive nitrogen, Hemisphere seasonal snow cover extent is virtually certain under        ozone and aerosols affect terrestrial vegetation and carbon cycle TS further global warming (Figure TS.12d). {2.3.2, 3.4.2, 8.3.2. 9.5.3,    through deposition and effects on large-scale radiation (high 12.4, 9.2, 11.2, Atlas 8.2}                                            confidence), but the magnitude of these effects on the land carbon sink, ecosystem productivity and indirect CO2 forcing remains The frequency and intensity of heavy precipitation events have          uncertain. {2.3.4, 3.6.1, 5.2.1, 6.4.5, 12.3.7, 12.4}
increased over a majority of land regions with good observational coverage since 1950 (high confidence, Box TS.6, Table TS.2). Human      Over the last century, there has been a poleward and upslope shift influence is likely the main driver of this change (Table TS.2). It is  in the distribution of many land species (very high confidence) extremely likely that on most land regions heavy precipitation will    as well as increases in species turnover within many ecosystems become more frequent and more intense with additional global            (high confidence). There is high confidence that the geographical warming (Table TS.2, Figure TS.12b). The projected increase in heavy    distribution of climate zones has shifted in many parts of the precipitation extremes translates to an increase in the frequency and  world in the last half century. The SRCCL concluded that continued magnitude of pluvial floods (high confidence) (Table TS.2). {Cross-    warming will exacerbate desertification processes (medium Chapter Box 3.2, 8.4.1, 11.4.2, 11.4.4, 11.5.5, 12.4}                  confidence) and that ecosystems will become increasingly exposed to climates beyond those that they are currently adapted to (high The probability of compound extreme events has likely increased        confidence). There is medium confidence that climate change will due to human-induced climate change. Concurrent heatwaves and          increase disturbance by, for example, fire and tree mortality, across droughts have become more frequent over the last century, and this      several ecosystems. Increases are projected in drought, aridity and trend will continue with higher global warming (high confidence).      fire weather in some regions (Section TS.4.3; high confidence).
The probability of compound flooding (storm surge, extreme              There is low confidence in the magnitude of these changes, but rainfall and/or river flow) has increased in some locations and will    the probability of crossing uncertain regional thresholds (e.g., fires, continue to increase due to both sea level rise and increases in heavy  forest dieback) increases with further warming (high confidence).
precipitation, including changes in precipitation intensity associated  The response of biogeochemical cycles to the anthropogenic with tropical cyclones (high confidence). {11.8.1, 11.8.2, 11.8.3}      perturbation can be abrupt at regional scales, and irreversible on decadal to century time scales (high confidence). {2.3.4, 5.4.3, Changes in key aspects of the terrestrial biosphere, such as an        5.4.9, 11.6, 11.8, 12.5, SRCCL 2.2, SRCCL 2.5, SR1.5 3.4}
increase of the growing season length in much of the Northern 84
 
Technical Summary Box TS.6 l Water Cycle Human-caused climate change has driven detectable changes in the global water cycle since the mid-20th century (high confidence), and it is projected to cause substantial further changes at both global and regional scales (high confidence).
Global land precipitation has likely increased since 1950, with a faster increase since the 1980s (medium confidence).
Atmospheric water vapour has increased throughout the troposphere since at least the 1980s (likely). Annual global land precipitation will increase over the 21st century as global surface temperature increases (high confidence).
Human influence has been detected in amplified surface salinity and precipitation minus evaporation (P-E) patterns over the ocean (high confidence).
The severity of very wet and very dry events increase in a warming climate (high confidence), but changes in atmospheric circulation patterns affect where and how often these extremes occur. Water cycle variability and related extremes are projected to increase faster than mean changes in most regions of the world and under all emissions                          TS scenarios (high confidence).
Over the 21st century, the total land area subject to drought will increase and droughts will become more frequent and severe (high confidence). Near-term projected changes in precipitation are uncertain mainly because of internal variability, model uncertainty and uncertainty in forcings from natural and anthropogenic aerosols (medium confidence).
Over the 21st century and beyond, abrupt human-caused changes to the water cycle cannot be excluded (medium confidence). {2.3, 3.3, 4.3, 4.4, 4.5, 4.6, 8.2, 8.3, 8.4, 8.5, 8.6, 11.4, 11.6, 11.9}
There is high confidence that the global water cycle has intensified since at least 1980 expressed by, for example, increased atmospheric moisture fluxes and amplified precipitation minus evaporation patterns. Global land precipitation has likely increased since 1950, with a faster increase since the 1980s (medium confidence), and a likely human contribution to patterns of change, particularly for increases in high-latitude precipitation over the Northern Hemisphere. Increases in global mean precipitation are determined by a robust response to global surface temperature (very likely 2-3% per &deg;C) that is partly offset by fast atmospheric adjustments to atmospheric heating by greenhouse gases (GHGs) and aerosols (Section TS.3.2.2). The overall effect of anthropogenic aerosols is to reduce global precipitation through surface radiative cooling effects (high confidence). Over much of the 20th century, opposing effects of GHGs and aerosols on precipitation have been observed for some regional monsoons (high confidence) (Box TS.13). Global annual precipitation over land is projected to increase on average by 2.4% (-0.2% to +4.7% likely range) under SSP1-1.9, 4.6% (1.5%
to 8.3% likely range) under SSP2-4.5, and 8.3% (0.9% to 12.9% likely range) under SSP5-8.5 by 2081-2100 relative to 1995-2014 (Box TS.6, Figure 1). Inter-model differences and internal variability contribute to a substantial range in projections of large-scale and regional water cycle changes (high confidence). The occurrence of volcanic eruptions can alter the water cycle for several years (high confidence). Projected patterns of precipitation change exhibit substantial regional differences and seasonal contrast as global surface temperature increases over the 21st century (Box TS.6, Figure 1). {2.3.1, 3.3.2, 3.3.3, 3.5.2, 4.3.1, 4.4.1, 4.5.1, 4.6.1, Cross-Chapter Box 4.1, 8.2.1, 8.2.2, 8.2.3, Box 8.1, 8.3.2.4, 8.4.1, 8.5.2, 10.4.2}
Global total column water vapour content has very likely increased since the 1980s, and it is likely that human influence has contributed to tropical upper tropospheric moistening. Near-surface specific humidity has increased over the ocean (likely) and land (very likely) since at least the 1970s, with a detectable human influence (medium confidence). Human influence has been detected in amplified surface salinity and precipitation minus evaporation (P-E) patterns over the ocean (high confidence). It is virtually certain that evaporation will increase over the ocean and very likely that evapotranspiration will increase over land, with regional variations under future surface warming (Box TS.6, Figure 1). There is high confidence that projected increases in precipitation amount and intensity will be associated with increased runoff in northern high latitudes (Box TS.6, Figure 1). In response to cryosphere changes (Section TS.2.5),
there have been changes in streamflow seasonality, including an earlier occurrence of peak streamflow in high-latitude and mountain catchments (high confidence). Projected runoff (Box TS.6, Figure 1c) is typically decreased by contributions from small glaciers because of glacier mass loss, while runoff from larger glaciers will generally increase with increasing global warming levels until their mass becomes depleted (high confidence). {2.3.1, 3.3.2, 3.3.3, 3.5.2, 8.2.3, 8.4.1, 11.5}
85
 
Technical Summary Box TS.6 (continued)
Warming over land drives an increase in atmospheric evaporative demand and in the severity of drought events (high confidence). Greater warming over land than over the ocean alters atmospheric circulation patterns and reduces continental near-surface relative humidity, which contributes to regional drying (high confidence). A very likely decrease in relative humidity has occurred over much of the global land area since 2000. Projected increases in evapotranspiration due to growing atmospheric water demand will decrease soil moisture over the Mediterranean region, south-western North America, South Africa, South-Western South America and south-western Australia (high confidence) (Box TS.6, Figure 1). Some tropical regions are also projected to experience enhanced aridity, including the Amazon basin and Central America (high confidence). The total land area subject to increasing drought frequency and severity will expand (high confidence), and in the Mediterranean, South-Western South America, and Western North America, future aridification will far exceed the magnitude of change seen in the last millennium (high confidence). {4.5.1, 8.2.2, 8.2.3, 8.4.1, Box 8.2, 11.6, 11.9}
Land-use change and water extraction for irrigation have influenced local and regional responses in the water cycle (high confidence).
TS    Large-scale deforestation likely decreases evapotranspiration and precipitation and increases runoff over the deforested regions relative to the regional effects of climate change (medium confidence). Urbanization increases local precipitation (medium confidence) and runoff intensity (high confidence) (Box TS.14). Increased precipitation intensities have enhanced groundwater recharge, most notably in tropical regions (medium confidence). There is high confidence that groundwater depletion has occurred since at least the start of the 21st century, as a consequence of groundwater withdrawals for irrigation in agricultural areas in drylands. {8.2.3, 8.3.1, 11.1.6, 11.4, 11.6, FAQ 8.1}
Water cycle variability and related extremes are projected to increase faster than mean changes in most regions of the world and under all emissions scenarios (high confidence). A warmer climate increases moisture transport into weather systems, which intensifies wet seasons and events (high confidence). The magnitudes of projected precipitation increases and related extreme events depend on model resolution and the representation of convective processes (high confidence). Increases in near-surface atmospheric moisture capacity of about 7% per 1&#xba;C of warming lead to a similar response in the intensification of heavy precipitation from sub-daily up to seasonal time scales, increasing the severity of flood hazards (high confidence). The average and maximum rain-rates associated with tropical and extratropical cyclones, atmospheric rivers and severe convective storms will therefore also increase with future warming (high confidence). For some regions, there is medium confidence that peak tropical cyclone rain-rates will increase by more than 7% per 1&deg;C of warming due to increased low-level moisture convergence caused by increases in wind intensity. In the tropics year-round and in the summer season elsewhere, interannual variability of precipitation and runoff over land is projected to increase at a faster rate than changes in seasonal mean precipitation (Figure TS.12e,f) (medium confidence). Sub-seasonal precipitation variability is also projected to increase, with fewer rainy days but increased daily mean precipitation intensity over many land regions (high confidence). {4.5.3, 8.2.3, 8.4.1, 8.4.2, 8.5.1, 8.5.2, 11.4, 11.5, 11.7, 11.9}
86
 
Technical Summary Box TS.6 (continued)
Long-term water cycle variables changes for SSP2-4.5 (2081-2100 vs 1995-2014)
(a) Precipitation                                                              (b) Evapotranspiration TS (c) Runoff                                                              (d) Surface soil moisture Box TS.6, Figure 1 l Projected water cycle changes. The intent of this figure is to give a geographical overview of changes in multiple components of the global water cycle using an intermediate emissions scenario. Important key message: without drastic reductions in greenhouse gas emissions, human-induced global warming will be associated with widespread changes in all components of the water cycle. Long-term (2081-2100) projected annual mean changes (%) relative to present-day (1995-2014) in the SSP2-4.5 emissions scenario for (a) precipitation, (b) surface evapotranspiration, (c) total runoff and (d) surface soil moisture. Numbers in top right of each panel indicate indicate the number of Coupled Model Intercomparison Project Phase 6 (CMIP6) models used for estimating the ensemble mean. For other scenarios, please refer to relevant figures in Chapter 8. Uncertainty is represented using the simple approach: No overlay indicates regions with high model agreement, where 80% of models agree on sign of change; diagonal lines indicate regions with low model agreement, where <80% of models agree on sign of change. For more information on the simple approach, please refer to the Cross-Chapter Box Atlas.1. {8.4.1; Figures 8.14, 8.17, 8.18, and 8.19}
87
 
Technical Summary Infographic TS.1 l Climate Futures Climate futures The climate change that people will experience this century and beyond depends on our greenhouse gases emissions, how much global warming this will cause and the response of the climate system to this warming.
Emissions pathways Different social and economic developments can lead to substantially different future emissions of carbon dioxide (CO 2 ), other greenhouse gases and air pollutants for the rest of the century.
Today CO2 peak TS                                                                              140 Very high CO2 emissions (billion tonnes CO2 per year) 120 100                                                                  CO2 doubled CO2 doubled High 80 60 CO2 peak 40 CO2 halved 20          CO2 peak CO2 peak                                                                Medium 0
Low Net CO2 removal                                  Very low
                                                                                -20 2000              2020            2040              2060    2080            2100 Effect on surface temperature For temperature to stabilize, CO2 emissions need to reach net zero.
Global warming since 1850-1900 (&deg;C) 0 0.5 1 1.5 2    3    4 Very high High Medium Low Very low 1980                                                          2000              2020            2040              2060    2080 Today Short-term effect: Natural variability Over short time scales (typically a decade), natural variability can temporarily dampen or accentuate global warming trends resulting from emissions.
Infographic TS.1 l Climate Futures. The intent of this figure is to show possible climate futures: The climate change that people will experience this century and beyond depends on our greenhouse gas emissions, how much global warming this will cause and the response of the climate system to this warming.
(top left) Annual emissions of CO2 for the five core Shared Socio-economic Pathway (SSP) scenarios (very low: SSP1-1.9, low: SSP1-2.6, intermediate: SSP2-4.5, high: SSP3-7.0, very high: SSP5-8.5).(bottom left) Projected warming for each of these emissions scenarios.
88
 
Technical Summary Climate futures Response of the climate system relative to 1850-1900 Many aspects of the climate system react quickly to temperature changes.
At progressively higher levels of global warming there are greater consequences (min/max range shown).
                                        +1.1&deg;C                +1.5&deg;C                                      +2&deg;C                                      +4&deg;C Today                                                                                                                                      TS Temperature                                                        +1.9&deg;C                                  +2.6&deg;C                                      +5.1&deg;C Hottest day in                                                  (+1.3 to 2.3&deg;C)                          (+1.8 to 3.1&deg;C)                            (+4.3 to 5.8&deg;C) a decade (+&deg;C)                  +1.2&deg;C
(+0.7 to 1.5&deg;C)
Drought A drought that used to occur                                        x2.0                                    x2.4                                        x4.1 once in a decade now                                            (x1.0 to 5.1)                            (x1.3 to 5.8)                                (x1.7 to 7.2) happens x times more x1.7 (x0.7 to 4.1)
Precipitation What used to be a wettest                                            x1.5                                    x1.7                                        x2.7 day in a decade now                                              (x1.4 to 1.7)                            (x1.6 to 2.0)                              (x2.3 to 3.6) happens x times more                x1.3 (x1.2 to1.4)
Snow Snow cover extent                                                    -5%                                      -9%                                        -26%
change (%)
                                              -1%                                                                    (-13 to 2)
(-3 to 1)                        (-7 to 2)                                                                            (-35 to -15)
Tropical cyclones Proportion of intense tropical                                      +10%                                    +13%                                        +30%
cyclones (%)
Long-term consequences: Sea level rise Today, sea level has already
                                                                +1.5&deg;C                                      +2&deg;C                                      +4&deg;C Metre rise                                Metre rise                                Metre rise increased by 20 cm and                                                                                                                                    33m will increase an additional 30 cm to 1 m or more by 2100, depending on future emissions.
16m                19m 13m Sea level reacts very slowly to                                        7m 12m 6m global warming so, once started,                    3m                6m 8m the rise continues for thousands                    2m                                      2m 2000-year        10,000-year              2000-year        10,000-year              2000-year      10,000-year of years.                                      commitment      commitment              commitment        commitment              commitment        commitment The future...
The climate we and the young generations will experience depends on future emissions.
Reducing emissions rapidly will limit further changes, but continued emissions will trigger larger, faster changes that will increasingly affect all regions. Some changes will persist for hundreds or thousands of years, so todays choices will have long-lasting consequences.
(top right) Response of some selected climate variables to four levels of global warming (&deg;C). Changes in the Today column are based on a global warming level of 1&deg;C.
(bottom right) The long-term effect of each global warming level on sea level. See Section TS.1.3.1 for more detail on the SSP climate change scenarios.
This infographic builds from several figures in the Technical Summary: Figure TS.4 (for top left panel), Figure TS.6 (bottom left), Figure TS.12 (top right) and Box TS.4, Figure 1b (bottom right).
89
 
Technical Summary TS.3            Understanding the Climate System                                                        TS.3.1          Radiative Forcing and Energy Budget Response and Implications for Limiting Global Warming                                                                              Since AR5, the accumulation of energy in the Earth system, quantified by observations of warming of the ocean, This section summarizes advances in our knowledge of Earths                                                atmosphere, and land and melting of ice, has become energy budget, including the time evolution of forcings and climate                                        established as a robust measure of the rate of global climate feedbacks that lead to the climate system responses summarized                                              change on interannual-to-decadal time scales. Compared to in Section TS.2. It assesses advances since AR5 and SR1.5 in the                                            changes in global surface temperature, the increase in the estimation of remaining carbon budgets, the Earth system response to                                        global energy inventory exhibits less variability, and thus carbon dioxide removal, and the quantification of metrics that allow                                        better indicates underlying climate trends.
comparisons of the relative effects of different forcing agents. The section also highlights: future climate and air pollution responses due                                    The global energy inventory increased by 282 [177 to to projected changes in short-lived climate forcers (SLCFs); the state                                      387] zettajoules (ZJ, equal to 1021 Joules) for the period of understanding of the climate response to potential interventions                                        1971-2006 and 152 [100 to 205] ZJ for the period 2006-TS related to solar radiation modification (SRM); and irreversibility,                                        2018 (Figure TS.13), with more than 90% accounted for tipping points and abrupt changes in the climate system.
(a) Global energy inventory                                      (b) Integrated radiative forcing                                (c) Integrated radiative response 2018                                                            2018                                                          2018 (d) Energy inventory components                                  (e) Radiative forcing components                                (f ) Energy budget 1971-2018 2018                                                            2018 Figure TS.13 l Estimates of the net cumulative energy change (ZJ = 1021 Joules) for the period 1971-2018 associated with (a) observations of changes in the global energy inventory, (b) integrated radiative forcing, and (c) integrated radiative response. The intent is to show assessed changes in energy budget and effective radiative forcings (ERFs). Black dotted lines indicate the central estimate with likely and very likely ranges as indicated in the legend. The grey dotted lines indicate the energy change associated with an estimated 1850-1900 Earth energy imbalance of 0.2 W m-2 (panel a) and an illustration of an assumed pattern effect of -0.5 W m-2 &deg;C-1 (panel c). Background grey lines indicate equivalent heating rates in W m-2 per unit area of Earths surface. Panels (d) and (e) show the breakdown of components, as indicated in the legend, for the global energy inventory and integrated radiative forcing, respectively. Panel (f) shows the global energy budget assessed for the period 1971-2018, that is, the consistency between the change in the global energy inventory relative to 1850-1900 and the implied energy change from integrated radiative forcing plus integrated radiative response under a number of different assumptions, as indicated in the figure legend, including assumptions of correlated and uncorrelated uncertainties in forcing plus response. Shading represents the very likely range for observed energy change relative to 1850-1900 and likely range for all other quantities. Forcing and response time series are expressed relative to a baseline period of 1850-1900. {Box 7.2, Figure 1}
90
 
Technical Summary by ocean warming. To put these numbers in context, the                                estimates of the global energy imbalance, and closure of the global 2006-2018 average Earth energy imbalance is equivalent                                sea level budget have led to a strengthened assessment relative to approximately 20 times the annual rate of global energy                            to AR5. (high confidence) {7.2.2, 7.5.2.3, Box 7.2, Table 7.1, 9.6.1, consumption in 2018. The accumulation of energy is driven                              Cross-Chapter Box 9.1, Table 9.5}
by a positive total anthropogenic effective radiative forcing (ERF) relative to 1750.                                                                As in AR5, the perturbations to Earths top-of-atmosphere energy budget are quantified using ERFs (see also Section TS.2.2). These The best estimate ERF of 2.72 W m2 has increased by 0.43                              include any consequent adjustments to the climate system (e.g.,
W m-2 relative to that given in AR5 (for 1750-2014) due                                from changes in atmospheric temperatures, clouds and water vapour to an increase in the greenhouse gas ERF that is partly                                as shown in Figure TS.14), but exclude any surface temperature compensated by a more negative aerosol ERF compared                                    response. Since AR5, ERFs have been estimated for a larger number to AR5. The greenhouse gas ERF has been revised due to                                of forcing agents and shown to be more closely related to the changes in atmospheric concentrations and updates to                                  temperature response than the stratospheric-temperature-adjusted forcing efficiencies, while the revision to aerosol ERF is due                        radiative forcing. (high confidence) {7.3.1}
to increased understanding of aerosol-cloud interactions                                                                                                                  TS and is supported by improved agreement between                                        Improved quantifications of ERF, the climate system radiative different lines of evidence. Improved quantifications                                  response, and the observed energy increase in the Earth system of ERF, the climate system radiative response, and the                                for the period 1971-2018 demonstrate improved closure of the observed energy increase in the Earth system for the                                  global energy budget relative to AR5 (Figure TS.13). Combining period 1971-2018 demonstrate improved closure of the                                  the likely range of ERF over this period with the central estimate global energy budget (i.e., the extent to which the sum                                of radiative response gives an expected energy gain of 340 [47 to of the integrated forcing and the integrated radiative                                662] ZJ. Both estimates are consistent with an independent response equals the energy gain of the Earth system)                                  observation-based assessment of the global energy increase of compared to AR5 (high confidence). (See FAQ 7.1). {7.2.2,                              284 [96 to 471] ZJ (very likely range), expressed relative to the 7.3.5, 7.5.2, Box 7.2, Table 7.1}                                                      estimated 1850-1900 Earth energy imbalance. (high confidence)
{7.2.2, 7.3.5, Box 7.2}
The global energy inventory change for the period 1971-2006 corresponds to an Earth energy imbalance (Box TS.1) of 0.50 [0.32                          The assessed greenhouse gas ERF over the 1750-2019 period to 0.69] W m-2, increasing to 0.79 [0.52 to 1.06] W m-2 for the period                    (Section TS.2.2) has increased by +0.59 W m2 over AR5 estimates 2006-2018. Ocean heat uptake is by far the largest contribution and                        for 1750-2011. This increase includes +0.34 W m-2 from increases in accounts for 91% of the total energy change. Land warming, melting                        atmospheric concentrations of well-mixed greenhouse gases (including of ice and warming of the atmosphere account for about 5%, 3% and                          halogenated species) since 2011, +0.15 W m -2 from upwards revisions 1% of the total change, respectively. More comprehensive analysis of                      of their radiative efficiencies and +0.10 W m -2 from re-evaluation of inventory components, cross-validation of satellite and in situ-based                      the ozone and stratospheric water vapour ERF. {7.3.2, 7.3.4, 7.3.5}
Atmospheric energy budget                    Instantaneous changes                    Adjustments                            Feedback responses perturbation to radiative balance                    to meteorology and composition Baseline                                  1                                      2                                        3 Adjustments in                    Biogeochemical        O chemical composition  O feedback    O    O O    O Higher clouds Absorbed                                                                                                                      Cloud sunlight                                                                                                                    feedback Adjustments in clouds, water vapour, temperature fewer low clouds Longwave                        Increased greenhouse                    Increased greenhouse                  Increased greenhouse radiative                      gases or aerosols                      gases or aerosols                      gases or aerosols emission                                                                          Surface and                            Surface albedo and vegetation                            biogeochemical response                              feedback Increased water vapour No change in meteorology or surface    No change in surface temperature        Increased surface temperature temperature Figure TS.14 l Schematic representation of changes in the top-of-atmosphere (TOA) radiation budget following a perturbation. The intent of this figure is to illustrate the concept of adjustments in the climate system following a perturbation in the radiation budget. The baseline TOA energy budget (a) responds instantaneously to perturbations (b), leading to adjustments in the atmospheric meteorology and composition and land surface that are independent of changes in surface temperature (c). Surface temperature changes (here using an increase as an example) lead to physical, biogeophysical and biogeochemical feedback processes (d). Long-term feedback processes, such as those involving ice sheets, are not shown here. {adapted from Figure 7.2; FAQ 7.2, Figure 1; and Figure 8.3}
91
 
Technical Summary (a) Effective radiative forcing                                                              (b) Change in global surface temperature 1750 to 2019                                                                                1750 to 2019 Emitted Components Climate effect through:
TS (c) Aerosol effective radiative forcing Figure TS.15 l Contribution to (a) effective radiative forcing (ERF) and (b) global surface temperature change from component emissions for 1750-2019 based on Coupled Model Intercomparison Project Phase 6 (CMIP6) models and (c) net aerosol ERF for 1750-2014 from different lines of evidence. The intent of this figure is to show advances since AR5 in the understanding of (a) emissions-based ERF, (b) global surface temperature response for short-lived climate forcers as estimated in Chapter 6, and (c) aerosol ERF from different lines of evidence as assessed in Chapter 7. In panel (a), ERFs for well-mixed greenhouse gases (WMGHGs) are from the analytical formulae. ERFs for other components are multi-model means based on Earth system model simulations that quantify the effect of individual components.
The derived emissions-based ERFs are rescaled to match the concentration-based ERFs in Figure 7.6. Error bars are 5-95% and for the ERF account for uncertainty in radiative efficiencies and multi-model error in the means. In panel (b), the global mean temperature response is calculated from the ERF time series using an impulse response function.
In panel (c), the AR6 assessment is based on energy balance constraints, observational evidence from satellite retrievals, and climate model-based evidence. For each line of evidence, the assessed best-estimate contributions from ERF due to aerosol-radiation interactions (ERFari) and aerosol-cloud interactions (ERFaci) are shown with darker and paler shading, respectively. Estimates from individual CMIP Phase 5 (CMIP5) and CMIP6 models are depicted by blue and red crosses, respectively. The observational assessment for ERFari is taken from the instantaneous forcing due to aerosol-radiation interactions (IRFari). Uncertainty ranges are given in black bars for the total aerosol ERF and depict very likely ranges. {6.4.2, Figure 6.12, 7.3.3, Cross-Chapter Box 7.1, Table 7.8, Figure 7.5}
For CO2, CH4, N2O, and chlorofluorocarbons, there is now evidence to                        emissions make the dominant contribution to the ERF from aerosol-quantify the effect on ERF of tropospheric adjustments. The assessed                        cloud interactions (high confidence). Over the 1750-2019 period, ERF for a doubling of CO2 compared to 1750 levels (3.9 +/- 0.5 Wm-2) is                      the contributions from the emitted compounds to global surface larger than in AR5. For CO2, the adjustments include the physiological                      temperature changes broadly match their contributions to the ERF effects on vegetation. The reactive well-mixed greenhouse gases                            (high confidence) (Figure TS.15b). Since a peak in emissions-induced (CH4, N2O, and halocarbons) cause additional chemical adjustments                          SO2 ERF has already occurred recently (Section TS.2.2) and since there to the atmosphere through changes in ozone and aerosols (Figure                            is a delay in the full global surface temperature response owing to TS.15a). The ERF due to CH4 emissions is 1.19 [0.81 to 1.58] W m-2, of                      the thermal inertia in the climate system, changes in SO2 emissions which 0.35 [0.16 to 0.54] W m-2 is attributed to chemical adjustments                      have a slightly larger contribution to global surface temperature mainly via ozone. These chemical adjustments also affect the                                change compared with changes in CO2 emissions, relative to their emissions metrics (Section TS.3.3.3). Changes in sulphur dioxide (SO2)                      respective contributions to ERF. {6.4.2, 7.3.2}
92
 
Technical Summary Aerosols contributed an ERF of -1.3 [-2.0 to -0.6] W m-2 over the        evolves and global surface temperature increases, leading to an ECS period 1750 to 2014 (medium confidence). The ERF due to aerosol-        that is higher than was inferred in AR5 based on warming over the cloud interactions (ERFaci) contributes most to the magnitude of        instrumental record (high confidence). Historical surface temperature the total aerosol ERF (high confidence) and is assessed to be -1.0      change since 1870 has shown relatively little warming in several key
[-1.7 to -0.3] W m-2 (medium confidence), with the remainder due        regions of positive feedbacks, including the eastern equatorial Pacific to aerosol-radiation interactions (ERFari), assessed to be -0.3 [-0.6    Ocean and the Southern Ocean, while showing greater warming to 0.0] W m-2 (medium confidence). There has been an increase in        in key regions of negative feedbacks, including the western Pacific the estimated magnitude - but a reduction in the uncertainty - of        warm pool. Based on process understanding, climate modelling, and the total aerosol ERF relative to AR5, supported by a combination        paleoclimate reconstructions of past warm periods, it is expected of increased process-understanding and progress in modelling and        that future warming will become enhanced over the eastern Pacific observational analyses (Figure TS.15c). Effective radiative forcing      Ocean (medium confidence) and Southern Ocean (high confidence) estimates from these separate lines of evidence are now consistent      on centennial time scales. This new understanding, along with with each other, in contrast to AR5, and support the assessment that    updated estimates of historical temperature change, ERF, and energy it is virtually certain that the total aerosol ERF is negative. Compared imbalance, reconciles previously disparate ECS estimates. {7.4.4, to AR5, the assessed magnitude of ERFaci has increased, while that      7.5.2, 7.5.3}                                                              TS of ERFari has decreased. {7.3.3, 7.3.5}
The AR6 best estimate of ECS is 3&deg;C, the likely range is 2.5&deg;C to 4&deg;C and the very likely range is 2&deg;C to 5&deg;C. There is a high level TS.3.2        Climate Sensitivity and Earth System Feedbacks            of agreement among the four main lines of evidence listed above (Figure TS.16b), and altogether it is virtually certain that ECS is larger TS.3.2.1 Equilibrium Climate Sensitivity, Transient Climate              than 1.5&deg;C, but currently it is not possible to rule out ECS values Response, and Transient Climate Response to                above 5&deg;C. Therefore, the 5&deg;C upper end of the very likely range is Cumulative Carbon-dioxide Emissions                        assessed with medium confidence and the other bounds with high confidence. {7.5.5}
Since AR5, substantial quantitative progress has been made in combining new evidence of Earths climate                    Based on process understanding, warming over the instrumental sensitivity with improvements in the understanding and              record, and emergent constraints, the best estimate of TCR is 1.8&deg;C, quantification of Earths energy imbalance, the instrumental        the likely range is 1.4&deg;C to 2.2&deg;C and the very likely range is 1.2&deg;C to record of global surface temperature change, paleoclimate            2.4&deg;C. There is a high level of agreement among the different lines of change from proxy records, climate feedbacks and their              evidence (Figure TS.16c) (high confidence). {7.5.5}
dependence on time scale and climate state. A key advance is the broad agreement across these multiple lines of                On average, CMIP6 models have higher mean ECS and TCR values evidence, supporting a best estimate of equilibrium climate          than the CMIP5 generation of models and also have higher mean sensitivity of 3&deg;C, with a very likely range of 2&deg;C to 5&deg;C. The      values and wider spreads than the assessed best estimates and very likely range of 2.5&deg;C to 4&deg;C is narrower than the AR5 likely        likely ranges within this Report. These higher mean ECS and TCR range of 1.5&deg;C to 4.5&deg;C. {7.4, 7.5}                                  values can be traced to a positive net cloud feedback that is larger in CMIP6 by about 20%. The broader ECS and TCR ranges from CMIP6 Constraints on equilibrium climate sensitivity (ECS) and transient      also lead the models to project a range of future warming that climate response (TCR) (see Glossary) are based on four main lines      is wider than the assessed future warming range, which is based of evidence: feedback process understanding, climate change and          on multiple lines of evidence (Cross-Section Box TS.1). However, variability seen within the instrumental record, paleoclimate evidence,  some of the high-sensitivity CMIP6 models (Section TS.1.2.2) are and so-called emergent constraints, whereby a relationship            less consistent with observed recent changes in global warming between an observable quantity and either ECS or TCR established        and with paleoclimate proxy records than models with ECS within within an ensemble of models is combined with observations to            the very likely range. Similarly, some of the low-sensitivity models derive a constraint on ECS or TCR. In reports up to and including        are less consistent with the paleoclimate data. The CMIP6 models the IPCC Third Assessment Report, ECS and TCR derived directly          with the highest ECS and TCRs values provide insights into low-from ESMs were the primary line of evidence. However, since AR4,        likelihood, high-impact futures, which cannot be excluded based on historical warming and paleoclimates provided useful additional          currently available evidence (Cross-Section Box TS.1). {4.3.1, 4.3.4, evidence (Figure TS.16a). This Report differs from previous reports in  7.4.2, 7.5.6}
not directly using climate model estimates of ECS and TCR in the assessed ranges of climate sensitivity. {1.5, 7.5}                      Uncertainties regarding the true value of ECS and TCR are the dominant source of uncertainty in global temperature projections over the It is now clear that when estimating ECS and TCR, the dependence        21st century under moderate to high GHG concentrations scenarios.
of feedbacks on time scales and the climate state must be accounted      For scenarios that reach net zero CO2 emissions (Section TS.3.3), the for. Feedback processes are expected to become more positive            uncertainty in the ERF values of aerosol and other SLCFs contribute overall (more amplifying of global surface temperature changes) on      substantial uncertainty in projected temperature. Global ocean heat multi-decadal time scales as the spatial pattern of surface warming      uptake is a smaller source of uncertainty in centennial warming. {7.5.7}
93
 
Technical Summary (a) Evolution of equilibrium climate sensitivity assessments from Charney to AR6 p < 10%
6                                                                                                                                                              6 Equilibrium climate sensitivity (&deg;C) 5                                                                                                                              Very likely: 2-5&deg;C              5 FAR      SAR                      AR4          AR5 Charney 4                                                            TAR                                                              Likely: 2.5-4&deg;C                4 Likely: p > 66%
AR6 3                                                                                                                              Best estimate: 3&deg;C              3 2                                                                                                                                                              2 AR6 combines evidence from:
* Process understanding
* Instrumental record 1                Primarily model evidence                                                p < 5%
* Paleoclimates 1
Emergent constraints Also considers instrumental record and paleoclimates TS 1980                1990                  2000                  2010                            2020                  2030 Year of assessment (b) Equilibrium climate sensitivity (&#xba;C) assessed in AR6 and                                                                (c) Transient climate response (&#xba;C) assessed in AR6 and simulated by CMIP6 ESMs                                                                                                      simulated by CMIP6 ESMs Process understanding                                                                                                        Process understanding Instrumental record                                                                                                          Instrumental record Paleoclimates                                                                                                                Paleoclimates Emergent constraints                                                                                                        Emergent constraints Combined assessment                                                                                                          Combined assessment CMIP6 ESMs                                                                                                                  CMIP6 ESMs 0          1  2    3  4    5    6      7    8      9                                            0              1              2          3                4 Best estimate range or value                      Likely range or limit                                      Very likely range or limit            Extremely likely limit Figure TS.16 l (a) Evolution of equilibrium climate sensitivity (ECS) assessments from the Charney Report through a succession of IPCC Assessment Reports to AR6, and lines of evidence and combined assessment for (b) ECS and (c) transient climate response (TCR) in AR6. The intent of this figure is to show the progression in estimates of ECS, including uncertainty and the lines of evidence used for assessment, and to show the lines of assessment used to assess ECS and TCR in AR6. In panel (a), the lines of evidence considered are listed below each assessment. Best estimates are marked by horizontal bars, likely ranges by vertical bars, and very likely ranges by dotted vertical bars. In panel (b) and (c), assessed ranges are taken from Tables 7.13 and 7.14 for ECS and TCR respectively. Note that for the ECS assessment based on both the instrumental record and paleoclimates, limits (i.e., one-sided distributions) are given, which have twice the probability of being outside the maximum/minimum value at a given end, compared to ranges (i.e., two tailed distributions) which are given for the other lines of evidence. For example, the extremely likely limit of greater than 95% probability corresponds to one side of the very likely (5% to 95%) range. Best estimates are given as either a single number or by a range represented by grey box. Coupled Model Intercomparison Project Phase 6 (CMIP6) Earth system model (ESM) values are not directly used as a line of evidence but are presented on the figure for comparison. {1.5, 7.5; Tables 7.13 and 7.14; Figure 7.18}
The transient climate response to cumulative CO2 emissions (TCRE)                                                                    of TCR. Beyond this century, there is low confidence that the TCRE is the ratio between globally averaged surface temperature increase                                                                  alone remains an accurate predictor of temperature changes in and cumulative CO2 emissions (see Glossary). This Report reaffirms                                                                    scenarios of very low or net negative CO2 emissions because of with high confidence the finding of AR5 that there is a near-linear                                                                  uncertain Earth system feedbacks that can result in further changes relationship between cumulative CO2 emissions and the increase in                                                                    in temperature or a path dependency of warming as a function of global average temperature caused by CO2 over the course of this                                                                      cumulative CO2 emissions. {4.6.2, 5.4, 5.5.1}
century for global warming levels up to at least 2&deg;C relative to 1850-1900. The TCRE falls likely in the 1.0&deg;C-2.3&deg;C per 1000 PgC range,                                                                    TS.3.2.2 Earth System Feedbacks with a best estimate of 1.65&deg;C per 1000 PgC. This is equivalent to a 0.27&deg;C-0.63&deg;C range with a best estimate of 0.45&deg;C when expressed                                                                    The combined effect of all climate feedback processes is to amplify in units per 1000 GtCO2. This range is about 15% narrower than the                                                                    the climate response to forcing (virtually certain). While major 0.8&deg;-2.5&deg;C per 1000 PgC assessment of AR5 because of a better                                                                        advances in the understanding of cloud processes have increased integration of evidence across chapters, in particular the assessment                                                                the level of confidence and decreased the uncertainty range for the 94
 
Technical Summary cloud feedback by about 50% compared to AR5, clouds remain the                                    Natural sources and sinks of non-CO2 greenhouse gases such as largest contribution to overall uncertainty in climate feedbacks (high                            methane (CH4) and nitrous oxide (N2O) respond both directly and confidence). Uncertainties in the ECS and other climate sensitivity                                indirectly to atmospheric CO2 concentration and climate change, metrics, such as the TCR and TCRE, are the dominant source of                                      and thereby give rise to additional biogeochemical feedbacks uncertainty in global temperature projections over the 21st century                                in the climate system. Many of these feedbacks are only partially under moderate to high GHG emissions scenarios. CMIP6 models                                      understood and are not yet fully included in ESMs. There is medium have higher mean values and wider spreads in ECS and TCR than the                                  confidence that the net response of natural ocean and land CH4 and assessed best estimates and very likely ranges within this Report,                                N2O sources to future warming will be increased emissions, but the leading the models to project a range of future warming that is                                    magnitude and timing of the responses of each individual process is wider than the assessed future warming range (Section TS.2.2). {7.1,                              known with low confidence. {5.4.7}
7.4.2, 7.5}
Non-CO2 biogeochemical feedbacks induced from changes in Earth system feedbacks can be categorized into three broad groups:                                emissions, abundances or lifetimes of SLCFs mediated by natural physical feedbacks, biogeophysical and biogeochemical feedbacks,                                  processes or atmospheric chemistry are assessed to decrease and feedbacks associated with ice sheets. In previous assessments,                                ECS (Figure TS.17b). These non-CO2 biogeochemical feedbacks            TS the ECS has been associated with a distinct set of physical feedbacks                              are estimated from ESMs, which since AR5 have advanced to (Planck response, water vapour, lapse rate, surface albedo, and cloud                              include a consistent representation of biogeochemical cycles and feedbacks). In this assessment, a more general definition of ECS is                                atmospheric chemistry. However, process-level understanding adopted whereby all biogeophysical and biogeochemical feedbacks                                    of many biogeochemical feedbacks involving SLCFs, particularly that do not affect the atmospheric concentration of CO2 are included.                              natural emissions, is still emerging, resulting in low confidence in These include changes in natural CH4 emissions, natural aerosol                                    the magnitude and sign of the feedbacks. The central estimate of emissions, N2O, ozone, and vegetation, which all act on time scales of                            the total biogeophysical and non-CO2 biogeochemical feedback is years to decades and are therefore relevant for temperature change                                assessed to be 0.01 [-0.27 to +0.25] W m-2 &deg;C-1 (Figure TS.17a).
over the 21st century. Because the total biogeophysical and non-CO2                                {5.4.7, 5.4.8, 6.2.2, 6.4.5, 7.4, Table 7.10}
biogeochemical feedback is assessed to have a central value that is near zero (low confidence), including it does not affect the assessed                              The combined effect of all known radiative feedbacks (physical, ECS but does contribute to the net feedback uncertainty. The                                      biogeophysical, and non-CO2 biogeochemical) is to amplify the biogeochemical feedbacks that affect the atmospheric concentration                                base climate response (in the absence of feedbacks), also known of CO2 are not included because ECS is defined as the response to                                  as the Planck temperature response20 (virtually certain). Combining a sustained doubling of CO2. Moreover, the long-term feedbacks                                    these feedbacks with the Planck response, the net climate feedback associated with ice sheets are not included in the ECS owing to their                              parameter is assessed to be -1.16 [-1.81 to -0.51] W m-2 &deg;C-1, long time scales of adjustment. {5.4, 6.4, 7.4, 7.5, Box 7.1}                                      which is slightly less negative than that inferred from the overall ECS assessment. The combined water vapour and lapse rate feedback The net effect of changes in clouds in response to global warming is                              makes the largest single contribution to global warming, whereas to amplify human-induced warming, that is, the net cloud feedback                                  the cloud feedback remains the largest contribution to overall is positive (high confidence). Compared to AR5, major advances                                    uncertainty. Due to the state-dependence of feedbacks, as evidenced in the understanding of cloud processes have increased the level                                  from paleoclimate observations and from models, the net feedback of confidence and decreased the uncertainty range in the cloud                                    parameter will increase (become less negative) as global temperature feedback by about 50% (Figure TS.17a). An assessment of the low-                                  increases. Furthermore, on long time scales the ice-sheet feedback altitude cloud feedback over the subtropical ocean, which was                                      parameter is very likely positive, promoting additional warming on previously the major source of uncertainty in the net cloud feedback,                              millennial time scales as ice sheets come into equilibrium with the is improved owing to a combined use of climate model simulations,                                  forcing. (high confidence) {7.4.2, 7.4.3, Figure 7.14, Table 7.10}
satellite observations, and explicit simulations of clouds, altogether leading to strong evidence that this type of cloud amplifies global                                The carbon cycle provides for additional feedbacks on climate owing warming. The net cloud feedback is assessed to be +0.42 [-0.10                                    to the sensitivity of land-atmosphere and ocean-atmosphere carbon to 0.94] W m -2 &deg;C-1. A net negative cloud feedback is very unlikely.                              fluxes and storage to changes in climate and in atmospheric CO2 The CMIP5 and CMIP6 ranges of cloud feedback are similar to this                                  (Figure TS.17c). Because of the time scales associated with land and assessed range, with CMIP6 having a slightly more positive median                                  ocean carbon uptake, these feedbacks are known to be scenario cloud feedback (high confidence). The surface albedo feedback and                                  dependent. Feedback estimates deviate from linearity in scenarios combined water vapour-lapse rate feedback are positive (Figure                                    of stabilizing or reducing concentrations. With high confidence, TS.17a), with high confidence in the estimated value of each based                                increased atmospheric CO2 will lead to increased land and ocean on multiple lines of evidence, including observations, models and                                  carbon uptake, acting as a negative feedback on climate change. It theory (Box TS.6). {7.4.2, Figure 7.14, Table 7.10}                                                is likely that a warmer climate will lead to reduced land and ocean carbon uptake, acting as a positive feedback (Box TS.5). {4.3.2, 5.4.1-5}
20  For reference, the Planck temperature response for a doubling of atmospheric CO2 is approximately 1.2&deg;C at equilibrium.
95
 
Technical Summary (a) Feedbacks in the climate system Negative feedbacks diminish the                      Positive feedbacks amplify the initial climate response to radiative forcing      initial climate response to radiative forcing Mean [very likely range]
Total                                                                                                              1.16 [1.81 to 0.51]
Planck                                                                                                              3.22 [3.39 to 3.05]
Water vapour and lapse rate                                                                                                                1.30 [ 1.13 to 1.47]
Surface albedo                                                                                                                0.35 [ 0.10 to 0.60]
Clouds                                                                                                                0.42 [0.10 to 0.94]
Biogeophysical and                                                                                                                0.01 [0.27 to 0.25]
nonCO2 biogeochemical (Total from panel (b))
TS                                        3.5    3.0      2.5  2.0  1.5  1.0    0.5  0.0    0.5      1.0    1.5    2.0  2.5      3.0  3.5 Climate feedback parameter (Wm-2 &deg;C-1)
(b) Biogeophysical and non-CO2 biogeochemical climate feedbacks                                              Mean [595% range]
Total (without permafrost feedback)                                                                                                                0.01 [0.27 to 0.25]
Permafrost feedback CH4 source response to climate                                                                                                                  0.03 [ 0.01 to 0.05]
Atm. CH4 lifetime response to climate                                                                                                                0.03 [0.12 to 0.06]
N2O source response to climate                                                                                                                  0.01 [0.01 to 0.02]
Other nonCO2 biogeochemical                                                                                                                  0.17 [0.36 to 0.02]
Biogeophysical                                                                                                                0.15 [ 0.00 to 0.30]
0.4          0.3        0.2        0.1        0.0            0.1          0.2          0.3        0.4 Climate feedback parameter (Wm-2 &deg;C-1)
(c) Carbon-cycle climate feedbacks                                                                          Mean [595% range]
Land carbon response to CO2                                                                                                                  0.78 [1.28 to 0.28]
Ocean carbon response to CO2                                                              Permafrost feedback                                0.68 [0.98 to 0.39]
Land carbon response to climate                                                                                                                  0.25 [0.03 to 0.54]
Ocean carbon response to climate                                                                                                                  0.08 [0.04 to 0.12]
3.5    3.0      2.5  2.0  1.5  1.0    0.5  0.0    0.5      1.0    1.5    2.0  2.5      3.0  3.5 Climate feedback parameter (Wm-2 &deg;C-1)
Figure TS.17 l An overview of physical and biogeochemical feedbacks in the climate system. The intent of this figure is to summarize assessed estimates of physical, biogeophysical and biogeochemical feedbacks on global temperature based on Chapters 5, 6 and 7. (a) Synthesis of physical, biogeophysical and non-carbon dioxide (CO2) biogeochemical feedbacks that are included in the definition of equilibrium climate sensitivity (ECS) assessed in this Technical Summary. These feedbacks have been assessed using multiple lines of evidence including observations, models and theory. The net feedback is the sum of the Planck response, water vapour and lapse rate, surface albedo, cloud, and biogeophysical and non-CO2 biogeochemical feedbacks. Bars denote the mean feedback values, and uncertainties represent very likely ranges; (b) Estimated values of individual biogeophysical and non-CO2 biogeochemical feedbacks. The atmospheric methane (CH4) lifetime and other non-CO2 biogeochemical feedbacks have been calculated using global Earth system model simulations from AerChemMIP, while the CH4 and nitrous oxide (N2O) source responses to climate have been assessed for the year 2100 using a range of modelling approaches using simplified radiative forcing equations. The estimates represent the mean and 5-95% range. The level of confidence in these estimates is low owing to the large model spread. (c) Carbon-cycle feedbacks as simulated by models participating in the C4MIP of the Coupled Model Intercomparison Project Phase 6 (CMIP6). An independent estimate of the additional positive carbon-cycle climate feedbacks from permafrost thaw, which is not considered in most C4MIP models, is added.
The estimates represent the mean and 5-95% range. Note that these feedbacks act through modifying the atmospheric concentration of CO2 and thus are not included in the definition of ECS, which assumes a doubling of CO2, but are included in the definition and assessed range of the transient climate response to cumulative CO2 emissions (TCRE).
{5.4.7, 5.4.8, Box 5.1, Figure 5.29, 6.4.5, Table 6.9, 7.4.2, Table 7.10}
96
 
Technical Summary Thawing terrestrial permafrost will lead to carbon release (high                                                                                                                            TS.3.3                      Temperature Stabilization, Net Zero Emissions confidence), but there is low confidence in the timing, magnitude                                                                                                                                                      and Mitigation and the relative roles of CO2 versus CH4 as feedback processes. An ensemble of models projects CO2 release from permafrost to be                                                                                                                              TS.3.3.1 Remaining Carbon Budgets and Temperature 3-41 PgC per 1&#xba;C of global warming by 2100, leading to warming                                                                                                                                      Stabilization strong enough that it must be included in estimates of the remaining carbon budget but weaker than the warming from fossil fuel burning.                                                                                                                                    The near-linear relationship between cumulative CO2 However, the incomplete representation of important processes,                                                                                                                                          emissions and maximum global surface temperature increase such as abrupt thaw, combined with weak observational constraints,                                                                                                                                      caused by CO2 implies that stabilizing human-induced global only allow low confidence in both the magnitude of these estimates                                                                                                                                      temperature increase at any level requires net anthropogenic and in how linearly proportional this feedback is to the amount of                                                                                                                                      CO2 emissions to become zero. This near-linear relationship global warming. There is emerging evidence that permafrost thaw                                                                                                                                        further implies that mitigation requirements for limiting and thermokarst give rise to increased CH4 and N2O emissions, which                                                                                                                                    warming to specific levels can be quantified in terms of a leads to the combined radiative forcing from permafrost thaw being                                                                                                                                      carbon budget (high confidence). Remaining carbon budget larger than from CO2 emissions only. However, the quantitative                                                                                                                                          estimates have been updated since AR5 with methodological                                          TS understanding of these additional feedbacks is low, particularly for                                                                                                                                    improvements, resulting in larger estimates that are N2O. These feedbacks, as well as potential additional carbon losses                                                                                                                                    consistent with SR1.5. Several factors, including estimates due to climate-induced fire feedback are not routinely included in                                                                                                                                      of historical warming, future emissions from thawing Earth system models. {Box 5.1, 5.4.3, 5.4.7, 5.4.8}                                                                                                                                                    permafrost, variations in projected non-CO2 warming, and the global surface temperature change after cessation of CO2 emissions, affect the exact value of carbon budgets (high confidence). {1.3.5, Box 1.2, 4.7.1, 5.5}
(a)                                                                  Cumulative carbon emissions since 1850 (PgC)
(b) 0      200      400    600  800    1000  1200  1400  1600  1800  2000 7
Global mean temperature increase since 1850-1900 (&deg;C)
Historical observations                                                                                                                                                                                            TCRE Human-induced warming estimate                                                                                            global warming limit of interest                                                (Section 5.5.1.4)
Temperature increase since pre-industrial levels (&deg;C) 6                Human-induced warming assessment                                                                                                    zero emissions commitment (ZEC, Section 5.5.2.2.4)
Assessed TCRE range non-CO2 contribution (Section 5.5.2.2.3) 5        Scenarios:
SSP1-1.9 SSP1-2.6 4        SSP2-4.5 remaining SSP3-7.0 SSP5-8.5 3
2                                                                                                                                          allowable warming 1
historical 0                                                                                                                                                                      human-induced warming 0                                    (Section 5.5.2.2.2)
                                                        -1 0        1000          2000  3000      4000      5000 6000    7000                                                                                        0                      remaining          unrepresented Earth system carbon budget          feedbacks (Section 5.5.2.2.5)
Cumulative carbon dioxide emissions since 1850 (GtCO2)
Cumulative CO2 emissions from today (GtCO2)
Figure TS.18 l Illustration of (a) relationship between cumulative emissions of carbon dioxide (CO2) and global mean surface air temperature increase and (b) the assessment of the remaining carbon budget from its constituting components based on multiple lines of evidence. The intent of this figure is to show (i) the proportionality between cumulative CO2 emissions and global surface air temperature in observations and models as well as the assessed range of the transient climate response to cumulative CO2 emissions (TCRE), and (ii) how information is combined to derive remaining carbon budgets consistent with limiting warming to a specific level. Carbon budgets consistent with various levels of additional warming are provided in Table 5.8 and should not be read from the illustrations in either panel. In panel (a) thin black line shows historical CO2 emissions together with the assessed global surface temperature increase from 1850-1900 as assessed in Chapter 2 (Box 2.3). The orange-brown range with its central line shows the estimated human-induced share of historical warming. The vertical orange-brown line shows the assessed range of historical human-induced warming for the 2010-2019 period relative to 1850-1900 (Chapter 3). The grey cone shows the assessed likely range for the TCRE (Section 5.5.1.4), starting from 2015. Thin coloured lines show Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations for the five scenarios of the WGI core set (SSP1-1.9, green; SSP1-2.6, blue; SSP2-4.5, yellow; SSP3-7.0, red; SSP5-8.5, maroon), starting from 2015 and until 2100. Diagnosed carbon emissions are complemented with estimated land-use change emissions for each respective scenario. Coloured areas show the Chapter 4 assessed very likely range of global surface temperature projections and thick coloured central lines show the median estimate, for each respective scenario. These temperature projections are expressed relative to cumulative CO2 emissions that are available for emissions-driven CMIP6 ScenarioMIP experiments for each respective scenario. For panel (b), the remaining allowable warming is estimated by combining the global warming limit of interest with the assessed historical human-induced warming (Section 5.5.2.2.2), the assessed future potential non-CO2 warming contribution (Section 5.5.2.2.3) and the zero emissions commitment (ZEC; Section 5.5.2.2.4). The remaining allowable warming (vertical blue bar) is subsequently combined with the assessed TCRE (Sections 5.5.1.4 and 5.5.2.2.1) and contribution of unrepresented Earth system feedbacks (Section 5.5.2.2.5) to provide an assessed estimate of the remaining carbon budget (horizontal blue bar, Table 5.8). Note that contributions in panel (b) are illustrative and are not to scale. For example, the central ZEC estimate was assessed to be zero. {Box 2.3, 5.2.1, 5.2.2, Figure 5.31}
97
 
Technical Summary Limiting further climate change would require substantial and                              but they may vary by an estimated +/-220 GtCO2 depending on sustained reductions of GHG emissions. Without net zero CO2                                how deeply future non-CO2 emissions are assumed to be reduced emissions, and a decrease in the net non-CO2 forcing (or sufficient                        (Table TS.3). {5.5.2, 5.6, Box 5.2, 7.6}
net negative CO2 emissions to offset any further warming from net non-CO2 forcing), the climate system will continue to warm.                            There is high confidence that several factors, including estimates of There is high confidence that mitigation requirements for limiting                        historical warming, future emissions from thawing permafrost, and warming to specific levels over this century can be estimated using                        variations in projected non-CO2 warming, affect the value of carbon a carbon budget that relates cumulative CO2 emissions to global                            budgets but do not change the conclusion that global CO2 emissions mean temperature increase (Figure TS.18, Table TS.3). For the period                      would need to decline to net zero to halt global warming. Estimates 1850-2019, a total of 2390 +/- 240 GtCO2 of anthropogenic CO2 has                            may vary by +/-220 GtCO2 depending on the level of non-CO2 emissions been emitted. Remaining carbon budgets (starting from 1 January                            at the time global anthropogenic CO2 emissions reach net zero levels.
2020) for limiting warming to 1.5&deg;C, 1.7&deg;C and 2.0&deg;C are estimated                        This variation is referred to as non-CO2 scenario uncertainty and at 500 GtCO2, 850 GtCO2 and 1350 GtCO2, respectively, based on                            will be further assessed in the AR6 Working Group III Contribution.
the 50th percentile of TCRE. For the 67th percentile, the respective                      Geophysical uncertainties surrounding the climate response to these TS values are 400 GtCO2, 700 GtCO2 and 1150 GtCO2. The remaining                              non-CO2 emissions result in an additional uncertainty of at least carbon budget estimates for different temperature limits assume                            +/-220 GtCO2, and uncertainties in the level of historical warming that non-CO2 emissions are mitigated consistent with the median                            result in a +/-550 GtCO2 uncertainty. {5.4, 5.5.2}
reductions found in scenarios in the literature as assessed in SR1.5, Table TS.3 l Estimates of remaining carbon budgets and their uncertainties. Assessed estimates are provided for additional human-induced warming, expressed as global surface temperature, since the recent past (2010-2019), likely amounted to 0.8&deg; to 1.3&deg;C with a best estimate of 1.07&deg;C relative to 1850-1900. Historical CO2 emissions between 1850 and 2014 have been estimated at about 2180 +/- 240 GtCO2 (1-sigma range), while since 1 January 2015, an additional 210 GtCO2 has been emitted until the end of 2019. GtCO2 values to the nearest 50. {Table 3.1, 5.5.1, 5.5.2, Box 5.2, Table 5.1, Table 5.7, Table 5.8}
Global            Global              Estimated remaining carbon surface          surface            budgets starting from 1 January Scenario temperature        temperature            2020 and subject to variations                                                Geophysical uncertaintiesd variation change since      change since        and uncertainties quantified in the 2010-2019          1850-1900a                columns on the right Non-CO2                          Zero CO2 Non-CO2                            Historical                          Recent Percentiles of TCREb                                    forcing and                        emissions
            &deg;C                &deg;C                                                          scenario                          temperature                          emissions GtCO2                                              response                        commitment variationc                        uncertaintya                        uncertaintye uncertainty                        uncertainty 17th    33rd    50th      67th    83rd          GtCO2                GtCO2          GtCO2            GtCO2            GtCO2 0.43              1.5          900      650      500      400    300                          Values can Values can vary by at least 0.53              1.6          1200      850      650      550    400      vary by at least
                                                                                                            +/-220 due to 0.63              1.7          1450    1050      850      700    550      +/-220 due to uncertainty in choices related                            +/-550              +/-420            +/-20 0.73              1.8          1750    1250    1000      850    650                          the warming to non-CO2 response to 0.83              1.9          2000    1450    1200      1000    800      emissions future non-CO2 mitigation 0.93                2          2300    1700    1350      1150    900                          emissions a
Human-induced global surface temperature increase between 1850-1900 and 2010-2019 is assessed at 0.8-1.3&deg;C (likely range; Cross-Section Box TS.1) with a best estimate of 1.07&deg;C. Combined with a central estimate of TCRE (1.65&deg;C per 1000 PgC) this uncertainty in isolation results in a potential variation of remaining carbon budgets of +/-550 GtCO2, which, however, is not independent of the assessed uncertainty of TCRE and thus not fully additional.
b TCRE: transient climate response to cumulative emissions of carbon dioxide, assessed to fall likely between 1.0-2.3&deg;C per 1000 PgC with a normal distribution, from which the percentiles are taken. Additional Earth system feedbacks are included in the remaining carbon budget estimates as discussed in Section 5.5.2.2.5.
c Estimates assume that non-CO2 emissions are mitigated consistent with the median reductions found in scenarios in the literature as assessed in SR1.5. Non-CO2 scenario variations indicate how much remaining carbon budget estimates vary due to different scenario assumptions related to the future evolution of non-CO2 emissions in mitigation scenarios from SR1.5 that reach net zero CO2 emissions. This variation is additional to the uncertainty in TCRE. The Working Group III Contribution to AR6 will reassess the potential for non-CO2 mitigation based on literature since SR1.5.
d Geophysical uncertainties reported in these columns and TCRE uncertainty are not statistically independent, as uncertainty in TCRE depends on uncertainty in the assessment of historical temperature, non-CO2 versus CO2 forcing, and uncertainty in emissions estimates. These estimates cannot be formally combined, and these uncertainty variations are not directly additional to the spread of remaining carbon budgets due to TCRE uncertainty reported in columns three to seven.
e Recent emissions uncertainty reflects the +/-10% uncertainty in the historical CO2 emissions estimate since 1 January 2015.
98
 
Technical Summary Methodological improvements and new evidence result in updated              to be removed to compensate for an emission of a given remaining carbon budget estimates. The assessment in AR6 applies            magnitude to attain the same change in atmospheric CO2 the same methodological improvements as in SR1.5, which uses                (medium confidence). CDR methods have wide-ranging side-a recent observed baseline for historic temperature change and              effects that can either weaken or strengthen the carbon cumulative emissions. Changes compared to SR1.5 are therefore                sequestration and cooling potential of these methods and small: the assessment of new evidence results in updated median              affect the achievement of sustainable development goals remaining carbon budget estimates for limiting warming to 1.5&deg;C              (high confidence). {4.6.3, 5.6}
and 2&deg;C being the same and about 60 GtCO2 smaller, respectively, after accounting for emissions since SR1.5. Meanwhile, remaining          Carbon dioxide removal (CDR) refers to anthropogenic activities carbon budgets for limiting warming to 1.5&deg;C would be about              that deliberately remove CO2 from the atmosphere and durably 300-350 GtCO2 larger if evidence and methods available at the            store it in geological, terrestrial or ocean reservoirs, or in products.
time of AR5 would be used. If a specific remaining carbon budget          Carbon dioxide is removed from the atmosphere by enhancing is exceeded, this results in a lower probability of keeping warming      biological or geochemical carbon sinks or by direct capture of CO2 below a specified temperature level and higher irreversible global        from air. Emissions pathways that limit global warming to 1.5&deg;C or warming over decades to centuries, or alternatively a need for net        2&deg;C typically assume the use of CDR approaches in combination            TS negative CO2 emissions or further reductions in non-CO2 greenhouse        with GHG emissions reductions. CDR approaches could be used to gases after net zero CO2 is achieved to return warming to lower levels    compensate for residual emissions from sectors that are difficult or in the long term. {5.5.2, 5.6, Box 5.2}                                  costly to decarbonize. CDR could also be implemented at a large scale to generate global net negative CO2 emissions (i.e., anthropogenic Based on idealized model simulations that explore the climate            CO2 removals exceeding anthropogenic emissions), which could response once CO2 emissions have been brought to zero, the                compensate for earlier emissions as a way to meet long-term climate magnitude of the zero CO2 emissions commitment (ZEC, see Glossary)        stabilization goals after a temperature overshoot. This Report assesses is assessed to be likely smaller than 0.3&deg;C for time scales of about      the effects of CDR on the carbon cycle and climate. Co-benefits and half a century and cumulative CO2 emissions broadly consistent with      trade-offs for biodiversity, water and food production are briefly global warming of 2&deg;C. However, there is low confidence about its        discussed for completeness, but a comprehensive assessment of the sign on time scales of about half a century. For lower cumulative CO2    ecological and socio-economic dimensions of CDR options is left to emissions, the range would be smaller yet with equal uncertainty          the WGII and WGIII reports. {4.6.3, 5.6}
about the sign. If the ZEC is positive on decadal time scales, additional warming leads to a reduction in the estimates of remaining carbon        CDR methods have the potential to sequester CO2 from the budgets, and vice versa if it is negative. {4.7.1, 5.5.2}                atmosphere (high confidence). In the same way part of current anthropogenic net CO2 emissions are taken up by land and ocean Permafrost thaw is included in estimates together with other              carbon stores, net CO2 removal will be partially counteracted by CO2 feedbacks that are often not captured by models. Limitations in          release from these stores, such that the amount of CO2 sequestered modelling studies combined with weak observational constraints            by CDR will not result in an equivalent drop in atmospheric CO2 (very only allow low confidence in the magnitude of these estimates            high confidence). The fraction of CO2 removed from the atmosphere (Section TS.3.2.2). Despite the large uncertainties surrounding the      that is not replaced by CO2 released from carbon stores - a measure quantification of the effect of additional Earth system feedback          of CDR effectiveness - decreases slightly with increasing amounts of processes, such as emissions from wetlands and permafrost thaw,          removal (medium confidence) and decreases strongly if CDR is applied these feedbacks represent identified additional risk factors that        at lower atmospheric CO2 concentrations (medium confidence). The scale with additional warming and mostly increase the challenge of        reduction in global surface temperature is approximately linearly limiting warming to specific temperature levels. These uncertainties      related to cumulative CO2 removal (high confidence). Because of this do not change the basic conclusion that global CO2 emissions would        near-linear relationship, the amount of cooling per unit CO2 removed need to decline to net zero to halt global warming. {5.4.8, 5.5.2,        is approximately independent of the rate and amount of removal Box 5.1}                                                                  (medium confidence). {4.6.3, 5.6.2.1, Figure 5.32, Figure 5.34}
TS.3.3.2 Carbon Dioxide Removal                                          Due to non-linearities in the climate system, the century-scale climate-carbon cycle response to a CO2 removal from the atmosphere Deliberate carbon dioxide removal (CDR) from the                      is not always equal and opposite to its response to a simultaneous atmosphere has the potential to compensate for residual                CO2 emission (medium confidence). For CO2 emissions of 100 PgC CO2 emissions to reach net zero CO2 emissions or to                    released from a state in equilibrium with pre-industrial atmospheric generate net negative CO2 emissions. In the same way                  CO2 levels, CMIP6 models simulate that 27+/- 6% (mean +/- 1 standard that part of current anthropogenic net CO2 emissions                  deviation) of emissions remain in the atmosphere 80-100 years are taken up by land and ocean carbon stores, net CO2                  after the emissions, whereas for removals of 100 PgC only 23 removal will be partially counteracted by CO2 release from            +/- 6% of removals remain out of the atmosphere. This asymmetry these stores (very high confidence). Asymmetry in the                  implies that an extra amount of CDR is required to compensate for carbon cycle response to simultaneous CO2 emissions and                a positive emission of a given magnitude to attain the same change removals implies that a larger amount of CO2 would need                in atmospheric CO2. Due to low agreement between models, there 99
 
Technical Summary is low confidence in the sign of the asymmetry of the temperature                                                                                                Carbon dioxide removal methods have a range of side effects that response to CO2 emissions and removals. {4.6.3, 5.6.2.1, Figure 5.35}                                                                                            can either weaken or strengthen the carbon sequestration and cooling potential of these methods and affect the achievement Simulations with ESMs indicate that under scenarios where CO2                                                                                                    of sustainable development goals (high confidence). Biophysical emissions gradually decline, reach net zero and become net negative                                                                                              and biogeochemical side-effects of CDR methods are associated during the 21st century (e.g., SSP1-2.6), land and ocean carbon sinks                                                                                            with changes in surface albedo, the water cycle, emissions of CH4 begin to weaken in response to declining atmospheric CO2                                                                                                        and N2O, ocean acidification and marine ecosystem productivity concentrations, and the land sink eventually turns into a source                                                                                                (high confidence). These side-effects and associated Earth system (Figure TS.19). This sink-to-source transition occurs decades to a few                                                                                          feedbacks can decrease carbon uptake and/or change local and centuries after CO2 emissions become net negative. The ocean                                                                                                    regional climate and in turn limit the CO2 sequestration and cooling remains a sink of CO2 for centuries after emissions become net                                                                                                  potential of specific CDR methods (medium confidence). Deployment negative. Under scenarios with large net negative CO2 emissions                                                                                                  of CDR, particularly on land, can also affect water quality and quantity, (e.g., SSP5-3.4-OS) and rapidly declining CO2 concentrations, the land                                                                                          food production and biodiversity (high confidence). These effects are source is larger than for SSP1-2.6 and the ocean also switches to a                                                                                              often highly dependent on local context, management regime, prior TS  source. While the general response is robust across models, there is                                                                                            land use, and scale (high confidence). The largest co-benefits are low confidence in the timing of the sink-to-source transition and the                                                                                            obtained with methods that seek to restore natural ecosystems or magnitude of the CO2 source in scenarios with net negative CO2                                                                                                  improve soil carbon sequestration (medium confidence). The climate emissions. Carbon dioxide removal could reverse some aspects                                                                                                    and biogeochemical effects of terminating CDR are expected to be climate change if CO2 emissions become net negative, but some                                                                                                    small for most CDR methods (medium confidence). {4.6.3, 5.6.2.2, changes would continue in their current direction for decades to                                                                                                Figure 5.36, 8.4.3, 8.6.3}
millennia. For instance, sea level rise due to ocean thermal expansion would not reverse for several centuries to millennia (high confidence)
(Box TS.4). {4.6.3, 5.4.10, 5.6.2.1, Figure 5.30, Figure 5.33}
Different emission stages in SSP1-2.6 scenario, characterized by:
(a) Large net positive (b) Small net positive                                                    (c) Net negative        (d) Net negative                No          (e) Net zero CO2 emissions          CO2 emissions                                                            CO2 emissions            CO2 emissions              significant    CO2 emissions (emissions exceed              (emissions exceed                                                (removals exceed        (removals exceed              change        and removals removals)                      removals)                                                        emissions)              emissions)                  within this period CO2 in atmosphere Net land flux 550 Net ocean flux Carbon dioxide concentration (ppm) 203 -41 500
                                                                  -61                            33 -23
                                                                                                                                -33 Net CO2 emissions 450
                                                                                                                                                      -30  2 4                                    3      0 and removals 400                          469 446                -7          -5          -7 411                                                          -7 403              400                      396 350  368
                                                      +101 ppm                                    -23 ppm                                              -35 ppm                  -8 ppm                                        -4 ppm 300 2000                  2050                                                                2100                          2150                    2200            2250                      2300 Figure TS.19 l Carbon sink response in a scenario with net carbon dioxide (CO2) removal from the atmosphere. The intent of this figure is to show how atmospheric CO2 evolves under negative emissions and its dependence on the negative emissions technologies. It also shows the evolution of the ocean and land sinks. Shown are CO2 flux components from concentration-driven Earth system model (ESM) simulations during different emissions stages of SSP1-2.6 and its long-term extension. (a) Large net positive CO2 emissions, (b) small net positive CO2 emissions, (c-d) net negative CO2 emissions, and (e) net zero CO2 emissions. Positive flux components act to raise the atmospheric CO2 concentration, whereas negative components act to lower the CO2 concentration. Net CO2 emissions and land and ocean CO2 fluxes represent the multi-model mean and standard deviation (error bar) of four ESMs (CanESM5, UKESM1, CESM2-WACCM, IPSL-CM6a-LR) and one Earth system model of intermediate complexity (Uvic ESCM). Net CO2 emissions are calculated from concentration-driven ESM simulations as the residual from the rate of increase in atmospheric CO2 and land and ocean CO2 fluxes.
Fluxes are accumulated over each 50-year period and converted to concentration units (parts per million, or ppm). {5.6.2.1, Figure 5.33}
100
 
Technical Summary TS.3.3.3 Relating Different Forcing Agents                            future global surface temperature outcomes (high confidence) {7.6.1, Box 7.3}
When including other GHGs, the choice of emissions metric affects the quantification of net zero GHG emissions and            Emissions metrics are needed to aggregate baskets of gases to their resulting temperature outcome (high confidence).              determine net zero GHG emissions. Generally, achieving net zero CO2 Reaching and sustaining net zero GHG emissions typically            emissions and declining non-CO2 radiative forcing would halt human-leads to a peak and decline in temperatures when                    induced warming. Reaching net zero GHG emissions quantified by quantified with the global warming potential over a 100-            GWP-100 typically leads to declining temperatures after net zero year period (GWP-100). Carbon-cycle responses are more              GHGs emissions are achieved if the basket includes short-lived gases, robustly accounted for in emissions metrics compared to            such as CH4. Net zero GHG emissions defined by CGTP or GWP*
AR5 (high confidence). New emissions metric approaches              imply net zero CO2 and other long-lived GHG emissions and constant can be used to generate equivalent cumulative emissions            (CGTP) or gradually declining (GWP*) emissions of short-lived gases.
of CO2 for short-lived greenhouse gases based on their rate        The warming evolution resulting from net zero GHG emissions of emissions. {7.6.2}                                              defined in this way corresponds approximately to reaching net zero CO2 emissions, and would thus not lead to declining temperatures      TS Over 10- to 20-year time scales, the temperature response          after net zero GHG emissions are achieved but to an approximate to a single years worth of current emissions of short-            temperature stabilization (high confidence). The choice of emissions lived climate forcers (SLCFs) is at least as large as that of      metric hence affects the quantification of net zero GHG emissions, CO2, but because the effects of SLCFs decay rapidly over            and therefore the resulting temperature outcome of reaching and the first few decades after emission, the net long-term            sustaining net zero GHG emissions levels (high confidence). {7.6.1.4, temperature response to a single years worth of emissions          7.6.2, 7.6.3}
is predominantly determined by cumulative CO2 emissions.
As pointed out in AR5, ultimately, it is a matter for policymakers Emissions reductions in 2020 associated with COVID-19              to decide which emissions metric is most applicable to their needs.
containment led to small and positive global ERF; however,          This Report does not recommend the use of any specific emissions global and regional climate responses to the forcing are            metric, as the most appropriate metric depends on the policy goal undetectable above internal variability due to the temporary        and context (see Chapter 7, Section 7.6). A detailed assessment of nature of emissions reductions. {6.6, Cross-Chapter Box 6.1}        GHG metrics to support climate change mitigation and associated policy contexts is provided in the WGIII contribution to the AR6.
The relative climate effects of different forcing agents are typically quantified using emissions metrics that compare the effects of        The global surface temperature response following a climate change an idealised pulse of 1 kg of some climate forcing agent against      mitigation measure that affects emissions of both short- and long-a reference climate forcing agent, almost always CO2. The two most    lived climate forcers depends on their lifetimes, their ERFs, how prominent pulse emissions metrics are the global warming potential    fast and for how long the emissions are reduced, and the thermal (GWP) and global temperature change potential (GTP) (see Glossary). inertia in the climate system. Mitigation, relying on emissions The climate responses to CO2 emissions by convention include the      reductions and implemented through new legislation or technology effects of warming on the carbon cycle, so for consistency these also  standards, implies that emissions reductions occur year after year.
need to be determined for non-CO2 emissions. The methodology          Global temperature response to a years worth of current emissions for doing this has been placed on a more robust scientific footing    from different sectors informs about the mitigation potential (Figure compared to AR5 (high confidence). Methane from fossil fuel sources    TS.20). Over 10- to 20-year time scales, the influence of SLCFs is has slightly higher emissions metric values than those from biogenic  at least as large as that of CO2, with sectors producing the largest sources since it leads to additional fossil CO2 in the atmosphere      warming being fossil fuel production and distribution, agriculture, (high confidence). Updates to the chemical adjustments for CH4 and    and waste management. Because the effects of the SLCFs decay N2O emissions (Section TS.3.1) and revisions in their lifetimes result rapidly over the first few decades after emission, the net long-term in emissions metrics for GWP and GTP that are slightly lower than      temperature effect from a single years worth of current emissions is in AR5 (medium confidence). Emissions metrics for the entire suite    predominantly determined by CO2. Fossil fuel combustion for energy, of GHGs assessed in the AR6 have been calculated for various time      industry and land transportation are the largest contributing sectors horizons. {7.6.1, Table 7.15, Table 7.SM.7}                            on a 100-year time scale (high confidence). Current emissions of CO2, N2O and SLCFs from East Asia and North America are the largest New emissions metric approaches, such as GWP* and Combined-            regional contributors to additional net future warming on both GTP (CGTP), relate changes in the emissions rate of short-lived        short (medium confidence) and long time scales (10 and 100 years, greenhouse gases to equivalent cumulative emissions of CO2            respectively) (high confidence). {6.6.1, 6.6.2, Figure 6.16}
(CO2-e). Global surface temperature response from aggregated emissions of short-lived greenhouse gases over time is determined      COVID-19 restrictions led to detectable reductions in global by multiplying these cumulative CO2-e by TCRE (see Section TS.3.2.1). anthropogenic emissions of nitrogen oxides (NOx) (about 35% in When GHGs are aggregated using standard metrics such as GWP or        April 2020) and fossil CO2 (7%, with estimates ranging from 5.8% to GTP, cumulative CO2-e emissions are not necessarily proportional to    13.0%), driven largely by reduced emissions from the transportation 101
 
Technical Summary Effect of a one year pulse of present-day emissions on global surface temperature Response after 10 years Response after 100 years                                                                                                          Net effect, 5% to 95% range 0.06        0.05        0.04        0.03      0.02      0.01        0.00        0.01        0.02        0.03          0.04        0.05        0.06        0.07 Change in global surface temperature due to total anthropogenic emissions (&deg;C)
By sector, response after 10 years                                  By sector, response after 100 years Fossil fuel production                                                                                        CO2 and distribution                                                                                              CH4 Agriculture                                                                                                    N2O BC Waste management                                                                                              OC TS                                                                                                                SO2 Residential and commercial                                                                                    NOx CO Fossil fuel combustion                                                                                        VOC for energy                                                                                                    NH3 Residential and commercial                                                                                    Aviacontrail (biofuel use only)                                                                                            AviastratH2O HFCs Land transportation Open biomass burning Industry Shipping Aviation 0.010 0.005                0      0.005      0.010        0.015    0.010      0.005          0        0.005        0.010      0.015 Change in global surface temperature (&deg;C)                                Change in global surface temperature (&deg;C)
Figure TS.20 l Global surface temperature change 10 and 100 years after a one-year pulse of present-day emissions. The intent of this figure is to show the sectoral contribution to present-day climate change by specific climate forcers, including carbon dioxide (CO2) as well as short-lived climate forcers (SLCFs). The temperature response is broken down by individual species and shown for total anthropogenic emissions (top), and sectoral emissions on 10-year (left) and 100-year time scales (right).
Sectors are sorted by (high-to-low) net temperature effect on the 10-year time scale. Error bars in the top panel show the 5-95% range in net temperature effect due to uncertainty in radiative forcing only (calculated using a Monte Carlo approach and best estimate uncertainties from the literature). Emissions for 2014 are from the Coupled Model Intercomparison Project Phase 6 (CMIP6) emissions dataset, except for hydrofluorocarbons (HFCs) and aviation H2O, which rely on other datasets (see Section 6.6.2 for more details). CO2 emissions are excluded from open biomass burning and residential biofuel use. {6.6.2, Figure 6.16}
sector (medium confidence). There is high confidence that, with                          temporarily adding to the total anthropogenic climate influence, the exception of surface ozone, reductions in pollutant precursors                        with positive forcing (warming influence) from aerosol changes contributed to temporarily improved air quality in most regions of                        dominating over negative forcings (cooling influence) from CO2, NOx the world. However, these reductions were lower than what would                          and contrail cirrus changes. Consistent with this small net radiative be expected from sustained implementation of policies addressing                          forcing, and against a large component of internal variability, Earth air quality and climate change (medium confidence). Overall, the net                      system models show no detectable effect on global or regional global ERF from COVID-19 containment was likely small and positive                        surface temperature or precipitation (high confidence). {Cross for 2020 (with a temporary peak value less than 0.2 W m-2), thus                          Chapter Box 6.1}
102
 
Technical Summary Box TS.7 l Climate and Air Quality Responses to Short-lived Climate Forcers in Shared Socio-economic Pathways Future changes in emissions of short-lived climate forcers (SLCFs) are expected to cause an additional global mean warming, with a large diversity in the end-of-century response across the WGI core set of Shared Socio-economic Pathways (SSPs), depending upon the level of climate change and air pollution mitigation (Box TS.7, Figure 1). This additional warming is either due to reductions in cooling aerosols for air pollution regulation or due to increases in methane (CH4), ozone and hydrofluorocarbons (HFCs). This additional warming is stable after 2040 in SSPs associated with lower global air pollution as long as CH4 emissions are also mitigated, but the overall warming induced by SLCF changes is higher in scenarios in which air quality continues to deteriorate (induced by growing fossil fuel use and limited air pollution control) (high confidence).
Sustained CH4 mitigation reduces global surface ozone, contributing to air quality improvements, and also reduces surface temperature in the longer term, but only sustained CO2 emissions reductions allow long-term climate                            TS stabilization (high confidence). Future changes in air quality (near-surface ozone and particulate matter, or PM) at global and local scales are predominantly driven by changes in ozone and aerosol precursor emissions rather than climate (high confidence). Air quality improvements driven by rapid decarbonization strategies, as in SSP1-1.9 and SSP1-2.6, are not sufficient in the near term to achieve air quality guidelines set by the World Health Organization in some highly polluted regions (high confidence). Additional policies (e.g., access to clean energy, waste management) envisaged to attain United Nations Sustainable Development Goals bring complementary SLCF reduction. {4.4.4, 6.6.3, 6.7.3, Box 6.2}
The net effect of SLCF emissions changes on temperature will depend on how emissions of warming and cooling SLCFs will evolve in the future. The magnitude of the cooling effect of aerosols remains the largest uncertainty in the effect of SLCFs in future climate projections. Since the SLCFs have undergone large changes over the past two decades, the temperature and air pollution responses are estimated relative to the year 2019 instead of 1995-2014.
Temperature Response In the next two decades, it is very likely that SLCF emissions changes will cause a warming relative to 2019, across the WGI core set of SSPs (see Section TS.1.3.1), in addition to the warming from long-lived GHGs. The net effect of SLCF and HFC changes in global surface temperature across the SSPs is a likely warming of 0.06&deg;C-0.35&deg;C in 2040 relative to 2019. This near-term global mean warming linked to SLCFs is quite similar in magnitude across the SSPs due to competing effects of warming (CH4, ozone) and cooling (aerosols) forcers (Box TS.7, Figure 1). There is greater diversity in the end-of-century response among the scenarios. SLCF changes in scenarios with no climate change mitigation (SSP3-7.0 and SSP5-8.5) will cause a warming in the likely range of 0.4&deg;C-0.9&deg;C in 2100 relative to 2019 due to increases in CH4, tropospheric ozone and HFC levels. For the stringent climate change and pollution mitigation scenarios (SSP1-1.9 and SSP1-2.6), the cooling from reductions in CH4, ozone and HFCs partially balances the warming from reduced aerosols, primarily sulphate, and the overall SLCF effect is a likely increase in global surface temperature of 0.0&deg;C-0.3&deg;C in 2100, relative to 2019. With intermediate climate change and air pollution mitigations, SLCFs in SSP2-4.5 add a likely warming of 0.2&deg;C-0.5&deg;C to global surface temperature change in 2100, with the largest warming resulting from reductions in aerosols. {4.4.4, 6.7.3}
Assuming implementation and efficient enforcement of both the Kigali Amendment to the Montreal Protocol on Substances that Deplete the Ozone Layer and current national plans result in limiting emissions (as in SSP1-2.6), the effects of HFCs on global surface temperature, relative to 2019, would remain below +0. 02&deg;C from 2050 onwards versus about +0.04&deg;C-0.08&deg;C in 2050 and
+0.1&deg;C-0.3&deg;C in 2100 considering only national HFC regulations decided prior to the Kigali Amendment (as in SSP5-8.5) (medium confidence). {6.6.3, 6.7.3}
Air Quality Responses Air pollution projections range from strong reductions in global surface ozone and PM (e.g., SSP1-2.6, with stringent mitigation of both air pollution and climate change) to no improvement and even degradation (e.g., SSP3-7.0 without climate change mitigation and with only weak air pollution control) (high confidence). Under the SSP3-7.0 scenario, PM levels are projected to increase until 2050 over large parts of Asia, and surface ozone pollution is projected to worsen over all continental areas through 2100 (high confidence).
In SSP5-8.5, a scenario without climate change mitigation but with stringent air pollution control, PM levels decline through 2100, but high CH4 levels hamper the decline in global surface ozone at least until 2080 (high confidence). {6.7.1}
103
 
Technical Summary Box TS.7 (continued)
TS Box TS.7, Figure 1 l Effects of short-lived climate forcers (SLCFs) on global surface temperature and air pollution across the WGI core set of Shared Socio-economic Pathways (SSPs). The intent of this figure is to show the climate and air quality (surface ozone and particulate matter smaller than 2.5 microns in diameter, or PM2.5 ) response to SLCFs in the SSP scenarios for the near and long-term. Effects of net aerosols, tropospheric ozone, hydrofluorocarbons (HFCs; with lifetimes less than 50 years), and methane (CH4) are compared with those of total anthropogenic forcing for 2040 and 2100 relative to year 2019. The global surface temperature changes are based on historical and future evolution of effective radiative forcing (ERF) as assessed in Chapter 7 of this Report. The temperature responses to the ERFs are calculated with a common impulse response function (RT) for the climate response, consistent with the metric calculations in Chapter 7 (Box 7.1). The RT has an equilibrium climate sensitivity of 3.0&deg;C for a doubling of atmospheric CO2 concentration (feedback parameter of -1.31 W m-2 &deg;C-1).
The scenario total (grey bar) includes all anthropogenic forcings (long- and short-lived climate forcers, and land-use changes). Uncertainties are 5-95% ranges. The global changes in air pollutant concentrations (ozone and PM2.5) are based on multimodel Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations and represent changes in five-year mean surface continental concentrations for 2040 and 2098 relative to 2019. Uncertainty bars represent inter-model +/-1 standard deviation. {6.7.2, 6.7.3, Figure 6.24}
Box TS.8 l Earth System Response to Solar Radiation Modification Since AR5, further modelling work has been conducted on aerosol-based solar radiation modification (SRM) options such as stratospheric aerosol injection, marine cloud brightening, and cirrus cloud thinning21 and their climate and biogeochemical effects. These investigations have consistently shown that SRM could offset some of the effects of increasing greenhouse gases on global and regional climate, including the carbon and water cycles (high confidence).
However, there would be substantial residual or overcompensating climate change at the regional scales and seasonal time scales (high confidence), and large uncertainties associated with aerosol-cloud-radiation interactions persist.
The cooling caused by SRM would increase the global land and ocean CO2 sinks (medium confidence), but this would not stop CO2 from increasing in the atmosphere or affect the resulting ocean acidification under continued anthropogenic emissions (high confidence). It is likely that abrupt water cycle changes will occur if SRM techniques are implemented rapidly. A sudden and sustained termination of SRM in a high CO2 emissions scenario would cause rapid climate change (high confidence). However, a gradual phase-out of SRM combined with emissions reduction and carbon dioxide removal (CDR) would avoid these termination effects (medium confidence). {4.6.3, 5.6.3. 6.4.6, 8.6.3}.
21 Although cirrus cloud thinning aims to cool the planet by increasing longwave emissions to space, it is included in the portfolio of SRM options for consistency with AR5 and SR1.5. {4.6.3.3}
104
 
Technical Summary Box TS.8 (continued)
Solar radiation modification (SRM) refers to deliberate, large-scale climate intervention options that are studied as potential supplements to deep mitigation, for example, in scenarios that overshoot climate stabilization goals. SRM options aim to offset some of the warming effects of GHG emissions by modification of Earths shortwave radiation budget. Following SR1.5, the SRM assessed in this Report also includes some options, such as cirrus cloud thinning, that alter the longwave radiation budget.
SRM contrasts with climate change mitigation activities, such as emissions reductions and CDR, as it introduces a mask to the climate change problem by altering Earths radiation budget, rather than attempting to address the root cause of the problem, which is the increase in GHGs in the atmosphere. By masking only the climate effects of GHG emissions, SRM does not address other issues related to atmospheric CO2 increase, such as ocean acidification. This Report assesses physical understanding of the Earth system response to proposed SRM, and the assessment is based primarily on idealized climate model simulations. There are other important considerations, such as risk to human and natural systems, perceptions, ethics, cost, governance, and trans-boundary issues and their relationship to the United Nations Sustainable Development Goals - issues that the WGII (Chapter 16) and WGIII (Chapter 14) Reports            TS address. {4.6.3}
SRM options include those that increase surface albedo, brighten marine clouds by increasing the amount of cloud condensation nuclei, or reduce the optical depth of cirrus clouds by seeding them with ice nucleating particles. However, the most commonly studied approaches attempt to mimic the cooling effects of major volcanic eruptions by injecting reflective aerosols (e.g., sulphate aerosols) or their precursors (e.g., sulphur dioxide) into the stratosphere. {4.6.3, 5.6.3, 6.4.6}
SRM could offset some effects of greenhouse gas-induced warming on global and regional climate, but there would be substantial residual and overcompensating climate change at the regional scale and seasonal time scales (high confidence). Since AR5, more modelling work has been conducted with more sophisticated treatment of aerosol-based SRM approaches, but the uncertainties in cloud-aerosol-radiation interactions are still large (high confidence). Modelling studies suggest that it is possible to stabilize multiple large-scale temperature indicators simultaneously by tailoring the deployment strategy of SRM options (medium confidence) but with large residual or overcompensating regional and seasonal climate changes. {4.6.3}
SRM approaches targeting shortwave radiation are likely to reduce global mean precipitation, relative to future CO2 emissions scenarios, if all global mean warming is offset. In contrast, cirrus cloud thinning, targeting longwave radiation, is expected to cause an increase in global mean precipitation (medium confidence). If shortwave approaches are used to offset global mean warming, the magnitude of reduction in regional precipitation minus evapotranspiration (P-E) (Box TS.5), which is more relevant to freshwater availability, is smaller than precipitation decrease because of simultaneous reductions in both precipitation and evapotranspiration (medium confidence). {4.6.3, 8.2.1, 8.6.3}.
If SRM is used to cool the planet, it would cause a reduction in plant and soil respiration and slow the reduction of ocean carbon uptake due to warming (medium confidence). The result would be an enhancement of the global land and ocean CO2 sinks (medium confidence) and a slight reduction in atmospheric CO2 concentration relative to unmitigated climate change. However, SRM would not stop CO2 from increasing in the atmosphere or affect the resulting ocean acidification under continued anthropogenic emissions (high confidence). {5.6.3}
The effect of stratospheric aerosol injection on global temperature and precipitation is projected by models to be detectable after one to two decades, which is similar to the time scale for the emergence of the benefits of emissions reductions. A sudden and sustained termination of SRM in a high GHG emissions scenario would cause rapid climate change and a reversal of the SRM effects on the carbon sinks (high confidence). It is also likely that a termination of strong SRM would drive abrupt changes in the water cycle globally and regionally, especially in the tropical regions by shifting the Inter-tropical Convergence Zone and Hadley cells. At the regional scale, non-linear responses cannot be excluded, due to changes in evapotranspiration. However, a gradual phase-out of SRM combined with emissions reductions and CDR would avoid larger rates of changes (medium confidence). {4.6.3, 5.6.3, 8.6.3}.
105
 
Technical Summary Box TS.9 l Irreversibility, Tipping Points and Abrupt Changes The present rates of response of many aspects of the climate system are proportionate to the rate of recent temperature change, but some aspects may respond disproportionately. Some climate system components are slow to respond, such as the deep ocean overturning circulation and the ice sheets (Box TS.4). It is virtually certain that irreversible, committed change is already underway for the slow-to-respond processes as they come into adjustment for past and present emissions.
The paleoclimate record indicates that tipping elements exist in the climate system where processes undergo sudden shifts toward a different sensitivity to forcing, such as during a major deglaciation, where 1&deg;C degree of temperature change might correspond to a large or small ice-sheet mass loss during different stages (Box TS.2). For global climate indicators, evidence for abrupt change is limited, but deep ocean warming, acidification and sea level rise are committed to ongoing change for millennia after global surface temperatures initially stabilize and are irreversible on human time TS      scales (very high confidence). At the regional scale, abrupt responses, tipping points and even reversals in the direction of change cannot be excluded (high confidence). Some regional abrupt changes and tipping points could have severe local impacts, such as unprecedented weather, extreme temperatures and increased frequency of droughts and forest fires.
Models that exhibit such tipping points are characterized by abrupt changes once the threshold is crossed, and even a return to pre-threshold surface temperatures or to atmospheric carbon dioxide concentrations does not guarantee that the tipping elements return to their pre-threshold state. Monitoring and early warning systems are being put into place to observe tipping elements in the climate system. {1.3, 1.4.4, 1.5, 4.3.2, Table 4.10, 5.3.4, 5.4.9, 7.5.3, 9.2.2, 9.2.4, 9.4.1, 9.4.2, 9.6.3, Cross-chapter Box 12.1}
Understanding of multi-decadal reversibility (i.e., the system returns to the previous climate state within multiple decades after the radiative forcing is removed) has improved since AR5 for many atmospheric, land surface and sea ice climate metrics following sea surface temperature recovery. Some processes suspected of having tipping points, such as the Atlantic Meridional Overturning Circulation (AMOC), have been found to often undergo recovery after temperature stabilization with a time delay (low confidence).
However, substantial irreversibility is further substantiated for some cryosphere changes, ocean warming, sea level rise, and ocean acidification. {4.7.2, 5.3.3, 5.4.9, 9.2.2, 9.2.4, 9.4.1, 9.4.2, 9.6.3}
Some climate system components are slow to respond, such as the deep ocean overturning circulation and the ice sheets. It is likely that under stabilization of global warming at 1.5&deg;C, 2.0&deg;C or 3.0&deg;C relative to 1850-1900, the AMOC will continue to weaken for several decades by about 15%, 20% and 30% of its strength and then recover to pre-decline values over several centuries (medium confidence).
At sustained warming levels between 2&deg;C and 3&deg;C, there is limited evidence that the Greenland and West Antarctic ice sheets will be lost almost completely and irreversibly over multiple millennia; both the probability of their complete loss and the rate of mass loss increases with higher surface temperatures (high confidence). At sustained warming levels between 3&deg;C and 5&deg;C, near-complete loss of the Greenland Ice Sheet and complete loss of the West Antarctic Ice Sheet is projected to occur irreversibly over multiple millennia (medium confidence); with substantial parts or all of Wilkes Subglacial Basin in East Antarctica lost over multiple millennia (low confidence). Early-warning signals of accelerated sea level rise from Antarctica could possibly be observed within the next few decades. For other hazards (e.g., ice-sheet behaviour, glacier mass loss and global mean sea level change, coastal floods, coastal erosion, air pollution, and ocean acidification) the time and/or scenario dimensions remain critical, and a simple and robust relationship with global warming level cannot be established (high confidence). {4.3.2, 4.7.2, 5.4.3, 5.4.5, 5.4.8, 8.6, 9.2, 9.4, Box 9.3, Cross-Chapter Box 12.1}
For global climate indicators, evidence for abrupt change is limited. For global warming up to 2&deg;C above 1850-1900 levels, paleoclimate records do not indicate abrupt changes in the carbon cycle (low confidence). Despite the wide range of model responses, uncertainty in atmospheric CO2 by 2100 is dominated by future anthropogenic emissions rather than uncertainties related to carbon-climate feedbacks (high confidence). There is no evidence of abrupt change in climate projections of global temperature for the next century:
there is a near-linear relationship between cumulative CO2 emissions and maximum global mean surface air temperature increase caused by CO2 over the course of this century for global warming levels up to at least 2&deg;C relative to 1850-1900. The increase in global ocean heat content (Section TS.2.4) will likely continue until at least 2300 even for low emissions scenarios, and global mean sea level will continue to rise for centuries to millennia following cessation of emissions (Box TS.4) due to continuing deep ocean heat uptake and mass loss of the Greenland and Antarctic ice sheets (high confidence). {2.2.3; Cross-Chapter Box 2.1; 5.1.1; 5.4; Cross-Chapter Box 5.1; Figures 5.3, 5.4, 5.25, and 5.26; 9.2.2; 9.2.4}
The response of biogeochemical cycles to anthropogenic perturbations can be abrupt at regional scales and irreversible on decadal to century time scales (high confidence). The probability of crossing uncertain regional thresholds increases with climate change (high 106
 
Technical Summary Box TS.9 (continued) confidence). It is very unlikely that gas clathrates (mostly methane) in deeper terrestrial permafrost and subsea clathrates will lead to a detectable departure from the emissions trajectory during this century. Possible abrupt changes and tipping points in biogeochemical cycles lead to additional uncertainty in 21st century atmospheric GHG concentrations, but future anthropogenic emissions remain the dominant uncertainty (high confidence). There is potential for abrupt water cycle changes in some high emissions scenarios, but there is no overall consistency regarding the magnitude and timing of such changes. Positive land surface feedbacks, including vegetation, dust, and snow, can contribute to abrupt changes in aridity, but there is only low confidence that such changes will occur during the 21st century. Continued Amazon deforestation, combined with a warming climate, raises the probability that this ecosystem will cross a tipping point into a dry state during the 21st century (low confidence). (Section TS.3.2.2) {5.4.3, 5.4.5, 5.4.8, 5.4.9, 8.6.2, 8.6.3, Cross-Chapter Box 12.1}
TS TS.4        Regional Climate Change                                            requirements, and demand (very high confidence). The decision-making context, level of user engagement, This section focuses on how to generate regional climate change                and co-production between scientists, practitioners and information and its relevance for climate services; the drivers of regional    users are important determinants of the type of climate climate variability and change and how they are being affected by              service developed and its utility in supporting adaptation, anthropogenic factors; and observed, attributed and projected changes          mitigation and risk management decisions. {10.3, 10.6, in climate, including extreme events and climatic impact-drivers (CIDs),        Cross-Chapter Box 10.3, 12.6, Cross-Chapter Box 12.2}
across all regions of the world. There is a small set of CID changes common to all land or ocean regions and a specific set of changes from a broader range of CIDs seen in each region. This regional diversity        TS.4.1.1 Sources and Methodologies for Generating Regional results from regional climate being determined by a complex interplay                    Climate Information between the seasonal-to-multi-decadal variation of large-scale modes of climate variability, external natural and anthropogenic forcings, local  Climate change information at regional scale is generated using a range climate processes and related feedbacks.                                    of data sources and methodologies (Section TS.1.4). Understanding of observed regional climate change and variability is based on the availability and analysis of multiple observational datasets that are TS.4.1      Generation and Communication of Regional                        suitable for evaluating the phenomena of interest (e.g., extreme Climate Change Information                                      events), including accounting for observational uncertainty (Section TS.1.2.1). These datasets are combined with climate model simulations Climate change information at regional scale is generated                of observed changes and events to attribute causes of those changes using a range of data sources and methodologies. Multi-                  and events to large- and regional-scale anthropogenic and natural model ensembles and models with a range of resolutions                    drivers and to assess the performance of the models. Future simulations are important data sources, and discarding models that                    with many climate models (multi-model ensembles) are then used to fundamentally misrepresent relevant processes improves                    generate and quantify ranges of projected regional climate responses the credibility of ensemble information related to these                  (Section TS.4.2). Discarding models that fundamentally misrepresent processes. A key methodology is distillation - combining                  relevant processes improves the credibility of regional climate lines of evidence and accounting for stakeholder context                  information generated from these ensembles (high confidence).
and values - which helps ensure the information is relevant,              However, multi-model mean and ensemble spread are not a full useful and trusted for decision-making (see Core Concepts                measure of the range of projection uncertainty and are not sufficient Box) (high confidence).                                                  to characterize low-likelihood, high-impact changes (Box TS.3) or situations where different models simulate substantially different Since AR5, physical climate storylines have emerged as                    or even opposite changes (high confidence). Large single-model a complementary approach to ensemble projections for                      ensembles are now available and provide a more comprehensive generating more accessible climate information and                        spectrum of possible changes associated with internal variability (high promoting a more comprehensive treatment of risk. They                    confidence) (Section TS.1.2.3). {1.5.1, 1.5.4, 10.2, 10.3.3, 10.3.4, 10.4.1, have been used as part of the distillation process within                10.6.2, 11.2, Box 11.2, Cross-Chapter Box 11.1, 12.4, Atlas.1.4.1}
climate services to generate the required context-relevant, credible and trusted climate information.                                Depending on the region of interest, representing regionally important forcings (e.g., aerosols, land-use change and ozone concentrations)
Since AR5, climate change information produced for climate                and feedbacks (e.g., between snow and albedo, soil moisture and services has increased significantly due to scientific and                temperature, or soil moisture and precipitation) in climate models is a technological advancements and growing user awareness,                    prerequisite for them to reproduce past regional trends to underpin the 107
 
Technical Summary reliability of future projections (medium confidence) (Section TS.1.2.2). Methodologies such as statistical downscaling, bias adjustment and In some cases, even the sign of a projected change in regional climate      weather generators are beneficial as an interface between climate cannot be trusted if relevant regional processes are not represented, for    model projections and impact modelling and for deriving user-example, for variables such as precipitation and wind speed (medium          relevant indicators (high confidence). However, the performance of confidence). In some regions, either geographical (e.g., Central Africa,    these techniques depends on that of the driving climate model: in Antarctica) or typological (e.g., mountainous areas, Small Islands and      particular, bias adjustment cannot overcome all consequences of cities), and for certain phenomena, fewer observational records are          unresolved or strongly misrepresented physical processes, such as available or accessible, which limits the assessment of regional climate    large-scale circulation biases or local feedbacks (medium confidence).
change in these cases. {1.5.1, 1.5.3, 1.5.4, 8.5.1, 10.2, 10.3.3, 10.4.1,    {10.3.3, Cross-Chapter Box 10.2, 12.2, Atlas.2.2}
11.1.6, 11.2, 12.4, Atlas.8.3, Atlas.11.1.5, Cross-Chapter Box Atlas.2}
Box TS.10 l Event Attribution TS The attribution of observed changes in extremes to human influence (including greenhouse gas and aerosol emissions and land-use changes) has substantially advanced since AR5, in particular for extreme precipitation, droughts, tropical cyclones, and compound extremes (high confidence). There is limited evidence for windstorms and convective storms. Some recent hot extreme events would have been extremely unlikely to occur without human influence on the climate system. (Section TS.1) {Cross-Working Group Box: Attribution in Chapter 1, 11.2, 11.3, 11.4, 11.6, 11.7, 11.8}
Since AR5, the attribution of extreme weather events has emerged as a growing field of climate research with an increasing body of literature. It provides evidence that greenhouse gases and other external forcings have affected individual extreme weather events by disentangling anthropogenic drivers from natural variability. Event attribution is now an important line of evidence for assessing changes in extremes on regional scales. (Section TS.1) {Cross-Working Group Box: Attribution, 11.1.4}
The regional extremes and events that have been studied are geographically uneven (Section TS.4.1). A few events, for example, extreme rainfall events in the United Kingdom, heatwaves in Australia, or Hurricane Harvey that hit Texas in 2017, have been heavily studied. Many highly impactful extreme weather events have not been studied in the event attribution framework, particularly in the developing world where studies are generally lacking. This is due to various reasons, including lack of observational data, lack of reliable climate models, and lack of scientific capacity. While the events that have been studied are not representative of all extreme events that have occurred, and results from these studies may also be subject to selection bias, the large number of event attribution studies provide evidence that changes in the properties of these local and individual events are in line with expected consequences of human influence on the climate and can be attributed to external drivers. {Cross-Working Group Box: Attribution, 11.1.4, 11.2.2}
It is very likely that human influence is the main contributor to the observed increase in the intensity and frequency of hot extremes and the observed decrease in the intensity and frequency of cold extremes on continental scales. Some specific recent hot extreme events would have been extremely unlikely to occur without human influence on the climate system. Changes in aerosol concentrations have likely slowed the increase in hot extremes in some regions, in particular from 1950-1980. No-till farming, irrigation and crop expansion have similarly attenuated increases in summer hot extremes in some regions, such as central North America (medium confidence). {11.3.4}
Human influence has contributed to the intensification of heavy precipitation in three continents where observational data are most abundant: North America, Europe and Asia (high confidence). On regional scales, evidence of human influence on extreme precipitation is limited, but new evidence from attributing individual heavy precipitation events found that human influence was a significant driver of the events. {11.4.4}
There is low confidence that human influence has affected trends in meteorological droughts in most regions, but medium confidence that they have contributed to the severity of some specific events. There is medium confidence that human-induced climate change has contributed to increasing trends in the probability or intensity of recent agricultural and ecological droughts, leading to an increase of the affected land area. {11.6.4}
Event attribution studies of specific strong tropical cyclones provide limited evidence for anthropogenic effects on tropical cyclone intensifications so far, but high confidence for increases in precipitation. There is high confidence that anthropogenic climate change contributed to extreme rainfall amounts during Hurricane Harvey (in 2017) and other intense tropical cyclones. {11.7.3}
108
 
Technical Summary Box TS.10 (continued)
The number of evident attribution studies on compound events is limited. There is medium confidence that weather conditions that promote wildfires have become more probable in southern Europe, northern Eurasia, the USA, and Australia over the last century. In Australia a number of event attribution studies show that there is medium confidence of increase in fire weather conditions due to human influence. {11.8.3, 12.4.3.2}
Climate change is already aecting every inhabited region across the globe, with human in"uence contributing to many observed changes in weather and climate extremes (a) Synthesis of assessment of observed change in hot extremes and confidence in human contribution to the observed changes in the worlds regions Type of observed change in hot extremes                                      North America                      GIC          Europe Increase (41)                                            NWN NEN                                          NEU                    RAR WNA    CNA    ENA                              WCE    EEU    WSB      ESB    RFE        Asia                                  TS Decrease (0)
NCA                                            MED    WCA      ECA    TIB    EAS Low agreement in the type of change (2)                                        Small Islands Central      SCA    CAR                              SAH    ARP            SAS            SEA Limited data and/or literature (2)              America                                                                                                    PAC NWS        NSA            WAF    CAF  NEAF NAU Con"dence in human contribution                                                                                                                                    Small to the observed change                                                            SAM      NES          WSAF SEAF        MDG                                  Islands CAU    EAU High South        SWS      SES                  ESAF Medium                                                      America                          Africa Australasia SAU    NZ Low due to limited agreement                                                  SSA Low due to limited evidence Type of observed change since the 1950s (b) Synthesis of assessment of observed change in heavy precipitation and confidence in human contribution to the observed changes in the worlds regions Type of observed change in heavy precipitation                              North America                      GIC          Europe NWN NEN                                          NEU                    RAR Increase (19)
WNA    CNA    ENA                              WCE    EEU    WSB      ESB    RFE        Asia Decrease (0)
NCA                                            MED    WCA      ECA    TIB    EAS Small Low agreement in the type of change (8)                                        Islands Central      SCA    CAR                              SAH    ARP            SAS            SEA Limited data and/or literature (18)            America                                                                                                    PAC NWS        NSA            WAF    CAF  NEAF NAU Con"dence in human contribution                                                                                                                                    Small SAM      NES          WSAF SEAF        MDG                                  Islands to the observed change                                                                                                                              CAU    EAU High                                                        South        SWS      SES                  ESAF Africa                                              SAU Medium                                                      America Australasia            NZ Low due to limited agreement                                                  SSA Low due to limited evidence Type of observed change since the 1950s (c) Synthesis of assessment of observed change in agricultural and ecological drought and confidence in human contribution to the observed changes in the worlds regions Type of observed change in agricultural and ecological drought              North America                      GIC          Europe Increase (12)                                            NWN NEN                                          NEU                    RAR WNA    CNA    ENA                              WCE    EEU    WSB      ESB    RFE        Asia Decrease (1)
NCA                                            MED    WCA      ECA    TIB    EAS Low agreement in the type of change (28)                                      Small Islands Central      SCA    CAR                              SAH    ARP            SAS            SEA Limited data and/or literature (4)              America                                                                                                    PAC NWS        NSA            WAF    CAF  NEAF NAU Con"dence in human contribution                                                                                                                                    Small to the observed change                                                            SAM      NES          WSAF SEAF        MDG                                  Islands CAU    EAU High South        SWS      SES                  ESAF Medium                                                      America                          Africa                                              SAU    NZ Australasia Low due to limited agreement                                                  SSA Low due to limited evidence Type of observed change since the 1950s Each hexagon corresponds        IPCC AR6 WGI reference regions: North America: NWN (North-Western North America, NEN (North-Eastern North America), WNA to one of the IPCC AR6          (Western North America), CNA (Central North America), ENA (Eastern North America), Central America: NCA (Northern Central America),
WGI reference regions            SCA (Southern Central America), CAR (Caribbean), South America: NWS (North-Western South America), NSA (Northern South America), NES (North-Eastern South America), SAM (South American Monsoon), SWS (South-Western South America), SES (South-Eastern South America),
North-Western          SSA (Southern South America), Europe: GIC (Greenland/Iceland), NEU (Northern Europe), WCE (Western and Central Europe), EEU (Eastern NWN                            Europe), MED (Mediterranean), Africa: MED (Mediterranean), SAH (Sahara), WAF (Western Africa), CAF (Central Africa), NEAF (North Eastern North America Africa), SEAF (South Eastern Africa), WSAF (West Southern Africa), ESAF (East Southern Africa), MDG (Madagascar), Asia: RAR (Russian Arctic), WSB (West Siberia), ESB (East Siberia), RFE (Russian Far East), WCA (West Central Asia), ECA (East Central Asia), TIB (Tibetan Plateau),
EAS (East Asia), ARP (Arabian Peninsula), SAS (South Asia), SEA (South East Asia), Australasia: NAU (Northern Australia), CAU (Central Australia), EAU (Eastern Australia), SAU (Southern Australia), NZ (New Zealand), Small Islands: CAR (Caribbean), PAC (Paci"c Small Islands)
Box TS.10, Figure 1 l Synthesis of assessed observed and attributable regional changes.
109
 
Technical Summary Box TS.10 (continued)
Box TS.10, Figure 1 (continued): The IPCC AR6 WGI inhabited regions are displayed as hexagons of identical sizes in their approximate geographical location (see legend for regional acronyms). All assessments are made for each region as a whole and for the 1950s to the present. Assessments made on different time scales or more local spatial scales might differ from what is shown in the figure. The colours in each panel represent the four outcomes of the assessment on observed changes.
Striped hexagons (white and light-grey) are used where there is low agreement in the type of change for the region as a whole, and grey hexagons are used when there is limited data and/or literature that prevents an assessment of the region as a whole. Other colours indicate at least medium confidence in the observed change.
The confidence level for the human influence on these observed changes is based on assessing trend detection and attribution and event attribution literature, and it is indicated by the number of dots: three dots for high confidence, two dots for medium confidence and one dot for low confidence (single, filled dot: limited agreement; single, empty dot: limited evidence).
Panel (a) For hot extremes, the evidence is mostly drawn from changes in metrics based on daily maximum temperatures; regional studies using other indices (heatwave duration, frequency and intensity) are used in addition. Red hexagons indicate regions where there is at least medium confidence in an observed increase in hot extremes.
Panel (b) For heavy precipitation, the evidence is mostly drawn from changes in indices based on one-day or five-day precipitation amounts using global and regional studies. Green hexagons indicate regions where there is at least medium confidence in an observed increase in heavy precipitation.
Panel (c) Agricultural and ecological droughts are assessed based on observed and simulated changes in total column soil moisture, complemented by evidence TS      on changes in surface soil moisture, water balance (precipitation minus evapotranspiration) and indices driven by precipitation and atmospheric evaporative demand.
Yellow hexagons indicate regions where there is at least medium confidence in an observed increase in this type of drought and green hexagons indicate regions where there is at least medium confidence in an observed decrease in agricultural and ecological drought.
For all regions, Table TS.5 shows a broader range of observed changes besides the ones shown in this figure. Note that Southern South America (SSA) is the only region that does not display observed changes in the metrics shown in this figure, but is affected by observed increases in mean temperature, decreases in frost and increases in marine heatwaves.
(Table TS.5) {11.9, Atlas 1.3.3, Figure Atlas.2}
TS.4.1.2 Regional Climate Information Distillation and                                      changes in a key mode of variability (the Southern Annular Mode)
Climate Services                                                            and drought in Cape Town among different observation periods and in model simulations. {10.5.3, 10.6, 10.6.2, 10.6.4, Cross-Chapter The construction of regional climate information involves people with                      Box 10.3, 12.4}
a variety of backgrounds, from various disciplines, who have different sets of experiences, capabilities and values. The process of synthesizing                  Since AR5, physical climate storyline approaches have emerged as climate information from different lines of evidence from a number                          a complementary instrument to provide a different perspective, or of sources, taking into account the context of a user vulnerable to                        additional climate information, to facilitate communication of the climate variability and change and the values of all relevant actors, is                    information or provide a more flexible consideration of risk. Storylines called distillation. Distillation is conditioned by the sources available,                  that condition climatic events and processes on a set of plausible the actors involved, and the context, which all depend heavily on the                      but distinct large-scale climatic changes enable the exploration of regions considered, and is framed by the question being addressed.                          uncertainties in regional climate projections. For example, they can Distilling regional climate information from multiple lines of evidence                    explicitly address low-likelihood, high-impact outcomes, which would and taking the user context into account increases fitness, usefulness,                    be less emphasized in a probabilistic approach, and can be embedded relevance and trust in that information for use in climate services                        in a users risk landscape, taking account of socio-economic factors (Box TS.11) and decision-making (high confidence). {1.2.3, 10.1.4,                          as well as physical climate changes. Storylines can also be used to 10.5, Cross-Chapter Box 10.3, 12.6}                                                        communicate climate information by narrative elements describing and contextualizing the main climatological features and the relevant The distillation process can vary substantially, as it needs to consider                    consequences in the user context and, as such, can be used as part multiple lines of evidence on all physically plausible outcomes                            of a climate information distillation process. {1.4.4., Box 10.2, 11.2, (especially when they are contrasting) relevant to a specific decision                      Box 11.2, Cross-Chapter Box 12.2}
required in response to a changing climate. Confidence in the distilled regional climate information is enhanced when there is agreement across multiple lines of evidence, so the outcome can be limited if these are inconsistent or contradictory. For example, in the Mediterranean region the agreement between different lines of evidence, such as observations, projections by regional and global models, and understanding of the underlying mechanisms, provides high confidence in summer warming that exceeds the global average (see Box TS.12). In a less clear-cut case for Cape Town, South Africa, despite consistency among global model future projections, there is medium confidence in a projected future drier climate due to the lack of consistency in links between increasing greenhouse gases, 110
 
Technical Summary Box TS.11 l Climate Services Climate services involve providing climate information to assist decision-making, for example, about how extreme rainfall will change to inform improvements in urban drainage. Since AR5, there has been a significant increase in the range and diversity of climate service activities (very high confidence). The level of user-engagement, co-design and co-production are factors determining the utility of climate services, while resource limitations for these activities constrain their full potential. {12.6, Cross-Chapter Box 12.2}
Climate services include engagement from users and providers and an effective access mechanism; they are responsive to user needs and based on integrating scientifically credible information and relevant expertise. Climate services are being developed across regions, sectors, time scales and user-groups and include a range of knowledge brokerage and integration activities. These involve identifying knowledge needs; compiling, translating and disseminating knowledge; coordinating networks and building capacity through informed decision-making; analysis, evaluation and development of policy; and personal consultation.
TS Since AR5, climate change information produced in climate service contexts has increased significantly due to scientific and technological advancements and growing user awareness, requirements and demand (very high confidence). Climate services are growing rapidly and are highly diverse in their practices and products. The decision-making context, level of user engagement and co-production between scientists, practitioners and intended users are important determinants of the type of climate service developed and their utility for supporting adaptation, mitigation and risk management decisions. They require different types of user-producer engagement depending on what the service aims to deliver (high confidence), and these fall into three broad categories: website-based services, interactive group activities and focused relationships.
Realization of the full potential of climate services is often hindered by limited resources for the co-design and co-production process, including sustained engagement between scientists, service providers and users (high confidence). Further challenges relate to the development and provision of climate services, generation of climate service products, communication with users, and evaluation of their quality and socio-economic benefit. (Section TS.4.1) {1.2.3, 10.5.4, 12.6, Cross-Chapter Box 12.2, Glossary}
Box TS.12 l Multiple Lines of Evidence for Assessing Regional Climate Change and the Interactive Atlas A key novel element in the AR6 is the Working Group I Atlas, which includes the Interactive Atlas (https://interactive-atlas.ipcc.ch/). The Interactive Atlas provides the ability to explore much of the observational and climate model data used as lines of evidence in this assessment to generate regional climate information. {Atlas.2}
A significant innovation in the AR6 WGI Report is the Atlas. Part of its remit is to provide region-by-region assessment on changes in mean climate and to link with other WGI chapters to generate climate change information for the regions. An important component is the new online interactive tool, the Interactive Atlas, with flexible spatial and temporal analyses of much of the observed, simulated past and projected future climate change data underpinning the WGI assessment. This includes the ability to generate global maps and a number of regionally aggregated products (time series, scatter plots, tables, etc.) for a range of observations and ensemble climate change projections of variables (such as changes in the climatic impact-drivers summarized in Table TS.5) from the Coupled Model Intercomparison Project Phases 5 and 6 (CMIP5, CMIP6) and the Coordinated Regional Climate Downscaling Experiment (CORDEX). The data can be displayed and summarized under a range of SSP-RCP scenarios and future time slices and also for different global warming levels, relative to several different baseline periods. The maps and various statistics can be generated for annual mean trends and changes or for any user-specified season. A new set of WGI reference regions is used for the regional summary statistics and applied widely throughout the report (with the regions, along with aggregated datasets and the code to generate these, available at the ATLAS GitHub: https://github.com/IPCC-WG1/Atlas).
Box TS.12, Figure 1 shows how the Interactive Atlas products, together with other lines of evidence, can be used to generate climate information for an illustrative example of the Mediterranean summer warming. The lines of evidence include the understanding of relevant mechanisms, dynamic and thermodynamic processes and the effect of aerosols in this case (Box TS.12, Figure 1a); trends in observational datasets (which can have different spatial and temporal coverage; Box TS.12, Figure 1b, c); and attribution of these trends and temperature projections from global and regional climate models at different resolutions, including single-model initial-condition large ensembles (SMILEs; Box TS.12, Figure 1d, e). Taken together, this evidence shows there is high confidence that the 111
 
Technical Summary Box TS.12 (continued) projected Mediterranean summer temperature increase will be larger than the global mean, with consistent results from CMIP5 and CMIP6 (Box TS.12, Figure 1e). However, CMIP6 results project both more pronounced warming than CMIP5 for a given emissions scenario and time period and a greater range of changes (Box TS.12, Figure 1d). {10.6.4, Atlas.2, Atlas.8.4}
The Interactive Atlas allows for flexible spatial and temporal analyses of essential climate variables, extreme indices and climatic impact-drivers, including multiple lines of evidence to support the assessment of regional climate change:
                                                                                                                                                                                - Observations TS                                                                                                                                                                              - CMIP5
                                                                                                                                                                                - CMIP6
                                                                                                                                                                                - CORDEX Regional information is displayed when                                                                                                                                CORDEX is available for 12 clicking on one or                                                                                                                              continent-wide domains.
several subregions Regional (aggregated) information for reference and typological regions:
Time series Stripes Annual cycle plots Summary tabular information.
Scatter plots (e.g., precip. vs temp.)
Dimensions of analysis include time periods across scenarios and global warming levels (1&#xba;C, 2&#xba;C, 3&#xba;C and 4&#xba;C).
Available in the Interactive Atlas Not available from the Interactive Atlas (a) Mechanisms of enhanced Mediterranean warming                                                                                  (b) Station locations                                                        (c) Temperature trend distribution Past period (1960-2014)
OBS Warm Atlantic Ocean                Reduction of                Surface drying          Lapse rate difference                                                                                                                            Ensemble means aerosols                                              Over land stronger                                                                                                                                        CMIP5 Enhanced                      warming than over sea                                                                                                                                      CMIP6 sensible                                                                                                                                                                          HighResMIP Reduced                  Sea    Land                                                                                                                          CORDEX EUR-44 heat flux latent                                                                                                                                                        CORDEX EUR-11 heat flux                                                                                                                                                            MIROC6 Altitude Dynamic and                                                                                                                                                                                                                        CSIRO-Mk3-6-0 thermodynamic effects                                                                                                                                                                                                                  MPI-ESM d4PDF E-OBS          Donat et al. 2013                                                  0.2            0.4              0.6 Less dimming and        Reduction of latent heat flux and increase of solar radiation increase of sensible heat flux            Temperature                                                                                                            Trend (&deg;C/decade)
(d) Mediterranean temperature anomalies                                                                                              (e) Mediterranean summer vs global warming Baseline period is 1995-2014                                                      CMIP6 long term                                                                        Baseline period is 1861-1900 temperature change 10                    CRU TS                            CMIP5                                                                                      10                                          8      CMIP5 (N=21)
Mediterranean summer temperature (&deg;C)
HadCRUT5                          CMIP6                                                                                                                                                RCP8.5 7
8                Berkeley Earth                    HighResMIP                                                                                8                                                          RCP6.0 CORDEX EUR-44                                                                                                                          6 RCP4.5 6                                                  CORDEX EUR-11                                                                              6                                                          RCP2.6 5
4 4                                                                                                                                              4
(&deg;C) 3 CMIP6 (N=31) 2                                                                                                                                              2 2                                                  SSP5-8.5 1
SSP3-7.0 0                                                                                                                                              0 SSP1-2.6                                                                                                    SSP2-4.5 SSP2-4.5                                                0                                                  SSP1-2.6
                  -2                                                                                                          Long term              SSP3-7.0 SSP5-8.5                                                    0          1            2          3              4 2000                  2020                2040                2060                        2080                2100                                                                                        Global temperature (&deg;C)
Box TS.12, Figure 1 l Example of generating regional climate information from multiple lines of evidence for the case of Mediterranean summer warming.
112
 
Technical Summary Box TS.12 (continued)
Box TS.12, Figure 1 (continued): The intent of this figure is to provide an example of using different lines of evidence to assess the confidence in or likelihood of a projected change in regional climate and which of these lines of evidence are available to view and explore in the Interactive Atlas. (a) Mechanisms and feedbacks involved in enhanced Mediterranean summer warming. (b) Locations of observing stations from different datasets. (c) Distribution of 1960-2014 summer temperature trends (&deg;C per decade) for observations (black crosses), CMIP5 (blue circles), CMIP6 (red circles), HighResMIP (orange circles), CORDEX EUR-44 (light blue circles), CORDEX EUR-11 (green circles), and selected single model initial-condition large ensembles (SMILEs; grey boxplots, MIROC6, CSIRO-Mk3-6-0, MPI-ESM and d4PDF). (d) Time series of area averaged (25&deg;N-50&deg;N, 10&deg;W-40&deg;E) land point summer temperature anomalies (&deg;C, baseline period is 1995-2014): the boxplot shows long term (2081-2100) temperature changes of different CMIP6 scenarios in respect to the baseline period. (e) Projected Mediterranean summer warming in comparison to global annual mean warming of CMIP5 (RCP2.6, RCP4.5, RCP6.0 and RCP8.5) and CMIP6 (SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5) ensemble means (lines) and spread (shading). {Figure 10.20, Figure 10.21, Figure Atlas.8}
TS.4.2        Drivers of Regional Climate Variability                                    temperature is moderated or amplified by soil moisture feedback, and Change                                                                  snow/ice-albedo feedback, regional forcing from land-use/land-cover changes, forcing from aerosol concentrations, or decadal/                TS Anthropogenic forcing, including GHGs and aerosols, but                                multi-decadal natural variability. Changes in local and remote also regional land use and irrigation have all affected                                aerosol forcings lead to south-north gradients of the effective observed regional climate changes (high confidence) and                                radiative forcing (hemispherical asymmetry). Along latitudes, will continue to do so in the future (high confidence), with                          it is more uniform, with strong amplification of the temperature various degrees of influence and response times, depending                            response towards the Arctic (medium confidence). The decrease on warming levels, the nature of the forcing and the relative                          of SO2 emissions since the 1980s reduces the damping effect of importance of internal variability.                                                    aerosols, leading to a faster increase in surface air temperature that is most pronounced at mid- and high latitudes of the Northern Since the late 19th century, major modes of variability                                Hemisphere, where the largest emissions reductions have taken (MoVs) exhibited fluctuations in frequency and magnitude                              place (medium confidence). {1.3, 3.4.1, 6.3.4, 6.4.1, 6.4.3, 8.3.1, at multi-decadal time scales, but no sustained trends                                  8.3.2, Box 8.1, 10.4.2, 10.6, 11.1.6, 11.3}
outside the range of internal variability (Table TS.4). An exception is the Southern Annular Mode (SAM), which has                                Multi-decadal dimming and brightening trends in incoming solar become systematically more positive (high confidence) and                              radiation at Earths surface occurred at widespread locations (high is projected to be more positive in all seasons, except for                            confidence). Multi-decadal variation in anthropogenic aerosol December-January-February (DJF), in high CO2 emissions                                emissions are thought to be a major contributor (medium confidence),
scenarios (high confidence). The influence of stratospheric                            but multi-decadal variability in cloudiness may also have played a ozone forcing on the SAM trend has been reduced since the                              role. Volcanic eruptions affect regional climate through their spatially early 2000s compared to earlier decades, contributing to                              heterogeneous effect on the radiative budget as well as through the weakening of its positive trend as observed over 2000-                            triggering dynamical responses by favouring a given phase from 2019 (medium confidence).                                                              some MoVs, for instance. {1.4.1, Cross-Chapter Box 1.2, 2.2.1, 2.2.2, 3.7.1, 3.7.3, 4.3.1, 4.4.1, 4.4.4, Cross-Chapter Box 4.1, 7.2.2, 8.5.2, In the near term, projected changes in most of the MoVs                                10.1.4, 11.1.6, 11.3.1}
and related teleconnections will likely be dominated by internal variability. In the long term, it is very likely that                        Historical urbanization affects the observed warming trends in cities the precipitation variance related to El Nino-Southern                                and their surroundings (very high confidence). Future urbanization Oscillation will increase. Physical climate storylines,                                will amplify the projected air temperature under different background including the complex interplay between climate drivers,                              climates, with a strong effect on minimum temperatures that could MoVs, and local and remote forcing, increase confidence                                be as large as the global warming signal (very high confidence) in the understanding and use of observed and projected                                (Box TS.14). Irrigation and crop expansion have attenuated increases regional changes. {2.4, 3.7, 4.3, 4.4, 4.5, 6.4, 8.3, 8.4, 10.3,                      in summer hot extremes in some regions, such as central North 10.4, 11.3}                                                                            America (medium confidence) (Box TS.6). {Box 10.3, 11.1.6, 11.3}
TS.4.2.2 Modes of Variability and Regional Teleconnections TS.4.2.1 Regional Fingerprints of Anthropogenic and Natural Forcing                                                        Modes of variability (Annex IV, Table TS.4) have existed for millennia or longer (high confidence), but there is low confidence in detailed While anthropogenic forcing has contributed to multi-decadal                              reconstructions of most of them prior to direct instrumental records.
mean precipitation changes in several regions, internal variability                      MoVs are treated as a main source of uncertainties associated with can delay emergence of the anthropogenic signal in long-term                              internal dynamics, as they can either accentuate or dampen, even precipitation changes in many land regions (high confidence). At the                      mask, the anthropogenically forced responses. {2.4, 8.5.2, 10.4, 10.6, regional scale, the effect of human-induced GHG forcing on extreme                        11.1.5, Atlas.3.1}
113
 
Technical Summary Since the late 19th century, major MoVs (Table TS.4) show              contributed to observed temporal evolution in the Atlantic Multi-no sustained trends, exhibiting fluctuations in frequency and          decadal Variability (AMV) and associated regional teleconnections, magnitude at multi-decadal time scales, except for the Southern        especially since the 1960s, but there is low confidence in the Annular Mode (SAM), which has become systematically more              magnitude of this influence and the relative contributions of positive (high confidence) (Table TS.4). It is very likely that human  natural and anthropogenic forcings. Internal variability is the influence has contributed to this trend from the 1970s to the          main driver of Pacific Decadal Variability (PDV) observed since 1990s, and to the associated strengthening and southward shift        the start of the instrumental records (high confidence), despite of the Southern Hemispheric extratropical jet in austral summer.      some modelling evidence for potential external influence. There is The influence of stratospheric ozone forcing on the SAM trend has      medium confidence that the AMV will undergo a shift towards a been reduced since the early 2000s compared to earlier decades,        negative phase in the near term. {2.4, 3.7.6, 3.7.7, 8.5.2, 4.4.3}
contributing to the weakening of its positive trend observed over 2000-2019 (medium confidence). By contrast, the cause of the Northern Annular Mode (NAM) trend toward its positive phase since the 1960s and associated northward shifts of Northern TS Hemispheric extratropical jet and storm track in boreal winter is not well understood. The evaluation of model performance on simulating MoVs is assessed in Section TS.1.2.2. {2.3.3, 2.4, 3.3.3, 3.7.1, 3.7.2}
In the near term, the forced change in SAM in austral summer is likely to be weaker than observed during the late 20th century under all five SSPs assessed. This is because of the opposing influence in the near to mid-term from stratospheric ozone recovery and increases in other greenhouse gases on the Southern Hemisphere summertime mid-latitude circulation (high confidence). In the near term, forced changes in the SAM in austral summer are therefore likely to be smaller than changes due to natural internal variability. In the long term (2081-2100) under the SSP5-8.5 scenario, the SAM index is likely to increase in all seasons relative to 1995-2014. The CMIP6 multi-model ensemble projects a long-term (2081-2100) increase in the boreal wintertime NAM index under SSP3-7.0 and SSP5-8.5, but regional associated changes may deviate from a simple shift in the mid-latitude circulation due to a modified teleconnection resulting from interaction with a modified mean background state.
{4.3.3, 4.4.3, 4.5.1, 4.5.3, 8.4.2}
Human influence has not affected the principal tropical modes of interannual climate variability (Table TS.4) and their associated regional teleconnections beyond the range of internal variability (high confidence). It is virtually certain that the El Nino-Southern Oscillation (ENSO) will remain the dominant mode of interannual variability in a warmer world. There is no consensus from models for a systematic change in amplitude of ENSO sea surface temperature (SST) variability over the 21st century in any of the SSP scenarios assessed (medium confidence). However, it is very likely that rainfall variability related to ENSO will be enhanced significantly by the latter half of the 21st century in the SSP2-4.5, SSP3-7.0 and SSP5-8.5 scenarios, regardless of the amplitude changes in SST variability related to the mode. It is very likely that rainfall variability related to changes in the strength and spatial extent of ENSO teleconnections will lead to significant changes at regional scale. {3.7.3, 3.7.4, 3.7.5, 4.3.3, 4.5.3, 8.4.2, 10.3.3}
Modes of decadal and multi-decadal variability over the Pacific and Atlantic Ocean exhibit no significant changes in variance over the period of observational records (high confidence). There is medium confidence that anthropogenic and volcanic aerosols 114
 
Technical Summary Table TS.4 l Summary of the assessments on modes of variability (MoVs) and associated teleconnections. (a) Assessments on observed changes since the start of instrumental records, Coupled Model Intercomparison Project Phases 5 and 6 (CMIP5 and CMIP6) model performance, human influence on the observed changes, and near-term (2021-2040) and mid- to long-term (2041-2100) changes. Curves schematically illustrate the assessed overall changes, with the horizontal axis indicating time, and are not intended to precisely represent the time evolution. (b) Fraction of surface air temperature (SAT) and precipitation (pr) variance explained at interannual time scale by each MoV for each AR6 region (numbers in each cell; in percent). Values correspond to the average of significant explained variance fractions based on HadCRUT, GISTEMP, BerkeleyEarth and CRU-TS (for SAT) and GPCC and CRU-TS (for precipitation). Significance is tested based on F-statistics at the 95% level confidence, and a slash indicates that the value is not significant in more than half of the available data sets. The colour scale corresponds to the sign and values of the explained variance as shown at the bottom. The corresponding anomaly maps are shown in Annex IV. DJF: December-January-February. MAM: March-April-May. JJA: June-July-August. SON: September-October-November.
In (b), Northern Annular Mode (NAM) and El Nino-Southern Oscillation (ENSO) teleconnections are evaluated for 1959-2019, Southern Annular Mode (SAM) for 1979-2019, Indian Ocean Basin (IOB), Indian Ocean Dipole (IOD), Atlantic Zonal Mode (AZM) and Atlantic Meridional Mode (AMM) for 1958-2019, and Pacific Decadal Variability (PDV) and Atlantic Multi-decadal Variability (AMV) for 1900-2019. All data are linearly detrended prior to computation. (Section TS.1.2.2) {2.4, 3.7, 4.3.3, 4.4.3, 4.5.3, Table Atlas.1, Annex IV}
(a) Assessments on MoV.
NAM              SAM              ENSO            IOB              IOD          AZM            AMM              PDV            AMV 1990s            2000s 1960s 1950s                            Within proxy-    Within proxy-                                      Dominated      Dominated TS inferred          inferred        Limited          Limited        by multi-      by multi-Past changes since the                  2010s variability      variability    evidence        evidence          decadal        decadal start of observations              Boreal          Austral 1400- Since        range            range                                        fluctuations    fluctuations winter          summer 1850 1950s
{2.4.1.1}        {2.4.1.2}          {2.4.2}      {2.4.3}          {2.4.3}        {2.4.4}          {2.4.4}          {2.4.5}          {2.4.6}
High            High            Medium        Medium            Medium            Low              Low          Medium          Medium CMIP5 and CMIP6                performance    performance        performance  performance      performance    performance      performance    performance      performance model performance
{3.7.1}          {3.7.2}            {3.7.3}      {3.7.4}          {3.7.4}        {3.7.5}          {3.7.5}          {3.7.6}          {3.7.7}
Contributed through GHG                                                                                                          Contributed No robust                              Low        No robust                        No robust        No robust Human influence on                              (all seasons)                                      Not detected                                    Not detected        through evidence                          agreement      evidence                        evidence        evidence the observed changes                              & ozone                                                                                                            aerosols (DJF)
{3.7.1}          {3.7.2}            {3.7.3}      {3.7.4}          {3.7.4}        {3.7.5}          {3.7.5}          {3.7.6}          {3.7.7}
                                                                                                                                                                        +
Internal                            Internal No robust        No robust      No robust        No robust          Limited                  -
Near-term future                variability                          variability evidence        evidence        evidence        evidence        evidence      Phase shift changes (2021-2040)            dominates        All seasons        dominates except DJF                                                                                                          from + to -
{4.4.3.1}        {4.4.3.1}          {4.4.3.2}    {4.4.3.3}        {4.4.3.3}      {4.4.3.4}        {4.4.3.4}        {4.4.3.5}      {4.4.3.6}
SSP5-8.5 SSP3-7.0 SSP5-8.5 SSP3-7.0 All seasons                  No robust          increase      No robust        No robust Mid-to-long-term                              SSP1-1.9                                                                                                              No changes DJF                      Increase in    evidence        in extreme      evidence        evidence future changes                SSP1-2.6                                                                                                              Decrease in SSP1-1.9                            precipitation                      positive (2041-2100)                                                                                                                                            variance DJF                    variance                        events
{4.3.3.1;        {4.3.3.1;          {4.3.3.2;
{4.5.3.3}        {4.5.3.3}      {4.5.3.4}        {4.5.3.4}        {4.5.3.5}      {4.5.3.6}
4.5.3.1}          4.5.3.1}          4.5.3.2}
low confidence                          medium confidence                    high confidence more likely than not                      likely                                very likely 115
 
Technical Summary Table TS.4 (continued): (b) Regional climate anomalies associated with MoV.
Mode                        NAM            SAM                ENSO              IOB              IOD            AZM              AMM              PDV            AMV Season                        DJF              DJF              DJF            MAM              SON              JJA              JJA        annual        annual Variable                  SAT        pr  SAT        pr  SAT        pr  SAT        pr    SAT        pr  SAT        pr  SAT        pr  SAT      pr  SAT  pr Mediterranean          28          58                    7                                                                                                  19 Sahara              58                                                14                              10          19              12            9    12    25 Western Africa        25                                            15  45                              21              10                        6    6    23 Central Africa        19          8                10  14              50                              13                                          10  14    11 Africa                        North Eastern Africa      19          7                                14  36                          32                                    7              7 South Eastern Africa                                        14          22  36                          57                    10                        4    9 West Southern Africa                                        49          26  27        16    8                                                4        12  5 East Southern Africa                      13              75          34  35        7                                                      4        6 Madagascar                                              24              24        7    11        10                    9                              5 TS                                        West Siberia          45                                            7                                                9                                          11 East Siberia          52                                                                                                                    3                    11 Russian Far East        8          10                    11              6                                                                                5        5 West Central Asia                                                                      15              21                                              4 East Central Asia                                                                      38 Asia Tibetan Plateau                    15                                                      15        7                11                    6        5    9 East Asia                                              7          20              23                                9                              9    13 South Asia          9                                                  12                          8                    8                              5 South East Asia                                          39          31  73        6                48                                    5        12            7 Arabian Peninsula        32                                                10        24              20                                              5    13        7 Northern Australia                                        21          13  38                          19                                7    7        7 Australasia Central Australia                        14              21          12  18              22        20              7                7    6        5 East Australia                          22              20          11  18                9        8                7                    7        8 Southern Australia                                                    11                    23        40              8                              3 New Zealand                            16 Southern Central America                                      21          16  33              10        11                    17              6              6        7 Central & South America North-Western South America              7    14          16  82          17  54              18                              13          16  7        8 Northern South America      7                                56          58  61                                          22  17          24  9        12  7 North-Eastern South America                                    25              58        19    9        12                                8 South American Monsoon                                        54              31              22        7                                6    7 South-Western South America                                10  16          14  17              10        16                                    8 South-Eastern South America                                                21              13    21        10              12                              5              6 Southern South America                                  23                                    13        7                                                              9 Mediterranean          28          58                    7                                                                                                  19 Europe Western and Central Europe    28          18                                                      13        10                                              4              8 Eastern Europe          35                                                                                  7 Northern Europe        53          32                                                                                                                      6 Northern Central America                      10          26  13          27  18                              7          12  15          12            6    19 Western North America                                                                                                                            4              6        5 North America Central North America      17                          12              17                                    8                                          3    9        6 Eastern North America      12                                                                                  11          9                    4              9        4 North-Eastern North America  18          26                                                                      8                                          10  9        4 North-Western North America              14                    10          8    17                                                                8        4 Small                                Caribbean                                        10  15          18  26        8                10                    17          12  7                        5 Islands                                Pacific Greenland/Iceland        42          8                                                                                                    7                  44 Polar Terrestrial                  Russian Arctic        25          10                                                                                                                  6    11        8 Regions West Antarctica                                                                                            8                21 East Antarctica                          38 colderwarmer                                                    drierwetter Not significant in >50% of available data sets 40 30 20 0 20 30 40                                        40 30 20 0 20 30 40                                  Data unavailable in >50% of data sets Temperature anomalies and                                    Precipitation anomalies and explained variance (%)                                      explained variance (%)
116
 
Technical Summary TS.4.2.3 Interplay Between Drivers of Climate Variability                                                annual to decadal time scales (high confidence). The anthropogenic and Change at Regional Scales                                                                    signal in regional sea level change will emerge in most regions by 2100 (medium confidence). {9.2.4, 9.6.1, 10.4.1, 10.4.2, 10.4.3}
Anthropogenic forcing has been a major driver of regional mean temperature change since 1950 in many sub-continental regions of                                          Regional climate change is subject to the complex interplay between the world (virtually certain). At regional scales, internal variability                                  multiple external forcings and internal variability. Time evolution is stronger, and uncertainties in observations, models and external                                      of mechanisms operating at different time scales can modify the forcing are all larger than at the global scale, hindering a robust                                      amplitude of the regional-scale response of temperature, and assessment of the relative contributions of greenhouse gases,                                            both the amplitude and sign of the response of precipitation, to stratospheric ozone, and different aerosol species in most of the cases.                                  anthropogenic forcing (high confidence). These mechanisms include Multiple lines of evidence, combining multi-model ensemble global                                        non-linear temperature, precipitation and soil moisture feedbacks; projections with those coming from single-model initial-condition                                        slow and fast responses of SST patterns; and atmospheric circulation large ensembles, show that internal variability is largely contributing                                  changes to increasing GHGs. Land-use and aerosol forcings and to the delayed or absent emergence of the anthropogenic signal in                                        land-atmosphere feedback play important roles in modulating long-term regional mean precipitation changes (high confidence).                                          regional changes, for instance in weather and climate extremes (high                              TS Internal variability in ocean dynamics dominates regional patterns on                                    confidence). These can also lead to a higher warming of extreme Pathway to understanding past and assessing future climate changes at regional scale The South-Eastern South America (SES) case study (a) Identi"cation of climate drivers and phenomena for interpreting                          (b) Models simulations/evaluation of SES DJF precipitation over SES observed precipitation trend and variability in austral summer (DJF)                      the historical period and 21st century based on 7 large ensembles Aerosol                                                                        GPCC observations Anomalies in mm/month w.r.t 1995-2014 AMV                                          ell ley C Had Near    Mid            Long Term    Term            Term ENSO SES PDV GHG Rossby                                    + O3 depletion 83.6%                53.3%          14.7%        6.3%
Waves Internal variability uncertainty lar                      Model uncertainty Precipitation trend (GPCC/1950-2014)                  Po tex                                                                                                            5.0%
r Vo (mm/month/decade)
Scenario uncertainty                                            1.4%  41.7%              78.9%
14.9%
Near term            Mid term            Long term Figure TS.21 l Example of the interplay between drivers of climate variability and change at regional scale to understand past and projected changes. The figure intent is to show an illustrative pathway for understanding past, and anticipating future, climate change at regional scale in the presence of uncertainties. (a) Identification of the climate drivers and their influences on climate phenomena contributing through teleconnection to South-Eastern South America (SES) summer (December-January-February; DJF) precipitation variability and trends observed over 1950-2014. Drivers (red squares) include modes of variability as well as external forcing. Observed precipitation linear trend from GPCC is shown on continents (green-brown colour bar in mm month-1 per decade) and the SES AR6 WGI reference region is outlined with the thick black contour. Climate phenomena leading to local effects on SES are schematically presented (blue ovals). (b) Time series of decadal precipitation anomalies for DJF SES simulated from seven large ensembles of historical plus RCP8.5 simulations over 1950-2100. Shading corresponds to the 5-95th range of climate outcomes given from each large ensemble for precipitation (in mm month -1) and thick coloured lines stand for their respective ensemble mean. The thick time series in white corresponds to the multi-model multi-member ensemble mean, with model contribution being weighted according to their ensemble size. GPCC observation is shown in the light black line with squares over 1950-2014, and the 1995-2014 baseline period has been retained for calculation of anomalies in all datasets. (c) Quantification of the respective weight (in percent) between the individual sources of uncertainties (internal in grey, model in magenta and scenario in green) at near-term, mid-term and long-term temporal windows defined in AR6 and highlighted in (b) for SES DJF precipitation. All computations are done with respect to 1995-2014, taken as the reference period, and the scenario uncertainty is estimated from Coupled Model Intercomparison Project Phase 5 (CMIP5) using the same set of models as for the large ensembles that have run different Representative Concentration Pathway (RCP) scenarios. {Figure 10.12a}
117
 
Technical Summary temperatures compared to mean temperature (high confidence),              which are relevant for the region. In fact, local changes over SES and possibly cooling in some regions (medium confidence). The            in terms of moisture convergence, ascending motion and storm-soil moisture-temperature feedback was shown to be relevant for          track locations depend on these climate phenomena, and they are past and present-day heatwaves based on observations and model            overall responsible for the observed precipitation trends. Projections simulations. {10.4.3, 11.1.6, 11.3.1}                                    suggest continuing positive trends in rainfall over SES in the near-term in response to GHG emissions scenarios. Multi-model mean and South-Eastern South America (SES) is one of the AR6 WGI reference        ensemble spread are not sufficient to characterize situations where regions (outlined with black thick contour in Figure TS.21a), and        different models simulate substantially different or even opposite it is used here as an illustrative example of the interplay between      changes (high confidence). In such cases, physical climate storylines drivers of climate variability and change at regional scale. Austral      addressing possible outcomes for climate phenomena shown to summer (DJF) precipitation positive trends have been observed            play a role in the variability of the region of interest can aid the over the region during 1950-2014. Drivers of this change include          interpretation of projection uncertainties. In addition, single-model MoVs, such as AMV, ENSO, and PDV, as well as external forcing,            initial-condition large ensembles of many realizations of internal like GHG increases and ozone depletion together with aerosols (as        variability are required to separate internal variability from forced TS illustrated in Figure TS.21a). Modes of variability and external forcing  changes (high confidence) and to partition the different sources of collectively affect climate phenomena, such as the Hadley cell width      uncertainties as a function of future assessed periods. {10.3.4, 10.4.2, and strength, Rossby waves activity emerging from the large-scale        Figure 10.12a}
tropical SST anomalies, and the Southern Hemisphere polar vortex, Box TS.13 l Monsoons Global land monsoon precipitation decreased from the 1950s to the 1980s, partly due to anthropogenic aerosols, but has increased since then in response to GHG forcing and large-scale multi-decadal variability (medium confidence).
Northern Hemispheric anthropogenic aerosols weakened the regional monsoon circulations in South Asia, East Asia and West Africa during the second half of the 20th century, thereby offsetting the expected strengthening of monsoon precipitation in response to GHG-induced warming (high confidence).
During the 21st century, global land monsoon precipitation is projected to increase in response to GHG warming in all time horizons and scenarios (high confidence). Over South and South East Asia, East Asia and the central Sahel, monsoon precipitation is projected to increase, whereas over North America and the far western Sahel it is projected to decrease (medium confidence). There is low confidence in projected precipitation changes in the South American and Australian-Maritime Continent monsoons. At global and regional scales, near-term monsoon changes will be dominated by the effects of internal variability (medium confidence). {2.3, Cross-Chapter Box 2.4, 3.3, 4.4, 4.5, 8.2, 8.3, 8.4, 8.5, Box 8.1, Box 8.2, 10.6}
Global Monsoon Paleoclimate records indicate that during warm climates, like the mid-Pliocene Warm Period, monsoon systems were stronger (medium confidence). In the instrumental records, global summer monsoon precipitation intensity has likely increased since the 1980s, dominated by Northern Hemisphere summer trends and large multi-decadal variability. Contrary to the expected increase of precipitation under global warming, the Northern Hemisphere monsoon regions experienced declining precipitation from the 1950s to 1980s, which is partly attributable to the influence of anthropogenic aerosols (medium confidence) (Box TS.13, Figure 1). {2.3.1, Cross-Chapter Box 2.4, 3.3.2, 3.3.3}
With continued global warming, it is likely that global land monsoon precipitation will increase during this century (Box TS.13, Figure 1),
particularly in the Northern Hemisphere, although the monsoon circulation is projected to weaken. A slowdown of the tropical circulation with global warming can partly offset the warming-induced strengthening of precipitation in monsoon regions (high confidence). In the near term, global monsoon changes are likely to be dominated by the effects of internal variability and model uncertainties (medium confidence). In the long term, global monsoon rainfall change will feature a robust north-south asymmetry characterized by a greater increase in the Northern Hemisphere than in the Southern Hemisphere and an east-west asymmetry characterized by enhanced Asian-African monsoons and a weakened North American monsoon (medium confidence). {4.4.1, 4.5.1, 8.4.1}
Regional Monsoons Paleoclimate reconstructions indicate stronger monsoons in the Northern Hemisphere but weaker ones in the Southern Hemisphere during warm periods, particularly for the South and South East Asian, East Asian, and North and South American monsoons, with the opposite occurring during cold periods (medium confidence). It is very likely that Northern Hemispheric anthropogenic aerosols weakened the regional monsoon circulations in South Asia, East Asia and West Africa during the second half of the 20th century, 118
 
Technical Summary Box TS.13 (continued)
(a) Global and regional monsoon domains 2014 NAmerM                                                  SAsia M EAsia M WAfriM EqAmer SAfri SAmer M                                AusMCM TS (b) Historical trend in monsoon precipitation Trend in precipitation (mm/day/decade)
Simulations: NAT ALL GHG AER 0.12                                                                      Observations: x APHRO x CRU x GPCP x
0.08                                                                x x
0.04      xx x
0                              x x
xx x
x x                    x
                                          -0.04                            x x
                                          -0.08                  x x
                                          -0.12 NAmerM    WA fri M    SAsiaM      EAsiaM    EqAmer    SAmerM      SAfri  AusMCM        GM (c) Projected future change in monsoon precipitation (SSP2-4.5) 20 Change in precipitation (%)
10 0
Near term
                                          -10                                                                                              Mid term Long term
                                          -20 NAmerM    WA fri M    SAsiaM      EAsiaM    EqAmer    SAmerM      SAfri  AusMCM        GM Box TS.13, Figure 1 l Global and regional monsoons: past trends and projected changes. The intent of this figure is to show changes in precipitation over regional monsoon domains in terms of observed past trends, how greenhouse gases and aerosols relate to these changes, and in terms of future projections in one intermediate emissions scenario in the near, medium and long term. (a) Global (black contour) and regional monsoons (colour shaded) domains. The global monsoon (GM) is defined as the area with local summer-minus-winter precipitation rate exceeding 2.5 mm day-1 (see Annex V). The regional monsoon domains are defined based on published literature and expert judgement (see Annex V) and accounting for the fact that the climatological summer monsoon rainy season varies across the individual regions. Assessed regional monsoons are South and South East Asia (SAsiaM, Jun-July-August-September), East Asia (EAsiaM, June-July-August),
West Africa (WAfriM, June-July-August-September), North America (NAmerM, July-August--September), South America (SAmerM, December-January-February),
Australia and Maritime Continent Monsoon (AusMCM, December-January-February). Equatorial South America (EqSAmer) and South Africa (SAfri) regions are also shown, as they receive unimodal summer seasonal rainfall although their qualification as monsoons is subject to discussion. (b) Global and regional monsoons precipitation trends based on DAMIP CMIP6 simulations with both natural and anthropogenic (ALL), greenhouse gas only (GHG), aerosols only (AER) and natural only (NAT) radiative forcing. Weighted ensemble means are based on nine Coupled model Intercomparison Project Phase 6 (CMIP6) models contributing to the MIP (with at least three members). Observed trends computed from CRU, GPCP and APHRO (only for SAsiaM and EAsiaM) datasets are shown as well. (c) Percentage change in projected seasonal mean precipitation over global and regional monsoons domain in the near term (2021-2040), mid-term (2041-2060), and long term (2081-2100) under SSP2-4.5 based on 24 CMIP6 models. {Figures 8.11 and 8.22}
119
 
Technical Summary Box TS.13 (continued) thereby offsetting the expected strengthening of monsoon precipitation in response to GHG-induced warming (Box TS.13, Figure 1).
Multiple lines of evidence explain this contrast over South Asia, with the observed trends dominated by the effects of aerosols, while future projections are mostly driven by GHG increases. The recent partial recovery and enhanced intensity of monsoon precipitation over West Africa is related to the growing influence of GHGs with an additional contribution due to the reduced cooling effect of anthropogenic aerosols, emitted largely from North America and Europe (medium confidence). For other regional monsoons, that is, North and South America and Australia, there is low confidence in the attribution of recent changes in precipitation (Box TS.13, Figure 1) and winds. {2.3.1, 8.3.1, 8.3.2, Box 8.1, 10.6.3}
Projections of regional monsoons during the 21st century indicate contrasting (region-dependent) and uncertain precipitation and circulation changes. The annual contrast between the wettest and driest month of the year is likely to increase by 3-5% per degree Celsius in most monsoon regions in terms of precipitation, precipitation minus evaporation, and runoff (medium confidence). For the TS    North American monsoon, projections indicate a decrease in precipitation, whereas increased monsoon rainfall is projected over South and South East Asia and over East Asia (medium confidence) (Box TS.13, Figure 1). West African monsoon precipitation is projected to increase over the central Sahel and decrease over the far western Sahel (medium confidence). There is low confidence in projected precipitation changes in the South American and Australian-Maritime Continent regional monsoons (for both magnitude and sign)
(Box TS.13, Figure 1). There is medium confidence that the monsoon season will be delayed in the Sahel and high confidence that it will be delayed in North and South America. {8.2.2, 8.4.2.4, Box 8.2}
Building the Assessment from Multiple Lines of Evidence Large natural variability of monsoon precipitation across different time scales, found in both paleoclimate reconstructions and instrumental measurements, poses an inherent challenge for robust quantification of future changes in precipitation at regional and smaller spatial scales. At both global and regional scales, there is medium confidence that internal variability contributes the largest uncertainty related to projected changes, at least in the near term (2021-2040). A collapse of the Atlantic Meridional Overturning Circulation could weaken the African and Asian monsoons but strengthen the Southern Hemisphere monsoons (high confidence).
{4.4.4, 4.5.1, Cross-Chapter Box 4.1, 8.5.2, 8.6.1, 9.2.3, 10.6.3}
Overall, long-term (2081-2100) future changes in regional monsoons like the South and South East Asian monsoon are generally consistent across global (including high-resolution) and regional climate models and are supported by theoretical arguments.
Uncertainties in simulating the observed characteristics of regional monsoon precipitation are related to varying complexities of regional monsoon processes and their responses to external forcing, internal variability, and deficiencies in representing monsoon warm rain processes, organized tropical convection, heavy orographic rainfall and cloud-aerosol interactions. {8.3.2, 8.5.1, 10.3.3, 10.6.3}
TS.4.3    Regional Climate Change and Implications for                    Relative sea level rise is very likely to virtually certain Climate Extremes and Climatic Impact-Drivers                    (depending on the region) to continue during the 21st century, contributing to increased coastal flooding in low-Current climate in all regions is already distinct from the              lying areas (high confidence) and coastal erosion along climate of the early or mid-20th century with respect to                  most sandy coasts (high confidence). Sea level will continue several climatic impact-drivers (CIDs), resulting in shifting            to rise beyond 2100 (high confidence) (Box TS.4).
magnitude, frequency, duration, seasonality and spatial extent of associated climate indices (high confidence).                  Every region of the world will experience concurrent It is very likely that mean temperatures have increased                  changes in multiple CIDs by mid-century or at 2&deg;C global in all land regions and will continue to increase at rates                warming and above (high confidence). Even for the current greater than the global average (high confidence). The                    climate, climate change-induced shifts in CID distributions frequency of heat and cold extremes have increased and                    and event probabilities, some of which have occurred over decreased, respectively. These changes are attributed to                  recent decades, are relevant for risk assessments. {11.9, human influence in almost all regions (medium to high                    12.1, 12.2, 12.4, 12.5, Atlas.3-Atlas.11}
confidence) and will continue through the 21st century (high confidence). In particular, extreme heat would exceed            An overview of changes in regional CIDs (introduced in Section TS.1) critical thresholds for health, agriculture and other sectors          is given in Table TS.5, which summarizes multiple lines of evidence more frequently by the mid 21st century with 2&deg;C of global            on regional climate change derived from observed trends, attribution warming (high confidence).                                            of these trends and future projections. The level of confidence and 120
 
Technical Summary the amplitude in the projected direction of change in CIDs at a given time horizon depends on climate change mitigation efforts over the 21st century. It is evident from Table TS.5 that many heat, cold, snow and ice, coastal, and oceanic CID changes are projected with high confidence in most regions starting from a global warming level (GWL) of 2&deg;C, indicating worldwide challenges. Changes in many other regional CIDs have higher confidence later in the 21st century or at higher GWLs (high confidence), and another small subset are projected with high confidence for the 1.5&deg;C GWL. This section focuses on the 2&deg;C GWL and mid-century time period because the signal emerges from natural variability for a wider range of CIDs at this higher warming level. Figure TS.22 shows the geographical location of regions belonging to one of five groups characterized by a specific combination of changing CIDs. The Regional Synthesis component of the Interactive Atlas provides comprehensive synthesis                      TS information about changes in all of the individual CIDs across all of the AR6 WGI reference regions. {10.5, Cross-Chapter Box 10.3, 11.1, 11.9, Box 11.1, 12.1, 12.2, 12.4, 12.5}
121
 
TS Table TS.5 l Summary of confidence for climatic impact-driver changes in each AR6 WGI reference region (illustrated in Figure TS.25) across multiple lines of evidence for observed, attributed and projected Technical Summary 122  directional changes. The colours represent their projected aggregate characteristic changes for the mid-21st century, considering scenarios RCP4.5, SSP2-4.5, SRES A1B, or above (RCP6.0, RCP8.5, SSP3-7.0, SSP5-8.5, SRES A2), which approximately encompasses global warming levels of 2.0&deg;C to 2.4&deg;C. Arrows indicate medium to high confidence trends derived from observations, and asterisks indicate medium and high confidence in attribution of observed changes. (North Africa is not an AR6 WGI reference region, but assessment here is based upon the African portion of the Mediterranean reference region). {Tables 12.3-12.11 and Tables 11.4-11.21}
Climatic Impact-driver Heat and Cold                                                                                          Wet and Dry                                                                                                                      Wind                                                                                          Snow and Ice                                                                                          Coastal and Oceanic                                                                                    Other Mean air temperature Extreme heat Cold spell Frost Mean precipitation River flood    Heavy precipitation and Landslide Aridity Hydrological drought Agricultural and ecological drought Fire weather Mean wind speed  Severe wind storm Tropical cyclone Sand and dust storm Snow, glacier and ice sheet Permafrost Lake, river and sea ice Heavy snowfall and ice storm Hail Snow avalanche  Relative sea level Coastal flood Coastal erosion Marine heatwave Ocean and lake acidity  Air pollution weather Atmospheric CO2 at surface Radiation at surface pluvial flood Africa North Africa                                    ***  ***                                                                                                                                                                                                                    3                                                                                                                                                                                                                                4 Sahara                                          **  **                                                                                                                                                                                                                                                                                                                                                                                                                                                          4 Western Africa                                  **  **                                      1                                                                1        1 1                                                                                                                                                                                                                                                                                                                                  4 Central Africa                                                                        1,2                                                                                                                                                                                                                                                                                                                                                                                                                        4 North Eastern Africa                                          **  **                                                                                                            1              1                          1                                                                                                                                                                                                                                                                                                4 South Eastern Africa                                          **                                                                                                                  1              1                          1                                                                                    3                                                                                                                                                                                                          4 West Southern Africa                                          ***  ***                                                                                                                                                                                                                                                                                                                                                                                                                                                    4 East Southern Africa                                          ***  ***                                                                                                                                                                                                                                        3                                                                                                                                                                                                            4,5 Madagascar                                                                                                                                                                                                                                                                                          3                                                                                                                                                                                                            4,5 Note: There are several region-specific qualifiers/exceptions attached to some of the directions of change/confidence levels indicated above. {12.4}
Key for observational trend evidence                                                Past upward trend (medium or higher confidence)                                                                                                              Past downward trend (medium or higher confidence)
Key for attribution evidence                                    *** High confidence (or more)                                                        ** Medium confidence High confidence                                                              Medium confidence                                                                                        Low confidence                                                                                                Medium confidence                                                            High confidence                                                                      Not broadly Key for level of confidence in future changes of increase (or more)                                                        of increase (or more)                                                                                    in direction of change                                                                                        of decrease                                                                  of decrease                                                                          relevant
 
Table TS.5 (continued)
Climatic Impact-driver Heat and Cold                                                                                      Wet and Dry                                                                                                                      Wind                                                                                        Snow and Ice                                                                                          Coastal and Oceanic                                                                                      Other Mean air temperature Extreme heat Cold spell Frost Mean precipitation River flood  Heavy precipitation and Landslide Aridity Hydrological drought Agricultural and ecological drought Fire weather Mean wind speed  Severe wind storm Tropical cyclone Sand and dust storm Snow, glacier and ice sheet Permafrost Lake, river and sea ice Heavy snowfall and ice storm Hail Snow avalanche  Relative sea level Coastal flood Coastal erosion Marine heatwave Ocean and lake acidity  Air pollution weather Atmospheric CO2 at surface Radiation at surface pluvial flood Asia Arabian Peninsula                              ***  **                                                                                                                                                                                                                                                                                                                                                                                                                                                    1 West Central Asia                              ***  ***                                  5                                                                                                                                                                                                                                                                                                                                                                                                            1,2 West Siberia                                  ***  ***
East Siberia                                  ***  ***
Russian Far East                              ***  ***                                                                                                                                                                                                                                                                                                                                                                                                                                                1,2 East Asia                                      ***  ***                                                                                                                                                                                                                          3                                                                                                                                                                                                                  1,2 East Central Asia                              ***  ***
Tibetan Plateau                                ***  ***
South Asia                                    ***  ***                                                                                                                                                                                                                                                                                                                                                                                                                                              1 South East Asia                                ***  ***                                  4                                                                                                                                                                                          3                                                                                                                                                                                                                  1,2 Technical Summary Note: There are several region-specific qualifiers/exceptions attached to some of the directions of change/confidence levels indicated above. {12.4}
Key for observational trend evidence                                              Past upward trend (medium or higher confidence)                                                                                                          Past downward trend (medium or higher confidence)
Key for attribution evidence      *** High confidence (or more)        ** Medium confidence High confidence                Medium confidence                                                                                                                                                                                                Low confidence                                                                                  Medium confidence                                                                            High confidence                                                                        Not broadly Key for level of confidence in future changes of increase (or more)          of increase (or more)                                                                                                                                                                                            in direction of change                                                                          of decrease                                                                                  of decrease                                                                            relevant 123 TS
 
TS Table TS.5 (continued)
Technical Summary 124 Climatic Impact-driver Heat and Cold                                                                                      Wet and Dry                                                                                                                        Wind                                                                                        Snow and Ice                                                                                          Coastal and Oceanic                                                                                      Other Mean air temperature Extreme heat Cold spell Frost Mean precipitation River flood  Heavy precipitation and Landslide Aridity Hydrological drought Agricultural and ecological drought Fire weather Mean wind speed  Severe wind storm Tropical cyclone Sand and dust storm Snow, glacier and ice sheet Permafrost Lake, river and sea ice Heavy snowfall and ice storm Hail Snow avalanche  Relative sea level Coastal flood Coastal erosion Marine heatwave Ocean and lake acidity  Air pollution weather Atmospheric CO2 at surface Radiation at surface pluvial flood Australasia Northern Australia                              ***  ***                                                                                                                                                                                                                                5                                                                                                                                                                                                                  7 Central Australia                              ***  ***                                                                                                                                                                                                                                                                                                                                                                                                                                                      7 Eastern Australia                              ***  ***                                                                                                                                                                                                                                                                                                                                                                                                                                                      7 Southern Australia                              ***  ***                                  1                                                                3                                                                    **                  7                                                                                                                                                                                                                                                  7 New Zealand                                                    **                          2                                                                  4                                                                                          8                                                                      6                                                                                                                                                                          7 Note: There are several region-specific qualifiers/exceptions attached to some of the directions of change/confidence levels indicated above. {12.4}
Key for observational trend evidence                                                Past upward trend (medium or higher confidence)                                                                                                            Past downward trend (medium or higher confidence)
Key for attribution evidence                                    *** High confidence (or more)                                                      ** Medium confidence High confidence                                                                  Medium confidence                                                                                          Low confidence                                                                                    Medium confidence                                                                            High confidence                                                                        Not broadly Key for level of confidence in future changes of increase (or more)                                                            of increase (or more)                                                                                      in direction of change                                                                            of decrease                                                                                  of decrease                                                                            relevant
 
Table TS.5 (continued)
Climatic Impact-driver Heat and Cold                                                                                        Wet and Dry                                                                                                                      Wind                                                                                          Snow and Ice                                                                                          Coastal and Oceanic                                                                                    Other Mean air temperature Extreme heat Cold spell Frost Mean precipitation River flood  Heavy precipitation and Landslide Aridity Hydrological drought Agricultural and ecological drought Fire weather Mean wind speed  Severe wind storm Tropical cyclone Sand and dust storm Snow, glacier and ice sheet Permafrost Lake, river and sea ice Heavy snowfall and ice storm Hail Snow avalanche  Relative sea level Coastal flood Coastal erosion Marine heatwave Ocean and lake acidity  Air pollution weather Atmospheric CO2 at surface Radiation at surface pluvial flood Central and South America Southern Central America                                        **  **                                                                                                                                                                                                                                            2                                                                                                                                                                                                              3 North-Western South America                                  ***  ***                                                                                                                                                                                                                                                                                                                                                                                                                                                        3,4 Northern South America                                        **  **                                                                                                                                                                                                                                            2                                                                                                                                                                                                              3,4 South American Monsoon                                        **  **                                                  1 North-Eastern South America                                  **  **                                                                                                                                                                                                                                                                                                                                                                                                                                                        3,4 South-Western South America                                  **  **                                                                                                                                                                                                                                                                                                                                                                                                                                                        3 South-Eastern South America                                  ***  ***                                                                                                                                                                                                                                                                                                                                                                                                                                                    3 Southern South America                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          3 Note: There are several region-specific qualifiers/exceptions attached to some of the directions of change/confidence levels indicated above. {12.4}
Key for observational trend evidence                                                Past upward trend (medium or higher confidence)                                                                                                          Past downward trend (medium or higher confidence)
Key for attribution evidence                                    *** High confidence (or more)                                                      ** Medium confidence Technical Summary High confidence                                                              Medium confidence                                                                                            Low confidence                                                                                        Medium confidence                                                                    High confidence                                                                      Not broadly Key for level of confidence in future changes of increase (or more)                                                        of increase (or more)                                                                                        in direction of change                                                                                of decrease                                                                          of decrease                                                                          relevant 125 TS
 
TS Table TS.5 (continued)
Technical Summary 126 Climatic Impact-driver Heat and Cold                                                                                        Wet and Dry                                                                                                                        Wind                                                                                        Snow and Ice                                                                                          Coastal and Oceanic                                                                                    Other Mean air temperature Extreme heat Cold spell Frost Mean precipitation River flood    Heavy precipitation and Landslide Aridity Hydrological drought Agricultural and ecological drought Fire weather Mean wind speed  Severe wind storm Tropical cyclone Sand and dust storm Snow, glacier and ice sheet Permafrost Lake, river and sea ice Heavy snowfall and ice storm Hail Snow avalanche  Relative sea level Coastal flood Coastal erosion Marine heatwave Ocean and lake acidity  Air pollution weather Atmospheric CO2 at surface Radiation at surface pluvial flood Europe Mediterranean                                  ***  ***                                                                        5                                      **  **                                                                      6                        7                                                                                                                                                                                                                              2 Western and Central Europe                                          ***  ***                                                                                          4                                                                                                                                                                                                                                                                                                                                                      2 Eastern Europe                                  ***  ***
Northern Europe                                ***  ***                                              1              ***                                                                                                                                                                                                                                                                                                                                                                8                2,3 North America North Central America                                        **  **                                                                                                                                                                                                                                                                                                                                                                                                                                                        2 Western North                                                                                                                                                                                                    6,7 America                                        **  **                                      3                                    5                  5          4,7                            **  6.7                                                                      8                              6                    1                                                                                1                              1                              5                      2 Central North America                                                                                                                  **                                      7                                            7                              7                                8                                        4                                                                                                                                                                                  2 Eastern North America                                                                              5                                                                                                                                                        7                                8                                                    1                                                                                1                              1                                                    2 North-Eastern North America                                  ***  ***                                  5                                                                    5                                            6,7                            6,7                                8                                                      1,6                                                                                                            1                  4            4,6                  2,6 North-Western North America                                  ***  ***                                  5                                                        6          5                                            6,7                      6,7                                    8                                                    1                                                                                                              1,6      9                                          2 Note: There are several region-specific qualifiers/exceptions attached to some of the directions of change/confidence levels indicated above. {12.4}
Key for observational trend evidence                                                Past upward trend (medium or higher confidence)                                                                                                              Past downward trend (medium or higher confidence)
Key for attribution evidence                                    *** High confidence (or more)                                                      ** Medium confidence High confidence                                                                        Medium confidence                                                                                Low confidence                                                                                                      Medium confidence                                                      High confidence                                                                      Not broadly Key for level of confidence in future changes of increase (or more)                                                                  of increase (or more)                                                                            in direction of change                                                                                              of decrease                                                            of decrease                                                                          relevant
 
Table TS.5 (continued)
Climatic Impact-driver Heat and Cold                                                                                      Wet and Dry                                                                                                                      Wind                                                                                          Snow and Ice                                                                                          Coastal and Oceanic                                                                                    Other Mean air temperature Extreme heat Cold spell Frost Mean precipitation River flood    Heavy precipitation and Landslide Aridity Hydrological drought Agricultural and ecological drought Fire weather Mean wind speed  Severe wind storm Tropical cyclone Sand and dust storm Snow, glacier and ice sheet Permafrost Lake, river and sea ice Heavy snowfall and ice storm Hail Snow avalanche  Relative sea level Coastal flood Coastal erosion Marine heatwave Ocean and lake acidity  Air pollution weather Atmospheric CO2 at surface Radiation at surface pluvial flood Small Islands Caribbean                                      **                                                                                                                                                                                                                                                5                                                                                                                                                                                                            6 Pacific                                        **1                                        2                                    3                              4                                                                                                                                5                                                                                                                                                                                                            6 Polar Terrestrial Regions Greenland and Iceland                                        **  **                                                                                                    3                                                                          2,3                                                                                      1                                                                                                                          5 Arctic North Europe                                                                                                                                                      3                                                                          2,3                                                                                      1                                                                                                                          6                                            7 Russian Arctic                                  **  **                                                                                                      3                                                                          2,3                                                                                      1,4                                                                                                                                                                      7 Arctic North-Western North America                                                                                                                                                3                                                                          2,3                                                                                      1,4                                                                                                                                                                    7 Arctic North-East North America                                                                                                                                                3                                                                          2,3                                                                                      1,4 West Antarctica                                                                                                                                                                                                                                                                                                                        1,4 East Antarctica Technical Summary Note: There are several region-specific qualifiers/exceptions attached to some of the directions of change/confidence levels indicated above. {12.4}
Key for observational trend evidence                                                                         Past upward trend (medium or higher confidence)                                                                                                          Past downward trend (medium or higher confidence)
Key for attribution evidence                                                                               *** High confidence (or more)                                                                                ** Medium confidence Key for level of confidence in future changes                                                                            High confidence                                                                        Medium confidence                                                                              Low confidence                                                                                                        Medium confidence                                                    High confidence                                                                        Not broadly 127                                                                                                                            of increase (or more)                                                                  of increase (or more)                                                                          in direction of change                                                                                                of decrease                                                          of decrease                                                                            relevant TS
 
TS Table TS.5 (continued)
Technical Summary 128 Climatic Impact-driver Mean ocean temperature Marine heatwave Ocean acidity  Ocean salinity Dissolved oxygen Sea ice Oceans Arctic Ocean                                                                                                                                              ***
South Pacific Ocean Equatorial Pacific Ocean North Pacific Ocean South Atlantic Ocean Equatorial Atlantic Ocean North Atlantic Ocean Equatorial Indian Ocean South Indian Ocean Arabian Sea Bay of Bengal Southern Ocean Note: There are several region-specific qualifiers/exceptions attached to some of the directions of change/confidence levels indicated above. {12.4}
Key for observational trend evidence                                         Past upward trend (medium or higher confidence)                                        Past downward trend (medium or higher confidence)
Key for attribution evidence                                               *** High confidence (or more)                          ** Medium confidence High confidence                          Medium confidence                          Low confidence                Medium confidence  High confidence  Not broadly Key for level of confidence in future changes of increase (or more)                    of increase (or more)                      in direction of change        of decrease        of decrease      relevant
 
Technical Summary Notes:
Africa (projections)
: 1. Contrasted regional signal: drying in western portions and wetting in eastern portions
: 2. Likely increase over the Ethiopian Highlands
: 3. Medium confidence of decrease in frequency and increase in intensity
: 4. Along sandy coasts and in the absence of sufficient sediment supply from terrestrial or offshore sources
: 5. Substantial parts of the East Southern Africa and Madagascar coast are projected to prograde if present-day ambient shoreline change rates continue Asia (projections)
: 1. Along sandy coasts and in the absence of additional sediment sinks/sources or any physical barriers to shoreline retreat.
: 2. Substantial parts of the coasts in these regions are projected to prograde if present-day ambient shoreline change rates continue
: 3. Tropical cyclones decrease in number but increase in intensity
: 4. High confidence of decrease in Indonesia (Atlas.5.4.5)
: 5. Medium confidence of decreasing in summer and increasing in winter Australasia (projections)
: 1. High confidence of decrease in the south-west of the state of Western Australia
: 2. Medium confidence of decrease in north and east and increase in south and west                                                                                          TS
: 3. High confidence of increase in the south-west of the state of Western Australia
: 4. Medium confidence of increase in the north and east and decrease in south and west
: 5. Low confidence of increasing intensity, and high confidence of decreasing occurrence
: 6. High confidence of decrease in glacier volume, medium confidence of decrease in snow
: 7. Along sandy coasts and in the absence of additional sediment sinks/sources or any physical barriers to shoreline retreat Central and South America (projections)
: 1. Increase in extreme flow in the Amazon basin
: 2. Tropical cyclones decrease in number but increase in intensity
: 3. Along sandy coasts and in the absence of additional sediment sinks/sources or any physical barriers to shoreline retreat.
: 4. Substantial parts of the North-Western South America, Northern South America and North-Eastern South America coasts are projected to prograde if present-day ambient shoreline change rates continue Europe (projections)
: 1. Excluding southern United Kingdom
: 2. Along sandy coasts and in the absence of additional sediment sinks/sources or any physical barriers to shoreline retreat
: 3. The Baltic Sea shoreline is projected prograde if present-day ambient shoreline change rates continue.
: 4. For the Alps, conditions conducive to landslides are expected to increase
: 5. Low confidence of decrease in the southernmost part of the region
: 6. General decrease except in Aegean Sea
: 7. Medium confidence of decrease in frequency and increase in intensities
: 8. Except in the Northern Baltic Sea region North America (projections)
: 1. Snow may increase in some high elevations and during the cold season and decrease in other seasons and at lower elevations
: 2. Along sandy coasts and in the absence of additional sediment sinks/sources or any physical barriers to shoreline retreat.
: 3. Increasing in northern regions and decreasing toward the south
: 4. Decreasing in northern regions and increasing toward the south
: 5. Higher confidence in northern regions and lower toward the south
: 6. Higher confidence in southern regions and lower toward the north
: 7. Higher confidence in increase for some climatic impact-driver indices during summertime
: 8. Increase in convective conditions but decrease in winter extratropical cyclones
: 9. Relative sea level rise reduced given land uplift in Southern Alaska Small Islands (projections)
: 1. Very high confidence in the direction of change, but low to medium confidence in the magnitude of change due to model uncertainty
: 2. Decrease in eastern Pacific and southern Pacific subtropics, but increase in parts of western and equatorial Pacific; with seasonal variation in future changes
: 3. High confidence in increase in extreme rain frequency and intensity in western tropical Pacific; low confidence in magnitude of change due to model bias
: 4. Increase in southern Pacific
: 5. Increase in intensity; decrease in frequency except over central North Pacific.
: 6. Along sandy coasts and in the absence of additional sediment sinks/sources or any physical barriers to shoreline retreat.
Polar Terrestrial Regions (projections)
: 1. Snow may increase in some high elevations and during the cold season and decrease in other seasons and at lower elevations
: 2. Higher confidence in southern regions and lower toward the north
: 3. Higher confidence in increase for some climatic impact-driver indices during summertime
: 4. Glaciers decline even as some regional snow climatic impact-driver indices increase
: 5. Decreasing in west and increasing in east
: 6. Except for Northern Baltic Sea coasts where relative sea levels fall
: 7. Along sandy coasts and in the absence of additional sediment sinks/sources or any physical barriers to shoreline retreat 129
 
Technical Summary While changes in climatic impact-drivers are projected everywhere, there is a specific combination of changes each region would experience (a) World regions grouped into five clusters, each one based on a combination of changes in climatic impact-drivers Assessed future changes: Changes refer to a 20-30 year period centred around 2050 and/or consistent with 2&deg;C global warming compared to a similar period within 1960-2014 or 1850-1900.
GIC                            RAR                                            1) Hotter and drier NWN            NEN                        NEU EEU  WSB        ESB    RFE                        2) Hotter and drier and in WCE                                                        some regions wetter WNA              ENA                                                  ECA                                    extremes CNA                                  MED          WCA TIB    EAS PAC                  NCA                                                                                                      3) Hotter and wetter SAH          ARP      SAS                                      extremes and in some CAR SCA                                                                                                regions more precipitation or WAF            NEAF TS                                            NSA                                                      SEA                            fire weather CAF                                          PAC NWS                                      SEAF NES                                                                                4) Hotter and wetter and in PAC                              SAM                              ESAF                          NAU                      some regions more flooding MDG                    CAU SES                    WSAF                                        EAU SWS
: 5) Hotter and in some SAU        NZ              regions wetter extremes or more precipitation SSA All coastal regions except North-East North America (NEN) and                                      6) Increase in Tropical Greenland/Iceland (GIC) will be exposed to at least two among                                      cyclones intensity or Severe increases in relative sea level, coastal flood and coastal erosion                                winds Combinations of future changes in climatic impact-drivers (CIDs)
LEGEND Heat increase in mean temperature, extreme heat CIDs changing in all the regions of the cluster with high confidence Cold decrease in cold spell, frost CIDs changing in some (1)                                            (2)                                            regions of the cluster with                Drought Hotter                                        Hotter and more fire weather                    high and sometimes with      increase in aridity, hydrological and in some regions more                      and in some regions more pluvial                medium confidence                            and fire weather or drought or                    flooding or drought or less mean                                                    agricultural drought both                                          precipitation or snow and ice or combinations of these Snow and Ice decrease in snow, glacier and ice sheet Mean            Mean precipitation precipitation decrease        increase River            Pluvial flooding          flooding increase          increase (3)                                            (4)                                        (5)
Hotter, less snow and ice and                  Hotter and less snow/ice                  Hotter more pluvial flooding                          and in some regions more                  and in some regions more and in some regions more                        pluvial flooding or river                  pluvial flooding or mean                        Fire precipitation or fire weather or                flooding or mean precipita-                precipitation or both                        weather both                                            tion or both                                                                            increase 130
 
Technical Summary (b) Number of land & coastal regions (i) and open-ocean regions (ii) where each climatic impact-driver (CID) is projected to increase or decrease with high confidence (dark shade) or medium confidence (light shade)
(i)                                                                                                                                                                                      (ii)
Heat & Cold                        Wet & Dry                    Wind              Snow & Ice        Other                            Coastal                                                            Open Ocean Extreme heat
& COASTAL REGIONS Cold spell Sand and dust storm NUMBER OF OPEN-OCEAN REGIONS Frost Mean surface temperature Snow, glacier and ice sheet Mean ocean temperature Mean precipitation Permafrost                                                                                                                                                    TS River flood Radiation at surface Lake, river and sea ice Heavy precipitation and pluvial flood  Fire weather        Heavy snowfall and ice storm                            Relative sea level                                                          Marine heatwave Landslide                              Mean wind speed                                                            Coastal flood Hail                                                                                                                                Ocean acidity NUMBER OF LAND Aridity                                                    Snow avalanche                                          Coastal erosion Hydrological drought                    Severe wind storm                                                                                                                                        Ocean salinity Air pollution weather                                  Marine heatwave Agricultural and ecological drought    Tropical cyclone    Atmospheric CO2 at surface                              Ocean acidity                                                                Dissolved oxygen 55 45 35 25 15                                                                                                                                                                                          15 5                                                                                                                                                                                              5 5                                                                                                                                                                                              5 15                                                                                                                                                                                          15 25 35 45 55 BAR CHART LEGEND                                                          LIGHTER-SHADED ENVELOPE LEGEND Regions with high confidence increase              The height of the lighter shaded envelope behind each bar represents the Regions with medium confidence increase            maximum number of regions for which each CID is relevant. The envelope Regions with high confidence decrease              is symmetrical about the x-axis showing the maximum possible number of Regions with medium confidence decrease            relevant regions for CID increase (upper part) or decrease (lower part).
Climatic impact-drivers (CIDs) are physical climate system conditions (e.g., means, events, extremes) that affect an element of society or ecosystems. Depending on system tolerance, CIDs and their changes can be detrimental, beneficial, neutral, or a mixture of each across interacting system elements and regions. The CIDs are grouped into seven types, which are summarized under the icons in sub-panels (i) and (ii). All regions are projected to experience changes in at least 5 CIDs. Almost all (96%) are projected to experience changes in at least 10 CIDs and half in at least 15 CIDs. For many CID changes, there is wide geographical variation, and so each region is projected to experience a specific set of CID changes. Each bar in the chart                                                            interactive-atlas.ipcc.ch represents a specific geographical set of changes that can be explored in the WGI Interactive Atlas.
Figure TS.22 l Synthesis of the geographical distribution of climatic impact-drivers changes and the number of AR6 WGI reference regions where they are projected to change.
131
 
Technical Summary Figure TS.22 (continued): Panel (a) shows the geographical location of regions belonging to one of five groups characterized by a specific combination of changing climatic impact-drivers (CIDs). The five groups are represented by the five different colours, and the CID combinations associated with each group are represented in the corresponding fingerprint and text below the map. Each fingerprint comprises a set of CIDs projected to change with high confidence in every region in the group and a second set of CIDs, one or more of which are projected to change in each region with high or medium confidence. The CID combinations follow a progression from those becoming hotter and drier (group 1) to those becoming hotter and wetter (group 5). In between (groups 2-4), the CIDs that change include some becoming drier and some wetter and always include a set of CIDs which are getting hotter. Tropical cyclones and severe wind CID changes are represented on the map with black dots in the regions affected. Regions affected by coastal CID changes are described by text on the map. The five groups are chosen to provide a reasonable level of detail for each region while not overwhelming the map with a full summary of all aspects of the assessment, which is available in Table TS.5 and can be visualized in the Regional Synthesis component of the Interactive Atlas. The CID changes summarized in the figure represent high and medium confidence changes for the mid-21st century, considering scenarios SSP2-4.5, RCP4.5, SRES A1B, or above (SSP3-7.0, SSP5-8.5, RCP6.0, RCP8.5, SRES A2), which approximately encompasses global warming levels of 2.0&deg;C to 2.4&deg;C.
The bar chart in panel (b) shows the numbers of regions where each CID is increasing or decreasing with medium or high confidence for all land regions and ocean regions listed in Table TS.5. The colours represent the direction of change and the level of confidence in the change: purple indicates an increase while brown indicates a decrease; darker and lighter shades refer to high and medium confidence, respectively. Lighter background colours represent the maximum number of regions for which each CID is broadly relevant. Sub-panel (i) shows the 30 CIDs relevant to the land and coastal regions while sub-panel (ii) shows the 5 CIDs relevant to the open ocean regions. Marine heatwaves and ocean acidity are assessed for coastal ocean regions in panel (i) and for open ocean regions in panel (ii). Changes refer to a 20- to 30-year period centred around 2050 and/or consistent with 2&deg;C global warming compared to a similar period within 1960-2014, except for hydrological drought and agricultural and ecological drought, which is compared to 1850-1900. Definitions of the regions are provided in Atlas.1, the Interactive Atlas (https://interactive-atlas.ipcc.ch/) and Chapter 12. (Table TS.5, Figure TS.24)
{11.9, 12.2, 12.4, Atlas.1}
TS TS.4.3.1 Common Regional Changes in Climatic Impact-Drivers                                  regions (high confidence). In many tropical regions, the number of days per year where a heat index of 41&deg;C is exceeded would increase Heat and cold: Changes in temperature-related CIDs such as mean                              by more than 100 days relative to the recent past under SSP5-8.5, temperatures, growing season length, and extreme heat and frost have                          while this increase will be limited to less than 50 days under SSP1-2.6 already occurred (high confidence), and many of these changes have                            (high confidence) (Figure TS.6). The number of days per year where been attributed to human activities (medium confidence). Over all                            temperature exceeds 35&deg;C would increase by more than 150 days in land regions with sufficient data (i.e., all except Antarctica), observed                    many tropical areas, such as the Amazon basin and South East Asia, changes in temperature have already clearly emerged outside the                              by the end of century for the SSP5-8.5 scenario, while it is expected to range of internal variability, relative to 1850-1900 (Figure TS.23). In                      increase by less than 60 days in these areas under SSP1-2.6 (except tropical regions, recent past temperature distributions have already                          for the Amazon Basin) (high confidence) (Figure TS.24). {4.6.1, 11.3, shifted to a range different to that of the early 20th century (high                          11.9, 12.4, 12.5.2, Atlas}
confidence) (Section TS.1.2.4). Most land areas have very likely warmed by at least 0.1&deg;C per decade since 1960, and faster in recent                          Wet and dry: Compared to the global scale, precipitation internal decades. On regional-to-continental scales, trends of increased                              variability is stronger at the regional scale while uncertainties in frequency of hot extremes and decreased frequency of cold extremes                            observations, models and external forcing are all larger. However, are generally consistent with the global-scale trends in mean                                GHG forcing has driven increased contrasts in precipitation amounts temperature (high confidence). In a few regions, trends are difficult                        between wet and dry seasons and weather regimes over tropical land to assess due to limited data availability. {2.3.1.1, 11.3, 11.9, 12.4,                      areas (medium confidence), with a detectable precipitation increase Atlas.3.1}                                                                                    in the northern high latitudes (high confidence) (Box TS.6). The frequency and intensity of heavy precipitation events have increased Warming trends observed in recent decades are projected to continue                          over a majority of land regions with good observational coverage over the 21st century and over most land regions at a rate higher                            (high confidence). A majority of land areas have experienced than the global average (high confidence). For given global warming                          decreases in available water in dry seasons due to human-induced levels, model projections from CMIP6 show future regional warming                            climate change associated with changes in evapotranspiration changes that are similar to those projected by CMIP5. However,                                (medium confidence). Global hydrological models project a larger projected regional warming in CMIP6 for given time periods and                                fraction of land areas to be affected by an increase rather than by emissions scenarios has a wider range with a higher upper limit                              a decrease in river floods (medium confidence). Extreme precipitation compared to CMIP5 because of the higher climate sensitivity in some                          and pluvial flooding will increase in many regions around the world CMIP6 models and differences in the forcings. {Atlas.3-Atlas.11}                              on almost all continents (high confidence), but regional changes in river floods are more uncertain than changes in pluvial floods Under RCP8.5/SSP5-8.5, it is likely that most land areas will experience                      because complex hydrological processes, including land cover and further warming of at least 4&deg;C compared to a 1995-2014 baseline                              human water management, are involved. {8.2.2.1, 8.3.1, Box 8.2, by the end of the 21st century, and in some areas significantly more.                        10.4.1, 11.5, 11.6, 11.9, 12.4, 12.5.1, Atlas.3.1}
At increasing warming levels, extreme heat will exceed critical thresholds for health, agriculture and other sectors more frequently                          Wind: Mean wind speed has decreased over most land areas with (high confidence), and it is likely that cold spells will become less                        good observational coverage (medium confidence). It is likely that frequent towards the end of the century. For example, by the end                              the global proportion of major tropical cyclone (TC) intensities of the 21st century, dangerous humid heat thresholds, such as the                            (Categories 3-5) over the past four decades has increased. The National Oceanic and Atmospheric Administration (NOAA) heat index                            proportion of intense TCs, average peak TC wind speeds, and peak (HI) threshold of 41&deg;C, will be exceeded much more frequently under                          wind speeds of the most intense TCs will increase on the global scale the SSP5-8.5 scenario than under SSP1-2.6 and will affect many                                with increasing global warming (high confidence). {11.7.1}
132
 
Technical Summary Year of signi"cant emergence of changes in temperature over land regions (S/N>2)
TS Dataset: Berkeley Earth. Temperature changes relative to 1850-1900.
Before 1981                1981-1988                    1989-1996                  1997-2004                  2005-2012                    2013-2020 Year of signi"cant emergence of changes in temperature over land regions (S/N>2)
Dataset: CRUTEM5. Temperature changes relative to 1850-1900. Grey: not enough data.
Before 1981                1981-1988                    1989-1996                  1997-2004                  2005-2012                    2013-2020 Figure TS.23 l Time period during which the signals of temperature change in observed data aggregated over the reference regions emerged from the noise of annual variability in the respective aggregated data, using a signal-to-noise ratio of two as the threshold for emergence. The intent of this figure is to show, for the AR6 WGI reference regions, when a signal of annual mean surface temperature change emerged from the noise of annual variability in two global datasets and thus also provide some information on observational uncertainty. Emergence time is calculated for two global observational datasets: (a) Berkeley Earth and (b) CRUTEM5.
Regions in the CRUTEM5 map are shaded grey when data are available over less than 50% of the area of the region. (Section TS.1.2.4) {Figure Atlas.11}
133
 
Technical Summary CMIP5                                              CMIP6                                                    CORDEX (a) TX35 for 20412060 (RCP2.6) rel. to 19952014 (e) TX35 for 20412060 (SSP1-2.6) rel. to 19952014      (i) TX35 for 20412060 (SSP1-2.6) rel. to 19952014 (b) TX35 for 20812100 (RCP2.6) rel. to 19952014 (f) TX35 for 20812100 (SSP1-2.6) rel. to 19952014      (j) TX35 for 20812100 (SSP1-2.6) rel. to 19952014 TS (c) TX35 for 20412060 (RCP8.5) rel. to 19952014 (g) TX35 for 20412060 (SSP5-8.5) rel. to 19952014      (k) TX35 for 20412060 (SSP5-8.5) rel. to 19952014 (d) TX35 for 20812100 (RCP8.5) rel. to 19952014 (h) TX35 for 20812100 (SSP5-8.5) rel. to 19952014      (l) TX35 for 20812100 (SSP5-8.5) rel. to 19952014 Change (days)                      Colour High model agreement                      No data Low model agreement
                                                                    -25 0      50    100  150      200    250 Figure TS.24 l Projected change in the mean number of days per year with maximum temperature exceeding 35&deg;C for Coupled Model Intercomparison Project Phase 5 (CMIP5; first column), Phase 6 (CMIP6; second column) and Coordinated Regional Climate Downscaling Experiment (CORDEX; third column) ensembles. The intent of this figure is to show that there is a consistent message about the patterns of projected change in extreme daily temperatures from the CMIP5, CMIP6 and CORDEX ensembles. The map shows the median change in the number of days per year between the mid-century (2041-2060) or end-century (2081-2100) and historical (1995-2014) periods for the CMIP5 and CORDEX RCP8.5 and RCP2.6 scenario ensembles and the CMIP6 SSP5-8.5 and SSP1-2.6 scenario ensembles. Hatching indicates areas where less than 80% of the models agree on the sign of change. {Interactive Atlas}
Snow and ice: Many aspects of the cryosphere either have seen                            Coastal and oceanic: There is high confidence that SST will significant changes in the recent past or will see them during the                        increase in all oceanic regions except the North Atlantic. Regional 21st century (high confidence). Glaciers will continue to shrink                          sea level change has been the main driver of changes in extreme and permafrost to thaw in all regions where they are present (high                        sea levels across the quasi-global tide gauge network over the confidence). Also, it is virtually certain that snow cover will experience                20th century (high confidence). With the exception of a few regions a decline over most land regions during the 21st century, in terms                        with substantial land uplift, relative sea level rise is very likely to of water equivalent, extent and annual duration. There is high                            virtually certain (depending on the region) to continue during the confidence that the global warming-induced earlier onset of spring                        21st century, contributing to increased coastal flooding in low-lying snowmelt and increased melting of glaciers have already contributed                      areas (high confidence) and coastal erosion along most sandy to seasonal changes in streamflow in high-latitude and low-elevation                      coasts (high confidence) over the 21st century. In the open ocean, mountain catchments. Nevertheless, it is very likely that some high-                      acidification, changes in sea ice, and deoxygenation have already latitude regions will experience an increase in winter snow water                        emerged in many areas (high confidence). Marine heatwaves are equivalent due to the effect of increased snowfall prevailing over                        also expected to increase around the globe over the 21st century warming-induced increased snowmelt. (Section TS.2.5) {8.2.2.1,                            (high confidence). (Section TS.2.4) {Box 9.2, 9.2.1.1, 9.6, 9.6.4, 8.3.1, Box 8.2, 9.4, 9.5.1, 9.5.2, 12.4, Atlas.4-Atlas.9, Atlas.11}                      9.6.4.2, 12.4}
134
 
Technical Summary Other variables and concurrent CID changes: It is virtually certain    In addition to the main changes summarized above and in that atmospheric CO2 and oceanic pH will increase in all climate        Section TS.4.3.1, additional details per CID are given below.
scenarios, until net zero CO2 emissions are achieved (Section TS.2.2).
In nearly all regions, there is low confidence in changes in hail, ice  Heat and cold: Observed and projected increases in mean storms, severe storms, dust storms, heavy snowfall, and avalanches,    temperature and a shift toward heat extreme characteristics are although this does not indicate that these CIDs will not be affected    broadly similar to the generic pattern described in Section TS.4.3.1.
by climate change. For such CIDs, observations are often short-        {2.3.1.1.2, 11.3, 11.9, 12.4.1.1, Atlas.4.2, Atlas.4.4}
term or lack homogeneity, and models often do not have sufficient resolution or accurate parametrizations to adequately simulate them    Wet and dry: Mean precipitation changes have been observed over climate change time scales. The probability of compound events    over Africa, but the historical trends are not spatially coherent (high has increased in the past due to human-induced climate change and      confidence). North Eastern Africa, East Southern Africa and Central will likely continue to increase with further global warming, including Africa have experienced a decline in rainfall since about 1980 and for concurrent heatwaves and droughts, compound flooding, and the      parts of West Africa an increase (high confidence). Increases in the possibility of connected sectors experiencing multiple regional extreme frequency and/or the intensity of heavy rainfall have been observed in events at the same time (for example, in multiple breadbaskets) (high  East and West Southern Africa, and the eastern Mediterranean region    TS confidence). {5.3.4.2, 11.8, Box 11.3, Box 11.4, 12.4}                  (medium confidence). Increasing trends in river flood occurrence can be identified beyond 1980 in East and West Southern Africa (medium TS.4.3.2 Region-by-Region Changes in Climatic Impact-Drivers            confidence) and Western Africa (high confidence). However, Northern Africa and West Southern Africa are likely to have a reduction in This section provides a continental synthesis of changes in CIDs,      precipitation. Over West Africa, rainfall is projected to decrease some examples of which are presented in Figure TS.25.                  in the western Sahel subregion and increase along the Guinea Coast subregion (medium confidence). Rainfall is projected to increase over With 2&deg;C global warming, and as early as the mid-21st                Eastern Africa (medium confidence). {8.3.1.6, 11.4, 11.9, 12.4.1.2, century, a wide range of CIDs, particularly related to the          Atlas.4.2, Atlas.4.4, Atlas.4.5}
water cycle and storms, are expected to show simultaneous region-specific changes relative to the recent past with high        Precipitation declines and aridity trends in Western Africa, Central or medium confidence. In a number of regions (Southern              Africa, Southern Africa and the Mediterranean co-occur with Africa, the Mediterranean, North Central America, Western            trends towards increased agricultural and ecological droughts in North America, the Amazon regions, South-Western South              the same regions (medium confidence). Trends towards increased America, and Australia), increases in one or more of drought,        hydrological droughts have been observed in the Mediterranean aridity and fire weather (high confidence) will affect a wide        (high confidence) and Western Africa (medium confidence). These range of sectors, including agriculture, forestry, health and        trends correspond with projected regional increases in aridity and ecosystems. In another group of regions (North-Western,              fire weather conditions (high confidence). {8.3.1.6, 8.4.1.6, 11.6, Central and Eastern North America, Arctic regions, North-            11.9, 12.4.1.2}
Western South America, Northern, Western and Central and Eastern Europe, Siberia, Central, South and East Asia,          Wind: Mean wind, extreme winds and the wind energy potential in Southern Australia and New Zealand), decreases in snow              North Africa and the Mediterranean are projected to decrease across and/or ice or increases in pluvial/river flooding (high              all scenarios (high confidence). Over Western Africa and Southern confidence) will affect sectors such as winter tourism,              Africa, a future significant increase in wind speed and wind energy energy production, river transportation and infrastructure.          potential is projected (medium confidence). There is a projected
{11.9, 12.3, 12.4, 12.5, Table 12.2}                                decrease in the frequency of tropical cyclones making landfall over Madagascar, East Southern Africa and East Africa (medium TS.4.3.2.1 Africa                                                      confidence). {12.4.1.3}
Additional regional changes in Africa, besides those                Snow and ice: There is high confidence that African glaciers and described in Section TS.4.3.1, include a projected decrease          snow have very significantly decreased in the last decades and that in total precipitation in the northernmost and southernmost          this trend will continue in the 21st century. {12.4.1.4}
regions (high confidence), with Western Africa having a west-to-east pattern of decreasing-to-increasing precipitation            Coastal and oceanic: Relative sea level has increased at a higher (medium confidence). Increases in heavy precipitation that          rate than GMSL around Africa over the last 3 decades. The present can lead to pluvial floods (high confidence) are projected          day 1-in-100-year extreme total water level (ETWL) is between for most African regions, even as increasing dry CIDs                0.1 m and 1.2 m around Africa, with values around 1 m or above (aridity; hydrological, agricultural and ecological droughts;        along the East and West Southern and Central Eastern Africa coasts.
fire weather) are projected in the western part of Western          Satellite-derived shoreline retreat rates up to 1 m yr -1 have been Africa, Southern Africa and Northern Africa and the                  observed around the continent from 1984 to 2015, except in South Mediterranean regions (medium to high confidence). {8.4,            Eastern Africa, which has experienced a shoreline progradation 11.3, 11.6, 11.9, 12.4, Atlas.4}                                    (growth) rate of 0.1 m yr -1 over the same period. {12.4.1.5}
135
 
Technical Summary (a)                                                                    Change in number of days with SWE>100 mm NWN                                NEN                                  NWA                            CNA 100 Days with SWE>100 mm (days yr-1) 80 60 40 20 0
ENA 100 80 60 TS                                                  40                                                                                    NWN                              NEN 20 0
1.5 2 4              mid. long.
r.past.
GWL                Time slices                                                    WNA      CNA              ENA NOAA Heat Index exceeding 41&deg;C NCA                              SCA                          CAR                    NCA 300                                                                                                                            CAR SCA days yr-1 200 100                                                                                                                                    NSA NWS 0                                                                                                                                                NES 1.5 2
4 r.past            mid. long.
1.5 2
4 r.past          mid. long.
1.5 2
4 r.past            mid. long.                                  SAM GWL Time slices              GWL Time slices                GWL Time slices 100-yr return period stream flow                                                                                  SES SWS NWS                                NSA                              NES 2.5 0.8 2                                                                                                      0.6                        SSA Recent past value (m3 s-1 km-2) 1.5                                                                                                      0.4 0.2 Change from recent past (m3 s-1 km-2) 1 0.5                                                                                                      0 0.2 0
SWS                                SAM                              SES                                SSA 2.5 0.8 2                                                                                                                                        0.6 1.5                                                                                                                                        0.4 1                                                                                                                                        0.2 0.5                                                                                                                                        0 0.2 0
1.5 2 4 mid. long.                1.5 2 4 mid. long.                1.5 2 4 mid. long.                1.5 2 4 mid. long.
r.past.                            r.past.                            r.past.                          r.past.
GWL Time slices                    GWL Time slices                    GWL Time slices                  GWL Time slices Legend              GWL                                          RCPs/SSPs                              90 th p
                                                                                          +1.5 &deg; C +2 &deg; C            +4 &deg; C        r.past          2.6    8.5 CMIP6                                                                                                median CMIP5 CORDEX                                                                                                  10 th p 136
 
Technical Summary (b)                                                    100-yr return period stream flow                                                                                                                                      Maximum temperature exceeding 35&deg;C NEU                              WCE                              MED                                                                                                            EEU                                      WSB                            RAR 0.8                                                                                                                  0.1 300                                                                                                          300 Change from past Recent past value days yr -1 0.6                                                                                                                                                                                  200                                                                                                          200 0
(m3 s-1 km-2) 0.4                                                                                                                                                                                  100                                                                                                          100 (m3 s-1 km-2) 0.2                                                                                                                              0.1                                                  0                                                                                                          0 ESB                                      ECA                            TIB 0            1.5 2 4 mid. long.                  1.5 2 4 mid. long.              1.5 2 4 mid. long.                                                                            300                                                                                                          300 days yr -1 r.past. GWL Time slices          r.past. GWL Time slices      r.past. GWL Time slices 200                                                                                                          200 100                                                                                                          100 0                                                                                                          0 RAR                                                                                                            1.5                                          1.5                            1.5 mid.        long.                  mid. long.                mid. long.
2 4                                            2 4                              2 4
NEU                                                                                                                                                                    r.past                                      r.past                        r.past EEU                    WSB              ESB                        RFE                                                                                    GWL          Time slices                    GWL      Time slices          GWL    Time slices WCE                                                                                                                                                                                                                                                                                  TS ECA                                                                                                                                                          Shoreline position change MED                WCA TIB                    EAS                                                                                                                                              EAS            RFE Average shoreline 200 SAH                  ARP                      SAS 0
WAF                                                                                                                                                                                                                              -200 position change (m)
NEAF                                                SEA CAF                                                                                                                                                                                                          -400 SEAF                                                                                                                                                                                                -600 NAU                                                                                                                                      mid- long-      mid- long-term term      term term WSAF    ESAF  MDG                                                          CAU    EAU NOAA Heat Index exceeding 41&deg;C SAU                                                  NZ                                                                        ARP                            WCA 1.5 SAH                                WAF                                                                                                                                300 days yr -1 0.4                                                                                                        200 Change from recent past (m3 s-1 km-2) 1                                                                    0.2                                                                                                        100 0                                                                                                            0 0.5                                                                                                                                                                                                                                          SEA SAS
                                                                                                                            -0.2 0                                                                                                                                                                                300 Recent past value (m3 s-1 km-2) days yr -1 1.5 CAF                          NEAF                      SEAF                                                                                                              200 0.4                                                                                          100 1                                                                                                      0.2                                                                                            0 1.5                                      1.5 mid. long.                  mid. long.
2                                        2 0.5                                                                                                    0                                                                                                      4 r.past 4
r.past
                                                                                                                                          -0.2                                                                                                  GWL Time slices                        GWL Time slices 0
1.5 WSAF                              ESAF          2.5 MDF                                                                                                                  Shoreline position change 0.4                                                        0.8 NAU                        CAU                  EAU              SAU              NZ Average shoreline 2                                                        0.6 1                                                              0.2 1.5                                                                                                          200 0.4 0                                                                                                                  0 0.5                                                                    1                                                        0.2
                                                                                                                                                                                                                    -200 position change (m)
                                                                                                    -0.2 0.5                                                        0 0                                                                                                                              -0.2                                            -400 1.5 2 4 mid. long.                1.5 2 4 mid. long.        0 1.5 2 4 mid. long.                                                                    -600 r.past. GWL Time slices        r.past. GWL Time slices                                                                                                                            mid- long-                  mid- long-              mid- long-          mid- long- mid- long-r.past. GWL Time slices                                                                              term term                  term term              term term          term term term term Legend GWL                                              RCPs/SSPs                                90 th p                                                                                                95 p th
                                                                  +1.5 &deg;C +2 &deg;C              +4 &deg;C      r.past        2.6          8.5                                                                                                                                                  CMIP5 RCP8.5 CMIP6                                                                                          median                                                                                                  median CMIP5                                                                                                                  th                                                                                  th                          CMIP5 RCP4.5 CORDEX                                                                                            10 p                                                                                                    5 p Figure TS.25 l Distribution of projected changes in selected climatic impact-driver (CID) indices for selected regions for Coupled Model Intercomparison Project Phases 5 and 6 (CMIP6, CMIP5) and Coordinated Regional Downscaling Experiment (CORDEX) model ensembles. The intent of this figure is to show that many CID projections for multiple global warming levels and scenarios time slices are available for all the AR6 WGI reference regions and are based on both global (CMIP5, CMIP6) and regional (CORDEX) model ensembles. Different indices are shown for different region: for Eastern Europe and North Asia, the mean number of days per year with maximum temperature exceeding 35&deg;C; for Central America, the Caribbean, South West Asia, South Asia and South East Asia, the mean number of days per year with the National Oceanic and Atmospheric Administration (NOAA) Heat Index exceeding 41&deg;C; for Australasia, East Asia and Russian Far East, the average shoreline position change; for South America, Europe and Africa, the mean change in 1-in-100-year river discharge per unit catchment area (m3 s-1 km-2); and for North America, the median change in the number of days with snow water equivalent (SWE) over 100 mm. For each box plot, the changes or the climatological values are reported with respect to, or compared to, the recent past (1995-2014) period for 1.5&deg;C, 2&deg;C and 4&deg;C global warming levels and for mid-century (2041-2060) or end-century (2081-2100) periods for the CMIP5 and CORDEX RCP8.5 and RCP2.6 and CMIP6 SSP5-8.5 and SSP1-2.6 scenarios ensembles. {Figures 12.5, 12.6, 12.9, 12.SM.1, 12.SM.2, and 12.SM.6}
137
 
Technical Summary TS.4.3.2.2 Asia                                                      {2.3, 8.3, 8.4, 9.5, 9.6, 10.6, Box 10.4, 11.4, 11.5, 11.7, 11.9, 12.4.2, Atlas.3.1, Atlas.5, Atlas.5.2, Atlas.5.3, Atlas.5.4, Due to the high climatological and geographical heterogeneity of    Atlas.5.5}
Asia, some assessment findings below are summarized over five sub-continental areas comprising one or more of the AR6 WGI      In addition to the main changes summarized above and in Section reference regions (Box TS.12): East Asia (EAS+ECA), North Asia  TS.4.3.1, further details are given below.
(WSB+ESB+RFE), South Asia (SAS), South East Asia (SEA) and South West Asia (ARP+WCA).                                            Heat and cold: Over all regions of Asia, observed and projected increases in mean temperature and a shift toward heat extreme Additional regional changes in Asia, besides those features  characteristics are broadly similar to the generic pattern described described in Section TS.4.3.1, include historical trends      in Section TS.4.3.1. Over South East Asia, annual mean surface of annual precipitation that show considerable regional      temperature will likely increase by a slightly smaller amount than the differences (high confidence). East Asian Monsoon            global average. {Atlas.5.4.4}
precipitation has changed, with drying in the north and TS    wetting in the south since the 1950s, and annual mean        Wet and dry: Over East Asia, historical trends of annual precipitation precipitation totals very likely have increased over          show considerable regional differences but with increases over north-most territories of North Asia since the mid-1970s (high      west China and South Korea (high confidence). Daily precipitation confidence). South Asian summer monsoon precipitation        extremes have increased over part of the region (high confidence).
decreased over several areas since the mid-20th century      Extreme hydrological drought frequency has increased in a region (high confidence) but is likely to increase during the 21st  extending from south-west to north-east China, with projected century, with enhanced interannual variability. (Box TS.13)  increases of agricultural and ecological drought for 4&deg;C GWL and fire weather for 2&deg;C and above (medium confidence). {8.3.2, 8.4.2, Increases in precipitation and river floods are projected    11.4.4, 11.4.5, 11.9, 12.4.2.2, Atlas.5.1.2}
over much of Asia: in the annual mean precipitation in East, North, South and South East Asia (high confidence);    Over North Asia, annual mean precipitation totals have very likely for extremes in East, South, West Central, North and South    increased, causing more intense flooding events, and there is East Asia (high confidence) and Arabian Peninsula (medium    medium confidence that the number of dry days has decreased.
confidence); and for river floods in East, South and South    Concurrently, total soil moisture is projected to decline extensively East Asia and East Siberia (medium confidence). Aridity      (medium confidence). {8.3.1.3, 8.4.1.6, 11.4.5, 11.5.2, 11.5.5, in East and West Central Asia is projected to increase,      12.4.2.2, Atlas.5.2.2}
especially beyond the middle of the 21st century and global warming levels beyond 2&deg;C (medium confidence). Fire          Over South Asia, the summer monsoon precipitation decreased weather seasons are projected to lengthen and intensify      over several areas since the mid-20th century (high confidence),
everywhere except South East Asia, Tibetan Plateau and        while it increased in parts of the western HKH and decreased over Arabian Peninsula (medium confidence).                        eastern-central HKH (medium confidence). The frequency of heavy precipitation and flood events has increased over several areas Surface wind speeds have been decreasing in Asia (high        during the last few decades (medium confidence). {8.3.1.3, 8.3.2.4.1, confidence), but there is a large uncertainty in future      8.4.1.5, 8.4.2.4.1, 10.6.3.3, 10.6.3.5, 10.6.3.6, 10.6.3.8, Cross-trends, with medium confidence that mean wind speeds          Chapter Box 10.4, 11.4.1, 11.4.2, 11.4.5, 11.5.5, 12.4.2.2, Box 10.4, will decrease in North Asia, East Asia and Tibetan Plateau    Atlas 5.3.2}
and that tropical cyclones will have decreasing frequency and increasing intensity overall in South East and East      Over South East Asia, mean precipitation trends are not spatially Asia.                                                        coherent or consistent across datasets and seasons (high confidence).
Most of the region has experienced an increase in rainfall intensity Over North Asia, increases in permafrost temperature and      but with a reduced number of wet days (medium confidence). Rainfall its thawing have been observed over recent decades (high      is projected to increase in the northern parts of South East Asia and confidence). Future projections indicate continuing decline  decrease in areas in the Maritime Continent (medium confidence).
in seasonal snow duration, glacial mass, and permafrost area  {8.4.1, 11.4.2, 11.5.5, 11.9, 12.4.2.2, Atlas.3.1, Atlas.5.4.2, Atlas.5.4.4}
by mid-century (high confidence). Snow-covered areas and snow volumes will decrease in most regions of the Hindu      Over South West Asia, an observed annual precipitation decline Kush Himalaya (HKH) during the 21st century, and snowline    over the Arabian Peninsula since the 1980s of 6.3 mm per decade elevations will rise (high confidence) and glacier volumes    is contrasted with observed increases between 1.3 mm and 4.8 mm are likely to decline with greater mass loss in higher CO2    per decade during 1960-2013 over the elevated part of eastern emissions scenarios. Heavy snowfall is increasing in East    West Central Asia (very high confidence), along with an increase of Asia and North Asia (medium confidence) but with limited      the frequency and intensity of extreme precipitation. {Figure 8.19, evidence on future changes in hail and snow avalanches.      Figure 8.20, 8.3.1.6, 8.4.1.6, 11.9, Table 11.2A, 12.4.2.2, Atlas.5.5}
138
 
Technical Summary Wind: Over East Asia, the terrestrial near-surface wind speed has          TS.4.3.2.3 Australasia decreased and is projected to decrease further in the future (medium confidence). Since the mid 1980s, there has been an increase in the            Additional regional changes in Australasia, besides those number and intensification rate of intense TCs (medium confidence),            features described in Section TS.4.3.1, include a significant with a significant north-westward shift in tracks and a northward              decrease in April to October rainfall in the south-west of shift in their average latitude, increasing exposure over East China,          the state of Western Australia, observed from 1910 to 2019 the Korean Peninsula and the Japanese Archipelago (medium                      and attributable to human influence (high confidence),
confidence). {11.7.1, 12.4.2.3}                                                which is very likely to continue in future. Agricultural and ecological droughts and hydrological droughts have Over North Asia, there is medium confidence for a decreasing trend              increased over Southern Australia (medium confidence),
in wind speed during 1979-2018 and for projected continuing                    and meteorological droughts have decreased over Northern decreases of terrestrial near-surface wind speed. {2.3.1.4.4, 12.4.2.3}        and Central Australia (medium confidence). Relative sea level has increased over the period 1993-2018 at a rate Over South East Asia, although there is no significant long-term trend          higher than GMSL around Australasia (high confidence).
in the number of TCs, fewer but more extreme TCs have affected the              Sandy shorelines have retreated around the region, except          TS Philippines during 1951-2013. {11.7.4, 12.4.2.3}                                in Southern Australia, where a shoreline progradation rate of 0.1 m yr -1 has been observed.
Snow and ice: Over East Asia, decreases have been observed in the frequency, and increases in the mean intensity, of snowfall in north-          In the future, heavy precipitation and pluvial flooding are western, north-eastern and south-eastern China and the eastern                  very likely to increase over Northern Australia and Central Tibetan Plateau since the 1960s. Heavy snowfall is projected to                Australia, and they are likely to increase elsewhere in occur more frequently in some parts of Japan (medium confidence).              Australasia for global warming levels (GWLs) exceeding 2&deg;C
{12.4.2.4, Atlas.5.1.2}                                                        and with medium confidence for a 2&deg;C GWL. Agricultural and ecological droughts are projected to increase in Southern Over North Asia, seasonal snow duration and extent have decreased              and Eastern Australia (medium confidence) for a 2&deg;C GWL.
in recent decades (high confidence), and maximum snow depth likely              Fire weather is projected to increase throughout Australia has increased since the mid-1970s, particularly over the south of the          (high confidence) and New Zealand (medium confidence).
Russian Far East. {2.3.2.5, 8.3.1.7.2, 9.5, 12.4.2.4, Atlas.5.2, Atlas.5.4}    Snowfall is expected to decrease throughout the region at high altitudes in both Australia (high confidence) and Over South Asia, snow cover has reduced over most of the HKH since              New Zealand (medium confidence), with glaciers receding the early 21st century, and glaciers have thinned, retreated, and lost          in New Zealand (high confidence). {11.4, Table 11.6, 12.3, mass since the 1970s (high confidence), although the Karakoram                  12.4.3, Atlas.6.4, Atlas.6.5}
glaciers have either slightly gained mass or are in an approximately balanced state (medium confidence). {8.3.1.7.1, Cross-Chapter              In addition to the main changes summarized above and in Section Box 10.4}                                                                  TS.4.3.1, further details are given below.
Over South West Asia, mountain permafrost degradation at high              Heat and cold: Observed and projected increases in mean altitudes has increased the instability of mountain slopes in the past      temperature and a shift toward heat extreme characteristics are decade (medium confidence). More than 60% of glacier mass in the            broadly similar to the generic pattern described in Section TS.4.3.1.
Caucasus is projected to disappear under RCP8.5 emissions by the            {11.9, 12.4.3.1, Atlas.6}
end of the 21st century (medium confidence). {9.5.1, 9.5.3, 12.4.2.4}
Wet and dry: There is medium confidence that heavy precipitation has Coastal and oceanic: Over the last three decades, relative sea              increased in Northern Australia since 1950. Annual mean precipitation level has increased at a rate higher than GMSL around Asia (high            is projected to increase in the south and west of New Zealand (medium confidence). Gross coastal area loss and shoreline retreat has been        confidence) and is projected to decrease in south-west Southern observed over 1984-2015, but with localized shoreline progradation          Australia (high confidence), Eastern Australia (medium confidence), and in the Russian Far East, East and South East Asia. {12.4.2.5}              in the north and east of New Zealand (medium confidence) for a GWL of 2&deg;C. There is medium confidence that river flooding will increase Projections show that regional mean sea level continues to rise (high      in New Zealand and Australia, with higher increases in Northern confidence), ranging from 0.4-0.5 m under SSP1-2.6 to 0.8-1.0 m            Australia. Aridity is projected to increase with medium confidence in under SSP5-8.5 for 2081-2100 relative to 1995-2014 (median                  Southern Australia (high confidence in south-west Southern Australia),
values). This will contribute to more frequent coastal flooding and        Eastern Australia (medium confidence) and in the north and east of higher ETWL in low-lying areas and coastal erosion along sandy              New Zealand (medium confidence) for GWLs around 2&deg;C. {11.4, 11.9, beaches (high confidence). There is high confidence that compound          Table 11.6, 12.4.3.2, Atlas.6.2}
effects of climate change, land subsidence, and human factors will lead to higher flood levels and prolonged inundation in the Mekong          Wind: Mean wind speeds are projected to increase in parts of north-Delta and other Asian coasts. {9.6.1, 9.6.3, 12.4.2.5}                      eastern Australia (medium confidence) by the end of the 21st century 139
 
Technical Summary under high CO2 emissions scenarios. TCs in north-eastern and north    South America and Southern Central America (medium confidence).
Australia are projected to decrease in number (high confidence) but  In Northern South America and Southern Central America, aridity increase in intensity except for east coast lows (low confidence). and agricultural and ecological droughts are increasing with
{12.4.3.3}                                                            medium confidence. Fire weather is projected to increase over Southern Central America and Southern South America with medium Snow and ice: Observations in Australia show that the snow season    confidence. {8.3.1.3, 8.4.2.4.5, 11.4.2, 11.9, Table 11.14, Table 11.15, length has decreased by 5% in the last five decades. Furthermore, the 12.4.4.2, Atlas.7.2.2, Atlas.7.2.4}
date of peak snowfall in Australia has advanced by 11 days over the last 5 decades. Glacier ice volume in New Zealand has decreased by    Wind: Climate projections indicate an increase in mean wind speed 33% from 1977 to 2018. {12.4.3.4, Atlas.6.2}                          and in wind power potential over the Amazonian region (Northern South America, South American Monsoon, North-Eastern South Coastal and oceanic: Observed changes in marine heatwaves            America) (medium confidence). {12.4.4.3}
(MHWs) over the 20th century in the region show an increase in their occurrence frequency, except along the south-east coast of New        Snow and ice: Glacier volume loss and permafrost thawing will TS Zealand, an increase in duration per event, and the total number of  likely continue in the Andes Cordillera under all climate scenarios, MHW days per decade, with the change being stronger in the Tasman    causing important reductions in river flow and potentially high-Sea than elsewhere. The present day 1-in-100-year ETWL is between    magnitude glacial lake outburst floods. {9.5.1.1, 12.4.4.4}
0.5-2.5 m around most of Australia, except the north-western coast where 1-in-100-year ETWL can be as high as 6-7 m. {Box 9.1,    Coastal and oceanic: Around Central and South America, relative 12.3.1.5, 12.4.3.5}                                                  sea level has increased at a higher rate than GMSL in the South Atlantic and the subtropical North Atlantic, and at a rate lower than TS.4.3.2.4 Central and South America                                  GMSL in the East Pacific over the last 3 decades. The present day 1-in-100-year ETWL is highest in Southern and South-Western South Additional regional changes in Central and South America,          America subregions, where it can be as large as 5 to 6 m. Satellite besides those features described in Section TS.4.3.1, include      observations for 1984-2015 show shoreline retreat rates along the increases in mean and extreme precipitation in South-              sandy coasts of Southern Central America, South-Eastern South Eastern South America since the 1960s (high confidence)            America and Southern South America, while shoreline progradation (Section TS.4.2.3). Decreasing trends in mean precipitation        rates have been observed in North-Western South America and and increasing trends in agricultural and ecological drought      Northern South America. Over the period 1982-2016, the coastlines are observed over North-Eastern South America (medium              experienced at least one MHW per year, and more along the Pacific confidence). The intensity and frequency of extreme                coast of North Central America and the Atlantic coast of South-precipitation and pluvial floods is projected to increase          Eastern South America. {12.4.4.5}
over South-Eastern South America, Southern South America, Northern South America, South American Monsoon and                TS.4.3.2.5 Europe North-Eastern South America (medium confidence) for a 2&deg;C GWL and above. Increases of agricultural and ecological              Additional regional changes in Europe, besides those features drought are projected in South America Monsoon and                    described in Section TS.4.3.1, include observed increases Southern South America, and fire weather is projected to              in pluvial flooding in Northern Europe and hydrological increase over several regions (Northern South America, the            and agricultural/ecological droughts in the Mediterranean South American Monsoon, North-Eastern South America                  (high confidence), which have been attributed to human and South-Western South America) (high confidence). {8.3,            influence with high and medium confidence, respectively.
8.4, 11.3, 11.4, 11.9, Table 11.13, Table 11.14, Table 11.15,        Increased mean precipitation amounts at high latitudes 12.4.4.2, Atlas.7.1, Atlas.7.2}                                      in boreal winter and reduced summer precipitation in southern Europe are projected starting from a 2&deg;C GWL In addition to the main changes summarized above and in Section          (high confidence). Aridity, agricultural and hydrological TS.4.3.1, further details are given below.                              droughts and fire weather conditions will increase in the Mediterranean region starting from 2&deg;C GWL (high Heat and cold: Observed and projected increases in mean                  confidence). Pluvial flooding will increase everywhere temperature and a shift toward heat extreme characteristics are          with high confidence except for medium confidence in the broadly similar to the generic pattern described in Section TS.4.3.1. Mediterranean; in Western and Central Europe this also
{11.3.2, 11.3.5, Table 11.13, 12.4.4.1, Atlas.7.1.2, Atlas.7.2.2,        applies to river flooding starting from a 2&deg;C GWL (high Atlas.7.2.4}                                                            confidence). Most periglacial processes in Northern Europe are projected to disappear by the end of the 21st century, Wet and dry: Mean precipitation is projected to change in a dipole      even for a low warming scenario (medium confidence). {8.3, pattern with increases in North-Western and South-Eastern South          11.3, 11.9, 12.4.5, 12.5.2, Atlas.8.2, Atlas.8.4}
America and decreases in North-Eastern and South-Western South America (high confidence) and with further decreases in Northern 140
 
Technical Summary In addition to the main changes summarized above and in Section          America and Northern Central America (from medium to TS.4.3.1, further details are given below.                              high confidence). Severe wind storms, tropical cyclones and dust storms in North America are shifting toward more Heat and cold: Observed and projected increases in mean                  extreme characteristics (medium confidence), and both temperature and a shift toward heat extreme characteristics are          observations and projections point to strong changes in the broadly similar to the generic pattern described in Section TS.4.3.1. seasonal and geographic range of snow and ice conditions
{11.3, 11.9, 12.4.5.1, 12.5.2, Atlas.8.2, Atlas.8.4}                    in the coming decades (very high confidence). General findings for relative sea level, coastal flooding and erosion Wet and dry: There is medium confidence that annual mean                will not apply for areas with substantial land uplift around precipitation has increased in Northern Europe, West and Central        the Hudson Bay and Southern Alaska. {8.4, 11.4, 11.5, 11.7, Europe, and Eastern Europe since the early 20th century and              11.9, 12.4, Atlas.9.4}
high confidence for increases in extreme precipitation. In the European Mediterranean, the magnitude and sign of observed land      In addition to the main changes summarized above and in Section precipitation trends depend on time period and exact study region    TS.4.3.1, further details are given below.
(medium confidence). There is medium confidence that river floods                                                                            TS will decrease in Northern, Eastern and southern Europe for high      Heat and cold: Observed and projected increases in mean warming levels. {8.3.1.3, 11.3, 11.9, 12.4.5.2, Atlas.8.2, Atlas.8.4} temperature and a shift toward heat extreme characteristics are broadly similar to the generic pattern described in Section TS.4.3.1.
Wind: Mean wind speed over land has decreased (medium                {11.3, 11.9, 12.4.6.1, Atlas.9.2, Atlas.9.4}
confidence), but the role of human-induced climate change has not been established. There is high confidence that mean wind speeds      Wet and dry: Annual precipitation increased over parts of Eastern will decrease in Mediterranean areas and medium confidence for        and Central North America during 1960-2015 (high confidence) and such decreases in Northern Europe for GWLs exceeding 2&deg;C. The        has decreased in parts of south-western United States and north-frequency of Medicanes (tropical-like cyclones in the Mediterranean)  western Mexico (medium confidence). River floods are projected to is projected to decrease (medium confidence). {11.9, 12.4.5.3}        increase for all North American regions other than Northern Central America (medium confidence). {8.4.2.4, 11.4, 11.5, 11.9, 12.4.6.2, Snow and ice: In the Alps, snow cover will decrease below elevations  Atlas.9.2, Atlas.9.4}
of 1500-2000 m throughout the 21st century (high confidence).
A reduction of glacier ice volume is projected in the European Alps  Agricultural and ecological drought increases have been observed in and Scandinavia with high confidence and with medium confidence      Western North America (medium confidence), and aridity is projected for the timing and mass change rates. {9.5.2, 12.4.5.4}              to increase in the south-western United States and Northern Central America, with lower summer soil moisture across much of the Coastal and oceanic: Over the last three decades, relative sea level  continental interior (medium confidence). {8.4.1, 11.6.2, 12.4.6.2}
has increased at a lower rate than GMSL in the sub-polar North Atlantic coasts of Europe. The present-day 1-in-100-year ETWL is      Wind: Projections indicate a greater number of the most intense between 0.5-1.5 m in the Mediterranean basin and 2.5-5.0 m in        TCs, with slower translation speeds and higher rainfall potential for the western Atlantic European coasts, around the United Kingdom      Mexicos Pacific Coast, the Gulf Coast and the United States East and along the North Sea coast, and lower at 1.5-2.5 m along the      Coast (medium confidence). Mean wind speed and wind power Baltic Sea coast. Satellite-derived shoreline change estimates over  potential are projected to decrease in Western North America (high 1984-2015 indicate shoreline retreat rates of around 0.5 m yr -1      confidence), with differences between global and regional models along the sandy coasts of Central Europe and the Mediterranean        lending low confidence elsewhere. {11.4, 11.7, 12.4.6.3}
and more or less stable shorelines in Northern Europe. Over the period 1982-2016, the coastlines of Europe experienced on average    Snow and ice: It is likely that some high-latitude regions will more than 2.0 MHW per year, with the eastern Mediterranean and        experience an increase in winter snow water equivalent due to the Scandinavia experiencing 2.5-3 MHWs per year. {12.4.5.5}              snowfall increase prevailing over the warming trend. At sustained GWLs between 3&deg;C and 5&deg;C, nearly all glacial mass in Western TS.4.3.2.6 North America                                              Canada and Western North America will disappear (medium confidence). {9.5.1, 9.5.3, 12.4.6.4, Atlas.9.4}
Additional regional changes in North America, besides those features described in Section TS.4.3.1, include changes            Coastal and oceanic: Around North America, relative sea level has in North American wet and dry CIDs, which are largely              increased over the last three decades at a rate lower than GMSL organized by the north-east (more wet) to south-west (more        in the subpolar North Atlantic and in the East Pacific, while it has dry) pattern of mean precipitation change, although heavy          increased at a rate higher than GMSL in the subtropical North Atlantic.
precipitation increases are widespread (high confidence).          Observations indicate that episodic coastal flooding is increasing Increasing evaporative demand will expand agricultural            along many coastlines in North America. Shoreline retreat rates of and ecological drought and fire weather (particularly in          around 1 m yr -1 have been observed during 1984-2015 along the summertime) in Central North America, Western North                sandy coasts of North-Western North America and Northern Central 141
 
Technical Summary America, while portions of the United States Gulf Coast have seen          Permafrost warming, loss of seasonal snow cover, and a retreat rate approaching 2.5 m yr -1. Sandy shorelines along Eastern      glacier melt will be widespread (high confidence). There is North America and Western North America have remained more or              high confidence that both the Greenland and Antarctic ice less stable during 1984-2014, but a shoreline progradation rate of          sheets have lost mass since 1992 and will continue to lose around 0.5 m yr -1 has been observed in North-Eastern North America.        mass throughout this century under all emissions scenarios.
{12.4.6.5}                                                                  Relative sea level and coastal flooding are projected to increase in areas other than regions with substantial land TS.4.3.2.7 Small Islands                                                    uplift (medium confidence). {2.3, 3.4, 4.3, 4.5, 7.4, 8.2, 8.4, Box 8.2, 9.5, 12.4.9, Atlas.11.1, Atlas.11.2}
Additional regional changes in Small Islands, besides those features described in Section TS.4.3.1, include a likely              In addition to the main changes summarized above and in Section decrease in rainfall during boreal summer in the Caribbean            TS.4.3.1, further details are given below.
and in some parts of the Pacific islands poleward of 20&deg; latitude in both the Northern and Southern Hemispheres.              Heat and cold: Changes in Antarctica showed larger spatial TS    These drying trends will likely continue in coming decades.          variability, with very likely warming in the Antarctic Peninsula since Fewer but more intense tropical cyclones are projected                the 1950s and no overall trend in East Antarctica. Less warming starting from a 2&deg;C GWL (medium confidence). {9.6, 11.3,              and weaker polar amplification are projected as very likely over 11.4, 11.7, 11.9, 12.4.7, Atlas.10.2, Atlas.10.4, Cross-Chapter      the Antarctic than in the Arctic, with a weak polar amplification Box Atlas.2}                                                          projected as very likely by the end of the 21st century. {4.3.1, 4.5.1, 7.4.4, 12.4.9.1, Atlas.11.1, Atlas.11.2}
In addition to the main changes summarized above and in Section TS.4.3.1, further details are given below.                              Wet and dry: Recent decades have seen a general decrease in Arctic aridity (high confidence), with increased moisture transport Heat and cold: It is very likely that most Small Islands have warmed    leading to higher precipitation, humidity and streamflow and a over the period of instrumental records, and continued temperature      corresponding decrease in dry days. Antarctic precipitation showed increases in the 21st century will further increase heat stress in these a positive trend during the 20th century. The water cycle is projected regions. {11.3.2, 11.9, 12.4.7.1, Atlas.10.2, Atlas.10.4, Cross-Chapter  to intensify in both polar regions, leading to higher precipitation Box Atlas.2}                                                            totals (and a shift to more heavy precipitation) and higher fraction of precipitation falling as rain. In the Arctic, this will result in higher Wet and dry: Observed and projected rainfall trends vary spatially      river flood potential and earlier meltwater flooding, altering seasonal across the Small Islands. Higher evapotranspiration under a warming      characteristics of flooding (high confidence). A lengthening of the fire climate can partially offset future increases or amplify future          season (medium confidence) and encroachment of fire regimes into reductions in rainfall, resulting in increased aridity as well as more  tundra regions (high confidence) are projected. {8.2.3, 8.4.1, Box 8.2, severe agricultural and ecological drought in the Caribbean (medium      9.4.1, 9.4.2, 12.4.9.2, Atlas.11.1, Atlas.11.2}
confidence). {11.4.2, 11.9, 12.4.7.2, Atlas.10.2, Atlas.10.4, Cross-Chapter Box Atlas.2}                                                    Wind: There is medium confidence in mean wind decrease over the Russian Arctic and Arctic North-East North America, but low Wind: Global changes indicate that Small Islands will face fewer        confidence of changes in other Arctic regions and Antarctica.
but more intense TCs, with spatial inconsistency in projections given    {12.4.9.3}
poleward shifts in TC tracks (medium confidence). {11.7.1.2, 11.7.1.5, 12.4.7.3}                                                                Snow and ice: Reductions in spring snow cover extent have occurred across the Northern Hemisphere since at least 1978 (very Coastal and oceanic: Continued relative sea level rise is very likely    high confidence). Permafrost warming and thawing have been in the ocean around Small Islands and, along with storm surges          widespread in the Arctic since the 1980s (high confidence), causing and waves, will exacerbate coastal inundation with the potential to      strong heterogeneity in surface conditions. There is high confidence increase saltwater intrusion into aquifers in small islands. Shoreline  in future glacier- and ice-sheet loss, permafrost warming, decreasing retreat is projected along sandy coasts of most small islands (high      permafrost extent and decreasing seasonal duration and extent of confidence). {9.6.3.3, 12.4.7.4, Cross-Chapter Box Atlas.2}              snow cover in the Arctic. Decline in seasonal sea ice coverage along the majority of the Arctic coastline in recent decades is projected to TS.4.3.2.8 Polar                                                        continue, contributing to an increase in coastal hazards (including open water storm surge, coastal erosion and flooding). {2.3.2, 3.4.2, It is virtually certain that surface warming in the Arctic            3.4.3, 9.4.1, 9.4.2, 9.5, 12.4.6, 12.4.9, Atlas.11.2}
will continue to be more pronounced than the global average warming over the 21st century. An intensification            Coastal and oceanic: Higher sea levels contribute to high of the polar water cycle will increase mean precipitation,            confidence for projected increases of Arctic coastal flooding and with precipitation intensity becoming stronger and more              higher coastal erosion (aided by sea ice loss) (medium confidence),
likely to be rainfall rather than snowfall (high confidence).
142
 
Technical Summary with lower confidence for those regions with substantial land uplift      TS.4.3.2.10 Other Typological Domains (Arctic North-East North America and Greenland). {12.4.9.5}
Some types of regions found in different continents face TS.4.3.2.9 Ocean                                                              common climate challenges regardless of their location.
These include biodiversity hot spots that will very likely The Indian Ocean, western equatorial Pacific Ocean and                    see even more extreme heat and droughts, mountain areas western boundary currents have warmed faster than the                      where a projected raising in the freezing level height will global average (very high confidence), with the largest                    alter snow and ice conditions (high confidence), and tropical changes in the frequency of marine heatwaves (MHWs)                        forests that are increasingly prone to fire weather (medium projected in the western tropical Pacific and the Arctic                  confidence). {8.4, Box 8.2, 9.5, 12.3, 12.4}
Ocean (medium confidence). The Pacific and Southern Ocean are projected to freshen and the Atlantic to become more                Biodiversity hotspots located around the world will each face unique saline (medium confidence). Anthropogenic warming is very              challenges in CID changes. Heat, drought and length of dry season, likely to further decrease ocean oxygen concentrations, and            wildfire weather, sea surface temperature and deoxygenation are this deoxygenation is expected to persist for thousands                relevant drivers to terrestrial and freshwater ecosystems and have        TS of years (medium confidence). Arctic sea ice losses are                marked increasing trends. {12.3, 12.4.10.1}
projected to continue, leading to a practically ice-free Arctic in September by the end of the 21st century under high                  Desert and semi-arid areas are strongly affected by CIDs such as CO2 emissions scenarios (high confidence). {2.3, 5.3, 9.2, 9.3,        extreme heat, drought and dust storms, with large-scale aridity Box 9.2, 12.3.6, 12.4.8}                                                trends contributing to expanding drylands in some regions (high confidence). {12.3, 12.4.10.3}
In addition to the main changes summarized above and in Section TS.4.3.1, further details are given below.                                Average warming in mountain areas varies with elevation, but the pattern is not globally uniform (medium confidence). Extreme Ocean surface temperature: The Southern Ocean, the eastern                precipitation is projected to increase in major mountainous regions equatorial Pacific, and the North Atlantic Ocean have warmed more          (medium to high confidence depending on location), with potential slowly than the global average or slightly cooled. Global warming          cascading consequences of floods, landslides and lake outbursts of 2&deg;C above 1850-1900 levels would result in the exceedance of            in all scenarios (medium confidence). {8.4.1.5, Box 8.2, 9.5.1.3, numerous hazard thresholds for pathogens, seagrasses, mangroves,          9.5.3.3, 9.5.2.3, Cross-Chapter Box 10.4, 11.5.5, 12.3, 12.4.1-12.4.6, kelp forests, rocky shores, coral reefs and other marine ecosystems        12.4.10.4}
(medium confidence). {9.2.13, 12.4.8}
Most tropical forests are challenged by a mix of emerging warming Marine heatwaves: Moderate increases in MHW frequency are                  trends that are particularly large in comparison to historical variability projected for mid-latitudes, and only small increases are projected        (medium confidence). Water cycle changes bring prolonged drought, for the Southern Ocean (medium confidence). Under the SSP5-8.5            longer dry seasons and increased fire weather to many tropical scenario, permanent MHWs (more than 360 days per year) are                forests (medium confidence). {10.5, 12.3, 12.4}
projected to occur in the 21st century in parts of the tropical ocean, the Arctic Ocean, and around 45&deg;S; however, the occurrence of such permanent MHWs can be largely avoided under the SSP1-2.6 scenario. {Box 9.2, 12.4.8}
Ocean acidity: With the rising CO2 concentration, the ocean surface pH has declined globally over the past four decades (virtually certain).
{2.3.3.5, 5.3.3.2, 12.4.8}
Ocean salinity: At the basin scale, it is very likely that the Pacific and the Southern Ocean have freshened while the Atlantic has become more saline. {2.3.3.2, 9.2.2.2, 12.4.8}
Dissolved oxygen: In recent decades, low oxygen zones in ocean ecosystems have expanded. {2.3.4.2, 5.3.3.2, 12.4.8}
Sea ice: Arctic perennial sea ice is being replaced by thin, seasonal ice, with earlier spring melt and delayed fall freeze up. There is no clear trend in the Antarctic sea ice area over the past few decades and low confidence in its future change. {2.3.2.1.1, 9.3.1.1, 12.4.8, 12.4.9}
143
 
Technical Summary Box TS.14 l Urban Areas With global warming, urban areas and cities will be affected by more frequent occurrences of extreme climate events, such as heatwaves, with more hot days and warm nights as well as sea level rise and increases in tropical cyclone storm surge and rainfall intensity that will increase the probability of coastal city flooding (high confidence). {Box 10.3, 11.3, 11.5, 12.3, 12.4}
Urban areas have special interactions with the climate system, for instance in terms of heat islands and altering the water cycle, and thereby will be more affected by extreme climate events such as extreme heat (high confidence). With global warming, increasing relative sea level compounded by increasing tropical cyclone storm surge and rainfall intensity will increase the probability of coastal city flooding (high confidence). Arctic coastal settlements are particularly exposed to climate change due to sea ice retreat (high confidence). Improvements in urban climate modelling and climate monitoring networks have contributed to understanding the mutual interaction between regional and urban climate (high confidence). {Box 10.3, 11.3, 11.5, 12.3, 12.4}
TS Despite having a negligible effect on global surface temperature (high confidence), urbanization has exacerbated the effects of global warming through its contribution to the observed warming trend in and near cities, particularly in annual mean minimum temperature (very high confidence) and increases in mean and extreme precipitation over and downwind of the city, especially in the afternoon and early evening (medium confidence). {2.3, Box 10.3, 11.3, 11.4, 12.3, 12.4}
Combining climate change projections with urban growth scenarios, future urbanization will amplify (very high confidence) the projected local air temperature increase, particularly by strong influence on minimum temperatures, which is approximately comparable in magnitude to global warming (high confidence). Compared to present day, large implications are expected from the combination of future urban development and more frequent occurrence of extreme climate events, such as heatwaves, with more hot days and warm nights adding to heat stress in cities (very high confidence). {Box 10.2, 11.3, 12.4}
Both sea levels and air temperatures are projected to rise in most coastal settlements (high confidence). There is high confidence in an increase in pluvial flood potential in urban areas where extreme precipitation is projected to increase, especially at high global warming levels. {11.4, 11.5, 12.4}
144
 
ATTACHMENT D SPM 9 Ocean, Cryosphere and Sea Level Change Coordinating Lead Authors:
Baylor Fox-Kemper (United States of America), Helene T. Hewitt (United Kingdom), Cunde Xiao (China)
Lead Authors:
Gufinna Aalgeirsd&#xf3;ttir (Iceland), Sybren S. Drijfhout (The Netherlands), Tamsin L. Edwards (United Kingdom), Nicholas R. Golledge (New Zealand/United Kingdom), Mark Hemer (Australia),
Robert E. Kopp (United States of America), Gerhard Krinner (France/Germany, France), Alan Mix (United States of America), Dirk Notz (Germany), Sophie Nowicki (United States of America/
France, United States of America), Intan Suci Nurhati (Indonesia), Lucas Ruiz (Argentina),
Jean-Baptiste Sall&#xe9;e (France), Aimee B.A. Slangen (The Netherlands), Yongqiang Yu (China)
Contributing Authors:
Cecile Agosta (France), Kyle Armour (United States of America), Mathias Aschwanden (Switzerland),
Jonathan L. Bamber (United Kingdom), Sophie Berger (France/Belgium), F&#xe1;bio Boeira Dias (Finland/
Brazil), Jason E. Box (Denmark/United States of America), Eleanor J. Burke (United Kingdom),
Kevin D. Burke (United States of America), Xavier Capet (France), John A. Church (Australia),
Lee de Mora (United Kingdom), Chris Derksen (Canada), Catia M. Domingues (Australia, United Kingdom/Brazil), Jakob Drr (Norway/Germany), Paul J. Durack (United States of America/
Australia), Thomas L. Frlicher (Switzerland), Thian Y. Gan (Canada/Malaysia), Gregory G. Garner (United States of America), Sebastian Gerland (Norway/Germany), Heiko Goelzer (Norway/
Germany), Natalya Gomez (Canada), Irina V. Gorodetskaya (Portugal/Belgium, The Russian Federation), Jonathan M. Gregory (United Kingdom), Robert Hallberg (United States of America),
F. Alexander Haumann (United States of America/Germany), Tim H. J. Hermans (The Netherlands),
Emma M. Hill (Singapore/United States of America, United Kingdom), Regine Hock (United States of America, Norway/Germany), Stefan Hofer (Norway/Austria), Romain Hugonnet (France, Switzerland/
France), Philippe Huybrechts (Belgium), A.K.M. Saiful Islam (Bangladesh), Laura C. Jackson (United Kingdom), Nicolas C. Jourdain (France), Andreas Kb (Norway/Germany), Nicole S. Khan (China/
United States of America), Shfaqat Abbas Khan (Denmark), Matthew Kirwan (United States of America), Roxy Mathew Koll (India), James Kossin (United States of America), Anders Levermann (Germany), Sophie Lewis (Australia), Shiyin Liu (China), Daniel Lowry (New Zealand/United States of America), Marta Marcos (Spain), Ben Marzeion (Germany), Matthew Menary (France/United Kingdom), Sebastian H. Mernild (Norway, Denmark/Norway), Philip Orton (United States of America),
Matthew D. Palmer (United Kingdom), Frank Pattyn (Belgium), Brodie Pearson (United States of America/United Kingdom), C&#xe9;cile Pellet (Switzerland), Chris Perry (United Kingdom), Mark D. Pickering 1211
 
Chapter 9                                                                      Ocean, Cryosphere and Sea Level Change (United Kingdom), Johannes Quaas (Germany), Roshanka Ranasinghe (The Netherlands/Sri Lanka, Australia), Roelof Rietbroek (The Netherlands), Malcolm J. Roberts (United Kingdom), Alessio Rovere (Germany/Italy), Maria Santolaria Otin (Spain, France/Spain), Abhishek Savita (Australia/India),
Alex Sen Gupta (Australia/United Kingdom, Australia), Helene Seroussi (United States of America/
France), Sharon L. Smith (Canada), Olga N. Solomina (The Russian Federation), Esther Stouthamer (The Netherlands), Fiametta Straneo (United States of America/Italy, United States of America),
William V. Sweet (United States of America),Thomas Wahl (United States of America/Germany), Lisan Yu (United States of America), Jiacan Yuan (United States of America/China), Jan David Zika (Australia)
Review Editors:
Unnikrishnan Alakkat (India), Benjamin P. Horton (Singapore/United Kingdom), Simon Marsland (Australia)
Chapter Scientists:
9            Gregory G. Garner (United States of America),Tim H. J. Hermans (The Netherlands), Lijuan Hua (China),
Tamzin Palmer (United Kingdom), Brodie Pearson (United States of America/United Kingdom)
This chapter should be cited as:
Fox-Kemper, B., H.T. Hewitt, C. Xiao, G. Aalgeirsd&#xf3;ttir, S.S. Drijfhout, T.L. Edwards, N.R. Golledge, M. Hemer, R.E. Kopp, G. Krinner, A. Mix, D. Notz, S. Nowicki, I.S. Nurhati, L. Ruiz, J.-B. Sall&#xe9;e, A.B.A. Slangen, and Y. Yu, 2021: Ocean, Cryosphere and Sea Level Change. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P.
Zhai, A. Pirani, S.L. Connors, C. P&#xe9;an, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E.
Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelek&#xe7;i, R. Yu, and B. Zhou (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 1211-1362, doi:10.1017/9781009157896.011.
1212
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                                                                                    Chapter 9 Table of Contents Executive Summary                  1214 9.6      Sea Level Change                          1287 9.6.1      Global and Regional Sea Level Change 9.1      Introduction                1218                in the Instrumental Era  1287 Box 9.1 l Key Processes Driving Sea Level Change  1220                                                                    Cross-Chapter Box 9.1 l Global Energy Inventory and Sea Level Budget  1291 9.2      Oceans        1221    9.6.2      Paleo Context of Global and Regional 9.2.1    Ocean Surface                      1221                Sea Level Change  1292 9.6.3      Future Sea Level Changes                                     1295 Box 9.2 l Marine Heatwaves                                    1227 9.2.2    Changes in Heat and Salinity                                         1228    Box 9.4 l High-end Storyline                                                                                              9 of 21st-century Sea Level Rise                                      1308 9.2.3    Regional Ocean Circulation                                        1236 9.6.4      Extreme Sea Levels: Tides, Surges and Waves  1309 9.2.4    Steric and Dynamic Sea Level Change                                                     1244 9.7      Final Remarks                    1314 9.3      Sea Ice        1247 9.3.1    Arctic Sea Ice                  1247 Acknowledgements                      1315 9.3.2    Antarctic Sea Ice                        1251 Frequently Asked Questions 9.4      Ice Sheets            1254    FAQ 9.1 l Can Continued Melting of the Greenland and 9.4.1    Greenland Ice Sheet                              1254                Antarctic Ice Sheets Be Reversed?
How Long Would It Take for Them Box 9.3 l Insights into Land Ice Evolution                                                                                              to Grow Back?.................................................................. 1316 From Model Intercomparison Projects  1261                                                      FAQ 9.2 l How Much Will Sea Level Rise 9.4.2    Antarctic Ice Sheet                          1263                in the Next Few Decades?...................................... 1318 FAQ 9.3 l Will the Gulf Stream Shut Down?                                                          ................. 1320 9.5      Glaciers, Permafrost and Seasonal Snow Cover                                        1273 References      1322 9.5.1    Glaciers            1273 9.5.2    Permafrost                  1280 9.5.3    Seasonal Snow Cover                                  1283 1213
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change Executive Summary                                                          Marine heatwaves - sustained periods of anomalously high near-surface temperatures that can lead to severe and persistent This chapter assesses past and projected changes in the ocean,            impacts on marine ecosystems - have become more frequent cryosphere and sea level using paleoreconstructions, instrumental          over the 20th century (high confidence). Since the 1980s, they observations and model simulations. In the following summary, we          have approximately doubled in frequency (high confidence) and update and expand the related assessments from the IPCC Fifth              have become more intense and longer (medium confidence).
Assessment Report (AR5), the Special Report on Global Warming of          This trend will continue, with marine heatwaves at global scale 1.5&deg;C (SR1.5) and the Special Report on Ocean and Cryosphere in            becoming four times [2 to 9, likely range] more frequent in 2081-2100 a Changing Climate (SROCC). This chapter covers major advances since      compared to 1995-2014 under SSP12.6, and eight times [3 to 15, likely SROCC, including the synthesis of extended and new observations.          range] more frequent under SSP58.5. The largest changes will occur in These advances allow for improved assessment of past change,              the tropical ocean and the Arctic (medium confidence). {Box 9.2}
processes and budgets for the last century, and the use of a hierarchy of models and emulators, which provide improved projections and            The upper ocean has become more stably stratified since uncertainty estimates of future change. In addition, the systematic use    at least 1970 over the vast majority of the globe (virtually 9 of model emulators makes our projections of ocean heat content, land      certain), primarily due to surface-intensified warming and ice loss and sea level rise fully consistent with each other and with      high-latitude surface freshening (very high confidence). Changes the assessed equilibrium climate sensitivity and projections of global    in ocean stability affect vertical exchanges of surface waters with surface air temperature across the entire report. In this executive        the deep ocean and large-scale ocean circulation. Based on recent summary, uncertainty ranges are reported as very likely ranges and        refined analyses of the available observations, the global 0-200 m expressed by square brackets, unless otherwise noted.                      stratification is now assessed to have increased about twice as much as reported by SROCC, with a 4.9 +/- 1.5% increase from 1970 to 2018 (high confidence) and even higher increases at the base of the surface Ocean Heat and Salinity                                                    mixed layer. Upper-ocean stratification will continue to increase throughout the 21st century (virtually certain). {9.2.1}
At the ocean surface, temperature has, on average, increased by 0.88 [0.68 to 1.01] &deg;C between 1850-1900 and 2011-2020, with 0.60 [0.44 to 0.74] &deg;C of this warming having occurred since 1980.        Ocean Circulation The ocean surface temperature is projected to increase between 1995 to 2014 and 2081 to 2100 on average by 0.86 [0.43 to                  The Atlantic Meridional Overturning Circulation (AMOC) will 1.47, likely range] &deg;C in SSP12.6 and by 2.89 [2.01 to 4.07,              very likely decline over the 21st century for all SSP scenarios.
likely range] &deg;C in SSP58.5. Since the 1950s, the fastest surface        There is medium confidence that the decline will not involve warming has occurred in the Indian Ocean and in western boundary          an abrupt collapse before 2100. For the 20th century, there is low currents, while ocean circulation has caused slow warming or surface      confidence in reconstructed and modelled AMOC changes because of cooling in the Southern Ocean, equatorial Pacific, North Atlantic, and    their low agreement in quantitative trends. The low confidence also coastal upwelling systems (very high confidence). At least 83% of the      arises from new observations that indicate missing key processes in ocean surface will very likely warm over the 21st century in all Shared    both models and measurements used for formulating proxies and Socio-economic Pathways (SSP) scenarios. {2.3.3, 9.2.1}                    from new evaluations of modelled AMOC variability. This results in low confidence in quantitative projections of AMOC decline in the The heat content of the global ocean has increased since at                21st century, despite the high confidence in the future decline as least 1970, and will continue to increase over the 21st century            a qualitative feature based on process understanding. {9.2.3}
(virtually certain). The associated warming will likely continue until at least 2300, even for low-emissions scenarios, because            Southern Ocean circulation and associated temperature of the slow circulation of the deep ocean. Ocean heat content              changes in Antarctic ice-shelf cavities are sensitive to changes has increased from 1971 to 2018 by 0.396 [0.329 to 0.463, likely          in wind patterns and increased ice shelf melt (high confidence).
range] yottajoules and will likely increase until 2100 by two to four      However, limitations in understanding feedback mechanisms times that amount under SSP12.6 and four to eight times that amount      involving the ocean, atmosphere and cryosphere, which are not fully under SSP58.5. The long time scale also implies that the amount of        represented in the current generation of climate models, generally deep-ocean warming will only become scenario-dependent after about        limit our confidence in future projections of the Southern Ocean and 2040 (medium confidence), and that the warming is irreversible over        of its forcing on Antarctic sea ice and ice shelves. {9.2.3, 9.3.2, 9.4.2}
centuries to millennia (very high confidence). On annual to decadal time scales, the redistribution of heat by the ocean circulation dominates Many ocean currents will change in the 21st century as spatial patterns of temperature change (high confidence). At longer time  a response to changes in wind stress associated with scales, the spatial patterns are dominated by additional heat, primarily  anthropogenic warming (high confidence). Western boundary stored in water masses formed in the Southern Ocean, and by weaker        currents have shifted poleward since 1993 (medium confidence),
warming in the North Atlantic where heat redistribution caused by          consistent with a poleward shift of the subtropical gyres. Of the four changing circulation counteracts the additional heat input through the    eastern boundary upwelling systems, only the California Current surface (high confidence). {9.2.2, 9.2.4, 9.6.1, Cross-Chapter Box 9.1}    system has experienced some large-scale upwelling-favourable 1214
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 wind intensification since the 1980s (medium confidence). In the            The Antarctic Ice Sheet has lost 2670 [1800 to 3540] Gt 21st century, consistent with projected changes in the surface winds,      mass over the period 1992-2020, equivalent to 7.4 [5.0 to the East Australian Current Extension and Agulhas Current Extension        9.8] mm global mean sea level rise. The mass-loss rate was, on will intensify, while the Gulf Stream and Indonesian Throughflow            average, 49 [-2 to +100] Gt yr -1 over the period 1992-1999, will weaken (medium confidence). Eastern boundary upwelling                70 [22 to 119] Gt yr -1 over the period 2000-2009 and 148 [94 to systems will change, with a dipole spatial pattern within each system      202] Gt yr -1 over the period 2010-2019. Mass losses from West of reduction at low latitude and enhancement at high latitude              Antarctic outlet glaciers outpaced mass gain from increased snow (high confidence). {9.2.1, 9.2.3}                                          accumulation on the continent and dominated the ice-sheet mass losses since 1992 (very high confidence). These mass losses from the West Antarctic outlet glaciers were mainly induced by ice-shelf basal Sea Ice                                                                    melt (high confidence) and locally by ice-shelf disintegration preceded by strong surface melt (high confidence). Parts of the East Antarctic The Arctic Ocean will likely become practically sea ice free1              Ice Sheet have lost mass in the last two decades (high confidence).
during the seasonal sea ice minimum for the first time before              {2.3.2, 9.4.2, Atlas.11.1}
2050 in all considered SSP scenarios. There is no tipping                                                                                          9 point for this loss of Arctic summer sea ice (high confidence).            Both the Greenland Ice Sheet (virtually certain) and the The practically ice-free state is projected to occur more often with higher Antarctic Ice Sheet (likely) will continue to lose mass greenhouse gas concentrations, and it will become the new normal for        throughout this century under all considered SSP scenarios.
high-emissions scenarios by the end of this century (high confidence).      The related contribution to global mean sea level rise until Based on observational evidence, Coupled Model Intercomparison              2100 from the Greenland Ice Sheet will likely be 0.01 to Project Phase 6 (CMIP6) models and conceptual understanding, the            0.10 m under SSP12.6, 0.04 to 0.13 m under SSP24.5 and substantial satellite-observed decrease of Arctic sea ice area over        0.09-0.18 m under SSP58.5, while the Antarctic Ice Sheet will the period 1979-2019 is well described as a linear function of global      likely contribute 0.03 to 0.27 m under SSP12.6, 0.03 to 0.29 m mean surface temperature, and thus of cumulative anthropogenic              under SSP24.5, and 0.03 to 0.34 m under SSP58.5. The loss of carbon dioxide (CO2) emissions, with superimposed internal variability      ice from Greenland will become increasingly dominated by surface (high confidence). According to both process understanding and              melt, as marine margins retreat and the ocean-forced dynamic CMIP6 simulations, a practically sea ice-free state will likely be          response of ice-sheet margins diminishes (high confidence). In the observed some years before additional (post-2020) cumulative                Antarctic, dynamic losses driven by ocean warming and ice-shelf anthropogenic CO2 emissions reach 1000 GtCO2. {4.3.2, 9.3.1}                disintegration will likely continue to outpace increasing snowfall this century (medium confidence). Beyond 2100, total mass loss from For Antarctic sea ice, regionally opposing trends and large                both ice sheets will be greater under high-emissions scenarios than interannual variability result in no significant trend in satellite-        under low-emissions scenarios (high confidence). The assessed likely observed sea ice area from 1979 to 2020 in both winter and                  ranges consider those ice-sheet processes in whose representation summer (high confidence). The regionally opposing trends result            in current models we have at least medium confidence, including primarily from changing regional wind forcing (medium confidence).          surface mass balance and grounding-line retreat in the absence of There is low confidence in model simulations of future Antarctic sea        instabilities. Under high-emissions scenarios, poorly understood ice decrease, and lack of decrease, due to deficiencies of process          processes related to marine ice sheet instability and marine ice cliff representation, in particular at the regional level. {2.3.2, 9.2.3, 9.3.2}  instability, characterized by deep uncertainty, have the potential to strongly increase Antarctic mass loss on century to multi-century time scales. {9.4.1, 9.4.2, 9.6.3, Box 9.3, Box 9.4}
Ice Sheets The Greenland Ice Sheet has lost 4890 [4140 to 5640] Gt                    Glaciers mass over the period 1992-2020, equivalent to 13.5 [11.4 to 15.6] mm global mean sea level rise. The mass-loss rate was                Glaciers lost 6200 [4600 to 7800] Gt of mass (17.1 [12.7 to on average 39 [-3 to +80] Gt yr -1 over the period 1992-1999,              21.5] mm global mean sea level equivalent) over the period 175 [131 to 220] Gt yr -1 over the period 2000-2009 and                    1993-2019 and will continue losing mass under all SSP scenarios 243 [197 to 290] Gt yr -1 over the period 2010-2019. This mass              (very high confidence). During the decade 2010-2019, glaciers loss is driven by both discharge and surface melt, with the latter          lost more mass than in any other decade since the beginning of increasingly becoming the dominating component of mass loss with            the observational record (very high confidence). For all regions high interannual variability in the last decade (high confidence). The      with long-term observations, glacier mass in the decade 2010-2019 largest mass losses occurred in the north-west and the south-east of        is the smallest since at least the beginning of the 20th century Greenland (high confidence). {2.3.2, 9.4.1}                                (medium confidence). Because of their lagged response, glaciers will continue to lose mass at least for several decades even if global temperature is stabilized (very high confidence). Glaciers will lose 1  Sea ice area below 1 million km2.
1215
 
Chapter 9                                                                                      Ocean, Cryosphere and Sea Level Change 29,000 [9000 to 49,000] Gt and 58,000 [28,000 to 88,000] Gt over      Ocean thermal expansion (38%) and mass loss from glaciers (41%)
the period 2015-2100 for RCP2.6 and RCP8.5, respectively (medium      dominate the total change from 1901 to 2018. The contribution of confidence), which represents 18 [5 to 31] % and 36 [16 to 56] %      Greenland and Antarctica to GMSL rise was four times larger during of their early-21st-century mass, respectively. {2.3.2, 9.5.1, 9.6.1,  2010-2019 than during 1992-1999 (high confidence). Because of 9.6.3, 12.4}                                                          the increased ice-sheet mass loss, the total loss of land ice (glaciers and ice sheets) was the largest contributor to global mean sea level rise over the period 2006-2018 (high confidence). {2.3.3, 9.6.1, 9.6.2, Permafrost                                                            Cross-Chapter Box 9.1, Table 9.A.1, Box 7.2}
Increases in permafrost temperature have been observed over            At the basin scale, sea levels rose fastest in the Western Pacific the past three to four decades throughout the permafrost              and slowest in the Eastern Pacific over the period 1993-2018 regions (high confidence), and further global warming will            (medium confidence). Regional differences in sea level arise from:
lead to near-surface permafrost volume loss (high confidence).        ocean dynamics; changes in Earth gravity, rotation and deformation Complete permafrost thaw in recent decades is a common                due to land ice and land-water changes; and vertical land motion.
9 phenomenon in discontinuous and sporadic permafrost regions            Temporal variability in ocean dynamics dominates regional patterns (medium confidence). Permafrost warmed globally by 0.29 [0.17 to      on annual to decadal time scales (high confidence). The anthropogenic 0.41, likely range] &deg;C between 2007 and 2016 (medium confidence).      signal in regional sea level change will emerge in most regions An increase in the active layer thickness is a pan-Arctic phenomenon  by 2100 (medium confidence). {9.2.4, 9.6.1}
(medium confidence), subject to strong heterogeneity in surface conditions. The volume of perennially frozen soil within the upper 3 m Regional sea level change has been the main driver of changes of the ground will decrease by about 25% per 1&deg;C of global surface    in extreme still water levels across the quasi-global tide gauge air temperature change (up to 4&deg;C above pre-industrial temperature)    network over the 20th century (high confidence) and will be (medium confidence). {9.5.2}                                          the main driver of a substantial increase in the frequency of extreme still water levels over the next century (medium confidence). Observations show that high-tide flooding events that Snow                                                                  occurred five times per year during the period 1960-1980 occurred, on average, more than eight times per year during the period Northern Hemisphere spring snow cover extent has been                  1995-2014 (high confidence). Under the assumption that other decreasing since 1978 (very high confidence), and there is            contributors to extreme sea levels remain constant (e.g., stationary high confidence that this trend extends back to 1950. Further          tides, storm-surge, and wave climate), extreme sea levels that decrease of Northern Hemisphere seasonal snow cover                    occurred once per century in the recent past will occur annually or extent is virtually certain under further global warming. The          more frequently at about 19-31% of tide gauges by 2050 and at observed sensitivity of Northern Hemisphere snow cover extent to      about 60% (SSP12.6) to 82% (SSP58.5) of tide gauges by 2100 Northern Hemisphere land surface air temperature for 1981-2010        (medium confidence). In total, such extreme sea levels will occur is -1.9 [-2.8 to -1.0, likely range] million km2 per 1&deg;C throughout    about 20 to 30 times more frequently by 2050 and 160 to 530 times the snow season. It is virtually certain that Northern Hemisphere      more frequently by 2100 compared to the recent past, as inferred snow cover extent will continue to decrease as global climate          from the median amplification factors for SSP12.6, SSP24.5, and continues to warm, and process understanding strongly suggests        SSP58.5 (medium confidence). Over the 21st century, the majority that this also applies to Southern Hemisphere seasonal snow            of coastal locations will experience a median projected regional cover (high confidence). Northern Hemisphere spring snow cover        sea level rise within +/-20% of the median projected GMSL change extent will decrease by about 8% per 1&deg;C of global surface air        (medium confidence). {9.6.3, 9.6.4}
temperature change (up to 4&deg;C above pre-industrial temperature)
(medium confidence). {9.5.3}                                          It is virtually certain that GMSL will continue to rise until at least 2100, because all assessed contributors to GMSL are likely to virtually certain to continue contributing Sea Level                                                              throughout this century. Considering only processes for which projections can be made with at least medium confidence, Global mean sea level (GMSL) rose faster in the 20th century          relative to the period 1995-2014, GMSL will rise by 2050 than in any prior century over the last three millennia                between 0.18 [0.15 to 0.23, likely range] m (SSP11.9) and (high confidence), with a 0.20 [0.15 to 0.25] m rise over              0.23 [0.20 to 0.29, likely range] m (SSP58.5), and by 2100 the period 1901-2018 (high confidence). GMSL rise has                  between 0.38 [0.28 to 0.55, likely range] m (SSP11.9) and accelerated since the late 1960s, with an average rate of              0.77 [0.63 to 1.01, likely range] m (SSP58.5). This GMSL 2.3 [1.6 to 3.1] mm yr -1 over the period 1971-2018 increasing        rise is primarily caused by thermal expansion and mass loss from to 3.7 [3.2 to 4.2] mm yr -1 over the period 2006-2018                glaciers and ice sheets, with minor contributions from changes in (high confidence). New observation-based estimates published          land-water storage. These likely range projections do not include since SROCC lead to an assessed sea level rise over the period        those ice-sheet-related processes that are characterized by deep 1901-2018 that is consistent with the sum of individual components. uncertainty. {9.6.3}
1216
 
Ocean, Cryosphere and Sea Level Change                                                                                                    Chapter 9 Higher amounts of GMSL rise before 2100 could be caused                    At sustained warming levels between 2&deg;C and 3&deg;C, the Arctic by earlier-than-projected disintegration of marine ice shelves,            Ocean will be practically sea ice free throughout September in the abrupt, widespread onset of marine ice sheet instability                most years (medium confidence); there is limited evidence that the and marine ice cliff instability around Antarctica, and faster-            Greenland and West Antarctic ice sheets will be lost almost completely than-projected changes in the surface mass balance and                      and irreversibly over multiple millennia; both the probability of their discharge from Greenland. These processes are characterized by              complete loss and the rate of mass loss will increase with higher deep uncertainty arising from limited process understanding, limited        temperatures (high confidence); about 50 to 60% of current glacier availability of evaluation data, uncertainties in their external forcing    mass outside Antarctica will be lost (low confidence); Northern and high sensitivity to uncertain boundary conditions and parameters.      Hemisphere spring snow cover extent will decrease by up to 30%
In a low-likelihood, high-impact storyline, under high emissions such      relative to 1995-2014 (medium confidence); permafrost volume processes could in combination contribute more than one additional          in the top 3 m will decrease by up to 75% relative to 1995-2014 metre of sea level rise by 2100. {9.6.3, Box 9.4}                          (medium confidence). Committed GMSL rise over 2000 years will be about 4 to 10 m with 3&deg;C of peak warming (medium agreement, Beyond 2100, GMSL will continue to rise for centuries due                  limited evidence). {9.3.1, 9.4.1, 9.4.2, 9.5.1, 9.5.2, 9.5.3, 9.6.3}
to continuing deep-ocean heat uptake and mass loss of the                                                                                            9 Greenland and Antarctic ice sheets, and will remain elevated for            At sustained warming levels between 3&deg;C and 5&deg;C, the Arctic thousands of years (high confidence). Considering only processes            Ocean will become practically sea ice free throughout several months for which projections can be made with at least medium confidence          in most years (high confidence); near-complete loss of the Greenland and assuming no increase in ice-mass flux after 2100, relative to the      Ice Sheet and complete loss of the West Antarctic Ice Sheet will occur period 1995-2014, by 2150, GMSL will rise between 0.6 [0.4 to 0.9,          irreversibly over multiple millennia (medium confidence); substantial likely range] m (SSP11.9) and 1.4 [1.0 to 1.9, likely range] m (SSP58.5). parts or all of Wilkes Subglacial Basin in East Antarctica will be lost By 2300, GMSL will rise between 0.3 m and 3.1 m under SSP12.6,            over multiple millennia (low confidence); 60 to 75% of current glacier between 1.7 m and 6.8 m under SSP58.5 in the absence of marine ice        mass outside Antarctica will disappear (low confidence); nearly all cliff instability, and by up to 16 m under SSP58.5 considering marine      glacier mass in low latitudes, Central Europe, Caucasus, western ice cliff instability (low confidence). {9.6.3}                            Canada and the USA, North Asia, Scandinavia and New Zealand will likely disappear; Northern Hemisphere spring snow cover extent will decrease by up to 50% relative to 1995-2014 (medium Cryospheric Changes and Sea Level Rise                                      confidence); permafrost volume in the top 3 m will decrease by up at Specific Levels of Global Warming                                        to 90% compared to 1995-2014 (medium confidence). Committed GMSL rise over 2000 years will be about 12 to 16 m with 4&deg;C of At sustained warming levels between 1.5&deg;C and 2&deg;C, the Arctic              peak warming and 19 to 22 m with 5&deg;C of peak warming (medium Ocean will become practically sea ice-free in September in some            agreement, limited evidence). {9.3.1, 9.4.1, 9.4.2, 9.5.1, 9.5.2, years (medium confidence); the ice sheets will continue to lose mass        9.5.3, 9.6.3}
(high confidence), but will not fully disintegrate on time scales of multiple centuries (medium confidence); there is limited evidence that the Greenland and West Antarctic ice sheets will be lost almost completely and irreversibly over multiple millennia; about 50 to 60%
of current glacier mass excluding the two ice sheets and the glaciers peripheral to the Antarctic Ice Sheet will remain, predominantly in the polar regions (low confidence); Northern Hemisphere spring snow cover extent will decrease by up to 20% relative to 1995-2014 (medium confidence); the permafrost volume in the top 3 m will decrease by up to 50% relative to 1995-2014 (medium confidence).
Committed GMSL rise over 2000 years will be about 2 to 6 m with 2&deg;C of peak warming (medium agreement, limited evidence).
{9.3.1, 9.4.1, 9.4.2, 9.5.1, 9.5.2, 9.5.3, 9.6.3}
1217
 
Chapter 9                                                                                                                                    Ocean, Cryosphere and Sea Level Change Chapter 9: Ocean, cryosphere and sea level                                                                    Chapter 9: Quick guide Chapter 9 assesses the physical processes underlying global and regional                                      Key topics and corresponding sub-sections changes in the ocean, cryosphere and sea level.
Arctic Section 9.1                                                                                                        9.3.1 l 9.4.1 Introduction Antarctic 9.2.3.2 l 9.3.2 l 9.4.2 Section 9.2                    Section 9.3                  Section 9.4              Section 9.5 Ocean                          Sea ice                      Ice sheets                Glaciers,                    Atlantic Overturning Meridional Circulation permafrost, snow            9.2.3.1 l FAQ 9.3 Section 9.6                                                                                                        Commitment Sea level                                                                                                          9.6.3.5 l FAQ 9.1 Marine extremes Section 9.7 9.6.4 l Box 9.2 Final remarks 9                                                                                                                              Paleo evidence 9.2 l 9.3.1.1 l 9.5.1 l 9.6.2 Sea level contributions 9.2.4 l 9.4 l 9.5.1 l 9.6 l Boxes 9.1, 9.3, 9.4 l CC Box 9.1 l FAQ 9.2 FAQs Warming levels 9.3.1 l 9.4.1 l 9.4.2 l 9.5.1 l 9.5.2 l 9.5.3 l 9.6.3 Boxes                                                                                                                    Cross-chapter boxes Box 9.1                        Box 9.2                    Box 9.3                              Box 9.4                CC Box 9.1 Drivers of sea level            Marine                    Model intercomparisons              High-end              Global energy inventory and sea level budget change                          heatwaves                  of land ice                          sea level rise Figure 9.1 l Visual guide to Chapter 9. Sections dealing with the cryosphere are highlighted with a snowflake.
9.1              Introduction                                                                          There are two major advances of this chapter compared with AR5 and SROCC facilitated by community efforts. The first is the temporal This chapter provides a holistic assessment of the physical processes                                  and spatial increase in observations of both the ocean and the underlying global and regional changes in the ocean, cryosphere and                                    cryosphere (Section 1.5.1.1). In particular, extended observations sea level, as well as improved understanding of observed, attributed                                  have allowed for improved assessment of past change and closure and projected future changes since the IPCC Fifth Assessment Report                                    of both the energy and sea level budgets in a consistent way (AR5) and the Special Report on the Ocean and Cryosphere in                                            (Cross-Chapter Box 9.1) and the sea level budget for the last century a Changing Climate (SROCC; see outline in Figure 9.1). The ocean and                                  (Section 9.6.1.1). Higher resolution observations have revealed the cryosphere (defined as the frozen components of the Earth system                                      details of the Atlantic Meridional Overturning Circulation (AMOC; such as sea ice, ice sheets, glaciers, permafrost and snow) exchange                                  Section 9.2.3.1) and globally resolved glacier changes for the first time heat and freshwater with the atmosphere and each other (Figure 9.2).                                  (Section 9.5.1.1). Improved methodology has resulted in a doubling In a warming climate, the combined effects of thermal expansion of                                    of the assessed level of observed increase in global ocean 0-200 m seawater and melting of the terrestrial cryosphere result in global                                    stratification compared to SROCC assessment (Section 9.2.1.3).
mean sea level rise (Box 9.1).
The second advance is the use of a hierarchy of models and Ocean acidification and deoxygenation are covered in Chapter 5,                                        emulators to update projections of oceanic, cryospheric and sea and regional changes to the ocean and cryosphere are covered in                                        level change arising from Coupled Model Intercomparison Project Chapter 12 and the Atlas. Ecosystem range shifts and climate risk                                      Phase 6 (CMIP6) and related projects (Section 1.5.4.3, Table 1.3, for marine biodiversity associated with ocean change are assessed                                      and Annex II).2 The CMIP6 included an ice-sheet modelling in AR6 Working Group II (WGII). The notion of climate velocity often                                intercomparison for the first time. Particular modelling advances used in impact studies, which is defined as the speed and direction                                    relevant to this chapter are the increase in ocean resolution in the at which a climate variable moves across a corresponding spatial                                      High Resolution Model Intercomparison Project (HighResMIP) and field, is underpinned by the assessment of changes in the physical                                    Ocean Model Intercomparison Project phase 2 (OMIP-2) experiments characteristics of the ocean provided in this chapter.                                                (Sections 1.5.3.1 and 9.2), projections of future glacier (GlacierMIP) and ice sheet (ISMIP6) and Linear Antarctic Response Model 2    In particular, this range of tools leads to advances in the evaluation of confidence in projections. When CMIP6 models are used without additional evidence, the 5-95% confidence range of projections is assigned to a likely range to acknowledge that there are uncertainty sources not reflected by model spread, consistent with Chapter 4.
1218
 
Ocean, Cryosphere and Sea Level Change                                                                                                                        Chapter 9 (a) 9 (b)
Ice sheet Snow Sea ice Permafrost Glaciers Ocean (Speed)
Figure 9.2 l Components of ocean, cryosphere and sea level assessed in this chapter. (a) Schematic of processes (mCDW=modified Circumpolar Deep Water, GIA=Glacial Isostatic Adjustment). White arrows indicate ocean circulation. Pinning points indicate where the grounding line is most stable and ice-sheet retreat will slow.
(b) Geographic distribution of ocean and cryosphere components (numbers indicate glacierized regions (RGI Consortium, 2017)). See Figures 9.20 and 9.21 for labels. Sea ice shaded to indicate the annual mean concentration. Green ocean colours indicate larger surface current speed. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1219
 
Chapter 9                                                                                                                                Ocean, Cryosphere and Sea Level Change Intercomparison Project (LARMIP-2) response from multi-model                                          There are other advances in scientific understanding. In the cryosphere, studies (Sections 9.5.1 and 9.4, and Box 9.3), and new methods to                                      this chapter assesses how fast-responding elements (sea ice, permafrost synthesize ocean and cryosphere models into sea level projections for                                  and snow; Sections 9.3, 9.5.2 and 9.5.3) track warming levels across all Shared Socieo-economic Pathway scenarios (SSPs; Sections 1.6.1,                                    observations and projections independent of scenario, process 9.4.1.3, 9.4.2.5 and 9.6.3, and Cross-Chapter Box 1.4) and warming                                    understanding of uncertainty in Antarctic Ice Sheet projections levels (Sections 9.6.3 and 1.6.2, and Cross-Chapter Box 11.1).                                        (Section 9.4.2 and Box 9.4) and new insight into thresholds for Arctic In particular, sea level projections and the individual contributions                                  sea ice (Section 9.3.1.1) and Greenland and West Antarctic ice sheets (Section 9.6.3.3) are consistent with equilibrium climate sensitivity                                  (Sections 9.4.1.4 and 9.4.2.6). In the ocean, process understanding and surface temperature assessments across this Report (Box 4.1 and                                    of ocean heat uptake (Section 9.2.2.1 and Cross-Chapter Box 5.3)
Cross-Chapter Box 7.1).                                                                                and observed changes in ocean stratification (Section 9.2.1.3) have implications for ocean biogeochemistry are also important.
9      Box 9.1 l Key Processes Driving Sea Level Change Sea level change arises from processes acting on a range of spatial and temporal scales, in the ocean, cryosphere, solid Earth, atmosphere and on land (Figure 9.2). Relative sea level (RSL) change is the change in local mean sea surface height relative to the sea floor, as measured by instruments that are fixed to the Earths surface (e.g., tide gauges). This reference frame is used when considering coastal impacts, hazards and adaptation needs. In contrast, geocentric sea level change is the change in local mean sea surface height with respect to the terrestrial reference frame, and is the sea level change observed with instruments from space.
This box provides a brief summary of sea level processes using standard terminology (Gregory et al., 2019).
Global processes Global mean sea level change (Sections 9.6 and 2.3.3.3) is the change in volume of the ocean divided by the ocean surface area.
It is the sum of changes in ocean density (global mean thermosteric sea level change) and changes in the ocean mass as a result of changes in the cryosphere or land-water storage (barystatic sea level change).
Steric sea level change is caused by changes in the ocean density and is composed of thermosteric sea level change and halosteric sea level change. Thermosteric sea level change (also referred to as thermal expansion) occurs as a result of changes in ocean temperature: increasing temperature reduces ocean density and increases the volume per unit of mass. Halosteric sea level change occurs as a result of salinity variations: higher salinity leads to higher density and decreases the volume per unit of mass. Although both processes can be relevant on regional to local scales, thermosteric changes contribute to global mean sea level change, whereas global mean halosteric change is negligible (Gregory et al., 2019). There is high confidence in the understanding of processes causing thermosteric sea level change (Section 9.2.4.1).
The Greenland and Antarctic ice sheets are the largest reservoirs of frozen freshwater and therefore potentially the largest contributors to sea level rise. Fluctuations in ice-sheet volume arise from the imbalance between accumulation (either at the ice-sheet surface or on the underside of ice shelves) and loss from sublimation, surface and basal melting, and iceberg calving. Ice sheets discharge the majority of their mass through marine-terminating ice streams that are in some cases buttressed by floating ice shelves. Changes in the thickness and extent of the ice shelves due to melt from below, calving, or disintegration, as a result of surface meltwater penetrating crevasses, can affect the flow of the inland ice streams. There is medium confidence in ice-sheet processes but low confidence in their forcing (ocean changes and ice-shelf collapse) and in instability processes (Sections 9.4.1 and 9.4.2).3 Glaciers contribute to sea level change via an imbalance between mass gain and mass loss processes, which leads to adjustments in the glacier geometry over an extended period of time, called the response time. The response time may range from a few years to a few hundred years. The glacial meltwater does not all flow immediately into the ocean: it can refreeze, feed rivers (where it may be extracted for domestic use), evaporate, or be stored in (proglacial) lakes or closed basins. There is medium to high confidence in the understanding of processes leading to sea level contributions from glaciers (Section 9.5.1).
Land-water storage includes surface water, soil moisture, groundwater storage and snow, but excludes water stored in glaciers and ice sheets. Changes in land-water storage can be caused either by direct human intervention in the water cycle (e.g., storage of water in reservoirs by building dams in rivers, groundwater extraction for consumption and irrigation, or deforestation) or by climate variations (e.g., changes in the amount of water in internally drained lakes and wetlands, the canopy, the soil, the permafrost and 3    The conversion of land ice mass loss to global mean sea level rise used in this Report - the sea level equivalent (SLE) - is 362.5 gigatonnes (Gt) of ice loss for 1 mm of sea level rise.
1220
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 Box 9.1 (continued) the snowpack). Land-water storage changes caused by climate variations may be indirectly affected by anthropogenic influences.
It is difficult to assign a single confidence level to land-water storage as understanding can vary from low confidence in groundwater recharge processes to high confidence in water storage via snowpack changes (Sections 8.2.3 and 8.3.1.7).
Regional and local processes Ocean dynamic sea level change refers to the change in mean sea level relative to the geoid and is associated with the circulation and density-driven changes in the ocean. Ocean dynamic sea level change varies regionally but by definition has a zero global mean. It includes the depression of the sea surface by atmospheric pressure. There is medium confidence in the understanding of ocean processes leading to dynamic sea level change (Section 9.2.4.2).
Changes in Earth gravity, Earth rotation and viscoelastic solid Earth deformation (GRD) - result from the redistribution of mass between terrestrial ice and water reservoirs and the ocean. Contemporary terrestrial mass loss leads to elastic solid Earth uplift          9 and a nearby RSL fall. (For a single source of terrestrial mass loss, this is within about 2000 km; for multiple sources, the distance depends on the interaction of the different RSL patterns.) Farther away (around more than 7000 km for a single source of terrestrial mass loss), RSL rises more than the global average, due to first-order gravitational effects. Earth deformation associated with adding water to the ocean and a shift of the Earths rotation axis towards the source of terrestrial mass loss leads to second-order effects that increase spatial variability of the pattern globally. GRD effects due to the redistribution of ocean water within the ocean itself are referred to as self-attraction and loading effects. There is high confidence in the understanding of GRD processes.
Glacial isostatic adjustment is ongoing GRD in response to past changes in the distribution of ice and water on Earths surface.
On a time scale of decades to tens of millennia following mass redistribution, Earths mantle flows viscously as it evolves toward isostatic equilibrium, causing solid Earth movement and geoid changes, which can result in regional to local sea level variations. There is medium confidence in the understanding of glacial isostatic adjustment processes.
Vertical land motion is the change in height of the land surface or the sea floor and can have several causes in addition to elastic deformation associated with contemporary GRD and viscoelastic deformation associated with glacial isostatic adjustment. Subsidence (sinking of the land surface or sea floor) can occur through compaction of alluvial sediments in deltaic regions, removal of fluids such as gas, oil, and water, or drainage of peatlands. Tectonic deformation of the Earths crust can occur as a result of earthquakes and volcanic eruptions. There is medium confidence in the understanding of vertical land motion processes.
Extreme sea level is an exceptionally low or high local sea surface height arising from combined short-term phenomena (e.g., storm surges, tides and waves). RSL changes affect extreme sea levels directly by shifting the mean water levels, and indirectly by modulating the depth for propagation of tides, waves and/or surges. Extreme sea levels can be influenced by changes in the frequency, tracks, or strength of weather systems, or anthropogenic changes such as dredging. Extreme still water level refers to the combined contribution of RSL change, tides and storm surges. Wind-generated waves also contribute to coastal sea level. Extreme total water level is the extreme still water level plus wave setup (time-mean sea level elevation due to wave energy dissipation). When considering coastal impacts, swash (vertical displacement up the shore-face induced by individual waves) is also important and included in Extreme coastal water level. There is low to medium confidence in the understanding of extreme sea level processes (Sections 9.6.4 and 12.4).
9.2          Oceans                                                        with buoy-based observations and improved treatment of sea ice, have had important consequences for key climate change indicators 9.2.1        Ocean Surface                                                  such as global mean surface temperature (GMST), global surface air temperature (GSAT), and SST (Cross-Chapter Box 2.3). The AR5 9.2.1.1      Sea Surface Temperature                                        assessment is confirmed, and it is now very likely that global mean SST changed by 0.88 [0.68 to 1.01] &deg;C from 1850-1900 to 2011-2020, and The IPCC Fifth Assessment Report (AR5; Hartmann et al., 2013)              0.60 [0.44 to 0.74] &deg;C from 1980 to 2020 (Figure 9.3 and Table 2.4).
assessed that it is virtually certain that global sea surface temperature (SST) has increased since the beginning of the 20th century (very high      Regions vary in the rate of SST warming, with slight cooling in some confidence). The Special Report on Ocean and Cryosphere in                  regions (Figure 9.3). The SROCC (Collins et al., 2019) and Section 7.4.4 a Changing Climate (SROCC) did not assess past SST change. Since            assess SST changes over specific regions, which are consistent with the AR5, improvements in the understanding of recent SST biases in the          changes reported here. The tropical ocean has been warming faster observational records, especially extending ship-based observations        than other regions since 1950, with the fastest warming in regions 1221
 
Chapter 9                                                                                                                Ocean, Cryosphere and Sea Level Change of the tropical Indian and western Pacific oceans (Figure 9.3), due to                      subpolar North Atlantic Ocean and Southern Ocean have warmed a combination of local atmosphere-ocean coupling, the Indonesian                            more slowly than the global average or cooled (Figure 9.3). Surface Throughflow (Section 9.2.3.4 and Figure 9.11), and trends in the                            warming in the subpolar Southern Ocean has been slower than Walker circulation (Sections 2.3.1.4.1 and 3.3.3.1, and Figure 3.16).                        the global average since the 1950s, and this pattern is consistent The western boundary currents of the subtropical gyres have warmed                          with the upwelling around Antarctica renewing surface water with faster than the global mean over the past century. There remains low                        pre-industrial, deeper water masses (Section 9.2.3.2; Frlicher agreement in the changes of the location and the dynamical changes                          et al., 2015; J. Marshall et al., 2015; Armour et al., 2016). New evidence in western boundary current extensions (Sections 2.3.3.4.2 and                              since SROCC (Meredith et al., 2019) confirms slight cooling since 9.2.3.4, and Figure 9.3). In the Arctic, the mean SST increase over the                      the 1980s around the subpolar Southern Ocean, contrasting with last two decades is similar to, or only slightly higher than, the global                    marked warming directly northward of it (Section 9.2.3.2; Haumann average (J.-L Chen et al., 2019). In contrast, the eastern Pacific Ocean,                    et al., 2020; Rye et al., 2020; Auger et al., 2021). In eastern boundary Sea surface temperature (SST) anomalies and maps 9      Observation-based estimated and CMIP6 multi-model means, biases and projected changes 5                                                                                                                                    15 Extended Paleo GMSST                  Global Mean SST (GMSST); modern history and model projections to 2100 4  Mean & 95% conf.                                                                                                                                                projections interval 4                                                                                          SSP5-8.5 22 SSP5-8.5 3 2
10 3                                                                                  SSP3-7.0 19 o                                                Mean GMSST C
0                      &deg;C                    CMIP6 Likely (17-83%) ranges
                                                                                                                                                                        &deg;C 2
Observations HighResMIP                                                                  5
          -2 Models (PMIP) 1                                                    (historical /SSP5-8.5 )
Observational                    9              5                        SSP2-4.5 23                      SSP1-2.6 3 CMIP                                                                          SSP1-2.6 23 40 models        Reanalyses                                                                            CMIP6
          -4                        0                                                                                                                  2100 means &
(very) likely Baseline Period                                        ranges 0
1850                    1900                    1950                  2000            2050                    2100                    2100          2300 Observation-based                                                                                                Color High model agreement (80%)
Observation-based                                                Low model agreement (<80%)
SSP5-8.5 SSP5-8.5 Figure 9.3 l Sea surface temperature (SST) and its changes with time. (a) Time series of global mean SST anomaly relative to 1950-1980 climatology. Shown are paleoclimate reconstructions and PMIP models, observational reanalyses (HadISST) and multi-model means from the Coupled Model Intercomparison Project (CMIP) historical simulations, CMIP projections, and HighResMIP experiment. (b) Map of observed SST (1995-2014 climatology HadISST). (c) Historical SST changes from observations.
(d) CMIP 2005-2100 SST change rate. (e) Bias of CMIP. (f) CMIP change rate. (g) 2005-2050 change rate for SSP58.5 for the CMIP ensemble. (h) Bias of HighResMIP (bottom left) over 1995-2014. (i) HighResMIP change rate for 1950-2014. (j) 2005-2050 change rate for SSP58.5 for the HighResMIP ensemble. No overlay indicates regions with high model agreement, where 80% of models agree on sign of change. Diagonal lines indicate regions with low model agreement, where <80% of models agree on sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1222
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 upwelling systems, SROCC (Bindoff et al., 2019) reported low                direction of SST changes in observations are consistent with CMIP6 agreement between SST trends in recent decades, due to varying              only when including internal variability (Olonscheck et al., 2020).
spatio-temporal resolution and interannual to multi-decadal                This is notably the case in the equatorial Pacific, North Atlantic, variability. Satellite evidence not included in SROCC shows that            and Southern Ocean, which are regions where SST is of known 92% of these regions warmed more slowly than neighbouring                  importance in controlling heat uptake (Section 9.2.2.1) and the global offshore locations between 1982 and 2015, so upwelling may buffer          radiative feedback parameter (Section 7.4.4.3). Overall, despite some the near shore from warming (Section 9.2.3.5; Varela et al., 2018).        persistent regional biases, CMIP6 coupled climate models reproduce Coupled ocean-atmospheric modes of variability strongly affect              the observed SST trends or high internal variability over the past regional SST (Cross-Chapter Box 3.1 and Annex IV). In summary,              century over a range of different multi-decadal periods (Figure 9.3; a positive SST trend since 1950 is evident globally, but there is very      Olonscheck et al., 2020; Watanabe et al., 2021), highlighting their skill high confidence that the Indian Ocean, western equatorial Pacific          to inform future large-scale SST changes at regional scale. Warming Ocean, and western boundary currents have warmed faster than the            is projected at varying rates in all regions by 2050, except the North global average, while the Southern Ocean, the eastern equatorial            Atlantic Subpolar Region, the equatorial Pacific, and the Southern Pacific, and the North Atlantic Ocean have warmed more slowly, or          Ocean where models disagree (high confidence).
have slightly cooled.                                                                                                                                9 It is virtually certain that SST will continue to increase in the In AR5 (Flato et al., 2013), a marginal improvement was noted in            21st century, at a rate depending on future emissions scenarios.
Coupled Model Intercomparison Project Phase 5 (CMIP5) climate              The future global mean SST increase projected by CMIP6 models model SST biases compared to Phase 3 (CMIP3) models in AR4, with            for the period 1995-2014 to 2081-2100 is 0.86 [5-95% range:
a reduction in the magnitude of biases. The AR5 noted that, in several      0.43-1.47] &deg;C under SSP12.6, 1.51 [1.02 to 2.19] &deg;C under SSP24.5, regions, large SST biases are symptomatic of errors in the representation  2.19 [1.56 to 3.30] &deg;C under SSP37.0, and 2.89 [2.01 to 4.07] &deg;C under of important processes, such as dynamics in the equatorial Pacific and      SSP58.5 (Figure 9.3). While under SSP12.6, the CMIP6 ensemble North Atlantic, and Southern Ocean. Common regional biases in SST          consistently projects that it is very likely at least 83% of the world or historical SST trends are not exclusively linked to the representation  ocean surface will have warmed by 2100, and under SSP58.5, at least of the ocean (high confidence), but can have multiple causes,              98% of the world ocean surface will have warmed. The spatial pattern including: errors in the representation of long-term historical trends      of future change is consistent with observed SST change over the in equatorial winds (Section 9.2.1.2); misrepresentation of the forced      20th century, though with notable regional differences (Figure 9.3).
equatorial ocean response (Karnauskas et al., 2012; Kohyama et al.,        Long-term change in SST patterns is important for regional impacts 2017; Coats and Karnauskas, 2018); thermocline depth errors (Linz          but also affects radiative feedbacks, and therefore long-term change et al., 2014); errors in atmospheric model cloud-related shortwave          in climate sensitivity (Section 7.4.4.3). In the Southern Ocean, CMIP6 radiation (Hyder et al., 2018); biases in ocean circulation variability (C. models project that SSTs will eventually consistently increase in the Wang et al., 2014); and deficiencies in upper ocean (Q. Li et al., 2019)    21st century, at a rate dependent on future scenarios (Figure 9.3 and atmospheric (Bates et al., 2012) boundary layer parametrizations.      and Section 9.2.3.2; Bracegirdle et al., 2020). Yet, there is only low In CMIP6, the mid-latitude biases in the Northern Hemisphere are            confidence that this Southern Ocean warming will emerge by the improved in the multi-model mean, and the inter-model standard              end of the century (Section 7.4.4.1), due to the inconsistent historical deviation of the zonal mean SST error is significantly decreased            and near-term simulations and observations over the 20th century in the northern Hemisphere south of 50&deg;N compared to CMIP5,                (Figure 9.3). Furthermore, the equilibrium SST pattern from proxy though biases in equatorial regions remain essentially unchanged            records or simulated by climate models under CO2 forcing stand in (Section 3.5.1.1 and Figures 3.23, 3.24 and 9.3). Some long-standing        contrast with the cooling trends in the Southern Ocean observed over ocean model biases have been reduced through increases in model            the past decades (Section 7.4.4.1.2). Similarly, the SST change pattern resolution in CMIP6 (Bock et al., 2020) and improved parametrizations      observed in the tropical Pacific Ocean will transition on centennial (Fox-Kemper et al., 2011; Q. Li et al., 2016; Qiao et al., 2016; Reichl    time scales to a mean pattern resembling the El Nino pattern (medium and Hallberg, 2018). The High Resolution Model Intercomparison              confidence) (Annex IV). However, it is difficult to delineate a climate Project (HighResMIP) ensemble (Figure 9.3) has smaller cold biases          change trend ressembling an El Nino pattern and El Nino variability in the North Atlantic and the tropical Pacific, and smaller warm biases    (Wittenberg, 2009; Collins et al., 2010) without large ensembles in the upwelling regions off the western coasts of Africa, North and        (Kay et al., 2015). Several Pliocene SST reconstructions indicate South America (Roberts et al., 2018, 2019; Caldwell et al., 2019;          enhanced warming in the centre of the eastern Pacific equatorial Docquier et al., 2019). In summary, CMIP6 models show persistent            cold tongue upwelling region, consistent with reconstruction of regional biases in representing the climatological SST state (very high    enhanced subsurface warming and enhanced warming in coastal confidence), but higher resolution reduces some biases, particularly        upwelling regions (Section 7.4.4.2.2). The North Atlantic subpolar in the North Atlantic and eastern boundary upwelling systems                gyre is projected to continue to warm more slowly than surrounding (Figure 9.3; high confidence).                                              regions (Suo et al., 2017), as the Gulf Stream concurrently warms rapidly (Figure 9.3; Cheng et al., 2013) and the Atlantic Meridional The CMIP6 models represent the observed trends in SST patterns with        Overturning Circulation further declines under greenhouse gas forcing, greater fidelity than CMIP5, with the ocean area that is inconsistent      although models disagree about the rate of change (Figure 9.3 and with the observed trends decreasing by about three quarters from            Section 9.2.3.1). In summary, CMIP6 models show a future pattern of CMIP5 to CMIP6 (Olonscheck et al., 2020). In some regions, the              SST change comparable to historical trends with intensity depending 1223
 
Chapter 9                                                                                                                  Ocean, Cryosphere and Sea Level Change on future emissions scenario, and some of the observed cooling trends                    The SROCC (Abram et al., 2019) and AR5 (Rhein et al., 2013) assessed over the 20th century will eventually transition to a warming SST                        that observations of air-sea fluxes had not yet reached the density on centennial time scales, in particular in the Southern Ocean (high                      or accuracy to directly detect trends beyond the noise. New evidence confidence) and in the equatorial Pacific (medium confidence), while                      since SROCC confirms that direct heat and freshwater flux trends the North Atlantic subpolar gyre will continue to warm more slowly                        have not emerged yet as spatial (Figure 9.4), annual (Yu, 2019), and than the global average (high confidence).                                                decadal (Zanna et al., 2019) variability overwhelm detection. Since AR5, comprehensive comparisons (Bentamy et al., 2017; Valdivieso 9.2.1.2      Air-Sea Fluxes                                                              et al., 2017; Yu et al., 2017) have used updated and new surface flux products to improve surface flux uncertainty estimates, and Air-sea fluxes of energy, freshwater, and momentum (wind stresses)                        these comparisons note that implied global energy imbalances are difficult to observe directly (Cronin et al., 2019), so estimates of                  often exceed the observed ocean warming. Flux estimates using the global mean net air-sea heat flux are inferred from observed                          top of atmosphere observations and atmospheric fluxes from ocean warming (Section 2.3.3.1, Box 7.2, and Cross-Chapter Box 9.1).                      reanalysis have improved over past products (Trenberth and Fasullo, Air-sea heat fluxes resemble the warming patterns of CMIP3                                2018) but require consistency adjustments (Trenberth et al., 2019) 9 (Domingues et al., 2008; Levitus et al., 2012) and are consistent with                    as the energy budget is not closed. Adjustments are needed for the ensemble mean warming rate of CMIP5 (Cheng et al., 2017,                              all flux products, and they remain less accurate than direct ocean 2019) and CMIP6 models (Section 3.5.1.3). Regional air-sea fluxes                        heat content change measurements (Cheng et al., 2017). Some in models remain a key driver of uncertainty (Huber and Zanna,                            regional changes are likely robust in both satellite observations and 2017; Tsujino et al., 2020). A substantial part of the upper 700 m                        projections (Figure 9.4). Recent satellite-based surface flux products energy increase is very likely attributed to anthropogenic forcing via                    with improved retrieval algorithms and new satellites, for example, increasing radiative forcing (Sections 3.5.1.3, 7.2 and 7.3).                            J-OFURO3 (Tomita et al., 2019) and OAFlux-HR (Yu, 2019), provide Figure 9.4 l Global maps of observed mean fluxes (a, d, g), the observed trends in these fluxes (b, e, h) and the projected rate of change in these fluxes from SSP58.5 (c, f, i). Shown are the freshwater flux (a-c), net heat flux (d-f), and momentum flux or wind stress magnitude (g-i), with positive numbers indicating ocean freshening, warming, and accelerating respectively. The means and observed trends are calculated between 1995-2014 (freshwater and wind stress) or 2001-2014 (heat).
The SSP58.5 projected rates are between 1995-2100 using 20-year averages at each end of the time period. Observations show objective interpolation from Clouds and the Earths Radiant Energy System (CERES) Energy Balanced and Filled (EBAF) v4 (Kato et al., 2018), Objectively Analyzed air-sea Fluxes-High Resolution (OAFlux-HR) (Yu, 2019),
and Global Precipitation Climatology Project (GPCP) (Adler et al., 2003) of fluxes and flux trends (b, e, h). Observed trends with no overlay indicate regions where the trends are significant at p = 0.34 level. Crosses indicate regions where trends are not significant. For (c, f, i) projections, no overlay indicates regions with high model agreement, where 80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where <80% of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1224
 
Ocean, Cryosphere and Sea Level Change                                                                                                Chapter 9 a complete suite of turbulent fluxes including heat, moisture, and    Air-sea flux biases result from common causes in most models, and momentum. When combined with satellite-based surface radiation        many are the same as during AR5 (Rhein et al., 2013). Important from Clouds and the Earths Radiant Energy System (CERES) Energy      currents (e.g., Gulf Stream, Kuroshio, Antarctic Circum-polar Current Balanced and Filled (EBAF; Kato et al., 2018) and precipitation from  patterns) are often found in erroneous locations in models, affecting Global Precipitation Climatology Project (GPCP; Adler et al., 2003),  SST and flux signatures (Bates et al., 2012; Beadling et al., 2020; J.-
full ocean-surface forcing is available since 1987 (Figure 9.4). These L.F. Li et al., 2020), but their locations are improved in high-resolution products agree with sparse buoy and ship observations within          ocean models (Chassignet et al., 2017, 2020; Hewitt et al., 2020),
30 W m-2 (Bentamy et al., 2017; Cronin et al., 2019). While patterns  and high-resolution coupled models reduce the mean air-sea flux agree between models and satellites in net fluxes (Figure 9.4),        biases (Delworth et al., 2012; Sakamoto et al., 2012; Small et al.,
the trend magnitudes are substantially weaker in models. The          2014; Haarsma et al., 2016; Caldwell et al., 2019; L.C Jackson fluxes tending to warm the North Atlantic and Southern Ocean          et al., 2020). Oceanic variability stems either from internal chaotic are consistent with the largest changes observed in the surface        variability or atmospheric forcing (Hasselmann, 1976; S&#xe9;razin properties and water masses (Sections 9.2.1.1, 9.2.2.1 and 9.2.2.3). et al., 2016, 2017). Large-scale variability in the ocean tends to The observed trend toward a saltier Atlantic Ocean and a fresher      follow atmospheric forcing in low-resolution models, while in high-Indian Ocean, as well as trends in evaporation minus precipitation    resolution coupled models ocean variability drives atmospheric            9 (E-P) patterns in the equatorial Pacific (see also Section 8.3.1)      variability on small scales (Bishop et al., 2017; Small et al., 2019),
enhance the present mean pattern of wetting and drying. Elsewhere      allowing these high-resolution models to mimic the coupling with patterns are less clear, with only partial, large-scale agreement      clouds, precipitation, and atmospheric and oceanic boundary layers with the wet gets wetter simplification (Sections 3.3.2.3, 4.4.1    apparent in observations (Chelton and Xie, 2010; Frenger et al.,
and 4.5.1). In summary, globally integrated and large-scale fluxes    2013). Even coarse-resolution models, such as the ocean and sea ice are more reliably inferred from heat content and salinity change,      components used in CMIP6, show significant sensitivity in the mean while regional trends are rarely robust in observations; where they    and variability of SST and sea ice to modest changes in flux forcing are robust, they tend to be underestimated or in disagreement          (Tsujino et al., 2020). Finally, there is still considerable disagreement in models (very high confidence).                                      between different parametrizations of air-sea fluxes used in models and strong scatter in direct observations (Renault et al., 2016; There is low confidence in long-term wind stress trends in most        Brodeau et al., 2017). In summary, there is very high confidence that regions, but a few locations have likely trends over the scatterometer air-sea heat flux and stress biases are reduced in coupled models era and in projections, as shown in Figure 9.4 (Desbiolles et al.,    with high ocean resolution over coarse-resolution models, although 2017; Young and Ribal, 2019; Yu, 2019). The AR5 (Rhein et al., 2013)  the effect on trends remain unclear.
assessed with medium confidence that zonal wind stress over the Southern Ocean increased from the early 1980s to the 1990s (medium    9.2.1.3        Upper-ocean Stratification and Surface Mixed Layers confidence) (Figure 9.4). Over 1995-2014, the zonal wind stress over the Southern Ocean continued to increase, westerly winds in      The density difference from surface to deep ocean is the upper-ocean the North Pacific and North Atlantic weakened, while the easterly      stratification. The AR5 (Rhein et al., 2013) assessed that it is very equatorial Pacific winds of the Walker circulation strengthened        likely that the thermal contribution to stratification over the fixed (Figure 9.4). In historical simulations, CMIP5 models projected        0-200 m layer increased by about 1% per decade between 1971 and annular modes (Annex IV) to move poleward and strengthen in both      2010 (based on linear trend consistently across reports). The SROCC hemispheres (Yang et al., 2016), while in CMIP6 models westerlies      (Bindoff et al., 2019) found it very likely that density stratification only strengthen over the Southern Ocean, with a weaker trend than      increased by 0.46-0.51% per decade between 60&deg;S and 60&deg;N from recently observed (Figure 9.4 and Sections 4.5.1 and 4.5.3). In the    1970 to 2017). New published estimates based on a variety of tropical Pacific Ocean, a weakening trend in easterly winds and        different interpolated observations show that SROCC assessed rate Walker circulation in the 20th century has been inferred based on      is too low, even using the same data and methods (Li et al., 2020).
observed sea level pressure data (Vecchi et al., 2006; Vecchi and      The 1960-2018 stratification increase is estimated at 1.2 +/- 0.1%
Soden, 2007) and coral proxies (Carilli et al., 2014) and is projected per decade from the IAP dataset, 1.2 +/- 0.4% per decade from the to continue by CMIP6 models (Figure 9.4). Yet, over 1995-2014          Ishii product, 0.7 +/- 0.5% per decade from the EN4 dataset, 0.9 observed winds have strengthened (Figure 9.4). The observed            +/- 0.5% per decade from ORAS4, and 1.2 +/-0.3% per decade from the strengthening may have been influenced by a combination of factors    National Centers for Environmental Information (NCEI) dataset (G. Li (Section 7.4.4.2.1), but there is low confidence in the attribution of et al., 2020). The improved methodology for computing stratification this signal to anthropogenic warming (Section 3.3.3.1) and medium      change on individual profiles before gridding yields a global annual confidence that it reflects internal variability (Section 8.3.2.3). mean increase of 0-200 m stratification change of 0.8 +/- 0.2% per Near-term projected changes over the Southern Ocean result from        decade between 1960 and 2018 (Yamaguchi and Suga, 2019) and ozone recovery and greenhouse gases (Sections 4.3.3 and 4.4.3).        a global summer mean increase of 0-200 m stratification change Overall, there is only low confidence in observed and projected wind  of 1.3 +/- 0.3% per decade between 1970 and 2018 (Sall&#xe9;e et al.,
stress trends in most regions because trends in oceanic wind stresses  2021) is of a similar magnitude to the long-term trend (Yamaguchi during the satellite era have not emerged or are inconsistent with    and Suga, 2019; G. Li et al., 2020). In summary, there is limited historical simulated changes.                                          evidence that focusing on changes over a fixed depth range might 1225
 
Chapter 9                                                                                                              Ocean, Cryosphere and Sea Level Change Color High model agreement (80%)
Low model agreement (<80%)
SSP1-2.6                              SSP5-8.5 m                                        m                                                            m 9
m                                        m                                                            m Figure 9.5 l Mixed-layer depth in (a-d) winter and (e-h) summer. (a, e) Observed climatological mean mixed-layer depth (based on density threshold) from the Argo Mixed Layer Depth Climatology (Holte et al., 2017) using observations for 2000-2019. (b, f) Bias between the observation-based estimate (2000-2019) and the 1995-2014 Coupled Model Intercomparison Project Phase 6 (CMIP6) climatological mean mixed-layer depth. (c, d, g, h) Projected mixed-layer depth (MLD) change from 1995-2014 to 2081-2100 under (c, g) SSP12.6 and (d, h) SSP58.5 scenarios. The (a-d) winter row shows December-January-February (DJF) in the Northern Hemisphere and June-July-August (JJA) in the Southern Hemisphere; the (e-h) summer row shows JJA in the Northern Hemisphere and DJF in the Southern Hemisphere. The mixed-layer depth is the depth where the potential density is 0.03 kg m-3 denser than at 10 m. No overlay indicates regions with high model agreement, where 80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where <80% of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
hide larger increases occurring at the seasonally and regionally                          The SROCC assessed that upper-ocean stratification will continue variable pycnocline depth. There is also limited evidence that summer                    to increase in the 21st century under increased radiative forcing (high stratification change within the pycnocline has occurred at a rate of                    confidence), due to increased surface temperature and high-latitude 8.9 +/- 2.7% per decade from 1970 to 2018, and limited evidence of                          surface freshening (Bindoff et al., 2019). New climate model a winter pycnocline stratification increase (Cummins and Ross, 2020;                      simulations concur with SROCC assessment of a future increase of Sall&#xe9;e et al., 2021).                                                                    the 0-200 m stratification under increased radiative forcing in all regions of the world ocean (Kwiatkowski et al., 2020). In addition, While AR5 and SROCC did not assess change in mixed-layer depth, the                      CMIP6 climate models project a shallowing of the mixed-layer reported changes in stratification can modulate the surface mixed-                        in summer and winter by the end of the century under increased layer depth, which is set by a balance between fluxes and dynamical                      radiative forcing (Figure 9.5; Kwiatkowski et al., 2020), with the mixing (winds, tides, waves, convection) acting against the background                    exception of the Arctic showing deepening of the mixed layer as stratification and restratification processes (solar and dynamical).                      a result of sea ice retreat (Figure 9.5; Lique et al., 2018). The regions Despite the large stratification increase observed at a global scale,                    of largest shallowing are associated with the deepest climatological new evidence shows that summer mixed-layer depth deepened                                mixed layer, in both winter and summer, particularly affecting the consistently over the globe at a rate of 2.9 +/- 0.5% per decade from                      North Atlantic and the Southern Ocean basins (Figure 9.5). While 1970 to 2018, with the largest deepening observed in the Southern                        CMIP6 models tend to project shallowing mixed layers under Ocean, corresponding to overall deepening from 3-15 m per decade                          a warming climate, except at high latitudes (Figure 9.5; Lique et al.,
depending on region (Somavilla et al., 2017; Sall&#xe9;e et al., 2021). While                  2018; Kwiatkowski et al., 2020), a deepening in the summer mixed-the shorter observational record in winter (compared to summer)                          layer depth by intensification of the surface winds and storms may does not allow global winter mixed-layer trends to be reliably                            explain inconsistency among models in many regions (Figure 9.5; assessed (Sall&#xe9;e et al., 2021), winter mixed-layer depths deepening                      Young and Ribal, 2019), although model mixed-layer biases are at rates of 10 m per decade have been reported at individual                              large in the summer in the Southern Ocean (Belcher et al., 2012; long-term mid-latitude monitoring sites (Somavilla et al., 2017).                        Sall&#xe9;e et al., 2013a; Q. Li et al., 2016; Tsujino et al., 2020). Lack of Projections agree that shoaling of mixed-layer depth is expected in                      observed ocean turbulence and climate model limitations do not allow the 21st century, but only for strong emissions scenarios, and only                      for direct assessment of ocean surface turbulence change and limit in some regions (Figure 9.5). In summary, there is limited observational                  confidence in past and future mixed-layer change. Understanding evidence that the mixed layer is globally deepening, while models                        of turbulent processes, their representation in ocean and climate show no emergence of a trend until later in the 21st century under                        models, and their effect on mixed-layer biases have been an active strong emissions.                                                                        and rapidly evolving topic of research since AR5 (Buckingham et al.,
2019; Q. Li et al., 2019). Small-scale mixed-layer processes are 1226
 
Ocean, Cryosphere and Sea Level Change                                                                                                    Chapter 9 not resolved in climate models (DAsaro, 2014; Buckingham et al.,            Boucher et al., 2020; Danabasoglu et al., 2020; Dunne et al., 2020; 2019; McWilliams, 2019) and despite significant improvements                  Kelley et al., 2020). In summary, the representation of upper-ocean in their parametrization over the last decade (Fox-Kemper et al.,            stratification and mixed layers has improved in CMIP6 compared 2011; Jochum et al., 2013; Q. Li et al., 2016, 2019; Qiao et al., 2016)      to CMIP5. While it is virtually certain that the global mean upper and significant improvement in some models (Li and Fox-Kemper,                ocean will continue to stratify in the 21st century, there is only 2017; Dunne et al., 2020), biases in mixed-layer representation              low confidence in the future evolution of mixed-layer depth, generally persist (Heuz&#xe9;, 2017; Williams et al., 2018; Cherchi et al.,        which is projected to mostly shoal under high emissions, except in 2019; Golaz et al., 2019; Voldoire et al., 2019; Yukimoto et al., 2019;      high-latitude regions where sea ice retreats.
Box 9.2 l Marine Heatwaves Marine heatwaves (MHW) are periods of extreme high sea temperature relative to the long-term mean seasonal cycle (Hobday et al.,
2016). Studies since the Special Report on the Ocean and Cryosphere in a Changing Climate (SROCC; Collins et al., 2019) confirm the              9 assessment that MHW can lead to severe and persistent impacts on marine ecosystems - from mass mortality of benthic communities, including coral bleaching, changes in phytoplankton blooms, shifts in species composition and geographical distribution, and toxic algal blooms, to decline in fisheries catch and mariculture (Smale et al., 2019; Cheung and Frlicher, 2020; Hayashida et al., 2020; Piatt et al.,
2020). Unlike synoptic atmospheric heatwaves (Section 11.3), MHWs can extend for millions of square kilometres, persist for weeks to months, and occur at subsurface (Bond et al., 2015; Schaeffer and Roughan, 2017; Perkins-Kirkpatrick et al., 2019; Laufktter et al., 2020).
The SROCC established that MHWs have occurred in all basins over the last decades. Additional evidence documenting widespread occurrence of marine heat waves in all basins and marginal seas continues to accumulate (Y. Li et al., 2019; Yao et al., 2020). The SROCC highlighted the role of large-scale climate modes of variability in amplifying or suppressing MHW occurrences, which has since been further corroborated, increasing confidence in climate modes as important drivers of MHWs (Holbrook et al., 2019; Sen Gupta et al., 2020). More generally, understanding of processes leading to MHWs has increased since SROCC, including air-sea heat flux (Section 9.2.1.2), increased horizontal heat advection, shoaling of the mixed-layer and suppressed mixing processes (Section 9.2.1.3),
reduced coastal upwelling and Ekman pumping (Section 9.2.3.5), changes in eddy activities and planetary waves, and the re-emergence of warm subsurface anomalies (Holbrook et al., 2020; Sen Gupta et al., 2020).
The SROCC reported with high confidence that MHWs - defined as days exceeding the 99th percentile in sea surface temperature (SST) from 1982 to 2016 - have very likely doubled in frequency between 1982 and 2016. Additional observation-based evidence and acquisition of longer observation time series since SROCC have confirmed and expanded on this assessment: since the 1980s MHWs have also become more intense and longer (Frlicher and Laufktter, 2018; Smale et al., 2019; Laufktter et al., 2020). Satellite observations and reanalyses of SST show an increase in intensity of 0.04&deg;C per decade from 1982 to 2016, an increase in spatial extent of 19% per decade from 1982 to 2016, and an increase in annual MHW days of 54% between the 1987-2016 period compared to 1925-1954 (Frlicher et al., 2018; Oliver, 2019). The SROCC assessed that 84-90% of all MHWs that occurred between 2006 and 2015 are very likely caused by anthropogenic warming. There is new evidence since SROCC that the frequency of the most impactful marine heatwaves over the last few decades has increased more than 20-fold because of anthropogenic global warming (Laufktter et al., 2020). In summary, there is high confidence that MHWs have increased in frequency over the 20th century, with an approximate doubling from 1982 to 2016, and medium confidence that they have become more intense and longer since the 1980s.
Consistent with SROCC, future MHWs are defined with reference to the historical climate conditions. The SROCC assessed that MHWs will very likely further increase in frequency, duration, spatial extent and intensity under future global warming in the 21st century. The CMIP6 projections allow us to confirm this assessment and quantify future change based on global mean probability ratio change (Box 9.2, Figure 1): they project MHWs will become four times (5-95% range: 2-9 times] more frequent in 2081-2100 compared to 1995-2014 under SSP12.6, or eight times (3-15 times) more frequent under SSP58.5. The SROCC highlighted that future change of MHWs will not be globally uniform, with the largest changes in the frequency of marine heatwaves being projected to occur in the western tropical Pacific and the Arctic Ocean (medium confidence). New evidence from the latest generation of climate models confirms and complements SROCC assessment (Box 9.2, Figure 1). Moderate increases are projected for mid-latitudes, and only small increases are projected for the Southern Ocean (medium confidence) (Hayashida et al., 2020). While under the SSP58.5 scenario, permanent MHWs (more than 360 days per year) are projected to occur in the 21st century in parts of the tropical ocean, the Arctic Ocean and around 45&deg;S, the occurrence of such permanent MHWs can largely be avoided under the SSP12.6 scenario (Frlicher et al., 2018; Oliver et al., 2019; Plecha and Soares, 2020). The resolution of current climate models (CMIP5 and CMIP6) capture the broad features of MHWs, but they may have a bias towards weaker and longer MHWs in the historical period (medium confidence) (Frlicher et al., 2018; Pilo et al., 2019; Plecha and Soares, 2020) and greater intensification in western boundary current regions (Hayashida et al., 2020).
1227
 
Chapter 9                                                                                                              Ocean, Cryosphere and Sea Level Change Box 9.2 (continued) 9 Box 9.2, Figure 1 l Observed and simulated regional probability ratio of marine heatwaves (MHWs) for the 1985-2014 period and for the end of the 21st century under two different greenhouse gas emissions scenarios. The probability ratio is the proportion by which the number of MHW days per year has increased relative to pre-industrial times. An MHW is defined as a deviation beyond the daily 99th percentile (11-day window) in the deseasonalized sea surface temperature. (a) The MHW probability ratio from satellite observations (NOAA OISST V2.1; Huang et al. 2020) during 1985-2014. The mean warming pattern (difference in ERSST5 (Huang et al. 2017) sea surface temperature between the 1985-2014 and 1854-1900 periods) has been added to the satellite observations to calculate the probability ratio. (b-d) Coupled Model Intercomparison Project Phase 6 (CMIP6) simulated multi-model mean probability ratio of the (b) 1985-2014 period, and 2081-2100 period in the (c) SSP1 2.6 and (d) SSP5 8.5 scenarios. The areas with grey diagonal lines in (d) indicate permanent MHWs (>360 heatwave days per year). These 14 CMIP6 models are included in the analysis: ACCESS-CM2, CESM2, CESM2-WACCM, CMCCCM2-SR5, CNRM-CM61, CNRM-ESM2-1, CanESM5, EC-Earth3, IPSL-CM6A-LR, MIROC6, MRI-ESM2-0, NESM3, NorESM2-LM, NorESM2-MM. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
9.2.2        Changes in Heat and Salinity                                                Ocean warming is not uniform with depth. The AR5 (Rhein et al.,
2013) assessed that, since 1971, ocean warming was virtually 9.2.2.1      Ocean Heat Content and Heat Transport                                        certain for the upper 700 m and likely for the 700-2000 m layer.
Both AR5 and SROCC (Bindoff et al., 2019) assessed that the deep Ocean warming - that is, increasing ocean heat content (OHC) - is                          ocean below 2000 m had likely warmed since 1992, especially an important aspect of energy on Earth: SROCC (Bindoff et al., 2019)                      in the Southern Ocean. Section 2.3.3.1 provides an updated reported that there is high confidence that ocean warming during                          assessment of ocean temperature change for different depth layers, 1971-2010 dominated the increase in the Earths energy inventory,                          time periods and observation-based reconstructions (Table 2.7).
which is confirmed by the Box 7.2 assessment that the ocean has                            Section 2.3.3.1 confirms the previous assessment that it is virtually stored 91% of the total energy gained from 1971 to 2018. As reported                      certain that the upper ocean (0-700 m) has warmed since 1971, in Sections 2.3.3.1, 3.5.1.3 and 7.2.2.2, Box 7.2 and Cross-Chapter                        that ocean warming at intermediate depths (700-2000 m) is very Box 9.1, confidence in the assessment of global OHC change since                          likely since 2006, and that it is likely that ocean warming has 1971 is strengthened compared to previous reports, and extended                            occurred below 2000 m since 1992. Section 3.5.1.3 assessed that it backward to include likely warming since 1871. Table 7.1 updates                          is extremely likely that human influence was the main driver of the the estimates of total ocean heat gains from 1971 to 2018, 1993 to                        ocean heat content increase observed since the 1970s, which extends 2018 and 2006 to 2018. Section 3.5.1.3 assesses that it is extremely                      into the deeper ocean (very high confidence), and shows that biases likely that anthropogenic forcing was the main driver of the OHC                          in potential temperature have a complex pattern (Figure 3.25). In the increase over the historical period. Section 2.3.3.1 reports that                          present section, we assess the regional patterns of this warming and current multi-decadal to centennial rates of OHC gain are greater                          associated processes driving regional ocean warming.
than at any point since the last deglaciation (medium confidence).
1228
 
Ocean, Cryosphere and Sea Level Change                                                                                                                          Chapter 9 The rate of ocean warming varies regionally, with some regions having                    variability in redistributing heat, driving interhemispheric asymmetry experienced slight cooling (Figure 9.6). The SROCC (Bindoff et al.,                      of the recent rate of ocean warming (Rathore et al., 2020; L. Wang 2019) assessed that ocean warming in the 0-700 m depth is                                et al., 2021). Since SROCC, one new study shows that the subpolar globally widespread, with slower than global average warming in                          North Atlantic warming hole observed since the 1980s has emerged the subpolar North Atlantic. The SROCC (Meredith et al., 2019) also                      from internal climate variability and can be attributed to greenhouse estimated that the Southern Ocean accounted for around 75% of                            gas emissions (Chemke et al., 2020). A new analysis of a suite of global ocean heat uptake during 1870-1995 and that 35-43% of the                          climate models (Hobbs et al., 2021) confirms SROCC assessment, upper 2000 m global ocean warming occurred in the Southern Ocean                          based on one paper (Swart et al., 2018), attributing the observed over 1970-2017 (45-62% for 2005-2017). The SROCC noted that this                          Southern Ocean warming to anthropogenic forcing. Given the large interhemispheric asymmetry might (at least partially) be explained by                    fraction of global ocean warming in the Southern Ocean and the high concentrations of aerosols in the Northern Hemisphere. Here,                        sparse observations there before 2005, there is limited evidence that we confirm these assessments, bring new evidence attributing these                        global OHC increase since 1971 might have been underestimated regional trends, and discuss the role of decadal ocean circulation                        (Cheng and Zhu, 2014; Durack et al., 2014). Cross-Chapter Box 9.1 9
(a)    20 Paleo OHC                      3    Global OHC (0-2000m depth); Modern history and model projections to 2100 Observed means and 10  (very) likely ranges 2              Global Mean OHC SSP5-8.5 CMIP6 17-83% ranges                                                                        18 24 0                        10    J                Observations (Ishii)
Hybrid (Zanna)                                                              SSP3-7.0 18 1              Hybrid (Cheng)
                                                              +/- 1 (Obs. & Hybrid)                    Observations                                        SSP2-4.5 CMIP6
          -10                                                                                          & Hybrid                              SSP1-2.6            17    2100 means &
0                                                                                                        17                (very) likely ranges CMIP 30 models            Baseline
          -20                                -1                                                                                  Period 1850                    1900                  1950                      2000                    2050                    2100 ColorHigh Color      model High modelagreement agreement (80%)
(80%)
Low Lowmodel agreement model  agreement(<80%)
(<80%)
Figure 9.6 l Ocean heat content (OHC) and its changes with time. (a) Time series of global OHC anomaly relative to a 2005-2014 climatology in the upper 2000 m of the ocean. Shown are observations (Ishii et al., 2017; Baggenstos et al., 2019; Shackleton et al., 2020), model-observation hybrids (Cheng et al., 2019; Zanna et al., 2019),
and multi-model means from the Coupled Model Intercomparison Project Phase 6 (CMIP6) historical (29 models) and Shared Socio-economic Pathway (SSP) scenarios (label subscripts indicate number of models per SSP). (b-g) Maps of OHC across different time periods, in different layers, and from different datasets/experiments. Maps show the CMIP6 ensemble bias and observed (Ishii et al., 2017) trends of OHC for (b, c) 0-700 m for the period 1971-2014, and (e, f) 0-2000 m for the period 2005-2017. CMIP6 ensemble mean maps show projected rate of change 2015-2100 for (d) SSP58.5 and (g) SSP12.6 scenarios. Also shown are the projected change in 0-700 m OHC for (d) SSP12.6 and (g) SSP58.5 in the CMIP6 ensembles, for the period 2091-2100 versus 2005-2014. No overlay indicates regions with high model agreement, where 80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where <80% of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1229
 
Chapter 9                                                                                                            Ocean, Cryosphere and Sea Level Change accounts for an increased error before 2005 in global OHC change.                      observational record may underestimate the rate of deep-ocean In summary, in the upper 2000 m since the 1970s, the subpolar                          warming from 1990 to 2010 by about 20% (Garry et al., 2019)
North Atlantic has been slowly warming, and the Southern Ocean                          which is included in the assessed OHC error (Cross-Chapter Box 9.1).
has stored a disproportionally large amount of anthropogenic heat                      There is still low agreement in deep-ocean changes from ocean data (medium confidence).                                                                    assimilation reanalyses (Palmer et al., 2017) and low confidence in such inferences. In summary, while observational coverage Below 2000 m, direct observations of full-depth ocean temperature                      below 2000 m is sparser than in the upper 2000 m, there is high change are limited to ship-based, high-quality deep-ocean                              confidence that deep-ocean warming below 2000 m has been larger temperature measurements. Such high-quality full-depth ship-based                      in the Southern Ocean than in other ocean basins due to widespread sampling has improved from 1990 to the present due to the World                        AABW warming.
Ocean Circulation Experiment (WOCE) and the Global Ocean Ship-based Hydrographic Investigations Program (GO-SHIP; Sloyan                        Different processes drive OHC patterns over a range of time scales.
et al., 2019). The SROCC (Bindoff et al., 2019) assessed that the likely                Recent literature has highlighted the role of ocean circulation warming of the ocean since the 1990s below 2000 m is associated                        variability in driving OHC patterns by decomposing the global pattern 9 with a marked regional pattern, with larger warming in the Southern                    of OHC change into a combination of added heat due to climate Ocean. In the deep North Atlantic, warming has reversed to cooling                      change taken up under fixed ocean circulation (added heat), and over the past decade, possibly due to internal variability fed by North                redistribution of heat associated with changing ocean currents Atlantic Deep Water (Section 9.2.2.3). Over the past decade, the                        (redistributed heat; Gregory et al., 2016; Bronselaer and Zanna, 2020; warming rate of Antarctic Bottom Water (AABW; Section 9.2.2.3) has                      Couldrey et al., 2021). Redistributed heat alters regional patterns of been dependent on origin: slower from the Weddell Sea and faster                        heat storage and carbon storage (Cross-Chapter Box 5.3; Bronselaer from the Ross Sea and Ad&#xe9;lie Land. One new study (Purkey et al.,                        and Zanna, 2020; Todd et al., 2020; Couldrey et al., 2021) but does 2019) strengthens confidence in AABW warming: below 4000 m                              not affect the global OHC. There is medium confidence that decadal a monotonic, basinwide, and multi-decadal temperature change                          variability of the ocean circulation strengthened the rate of ocean is found in the southern Pacific basin, with larger warming rates                      warming in the Southern Hemisphere compared to the Northern near the bottom water formation sites than further downstream.                          Hemisphere in the decade from 2005 (Rathore et al., 2020; L. Wang New analysis of one model provides limited evidence that the sparse                    et al., 2021; Zika et al., 2021). More generally, since 2005, the OHC Figure 9.7 l Meridional-depth profiles of zonal-mean potential temperature in the ocean and its rate of change in the upper 2000 m of the Global, Pacific, Atlantic and Indian oceans. Shown are (a, e, i, m) observed temperature (Argo climatology 2005-2014), (b, f, j, n) bias of the Coupled Model Intercomparison Project Phase 6 (CMIP6) ensemble over this period, and future changes under (c, g, k, o) SSP12.6 and (d, h, l, p) SSP58.5. No overlay indicates regions with high model agreement, where 80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where <80% of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1230
 
Ocean, Cryosphere and Sea Level Change                                                                                                                    Chapter 9 9
Figure 9.8 l Decomposition of simulated ocean heat content and northward ocean heat transport. (a, c, e) Total ocean heat content (0-2000 m) warming rate as observed and simulated by Coupled Model Intercomparison Project Phase 5 (CMIP5) models over the historical period (1972-2011) and under the RCP8.5 future (2021-2060) versus the associated decomposed (b, d, f) added heat contribution (neglecting changes in ocean circulation) to the total (Bronselaer and Zanna, 2020). (g) Relationship between northward heat transport and Atlantic Meridional Overturning Circulation (AMOC) in HighResMIP models (1950-2050) and observations during the RAPID period (2004-2018). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
pattern observed is predominantly due to heat redistribution with                      (Muilwijk et al., 2018; Q. Wang et al., 2019; Tsubouchi et al., 2021),
regions of both warming and cooling (Figure 9.6; Zika et al., 2021);                    similar to SROCC assessment, and consistent with observed increase however, extending analysis back to 1972 shows the importance of                        in OHC in the ice-free Arctic ocean (Mayer et al., 2019). It is estimated added heat setting a large-scale warming pattern with mid-latitude                      that an increase of 0.021 PW of OHT occurred after 2001 into the maxima consistent with subduction of water masses, particularly in                      Arctic, which is sufficient to account for the recent OHC change in the Southern Hemisphere Mode Waters (Section 9.2.2.3, and Figures 9.6                      northern seas (Tsubouchi et al., 2021). However, these trends cannot and 9.8; Bronselaer and Zanna, 2020). The longer the analysis                          yet be attributed to anthropogenic forcing due to potential internal window, the more added heat dominates over redistributed heat.                          variability (Muilwijk et al., 2018; Wang et al., 2019). New evidence This translates into more ocean area with statistically significant                    strengthens the case that El Nino-Southern Oscillation (ENSO) and the warming trends and less area with statistically significant cooling                    Northern Annular Mode affect interannual OHT variability (Trenberth trends (Johnson and Lyman, 2020). The region where added heat                          et al., 2019) and shows that a slowing AMOC reduces northward is most compensated for by redistributed cooling is in the northern                    OHT in the Atlantic at 26.5&deg;N (Section 9.2.3.1 and Figure 9.8; Bryden North Atlantic basin, where changes in the subpolar gyre circulation                    et al., 2020). Despite a decrease of AMOC northward heat (0.17 PW) and Atlantic Meridional Overturning Circulation (AMOC) result in                        and mass (2.5 Sverdrup (Sv); 1 Sv = 109 kg s-1) transport, OHT has cooling (Section 9.2.3.1; Williams et al., 2015; Piecuch et al., 2017;                  increased toward the Arctic through increased upper northern North Zanna et al., 2019; Bronselaer and Zanna, 2020). In summary, and                        Atlantic temperatures and stronger wind-driven gyres (medium strengthening SROCC assessment, ocean warming is not globally                          confidence) (Section 9.2.3.4 and Figure 9.11; Singh et al., 2017; uniform due to patterns of uptake predominantly along known water                      Oldenburg et al., 2018). In summary, OHT has increased toward the mass pathways, and due to changing ocean circulation redistributing                    Arctic in recent decades, which at least partially explains the recent heat within the ocean (high confidence).                                                OHC change in the Arctic (medium confidence).
While heat redistribution reflects changes in ocean circulation and is                  Major volcanic eruptions have caused interannual to decadal cooling a useful concept to understand the underlying processes driving OHC                    phases within the marked long-term increase in global OHC -
patterns, change in ocean heat transport (OHT) arises due to changes                    Mount Agung in 1963, El Chich&#xf3;n in 1982 and Mount Pinatubo in ocean circulation and ocean temperature and affects regional                        in 1991 (Cross-Chapter Box 4.1; Church et al., 2005; Fasullo et al.,
OHC change. The AR5 did not assess change in OHT and SROCC                              2016; Stevenson et al., 2016; Fasullo and Nerem, 2018). In the (Meredith et al., 2019) only assessed projected OHT increases into                      first few years following an eruption, heat exchange with the the Nordic Seas and the Arctic Ocean. New evidence of increasing                        subsurface ocean allows atmospheric cooling to be sequestered northward OHT into the Arctic has been observed in recent decades                      into the seasonal thermocline, therefore reducing the magnitude of 1231
 
Chapter 9                                                                                        Ocean, Cryosphere and Sea Level Change the peak atmospheric temperature anomaly (Gupta and Marshall,          0-2000 m layer will increase from 2017 to 2100 by 0.900 +/- 0.345 YJ 2018). However, while explosive volcanic eruptions only disturb        (1 YJ = 1024 Joules) under RCP2.6 and 2.150 +/- 0.540 YJ under the Earths radiative budget and surface fluxes for a few years,      RCP8.5. Updating SROCC estimates with CMIP6 projections gives the ocean preserves an anomaly in OHC in the upper 500 m (also        heat content increases and 17-83% ranges in the 0-2000 m layer affecting thermosteric sea level) many years after the eruption        between 1995-2014 and 2081-2100 of 1.06 (0.80-1.31) YJ, (Gupta and Marshall, 2018; Bilbao et al., 2019). The anomaly affects  1.35 (1.08-1.67) YJ, 1.62 (1.37-1.91) YJ, 1.89 (1.60-2.29) YJ under the atmosphere through air-sea heat fluxes with surface conditions    scenarios SSP12.6, SSP24.5, SSP37.0, and SSP58.5, respectively returning to normal only after several decades (Gupta and Marshall,    (Figure 9.6 and Table 9.1). The two-layer model used here to calculate 2018; Bilbao et al., 2019), or on centennial time scales in the case  thermosteric sea level rise (9.SM.4) and tuned for AR6-assessed of repeated eruptions (G.H. Miller et al., 2012; Atwood et al., 2016;  equilibrium climate sensitivity (ECS; Section 7.SM.2), provides Gupta and Marshall, 2018). In summary, there is medium confidence      consistent 17-83% ranges of 1.18 (0.99-1.42) YJ, 1.56 (1.33-1.86) YJ, that oceanic mechanisms buffer the atmospheric response to            1.90 (1.63-2.21) YJ, 2.23 (1.92-2.64) YJ under scenarios SSP12.6, volcanic eruptions on annual time scales by storing volcanic cooling  SSP24.5, SSP37.0, and SSP58.5, respectively (Table 9.1). Based in the subsurface ocean, affecting OHC and thermosteric sea level on  on CMIP6 models and the two-layer model, it is likely that, between 9 decadal to centennial time scales.                                    1995-2014 and 2081-2100, OHC will increase two to four times the amount of the 1971-2018 OHC increase under SSP12.6, and four CMIP5 and CMIP6 models simulate OHC changes that are consistent        to eight times that amount under SSP58.5. The CMIP6 models show with the updated observational and improved estimates of OHC          that OHC dependence on scenarios begins only after about 2040 over the period 1960 to 2018 (Figures 9.6, 9.7 and 9.8), and they      (Figure 9.6).
replicate the vertical partitioning of OHC change for the industrial era, although with a tendency to underestimate OHC gain                The OHC patterns projected by CMIP6 models (Figures 9.6 and 9.7) shallower than 2000 m and overestimate it deeper than 2000 m          are similar to the CMIP5 projections assessed in SROCC (Bindoff et al.,
(Section 3.5.1.3). The AR5 (Flato et al., 2013) assessed that climate  2019): faster warming in all water mass subduction regions models transport heat downward more than the real ocean. Since        (e.g., subtropical cells and mode waters); deeper penetration in the AR5, studies have shown that increasing the horizontal resolution of  centre of subtropical gyres; slower northern North Atlantic warming ocean models tends to increase agreement of vertical heat transport    due to slowing AMOC; and slower subpolar Southern Ocean with observations as the dependency on ad-hoc choices of eddy          warming due upwelled pre-industrial water masses. Decreased parametrizations is relaxed (Griffies et al., 2015; Chassignet et al., aerosol forcing will allow Northern Hemisphere ocean warming 2020). The magnitude of the AMOC and Indonesian Throughflow            to be faster and less dominated by Southern Hemisphere change affect future OHC change - for example, through overestimated          (Shi et al., 2018; Irving et al., 2019). Since SROCC, distinguishing modelled downward heat pumping (Kostov et al., 2014) - and there      between added and redistributed heat has aided in understanding are indications of greater model consistency in these transports at    projections (Bronselaer and Zanna, 2020; Dias et al., 2020; Couldrey higher resolution (Figure 9.10; Chassignet et al., 2020; L.C. Jackson  et al., 2021). The near-term decades will feature patterns strongly et al., 2020). Climate models tend to reproduce the observed added    influenced by heat redistribution and internal variability (Rathore heat, but redistributed heat is less well represented (Figure 9.8;    et al., 2020). Strengthening Southern Hemisphere westerlies are Bronselaer and Zanna, 2020; Dias et al., 2020; Couldrey et al., 2021). projected, except for stringent mitigation scenarios (Bracegirdle et al.,
Since redistributed heat dominates historical OHC change, historical  2020), and will cause a northward and downward OHT. There is low simulations poorly reproduce regional patterns, but as future OHC      agreement in future Southern Ocean warming across model results change will become dominated by added heat, more skill in future      due to uncertainties in the magnitude of westerly wind changes modelled OHC patterns is expected (Bronselaer and Zanna, 2020).        (Figure 9.4; Liu et al., 2018; He et al., 2019; Dias et al., 2020; Lyu et al.,
In summary, climate models have more skill in representing OHC        2020b) and the degree of eddy compensation of overturning across change from added heat than from ocean circulation change (high        different parametrizations and resolutions (Section 9.2.3.2; Beal and confidence). Since added heat dominates over redistributed heat        Elipot, 2016; Mak et al., 2017; Roberts et al., 2020). By 2100, however, on a centennial scale (especially under high-emissions scenarios)      the OHC change will be dominated by the added heat response, confidence in future modelled OHC patterns at the end of the          particularly for strong warming scenarios (Garuba and Klinger, 2018; 21st century is greater than at decadal scale.                        Bronselaer and Zanna, 2020) with added heat following unperturbed water mass pathways in the North Atlantic and Southern Ocean The SROCC (Bindoff et al., 2019) assessed that the ocean will continue (Figure 9.8; Dias et al., 2020; Couldrey et al., 2021). There is high to take up heat in the coming decades for all plausible scenarios,    confidence that projected weakening of the AMOC (Section 9.2.3.1) and here this assessment is confirmed with very high confidence.      will cause a decrease in northward OHT in the Northern Hemisphere The SROCC reported that, compared with the observed changes since      mid-latitudes (Figure 9.8 and Sections 9.2.3.1 and 4.3.2.3; Weijer et al.,
the 1970s, the warming of the ocean by 2100 would very likely double  2020) associated with a dipole pattern of Atlantic OHC redistributed to quadruple for low-emissions scenarios (RCP2.6) and increase five    from northern to low latitudes that may override added heating in to seven times for high-emissions scenarios (RCP8.5). The SROCC also  the northern North Atlantic (Figures 9.6, 9.7 and 9.8). Variations concluded with high confidence that the overall warming of the ocean  in the degree of AMOC redistributed heat (Menary and Wood, would continue this century, even after radiative forcing and mean    2018) causes large intermodel spread in SST (Figure 9.3) and OHC surface temperatures stabilize. The SROCC projected that OHC in the    change (Figure 9.6; Kostov et al., 2014; Bronselaer and Zanna, 2020; 1232
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                                  Chapter 9 Todd et al., 2020; Couldrey et al., 2021). In the 700-2000 m depth                                            from ice core rare gas elemental and isotopic ratios document a rise range, CMIP5 and CMIP6 models project the largest warming to be                                              in global OHC relative to the Last Glacial Maximum of >17,000 ZJ in the North Atlantic Deep Water and Antarctic Intermediate Water                                            (change in mean ocean temperature >3.1&deg;C; 1 ZJ = 1021 Joules)
(Figure 9.7) while below 2000 m, the North Atlantic cools in many                                            (Figure 9.9; Bereiter et al., 2018; Baggenstos et al., 2019; Shackleton models, and Antarctic Bottom Waters warm (Sall&#xe9;e et al., 2013b;                                              et al., 2019, 2020). This temperature increase is significantly larger Heuz&#xe9; et al., 2015). In summary, on decadal time scales, redistribution                                      than the modelled OHC changes associated with collapse of AMOC will dominate regional patterns of OHC change without affecting the                                          alone, and tracks rising Southern Ocean SST (Uemura et al., 2018),
globally integrated OHC; however, by 2100, particularly under strong                                          strengthening of the deep abyssal overturning cell (Du et al., 2020) warming scenarios, there is high confidence that regional patterns                                            and increased North Atlantic water in the Southern Ocean (Wilson of OHC change will be dominated by added heat entering the sea                                                et al., 2020). This underscores the importance of Antarctic abyssal surface, primarily in water mass formation regions in the subtropics;                                        ventilation on long-term oceanic heat budgets (Section 9.2.3.2).
and reduced aerosols will increase the relative rate of Northern                                              An ensemble of four intermediate-complexity models project Hemisphere heat uptake (medium confidence).                                                                  10,000-year future responses to CO2 emissions (Clark et al., 2016) with SST change peaking around 2300 and a varying scenario-dependent The SROCC assessed that the warming of the deep ocean is slow                                                magnitude approaching the scale of glacial-to-interglacial changes                      9 to manifest, with multi-century or longer response times, so global                                          in paleodata (Figure 9.9). Long-term OHC commitments relative to OHC (and global mean thermosteric sea level) will continue to                                                1850-1900 conditions are 2.6, 9.7, 15.2, 21.6, and 28.0 YJ (with rise for centuries (Figures 9.9 and 9.30). New studies show that                                              mean ocean temperature change as much as 5.1&deg;C) for emissions this continuation persists, even after cessation of greenhouse gas                                            of 0, 1280, 2560, and 3840 and 5120 Gt after 2000 CE respectively, emissions (Ehlert and Zickfeld, 2018). Ocean warming will continue,                                          with OHC peaking near 4000 CE, reflecting whole-ocean warming even after emissions reach zero because of slow ocean circulation                                            lagging SST by thousands of years. The exact timing is uncertain, (Larson et al., 2020). OHC will increase until at least 2300, even for                                        subject to rates of high-latitude meltwater input (Van Breedam et al.,
low-emissions scenarios, but with a scenario-dependent rate (Nauels                                          2020) and circulation time (Gebbie and Huybers, 2019). In summary, et al., 2017; Palmer et al., 2018) and depends on cumulative CO2                                              there is very high confidence that there is a long-term commitment emissions, as well as the time profile of emissions (Bouttes et al.,                                          to increased OHC in response to anthropogenic CO2 emissions, which 2013). Past long-term changes in total OHC illustrate adjustment                                              is essentially irreversible on human time scales.
relevant to expected future changes (Figure 9.9). Observational data Long-term trends of ocean heat content and surface temperature Observed and modelled historical data, and model projections under different emissions scenarios 3
Last Interglacial Observations          Deglacial/Holocene Observations OHC: Shackleton                          OHC: Baggenstos, Shackleton SST (Southern Ocean): Uemera, Proxies    SST (Southern Ocean): Uemera, Proxies OHC anomaly (1025 J = 10 4 ZJ) 2 Ocean Heat Content (OHC)
Surface Temperature 1
0 Model projections under different emissions OHC and atmospheric surface temperature:
1280 Gt, 2560 Gt, 3840 Gt, 5120 Gt
                                                -1
                                                -2 0.06 Modern Historical Data                                                                                                                6 0.03 OHC: Levitus (observed), HadCM3 (modelled)
SST (Southern Ocean): Uemera 3
1025 J                                0                                                                                                                                  0    C o
                                          -0.03                                                                                                                                            -3
                                          -0.06                                                                                                                                            -6 1500 CE                                  1600          1700                      1800                      1900                        2000 Figure 9.9 l Long-term trends of ocean heat content (OHC) and surface temperature. (a, b) Ice-core rare gas estimates of past mean OHC (ZJ), scaled to global mean ocean temperature (&deg;C), and to steric global mean sea level (GMSL) (m) per CCB-2 (red dashed line), compared to surface temperatures (black solid line, gold solid line; &deg;C rightmost axis). Southern Ocean sea surface temperature (SST) from multiple proxies in 11 sediment cores and from ice core deuterium excess (Uemura et al.,
2018). (a) Penultimate glacial interval to last interglacial, 150,000-100,000 yr B2K (before 2000) (Shackleton et al., 2020). (b) Last glacial interval to modern interglacial, 40,000-0 yr B2K (Baggenstos et al., 2019; Shackleton et al., 2019). Changes in OHC (dashed lines) track changes in Southern Ocean SST (solid lines). (c) Long-term projected (2000 to 12000 CE) changes of OHC (dashed lines) in response to four greenhouse gas emissions scenarios (Clark et al., 2016) scale similarly to large-scale paleo changes but lag projected global mean SST (solid lines). (d) model simulated 1500-1999 OHC (Gregory et al., 2006) and 1955-2019 observations (Levitus et al., 2012) updated by NOAA NODC. All data expressed as anomalies relative to pre-industrial time. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1233
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change 9.2.2.2      Ocean Salinity                                                Ocean circulation changes also affect salinity, largely on annual to decadal time scales (Du et al., 2019; Liu et al., 2019; Holliday The AR5 (Rhein et al., 2013) assessed that it was very likely that        et al., 2020). For instance, in the subpolar North Atlantic, increasing subsurface salinity changes reflect surface salinity change, and          northward transport of Atlantic waters entering the subpolar gyre that basin-scale regions of high salinity and evaporation had trended      from the South have compensated the salinity decrease expected more saline, while regions of low salinity and more precipitation had      from increased Greenland meltwater flux since the early 1990s trended fresher since the 1950s. The SROCC (Bindoff et al., 2019)          (Dukhovskoy et al., 2016, 2019; Stendardo et al., 2020). After assessment was consistent with AR5. Section 2.3.3.2 strengthens            the mid-2010s the trend reversed towards a broad freshening, evidence that subsurface salinity trends are connected to surface trends  the largest in 120 years, in the North Atlantic (Holliday et al.,
(very likely), which are, in turn, linked to an intensifying hydrological  2020). The long-term freshening in the Pacific Ocean has also been cycle (medium confidence). Increasing evidence from updated                subject to decadal variability, such as a marked salinification since observational records indicates that it is now virtually certain that      2005 associated with increased surface fluxes (G. Li et al., 2019).
surface salinity contrasts are increasing. At basin scale, Section 2.3.3.2 Local salinity anomalies forced by water cycle intensification can and AR5 concur that it is very likely that the Pacific and Southern Ocean  be weakened by rapid exchange between basins with opposing 9 have freshened, and the Atlantic has become more saline. Figures 3.25      trends, such as by water mass exchange in shallow wind-driven cells and 3.27 compare CMIP6 models to salinity observations.                    between the tropics and the subtropics (Levang and Schmitt, 2020).
Similarly, eddy exchanges between neighbouring gyres can partly Globally the mean salinity contrast at near-surface between high- and      counterbalance decadal time scale long-term subpolar freshening low-salinity regions increased 0.14 [0.07 to 0.20] from 1950 to            and affect deep convection (Levang and Schmitt, 2020). There is high 2019 (Section 2.3.3.2). At regional scale, SROCC (Meredith et al.,        confidence that, at annual to decadal time scales, regional salinity 2019) assessed an Arctic liquid freshwater trend of 600 +/- 300 km3 yr -1    changes are driven by ocean circulation change superimposed on (600 +/- 200 Gt yr -1) between 1992 and 2012, reflecting changes            longer-term trends.
associated with continental freshwater imports that affect ocean mass (land ice, rivers) as well as changes in sea ice volume. Since AR5,  The CMIP5 historical simulations have patterns similar to, but regional observation-based analyses not assessed in SROCC further          with greater spatial variability than, observed estimates and confirm the long-term, large-scale and regional patterns of salinity      correspondingly smaller amplitudes in the multi-model mean (Durack, change, both at the ocean surface and in the subsurface ocean,            2015; Cheng et al., 2020; Silvy et al., 2020). Section 3.5.2.1 reports, including almost 120 years of changes in the North Atlantic (Friedman      however, that the fidelity of ocean salinity simulation has improved et al., 2017) and 60 years of monitoring in the subpolar North Pacific    in CMIP6, and near-surface and subsurface biases have been reduced (Cummins and Ross, 2020). These longer time series also provide            (medium confidence), though the structure of the biases strongly context to detect large multi-annual change from 2012 to 2016 in          reflects those of CMIP5. At regional scale, salinity biases are at least the subpolar North Atlantic, unprecedented over the centennial            partially a result of inaccurate ocean dynamics (Levang and Schmitt, record (Holliday et al., 2020). In summary, there is high confidence      2020). Despite the regional limitations, Section 3.5.2.2 assesses that, that salinity trends have extended for more than 60 to 100 years in        at the global scale, it is extremely likely that human influence has the regions with long historical observation records, such as the North    contributed to observed surface and subsurface salinity changes Pacific and the North Atlantic basin.                                      since the mid-20th century (strengthened from the very likely AR5 assessment).
While there is low confidence in direct estimates of trends in surface freshwater fluxes (Sections 2.3.1.3.5, 8.3.1.1 and 9.2.1.2),      The SROCC (Bindoff et al., 2019) assessed that projected salinity as discussed in SROCC (Meredith et al., 2019), observational studies      changes in the subsurface ocean reflect changes in the rates of coupled with modelling studies suggest that surface flux changes          formation of water masses or their newly formed properties.
drive many observed near-surface salinity changes, on top of              Additional consistent newer evidence based on CMIP5 and regional changes specific to polar regions. Advances in salinity observations -    climate models confirms that 21st century projections adhere for example, the Argo program (Riser et al., 2016); Soil Moisture and      to the fresh gets fresher, salty gets saltier paradigm, through Ocean Salinity (SMOS), Aquarius and Soil Moisture Active Passive          subduction of freshening high-latitude waters into the ventilated (SMAP; Supply et al., 2018; Vinogradova et al., 2019) - combined          water masses in both hemispheres in the Pacific, Indian and with process studies (SPURS-1/2; Lindstrom et al., 2015; SPURS-2          Southern Ocean - especially the Arctic and upper Southern Ocean, Planning Group 2015) and methodological and numerical advances,            and saltier subtropical and Mediterranean surface waters - lead to have increased understanding of how subsurface salinity anomalies          saltier pycnoclines and North Atlantic mode water (Metzner et al.,
link to surface fluxes, and thus increase confidence that near-surface    2020; Parras-Berrocal et al., 2020; Silvy et al., 2020; Soto-Navarro and subsurface salinity pattern changes since the 1950s are linked        et al., 2020). Overall, projections confirm SROCC assessment to changing surface freshwater fluxes (Zika et al., 2018; Cheng et al.,    that fresh ocean regions will continue to get fresher and salty 2020) with an additional contribution from changes in sea ice and          ocean regions will continue to get saltier in the 21st century land ice discharge at high latitudes (Haumann et al., 2016; Purich        (medium confidence).
et al., 2018; Dukhovskoy et al., 2019; Rye et al., 2020). There is therefore medium confidence in the processes linking surface fluxes to surface and subsurface salinity change.
1234
 
Ocean, Cryosphere and Sea Level Change                                                                                                    Chapter 9 9.2.2.3      Water Masses                                                  Climate models from CMIP3 to CMIP5 generally simulated shallower and lighter SAMW and AAIW than is observed (Flato et al., 2013).
Water masses refer to connected bodies of ocean water, formed at the        New analysis of CMIP5 models suggests that the freshening of ocean surface with identifiable properties (temperature, salinity, density, these water masses is one of the most prominent projected salinity chemical tracers) resulting from the unique formation conditions of the    changes in the world ocean, and that this freshening emerged from overlying atmosphere and/or ice, before being transferred (subducted)      internal variability as early as the 1980s to 1990s (Silvy et al., 2020).
to the deeper ocean below the surface turbulent layer. As water masses subduct, they ventilate the subsurface ocean, transferring          Trends in North Atlantic Deep Water (NADW) are obscured by decadal characteristics acquired at the ocean surface to the subsurface.            variability (Rhein et al., 2013; Bindoff et al., 2019). The AR5 (Rhein By integrating surface flux changes, water masses provide higher            et al., 2013) assessed that it is very likely that the temperature, salinity, signal-to-noise ratios for detecting and monitoring climate change          and formation rate of the Upper NADW (formed by deep convection than surface fluxes (Bindoff and McDougall, 2000; Durack and Wijffels,      in the Labrador and Irminger Seas) is dominated by strong decadal 2010; Silvy et al., 2020).                                                  variability related to the North Atlantic Oscillation (NAO) and it is likely that Lower NADW (formed in the Nordic Seas and supplied to Subtropical mode waters (STMW) ventilate the main thermocline of            the North Atlantic by deep overflows over the sills between Scotland          9 the ocean at mid- to low-latitudes and have circulation time scales        and Greenland) cooled from 1955 to 2005. New insights from away from the surface of the order of years to decades. The SROCC          observations have emphasized the stability of the deep overflows (Bindoff et al., 2019) reported that warming in the subtropical gyres      associated with Lower NADW (Hansen et al., 2016; Jochumsen et al.,
penetrates deeper than in other gyres, following the density surfaces      2017; &#xd8;sterhus et al., 2019) and even slight warming in the Faroe in these gyres. Consistently, we assess that STMW have deepened            Bank Channel (Hansen et al., 2016). As a result, the AR5 assessment worldwide, with greatest deepening in the Southern Hemisphere              that Lower NADW likely cooled between 1955 and 2005 is revised (high confidence) (Hkkinen et al., 2016; Desbruyres et al., 2017).        to: it is likely that any observed changes in temperature, salinity, Subsurface warming in the Northern Hemisphere STMW is larger                and formation rate of the Lower NADW are dominated by decadal than at the surface (Sugimoto et al., 2017) because they are formed in      variability. For CMIP5 models, it was shown that AMOC variability is winter western boundary current extensions, where surface warming          linked to variability in NADW formation (Heuz&#xe9;, 2017) and projected is larger than the global average (Section 9.2.1.1). Variability            AMOC decline to decreased NADW formation (both Lower NADW in STMW thickness or temperature has a large imprint on OHC                and Upper NADW; Heuz&#xe9; et al., 2015). For CMIP6 models, projected (Section 9.2.2.1; Kolodziejczyk et al., 2019). STMW are observed to be      AMOC decline is also associated with a decline in NADW formation freshening in the North Pacific and associated with increased salinity      (Reintges et al., 2017; Weijer et al., 2020). The link between AMOC in the North Atlantic (Oka et al., 2017; Silvy et al., 2020), with large    and NADW formation appears insensitive to the large range in decadal variability (Oka et al., 2019; Wu et al., 2020). Anthropogenic      model bias in NADW water mass characteristics (Heuz&#xe9;, 2017). Many temperature and salinity changes in the STMW layer are projected            models may overestimate deep water formation in the Labrador Sea, to intensify in the future, with emergence from natural variability        but at least one new model is consistent with recent Overturning in around 2020 to 2040 (Silvy et al., 2020).                                  the Subpolar North Atlantic Program (OSNAP) observations showing very weak overturning in the western subpolar gyre, where Labrador Subantarctic mode water (SAMW) and Antarctic intermediate water            Sea water is formed (Menary et al., 2020a). The CMIP6 models show (AAIW) form at the Southern Ocean surface directly north of the            a reduced bias in NADW properties compared to CMIP5 models, but Antarctic Circumpolar Current and ventilate the upper 1000 m of            still feature varying locations of deep convection in the subpolar the Southern Hemisphere subtropics. The SROCC (Meredith et al.,            gyre: some convect only in the Labrador Sea (6/35 models), most in 2019) reported a freshening of these water masses between 1950              both the Labrador and Irminger Seas (26/35 models; as is observed),
and 2018, and they are projected to have the largest subsurface            and some only in the Irminger Sea (3/35 models), but in general, the temperature increase of the Southern Hemisphere oceans, along with          area where deep convection takes place has expanded relative to a continued freshening, in the 21st century. The SROCC connected            CMIP5, which appears unrealistic (Heuz&#xe9;, 2021). Models with most SAMW and AAIW to Southern Ocean temperature changes as the                  deep convection in the subpolar gyre feature the smallest bias in large Southern Ocean surface heat uptake is circulated and mixed            NADW characteristics, partly associated with NADW formed in the along with these water masses (high confidence). Close to its              Nordic Seas (as observed) being largely unable to leave the area formation region, SAMW is predominantly affected by air-sea flux            (Heuz&#xe9;, 2021) due to inaccurate overflows (Danabasoglu et al., 2010; changes, while further northward it is influenced by wind-forced            Deshayes et al., 2014; Wang et al., 2015). Despite the wide range in changes (Meredith et al., 2019). New evidence shows that a change          model bias, it remains very likely that any long-term (multi-decadal in SAMW heat content over the last decade is primarily attributable        or longer) decrease in AMOC is accompanied by a decline in NADW to its thickening (Kolodziejczyk et al., 2019). Over the past decade, the  formation, associated with lighter densities in the northern North SAMW and AAIW volumes have changed by thickening of the lighter            Atlantic and Arctic basins.
and thinning of the denser parts of SAMW and AAIW, leading to lightening of these ventilated ocean layers overall (Hong et al., 2020;    The SROCC (Meredith et al., 2019) assessed that the global volume Portela et al., 2020). Over the last decade, there is limited evidence      of Antarctic Bottom Water (AABW) had decreased and warmed of increased subduction of SAMW due to deepening mixed layers              since the 1980s, most noticeably near Antarctica. The SROCC also in the SAMW formation region (Section 9.2.1.3; Qu et al., 2020).            noted freshening in the Indian and Pacific sectors of the Southern 1235
 
Chapter 9                                                                                        Ocean, Cryosphere and Sea Level Change Ocean and a higher rate of freshening in the Indian Sector from the    facilitate access of CDW to the sub-ice-shelf cavities (Section 9.4.2.1).
2000s to 2010s than from the 1990s to 2000s (low confidence).          However, there is low confidence in the quantitification, importance Since SROCC, freshening of Indian Ocean AABW from 1974 to 2016          and the ability of present models, especially at coarse resolution, to has been revealed (Aoki et al., 2020). Additionally, interannual to    project changes in each of these processes (Section 9.4.2.2). Some decadal variability in AABW has been quantified to be larger than      studies have projected a possible shift from cold to warm sub-ice-shelf previously thought in terms of temperature, salinity and thickness,    cavities causing a sudden flush of warm water underneath ice and in volume transport (Abrahamsen et al., 2019; Purkey et al.,        shelves, but there is low confidence in the driving processes and the 2019; Gordon et al., 2020; Silvano et al., 2020). Multi-decadal to      threshold to trigger the shift (Box 9.4; Hellmer et al., 2012, 2017; centennial modes of variability could have driven the observed trends  Silvano et al., 2018; Hazel and Stewart, 2020).
of the lower cell over the past decades via the opening of a Weddell Sea Polynya (L. Zhang et al., 2019), although other studies find it contributed minimally to the observed abyssal warming (Zanowski        9.2.3        Regional Ocean Circulation et al., 2015; Zanowski and Hallberg, 2017). Therefore, there is limited evidence and low agreement in the role of open ocean polynyas in        9.2.3.1      Atlantic Meridional Overturning Circulation 9 driving past decadal observed trends of AABW. Beyond variability, all observational, theoretical, and numerical evidence supports SROCC      Atlantic Meridional Overturning Circulation (AMOC) is the main assessment that formation and export of AABW will continue to          overturning current system in the South and North Atlantic decrease due to warming and freshening of surface source waters        oceans. It transports warm upper-ocean water northwards, near the Antarctic continent. Consistent with Section 9.2.3.2,          and cold, deep water southwards, as part of the global ocean confidence in this assessment is increased to medium confidence        circulation system (Section 2.3.3.4.1). Changes in AMOC influence compared to SROCC.                                                      global ocean heat content (OHC) and transport (Section 9.2.2.1);
global ocean anthropogenic carbon uptake changes and climate Circumpolar Deep Water (CDW) lies in the Southern Ocean and forms      sensitivity (Cross-Chapter Box 5.3); and dynamical sea level by the mixing of NADW and AABW (Talley, 2013). The SROCC (Meredith      change (Section 9.2.4). Since AR5/SROCC, confidence in modelled et al., 2019) assessed with low confidence that mean southward          and reconstructed AMOC has decreased due to new observations and upward CDW transport is linked to decadal wind variability          and model disagreement. Confidence levels have been revisited in (Section 9.2.3.2), and that CDW has warmed south of the Antarctic      modelled AMOC evolution during the 20th century, the magnitude of Circumpolar Current (ACC) in the past decades. New evidence            21st-century AMOC decline, and the possibility of an abrupt collapse reinforces SROCC assessment: changes in Southern Ocean wind stress      before 2100.
have been confirmed to drive variability and increase the large-scale southward CDW transport (Waugh et al., 2019). In addition, growing      The AR5 (Flato et al., 2013) found that the mean AMOC strength evidence suggests that the upper-ocean stratification increase in      in CMIP5 models ranges from 15 to 30 Sv for the historical period.
the subpolar Southern Ocean since the 1970s (Section 9.2.1.3) has      The multi-model mean overturning at 26&deg;N in CMIP5 and CMIP6 is reduced the volume of CDW that is mixed to the surface, causing        comparable to the RAPID array measurements (Reintges et al., 2017),
subsurface CDW warming (Bronselaer et al., 2020; Haumann et al.,        but the inter-model spread in CMIP6 is as large (10-31 Sv) as in 2020; Jeong et al., 2020; Moorman et al., 2020). Large regions of the  CMIP5 (Section 3.5.4; Weijer et al., 2020). Biases in simulations of the Antarctic shelves are currently isolated from warm CDW (Thompson        present-day AMOC and associated deep convection in the subpolar et al., 2018; Jourdain et al., 2020). The SROCC (Meredith et al., 2019) gyre and Nordic Seas were large in CMIP5 models, with many models assessed that subsurface warming extends close to Antarctica and has    exhibiting ocean convection that is too deep, over too large an area, co-occurred with shoaling of the CDW since the 1980s, influencing      too far south, and occurring too frequently (Section 9.2.1.3 and the continental shelf most in the Amundsen-Bellingshausen Seas,        Figure 9.5; Heuz&#xe9;, 2017) related to biases in sea ice extent, overflows, Wilkes Land, and the Antarctic Peninsula. New evidence since SROCC      and freshwater forcing (Deshayes et al., 2014; H. Wang et al., 2015).
reinforces confidence in the importance of the role of winds in        As a result, the AMOC in CMIP5 was nearly always too shallow, with too transporting heat associated with CDW to continental shelves and        weak a temperature contrast between the northward and southward ice cavities in the Amundsen-Bellingshausen Seas (Dotto et al., 2019)  flowing branches. Deep convection errors are still large in CMIP6, and and via variable small-scale undercurrents to the Shirase Glacier      the shallow bias in AMOC persists (Weijer et al., 2020; Heuz&#xe9;, 2021).
Tongue in East Antarctica (Hirano et al., 2020; Kusahara et al., 2021). Since AR5, there is emerging evidence that enhancing horizontal There is limited evidence that increased greenhouse gas forcing has    resolution can reduce long-standing climate model biases in AMOC caused a slight mean change of the local winds from 1920 to 2018,      strength, where the magnitude and profile of northward heat transport facilitating CDW heat intrusion onto the Amundsen-Bellingshausen        at 26&deg;N become more comparable to observations (Chassignet et al.,
continental shelf and ice shelf melt (Holland et al., 2019). Multiple  2020; Roberts et al., 2020). The sensitivity of the AMOC to ocean lines of observational, numerical, theoretical, and paleo evidence      resolution, however, is model-dependent and can be positive as well as provide high confidence that changes in wind pattern (Spence et al.,    negative (Roberts et al., 2020). An increase in AMOC strength at 26&deg;N, 2014; Dotto et al., 2019; Holland et al., 2019), increased ice-shelf    with higher resolution in the ocean component, has been associated melt (Golledge et al., 2019; Moorman et al., 2020), reduction in sea    with too strong (deep) convection in the subpolar gyre and too deep ice production (Timmermann and Hellmer, 2013; Obase et al., 2017),      winter mixed layers (L.C. Jackson et al., 2020), which occurs in most and eddies (Stewart and Thompson, 2015; Thompson et al., 2018) can      CMIP6 models that are unable to overflow deep water formed in the 1236
 
Ocean, Cryosphere and Sea Level Change                                                                                              Chapter 9 Nordic Seas across the Greenland-Iceland-Scotland Ridge. Models        The AMOC is a potential driver of Atlantic Multi-decadal Variability with a correct AMOC strength may do so by compensating a lack of        (AMV), but there is new evidence that anthropogenic aerosol deep-water outflow from the Nordic Seas through too much deep          changes have contributed to observed AMV changes, and that convection and deep-water formation in the Labrador and Irminger        underestimation of the magnitude and duration of AMV changes Seas (Heuz&#xe9;, 2021).                                                    in CMIP5 is tempered in CMIP6 (Section 3.7.7 and Annex IV.2.7).
Comparison of observed AMOC variability at the RAPID section with Models and paleoreconstructions have often assumed a close relation    modelled variability reveals that CMIP5 models appear to largely between the AMOC and deep convection in the Labrador Sea; the          underestimate the interannual and decadal time scale variability Labrador Sea convection variability has been interpreted as connecting  (Roberts et al., 2014; Yan et al., 2018), and similar results are found to AMOC variability. Observational studies have been inconclusive on    when analysing CMIP6 models (Section 3.5.4.1). By underestimating whether this relation exists (Buckley and Marshall, 2016). New insight  the multi-decadal AMOC-AMV link and other low-frequency AMOC from observed overturning in the eastern and western subpolar          variability, climate models also underestimate internal variability gyre in the North Atlantic in OSNAP (Lozier et al., 2019; Petit et al., in subpolar SSTs that feed back on the North Atlantic Oscillation 2020) reveals that 15.6 +/- 3.1Sv takes place north of the OSNAP array    (NAO). This causes the NAO to lack variability on multi-decadal time between Greenland and Scotland, with only 2.1 +/- 0.9 Sv of overturning  scales (Kim et al., 2018). Despite the role of the AMOC in generating  9 occurring across the Labrador Sea, as found with the OSNAP 53&deg;N        AMV through subsurface temperatures in antiphase with SST and array spanning the mouth, calling into question the validity of the    downward heat fluxes into the ocean that anticorrelate with SSTs Labrador Sea convection-AMOC link (Lozier et al., 2019). Although      (R. Zhang et al., 2019), it is generally accepted that AMOC forcing these results are derived from only the first 21 months of data from    of SST variability exists alongside stochastic wind forcing and monitoring since 2014, hydrographic observations during 1990-1997      external forcing by aerosols (Bellomo et al., 2018; Haustein et al.,
previously found small overturning (1-2 Sv) in the Labrador Sea        2019; OReilly et al., 2019; Wills et al., 2019).
(Pickart and Spall, 2007). However, previous estimates of Labrador Sea Water formation (obtained with different techniques) suggest        The SROCC (Collins et al., 2019) assessed that in situ observations larger overturning (Haine et al., 2008). Part of this controversy could (2004-2017) and sea surface temperature reconstructions indicate be explained if a large fraction of newly formed Labrador Sea Water    that AMOC has weakened relative to 1850-1900 (medium is not exported from the Labrador Sea. The OSNAP observations are      confidence). However, SROCC also assessed that there is insufficient supported by previous hydrographic measurements in showing strong      data to quantify the magnitude of the weakening, or to properly east-west symmetry in isopycnal slope in the Labrador Sea in periods    attribute it to anthropogenic forcing, due to the limited length of of both strong and weak convection; this implies compensating          the observational record. Here, this assessment is adjusted to low northward and southward transport above and below the potential        confidence in the weakening (as also discussed in Sections 2.3.3.4.1 density surface that separates the upper and lower overturning          and 3.5.4.1). The CMIP5 multi-model mean showed no 20th century limbs (Lozier et al., 2019), despite large deep convection variability  trend in AMOC (Cheng et al., 2013). The CMIP6 multi-model mean (Yashayaev, 2007; Yashayaev and Loder, 2016). New observations          slightly opposes the reconstructed decline due to a strong increase of deep winter mixing in the Irminger Basin (de Jong et al., 2018;      in the 1940-1985 period (Menary et al., 2020b; Weijer et al., 2020),
Josey et al., 2019) support the assertion that the Irminger Sea, in    thought to be in response to aerosol forcing (Section 3.5.4.1),
addition to the Nordic Seas (Chafik and Rossby, 2019), are the main    followed by a smaller decline since the 1990s. Also, agreement sources of overturning in the eastern subpolar gyre, consistent with    between different proxy-based reconstructions is weak in many OSNAP (Petit et al., 2020). It is unclear to what extent models are in  details (Moffa-S&#xe1;nchez et al., 2019) and questions can be raised disagreement with this view of overturning in the subpolar gyre, as    regarding various proxies used in reconstructions (Section 2.3.3.4.1).
a direct comparison with OSNAP of model analyses partitioning the      For instance, SST-based proxies can be influenced by atmospheric overturning into a western and eastern part is mostly lacking, with    and other processes acting on different time scales (Moffa-S&#xe1;nchez a notable exception (Menary et al., 2020a). Other results give rise to  et al., 2019; Jackson and Wood, 2020). In addition, many proxies considerable uncertainty over veracity of the models in simulating      are indirect and based on AMOC-related processes assumed to be the overturning partitioning between east and west and the role of      similar to those found in models, such as the link between AMOC various drivers of AMOC variability, including: the analysis of water  and Labrador Sea convection, which has been questioned recently mass formation in CMIP6 models (Heuz&#xe9;, 2021); the analysis between      (see above). In addition, the subpolar gyre from which many AMOC Labrador Sea Water formation and AMOC in a suite of ocean-only          proxies are taken may vary independently of AMOC, with similar models (Danabasoglu et al., 2014); and the fact that when the OSNAP    patterns in SST and OHC driven by wind variability (Williams et al.,
observing system design was tested in an eddy-permitting ocean          2014; Piecuch et al., 2017). Finally, a new dynamic reconstruction model comparable amounts of overturning in the western and eastern      of the Atlantic inflow to the Nordic Seas suggests no slowdown subpolar gyre were found (Susan Lozier et al., 2017). Disagreement      over the past 70 to 100 years (Rossby et al., 2020), in contrast to between models and OSNAP observations may decrease in higher-          a new compilation of proxy reconstructions which suggests that resolution models (Menary et al., 2020a). In summary, multiple          AMOC is presently in its weakest state in the last millennium (Caesar lines of evidence provide medium agreement between models and          et al., 2021), reinforcing the evidence that motivated the previous observations on drivers of change and variability in the AMOC and, in  SROCC assessment. Section 3.5.4.1 also questions the veracity of particular, the role of Labrador Sea deep convection in constituting    the models forced AMOC response during the 20th century. Given AMOC variability.                                                      the large discrepancy between modelled and reconstructed AMOC 1237
 
Chapter 9                                                                                                              Ocean, Cryosphere and Sea Level Change in the 20th century, and the uncertainty over the realism of the                            It is unclear what causes a weakening of the deepest limb of AMOC 20th century modelled AMOC response (Section 3.5.4.1), we have                              at 26.5&deg;N, if the main sources for this flow farther north remain low confidence in both.                                                                    constant. Various estimates of AMOC and associated heat transport suggest an increase since the 1940s with a subsequent decrease The strength of AMOC has been measured directly since 2004 using                            since the 1990s (Section 2.3.3.4.1), supported by ocean reanalysis the RAPID Array (Section 2.3.3.4.1; Smeed et al., 2018). RAPID-based                        (Jackson et al., 2019), forced ocean model simulations (Robson et al.,
estimates show a large amount of variability compared to CMIP                              2012; Danabasoglu et al., 2016) and CMIP6 simulations (Menary models (Roberts et al., 2014). Observed changes since 2004 are                              et al., 2020a). This suggests that the observed AMOC-shift between too short for the evaluation of a long-term trend given the decadal                        2007 and 2011 may be part of a longer-term decrease (medium scale internal variability (Section 2.3.3.4.1). Nevertheless, Smeed                        confidence), which has been attributed to be part of multiannual et al. (2018) argue that, between 2007 and 2011, AMOC shifted to                            variability (Rhein et al., 2019).
a state of reduced overturning - decreasing from 18.8 Sv between 2004 and 2008 to 16.1 Sv after 2008. A shift in AMOC strength of                            The SROCC (Collins et al., 2019) found that AMOC will very likely this magnitude is not captured by CMIP5 and CMIP6 models, which                            weaken over the 21st century. In CMIP6 projections, the modelled 9 generally underestimate interannual to decadal AMOC variability                            decline starting in the 1990s continues in all future projections, almost (Section 3.5.4.1). Additional evidence since SROCC also raises the                          independent of the forcing scenario until about 2060, after which inconsistency between the RAPID weakening in the 3000-5000 m                                low-emissions scenarios show stabilization, while high-emissions depth range and the relative constancy of deep overflows from the                          scenarios continue to exhibit AMOC decline (Figure 9.10; Menary Arctic (&#xd8;sterhus et al., 2019), implying that the recent decrease in                        et al., 2020b; Weijer et al., 2020). Despite differences in overall AMOC AMOC at 26.5&deg;N (Smeed et al., 2018) is not caused by overflow                              strength, location and latitude of deep convection, sea ice and SST weakening or reduced overturning in the Nordic Seas, although the                          bias and representation of deep overflows, the model projections are weakening occurred almost exclusively in the 3000-5000 m depth                              qualitatively similar. This agreement suggests that AMOC decline may range associated with a reduction of Lower NADW (Section 9.2.2.3).                          be governed by large-scale constraints independent of the details of Figure 9.10 l Atlantic Meridional Overturning Circulation (AMOC) strength in simulations and sensitivity to resolution and forcing. (Top left) AMOC magnitude (units: Sverdrup (Sv) = 109 kg s-1) in Paleoclimate Modelling Intercomparison Project (PMIP) experiments. (Top right) Time series of AMOC from Coupled Model Intercomparison Project Phase 5 and 6 (CMIP5 and CMIP6) based on (Menary et al., 2020b). (Bottom left) Percent change in AMOC strength per year at different resolutions over the 1950-2050 period with colours for model families (Roberts et al., 2020). (Bottom right) A compilation of percentage changes in the simulated AMOC after applying an additional freshwater flux in the subpolar North Atlantic at the surface for a limited time (de Vries and Weber, 2005; Stouffer et al., 2006; Yin and Stouffer, 2007; Jackson, 2013; Liu and Liu, 2013; Jackson and Wood, 2018; Haskins et al., 2019). Symbols indicate whether the AMOC recovers within 200 years (circles), is starting to recover (upwards arrow), or does not recover within 200 years (downwards arrow). Symbol size indicates rate of freshwater input. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1238
 
Ocean, Cryosphere and Sea Level Change                                                                                              Chapter 9 the models. In theoretical models of the thermohaline circulation, the  climate models to represent the Southern Ocean persist due to most circulation strength is proportional to a density or pressure difference CMIP6 models still using parameterized mesoscale eddy processes, between the subpolar North Atlantic and subtropical South Atlantic      which are limited in projecting the future response of the horizontal (Kuhlbrodt et al., 2007; Weijer et al., 2019). In all models, the north- and vertical circulation under climate warming, and also because south pressure gradient decreases in the 21st century, as subpolar      of the continued absence of active ice-shelf and ice-sheet coupling waters warm faster than subtropical waters, and an enhanced              in the CMIP6 model suite, therefore ignoring basal meltwater hydrological cycle drives freshening at subpolar latitudes, while        and calving feedback on the circulation (Meredith et al., 2019). In subtropical latitudes feature more evaporation and salinification        addition, two important limitations of CMIP6 models of the Southern (Section 9.2.1). As a result, surface waters at subpolar latitudes      Ocean involve processes that were not assessed in SROCC. First, become more buoyant and more stable, so that deep water formation        the poor representation of dense overflows causes most of the driving the AMOC declines (Section 9.2.1.3). Projected AMOC decline      Antarctic Bottom Water (AABW) to be formed by spurious open by 2100 ranges from 24 [4 to 46] % in SSP12.6 to 39 [17-55] %          ocean convection rather than by dense overflows from the Antarctic in SSP58.5 (medium confidence) (Section 4.3.2.3). Note that            continental shelves that feed the lower overturning cell (Snow et al.,
these ranges are based on ensemble means of individual models,          2015; Dufour et al., 2017; Heuz&#xe9;, 2021). Second, Antarctic continental largely smoothing out internal variability. If single realizations are  shelf waters are poorly simulated because potentially important        9 considered, the ranges become wider, especially by lowering the low      controlling mechanisms tend to be too small and transient to end of the range (Section 4.3.2.3). In summary, it is very likely that  observe and resolve in CMIP ocean models. These small processes AMOC will decline in the 21st century, but there is low confidence in    include: the heterogeneity of observed sub-ice-shelf melt with the models projected timing and magnitude. In addition, freshwater      warm water driving narrow basal channels that cut underneath from the melting of the Greenland Ice Sheet (Sections 9.4.1.3 and        the ice (Drews, 2015; Alley et al., 2016; Marsh et al., 2016; Milillo 9.4.1.4) could further enhance the future weakening of AMOC in the      et al., 2019); eddies and tides (Stewart et al., 2018; Jourdain et al.,
21st century (Collins et al., 2019; Golledge et al., 2019).              2019; Hausmann et al., 2020), which can drive Circumpolar Deep Water (CDW) onto the continental shelves or dynamically increase Both AR5 (Collins et al., 2013) and SROCC (Collins et al., 2019)        melting (Section 9.2.3.6); and feedback mechanisms between assessed that an abrupt collapse of AMOC before 2100 was very            ocean, atmosphere and cryosphere that can weaken or amplify unlikely, but SROCC added that, by 2300, an AMOC collapse was            initial perturbations (Donat-Magnin et al., 2017; Spence et al., 2017; as likely as not for high-emissions scenarios. The SROCC also            Turner et al., 2017; Silvano et al., 2018; Webber et al., 2019; Hazel assessed that model bias may considerably affect the sensitivity of      and Stewart, 2020). In addition, the Southern Ocean in CMIP5 and the modelled AMOC to freshwater forcing. Tuning towards stability        CMIP6 models exhibit surface temperature biases (Section 9.2.1.1),
and model biases (Valdes, 2011; Liu et al., 2017; Mecking et al.,        which have been linked in CMIP5 models to errors in atmospheric 2017; Weijer et al., 2019) provides CMIP models a tendency toward        model cloud-related shortwave radiation (Hyder et al., 2018) and unrealistic stability (medium confidence). By correcting for existing    are somewhat improved in High Resolution Model Intercomparison salinity biases, Liu et al. (2017) demonstrated that AMOC behaviour      Project (HighResMIP) models (Figure 9.3). In summary, there is high may change dramatically on centennial to millennial time scales,        confidence that future change in the subpolar Southern Ocean region, and that the probability of a collapsed state increases. None of the    including sea ice cover and ocean temperature change on Antarctic CMIP6 models features an abrupt AMOC collapse in the 21st century,      continental shelves, depends on feedback mechanisms involving the but they neglect meltwater release from the Greenland Ice Sheet.        ocean, atmosphere and cryosphere that are poorly understood and Also, a recent process study reveals that a collapse of AMOC            not represented in the current generation of climate models. This can be induced, even by small-amplitude changes in freshwater            results in large uncertainty and low confidence in the future sea forcing (Lohmann and Ditlevsen, 2021). As a result, we change the        ice cover (Section 9.3.2) and in ocean temperature change on the assessment of an abrupt collapse before 2100 to medium confidence        Antarctic continental shelf (Section 9.4.2.3).
that it will not occur.
Despite these challenges, the CMIP6 ensemble does represent the 9.2.3.2      Southern Ocean                                              main Southern Ocean circulation characteristics: the simulated Antarctic Circumpolar Current (ACC) transport is generally lower than The changing Southern Ocean circulation system exerts a strong          observation-based values but consistent when considering ensemble influence on the global climate by modulating: (i) global OHC            spread, and the inter-model spread in ACC transport has greatly (Section 9.2.2.1); (ii) global ocean anthropogenic carbon uptake        reduced from previous generations of climate models from CMIP3 (Cross-chapter Box 5.3); global ocean overturning circulation            to CMIP6 (Beadling et al., 2019, 2020). The structure (but not the (Section 9.2.3.1); (iii) climate sensitivity (Section 7.4.4 and          magnitude) of the two-cell zonally averaged overturning is captured Cross-chapter Box 5.3); (iv) sea level through basal melt of ice shelves by most CMIP6 models (Russell et al., 2018; Beadling et al., 2019).
(9.4.2); and (v) Southern Hemisphere sea ice cover (Section 9.3.2).      In addition, while issues remain, CMIP6 climate models show clear improvements in their representation of AABW compared to CMIP5:
The SROCC (Meredith et al., 2019) had low confidence in all              several models correctly represent or parameterize Antarctic shelf CMIP5-based model projections due to their inability to explicitly      processes, fewer models exhibit Southern Ocean deep convection, resolve eddy processes, and their inability to properly consider future  bottom density biases are reduced, and abyssal overturning is more meltwater change from the Antarctic Ice Sheet. These limitations of      realistic (Heuz&#xe9;, 2021). In terms of atmospheric wind forcing, CMIP6 1239
 
Chapter 9                                                                                              Ocean, Cryosphere and Sea Level Change models show an improvement compared to CMIP5 models, with                    in the assessments of a long-term increase in upper-ocean overturning.
an overall reduction in the equatorward bias of the annual mean              Consistent with SROCC, the importance of eddy processes and winds westerly jet from 1.9&deg; in CMIP5 to 0.4&deg; in CMIP6, but in contrast,            in driving long-term change and variability have been reinforced, with they show no such overall improvements for their representation of            a potential fast wind response partially counteracted by a slower eddy the Amundsen Sea Low (Bracegirdle et al., 2020; Lyu et al., 2020a),          response (Doddridge et al., 2019; Waugh et al., 2019; Stewart et al.,
which can be critical in driving variability of water masses on the          2020). Eddy parametrizations affect the strength of overturning, its Antarctic continental shelf in west Antarctica, the Weddell Sea or            sensitivity to winds and the ACC transport (Mak et al., 2017). Even the Ross Sea (Holland et al., 2019; Silvano et al., 2020).                    in eddy-resolving simulations, sub-gridscale dissipation affects the overturning and ACC (Pearson et al., 2017). In addition, there has The SROCC (Meredith et al., 2019) established that, while trends in          been progress in understanding the importance of Antarctic Ice Shelf the atmospheric forcing of the Southern Ocean have been dominated            meltwater and sea ice, in driving the observed changes in the near by a strengthening of the Southern Hemisphere westerly winds in              surface and in the upper overturning cell over the past decades, on recent decades, there is medium confidence that ACC transport is              top of changes induced by winds and eddies (Bronselaer et al., 2020; weakly sensitive to changes in winds. It also reported that, instead of      Haumann et al., 2020; Jeong et al., 2020; Rye et al., 2020). In particular, 9 increasing the mean ACC transport, additional energy input associated        increased stratification caused by increased freshwater flux to the with increased wind stress cascades into the eddy field (medium              surface ocean (Section 9.2.1.3) can cause a shoaling and warming of confidence). In contrast with the AR5 assessment (Rhein et al., 2013),        the CDW layer, and create a positive feedback, enhancing basal melt SROCC evaluated that it was unlikely that there has been a net                of the Antarctic Ice Sheet (Section 9.4.2.1; Bronselaer et al., 2018; southward migration of the mean ACC position over the past 20 years.          Golledge et al., 2019; Schloesser et al., 2019; Sadai et al., 2020). There There is no additional evidence to revisit SROCC assessment on wind          is medium confidence in the existence of this feedback mechanism sensitivity. However, new evidence does suggest that air-sea buoyancy        but low agreement on the magnitude of the feedback. The SROCC forcing associated with idealized 4xCO2 forcing leads to an increase in      reported that CMIP5 models project that the overall transport ACC transport (limited evidence) (Shi et al., 2020). The SROCC noted          of upper-ocean overturning cell will increase by up to 20% in the that, if the general strengthening in westerly winds is sustained, then      21st century, and no new studies alter that assessment.
it is very likely that the eddy field will continue to increase in intensity, and it is likely that the mean position and strength of the ACC will          For the lower cell overturning circulation, SROCC assessed that remain only weakly sensitive to winds. In the future, the strength of        a slowdown of its transport is consistent with the observed decrease the Southern Hemisphere westerly wind jet results from a competition          in volume (medium confidence) of AABW in the global ocean between decrease due to ozone hole recovery and increase due to              (Section 9.2.2.3). Additional evidence since SROCC strengthens increased radiative forcing (Section 4.3.3.1). This competition results      confidence that increased glacial meltwater flux will reduce the in an increased atmospheric jet by 2100 compared to present day              density of bottom waters during the 21st century. It will eventually under SSP24.5, SSP37.0, and SSP58.5, but a decreased jet by 2100          reach a point where deep convection will be curtailed, and shelf under SSP12.6 (Bracegirdle et al., 2020). There is little inter-model        water will become too buoyant to sink to the ocean interior, thereby spread in the CMIP6 future response of the atmospheric westerly              slowing the lower cell overturning circulation (Bronselaer et al., 2018; jet, providing high confidence in this assessment (in contrast, CMIP6        Golledge et al., 2019; Lago and England, 2019; Moorman et al., 2020).
models show no consistency in their future projection of easterly wind        While such changes are consistent with the observed freshening change along the Antarctic continental shelf break; Bracegirdle et al.,      and decreased volume of the AABW layer reported in SROCC 2020). Paleo-oceanographic evidence suggests that ACC flow through            (as discussed in Section 9.2.2.3), new observation-based studies have Drake Passage was consistently stronger during warm intervals of the          highlighted how the lower cell overturning can episodically increase past (both during interstadials and interglacials), but with relatively      as a response to climate anomalies, temporally counteracting the little change and no consensus on the sign of change in other regions        tendency for melt to reduce AABW formation (Abrahamsen et al.,
(Lamy et al., 2015; Toyos et al., 2020). In summary, additional evidence      2019; Castagno et al., 2019; Gordon et al., 2020; Silvano et al., 2020).
since SROCC confirms that there is medium confidence that the ACC            In addition, while the opening of open ocean polynyas can affect has been weakly sensitive to Southern Hemisphere atmospheric jet              the lower cell on decadal to centennial time scales, there is limited increase in the past decades. New evidence since SROCC suggests that          evidence and low agreement in the role of open ocean polynyas there is high confidence that the Southern Hemisphere atmospheric jet        in driving observed trends of the lower cell in the last decade will increase in the 21st century for all scenarios (except for SSP11.9      (Section 9.2.2.3). Based on CMIP5 models, SROCC reported with low and SSP12.6; Section 4.3.3.1) with a greater increase for larger            confidence that formation and export of AABW associated with the radiative forcing. An increase in westerly winds will very likely force      lower overturning cell will decrease in the 21st century, and there an increase of the eddy field in the ACC, and while there is medium          is no new evidence to revisit that assessment from climate models.
confidence that the ACC is weakly sensitive to wind change, new              However, additional paleo evidence from marine sediments suggests advances since SROCC provide limited evidence that the ACC transport          that AABW formation/ventilation was vulnerable to freshwater will nevertheless increase in response to wind and buoyancy fluxes.          fluxes during past interglacials (Hayes et al., 2014; Huang et al.,
2020; Turney et al., 2020) and that AABW formation was strongly For the upper cell overturning circulation, SROCC concluded that:            reduced (Skinner et al., 2010; Gottschalk et al., 2016; Jaccard et al.,
its transport has experienced significant inter-decadal variability in        2016) or possibly totally curtailed (Huang et al., 2020) during the response to wind forcing since the 1990s; and there is low confidence        Last Glacial Maximum (LGM) and transient cold intervals of marine 1240
 
Ocean, Cryosphere and Sea Level Change                                                                                                    Chapter 9 isotope stages 2 and 3 (MIS2 and MIS3). Specifically, sedimentary            central Pacific, and thus the meridional temperature gradients that reconstructions show a transient reduction in AABW ventilation              drive tropical instability waves (Terada et al., 2020), along with in the Atlantic sector of the Southern Ocean during MIS5e, which            a weakening, flattening and shoaling of the tropical thermocline and is assessed to have been warmer than modern climate (Thomas                  equatorial undercurrent (Luo and Rothstein, 2011). A weak or absent et al., 2020). However, long multi-centennial or millennial model            equatorial undercurrent (Kuntz and Schrag, 2020) and a too-diffuse runs under higher-than-pre-industrial CO2 concentrations show that,          and incorrectly sloped tropical thermocline (Zhu et al., 2020) remain after 500-1000 years, ventilation in the Southern Ocean resumes,            issues in most CMIP6 models. In summary, while future changes in and possibly overshoots with enhanced convection in the Weddell              tropical modes of variability remain unclear, change in atmospheric and Ross seas, leading to enhanced bottom water ventilation globally        and ocean circulation will drive continued change in tropical ocean (Yamamoto et al., 2015; Frlicher et al., 2020). AABW ventilation            temperature in the 21st century (medium confidence), with part increased at the onset of the last deglacial transition, promoting          of the region experiencing drastic marine heat wave conditions the release of previously sequestered CO2 to the atmosphere on              (high confidence).
centennial to millennial time scales (Bauska et al., 2016; Jaccard et al.,
2016; Rae et al., 2018), concomitant with a southward shift of the          9.2.3.4    Gyres, Western Boundary Currents Southern Hemisphere westerly wind belt (Denton et al., 2010; Jaccard                    and Inter-basin Exchanges                                    9 et al., 2016) and reduced sea ice cover (Ferrari et al., 2014; Stein et al.,
2020). In summary, the combination of observational, numerical and          The AR5 (Rhein et al., 2013) assessed with medium to high confidence paleoclimate evidence provides us with medium confidence that the            that the North Pacific subpolar gyre, the South Pacific subtropical gyre, lower cell will continue decreasing in the 21st century as a result of      and the subtropical cells have intensified. They also reported that the increased basal melt from the Antarctic Ice Sheet.                          North Pacific subtropical gyre had expanded since the 1990s, and that, overall, the changes in gyre systems were likely predominantly 9.2.3.3    Tropical Oceans                                                  due to interannual-to-decadal variability. The SROCC (Meredith et al.,
2019) complemented the AR5 assessment by reporting that the polar The tropics are a tightly coupled ocean-atmosphere system with              Beaufort Gyre in the Arctic expanded to the north-west between tightly interconnected basins (Cai et al., 2019). The zonal atmospheric      2003 and 2014, contemporaneous with changes in its freshwater Walker Circulation and the Indonesian Throughflow (Figure 9.11) are          accumulation and alterations in wind forcing. Consistent with the key connections between the Pacific and Indian oceans, and variations        reported change over the gyres, both AR5 and SROCC (Bindoff et al.,
in the Walker and Hadley Circulations are tightly linked to the tropical    2019; Collins et al., 2019) reported that western boundary currents Pacific SST and currents. The tropics have a profound influence on the      (WBCs) have intensified (Figure 9.11), and expanded poleward, climate system through the multiple modes of variability they host,          except for the Gulf Stream and the Kuroshio. Section 2.3.3.4 which have widespread global influence at seasonal to annual time            provides an overall assessment of gyres and WBCs, including an scale (Annex IV).                                                            assessment of change from paleoclimate archives. Section 2.3.3.4 assesses that, while WBC strength is highly variable at multi-decadal The effect of tropical modes of variability on climate and their            scale (high confidence), WBCs and subtropical gyres have shifted long-term changes are reviewed in detail in Annex IV, while changes          poleward since 1993 (medium confidence), at a rate on the order of to the tropical ocean are assessed throughout the report and                0.04-0.1 degree per decade during 1993-2018. Figure 9.11 shows briefly summarized here. Section 2.4 concludes that a sustained              that CMIP5 and CMIP6 models agree in projecting a weaker Gulf shift beyond multi-centennial variability has not been observed              Stream and Gulf Stream Extension, while the Kuroshio changes less for El Nino-Southern Oscillation (ENSO) (medium confidence) and            (Sen Gupta et al., 2016).
that there is limited evidence and limited agreement about the long-term behaviour of other tropical modes. Section 3.7 assesses            Although the observed wind stress curl shows systematic with high confidence that human influence has not affected the              poleward shift in each basin as a result of anthropogenic warming principal tropical modes of interannual climate variability and their        (Section 2.3.1.4; Chen and Wu, 2012; Wu et al., 2012; Zhai et al., 2014),
associated regional teleconnections beyond the range of internal            which has caused a systematic shift of the WBCs and subtropical variability. Section 4.3.3.2 assesses with medium confidence                gyres since 1993 (Wu et al., 2012; Yang et al., 2016, 2020), the that there is no consensus from models for a systematic change              response of current strength is more complex and inconsistent across in the amplitude of ENSO sea surface temperature variability over            regions (Sloyan and OKane, 2015; Y.-L. Wang et al., 2016; Elipot and the 21st century. The related change in tropical SSTs is covered in          Beal, 2018; McCarthy et al., 2018; Wang and Wu, 2018; Dong et al.,
Section 9.2.1.1. The projected changes in SST have implications for          2019). The strength of WBCs and gyres exhibit inconsistent responses marine heat wave characteristics, which are assessed in Box 9.2.            because they are dependent on wind stress forcing and because SST changes in the tropics are related to changes in the atmospheric        multi-scale interaction and air-sea interaction have an important circulation, including surface equatorial easterly trade winds and          role in their long-term trends and variability (Zhang et al., 2020).
Walker Circulation (Section 4.5.3.2), and the weakening Indonesian          Observed changes in gyre circulation are dominated by interannual Throughflow and strengthening Agulhas Extension and leakage                  and decadal modes of variability globally (Qiu and Chen, 2012; Melzer (Section 9.2.3.4). Weakening trade winds under climate change                and Subrahmanyam, 2017; McCarthy et al., 2018; Hu et al., 2020).
(Vecchi and Soden, 2007) will tend to decrease upwelling, along              The North Atlantic subpolar gyre is strongly modulated by variability isopycnals in the eastern Pacific and diapycnal upwelling in the            associated with the NAO and AMV (Annex IV; Robson et al., 2016).
1241
 
Chapter 9                                                                                                                Ocean, Cryosphere and Sea Level Change Subpolar gyre systems can change abruptly due to a positive                                    mesoscale eddies, and also improves simulation of the strength and feedback between convective mixing and salinity transport (Born                                position of WBCs such as the Kuroshio Current, Gulf Stream, and et al., 2013, 2016) and air-sea interaction (Moffa-S&#xe1;nchez et al.,                              East Australian Current (very high confidence) (Sasaki et al., 2004; 2014; Moreno-Chamarro et al., 2017) within the gyre. In the Arctic,                            Chassignet and Marshall, 2008; Delworth et al., 2012; Yu et al., 2012; both the Beaufort gyre and mesoscale eddies strengthened between                                Small et al., 2014; Haarsma et al., 2016; Chassignet et al., 2017, 2003 and 2014 (Armitage et al., 2017), which might be partly due                                2020; Hewitt et al., 2020). Improved boundary current location to increased wind stress (Oldenburg et al., 2018) or reduced sea ice                            relates to improved recirculation regions (Jayne et al., 2009), mean thickness and changes in sea ice pack morphology (van der Linden                                path and variability, and existence of multiple stable paths (Qiu et al., 2019). Presently, there is limited evidence in attributing                              et al., 2005; Delman et al., 2015), air-sea fluxes (Small et al., 2014),
causality to these changes for any of the proposed mechanisms.                                  and related coastal weather patterns (Kaspi and Schneider, 2011).
In the North Pacific, there has been an increasing trend in the Alaska                          The wind-current feedback, implemented by considering relative Gyre from 1993 to 2017 (Cummins and Masson, 2018), which might                                  velocity of currents and wind, realistically dampens mesoscale be attributed to Pacific Decadal Oscillation (low confidence) (Hristova                        eddies and WBCs, through mesoscale air-sea interaction (Ma et al.,
et al., 2019). In the Southern Ocean, limited evidence indicates that                          2016; Renault et al., 2016, 2019), even though sub-mesoscale wind-9 the subpolar gyres respond to Southern Hemisphere atmospheric                                  current damping feedback is missing in these models (medium modes of variability at interannual time scale (Armitage et al., 2018;                          confidence) (Z. Zhang et al., 2016). As eddies potentially play a role Dotto et al., 2018).                                                                            in determining the strength of gyre circulations and their low-frequency variability (Fox-Kemper and Pedlosky, 2004; Berloff et al.,
All climate models reproduce WBCs and gyres, but eddy-                                          2007), it is expected that eddy-present and eddy-rich models will present or eddy-rich models (roughly 10-25 km and about                                        differ in their decadal variability and sensitivity to changes in the 10 km resolution, respectively) represent these currents more                                  wind stress of gyres from eddy-parameterized models (medium realistically than eddy-parameterized models (very high confidence)                            confidence). Nonetheless, important aspects of gyre strength (Small et al., 2014; Griffies et al., 2015; Chassignet et al., 2017,                            depend primarily on forcing and not resolution, allowing long-term 2020; Hewitt et al., 2017, 2020; Roberts et al., 2018). Compared                                changes in gyre strength to be investigated with low-resolution to observations or to eddy-present and eddy-rich models, the eddy-                              climate models (Hughes and de Cuevas, 2001; Yeager, 2015).
parameterized models from CMIP5 and CMIP6 simulate weaker and wider WBCs, as well as less realistic locations of subtropical and                              Under future scenarios RCP4.5 and RCP8.5, AR5 (Collins et al.,
subpolar gyre boundaries (Figure 9.11). Increased resolution admits                            2013) assessed an intensification and poleward extension of SSP5-8.5 Color High model agreement (80%)
Low model agreement (<80%)
Figure 9.11 l Simulated barotropic streamfunction, surface speed and major current transport in Coupled Model Intercomparison Project Phase 5 and 6 (CMIP5 and CMIP6). (a) Mean barotropic streamfunction (unit: 109 kg s-1; 1995-2014) and projected barotropic streamfunction change (109 kg s-1; 2018-2100 vs 1995-2014) under (b) SSP58.5. (d) Mean surface (0-100 m) speed (m s-1) and projected surface speed change (m s-1, 2081-2100) versus 1995-2014 under (e) SSP58.5. (c, f) Median and likely range of 1995-2014 and 2081-2100 transport of three currents with the largest transport change and four with the largest fractional change (Sen Gupta et al.,
2016). (c) Deep currents: Agulhas Extension (ACx), Gulf Stream (GS), Gulf Stream Extension (GSx), Tasman Leakage (TASL), East Australia Current Extension (EACx), Indonesian Throughflow (ITF), and Brazil Current (BC). (f) Shallow currents: as for deep but with New Guinea Current (NGC), and without ACx. No overlay indicates regions with high model agreement, where 80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where <80% of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1242
 
Ocean, Cryosphere and Sea Level Change                                                                                            Chapter 9 the southern Hemisphere subtropical gyres in the 21st century.          1990-2015. Realistic representation of the Bering Strait transport New evidence since AR5 further reinforces their conclusions, which      in the current generation of climate models is challenging because are now extended to all subtropical gyre systems in the Northern        the strait is narrow compared to the resolution of climate models and Southern hemispheres (Yang et al., 2016, 2020). CMIP6 models        (Clement Kinney et al., 2014; Aksenov et al., 2016). For the Atlantic project changes in WBCs that are consistent with projected changes      to Arctic transport, Section 2.3.3.4 reports that the major branches in the surface winds. Under strong radiative forcing, in scenario      of Atlantic Water inflow across the Greenland-Scotland Ridge have SSP58.5, CMIP6 models project that the East Australian Current        remained stable, with only the smaller pathway of Atlantic Water Extension, Agulhas Current Extension and Brazil Current will intensify  north of Iceland showing a strengthening trend during 1993-2018.
in the 21st century, while the Gulf Stream will weaken (Figure 9.11). Section 2.3.3.4 also assesses that the Arctic outflow remained stable Although CMIP5/CMIP6 are limited in resolution, medium confidence      from the mid-1990s to the mid-2010s. Future changes in these is given to changes in WBCs due to consistency across generations of    currents have not yet been studied in CMIP6 models.
climate models, including CMIP6, despite changes in model structure, resolution and parametrizations.                                        9.2.3.5    Eastern Boundary Upwelling Systems The SROCC (Collins et al., 2019) concluded with high confidence that    Eastern boundary upwelling systems (EBUS) exist where trade winds      9 Indonesian Throughflow (ITF) transport from the Pacific Ocean to        draw cold and generally low-pH/low-oxygen waters upward. Coastal the Indian Ocean has increased in the past two decades as a result      upwelling plays a key role in supplying the food chain with nutrients, (medium confidence) of an unprecedented intensification of the          hence the richness and productivity of EBUS (Bindoff et al., 2019).
equatorial Pacific trade wind system. Section 2.3.3.4 assesses that    The SROCC (Bindoff et al., 2019) assessed with high confidence that there is high confidence that the increase in the ITF over the past    three out of the four major EBUS have experienced large-scale wind two decades is linked to multi-decadal scale variability rather than    intensification in the past 60 years (only the trend for the Canary a longer-term trend. Consistently, in the future, as winds change      Current is considered uncertain). However, it also emphasized that under increased radiative forcing, most models project a decline of    various processes can also modulate, or even reverse, wind trends the ITF on the centennial time scale (Figure 9.11). One of the clearest locally (Bindoff et al., 2019). Here we revisit SROCC assessment changes of ocean current transport simulated by climate models is      (Bindoff et al., 2019) based on evidence showing low agreement a weakening of the Indonesian Throughflow, projected in CMIP5          between studies that have investigated trends over past decadess of simulations under RCP4.5 and RCP8.5 scenarios (Sen Gupta et al.,        upwelling-favourable winds (Varela et al., 2015). This low agreement 2016; Stellema et al., 2019), and in CMIP6 simulations under the        has been related to differences in wind products, season of interest, SSP58.5 scenario (high confidence, Figure 9.11).                      and length of the considered time series (Varela et al., 2015). Based on this, we assess that only the California Current system has The SROCC reports with high confidence that the Agulhas leakage        experienced large-scale upwelling-favorable wind intensification over from the Indian to the Atlantic Ocean has increased in the past two    the period 1982-2010, albeit with regional differences (Garc&#xed;a-Reyes decades (Collins et al., 2019), and there is no additional evidence    and Largier, 2010; Seo et al., 2012). In the Benguela, Canary, and since then allowing this assessment to be revisited (Biastoch et al.,  Humboldt systems, large-scale, upwelling-favourable wind trends 2015; Loveday et al., 2015; L&#xfc;bbecke et al., 2015). There is low        are ambiguous, owing to low confidence in long-term in situ confidence in future projections of Agulhas leakage because most        marine wind data (Cardone et al., 1990; Bakun et al., 2010) and low CMIP models cannot directly simulate it, due to coarse resolution.      agreement among available studies (Narayan et al., 2010; Sydeman However, there is medium evidence that the strength of the Southern    et al., 2014; Varela et al., 2015). Our assessment confirms SROCC Hemisphere westerlies controls Agulhas leakage (Durgadoo et al.,        assessment (Bindoff et al., 2019) in that high natural variability of 2013; Biastoch et al., 2015; Loveday et al., 2015), and high confidence EBUS and their inadequate representation by most climate models that the strength of the Southern Hemisphere westerlies will increase  gives low confidence in attribution of observed changes, while under increased radiative forcing, except in lower warming scenarios    anthropogenic changes are projected to emerge primarily in the (SSP11.9, SSP1.2-6; Section 4.3.3.1; Bracegirdle et al., 2020). There  second half of the 21st century (limited evidence: one model and one is also evidence that increasing Agulhas leakage is consistent with    study) (Brady et al., 2017).
observed change of the temperature and salinity structure in the Atlantic ocean, and with variability of the AMOC (Section 9.2.3.1;      Under increased radiative forcing, SROCC (Bindoff et al., 2019)
Biastoch et al., 2015). This range of indirect evidence provides medium assessed that climate models project, in the 21st century, confidence that the Agulhas leakage will increase in the 21st century,  a reduction of wind and upwelling intensity in EBUS at low latitudes, except for the strongest mitigation scenario (Figure 9.11).            and enhancement at high latitudes, under scenario RCP8.5, with an overall reduction in either upwelling intensity or extension. It also The SROCC assessed that the annual Bering Strait volume transport      highlighted that coastal warming and wind intensification may lead from the Pacific to the Arctic Ocean increased from 2001-2014,          to variable countervailing responses to upwelling intensification consistent with an estimated increased northward heat transport of      at local scales. Despite differences among EBUS (D. Wang et al.,
about 60% from 2001-2014, and an increased freshwater transport        2015), there is growing evidence since SROCC in this pattern of of 30 +/- 20 km3 yr -1 from 1991 to 2015 (Meredith et al., 2019).        change. While it has long been hypothesized that, for upwelling Section 2.3.3.4 assesses that volume transport from the Pacific to      winds, change is linked to air temperature contrast between ocean the Arctic has increased since the 1990s from 0.8 Sv to 1.0 Sv over    and land (Bakun, 1990), this hypothesis has increasingly been 1243
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change challenged. Changes in sea level pressure and wind fields in EBUS          circulations and mixing, leading to indirect connections between appear to be primarily tied to those affecting subtropical highs            the inner shelves and coastlines and offshore conditions. Coastal (Garc&#xed;a-Reyes et al., 2013). Poleward expansion of the Hadley cell          processes link to large-scale metrics of climate and regional effects, (Section 2.3.1.4.1; Staten et al., 2018) and the related poleward          from changing rivers and estuaries, melt and runoff to deep water, migration of subtropical highs (He et al., 2017; Cherchi et al., 2018),    to how changes offshore affect regional and coastal conditions.
produce robust patterns of changes of reduced upwelling at low latitude and enhanced upwelling at high latitude (Echevin et al.,          Shelf-deep ocean exchanges involve eddying, tidal, or turbulent 2012; Belmadani et al., 2014; Bettencourt et al., 2015; Rykaczewski        motions and small-scale topography such as submarine canyons; et al., 2015; Sousa et al., 2017; Lamont et al., 2018; Sylla et al., 2019). high-resolution observations and models are needed to capture These patterns are most apparent in summer in both hemispheres.            these effects (Greenberg et al., 2007; Capet et al., 2008; Allen and Synoptic variability of upwelling winds, important to the functioning      Durrieu de Madron, 2009; Colas et al., 2012; Trotta et al., 2017).
of upwelling ecosystems (Garc&#xed;a-Reyes et al., 2014), may also be            Example coastal processes that introduce uncertainty into large-scale affected by climate change (Aguirre et al., 2019). However, coarse          projections are exchange of CDW across the Antarctic shelf-break, resolution model projections of winds in upwelling regions may be          which affects AABW formation and Antarctic ice-shelf-ocean 9 more consistent than higher-resolution projections, as these regions        interaction (Sections 9.2.2.3 and 9.2.3.2; Stewart and Thompson, are highly sensitive to resolution (Small et al., 2015).                    2013, 2015), river and estuarine plumes and their responses to water level and hydrology change (Banas et al., 2009; Sun et al., 2017),
Projected future annual cumulative upwelling wind changes at most          fjord dynamics linked to glacial outflows (Straneo and Cenedese, locations, and seasons remain within +/-10-20% of present-day                2015; Torsvik et al., 2019), and changing formation of water masses values in the 21st century, even in the context of high-end                in marginal seas (Kim et al., 2001; Greene and Pershing, 2007; Giorgi emissions scenarios (4xCO2 or RCP8.5) (medium confidence).                  and Lionello, 2008; Renner et al., 2009). Downscaling projections Changes due to wind stress curl and alongshore pressure gradients          to the local level allows process detail (Foreman et al., 2014; tend to agree with alongshore wind changes (Oerder et al., 2015;            Mathis and Pohlmann, 2014; Meier, 2015; Tinker et al., 2016). Some Sylla et al., 2019). Direct estimation of oceanic upward transport          processes can only be simulated when coastal models are forced by (Oyarz&#xfa;n and Brierley, 2019; Sylla et al., 2019) and nutrient flux          larger-scale models of the atmosphere, cryosphere, or hydrosphere into the euphotic layer (Jacox et al., 2018) provide a meaningful          (Seo et al., 2007, 2008; Somot et al., 2008; Oerder et al., 2015; Renault estimator of upwelling, integrating all relevant processes, including      et al., 2016; Y. Zhang et al., 2016; Whlin et al., 2020), including the changes in wind stress curl. However, there is limited evidence            addition of tides (Janekovi and Powell, 2012; Timko et al., 2013; from vertical velocity of climate models and missing processes in          Tinker et al., 2015; Pickering et al., 2017; Hausmann et al., 2020).
coarse-resolution climate models that presently limit this approach.        Due to coastal process complexity and small scale, linking the Change in upper-ocean stratification (Section 9.2.1.3) is projected        effects of coastal ocean changes to global ocean changes requires to increase confinement of upwelling vertical velocities to near the        high-resolution modelling (Holt et al., 2017, 2018), two-way nesting, ocean surface (high confidence) (Oerder et al., 2015; Oyarz&#xfa;n and          or local mesh refinement (Fringer et al., 2006; Zhang and Baptista, Brierley, 2019).                                                            2008; Mason et al., 2010; Dietrich et al., 2012; Hellmer et al., 2012; Ringler et al., 2013; Q. Wang et al., 2014; Zngl et al., 2015; Y.J. Zhang In summary, SROCC and this Report conclude that the California              et al., 2016; Soto-Navarro et al., 2020). Coarse climate models and Current system has experienced some upwelling-favourable wind              HighResMIP models do not represent some coastal phenomena such intensification since the 1980s (high confidence), while low agreement      as cross-shelf exchanges and sub-mesoscale eddies, which require among reported wind changes in the Benguela, Canary, and Humboldt          1 km or finer resolution. Thus, there is low confidence in projecting systems prevent a similar assessment. As in SROCC, there is low            centennial scale coastal climate change where regional downscaling confidence in attribution of observed changes to anthropogenic or          or refinement is lacking. There is high confidence in the ability of natural causes. New evidence reinforces our confidence in SROCC            regional coupled models to improve coastal climate change process assessment that, under increased radiative forcing, EBUS winds will        understanding and provide regional information (Section 12.4), but change with a dipole spatial pattern within each EBUS of reduction          many sites globally await such projections.
(weaker and/or shorter) at low latitude, and enhancement (stronger and/or longer) at high latitude (high confidence). There is medium confidence that, across all scenarios, upwelling wind changes in            9.2.4        Steric and Dynamic Sea Level Change EBUS will remain moderate in the 21st century, within +/-10-20%
from present-day values.                                                    9.2.4.1      Global Mean Thermosteric Sea Level Change 9.2.3.6      Coastal Systems and Marginal Seas                              Changes in globally averaged ocean heat content (OHC) cause global mean thermosteric sea level (GMTSL) change (Box 9.1). The Beyond the worlds coastlines lie the shoreline, shallow estuaries,        observed increased OHC for 1971-2018 of 325 to 546 ZJ (very likely continental shelves, and deeper fjords and slopes, where depths            range) (Section 7.2, Box 7.2) has led to a GMTSL rise of 0.03 to 0.06 increase rapidly from the shelves to the deep-ocean floor. It is more      m out of a total global mean sea level (GMSL) of 0.07 to 0.15 m difficult to transport fluid across (rather than along) the shelf-break    (very likely range) (Section 2.3.3.3, Tables 2.7 and 9.5, and Cross-or slope (Brink, 2016), and estuaries and shelves have complex              Chapter Box 9.1).
1244
 
Ocean, Cryosphere and Sea Level Change                                                                                                              Chapter 9 Projections of GMTSL rise in AR5 (Church et al., 2013b) and SROCC        are storing the added heat, and the commitment time scale (Hallberg (Oppenheimer et al., 2019) were derived from the CMIP5 ensemble,          et al., 2013). For paleoclimate, a scaling for sea surface temperature after removing drift estimated based on pre-industrial control            (0.6 m &deg;C-1) or global surface air temperature (GSAT; see Cross-Chapter simulations. Differences between removing a linear and a quadratic        Box 2.3) can be estimated, but mean ocean temperature is in phase drift are small (Hobbs et al., 2016a; Hermans et al., 2021). These prior  with steric sea level change, while sea surface temperatures are not assessments filled in projections for models that did not provide GMTSL  (Figure 9.9; Shakun et al., 2012; Tierney et al., 2020). Thus, while rise for all scenarios, by calculating the heat content of the climate    conversions between OHC, mean ocean temperature and GMTSL system from global surface air temperature and net radiative flux, then  across applications are within uncertainty ranges (medium confidence) converting this to GMTSL rise using each models diagnosed expansion      (Table 9.1), little consistency is found when correlating these variables efficiency coefficient. In AR5, the associated uncertainties were derived to SST or GSAT, which may vary independently.
by assuming a normal distribution, with the 5th-95th percentile CMIP5 ensemble range taken as the likely range (+/-1 standard deviation).        Short-lived climate forcers (Sections 6.3 and 6.6.3) are associated with a sea level commitment, due to an OHC and mean ocean In this Report, global surface air temperature projections are not        temperature response that lasts substantially longer than their derived directly from the CMIP6 ensemble (Box 4.1). Therefore, to        atmospheric forcing and SST response, although not as long as the sea                    9 produce projections of OHC and GMTSL rise consistent with the            level commitment associated with CO2 emissions (Sections 9.2.1.1 Reports assessment of equilibrium climate sensitivity and transient      and 4.4.4). For example, Zickfeld et al. (2017) find that about 70% of climate response (Section 7.5.2.2), this chapter employs a two-layer      the thermosteric sea level rise associated with methane forcing energy budget emulator (Supplementary Materials 7.SM.2, 9.SM.4.3).        would persist 100 years after the elimination of methane emissions, Since AR5, climate model emulators have been increasingly used to        and 40% would persist for more than 500 years.
predict GMTSL (Cross-Chapter Box 7.1; Kostov et al., 2014; Palmer et al., 2018, 2020; Nauels et al., 2019). The expansion efficiency        In summary, consistent relationships between OHC (Section 9.2.2.1),
coefficient that relates GMTSL and OHC for the two-layer emulator        mean ocean temperature and GMTSL are found using two-layer has a mean and standard deviation of 0.113 +/- 0.013 m YJ-1                emulators, CMIP6 models, and modern and paleo observations (Supplementary Material 9.SM.4.3). This approach yields                  to provide medium confidence in the 0.113 +/- 0.013 m YJ-1, a likely thermosteric contribution between 1995-2014 and                  or 0.617 +/- 0.071 m &deg;C-1 likely ranges of assessed conversion values.
2100 that represents a minimal change from AR5 and SROCC                  It is possible to estimate relationships between SST or GSAT change (Table 9.8). The two-layer emulator GMTSL projected median and            and GMTSL rise, but conversions are not generally applicable and 17th-83rd percentile, or likely, range is 0.12 (0.09 to 0.15) m for      depend on time scale and application.
SSP11.9, 0.14 (0.11 to 0.18) m for SSP12.6, 0.20 (0.16 to 0.24) m for SSP24.5, 0.25 (0.21 to 0.30) m for SSP37.0, and 0.30 (0.24 to 0.36) m for SSP58.5 by 2100 (Section 9.6.3.2 and Tables 9.1, 9.8        Table 9.1 l Projected contributions to median and 17-83% (parentheses) and 9.9). The two-layer model heat content increases slightly faster      and 5-95% [square brackets] ranges of thermosteric sea level from AR5 than that of the total depth CMIP6 ensemble, which is related to          (Church et al., 2013b), CMIP6 (Jevrejeva et al., 2020; Hermans et al., 2021) and the two-layer energy balance model (described in Sections 7.SM.2, its role in the assessed energy balance (Section 7.SM.2), but with 9.SM.4 and Box 4.1) averaged over 2081-2100, with respect to a baseline a similar ensemble spread (Table 9.1). Projecting the likely factor by    of 1995-2014. Note that AR5 and SROCC interpret 5-95% range as the likely which 1995-2014 to 2081-2100 OHC change exceeds change over              range, while in this table square brackets are used for consistency.
1971 to 2018 in CMIP6 yields 3 to 5 for SSP12.6, 4 to 6 for SSP24.5, 5 to 7 for SSP37.0, and 5 to 8 for SSP58.5. The two-layer model                                                RCP2.6/          RCP4.5/          RCP8.5/
Study SSP12.6          SSP24.5          SSP58.5 likely equivalents are 2 to 3 for SSP12.6, 3 to 4 for SSP24.5, 4 to 5 for SSP37.0, and 4 to 6 for SSP58.5.                                    IPCC AR5 and SROCC GMTSL 0.13              0.18              0.26 (Church et al., 2013b;
[0.09 to 0.17] m  [0.13 to 0.22] m  [0.20 to 0.32] m Oppenheimer et al., 2019)
For reconstructions, the expansion efficiency coefficient is required for CMIP6 5-95% GMTSL                      0.14              0.18              0.26 the conversion between ocean temperature and steric sea level over (Hermans et al., 2021)            [0.08 to 0.17] m  [0.11 to 0.23] m  [0.17 to 0.33] m a specific time scale. Combining the assessed sea level and energy CMIP6 5-95% GMTSL                                        0.19              0.27 data over 1995 to 2014 (drawn from the analysis in Cross-Chapter                                                    -
(Jevrejeva et al., 2020)                            [0.13 to 0.24] m  [0.19 to 0.35] m Box 9.1) results in a coefficient of 0.1210 +/- 0.0014 m YJ-1, or 0.6607 +/- 0.0076 m &deg;C-1 in terms of mean ocean temperature.                Assessed GMTSL based 0.13              0.17              0.25 on two-layer model The two-layer emulator assessment used in AR6 results in                  17-83% and 5-95%
(0.11 to 0.16)    (0.14 to 0.21)    (0.20 to 0.30) 0.113 +/- 0.013 m YJ-1, or 0.617 +/- 0.071 m &deg;C-1 (Appendices 7.SM.2,                                            [0.09 to 0.19] m  [0.12 to 0.25] m  [0.18 to 0.35] m (Sections 7.SM.2, 9.SM.4) 9.SM.4). Both of these estimates are in line with an independent Total OHC 17-83% and estimate of 0.70 m/&deg;C (Hieronymus, 2019) and other estimates, for          5-95% from assessed 1.18              1.56              2.23 (0.99 to 1.42)    (1.33 to 1.86)    (1.92 to 2.64) example, 0.116 +/- 0.011 m YJ-1 (Kuhlbrodt and Gregory, 2012), but          two-layer model
[0.86 to 1.65] YJ [1.19 to 2.12] YJ [1.71 to 3.00] YJ are significantly larger than the temperature to sea level conversion      (Sections 7.SM.2, 9.SM.4) used in AR5 (0.42 m &deg;C-1 based on SST and the estimated range from        0-2000 m OHC 17-83%                    1.06              1.35              1.89 Levermann et al., 2013). The expansion coefficient is not fixed across    and 5-95% from CMIP6              (0.80 to 1.31)    (1.08 to 1.67)    (1.60 to 2.29) models, nor in time, as it varies depending on which water masses          (Figure 9.6)                      [0.66 to 1.64] YJ [0.90 to 1.84] YJ [1.28 to 2.58] YJ 1245
 
Chapter 9                                                                                                                Ocean, Cryosphere and Sea Level Change 9.2.4.2      Ocean Dynamic Sea Level Change                                              Ocean dynamic sea level change is strongly affected by internal variability (Section 9.6.1.4), partly from interannual to decadal coupled Projections of ocean dynamic sea level change (Box 9.1) on                                atmosphere-ocean modes of variability via wind-driven redistribution multi-annual time scales resemble the patterns of steric sea level                        (Annex IV; Griffies et al., 2014; Han et al., 2017) and partly from change in the open ocean (Figures 9.11 and 9.12; Lowe and Gregory,                        intrinsic ocean variability, particularly in higher-resolution simulations 2006; Pardaens et al., 2011; Couldrey et al., 2021). On shorter time                      (such as HighResMIP), which statistically resemble observations, even scales, especially in extratropical coastal areas, there may be an                        on short time scales (Figure 9.12; Griffies et al., 2014; S&#xe9;razin et al.,
important barotropic component (also called bottom pressure change)                      2016; Llovel et al., 2018; Chassignet et al., 2020). High-resolution due mostly to changes in wind-driven circulation and eddies apparent                      simulations are not used in relative sea level projections (Section 9.6.3) in the variance of ocean dynamic sea level (Figure 9.12; Roberts et al.,                  due to the limited range of forcing scenarios. The most marked feature 2016; Hughes et al., 2018). This component is highly sensitive to ocean                  of long-term regional sea level change in the continuous satellite model resolution (Chassignet et al., 2020). Steric sea level change is                    altimetry record, beginning in 1992, is the east-west dipole in the associated with local changes in temperature and salinity, which                          Pacific Ocean (rising more rapidly in the east, see also Section 9.6.1.3),
come about through changes in surface fluxes of heat and freshwater                      which persisted until 2015, and can be explained by anomalously 9 (Section 9.2.1.2) and through redistribution of existing water masses                    strong trade winds (Merrifield et al., 2012; England et al., 2014; Griffies by changed ocean circulation and mixing processes (Figure 9.12                            et al., 2014; Takahashi and Watanabe, 2016; Han et al., 2017) together and Sections 9.2.2.1 and 9.2.3). Redistribution of water masses                          with associated changes in surface heat flux (Piecuch et al., 2019).
often involves anticorrelated thermosteric and halosteric changes                        The most notable features of sub-annual variability in altimetry are (Figure 9.12), especially in the Atlantic (Pardaens et al., 2011; Bouttes                eddies and tides, which are directly simulated only in high-resolution et al., 2014; Durack et al., 2014; Griffies et al., 2014; Han et al., 2017).              models (Haigh et al., 2019; Chassignet et al., 2020).
SSP5-8.5 (17 models)
SSP1-2.6 (17 models)
Figure 9.12 l (a-f) Coupled Model Intercomparison Project Phase 6 (CMIP6) multi-model mean projected change contributions to relative sea level change in (a, d) steric sea level anomaly, (b, e) thermosteric sea level anomaly, and (c, f) halosteric sea level anomaly between 1995-2014 and 2081-2100 using a method that does not require a reference level (Landerer et al., 2007). Global mean change has been removed from these figures, consistent with the methods in Sections 9.6.3 and 9.SM.4 and the definitions of Gregory et al. (2019). (Gregory et al., 2019). See Figure 9.27 for global mean sea level (GMSL). (g-i) Standard deviation of ocean dynamic sea level change from (g) Aviso observations (10-day high-pass filter); (h) five-day mean of high-resolution Ocean Model Intercomparison Project phase 2 (OMIP-2) models forced with observed fluxes; and (i) five-day mean of low-resolution OMIP-2 models which are comparable in resolution to the models in (a-f). No overlay indicates regions with high model agreement, where 80% of models agree on the sign of change. Diagonal lines indicate regions with low model agreement, where <80%
of models agree on the sign of change (see Cross-Chapter Box Atlas.1 for more information). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1246
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 Projections of the pattern and amplitude of regional ocean dynamic      pronounced thermosteric sea level rise along the American coast sea level change in CMIP6 and previous model generations show            around 40&deg;N (Figures 9.12 and 9.26), leading to a relatively large a large model spread, of a similar size to the geographical spread      ocean dynamic sea level rise in this region (Yin, 2012; Bouttes et al.,
(Figure 9.12). The model spread derives from model dependence            2014; Slangen et al., 2014b; Little et al., 2019; Lyu et al., 2020a).
of changes both in surface fluxes (Section 9.2.1.2) and in the ocean response (Section 9.2.2). The spread is similar in CMIP6 and      In summary, ocean dynamic sea level change involves changes to CMIP5, and is largest in regions with large projected variations in      temperature and salinity and responses of currents to changing ensemble-mean ocean dynamic sea level change (Lyu et al., 2020a),        forcing, with significant variability driven by unforced oceanic such as the Southern Ocean Dipole with an ocean dynamic sea level        variability. Projections of dynamic sea level variability require fully rise north of the ACC and a fall to the south, the Atlantic Dipole      three-dimensional ocean models, and only high-resolution ocean with a sea level rise north of 40&deg;N and a fall in 20&deg;N-40&deg;N, the        models are statistically consistent on short time scales with satellite Northwest Pacific Dipole, and the large sea level rise in the Arctic    altimeter observations (very high confidence).
(Church et al., 2013b; Slangen et al., 2014a, 2015; Bilbao et al.,
2015; Gregory et al., 2016; C. Chen et al., 2019; Lyu et al., 2020a; Couldrey et al., 2021). Patterns of change are consistent between        9.3          Sea Ice                                                        9 model simulations and observations (medium confidence). The major model ensemble-mean features resemble thermosteric sea level            9.3.1        Arctic Sea Ice change, as expected from altered input of heat to the ocean without changing circulation, while model spread results from the        9.3.1.1      Arctic Sea Ice Coverage diversity in redistribution of the heat content of the unperturbed ocean (Section 9.2.2.1; Bouttes and Gregory, 2014; Gregory et al.,      The observed decrease of Arctic sea ice area is a key indicator of 2016; Huber and Zanna, 2017; Lyu et al., 2020b; Todd et al., 2020;      large-scale climate change (Section 2.3.2.1.1, Cross-Chapter Box 2.2).
Couldrey et al., 2021).                                                  The SROCC (Meredith et al., 2019) assesses that sea ice extent, which is the total area of all grid cells with at least 15% sea ice concentration, The Southern Ocean Meridional Dipole is driven by a northward            has declined since 1979 in each month of the year (very high advection of excess heat (from changes in surface fluxes) by the        confidence). In contrast to SROCC, we assess changes in sea ice area wind-driven circulation followed by subduction or diffusive uptake      (the actual area of the ocean covered by sea ice) rather than sea ice in mid-latitudes, northward redistribution of existing heat by the      extent, because sea ice area is geophysically more relevant and not strengthening of that circulation, and the meridional contrast in        grid-dependent (Notz, 2014; Ivanova et al., 2016; Notz et al., 2016; thermal expansivity due to its temperature-dependence (Armour et al.,    Notz and SIMIP Community, 2020). Arctic sea ice area is calculated 2016; Gregory et al., 2016; Lyu et al., 2020b; Todd et al., 2020;        based on measurements by passive microwave satellite sensors that Couldrey et al., 2021).                                                  provide near-continuous measurements of gridded, pan-Arctic sea ice concentration from 1979 onwards. Irreducible uncertainties in the The positive Arctic ocean dynamic sea level change is driven by          conversion of thermal microwave brightness temperature to sea ice increased freshwater input (Couldrey et al., 2021). The Northwest        concentration, and choices in algorithm design, cause uncertainties in Pacific Dipole is driven by the intensification of the Kuroshio Current  observed Arctic sea ice area, which are far smaller than the observed in response to reduced heat loss and in some models to wind stress      sea ice loss (e.g., Comiso et al., 2017a; Niederdrenk and Notz, 2018; change (C. Chen et al., 2019; Couldrey et al., 2021).                    Alekseeva et al., 2019; Kern et al., 2019; Meier and Stewart, 2019).
Sea ice area has decreased in every month of the year from 1979 to The North Atlantic sea level change dipole is forced by a reduction      the present (very high confidence) (Figure 9.13). The absolute and in heat loss from the ocean north of 40&deg;N (i.e., net heat uptake),      the relative ice losses are highest in late summer-early autumn (high which in all Earth system models leads to a weakening of the AMOC,      confidence) (Figure 9.13). Averaged over the decade 2010-2019, the although the magnitude has a large model spread (Section 9.2.3.1;        monthly Arctic sea ice area from August to October has been around Gregory et al., 2016; Huber and Zanna, 2017). The reduced northward      2 million km&#xb2; (or about 25%) smaller than during 1979-1988 (high transport of warm, salty water (Section 9.2.2) causes further ocean      confidence) (Figure 9.13).
dynamic sea level change, whose details are model-dependent. North of 40&deg;N, this redistribution leads to a sea level rise, predominantly    The SROCC discussed the regional distribution of Arctic sea ice loss, halosteric, reinforcing the thermosteric effect of heat uptake (Couldrey and the findings remain valid for the updated time series covering et al., 2021). Comparison of observed Atlantic OHC for 1955-2017        2019 (Figure 9.13). Sea ice loss in winter is strongest in the Barents with a reconstruction assuming no change in circulation indicates        Sea, while summer losses occur primarily at the summer sea ice that the thermosteric sea level change resulting from southward          region margins, in particular in the East Siberian, Chukchi, Kara and redistribution of heat may be detectable (Zanna et al., 2019). This      Beaufort Seas (Frey et al., 2015; Chen et al., 2016; Onarheim et al.,
redistribution causes a tendency for SST cooling north of 40&deg;N and      2018; Peng and Meier, 2018; Maksym, 2019). In the Bering Sea, anomalous heat input from the atmosphere, and thus a positive            expanding winter sea ice cover was observed until 2017 (Frey et al.,
feedback on AMOC weakening (Winton et al., 2013; Gregory et al.,        2015; Onarheim et al., 2018; Peng and Meier, 2018), but a marked 2016; Todd et al., 2020; Couldrey et al., 2021). Many climate and        reduction in sea ice concentration has occurred since then (high ocean models agree that the AMOC weakening is associated with            confidence) (Stabeno and Bell, 2019).
1247
 
Chapter 9                                                                                                              Ocean, Cryosphere and Sea Level Change Arctic sea-ice historical records and CMIP6 projections Anomaly time series, maps of seasonal sea-ice concentration and changes, and projected sea-ice metrics in SSP2-4.5 9
Figure 9.13 l Arctic sea ice historical records and Coupled Model Intercomparison Project Phase 6 (CMIP6) projections. (Left) Absolute anomaly of monthly-mean Arctic sea ice area during the period 1979 to 2019 relative to the average monthly-mean Arctic sea ice area during the period 1979 to 2008. (Right) Sea ice concentration in the Arctic for March and September, which usually are the months of maximum and minimum sea ice area, respectively. First column: Satellite-retrieved mean sea ice concentration during the decade 1979-1988. Second column: Satellite-retrieved mean sea ice concentration during the decade 2010-2019. Third column:
Absolute change in sea ice concentration between these two decades, with grid lines indicating non-significant differences. Fourth column: Number of available CMIP6 models that simulate a mean sea ice concentration above 15 % for the decade 2045-2054. The average observational record of sea ice area is derived from the UHH sea ice area product (Doerr et al., 2021), based on the average sea ice concentration of OSISAF/CCI (OSI-450 for 1979-2015, OSI-430b for 2016-2019) (Lavergne et al., 2019), NASA Team (version 1, 1979-2019) (Cavalieri et al., 1996) and Bootstrap (version 3, 1979-2019) (Comiso, 2017) that is also used for the figure panels showing observed sea ice concentration. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
With respect to seasonal changes in the sea ice cover, the winter sea                    The observed fluctuations and trends of the Arctic sea ice cover ice loss causes a decrease in the average sea ice age and fraction of                    arise from a combination of changes in natural external forcing and multi-year ice, as assessed by SROCC (very high confidence), and also                    anthropogenic forcing, internal variability and internal feedbacks of the ocean area covered intermittently by sea ice (Bliss et al., 2019).                (e.g., Notz and Stroeve, 2018; Halloran et al., 2020). New paleo-proxy In contrast, the seasonal ice zone (covered by sea ice in winter but not                  techniques indicate regional sea ice changes over epochs and in summer) has expanded regionally (Bliss et al., 2019) and over the                      millennia and allow possible drivers to be assessed. Biomarker IP25 whole Arctic (Steele and Ermold, 2015), because the loss of summer                        (Belt et al., 2007) together with other sedimentary biomarkers sea ice area is larger than the loss of winter sea ice area. Arctic sea                  (Belt, 2018) provide local temporal information on seasonal sea ice retreat includes an earlier onset of surface melt in spring and                      ice coverage, permanent sea ice coverage and ice-free waters, with a later freeze up in autumn, lengthening the open water season in                        occasional ambiguous contrasting results (Belt, 2019). These records the seasonal sea ice zone (Stroeve and Notz, 2018). However, there is                    and other proposed paleo proxies, including bromine in ice cores low agreement in quantification of regional trends of melt and freeze                    (Spolaor et al., 2016), dinocyst assemblages (e.g., De Vernal et al.,
onset between different observational products (Bliss et al., 2017;                      2013b) and driftwood (e.g., Funder et al., 2011), provide evidence of Smith and Jahn, 2019).                                                                    sea ice fluctuations that exceed internal variability (high confidence).
Reconstructions of Arctic sea ice coverage put the satellite period                      The inferred sea ice fluctuations over millennia can be related changes into centennial context. Direct observational data coverage                      to Northern Hemisphere temperature evolution and give rise to (Walsh et al., 2017) and model reconstructions (Brennan et al., 2020)                    Arctic-wide fluctuations in sea ice coverage in the paleorecord warrant high confidence that the low Arctic sea ice area of summer                        (Section 2.3.2.1.1). On a regional scale, fluctuations include decreased 2012 is unprecedented since 1850, and that the summer sea ice loss is                    sea ice cover during the Aller&#xf8;d warm period (14.7-12.9 ka) in the significant in all Arctic regions except for the Central Arctic (Cai et al.,              Laptev (Hrner et al., 2016) and Bering Seas (M&#xe9;heust et al., 2018);
2021). Direct winter observational data coverage before 1953 is too                      an extensive sea ice cover during the Younger Dryas (around 12 ka) sparse to reliably assess Arctic sea ice area. Since 1953, the years                      in the Bering (M&#xe9;heust et al., 2018), Kara (Hrner et al., 2018),
2015 to 2018 had the four lowest values of maximum Arctic sea ice                        Laptev (Hrner et al., 2016) and Barents (Belt et al., 2015) Seas, and area, which usually occurs in March (high confidence) (Figure 2.20).                      at the Yermak Plateau (Kremer et al., 2018); little sea ice during the Reconstructions of Arctic sea ice area before 1850 remain sparse,                        early Holocene, when Northern Hemisphere summer insolation was and as in SROCC, there remains medium confidence that the current                        higher than today (8000 to 9000 years before present), in the North sea ice levels in late summer are unique during the past 1 kyr                            Icelandic Shelf area (Cabedo-Sanz et al., 2016; Xiao et al., 2017), Sea (Section 2.3.2.1.1; Kinnard et al., 2011; De Vernal et al., 2013b).                      of Okhotsk (Lo et al., 2018), Canadian Arctic (Spolaor et al., 2016),
1248
 
Ocean, Cryosphere and Sea Level Change                                                                                                Chapter 9 Barents (Berben et al., 2017), Bering (M&#xe9;heust et al., 2018), and        reconstructions (SIA,Sep = 0.2 million km2) (medium confidence Chukchi (Stein et al., 2017) Seas, at the Yermak Plateau (Kremer        because of limited reliability of longer-term sea ice reconstructions) et al., 2018) and north of Greenland (Funder et al., 2011); increasing  (Brennan et al., 2020). Internal variability has been estimated to sea ice cover throughout much of the middle and late Holocene            have contributed 30 to 50% of the observed Arctic summer sea ice around Svalbard (Knies et al., 2017), in the North Icelandic Shelf      loss since 1979 (Kay et al., 2011; Stroeve et al., 2012; Ding et al.,
area (Cabedo-Sanz et al., 2016; Harning et al., 2019; Halloran et al.,  2017, 2019; England et al., 2019). However, this estimate from 2020), north of Greenland (Funder et al., 2011), and in the Western      models might be biased towards internal over forced variability Greenland (Kolling et al., 2018), Barents (Belt et al., 2015; Berben    because of the models high internal variability and because the et al., 2017), Chukchi (De Vernal et al., 2013a; Stein et al., 2017) and CMIP5 simulated September sea ice sensitivity to forcing is lower Laptev (Hrner et al., 2016) Seas. The consistent, Arctic-wide changes  than observed, even if internal variability is taken into account (Notz give high confidence in millennial-scale co-variability of the sea ice  and Stroeve, 2016; Rosenblum and Eisenman, 2017). Most CMIP6 cover with temperature fluctuation.                                      models fail to simulate the observed sensitivity of sea ice loss to CO2 emissions (as a proxy for time) and to temperature simultaneously.
The SROCC assessed that approximately half of the satellite-observed    However, they better capture the observed sensitivity of sea ice loss Arctic summer sea ice loss is driven by increased concentrations        to CO2 emissions than CMIP5 models (Section 3.4.1; Figure 9.14h;          9 of atmospheric greenhouse gases (medium confidence). Recent              Notz and SIMIP Community, 2020).
attribution studies now allow the strengthened assessment that it is very likely that more than half of the observed Arctic sea ice loss  The SROCC examined the different atmospheric and oceanic processes in summer is anthropogenic (Section 3.4.1.1). This assessment is        that caused the observed sea ice loss, with recent studies providing confirmed by process-based analyses of Arctic sea ice loss not assessed  new evidence for the importance of variations in air temperature by SROCC. Similar to the paleorecord, the satellite record of Arctic    (Olonscheck et al., 2019; Dahlke et al., 2020), wind patterns (Graham sea ice area from 1979 onwards is strongly and linearly correlated      et al., 2019), oceanic heat flux (Docquier et al., 2021) and riverine with global mean temperature on decadal and longer time scales          heat influx (Park et al., 2020). As in SROCC, the relative contribution of (Figures 9.14a,e) (e.g., Gregory et al., 2002; Rosenblum and Eisenman,  each physical cause to the sea ice loss cannot be robustly quantified 2017). The correlation holds across all months with R2 ranging from      because of disagreement among models (Burgard and Notz, 2017),
0.61 to 0.81 (Niederdrenk and Notz, 2018). However, in contrast          sparse observations, and limited understanding of the variation of to paleorecords, sea ice fluctuations during the satellite period are    each factor with global mean temperature. This is addressed by new only weakly correlated with Northern Hemisphere insolation (Notz        diagnostics available from CMIP6 simulations, which now allow for and Marotzke, 2012); modern Northern Hemisphere sea ice area            more detailed analyses of the drivers of sea ice loss at a process level is more strongly correlated with atmospheric carbon dioxide (CO2)        (Keen et al., 2021).
concentration (Johannessen, 2008; Notz and Marotzke, 2012) and cumulative anthropogenic CO2 emissions (Figures 9.14b,f; Zickfeld        In examining temperature thresholds for the loss of Arctic summer et al., 2012; Herrington and Zickfeld, 2014; Notz and Stroeve, 2016). sea ice, the Special Report on Global Warming of 1.5&deg;C (SR1.5; The R2 values of the correlation between sea ice area and cumulative    Hoegh-Guldberg et al., 2018) and SROCC assess that a reduction CO2 emissions range across all months from 0.76 to 0.92 (Stroeve        of September mean sea ice area to below 1 million km2, practically and Notz, 2018). In summary, there is high confidence that satellite-    a sea ice-free Arctic Ocean, is more probable for a global mean observed Arctic sea ice area is strongly correlated with global mean    warming of 2&deg;C compared to global mean warming of 1.5&deg;C temperature, CO2 concentration and cumulative anthropogenic              (high confidence). Analyses of CMIP6 simulations (Notz and SIMIP CO2 emissions.                                                          Community, 2020) confirm this result, as they show that, on decadal and longer time scales, Arctic summer sea ice area will remain highly In addition to changes in the external forcing, internal variability    correlated with global mean temperature until the summer sea substantially affects Arctic sea ice, evidenced from both paleorecords  ice has vanished (Figure 9.14a,e). Quantitatively, existing studies (e.g., Chan et al., 2017; Hrner et al., 2017; Kolling et al., 2018) and (Screen and Williamson, 2017; Jahn, 2018; Ridley and Blockley, 2018; satellites after 1979 (e.g., Notz and Stroeve, 2018; Roberts et al.,    Sigmond et al., 2018; Notz and SIMIP Community, 2020) also show 2020) (high confidence). Most of the internal variability on annual      that, for a warming between 1.5 and 2&deg;C, the Arctic will only be time scales is related to atmospheric temperature fluctuations,          practically sea ice free in September in some years, while at 3&deg;C for example linked to cyclone activities (Wernli and Papritz, 2018;      warming, the Arctic is practically sea ice free in September in most Olonscheck et al., 2019), while multi-decadal internal variability      years, with longer practically sea ice-free periods at higher warming is primarily related to changes in oceanic heat transport (Zhang,        levels (medium confidence). However, because of the CMIP5 and 2015; Halloran et al., 2020). These mechanisms are represented          CMIP6 models generally too low sensitivity of sea ice loss to global in current climate models (Olonscheck et al., 2019; Halloran et al.,    warming, there is only low confidence regarding the specific warming 2020), but the resulting internal variability of September sea          level at which the Arctic Ocean first becomes practically sea ice free ice area in CMIP5 and CMIP6 models, as given by the ensemble            (Section 4.3.2.1; Notz and SIMIP Community, 2020).
mean standard deviation SIA,Sep = 0.5 million km&#xb2; (Olonscheck and Notz, 2017; Notz and SIMIP Community, 2020), exceeds the            In contrast, CMIP6 models capture the observed sensitivity of estimated internal variability for the period 1850 to 1979 from          Arctic sea ice area to cumulative anthropogenic CO2 emissions well, both reanalyses (SIA,Sep = 0.3 million km2) and direct observational    providing high confidence that the Arctic Ocean will likely become 1249
 
Chapter 9                                                                                                              Ocean, Cryosphere and Sea Level Change 9
Figure 9.14 l Monthly mean March (a-d) and September (e-h) sea ice area as a function of global surface air temperature (GSAT) anomaly (a, e);
cumulative anthropogenic CO2 emissions (b, f); year (c, g) in Coupled Model Intercomparison Project Phase 6 (CMIP6) model simulations (shading, ensemble mean as bold line) and in observations (black dots). Panels (d) and (h) show the sensitivity of sea ice loss to anthropogenic CO2 emissions as a function of the modelled sensitivity of GSAT to anthropogenic CO2 emissions. In panels (d) and (h), the black dot denotes the observed sensitivity, while the shading around it denotes internal variability as inferred from CMIP6 simulations (after Notz and SIMIP Community, 2020). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
practically sea ice free in the September mean for the first time for                    There is an indication that CMIP6 simulations of Arctic sea ice future CO2 emissions of less than 1000 Gt and before the year 2050                      have improved relative to CMIP5 (Section 3.4.1.1), but detailed in all SSP scenarios (Notz and SIMIP Community, 2020). This new                          evaluation studies exist mainly for CMIP5 models. These studies assessment is consistent with an observation-based projection of                        found that CMIP5 model projections and reanalyses show a large a practically sea ice-free Arctic Ocean in September for additional                      spread of simulated regional Arctic sea ice concentration (Lalibert&#xe9; anthropogenic CO2 emissions of 800 +/- 330 GtCO2 beyond the                                et al., 2016; Chevallier et al., 2017), which remains true for CMIP6 year 2018 (Notz and Stroeve, 2018; Stroeve and Notz, 2018). This                        models (Shu et al., 2020; Wei et al., 2020). In addition, both CMIP5 estimate may, however, be too high due to neglecting possible                            and CMIP6 models show a large spread in the simulated seasonal future reduction in atmospheric aerosol load that would cause                            cycle of Arctic sea ice area, with too high a sea ice area in March in additional warming (Gagn&#xe9; et al., 2015a; Wang et al., 2018), and                        the ensemble mean (Notz and SIMIP Community, 2020). The CMIP5 is subject to the same constraints as the carbon budget analysis                        models have also had difficulty simulating realistic landfast sea ice for global mean temperature (see section 5.5 for details). Based on                      (Lalibert&#xe9; et al., 2018). These findings imply that both CMIP5 and CMIP6 simulations, it is very likely that the Arctic Ocean will remain                  CMIP6 models do not realistically capture the regional and seasonal sea ice covered in winter in all scenarios throughout this century                      processes governing observed Arctic sea ice evolution, causing low (Sections 4.3.2 and 4.4.2).                                                              confidence in the models projections of future regional sea ice evolution, including updated projections for shipping routes across the Northern Sea Route and Northwest Passage (Wei et al., 2020).
1250
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 The CMIP5 models also have issues with capturing the seasonal            For the more recent past, ice thickness can be directly estimated from cycle of observed changes in Arctic sea ice drift speed, which affects    satellite retrievals of sea ice freeboard (Kwok and Cunningham, 2015; their simulation of regional sea ice concentration patterns. Direct      Kwok, 2018). Based on these retrievals, there is medium confidence measurements of Arctic sea ice from drift buoys and satellites show      that Arctic sea ice volume has decreased since 2003. There is low that drift speed of Arctic sea ice has increased over the satellite      confidence in the amount of decrease over this period and over period in all seasons (e.g., Rampal et al., 2009; Docquier et al., 2017). the CryoSat-2 period from 2011 onwards, primarily because of In summer, CMIP5 models show a slowdown of Arctic sea ice drift          snow-induced uncertainties in the retrieval algorithms, the shortness rather than the observed acceleration (Tandon et al., 2018). In winter,  of the record, and the small identified trend (e.g., Bunzel et al., 2018; CMIP5 models generally capture the observed acceleration of Arctic        Petty et al., 2018, 2020).
drift speed. The drift acceleration is primarily caused by the decrease in concentration and thickness in the observational record (Rampal        Observations of regional changes in sea ice thickness vary in quality.
et al., 2009; Spreen et al., 2011; Olason and Notz, 2014; Docquier        Analysis of submarine data in the central Arctic Ocean suggests that et al., 2017) and, for winter, in CMIP5 models (Tandon et al., 2018). its sea ice has thinned by about 75 cm compared to the mid-1970s Changes in wind speed are less important for the observed large-scale    (Section 2.3.2.1.1). For smaller regions, data are too sparse to allow changes (Spreen et al., 2011; Vihma et al., 2012; Olason and Notz,        for quantitative estimates of long-term trends (King et al., 2017;          9 2014; Docquier et al., 2017; Tandon et al., 2018). In summary, there      Rsel et al., 2018), but a clear thinning signal over 10 to 20 years has is high confidence that Arctic sea ice drift has accelerated because of  been found for sea ice in the Fram Strait (Spreen et al., 2020), north of the decrease in sea ice concentration and thickness.                      Canada (Haas et al., 2017) and for landfast ice in the Kongsfjorden/
Svalbard Arctic border (Pavlova et al., 2019). The CMIP5 models and The SR1.5 assessed with high confidence that there is no hysteresis      reanalyses fail to capture the observed distribution (Stroeve et al.,
in the loss of Arctic summer sea ice. In addition, there is no tipping    2014; Shu et al., 2015) and evolution (Chevallier et al., 2017) of Arctic point or critical threshold in global mean temperature beyond which      sea ice thickness. Most CMIP6 models do not capture the observed the loss of summer sea ice becomes self-accelerating and irreversible    spatial distribution of sea ice thickness realistically (Wei et al., 2020).
(high confidence). This is because stabilizing feedbacks during winter    This leads to low confidence in estimates of thickness from reanalyses related to increased heat loss through thin ice and thin snow, and        and from CMIP5 and CMIP6 models and in their projections of sea increased emission of longwave radiation from open water, dominate        ice volume.
over the amplifying ice albedo feedback (see Section 7.4.2 for details on the individual feedbacks; e.g., Eisenman, 2012; Wagner and Eisenman, 2015; Notz and Stroeve, 2018). Observed and modelled            9.3.2        Antarctic Sea Ice Arctic summer sea ice and global mean temperature are linked with little temporal delay, and the summer sea ice loss is reversible    9.3.2.1      Antarctic Sea Ice Coverage on decadal time scales (Armour et al., 2011; Ridley et al., 2012; Li et al., 2013; Jahn, 2018). The loss of winter sea ice is reversible    The SROCC (Meredith et al., 2019) assessed that there was no as well, but the loss of winter sea ice area per degree of warming        significant trend in annual mean Antarctic sea ice area over the period in CMIP5 and CMIP6 projections increases as the ice retreats from        of reliable satellite retrievals starting in 1979 (high confidence). The the continental shore lines, because these limit the possible areal      updated time series is consistent with this assessment. It includes fluctuations (high confidence) (Section 4.3.2.1; Bathiany et al., 2016,  a maximum sea ice area in 2014, then a substantial decline until 2020; Meccia et al., 2020).                                              the minimum sea ice area in 2017, and an increase in sea ice area since 2017 (Figures 2.20 and 9.15; Schlosser et al., 2018; Maksym, 9.3.1.2      Arctic Sea Ice Volume and Thickness                          2019; Parkinson, 2019). As assessed in Section 2.3.2.1.2, the possible significance of the increase in mean Antarctic sea ice area over The SROCC assessed with very high confidence that Arctic sea ice          the shorter period 1979 to 2014 (Figure 2.20; Simmonds, 2015; has become thinner over the satellite period from 1979 onwards,          Comiso et al., 2017b) is unclear. This is because of observational and this assessment is confirmed for the updated time series              uncertainty (see Section 9.3.1.1), large year-to-year fluctuations in (Section 2.3.2.1.1). Sea ice area has also decreased substantially        all months (Figure 9.15), and limited understanding of the processes over this period (Section 9.3.1.1), leading to the assessment that        and reliability of year-to-year correlation of Antarctic sea ice area Arctic sea ice volume has also decreased with very high confidence        (Yuan et al., 2017).
over the satellite period since 1979. There is, however, only low confidence in quantitative estimates of the sea ice volume                As assessed by SROCC, the evolution of mean Antarctic sea ice area loss over this period because of a lack of reliable, long-term,          is the result of opposing regional trends (high confidence), with pan-Arctic observations and substantial spread in available              slightly decreasing sea ice cover during the period 1979 to 2019 in reanalyses (Chevallier et al., 2017). Current best estimates from        the Amundsen and Bellingshausen Seas, particularly during summer, reanalyses suggest a reduction of September Arctic sea ice volume        and slightly increasing sea ice cover in the eastern parts of the of 55 to 65% over the period 1979-2010, and of about 72% over            Weddell and Ross Seas (Figure 9.15). With the exception of the Ross the period 1979-2016, with the latter deemed a conservative              Sea, these trends are not significant, considering the large variability estimate (Schweiger et al., 2019).                                        of the time series (Yuan et al., 2017).
1251
 
Chapter 9                                                                                                                Ocean, Cryosphere and Sea Level Change The SROCC assessed that the regional trends are closely related                            1979 results from a long-term Southern Ocean surface cooling trend to meridional wind trends (high confidence). This is the case as the                      (e.g., Kusahara et al., 2019; Jeong et al., 2020) but the importance of regional trends in the maximum northward extent of the ice cover                          this mechanism for the observed sea ice evolution is unclear owing (Figure 9.15) are determined by the balance between the northward                          to intricate feedbacks between sea ice change and surface cooling advection of the ice that is formed in polynyas near the continental                      (Haumann et al., 2020). The importance of changing wave activity margin, and the lateral and subsurface melting through oceanic heat                        (Section 9.6.4.2; Kohout et al., 2014; Bennetts et al., 2017; Roach et al.,
fluxes. The advection of the sea ice is strongly correlated with winds                    2018b) on sea ice is unclear due to limited process understanding. In and cyclones (Schemm, 2018; Vichi et al., 2019; Alberello et al., 2020).                  summary, there is high confidence that regional Antarctic trends are Accordingly, the increasing sea ice area in the Ross Sea can be linked                    primarily caused by changes in sea ice drift and decay, with medium to a strengthening of the Amundsen Sea low (e.g., Holland et al.,                          confidence in a dominating role of changing wind pattern. The precise 2017b, 2018), while other regional sea ice trends in the austral autumn                    relative contribution of individual drivers remains uncertain because can be linked to changes in westerly winds, cyclone activity and the                      of limited observations, disagreement between models, unresolved Southern Annular Mode (SAM) in summer and spring (Doddridge and                            processes, and temporal and spatial remote linkages caused by sea ice Marshall, 2017; Holland et al., 2017a; Schemm, 2018). In addition to                      drift (Section 9.2.3.2; Pope et al., 2017).
9 the wind-driven changes, increased near-surface ocean stratification (Section 9.2.1.3) has contributed to the observed increase in sea ice                      Recent research has confirmed SROCC assessment of atmospheric and coverage (e.g., Purich et al., 2018; L. Zhang et al., 2019) as it tends                    oceanic drivers of the sea ice decline from 2014 to 2017, which can be to cool the surface ocean (Sections 9.2.1.1 and 9.2.3.2). The changes                      linked to changes in both subsurface ocean heat flux (Meehl et al., 2019; in stratification result partly from surface freshening (De Lavergne                      Purich and England, 2019) and atmospheric circulation, with the latter et al., 2014), associated with increased northward sea ice advection                      partly related to teleconnections with the tropics (Meehl et al., 2019; (Haumann et al., 2020) and/or melting of the Antarctic ice sheet (medium                  Purich and England, 2019; G. Wang et al., 2019). In the Weddell Sea, confidence) (e.g., Haumann et al., 2020; Jeong et al., 2020; Mackie                        these changes caused in 2017 the re-emergence of the largest polynya et al., 2020), and amplified by local ice-ocean feedbacks (Goosse and                      over the Maud Rise since the 1970s (Section 9.2.3.2; Campbell et al.,
Zunz, 2014; Lecomte et al., 2017; Goosse et al., 2018). In the Amundsen                    2019; Jena et al., 2019; Turner et al., 2020).
Sea, strong ice shelf melting can cause local sea ice melt next to the ice shelf front by entraining warm circumpolar deep water to the ice                      The AR5 (Collins et al., 2013) and SROCC found low confidence in future shelf cavity and surface ocean (medium confidence) (Sections 9.2.3.2                      projections of Antarctic sea ice. This includes the projected mitigation and 9.4.2.2; Jourdain et al., 2017; Merino et al., 2018). It has also been                of the sea ice loss by stratospheric ozone recovery (Smith et al., 2012) suggested that the observed regional increase in sea ice coverage since                    and by an increased freshwater input from melting of the Antarctic Antarctic sea-ice historical records and CMIP6 projections Anomaly time series, maps of seasonal sea-ice concenration and changes, and projected sea-ice metrics in SSP2-4.5 Figure 9.15 l Antarctic sea ice historical records and Coupled Model Intercomparison Project Phase 6 (CMIP6) projections. (Left) Absolute anomaly of observed monthly mean Antarctic sea ice area during the period 1979-2019 relative to the average monthly mean Antarctic sea ice area during the period 1979-2008.
(Right) Sea ice coverage in the Antarctic as given by the average of the three most widely used satellite-based estimates for September and February, which usually are the months of maximum and minimum sea ice coverage, respectively. First column: Mean sea ice coverage during the decade 1979-1988. Second column: Mean sea ice coverage during the decade 2010-2019. Third column: Absolute change in sea ice concentration between these two decades, with grid lines indicating non-significant differences. Fourth column: Number of available CMIP6 models that simulate a mean sea ice concentration above 15% for the decade 2045-2054. The average observational record of sea ice area is derived from the UHH sea ice area product (Doerr et al., 2021), based on the average sea ice concentration of OSISAF/CCI (OSI-450 for 1979-2015, OSI-430b for 2016-2019)
(Lavergne et al., 2019), NASA Team (version 1, 1979-2019) (Cavalieri et al., 1996) and Bootstrap (version 3, 1979-2019) (Comiso, 2017) that is also used for the figure panels showing observed sea ice concentration. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1252
 
Ocean, Cryosphere and Sea Level Change                                                                                              Chapter 9 Ice Sheet (Bronselaer et al., 2018). Compared to the interannual        Regionally, proxy data from ice cores consistently indicate that the variability during the satellite record from 1979 onwards, models      increase of sea ice area in the Ross Sea and the decrease of sea ice simulate too much variability in both CMIP5 (Zunz et al., 2013) and    area in the Bellingshausen Sea are part of longer centennial trends CMIP6 (Roach et al., 2020). The seasonal cycle in sea ice coverage is  and exceed internal variability on multi-decadal time scales (medium misrepresented in most CMIP5 (e.g., Holmes et al., 2019) and CMIP6      confidence) (e.g., Thomas et al., 2019; Tesi et al., 2020). These models (Roach et al., 2020), but the multi-model mean seasonal cycle    centennial trends are consistent with simulations from CMIP5 models in CMIP5 and CMIP6 agrees well with observations (Shu et al., 2015;    (Hobbs et al., 2016b; J.M. Jones et al., 2016; Kimura et al., 2017).
Roach et al., 2020). Most CMIP5 models do not realistically simulate the evolution of Antarctic sea ice volume (Shu et al., 2015) and        There is low confidence in the attribution of the observed changes consistently overestimate the amount of low concentration sea ice,      in Antarctic sea ice area (Section 3.4.1.2). Based on the available and underestimate the amount of high concentration sea ice (Roach      evidence, the lack of a negative trend of Antarctic sea ice area, et al., 2018a). In contrast, CMIP6 models simulate a more realistic    despite substantial global warming in recent decades, has been distribution of regional sea ice coverage (Roach et al., 2020). Most    attributed to internal variability in analyses of the observational CMIP5 models poorly represent Antarctic sea ice drift (e.g., Schroeter  record (Meier et al., 2013; Gallaher et al., 2014; Gagn&#xe9; et al.,
et al., 2018; Holmes et al., 2019), affecting simulated historical      2015b), reconstructions from early observations (Fan et al., 2014;      9 trends, with models that simulate a strong sea ice motion showing      Edinburgh and Day, 2016) and proxy data (Hobbs et al., 2016b) in more variability in sea ice coverage than models with weaker sea ice    model simulations (Turner et al., 2013; Zunz et al., 2013; L. Zhang motion (Schroeter et al., 2018). Owing to limited agreement between    et al., 2019). Nonetheless, without accurate simulations of observed model simulations and observations, limited reliable observations on    changes, the possible contribution of anthropogenic forcing to the a process level, and a lack of process understanding of the substantial regional changes in sea ice area remains unclear (Hosking et al.,
spread in CMIP5 and CMIP6 model simulations, there remains low          2013; Turner et al., 2013; Haumann et al., 2014; L. Zhang et al., 2019).
confidence in existing future projections of Antarctic sea ice decrease and lack of decrease.                                                  The attribution of the observed trends in atmospheric and oceanic forcing is also uncertain because of limited observational records and The discrepancy between the modelled and observed evolution            discrepancies between modelled and observed evolution of the sea of Antarctic sea ice has been related by SROCC to deficiencies in      ice cover. More specifically, there is contrasting evidence for a direct modelled stratification, freshening by ice-shelf meltwater, clouds, and role of stratospheric ozone depletion on the observed changes in other wind- and ocean-driven processes. Recent studies highlight the    atmospheric circulation (Haumann et al., 2014; England et al., 2016; possible mis-representation of freshwater fluxes from ice shelves      Landrum et al., 2017). In contrast, there is high confidence that (Jeong et al., 2020), and the possible effect of the low resolution of  multi-decadal variations in the tropical Pacific and in the Atlantic most models (Sidorenko et al., 2019), even though lower-resolution      affect the Amundsen Sea low (Li et al., 2014; Kwok et al., 2016; Meehl models are, in principle, capable of a realistic simulation of the      et al., 2016; Purich et al., 2016; Simpkins et al., 2016), while other seasonal sea ice budgets in the Southern Ocean (Holmes et al., 2019). modes of climate variability (Annex IV) affect, for example, Southern The relative importance of these possible reasons for the models      Ocean cyclone activity (Simpkins et al., 2012; Cerrone et al., 2017; shortcomings remains unclear (see Section 3.4.1.2 for details).        Schemm, 2018).
The analysis and understanding of the long-term evolution of the        9.3.2.2      Antarctic Sea Ice Thickness Antarctic sea ice cover is hindered by the scarcity of observational records before the satellite period, and the scarcity of paleorecords  The SROCC assessed that observations are too sparse to reliably (see Section 2.3.2.1.2 for further details). Such long records are      estimate long-term trends in Antarctic sea ice thickness. This remains particularly relevant given that the Southern Ocean response            true, and only qualitative statements on prevailing thicknesses are to external forcing takes longer than the length of the available      possible. Data from ICESat-1 laser altimetry (Kurtz and Markus, 2012),
direct observational record (Goosse and Renssen, 2001; Armour          from Operation IceBridge (Kwok and Kacimi, 2018), and long-term et al., 2016). There is only limited evidence for large-scale decadal  shipboard observations collected in the Antarctic Sea Ice Processes fluctuations in sea ice coverage caused by large-scale temperature      and Climate (ASPeCt) dataset (Worby et al., 2008) suggest that and wind forcing. Sparse direct pre-satellite observations suggest      sea ice thicker than 1 m prevails in regions of multi-year ice along a decrease in sea ice coverage from the 1950s to the 1970s (Fan        the eastern coast of the Antarctic Peninsula in the Weddell Sea, in the et al., 2014). Paleo-proxy data indicate that, on multi-decadal to      high-latitude embayment of the Weddell Sea, and along the coast multi-centennial time scales, sea ice coverage of the Southern          of the Amundsen Sea, with remaining regions dominated by thinner Ocean follows large-scale temperature trends (e.g., Crosta et al.,      first-year sea ice (high confidence). Regional patterns in ice thickness 2018; Chadwick et al., 2020; Lamping et al., 2020), for example        are affected by areas of high snow deposition and resulting snow-ice linked to fluctuations in the El Nino-Southern Oscillation and          formation (Massom et al., 2001; Maksym and Markus, 2008), and Southern Annular Mode (Crosta et al., 2021), and that during the        deformation, ridging, and rafting that regionally cause formation of Last Glacial Maximum, Antarctic sea ice extended to about the          very thick sea ice (Massom et al., 2006; G. Williams et al., 2015).
polar front latitude in most regions during winter, whereas the        In addition, near ice shelves a sub-ice platelet layer from supercooled extent during summer is less well understood (e.g., Benz et al.,        water can significantly increase sea ice thickness (Hoppmann et al.,
2016; Xiao et al., 2016; Nair et al., 2019).                            2020; Haas et al., 2021). Regarding snow thickness, observations 1253
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change are too sparse in space and time to reliably estimate changes across      Andresen et al., 2017; Khan et al., 2020; Vermassen et al., 2020).
Southern Ocean sea ice (Webster et al., 2018).                            Historical photographs (Khan et al., 2020) show large mass losses of Jakobshavn and Kangerlussuaq Glaciers in West Greenland There is low confidence in the long-term trend of Antarctic sea ice        from 1880 until the 1940s, exceeding their 21st-century mass loss, thickness. Both ASPeCt and ICESat-1 measurements are biased low            whereas the Helheim Glacier in East Greenland remained stable, in regions with thick ice (Kern and Spreen, 2015), compared to results    gained mass in the 1990s, then rapidly lost mass after 2000. Together, from reanalyses (Massonnet et al., 2013; Haumann et al., 2016) and        these three large outlet glaciers, draining about 12% of the ice sheet observations with autonomous vehicles under sea ice (G. Williams          surface area, have lost 22 +/- 3 Gt yr -1 in the period 1880-2012 (Khan et al., 2015). Estimates of sea ice thickness from CryoSat-2 do not        et al., 2020). Overall, these studies provide a variable picture of the substantially reduce uncertainty, primarily because of the unknown        Greenland Ice Sheet mass change in the 20th century. The updated snow thickness and radar scattering above the snow-ice interface          mass loss of Greenland Ice Sheet, including peripheral glaciers for (Bunzel et al., 2018; Kwok and Kacimi, 2018; Kacimi and Kwok,              the period 1901-1990, is 120 [70-170] Gt yr -1 (see Table 9.5 and 2020). Isolated in situ time series show no clear long-term trend          Figures 9.16 and 9.17).
in landfast ice thickness in the Weddell Sea (Arndt et al., 2020).
9 Reanalyses suggest overall increasing sea ice thickness and volume        Post-1992, SROCC stated that it is extremely likely that the rate between 1980 and 2010 (Holland et al., 2014; Zhang, 2014;                  of mass change of Greenland Ice Sheet was more negative during Massonnet et al., 2015), while CMIP5 (Shu et al., 2015; Schroeter          2012-2016 than during 1992-2001, with very high confidence et al., 2018) and CMIP6 models simulate a decrease in Antarctic sea        that summer melting has increased since the 1990s to a level ice volume over the historical period. Because of this discrepancy,        unprecedented over at least the last 350 years. Since SROCC, the and the unclear reliability of the reanalyses (Uotila et al., 2019), there updated synthesis of satellite observations by the Ice Sheet Mass is low confidence in CMIP5 and CMIP6 simulated future Antarctic            Balance Intercomparison Exercise (The IMBIE Team, 2020) and the sea ice thickness.                                                        GRACE Follow-On (GRACE-FO) Mission (Abich et al., 2019; Kornfeld et al., 2019), have confirmed the mass change record, and the record has been extended to 2020 (The IMBIE Team, 2021) as presented 9.4        Ice Sheets                                                    in 2.3.2.4. The Greenland Ice Sheet lost 4890 [4140-5640] Gt of ice between 1992 and 2020, causing sea level to rise by 13.5 [11.4 to 9.4.1      Greenland Ice Sheet                                            15.6] mm (The IMBIE Team, 2021; see also Section 2.3.2.4.1, Figure 9.16 and Table 9.5). The IMBIE Teams (2020) estimates are 9.4.1.1    Recent Observed Changes                                        consistent with other post-AR5 reviews (Figure 9.17, Table 9.SM.1; Bamber et al., 2018a; Cazenave et al., 2018; Mouginot et al., 2019; In this section we present regional mass change time series for the        Slater et al., 2021). Recent GRACE-FO data (Sasgen et al., 2020; Greenland Ice Sheet and assess the different processes that are            Velicogna et al., 2020) show that, after two cold summers in 2017 causing the increase in mass loss. The vast increase in observational      and 2018, with relatively moderate mass change of about -100 products from various platforms (e.g, GRACE, PROMICE, ESA-                Gt yr -1, the 2019 mass change (-532 +/- 58 Gt yr -1) was the largest CCI, NASA MEaSUREs) provide a consistent and clear picture of              annual mass loss in the record. The high agreement across a variety a shrinking Greenland Ice Sheet (Colgan et al., 2019; Mottram et al.,      of methods confirms SROCC and Chapter 2 assessments. The mass-2019; Mouginot et al., 2019; King et al., 2020; Mankoff et al., 2020;      loss rate was, on average, 39 [-3 to 80] Gt yr -1 over the period Moon et al., 2020; Sasgen et al., 2020; Velicogna et al., 2020; The        1992-1999, 175 [131 to 220] Gt yr -1 over the period 2000-2009 IMBIE Team, 2020). Section 2.3.2.4.1 provides an updated estimate          and 243 [197 to 290] Gt yr -1 over the period 2010-2019 (see Table 9.
of the total Greenland Ice Sheet mass change in a global context          SM.1).
(Figure 2.24). The estimated ice-sheet extent at different times is shown in Figure 9.17, and the paleo perspective on Greenland Ice          The SROCC assessed with high confidence that surface mass balance Sheet evolution is presented in Section 9.6.2.                            (SMB), rather than discharge, has started to dominate the mass loss of the Greenland Ice Sheet (due to increased surface melting and For the 20th century, SROCC (Meredith et al., 2019) presented              runoff), increasing from 42% of the total mass loss for 2000-2005 one reconstruction for 1900-1983 and estimated mass change for            to 68% for 2009-2012. While these estimates have been confirmed the Greenland Ice Sheet and its peripheral glaciers for the period        since SROCC (Mouginot et al., 2019), the new longer record, as 1901-1990. Since SROCC, a comprehensive new study has extended            well as further comprehensive studies (Khan et al., 2015; Colgan the satellite record back to 1972 (Figure 9.16; Mouginot et al.,          et al., 2019; Mottram et al., 2019; The IMBIE Team, 2020) and 2019). The rate of ice-sheet mass change was positive (i.e., it gained    detailed discharge records (King et al., 2020; Mankoff et al., 2020) mass) in 1972-1980 (47 +/- 21 Gt yr -1) and then negative (i.e., it lost    reveal a more complex picture than the continuous trajectory this mass; -51 +/- 17 Gt yr -1 and -41 +/- 17 Gt yr -1) in 1980-1990 and            statement may have implied. Discharge was relatively constant from 1990-2000, respectively. Other ice discharge time series starting in      1972-1999, varying by around 6% for the whole ice sheet, while SMB 1985 (King et al., 2018, 2020; Mankoff et al., 2019, 2020) agree with      varied by a factor of over two interannually, leading to either mass Mouginot et al. (2019) (see also Figure 9.16). There is limited evidence  gain or loss in a given year (Figure 9.16). During 2000-2005, the rate of temporally and spatially heterogeneous Greenland outlet glacier        of discharge increased by 18%, then remained fairly constant again evolution during the 20th century (Lea et al., 2014; L&#xfc;thi et al., 2016;  (increasing by 6% from 2006-2018). After 2000, SMB decreased 1254
 
Ocean, Cryosphere and Sea Level Change                                                                                                                            Chapter 9 9
Figure 9.16 l Mass changes and mass change rates for Greenland and Antarctic ice sheet regions. (a) Time series of mass changes in Greenland for each of the major drainage basins shown in the inset figure (Bamber et al., 2018b; Mouginot et al., 2019; The IMBIE Team, 2021) for the periods 1972-2016, 1992-2018, and 1992-2020.
(b) Time series of mass changes for three portions of Antarctica (Bamber et al., 2018b; The IMBIE Team, 2021) for the period 1992-2016 and 1992-2020. Estimates of mass change rates of surface mass balance, discharge and mass balance in (g) all of Greenland and (c-f, h-j) in seven Greenland regions (Bamber et al., 2018b; Mankoff et al.,
2019; Mouginot et al., 2019; King et al., 2020). Estimates of mass change rates of surface mass balance, discharge and mass balance for (k) all of Antarctica and (l-n) for three regions of Antarctica (Bamber et al., 2018b; The IMBIE Team, 2018; Rignot et al., 2019). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1255
 
Chapter 9                                                                                                                      Ocean, Cryosphere and Sea Level Change more rapidly than discharge increased. In summary, the consistent                                  show that SMB has been gradually decreasing in all regions, while temporal pattern in these longer datasets leads to high confidence                                the increase in discharge in the south-east, central east, north-west that the Greenland Ice Sheet mass losses are increasingly dominated                                and central west has been linked to retreating tidewater glaciers by SMB, but there is high confidence that mass loss varies strongly,                              (Figure 9.16). In summary, the detailed regional records show an due to large interannual variability in SMB.                                                      increase in mass loss in all regions after the 1980s, caused by both increases in discharge and decreases in SMB (high confidence),
On a regional scale, the surface elevation is lowering in all regions, and                        although the timing and patterns vary between regions. The largest widespread terminus and calving front retreats have been observed                                  mass loss occurred in the north-west and the south-east of Greenland (with no glaciers advancing; Mottram et al., 2019; Moon et al., 2020).                            (high confidence).
The largest mass losses have occurred along the west coast and in south-east Greenland (Figure 9.16), concentrated at a few major                                    The SROCC stated with high confidence that variability in large-scale outlet glaciers (Mouginot et al., 2019; Khan et al., 2020). This regional                          atmospheric circulation is an important driver of short-term SMB pattern is consistent with independent Global Navigation Satellite                                changes for the Greenland Ice Sheet. This effect of atmospheric System (GNSS) observations from the Greenland Global Positioning                                  circulation variability on both precipitation and melt rates (and SROCC 9 System (GPS) network which show elastic bedrock uplift of tens of                                  assessment) is confirmed by more recent publications (Vlisuo et al.,
centimetres between 2007-2019 as a result of ongoing ice mass loss                                2018; B. Zhang et al., 2019; Velicogna et al., 2020). The strong mass (Bevis et al., 2019). The regional time series (Figures 9.16; Atlas.30)                            loss in 2019 (Cullather et al., 2020; Hanna et al., 2020; Tedesco Greenland ice sheet cumulative mass change and equivalent sea level contribution 4                                                                                                                        -0.1 Modern & Projected Changes                                                                                                2100 medians, Box Range                                                                                                                66% and 90% ranges 1840-1972 Mouginot Box                              IMBIE                                                      SSP1-2.6                    ISMIP6  Emulator 0                                                                                                                        0 Bamber 4                                                                                                                            m 10 Gt
                                                    -4                                                                                                                        0.1 Emulator median (SSPs)
Emulator 90% range Emulator 66% range                                            SSP5-8.5 ISMIP6 models (RCPs/SSPs)                                                              0.2
                                                    -8 1980            2000                2020            2040              2060              2080              2100 Figure 9.17 l Greenland Ice Sheet cumulative mass change and equivalent sea level contribution. (a) A p-box (Section 9.6.3.2) based estimate of the range of values of paleo Greenland Ice Sheet mass and sea level equivalents relative to present day and the median over all central estimates (Simpson et al., 2009; Argus and Peltier, 2010; Colville et al., 2011; Dolan et al., 2011; Fyke et al., 2011; Robinson et al., 2011; Born and Nisancioglu, 2012; K.G. Miller et al., 2012; Dahl-Jensen et al., 2013; Helsen et al., 2013; Nick et al., 2013; Quiquet et al., 2013; Stone et al., 2013; Colleoni et al., 2014; Lecavalier et al., 2014; Robinson and Goelzer, 2014; Calov et al., 2015, 2018; Dutton et al., 2015; Koenig et al., 2015; Peltier et al., 2015; Stuhne and Peltier, 2015; Vizcaino et al., 2015; Goelzer et al., 2016; Khan et al., 2016; Yau et al., 2016; de Boer et al., 2017; Simms et al., 2019);
(b, left) cumulative mass loss (and sea level equivalent) since 2015 from 1972 (Mouginot et al., 2019) and 1992 (Bamber et al., 2018b; The IMBIE Team, 2020), the estimated mass loss from 1840 (Box and Colgan, 2013; Kjeldsen et al., 2015) indicated with a shaded box, and projections from Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6) to 2100 under RCP8.5/SSP58.5 and RCP2.6/SSP12.6 scenarios (thin lines from Goelzer et al. (2020); Edwards et al. (2021); Payne et al. (2021)) and ISMIP6 emulator under SSP5-8.5 and SSP1-2.6 to 2100 (shades and bold line; Edwards et al., 2021); (b, right) 17th-83rd and 5th-95th percentile ranges for ISMIP6 and ISMIP6 emulator at 2100.
Schematic interpretations of individual reconstructions (Lecavalier et al., 2014; Goelzer et al., 2016; Berends et al., 2019) of the spatial extent of the Greenland Ice Sheet are shown for the: (c) mid-Pliocene Warm Period; (d) the Last Interglacial; and (e) the Last Glacial Maximum: grey shading shows extent of grounded ice. Maps of mean elevation changes (f) 2010-2017 derived from CryoSat 2 radar altimetry (Bamber et al., 2018b) and (g) ISMIP6 model mean (2093-2100) projected changes for the MIROC5 climate model under the RCP8.5 scenario (Goelzer et al., 2020). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1256
 
Ocean, Cryosphere and Sea Level Change                                                                                                    Chapter 9 and Fettweis, 2020) was driven by highly anomalous atmospheric              of western Greenland, reducing meltwater retention capacity.
circulation patterns, both on daily (Cullather et al., 2020) and seasonal  Moreover, meltwater infiltration into firn can be strongly limited time scales (Tedesco and Fettweis, 2020). Although surface melt            by low-permeability ice slabs created by refreezing of infiltrated is anticorrelated with the summer North Atlantic Oscillation Index          meltwater (Machguth et al., 2016). Recent observations and modelling (Vlisuo et al., 2018; Ruan et al., 2019; Sherman et al., 2020), especially efforts indicate that rapidly expanding low-permeability layers have in West Greenland (Bevis et al., 2019), Greenland Ice Sheet melt is        led to an increase in runoff area since 2001 (MacFerrin et al., 2019).
more strongly correlated with the Greenland Blocking Index (Hanna          In summary, there is medium confidence that meltwater storage and et al., 2016, 2018) than with the summer North Atlantic Oscillation        refreezing can temporarily buffer a large-scale melt increase, but index (Huai et al., 2020).                                                  limiting factors have been identified.
The SROCC did not assess the role of cloud changes in detail. Studies      The SROCC reported that there was medium confidence that ocean since AR5 have shown that higher incident shortwave radiation in            temperatures near the grounding zone of tidewater glaciers are conjunction with reduced cloud cover leads to increased melt rates,        critically important to their calving rate, but there was low confidence particularly over the low-albedo ablation zone in the southern              in understanding their response to ocean forcing. The increase in ice part of the Greenland Ice Sheet (Hofer et al., 2017; Niwano et al.,        discharge in the late 1990s and early 2000s (Mouginot et al., 2019;      9 2019; Ruan et al., 2019). Conversely, an increase in cloud cover            King et al., 2020; Mankoff et al., 2020) has been associated with over the high-albedo central parts of the ice sheet, leading to            a period of widespread tidewater glacier retreat (Murray et al., 2015; higher downwelling longwave radiation, was shown to lead either            Wood et al., 2021) and speed up (Moon et al., 2020). Since SROCC, to increased melt (Bennartz et al., 2013) or reduced refreezing of          new studies provide strong evidence for rapid submarine melting meltwater (van Tricht et al., 2016). The elevation dependence of the        at tidewater glaciers (Sutherland et al., 2019; Wagner et al., 2019; cloud radiative effect and its control on surface meltwater generation      Bunce et al., 2020; R.H. Jackson et al., 2020). Changes in submarine and refreezing (W. Wang et al., 2019; Hahn et al., 2020) can induce        melting and subglacial meltwater discharge can trigger increased a spatially consistent response of the integrated Greenland Ice Sheet      ice discharge by reducing the buttressing to ice flow and promoting melt to dominant patterns of cloud and atmospheric variability.            calving (Benn et al., 2017; Todd et al., 2018; Ma and Bassis, 2019; The shortwave and longwave radiation effects on surface melt by            Mercenier et al., 2020); through undercutting (Rignot et al., 2015; clouds have been shown to compensate for each other during strong          D.A. Slater et al., 2017; Wood et al., 2018; Fried et al., 2019) and atmospheric river events, and the increase in melt is caused by            frontal incision (Cowton et al., 2019). Warming ocean waters have increased sensible heat fluxes during such events (Mattingly et al.,        been implicated in the recent thinning and breakup of floating ice 2020). In summary, there is medium confidence that cloud cover              tongues in north-eastern and north-western Greenland (Mouginot changes are an important driver of the increasing melt rates in the        et al., 2015; Wilson et al., 2017; Mayer et al., 2018; Washam et al.,
southern and western part of the Greenland Ice Sheet.                      2018; An et al., 2021; Wood et al., 2021). On decadal time scales, tidewater glacier terminus position correlates with submarine melting The SROCC stated with high confidence that positive albedo feedbacks        (Slater et al., 2019). Over shorter time scales, individual glaciers or contributed substantially to the post-1990s Greenland Ice Sheet melt        clusters of glaciers can behave differently and asynchronously (Bunce increase. Several (mostly positive) feedbacks involving surface albedo      et al., 2018; Vijay et al., 2019; An et al., 2021), and there are not operate on ice sheets (e.g., Fyke et al., 2018). Melt amplification        always clear associations between water temperature and glacier by the observed increase of bare ice exposure through snowline              calving rates (Motyka et al., 2017), retreat or speed-up (Joughin et al.,
migration to higher parts of the ice sheet since 2000 (Shimada et al.,      2020; Solgaard et al., 2020). Variations in ice m&#xe9;lange at the front 2016; Ryan et al., 2019) was five times stronger than the effect of        of a glacier, associated with changes in ocean and air temperature, hydrological and biological processes that lead to reduced bare            have also emerged as a plausible control on calving (Burton et al.,
ice albedo (Ryan et al., 2019). Impurities, in part biologically active    2018; Xie et al., 2019; Joughin et al., 2020). In summary, there is (Ryan et al., 2018), have been observed to lead to albedo reduction        high confidence that warmer ocean waters and increased subglacial (Stibal et al., 2017) and are estimated to have increased runoff from      discharge of surface melt at the margins of marine-terminating bare ice in the southwestern sector of the Greenland Ice Sheet by          glaciers increase submarine melt, which leads to increased ice about 10% (Cook et al., 2020). In summary, new studies confirm that        discharge. There is medium confidence that this contributed to the there is high confidence that the Greenland Ice Sheet melt increase        increased rate of mass loss from Greenland, particularly in the period since about 2000 has been amplified by positive albedo feedbacks,          2000-2010 when increased discharge was observed in the south-east with the expansion of bare ice extent being the dominant factor, and        and north-west.
albedo in the bare ice zone being primarily controlled by distributed biologically active impurities (see also Section 7.3.4.3).                  The SROCC reported that accurate bedrock topography is required for understanding and projecting the glacier response to ocean forcing.
The SROCC reported with medium confidence that around half of              Accurate bathymetry is essential for establishing which water masses the 1960-2014 Greenland Ice Sheet surface meltwater ran off, while          enter glacial fjords, and for reliable estimates of the submarine melt most of the remainder infiltrated firn and snow, where it either            rates experienced by tidewater glaciers (Schaffer et al., 2020; T. Slater refroze or accumulated in firn aquifers. Studies since SROCC show          et al., 2020; Wood et al., 2021). Subglacial and lateral topography a decrease of firn air content between 1998-2008 and 2010-2017              is known to strongly modulate tidewater glacier dynamics and the (Vandecrux et al., 2019) in the low-accumulation percolation area          sensitivity of tidewater glaciers to climatic forcing (Enderlin et al.,
1257
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change 2013; Catania et al., 2018). Bathymetric mapping around the ice            energy balance models (Punge et al., 2012; Cullather et al., 2014; sheet has greatly improved with direct and gravimetric surveys            van Kampenhout et al., 2017, 2020; Alexander et al., 2019) or use (Millan et al., 2018; An et al., 2019a, b; Jakobsson et al., 2020) leading elevation classes to compensate for their coarser resolution (Lipscomb to the improvement of Greenland-wide bathymetric and topographic          et al., 2013; Sellevold et al., 2019; Gregory et al., 2020; Muntjewerf mapping (e.g., Morlighem et al., 2017). However, large uncertainties      et al., 2020a, b). Resulting SMB simulations compare better with in ice thickness remain for around half of the outlet glaciers            regional climate models and observations (Alexander et al., 2019; (Mouginot et al., 2019; Wood et al., 2021) and sea ice covered and        van Kampenhout et al., 2020), but the remaining shortcomings iceberg-packed regions remain poorly sampled near glacier termini          lead to problems reproducing a present-day ice-sheet state close to (Morlighem et al., 2017). There is high confidence that bathymetry        observations. In summary, there is medium confidence in quantitative (governing the water masses that flow into fjord cavities) and            simulations of the present-day state of the Greenland Ice Sheet in ESMs.
fjord geometry and bedrock topography (controlling ice dynamics) modulate the response of individual glaciers to climate forcing.          The SROCC (Meredith et al., 2019) stated that there is low confidence in understanding coastal glacier response to ocean forcing because The AR5 assessed that it is likely that anthropogenic forcing has          submarine melt rates, calving rates, bed and fjord geometry and 9 contributed to the surface melting of Greenland since 1993 (Bindoff        the roles of ice m&#xe9;lange and subglacial discharge are poorly et al., 2013). Section 3.4.3.2 assesses that it is very likely that human  understood. Ice-ocean interactions remain poorly understood and influence has contributed to the observed surface melting of the          difficult to model, with parametrizations often used for calving of Greenland Ice Sheet over the past two decades. There is medium            marine-terminating glaciers (Mercenier et al., 2018) and submarine confidence of an anthropogenic contribution to recent mass loss            and plume-driven melt (Beckmann et al., 2019). Due to the from Greenland.                                                            difficulties of modelling the large number of marine-terminating glaciers and limited availability of high-resolution bedrock data, the 9.4.1.2    Model Evaluation                                              majority of recent modelling work on Greenland outlet glaciers is focused on individual or a limited number of glaciers (Krug et al.,
The SROCC (Oppenheimer et al., 2019) stated that substantial              2014; Bondzio et al., 2016, 2017; Morlighem et al., 2016b; Muresan challenges remained for modelling of the Greenland SMB and the            et al., 2016; Choi et al., 2017; Beckmann et al., 2019), or a specific dynamical ice sheet. Since SROCC, further insights into modelling of      region (Morlighem et al., 2019). Since SROCC, using a flowline model the Greenland ice sheet has come from model intercomparison studies        that includes calving and submarine melting, Beckmann et al. (2019) of the SMB (Fettweis et al., 2020) and dynamical ice sheets (Goelzer      concluded that the AR5 upscaling of contributions from four of the et al., 2020; Payne et al., 2021). Further aspects relevant to the forcing largest glaciers (Nick et al., 2013) overestimated the total glacier of the ice sheet from large scale global climate models and regional      contribution from the Greenland Ice Sheet, due to differences in climate models are discussed in Box 9.3 and Section Atlas.11.2.            response between large and small glaciers. The regional study of Morlighem et al. (2019) confirms that ice-ocean interactions have The SROCC stated that climate model simulations of Greenland SMB          the potential to trigger extensive glacier retreat over decadal time had improved since AR5, giving medium confidence in the ability of        scales, as indicated by observations (Section 9.4.1.1). One focus climate models to simulate changes in Greenland SMB. Since SROCC,          of continental ice-sheet models has been the improved treatment a multi-model intercomparison study (Fettweis et al., 2020) of regional    of marine-terminating glaciers via the inclusion of calving processes and global climate models has shown that the greatest inter-model          and freely moving calving fronts (Aschwanden et al., 2019; Choi spread occurs in the ablation zone, due to deficiencies in an accurate    et al., 2021). An improved bedrock topographic dataset (Morlighem model representation of the ablation zone extent and processes            et al., 2017) allows for ice discharge to be better captured for outlet related to surface melt and runoff, confirming SROCC statement that        glaciers in continental ice-sheet models, and simulations indicate there is large uncertainty in the bare ice model (Ryan et al., 2019).      that bedrock topography controls the magnitude and rate of retreat This intercomparison showed that simple, well-tuned SMB models            (Aschwanden et al., 2019; R&#xfc;ckamp et al., 2020). Overall, although using positive degree day melt schemes can perform as well as more        there is high confidence that the dynamic response of Greenland complex physically based models (Figure Atlas 30). Furthermore, the        outlet glaciers is controlled by bedrock topography, there is low ensemble mean of the models produced the best estimate of the              confidence in quantification of future mass loss from Greenland present-day SMB relative to observations (particularly in the ablation    triggered by warming ocean conditions, due to limitations in the zone). Further assessment of Greenland Ice Sheet regional SMB can          current understanding of ice-ocean interactions, its implementation be found in Section Atlas.11.2.3. Recent progress confirms SROCC          in ice-sheet models, and knowledge of bedrock topography.
assessment that there is medium confidence in the ability of climate models to simulate changes in Greenland SMB.                              The SROCC (Oppenheimer et al., 2019) noted the progress made in Greenland Ice Sheet models since AR5. New since SROCC is The SROCC noted increased use of coupled climate-ice sheet                a focus on improved representation of the present-day state of the models for simulating the Greenland ice sheet, but it also noted          ice sheet (Box 9.3; Goelzer et al., 2018, 2020). Improvements are that remaining deficiencies in coupling between models of climate          closely linked to the growing number and quality of observations and ice sheets (e.g., low spatial resolution) limited the adequate        (Section 9.4.1.1), new techniques to generate internally consistent representation of the feedbacks between them. Some Earth system            input datasets (Morlighem et al., 2014, 2016a), wider use of data models (ESMs) now incorporate multi-layer snow models and full            assimilation techniques (Larour et al., 2014, 2016; Perego et al.,
1258
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 2014; Goldberg et al., 2015; Lee et al., 2015; Schlegel et al., 2015;      0.073 m for RCP8.5. Although the ocean does not directly force the Mosbeux et al., 2016), increased model resolution (Aschwanden et al.,      ice-sheet models in these simulations, the new coupled models allow 2016) and tuning of key processes such as calving (Choi et al., 2021). for interactions between ice-sheet dynamics, SMB and local climate.
A remaining challenge is low confidence in reproducing historical          The coupled projections fall within the lower bounds of AR5 and mass changes of the Greenland Ice Sheet (Box 9.3). However, there          SROCC and, as these studies do not prescribe ocean forcing directly, it is medium confidence in ice-sheet models reproducing the present          is possible that the dynamic response is underestimated.
state of the Greenland Ice Sheet, leading to medium confidence in the current ability to accurately project its future evolution.            Since SROCC, projections of the Greenland Ice Sheet are also available from The Ice Sheet Model Intercomparison Project for 9.4.1.3      Projections to 2100                                          CMIP6 (ISMIP6) (Box 9.3; Annex II; Figure 9.17; Nowicki et al.,
2016, 2020a). ISMIP6 multi-model projections are corrected with an The AR5 and SROCC projected that changes in Greenland SMB will            assessment of the historical dynamical response to pre-2015 climate contribute to sea level in 2100 by 0.03 (0.01 to 0.07) m sea level        forcing (Box 9.3). For the period 2015-2100, the ISMIP6 uncorrected equivalent (SLE) under RCP2.6, and 0.07 (0.03 to 0.16) m SLE under        multi-model ensemble projects sea level contributions ranging from RCP8.5. New since SROCC are the projections of SMB obtained by            0.01 to 0.05 m under RCP2.6, 0.04 to 0.14 m under RCP8.5 (Goelzer        9 an ESM, two regional climate models, and reconstructions based on          et al., 2020), 0.02 to 0.06 m under SSP12.6, and 0.08 to 0.25 m temperature from the CMIP5 and CMIP6 ensembles (Hofer et al., 2020;        under SSP58.5 (Table 9.2; Payne et al., 2021). The higher mass loss Nol et al., 2021). The range of sea level contribution from Greenland    in the SSPs is attributed to a larger decrease in SMB due to the high SMB in Nol et al. (2021) is comparable to the AR5 assessment when        climate sensitivity of the models used (Payne et al., 2021). This finding either CMIP5 or CMIP6 models are used, while Hofer et al. (2020)          is confirmed by Choi et al. (2021), where CMIP6 SSP58.5 SMB leads find a greater mass loss across all CMIP6 emissions scenarios when        to larger ice loss than CMIP5 RCP8.5, while ice discharge is similar.
compared to CMIP5 scenarios. Using SSP58.5 instead of RCP8.5              As the ISMIP6 framework considers a subset of the RCPs/SSPs increases the mean projected sea level from 2005-2100 by up to            and CMIP models, SSP-based projections have been inferred from 0.06 m in the regional climate model simulations of Hofer et al. (2020)    multiple approaches. First, the ISMIP6 CMIP5-forced (Goelzer et al.,
who attribute the difference mainly to a greater Arctic amplification      2020) and CMIP6-forced (Payne et al., 2021) combined ensemble and associated cloud and sea ice feedbacks in the CMIP6 SSP58.5          projections were corrected with the historical trend (Box 9.3) using simulations. In summary, these new projections with fixed ice-sheet        bootstrapping. Second, an emulator of the ISMIP6 projections topography do not provide sufficient evidence to change the AR5 and        (Box 9.3; Edwards et al., 2021) is forced by distributions of global SROCC assessments.                                                        surface air temperature for each SSP from a two-layer energy budget emulator (Supplementary Material 7.SM.2) and then corrected with Reviewing modelling studies since AR5 (Church et al., 2013b), SROCC        the historical trend in the same way. These two approaches result (Oppenheimer et al., 2019) assessed Greenlands contribution to            in projections that are similar in their median values to AR5 and future sea level to be relatively similar to AR5 (Table 9.2). The baseline SROCC projections (Table 9.2), but differ in their range. Similar results for projections has shifted from 1986-2005 in SROCC, to 1995-2014          are obtained when the AR5 parametric fit is applied to the ISMIP6 in this Report. Adjusted to the new 1995-2014 baseline by subtracting      models (Table 9.2, Supplementary Material 9.SM.4.4), which is used 0.01 m, SROCC projected a likely contribution of 0.07 (0.0-0.11) m        to estimate rates of change and post-2100 projections (Sections SLE under RCP2.6, and 0.14 (0.08-0.27) m SLE under RCP8.5 by 2100.        9.4.1.4 and 9.6.3.2).
Since SROCC, new projections for the 21st century have included dynamic ice sheets coupled to ESMs (Muntjewerf et al., 2020a;              The SROCC noted that the study by Aschwanden et al. (2019) projects Van Breedam et al., 2020) or regional atmospheric models (Table 9.2;      a significantly higher Greenland contribution to sea level than the Le clech et al., 2019). The coupled ESM-ice-sheet model CESM2-            assessed likely range in AR5 and SROCC. Under RCP8.5, Aschwanden CISM2 (Community Earth System Model Version 2 and Community Ice            et al. (2019) found that Greenland could contribute up to 0.33 m to Sheet Model 2) projects a sea level rise of 0.109 m in 2100 relative      sea level by 2100 relative to 2000 (the ensemble member that best to 2015 under SSP58.5 (Muntjewerf et al., 2020a) and a similar            reproduces the 2000-2015 mean SMB from a regional climate model contribution under the idealized 1% yr -1 increase in CO2 scenario        projects Greenland mass losses of 0.08 m SLE under RCP2.6 and (Muntjewerf et al., 2020b). The CESM2-CISM2 simulations include            0.18 m SLE under RCP8.5). The SROCC noted that the potentially high ice-sheet-atmosphere interactions and ice-sheet surface meltwater          sea level contribution in this study could be due to the assumption routed to the ocean. The coupled regional atmospheric model and            of spatially uniform warming, which can overestimate surface melt ice-sheet model MAR-GRISLI (Modele Atmospherique Regional and          rates. However, it also reflects the deep uncertainty surrounding Grenoble ice sheet and land ice model) projects a sea level rise of        atmospheric forcing, surface processes, submarine melt, calving 0.079 m in 2100 relative to 2000 under RCP8.5 (Le Clech et al., 2019). and ice dynamics. Goelzer et al. (2020) ascribe 40% of the ISMIP6 An ESM of lower complexity coupled to an ice-sheet model gives            multi-model ensemble spread to ice-sheet model uncertainty, 40% to a sea level contribution of 0.025 to 0.064 m under RCP2.6 and 0.056        climate model uncertainty and 20% to ocean forcing uncertainty.
to 0.12 m under RCP8.5 (the range is due to four simulations with          We note that this finding reflects the current challenges associated different parameter sets for the atmosphere model) (Van Breedam            with the representation of ice-ocean interactions in models, and et al., 2020). Van Breedam et al. (2020) identify a simulation with        the uncertainty in basal conditions (Section 9.4.1.2). However, this a preferred parameter set that projects 0.034 m for RCP2.6 and            finding is consistent with the work of Aschwanden et al. (2019) 1259
 
Chapter 9                                                                                                                  Ocean, Cryosphere and Sea Level Change Table 9.2 l Projected sea level contributions in metres from the Greenland Ice Sheet by 2100 relative to 1995-2014, unless otherwise stated, for selected Representative Concentration Pathway (RCP) and Shared Socio-economic Pathways (SSP) scenarios. Italics denote partial contributions. Historical dynamic response omitted from ISMIP6 simulations is estimated to be 0.19 +/- 0.10 mm yr -1 (0.02 m +/- 0.01 m in 2100 relative to 2015). The climate forcing is described in Appendix 7.SM.2.
Representative Concentration Pathways (RCPs)
Study                            RCP2.6                RCP4.5                RCP8.5                                    Notes IPCC AR5 and SROCC                                      0.07                  0.08                  0.14        Median and likely (66% range) contributions in 2100 relative (Oppenheimer et al., 2019)                          (0.03 to 0.11)        (0.04 to 0.15)        (0.08 to 0.27)  to 1995-2014. Median of multiple studies ISMIP6 CMIP5-forced (Goelzer et al., 2020);                                                                        Range of multi-model contributions in 2100 relative to 2015 0.01 to 0.05              n/a                0.04 to 0.14 excludes historical dynamic response                                                                              from 1 ESM for RCP2.6 and 6 ESMs for RCP8.5 (see caption)
Coupled regional atmosphere-ice sheet n/a                    n/a                  0.079        Contribution in 2100 relative to 2000 from AR-GRISLI model model (Le clech et al., 2019)
Coupled Earth system model (ESM)                                                                                  Contribution in 2100 relative to 2000 from LOVECLIM-AGISM 0.034                                        0.073 of lower complexity-ice-sheet model                                            n/a                                model; preferred parameter set and range from four simulations (0.025 to 0.064)                              (0.056 to 0.12)
(Van Breedam et al., 2020)                                                                                        with different parameters for atmosphereodel 9
Shared Socio-economic Pathways (SSPs)
Study                          SSP12.6              SSP24.5              SSP58.5                                    Notes Coupled ESM-ice sheet model                                                                                        Contribution in 2100 relative to 2015 from coupled n/a                    n/a                  0.109 (Muntjewerf et al., 2020a)                                                                                        CESM2-CISM2 ISMIP6 CMIP6-forced (Payne et al., 2021);                                                                          Range of multi-model contributions in 2100 relative to 2015 0.02 to 0.06              n/a                0.08 to 0.25 excludes historical dynamic response                                                                              from one ESM for SSP12.6 and four ESMs for SSP58.5 ISMIP6 CMIP5 and CMIP6 forced ensemble          0.06 (0.05 to 0.07)                          0.11 (0.09 to 0.14) Median (66% range) [90% range] contribution from ISMIP6 n/a including historical dynamic response              [0.04 to 0.08]                                [0.07 to 0.17]  CMIP5- and CMIP6-forced multi-model ensembles ISMIP6 with AR5 parametric fit: used to 0.08 (0.06 to 0.10)    0.10 (0.08 to 0.13)    0.14 (0.11 to 0.18) Median (66% range) [90% range] contribution from AR5 estimate rates (Supplementary Material
[0.05 to 0.12]        [0.07 to 0.15]        [0.10 to 0.22]  parametric fit to ISMIP6 ensemble, relative to 1995-2014 9.SM.4.4) including historical dynamic response Median (66% range) [90% range] contribution in Emulated ISMIP6; excludes historical dynamic    0.03 (-0.01 to 0.08)    0.06 (0.01 to 0.10)    0.11 (0.06 to 0.16) 2100 relative to 2015 from emulator of ISMIP6 used response (Edwards et al., 2021)                    [-0.04 to 0.12]        [-0.02 to 0.15]        [0.03 to 0.21]
with Chapter 7: Climate Forcing 0.06 (0.01 to 0.10)    0.08 (0.04 to 0.13)    0.13 (0.09 to 0.18)  As above, but relative to 1995-2014 and including This assessment: emulated ISMIP6 total
[-0.02 to 0.15]        [0.01 to 0.18]        [0.05 to 0.23]    historical dynamic response and thus, there is medium confidence that uncertainty in mass loss                            In summary, it is virtually certain that the Greenland Ice Sheet will from the Greenland Ice Sheet is dominated by uncertainty in climate                          continue to lose mass this century under all emissions scenarios, scenario and surface processes, whereas uncertainty in calving and                            and high confidence that total mass loss by 2100 will increase with frontal melt play a minor role.                                                              cumulative emissions. The sea level assessment (Section 9.6.3.3) is based on the emulated ISMIP6 projections, allowing a more consistent The SROCC stated that surface processes, rather than ice                                      approach to a wider range of climate and ocean forcings. The Greenland discharged into the ocean, will dominate Greenland ice loss                                  Ice Sheet is likely to contribute 0.06 (0.01 to 0.10) m under SSP12.6 and over the 21st century, regardless of the emissions scenario (high                            0.13 (0.09 to 0.18) m under SSP58.5 by 2100 relative to 1995-2014.
confidence). This is confirmed by the ISMIP6 projections (Goelzer                            These projections (as well as those of AR5 and SROCC) are lower et al., 2020; Payne et al., 2021). The projected mass loss of Greenland                      than the study of Aschwanden et al. (2019) or the range of possible is predominantly due to increased surface meltwater and loss in                              sea level changes resulting from Structured Expert Judgement (SEJ; refreezing capacity resulting in decreasing SMB (high confidence),                            Section 9.6.3.2; Bamber et al., 2019), contributing to the deep uncertainty concurrent with rising temperatures and darkening of the ice-sheet                            in projected sea level (Box 9.4). There is, however, high confidence that surface (Fettweis et al., 2013; Vizcaino et al., 2015; Le Clech et al.,                      the loss from Greenland will become increasingly dominated by SMB 2019; Muntjewerf et al., 2020a, b; Sellevold and Vizca&#xed;no, 2020).                            and surface melt, as the ocean-forced dynamic response of glaciers will Mass changes due to SMB and outlet glacier dynamics are linked                                diminish as marine margins retreat to higher grounds.
(Goelzer et al., 2013; F&#xfc;rst et al., 2015; R&#xfc;ckamp et al., 2020), as mass loss by one process decreases mass loss by the other - for                              9.4.1.4        Projections Beyond 2100 example, SMB removes ice before it can reach the marine glacier terminus. There is medium confidence that the mass loss through ice                          The AR5 (Church et al., 2013b) assessed the contribution from discharge will decrease in the future (F&#xfc;rst et al., 2015; Aschwanden                        Greenland to sea level projections in 2300 as 0.15 m SLE in et al., 2019; Golledge et al., 2019), because an increase in mass loss                        low-emissions scenarios (about RCP2.6) and 0.31-1.19 m in high (via increased discharge or surface runoff) leads, in most areas, to                          scenarios (approximately RCP6.0/RCP8.5). The SROCC (Oppenheimer a retreat of the glacier margin onto land above sea level, isolating the                      et al., 2019) did not update AR5 estimates, given limited ice sheet from marine influence.                                                              evidence and low agreement from three new studies (Vizcaino et al.,
1260
 
Ocean, Cryosphere and Sea Level Change                                                                                                      Chapter 9 2015; Calov et al., 2018; Aschwanden et al., 2019). Since SROCC,            consistent with future-focused studies (Aschwanden et al. 2019, Le a new study gives a sea level contribution of 0.11 to 0.20 m in low-        Clech et al., 2019, Gregory et al., 2020).
emissions scenarios and 0.61 to 1.29 m in high-emissions scenarios (Van Breedam et al., 2020). The low-emissions projections by Van            The SROCC adopted the AR5 assessment that complete loss of Breedam et al. (2020) encompass AR5s assessed contribution, while          Greenland ice, contributing about 7 m to sea level, over a millennium the high emissions projections are higher than that from AR5. The            or more would occur for a sustained global mean surface temperature optimal ensemble member of Aschwanden et al. (2019) (see also              (GMST) between 1&deg;C (low confidence) and 4&deg;C (medium confidence)
Section 9.4.1.3) indicates that Greenland could contribute 0.25 m            above pre-industrial levels. New studies since SROCC (Gregory et al.,
under RCP2.6 and 1.74 m under RCP8.5. Structured expert judgement            2020; Van Breedam et al., 2020) confirm this assessment (see also (Bamber et al., 2019) projects Greenland losses of 0.54 (0.28-1.28) m        Figure 9.30). Clark et al. (2016) estimate a complete loss to take about under 2&deg;C warming and 0.97 (0.4-2.23) m under 5&deg;C warming. These            8000 years at 5.5&deg;C and about 3000 years at 8.6&deg;C. Based on the studies therefore agree that the AR5 and SROCC assessments are              agreement between new and previous studies, there is therefore high at the low end of the range of projections. In addition, observations        confidence that the rate at which Greenland Ice Sheet commitment is suggest that Greenland Ice Sheet losses are tracking the upper range        realized depends on the amount of warming.
of AR5 projections (T. Slater et al., 2020). Therefore, we update the likely                                                                              9 range for the contribution of the Greenland Ice Sheet to global mean        Accounting for more detailed feedbacks between the atmosphere sea level (GMSL) by 2300 to 0.11-0.25 m under RCP2.6/SSP1-2.6 and            and the ice sheet (Gregory et al., 2020) found a gradual relationship 0.31-1.74 m under RCP8.5/SSP5-8.5. However, given the uncertainty in        between sustained global mean warming and the corresponding near-climatic drivers used to project ice-sheet change over the 21st century      equilibrium ice-sheet volume, in contrast to a sharp threshold as found (Goelzer et al., 2020; Hofer et al., 2020; Nol et al., 2021) and the        by Robinson et al. (2012). Rather than a climatically controlled tipping large range in simulations since AR5 extending beyond 2100, we only          point for irreversible loss of the Greenland Ice Sheet, Gregory et al.
have low confidence in the contribution to GMSL by 2300 and beyond.          (2020) found a threshold of irreversibility linked to ice-sheet size, similar to previous work (Ridley et al., 2010). The results of Gregory et al. (2020)
The role of the elevation-mass feedback for future projections of            show that, if the ice sheet loses mass equivalent to about 3-3.5 m of Greenland can be assessed from paleo simulations. Ice-sheet model            sea level rise, it would not regrow to its present state, and 2 m of the sea simulations of the Laurentide (Gomez et al., 2015; Gregoire et al.,          level rise would be irreversible. The point in time at which the current 2016) and Eurasian (Alvarez-Solas et al., 2019) ice sheets invoke at        ice sheet might reach this critical volume depends on oceanic and least some contribution to last glacial termination mass loss from SMB      atmospheric conditions, ice dynamics, and climate-ice sheet feedbacks reduction, as a consequence of an elevation-mass balance feedback            (Gregory et al., 2020; Van Breedam et al., 2020). Therefore, projections (Levermann and Winkelmann, 2016). In a model spanning Meltwater              differ in the magnitude and rate of temperature change to cross the Pulse 1A, this mechanism increased mass loss by approximately 66%            threshold for irreversible loss. Projections from a large ensemble (Gregoire et al., 2016) but in Last Interglacial simulations, the effect    indicate that the mass threshold may be reached in as early as 400 years of this feedback is shown to depend on the surface scheme of the            under extended RCP8.5 if warming reaches 10&deg;C or more above present climate model employed (Plach et al., 2019). Given the agreement            levels (Aschwanden et al., 2019). In summary, there is high confidence between theoretical analyses and paleo-ice-sheet model experiments,          in the existence of threshold behaviour of the Greenland Ice Sheet in there is high confidence that the elevation-mass balance feedback            a warmer climate; however, there is low agreement on the nature of the is most relevant at multi-centennial and millennial time scales,            thresholds and the associated tipping points.
Box 9.3 l Insights into Land Ice Evolution From Model Intercomparison Projects Projections of ice sheets and glaciers in AR5 (Church et al., 2013b) and SROCC (Oppenheimer et al., 2019) were assessed by collecting single model studies - with the exception of glaciers in SROCC (Hock et al., 2019b). Community benchmark experiments (ISMIP-HOM; Pattyn et al., 2008) or Marine Ice Sheet Model Intercomparison Projects (MISMIP; Pattyn et al., 2012); MISMIP3d, (Pattyn and Durand, 2013); MISMIP+ (Asay-Davis et al., 2016; Cornford et al., 2020) have substantially advanced ice-sheet modelling since AR5. Model Intercomparison Projects (MIPs) now inform projections of both ice sheets and glaciers: the Ice Sheet MIP for CMIP6 (ISMIP6; Sections 9.4.1.3 and 9.4.2.5), the Linear Antarctic Response MIP (LARMIP-2; Section 9.4.2.5) and GlacierMIP (Section 9.5.1.3).
Regional forcing for land ice intercomparison projects Simulations of ice sheets and glaciers are dependent on forcing provided by atmosphere and ocean models. Despite progress in representing processes, reducing biases and increasing resolution, regional and global models still have difficulties reproducing observed regional air temperature, surface mass balance (SMB) and ocean changes (Sections 9.4.1.2 and 9.4.2.2, and Atlas.11). An assessment of CMIP5 and CMIP6 climate models, as forcing for land ice models, has been undertaken (Walsh et al., 2018; Barthel et al., 2020; Marzeion et al., 2020; Nowicki et al., 2020b) with the aim of selecting the best available historical forcings and sampling potential regional future climate changes. Despite improvement in simulation of atmospheric forcing, persistent biases remain in CMIP5 and CMIP6, which reduces the fidelity of historical and future simulations of land ice.
1261
 
Chapter 9                                                                                              Ocean, Cryosphere and Sea Level Change Box 9.3 (continued)
ISMIP6 initial state intercomparison projects The ISMIP6 initial state intercomparison projects (initMIP) for the Greenland (Goelzer et al., 2018) and Antarctic (Seroussi et al.,
2019) ice sheets were designed to understand the uncertainty in sea level projections resulting from the choice of initialization procedures used for projections of sea level (Nowicki et al., 2016). Participating modelling groups (Annex II) were free to decide on the initialization method used to bring ice-sheet models to a present-day state, with the effect of these choices captured in a control simulation (starting from the present-day state, with no further climate forcing applied), which measures intrinsic model drift.
Compared to the earlier SeaRISE intercomparison project (Bindschadler et al., 2013; Nowicki et al., 2013), the modelled present-day ice sheets are in closer agreement with observations, and the model drift has been reduced (Goelzer et al., 2018; Seroussi et al.,
2019). Nonetheless, historical simulations remain challenging for ice-sheet models, due to limited ice-sheet observations prior to the satellite era and biases in the historical atmospheric and oceanic forcings from climate models (Nowicki and Seroussi, 2018). ISMIP6 and LARMIP-2 therefore did not provide a protocol for the historical runs used to bring the ice sheets to present day, nor criteria for 9    sub-selecting models from the multi-model ensemble based on the ability to reproduce historical changes (Levermann et al., 2020; Nowicki et al., 2020a).
ISMIP6 projections for the Greenland and Antarctic ice sheets The ISMIP6 projection protocol (Nowicki et al., 2016, 2020a) was designed to sample the uncertainty in future sea level due to climate scenarios (via the use of high- and low-emissions scenarios and multiple climate models), ice-ocean interactions and inland response to ice-shelf collapse, and ice-sheet model diversity. The participating ice-sheet models are listed in Annex II. For each ice sheet, forcing was selected (Barthel et al., 2020) from the CMIP5 (Taylor et al., 2012) and CMIP6 (Eyring et al., 2016) models. Atmospheric forcing fields consisted of anomalies in SMB and surface air temperatures; these were generated directly from the CMIP models for the Antarctic Ice Sheet and downscaled using the regional climate model (MAR) for the Greenland Ice Sheet (Hofer et al., 2020). To sample the uncertainty due to ocean forcings, models used either a model-specific scheme with the ISMIP6-provided oceanic dataset or a standard ISMIP6 approach. For the Greenland Ice Sheet, the oceanic dataset consists of thermal forcing (temperature minus freezing temperature) extrapolated into fjords and subglacial runoff. The standard approach uses timelines of tidewater glacier retreat (D.A. Slater et al., 2019, 2020). For the Antarctic Ice Sheet, the oceanic dataset consists of salinity, thermal forcing and temperature added to an observationally derived climatology and extrapolated under ice shelves. The standard approach is a basal melt rate that depends quadratically on thermal forcing, adapted from Favier et al. (2019), with two different calibrations (Figure 9.19, Jourdain et al., 2020) that reproduce observed basal melt rates across Antarctica or Pine Island Glacier, respectively (Sections 9.4.2.2, 9.4.2.3). Antarctic ice-shelf disintegration datasets (Nowicki et al., 2020a) assume that ice shelves disintegrate when annual surface melt reaches a threshold (Trusel et al., 2015).
The ISMIP6 projections (Goelzer et al., 2020; Seroussi et al., 2020; Payne et al., 2021) are reported as experiment minus control and represent the sea level resulting from future climate change only. The control simulation, which has constant climate conditions starting in 2015 from the historical run, captures drift associated with the choices made for the initialization method and historical run. Subtraction of this control removes any long-term dynamic response of the ice sheet to pre-2015 climate change. This response has been assessed using dynamic discharge derived from observations over the last 40 years (Mouginot et al., 2019; Rignot et al.,
2019), under an assumption that it persists at the past rate until 2100, rather than diminishing. The dynamic response to historical forcing is estimated as 0.19 +/- 0.10 mm yr -1 for the Greenland Ice Sheet (Section 9.4.1.3) and 0.33 +/- 0.16 mm yr -1 for the Antarctic Ice Sheet (Section 9.4.2.5). Over the period 2015-2100, this leads to an additional sea level contribution of 1.7 cm for Greenland and 2.8 cm for Antarctica.
LARMIP-2 projections for the Antarctic Ice Sheet LARMIP-2 is focused on the uncertainty in the ocean forcing and associated ice-shelf melting (Levermann et al., 2014, 2020) with the majority of the models also participating in ISMIP6 (Annex II). The experiments start from present day and impose an additional basal ice-shelf melting of 8 m yr -1 at the beginning of the 100-year simulation. A control run is used to remove drift resulting from initialization. The time derivative of the ice-sheet response yields a linear response function, which is then convoluted with a forcing of basal shelf melt time series for five Antarctic regions. The forcing time series for RCP2.6, 4.5, 6.0 and 8.5 were obtained from a random combination of global mean temperature for each Representative Concentration Pathway (RCP) from MAGICC-6.0 (Meinshausen et al., 2011), a scaling factor and time delay for the relationship between global surface air temperature and subsurface ocean warming in a given sector of the Southern Ocean from one of 19 CMIP5 models (Taylor et al., 2012) and a basal melting sensitivity from the interval [7-16] m yr -1 &deg;C-1 to convert the regional subsurface warming into basal ice-shelf melting. This process is repeated 20,000 times to obtain a probability distribution of the sea level contribution for five Antarctic sectors. The linear response framework captures complex temporal responses of the ice sheets resulting from an increase in basal ice-shelf melting, but neglects the response to SMB and any self-dampening or self-amplifying processes, such as marine ice shelf instability (MISI). The LARMIP-2 method is 1262
 
Ocean, Cryosphere and Sea Level Change                                                                                                    Chapter 9 Box 9.3 (continued) applied to temperature projections for the Shared Socio-economic Pathways (SSPs; Supplementary Material 7.SM.2) and an estimate of SMB change from the AR5 parametric Antarctic Ice Sheet SMB model (Church et al., 2013b) is added to the results (Sections 9.4.2.4, 9.4.2.5 and 9.6.3.2). It is not necessary to add a long-term dynamic response to the LARMIP-2 projections, as this is incorporated in the basal melt time series.
GlacierMIP projections GlacierMIP (Marzeion et al., 2020) was designed to estimate the glacier contribution to sea level rise, including from peripheral glaciers in Greenland and Antarctica that can be considered to be dynamically decoupled, or entirely separate, from the ice sheets.
Glacier models are described in Annex II. Initial conditions were based on Randolph Glacier Inventory Version 6 (RGI Consortium, 2017) and initial ice thickness and volume were provided from an update of Huss and Farinotti (2012), although some glacier models used their own estimates. Forcings were taken from 10 different CMIP5 general circulation models, selected based on availability of multiple RCPs, the choice in a previous model intercomparison (Hock et al., 2019a), and performance in glacier-covered regions according to              9 Walsh et al. (2018). In addition, two global glacier models performed the same experiment with 13 CMIP6 models (Section 9.5.1.3).
Use of an emulator with ISMIP6 and GlacierMIP projections The ISMIP6 and GlacierMIP projections are primarily based on a limited number of CMIP5 RCPs and CMIP6 SSPs, and a limited sampling of ice-ocean interaction parameters and ice-shelf collapse simulations. Emulators provide a method for expanding these projections to a range of SSPs with more comprehensive sampling of climate, ice-sheet and glacier modelling uncertainties. Sections 9.4.1.3, 9.4.2.5 and 9.5.1.3 show estimates from the emulator of Edwards et al. (2021). This is a Gaussian Process, rather than a physically based (Cross-Chapter Box 7.1) model derived from the ISMIP6 and GlacierMIP simulations; projections use distributions of global surface air temperature (GSAT) from the two-layer emulator (Supplementary Material 7.SM.2) and ice-sheet parameters as inputs, and include estimates of the emulator uncertainty. Therefore, probability intervals are not inflated by a further factor, as is often the case for multi-model ensemble projections, to account for missing uncertainties (Section 9.6.3.2). The emulator is used in Section 9.6.3 to provide projections of the land ice contribution to sea level that are fully consistent with each other, ocean heat content, and the assessed equilibrium climate sensitivity and projections of GSAT across the entire report.
9.4.2        Antarctic Ice Sheet                                          Bamber et al., 2018b; Gardner et al., 2018; The IMBIE Team, 2018; Rignot et al., 2019).
9.4.2.1      Recent Observed Changes The SROCC reported with very high confidence that the acceleration, As stated in Section 2.3.2.4, satellite observations by Ice Sheet Mass    retreat and thinning of the principal West Antarctic outlet glaciers Balance Intercomparison Exercise (IMBIE) combining multi-team              has dominated the observed Antarctic mass loss over the last estimates based on altimetry, gravity anomalies (GRACE) and the            decades, and stated with high confidence that these losses were input-output method, already presented in SROCC (Meredith et al.,          driven by melting of ice shelves by warm ocean waters. The average 2019), are updated and extended to 2020 (The IMBIE Team, 2021).            West Antarctic Ice Sheet (WAIS) mass loss of 82 +/- 9 Gt yr -1 between The Antarctic Ice Sheet (AIS) lost 2670 [1800 to 3540] Gt mass over        1992 and 2017 (The IMBIE Team, 2021) leads to substantial observed the period 1992-2020, equivalent to 7.4 [5.0 to 9.8] mm GMSL rise          surface lowering (e.g., Schrder et al., 2019; Shepherd et al., 2019),
(for contribution to sea level budget, see Figures 9.16 and 9.18, and      particularly in coastal regions (Figure 9.18). Recent studies using Table 9.5). Within uncertainties, this estimate agrees with a review of    satellite altimetry (Schrder et al., 2019) and the input-output method post-AR5 studies up to 2016 (Bamber et al., 2018b) and is consistent      (Rignot et al., 2019) consistently show mass loss in these coastal with recent single studies based on satellite laser altimetry (Smith      regions since the late 1970s (Figure 9.16). Because of consistent et al., 2020), the input-output method (Rignot et al., 2019) and          multiple lines of evidence, there is high confidence in mass loss of the gravimetry (Velicogna et al., 2020). The mass-loss rate was on average    Totten Glacier in East Antarctica (Miles et al., 2013; X. Li et al., 2016; 49 [-2 to 100] Gt yr -1 over the period 1992-1999, 70 [22 to 119]          Mohajerani et al., 2018; Rignot et al., 2019; Schrder et al., 2019; Gt yr -1 over the period 2000-2009, and 148 [94 to 202] Gt yr -1 over      Shepherd et al., 2019) since about 2000, dominated by changes in the period 2010-2016 (see Figures 9.16 and 9.18, and Table 9.SM.1).        coastal ice dynamics (X. Li et al., 2016). It is currently unclear whether However, recent work suggests that the mass loss has not further          mass loss of the EAIS over the last three decades has been significant increased since 2016 because of regional mass gains in Dronning            (Rignot et al., 2019) or, at 5 +/- 46 Gt yr -1 between 1992 and 2017, Maud Land (Velicogna et al., 2020). Mass loss of the West Antarctic        essentially zero within uncertainties (The IMBIE Team, 2018).
and Antarctic Peninsula ice sheets has increased since about 2000          In summary, WAIS losses, through acceleration, retreat and thinning (very high confidence), essentially due to increased ice discharge        of the principal outlet glaciers, dominated the AIS mass losses over (Harig and Simons, 2015; Paolo et al., 2015; Forsberg et al., 2017;        the last decades (very high confidence) and there is high confidence 1263
 
Chapter 9                                                                                                                    Ocean, Cryosphere and Sea Level Change that this is the case since the late 1970s. Furthermore, parts of the                          with multi-decadal increases in the Antarctic Peninsula inferred since EAIS have lost mass in the last two decades (high confidence).                                the 1930s (Medley and Thomas, 2019), and dominate the interannual to decadal variability of the AIS mass balance (Rignot et al., 2019).
As stated in SROCC, snowfall and glacier flow are the largest                                  However, no significant continent-wide SMB trend is inferred since components determining AIS mass changes, with glacier flow                                    1979 (The IMBIE Team, 2018; Medley and Thomas, 2019; regional acceleration (dynamic thinning) on the WAIS and the Antarctic                                  changes of Antarctic SMB are assessed further in Atlas Section 11.1).
Peninsula driving total loss trends in recent decades (very high                              In summary, there is very high confidence that the observed AIS mass confidence), and a partial offset of the dominating dynamic-thinning                          loss since the early 1990s is primarily linked to ice-shelf changes.
losses by increased snowfall (high confidence). The SROCC attributed medium confidence to estimates of 20th-century snowfall increases                              The SROCC stated with high confidence that melting of ice shelves by equivalent to a sea level change of -7.7 +/- 4.0 mm on the EAIS, and                            warm ocean waters, leading to reduction of ice-shelf buttressing, has
  -2.8 +/- 1.7 mm on the WAIS, respectively (Medley and Thomas, 2019).                            driven the observed ongoing thinning of major WAIS outlet glaciers.
Loss of buttressing, which can be caused by ice-shelf thinning, gradual                        Since SROCC, digitized radar measurements have shown that the ice-shelf front retreat or ice-shelf disintegration, has been linked to                        eastern ice shelf of Thwaites Glacier in the Amundsen Sea Embayment 9 instantaneous ice velocity increases, and thus dynamic thinning, since                        thinned between 10 and 33% during the three decades after 1978 the early 1990s. This link is clearly evident in the Amundsen and, to                          (Schroeder et al., 2019), and the role of basal ice-shelf melting has a lesser degree, Bellingshausen sectors (Gudmundsson et al., 2019),                            been emphasized (Smith et al., 2020). Strong surface meltwater where passive shelf ice (ice that can be removed without major                                production has been noted as a precursor of ice-shelf disintegration effects on the ice-shelf dynamics) is very limited or absent (F&#xfc;rst                            in and since SROCC (Bell et al., 2018), and recent work placed strong et al., 2016). Surface mass balance (SMB) changes, dominated by                                meltwater production events (Lenaerts et al., 2017; Nicolas et al.,
snowfall, exhibit strong regional and temporal variability, for example                        2017; Wille et al., 2019) and seasons (Robel and Banwell, 2019)
Antarctic ice sheet cumulative mass change & equivalent sea level contribution (a)                                    (b) 4 Modern & Projected Changes                                                                                  -0.1 2100 medians, 66% and 90% ranges Rignot ISMIP6  Emulator LARMIP-2 Bamber 0                                                                                                            0 IMBIE m
SSP5-8.5                            0.1
                                                -4 4
Gt                                                                                            SSP1-2.6 0.2
                                                -8 0.3
                                                -12                                                  Emulator median (SSPs)
Emulator 90% range Emulator 66% range                                          0.4
                                                -16                                                  ISMIP6 models (RCPs/SSPs) 1980            2000              2020              2040            2060              2080              2100 Figure 9.18 l Antarctic Ice Sheet cumulative mass change and equivalent sea level contribution. (a) A p-box (Section 9.6.3.2) based estimate of the range of values of paleo Antarctic ice sheet mass and sea level equivalents relative to present day and the median over all central estimates (Bamber et al., 2009; Argus and Peltier, 2010; Dolan et al.,
2011; Mackintosh et al., 2011; Golledge et al., 2012, 2013, 2014, 2015, 2017b; K.G. Miller et al., 2012; Whitehouse et al., 2012; Ivins et al., 2013; Argus et al., 2014; Briggs et al.,
2014; Maris et al., 2014; de Boer et al., 2015, 2017; Dutton et al., 2015; Pollard et al., 2015; DeConto and Pollard, 2016; Gasson et al., 2016; Goelzer et al., 2016; Yan et al., 2016; Kopp et al., 2017; Simms et al., 2019); (b left) cumulative mass loss (and sea level equivalent) since 2015, with satellite observations shown from 1993 (Bamber et al., 2018a; The IMBIE Team, 2018; WCRP Global Sea Level Budget Group, 2018) and observations from 1979 (Rignot et al., 2019), and projections from Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6) to 2100 under RCP8.5/SSP58.5 and RCP2.6/SSP12.6 scenarios (thin lines from Seroussi et al., 2020; Edwards et al., 2021; Payne et al., 2021) and ISMIP6 emulator under SSP5-8.5 and SSP1-2.6 to 2100 (shades and bold line; Edwards et al., 2021); (b, right) 17th-83rd, 5th-95th percentile ranges for ISMIP6, ISMIP6 emulator, and LARMIP-2 including surface mass balance (SMB) at 2100. (c-e) Schematic interpretations of individual reconstructions (Anderson et al., 2002; Bentley et al., 2014; de Boer et al.,
2015; Goelzer et al., 2016) of the spatial extent of the Antarctic Ice Sheet are shown for the: (c) mid-Pliocene Warm Period, (d) Last Interglacial; and (e) Last Glacial Maximum (Fretwell et al., 2013): grey shading shows extent of grounded ice. (f-g) Maps of mean elevation changes (f) 1978-2017 derived from multi-mission satellite altimetry (Schrder et al., 2019) and (g) ISMIP6: 2061-2100 projected changes for an ensemble using the Norwegian Climate Centers Earth System Model (NorESM1-M) climate model under the RCP8.5 scenario (Seroussi et al., 2020). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1264
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 in this context. Antarctic ice-shelf basal meltwater flux varied          The SROCC stated with high confidence that ice-shelf disintegration between about 1100 +/- 150 Gt yr -1 in the mid-1990s and about              has driven dynamic thinning in the northern Antarctic Peninsula over 1570 +/- 140 Gt yr -1 in the late 2000s before decreasing to                recent decades, and expressed high confidence in current ongoing 1160 +/- 150 Gt yr -1 in 2018, and basal melt rates strongly vary with      mass loss from glaciers that fed now-disintegrated ice shelves.
geographical position and depth, as a function of the surrounding        However, the mass loss rate has decreased in the 20 years since water temperature (Adusumilli et al., 2020). Section 9.2.2.3 assesses    the immediate speed-up following ice-shelf disintegration in 1995 that the intrusion of warm Circumpolar Deep Water (CDW), which            and 2002. Observed flow speed of these tributary glaciers is still has warmed and shoaled since the 1980s, has been at least partially      26% higher than before the ice shelf disintegration (Seehaus et al.,
controlled by forcing with significant decadal variability. Limited      2018). Conversely, one study interpreted the increased flow speed evidence suggests that, beyond strong internal decadal wind              of the Scar Inlet Ice Shelfs tributary glaciers as a sign of evolving variability, increased greenhouse gas forcing has slightly modified the  instability of the currently intact ice shelf (Qiao et al., 2020).
mean local winds between 1920 and 2018, facilitating the intrusion of CDW heat on the Amundsen-Bellingshausen continental shelf,            Ongoing grounding line retreat, indicating dynamic thinning, and increased ice shelf melt (Section 9.2.2.3). However, theoretical      is observed with high confidence in many areas of Antarctica, understanding is still incomplete and in situ measurements within        and particularly on the WAIS, with the highest rates being in the        9 the ice-ocean boundary layer are sparse (Whlin et al., 2020).            Amundsen and Bellingshausen Sea areas, and around Totten Glacier Modelling, and therefore attribution of ice shelf basal melt, remains    in East Antarctica, as stated in SROCC. Research published since challenging because of insufficient process understanding, required      SROCC has evidenced grounding line retreat of the West Antarctic spatial resolution, the paucity of in situ observations (Dinniman et al., Berry Glacier on the Getz Coast (Millan et al., 2020) and on the East 2016; Asay-Davis et al., 2017; Turner et al., 2017), and uncertainties    Antarctic Denman Glacier (Brancato et al., 2020), both since 1996.
of bathymetric datasets under ice-shelf cavities (Goldberg et al.,        Furthermore observed grounding line retreat in excess of 1.5 km 2019, 2020; Morlighem et al., 2020). In summary, ice-shelf thinning,      between 2003 and 2015 has been reported for parts of Marie Byrd mainly driven by basal melt, is widespread around the Antarctic coast    Land (Christie et al., 2018). In summary, there is high confidence and particularly strong around the WAIS (high confidence), although      that grounding lines of marine-terminating glaciers are currently basal melt rates show substantial spatio-temporal variability.            retreating in many areas around Antarctica, particularly around the WAIS, and additional areas of grounding line retreat have been Satellite observations suggest that changes in sea ice coverage and      evidenced since SROCC.
thickness can modulate iceberg calving, ice shelf flow and glacier terminus position around Antarctica (Miles et al., 2013, 2016, 2017;      The SROCC stated with medium confidence that sustained mass Massom et al., 2015; Greene et al., 2018; Bevan et al., 2019), either    losses of several major glaciers in the Amundsen Sea Embayment through mechanical coupling or via changes to ocean stratification,      (ASE) are compatible with the onset of marine ice sheet instability influencing basal melting. A combined observational and modelling        (MISI). However, whether unstable WAIS retreat had begun, or was study (Massom et al., 2018) showed that regional loss of a protective    imminent, remained a critical uncertainty. New publications since sea ice buffer played a role in the rapid disintegration events of the    SROCC have not substantially clarified this question. One study Larsen A and B and Wilkins ice shelves in the Antarctic Peninsula        that combined satellite measurements with a numerical model and between 1995 and 2009, by exposing damaged (rifted) outer ice shelf      prescribed ice-shelf thinning (Gudmundsson et al., 2019) suggests margins to enhanced flexure by storm-generated ocean swells. One          that MISI is not required to explain the observed current mass observational study (Sun et al., 2019) suggests that the absence of sea  loss rates of the WAIS, because they are consistent with external ice in front of ice shelves, which leads to strengthened topographic      climate drivers. Furthermore, the fast grounding line retreat of the waves, favours higher ice-shelf basal melt rates by increasing the        Pine Island Glacier in the ASE, which was triggered in the 1940s baroclinic (depth varying) ocean heat flux which can enter the cavity    (Smith et al., 2017), observed after 1992 (Rignot et al., 2014) and (Whlin et al., 2020). Paleo evidence for sea ice control on ice sheets  previously interpreted as a sign of MISI (Favier et al., 2014), seems to is lacking, but geologic evidence shows a concordance between            have stabilized recently (Milillo et al., 2017; Konrad et al., 2018), and periods of ice-sheet growth and the expansion of sea ice (Patterson      its current flow patterns do not suggest ongoing or imminent MISI et al., 2014; Levy et al., 2019), both being favoured by reduced sea      (Bamber and Dawson, 2020). However, sustained fast grounding line surface temperatures. Modelling confirms that sea ice controls the        retreat has been observed for the Smith Glacier in the ASE (Scheuchl strength of ice m&#xe9;lange (Robel, 2017; Schlemm and Levermann, 2021)        et al., 2016), and an analysis of flow patterns and grounding line and thus influences ice-shelf flexure and calving rates and stability of  retreat of the ASE Thwaites Glacier between 1992 and 2017 (Milillo floating ice margins, but one model shows this had negligible effect      et al., 2019) showed sustained, albeit spatially heterogeneous, on AIS retreat rates during past warm periods (Pollard et al., 2018). grounding line retreat, highlighting ice-ocean interactions that Loss of ice-shelf-proximal sea ice is also associated with increased      lead to increased basal melt. In addition, Denman Glacier in East solar heating of surface waters and increased sub-shelf melting          Antarctica was shown to hold potential for unstable retreat (Brancato (Bendtsen et al., 2017; Stewart et al., 2019). In summary, although      et al., 2020). In summary, the observed evolution of the ASE glaciers in some cases sea ice decrease and glacier and ice-shelf flow and        is compatible with, but not unequivocally indicating an ongoing MISI terminus position changes can have the same common cause, there          (medium confidence).
is medium confidence that sea ice decrease ultimately favours the mass loss of nearby ice shelves through a variety of processes.
1265
 
Chapter 9                                                                                          Ocean, Cryosphere and Sea Level Change The SROCC reported limited evidence and medium agreement for              the advantage of connecting melt rates to emissions scenarios, anthropogenic forcing of the observed AIS mass balance changes.          a large variety of melt parametrizations exist (DeConto and Pollard, As stated in Section 3.4.3.2, there remains low confidence in            2016; Lazeroms et al., 2018; Reese et al., 2018; Hoffman et al., 2019; attributing the causes of the observed mass of loss from the AIS since    Pelle et al., 2019; Jourdain et al., 2020), and there is low agreement 1993, in spite of some additional process-based evidence to support      due to limited observational constraints (ocean temperature, salinity, attribution to anthropogenic forcing.                                    velocity, and ice shelf draft)(Jourdain et al., 2020), uncertainty in the physics of parametrized processes, missing processes (e.g., tides),
9.4.2.2    Model Evaluation                                              and uncertainty in the treatment of ice-sheet-climate feedbacks (Donat-Magnin et al., 2017; Bronselaer et al., 2018; Golledge et al.,
The AR5 (Church et al., 2013b; Flato et al., 2013) stated that regional  2019). Parametrizations are usually calibrated to present-day melt climate models and global models with bias-corrected SST and sea          rates, but can respond differently to projected ocean warming (Favier ice concentration tended to produce more accurate simulations of          et al., 2019; Jourdain et al., 2020). Two different calibrations were Antarctic SMB than coupled climate models. It also noted strong          used in ISMIP6 (Box 9.3; Jourdain et al., 2020; Nowicki et al., 2020b):
climate model temperature biases over the Antarctic, though the          one reproducing melt rates averaged around the whole continent 9 latter may reflect known biases in the reanalysis used (Fr&#xe9;ville et al.,  (MeanAnt: Figure 9.19), and the other reproducing melt rates near the 2014). Section Atlas.11.1 assesses that there is medium confidence        grounding line of Pine Island Glacier (PIGL; see Figure 9.19), leading to in the capacity of climate models to simulate Antarctic climatology      large differences in melt rates. Evaluation with observations and two and SMB changes.                                                          cavity-resolving models suggests that the MeanAnt parametrization better reproduces observed melt rates and projected increases in both Section 9.2.3.2 assesses that there is low confidence in simulations      the warm Amundsen Sea Embayment and cold Ronne-Filchner shelf of Southern Ocean temperature. Few ocean models resolve ice-shelf        cavity, as well as total Antarctic melting (Jourdain et al., 2020). The cavities, and biases in present-day melt rates can be substantial in      PIGL calibration represents the upper end for increased basal melt some sectors, including the key region of the Amundsen Sea (e.g., an      sensitivity that would be caused by continent-wide changes to ocean exception is the FESOM simulation in Figure 9.19 includes ice-            water properties and circulation under strong future forcing (Jourdain shelf cavities and simulates ice-shelf basal melting and refreezing)      et al., 2020). The basal sliding law also has a strong influence (Naughten et al., 2018). An increasing number of observational            on grounding line retreat and glacier acceleration in response to studies from which basal melt rates are calculated (Huhn et al., 2018;    perturbations, and varies spatially (Sun et al., 2020). Sliding laws Adusumilli et al., 2020; Das et al., 2020; Hirano et al., 2020; Stevens  (Joughin et al., 2019) can only be constrained with observations in et al., 2020), combined with improved understanding of influences        regions experiencing significant change, and with sufficiently long specific to water-masses and modes of melting or dissolving (Silvano      observational records.
et al., 2018; Adusumilli et al., 2020; Malyarenko et al., 2020; Whlin et al., 2020), may help to refine these models in the future. However,    The SROCC noted that AIS simulations are increasingly evaluated given the limited number of available models and their biases, there      or formally calibrated with modern observations and/or paleodata -
is currently low confidence in the sub-shelf melt rates simulated by      to obtain more realistic initial conditions (ice-sheet geometry, ocean models.                                                            velocity and forcing) and to constrain uncertainty in probabilistic projections. This trend continues (Nias et al., 2019; Gilford et al.,
Improvements in the representation of grounding line evolution in        2020; Hamlington et al., 2020b; Wernecke et al., 2020). However, ice-sheet models since AR5 (such as sub-grid schemes for basal friction  while the large-scale characteristics of the initial ice-sheet state and ice-shelf melt, and local grid refinement) means that most of the    have improved significantly (Box 9.3), capturing the smaller-scale model simulations presented in SROCC were dominated by physical          rates of change, including mass trends, remains challenging for processes. Since then, these advances have been applied in several        many models (Goldberg et al., 2015; Reese et al., 2020; Seroussi model intercomparison projects - such as ISMIP6 and LARMIP-2              et al., 2020; Siegert et al., 2020). This increases uncertainty in (see Box 9.3); MISMIP+ (Cornford et al. 2020); and ABUMIP (Sun et al. projections, especially for the 21st century (Section 9.4.2.5).
2020). All models participating in ISMIP6 and LARMIP-2 simulate          However, uncertainties in ice-sheet model simulations have been ice-shelf and grounding-line evolution, and include sub-shelf melt        much better quantified since AR5, through model intercomparison parametrization, which was not the case in the Sea-level Response to      projects (in particular, ISMIP6 and LARMIP-2; see Box 9.3),
Ice Sheet Evolution (SeaRISE) project intercomparison (Bindschadler      perturbed parameter ensembles, and increasing use of statistical et al., 2013; Nowicki et al., 2013). Simulations of grounding line        emulation (Gilford et al., 2020; Levermann et al., 2020; Wernecke evolution (Seroussi et al., 2017, 2020) have benefitted from improved    et al., 2020; DeConto et al., 2021; Edwards et al., 2021) to better bedrock topography (Morlighem et al., 2020). Treatment of sub-shelf      sample the parameter space. By exploring uncertainties more fully, melting, however, remains one of the causes of large differences in      these methods have the potential to identify better simulations of AIS models, particularly for partially floating grid cells in models with the historical period.
coarse resolution (Levermann et al., 2020; Edwards et al., 2021). Due to the limitations in resolving cavities in ocean models, as described    An important difficulty is how to evaluate simulations of processes above, basal melt rates are generally parameterized at the ice shelf      that are: not currently observed; or rare; or indirectly deduced - in base, based on ocean model simulations of temperatures and salinity      particular, the ice-shelf disintegrations and cliff failures that would instead (Nowicki et al., 2020b; Seroussi et al., 2020). While this has    drive the proposed marine ice cliff instability (MICI; Section 9.4.2.4 1266
 
Ocean, Cryosphere and Sea Level Change                                                                                                                          Chapter 9 9
Figure 9.19 l Ice-shelf basal melt rates for present-day (upper panels) and changes from present-day to the end of the 21st century under the RCP8.5 scenario (lower panels). Present-day melt rates were estimated through: the input-output method constrained by satellite observations and atmosphere/snow simulations (Rignot et al., 2013) and representative of 2003-2008 (upper left); the Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6) non-local-PIGL parametrization constrained by observation-based ocean properties (Jourdain et al., 2020) and representative of 1995-2014 (upper centre); the Finite Element Sea ice/Ice Shelf Ocean Model (FESOM) simulation over 2006-2015, forced by atmospheric conditions from a Coupled Model Intercomparison Project Phase 5 (CMIP5) multi-model mean (MMM) under the RCP8.5 scenario (Naughten et al., 2018) (upper right). Future anomalies are calculated as 2081-2100 minus present-day using the ISMIP6 non-local-MeanAnt and non-local-PIGL parametrizations (Jourdain et al., 2020) (lower left and centre, respectively) based on projections from the Norwegian Climate Centers Earth System Model (NorESM1-M)
CMIP5 model, and the FESOM-MMM projection (lower right). Note the symmetric-log colour bar (linear around zero, logarithmic for stronger negative and positive values). Inset highlights the Amundsen Sea Region. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
and Box 9.4; DeConto and Pollard, 2016; DeConto et al., 2021). Models                    9.4.2.3      Drivers of Future Antarctic Ice Sheet Change of ice-cliff failure can only be indirectly and partially evaluated, using existing (i.e., static) cliffs and laboratory experiments (Clerc et al.,                  9.4.2.3.1 Surface mass balance 2019). The SROCC stated that there was low agreement on the exact MICI mechanism and limited evidence of its occurrence in the present                      The AR5 projected a negative contribution from Antarctic surface or the past, and that the validity of MICI remains unproven. Only one                    mass balance (SMB) changes to sea level over the 21st century ice-sheet model represents MICI (Pollard et al., 2015; DeConto and                        (i.e., mitigating sea level rise), due to increased snowfall associated Pollard, 2016; DeConto et al., 2021). The mechanism has not been                          with warmer air temperatures. Sensitivity of SMB to Antarctic found to be essential for reproducing Mid Pliocene Warm Period and                        surface air temperature change varied from 3.7 to 7% &deg;C-1, and Last Interglacial reconstructions or satellite observations, though                      the sea level projections assumed a sensitivity of 5.1 +/- 1.5% &deg;C-1 Last Interglacial data slightly favours it in this model (Edwards et al.,                from CMIP3 era models (Gregory and Huybrechts, 2006) to estimate 2019; Gilford et al., 2020; DeConto et al., 2021).                                        SMB changes from Antarctic temperatures in the CMIP5 ensemble.
Since the AR5, analyses of CMIP5 and CMIP6 models have found In summary, there is now medium confidence in many ice-sheet                              Antarctic temperature sensitivity for accumulation (precipitation processes in ice-sheet models, including grounding line evolution.                        minus sublimation) of 3.5 to 8.7% &deg;C-1 (Frieler et al., 2015), for SMB However, there remains low confidence in the ocean forcing affecting                      of 6.0 to 9.9% &deg;C-1 (Previdi and Polvani, 2016) and for precipitation of the basal melt rates, and low confidence in simulating mechanisms                        around 4 to 9% &deg;C-1 (+/-1 standard deviation ranges; Bracegirdle et al.,
that have the potential to cause widespread, sustained and very                          2020). An accumulation sensitivity estimate derived from ice core rapid ice loss from Antarctica through MICI.                                              data lies in the middle of the range, around 6% &deg;C-1 (Frieler et al.,
2015). These are consistent, within uncertainties, with each other and AR5, under the approximation that SMB is dominated by snowfall.
1267
 
Chapter 9                                                                                          Ocean, Cryosphere and Sea Level Change The AR5 found that the median and likely sea level contributions      The response of sub-shelf melting to ocean warming is also poorly due to SMB from 1986-2005 to 2100 were -0.05 (-0.09 to                constrained. A key unknown is whether, and when, cold ice-shelf
  -0.02) m under RCP8.5 and -0.02 (-0.05 to 0.00) m under RCP2.6.        cavities might become more similar to the Amundsen Sea Embayment, The SROCC did not present a separate SMB contribution, instead        not only in ocean temperature but also ice-ocean heat exchange, showing total Antarctic projections derived from ice-sheet models      which depends on the cavity geometry and ocean circulation (Section 9.4.2.5). Projections of the SMB contribution to sea level    (Little et al., 2009). Only two ocean models with ice-shelf cavities have tend to be slightly more negative since AR5, due at least in part to  been used to make sub-shelf basal melting projections for Special the higher range in equilibrium climate sensitivity values in CMIP6    Report on Emissions Scenarios and Representative Concentration (Payne et al., 2021). Mean and +/-1 standard deviation ranges for        Pathway (RCP) scenarios (Hellmer et al., 2012; Timmermann and grounded Antarctic Ice Sheet SMB changes from 2000 to 2100            Hellmer, 2013; Timmermann and Goeller, 2017; Naughten et al.,
computed from CMIP5 models are -0.08 (-0.13 to -0.04) m sea            2018). The FESOM simulation, forced by a CMIP5 multi-model mean level equivalent (SLE) for RCP8.5 and, similarly for CMIP6 models,    under RCP8.5, projects a 90% increase in melting (Figure 9.19),
are -0.07 (-0.11 to -0.03) m for SSP58.5 (Gorte et al., 2020). The    although this could be overestimated due to an underestimation general circulation models (GCMs) used to drive ice-sheet models      of present-day melt rates (Section 9.4.2.2; Naughten et al., 2018).
9 in ISMIP6 (Box 9.3) project mean grounded AIS SMB changes from        The temperature-melt relationship was parameterized by ISMIP6 in 2005 to 2100 of -0.06 (range -0.08 to -0.03) m SLE under RCP8.5 for    terms of heat exchange velocity in m a-1, and by LARMIP-2 as basal the six CMIP5 models (Seroussi et al., 2020) and -0.09 (range -0.10    melt sensitivity in m a-1 &deg;C-1 (Box 9.3; Jourdain et al., 2020; Levermann to -0.07) m SLE under SSP58.5 for the four CMIP6 models, which        et al., 2020; Reese et al., 2020), and both vary widely around the have climate sensitivity values of 4.8&deg;C -5.3&deg;C (Payne et al., 2021). continent, depending on cavity type. Median values of ISMIP6 heat We apply the AR5 parametric AIS SMB model (Section 9.6.3.2) to        exchange velocity vary by a factor of 5-10 when calibrating to updated projections of global mean temperature from a two-layer        either mean Antarctic or high Pine Island Glacier observed melt rates energy budget emulator (Supplementary Material 7.SM.2), which          (Section 9.4.2.2; Box 9.3; Jourdain et al., 2020). Basal melt sensitivities gives a median -0.05 (5-95% range -0.07 to -0.02) m SLE for            near the grounding line estimated by Reese et al. (2020) with a box SSP58.5 (Section 9.4.2.5, Table 9.3), that is, similar to the AR5    model of ocean overturning range from 3.9 m a-1 &deg;C-1 for the Weddell assessment and slightly smaller than the CMIP6 estimate. This          Sea to 10.5 m a-1 &deg;C-1 for the Amundsen Sea region, with a continental estimate is used to augment the LARMIP-2 dynamic projections          mean of 5.3 m a-1 &deg;C-1. Similarly high Amundsen Sea sensitivities (Box 9.3) in Sections 9.4.2.5 and 9.4.2.6. Overall, CMIP5 and CMIP6    are estimated in coupled ice-ocean simulations of Thwaites Glacier GCM simulations of sea level fall by 2100 due to Antarctic SMB        (mean 9.4 m a-1 &deg;C-1; range 6-16 m a-1 &deg;C-1) (Seroussi et al., 2017).
increases are around 2-4 cm greater than estimates derived with        These large variations lead to large differences in basal melt rates and the statistical method used in AR5. Further details about projections  projected sea level contributions when applied to the whole ice sheet of Antarctic temperature, precipitation and SMB are provided in        in ISMIP6 and LARMIP-2 (Box 9.3). Projections of melt rates from the Section Atlas.11.1.4, which assesses that, due to the challenges of    two ISMIP6 calibrations are higher than those from FESOM, driven model evaluation (Section 9.4.2.2) and the possibility of increased    by a CMIP5 multi-model mean (Figure 9.19; Jourdain et al., 2020).
meltwater runoff (Kittel et al., 2021), there is only medium          The ISMIP6 ensemble mostly uses the mean Antarctic calibration, but confidence that the future contribution of Antarctic SMB to sea level  includes some simulations with the Pine Island Glacier calibration, this century will be negative under all greenhouse gas emissions      and the ISMIP6 emulator samples more of these higher values; scenarios. Longer time scales are discussed in 9.4.2.6.                LARMIP-2 uses basal melt sensitivities (7-16 m a-1 &deg;C-1) consistent with estimates for the Amundsen Sea Embayment. Due to the limited 9.4.2.3.2 Sub-shelf melting                                            availability of cavity-resolving ocean models, and the wide regional variation in estimates of basal melt sensitivity to ocean temperature, The SROCC highlighted that an important ongoing deficiency in          there is only low confidence in projected future sub-ice-shelf melt projections of Antarctic sub-shelf melting is the lack of ice-ocean    rates. The impact of this uncertainty on AIS model projections to 2100 coupling in most continental-scale studies. Increased basal melting is is discussed in Section 9.4.2.5.
mainly caused by warmer CDW (Section 9.2.2.3) on the continental shelves, and warming surface waters intruding under ice shelves        9.4.2.3.3 Ice-shelf disintegration (Naughten et al., 2018). Predicting whether or not open ocean water masses will freely penetrate ice shelf cavities, or will be partially  Antarctic ice shelves modulate grounded ice flow through buttressing, blocked by ocean density gradients, is complex (Whlin et al., 2020);  so their weakening or disintegration is crucial for the timing and while melting related to CDW inflow is currently dominant in the      magnitude of ice loss and onset of instabilities (Section 9.4.2.4; Amundsen Sea Embayment, melt in other embayments is limited            Box 9.4). Projections of ice-shelf disintegration are uncertain in terms by deep inflows of high-salinity shelf water or seasonally warmed      of atmospheric warming and the response of the shelf surface -
shallow incursions of Antarctic Surface Water (Stewart et al., 2019;  that is, surface melting, and whether shelves then disintegrate due Adusumilli et al., 2020). There is little consensus regarding future  to hydrofracturing and flexing, or are resilient through refreezing or change in CDW (Section 9.2.2.3), and more generally low confidence    drainage (Bell et al., 2018). The SROCC stated it is not expected that in future change in the temperature of Antarctic ice-shelf cavities    widespread ice-shelf loss will occur before the end of the 21st century, (Section 9.2.3.2).                                                    but this was based on only one study, using a regional climate model forced by five GCMs (Trusel et al., 2015), so there was low confidence 1268
 
Ocean, Cryosphere and Sea Level Change                                                                                                Chapter 9 in this assessment. The study of DeConto and Pollard (2016) projected  troughs in East Antarctica potentially vulnerable to MISI (Morlighem the appearance of extensive surface meltwater several decades          et al., 2020) has only been used by a few models (Seroussi et al.,
earlier than Trusel et al. (2015) and was therefore assessed to be too  2020; Sun et al., 2020), so current projections could underestimate uncertain to include in SROCC projections of the AIS.                  vulnerability in these regions. The sea level rise contribution of the AIS therefore crucially depends on the behaviour of individual ice Since SROCC, further studies have highlighted the modelling            shelves and outlet glacier systems and whether they enter MISI for uncertainties in this area. Coastal surface air temperature projections a given level of warming (Box 9.4; Pattyn and Morlighem, 2020).
in CMIP6 models show large inter-model differences driven by            As for Antarctic simulations generally (Sections 9.4.2.2 and 9.4.2.3),
sea ice retreat and exhibit more warming relative to global mean        there is medium confidence in simulating MISI but low confidence temperature under low emissions than high, due to delayed response      in projecting the sub-shelf melting and ice-shelf disintegration that of the Southern Ocean to stabilized emissions and stratospheric ozone  drive it.
recovery (Bracegirdle et al., 2020). The updated study of DeConto et al. (2021) includes improvements to the climate simulations          The SROCC noted limited evidence from geological records and relative to those in DeConto and Pollard (2016), and the resulting      ice-sheet modelling, suggesting that parts of the AIS experienced surface meltwater projections are now consistent with Trusel et al. rapid (centennial) retreat likely due to MISI between 20,000 and          9 (2015). However, the net effect of meltwater feedbacks on ice shelves  9,000 years ago, and also described more uncertain evidence for the is uncertain. Ice discharge is expected to lead to surface ocean and    Last Interglacial (LIG) and mid-Pliocene Warm Period (MPWP). Recent atmosphere cooling: this increases ocean stratification and sub-shelf  support for past MISI is provided by model simulations of the WAIS melting, but also reduces ice-shelf surface melting and delays          during the LIG (Clark et al., 2020), the British Ice Sheet during the last hydrofracturing (Golledge et al., 2019; Sadai et al., 2020; DeConto    termination (Gandy et al., 2018) and the Laurentide Ice Sheet during et al., 2021). The new studies are insufficient to change SROCCs low  the Younger Dryas (Pico et al., 2019), which show progressive retreat confidence assessment on ice-shelf loss. The consequence of this        despite declining temperatures, indicative of a true (ice dynamic) uncertainty on projections is discussed in Section 9.4.2.5 and Box 9.4. instability. Direct observational evidence of rapid paleo ice-sheet grounding line retreat is rare but, on the Larsen continental shelf, 9.4.2.4    Ice-sheet Instabilities                                    retreat rates of >10 km yr -1 during the deglaciation have been estimated (Dowdeswell et al., 2020). MISI has also been inferred from A major uncertainty in future Antarctic mass losses is the possibility  sedimentological evidence of ice loss from Wilkes Subglacial Basin, of rapid and/or irreversible ice losses through instability of marine  East Antarctica (Bertram et al., 2018; Wilson et al., 2018; Blackburn parts of the ice sheet, via the proposed mechanisms of marine ice      et al., 2020) but these reconstructions cannot unambiguously identify sheet instability (MISI) and marine ice cliff instability (MICI), and  unstable from progressive retreat. Therefore, there is limited evidence whether these processes will lead to a collapse of the West Antarctic  to identify the operation of instability mechanisms such as MISI Ice Sheet (WAIS).                                                      in paleo ice-sheet retreat.
MISI is a proposed self-reinforcing mechanism within marine ice        The SROCC assessed that ice-sheet interactions with the solid Earth sheets that lie on a bed that slopes down towards the interior of      are not expected to substantially slow sea level rise from marine-based the ice sheet, whereby, in the absence of ice-shelf buttressing,        ice in Antarctica over the 21st century (medium confidence), but the position of the grounding line is inherently unstable until        that these processes could become important on multi-century and reaching an upward sloping bed. The SROCC (Meredith et al., 2019)      longer time scales. More recent modelling of deglaciation of the Ross noted advances in modelling MISI since AR5, but that significant      Embayment by Lowry et al. (2020) is consistent with this assessment.
discrepancies remained in projections due to poor understanding        However, new projections for Pine Island Glacier (Kachuck et al.,
of mechanisms, and lack of observational data to constrain the          2020) support previous work (Barletta et al., 2018) suggesting that models. Since SROCC, modelling uncertainties have been more            lower mantle viscosity in this region leads to a negative feedback on thoroughly explored, rather than constrained (compatibility of          decadal time scales. Grounding line stabilization by the solid Earth current observations in the Amundsen Sea Embayment with MISI is        response may therefore occur over the 21st century in the Amundsen assessed in Section 9.4.2.1). Internal climate variability might either Sea Embayment, where most mass loss is occurring (Section 9.4.2.1),
slow (Hoffman et al., 2019) or amplify (Robel et al., 2019) MISI, and  but more generally occurs over multi-centennial to millennial time stable grounding line positions can be reached on downward sloping      scales (medium confidence).
beds if ice shelves provide buttressing (Sergienko and Wingham, 2019; Cornford et al., 2020). Ice-sheet model simulations that remove  The MICI hypothesis describes rapid, unmitigated calving triggered all Antarctic ice shelves (and prevent them from reforming) show        by ice-shelf collapse (Pollard et al., 2015). The SROCC noted that 2-10 m SLE Antarctic mass loss after 500 years due to MISI, of which    the MICI mechanism led one model (DeConto and Pollard, 2016)
WAIS collapse contributes 2-5 m (Sun et al., 2020), with the majority  to lose mass far more rapidly, but excluded the mechanism from of the mass loss in the first one to two centuries. Much of the multi-  its projections due to uncertainty in the timing of the ice-shelf model variation is due to the sliding law (Section 9.4.2.2). However,  disintegration (Section 9.4.2.3). They stated that MICI could lead it is not expected that widespread ice-shelf loss will occur before the to sea level contributions beyond 2100 considerably higher than end of the 21st century (Section 9.4.2.3; Box 9.4). A recent update    the likely range projected by other models. However, given the low of bed topography that unveiled large and overdeepened subglacial      agreement on the exact MICI mechanism and limited evidence of its 1269
 
Chapter 9                                                                                        Ocean, Cryosphere and Sea Level Change occurrence in the present or the past (Section 9.4.2.2), its potential  extensive exploration of uncertainties, and inclusion of a positive to affect future sea level rise was very uncertain. Since SROCC, new    feedback of meltwater on climate (Golledge et al., 2019). However, simulations show later ice-shelf disintegration, in agreement with      two of the projections did not include SMB changes that would other models (Section 9.4.2.3; DeConto et al., 2021), and therefore    offset dynamic losses (Levermann et al., 2014; Ritz et al., 2015),
lower projections at 2100 (Section 9.4.2.5). New theoretical            and the scenario dependence may have been further amplified by evidence suggests that ice-cliff collapse may only occur after very    highly sensitive sub-shelf melt parametrizations and use of simplified rapid ice shelf disintegration caused by unusually high meltwater      SMB schemes (Golledge et al., 2015, 2019; Bulthuis et al., 2019; production (Clerc et al., 2019; Robel and Banwell, 2019), and that      Oppenheimer et al., 2019).
the subsequent rate of retreat depends on the terminus geometry (Bassis and Ultee, 2019). As SROCC noted, only Crane Glacier on        Since SROCC, new projections have arisen from multi-model the Peninsula has shown retreat consistent with MICI, after the        intercomparison projects ISMIP6 and LARMIP-2 (Box 9.3) and one Larsen B ice shelf collapsed, and MICI-style behaviour at Jakobshavn    model that includes MICI (Section 9.4.2.4; Table 9.3; DeConto et al.,
and Helheim Glaciers in Greenland might not be representative of        2021). Corrections are added to allow comparison: all ISMIP6-derived wider Antarctic glaciers. Observations from Greenland show that        projections have an estimate of the historical dynamical response 9 steep cliffs commonly evolve into short floating extensions, rather    to pre-2015 climate forcing added, which increases contributions than collapsing catastrophically (Joughin et al., 2020). As assessed    (Box 9.3; Figure 9.18); the LARMIP-2 dynamic projections are in Section 9.4.2.2 and 9.4.2.3, there is therefore low confidence      combined with an estimate of SMB, which decreases contributions in simulating mechanisms that have the potential to cause              (Sections 9.4.2.3 and 9.6.3.2); and the ISMIP6 emulated and LARMIP-2 widespread, sustained and very rapid ice loss from Antarctica this      projections were re-estimate using the global surface air temperature century through MICI, and low confidence in projecting the driver of    distributions from the two-layer energy budget emulator described in ice-shelf disintegration.                                              Supplementary Material 7.SM.2. The majority of the new projections indicate that, under all emissions scenarios, the AIS will lose mass In summary, poorly understood processes of instabilities, characterized overall and contribute to sea level rise. Most thinning occurs in the by deep uncertainty, have the potential to strongly increase Antarctic  Amundsen Sea sector in WAIS and Totten Glacier in EAIS (Figure 9.18).
mass loss under high greenhouse gas emissions on century-to-            The most negative contribution is -0.02 m (5th percentile of ISMIP6 multicentury time scales (Box 9.4). These instabilities are therefore  combined RCP8.5 and SSP58.5 projections after correction) and considered separately in assessments of the future contribution        the largest contribution is 0.57 m SLE (95th percentile; Levermann to global mean sea level (GMSL; Sections 9.4.2.5, 9.4.2.6, 9.6.3.2      et al., 2020), or 0.63 m SLE with MICI (95th percentile; DeConto et al.,
and 9.6.3.5).                                                          2021). ISMIP6 ensemble ranges are wider for the high scenarios (RCP8.5/SSP58.5) than the low (RCP2.6/SSP12.6), in part because 9.4.2.5      Projections to 2100                                        more simulations were available. The ISMIP6 simulations that apply an ice-shelf collapse scenario based on exceedance of a surface The AR5 assessed the median and likely (66-100% probability) sea        meltwater threshold (Trusel et al., 2015), driven by CMIP5 models, level contributions of the AIS in 2100 relative to 1986-2005 to be      show only a small increase in mass loss (around 0-0.04 m), mostly 0.06 (-0.04 to +0.16) m SLE under RCP2.6 and 0.04 (-0.08 to +0.14) m    from the Peninsula, due in part to the small number of ice shelves SLE under RCP8.5 (Table 9.3; no change when using the AR6 baseline). predicted to collapse this century (Seroussi et al., 2020). Simulations The AR5 stated that only the collapse of the marine-based sectors of    driven by the CMIP5 model HadGEM2-ES, which has unusually the AIS, if initiated, could cause GMSL to rise substantially above the extreme warming in the Ross Sea (Barthel et al., 2020), show a larger likely range during the 21st century, with medium confidence that      mass loss (up to about 0.05 m) in East Antarctica under ice-shelf this would not exceed several tenths of a metre during this period. collapse (Edwards et al., 2021). The ISMIP6 projections do not include The assessment of the dynamical contribution had no dependence on      the efficient meltwater drainage or atmospheric feedbacks that could emissions scenarios, due to the lack of literature, so the decrease in  reduce mass loss further (Seroussi et al., 2020).
sea level contribution in the higher-emissions scenario was solely due to increased SMB (Section 9.4.2.3). The SROCC (Oppenheimer et al.,      The relationship between emissions scenario and AIS response 2019) assessed the total contribution based on five new ice-sheet      varies across the studies, with emulated ISMIP6 projections showing modelling studies that incorporated marine ice-sheet dynamics,          a slight negative scenario dependence in the median (-0.01 m) from combining their estimates and interpreting the 5-95th percentile        SSP12.6 to SSP58.5, and LARMIP-2-based projections showing range of the resulting distribution as the likely range (17-83%        a slight positive scenario-dependence in the median (0.02 m; probability interval, i.e., not open-ended as in the AR5). The median  Table 9.3). A lack of clear scenario dependence in the median masks and likely range contributions by 2100 were 0.04 (0.01-0.11) m          large individual variations across climate and ice-sheet models, under RCP2.6 and 0.12 (0.03-0.28) m under RCP8.5 (Table 9.3).          whereby the net AIS contribution response to emissions scenario The positive scenario-dependence in SROCC - where increases in          depends on the relative magnitudes of the atmosphere, ocean dynamic losses driven by ocean warming and ice-shelf disintegration    and ice-sheet responses (Barthel et al., 2020; Seroussi et al., 2020; under higher emissions (Section 9.4.2.3) dominate over increases in    Edwards et al., 2021). Climate and ice-sheet models do not project SMB - arose from a combination of physical processes and model          that the AIS response will be the same under high or low greenhouse limitations. Modelling improvements in these studies included          gas emissions in 2100; rather, there is no consensus on the sign of improved representations of grounding line response to drivers, more    the change. In contrast, strong scenario dependence is seen from 1270
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                  Chapter 9 Table 9.3 l Projected sea level contributions in metres from the Antarctic Ice Sheet in 2100 relative to 1995-2014, unless otherwise stated, for selected Representative Concentration Pathway (RCP) and Shared Socio-economic Pathways (SSP) scenarios. Italics denote partial contributions. The historical dynamic response omitted from ISMIP6 simulations is estimated to be 0.33 +/- 0.16 mm yr -1 (0.03 m +/- 0.01 m in 2100 relative to 2015; Box 9.3). The climate forcing is described in Supplementary Material 7.SM.2.
Representative Concentration Pathways (RCPs)
Study                              RCP2.6                  RCP4.5                    RCP8.5                                  Notes IPCC AR5 (Church et al., 2013b)              0.06 (-0.04 to +0.16)    0.05 (-0.05 to +0.15)    0.04 (-0.08 to +0.14)  Median and likely ( 66% range) contribution Median and likely (66% range) contribution.
IPCC SROCC (Oppenheimer et al., 2019)          0.04 (0.01 to 0.11)      0.06 (0.01 to 0.15)      0.12 (0.03 to 0.28)
Combination of five studies ISMIP6 CMIP5-forced (Seroussi et al., 2020);                                                                          Range of ISMIP6 multi-model contributions in 2100 relative
                                                  -0.01 to +0.16                  -                -0.08 to +0.30 excludes historical dynamic response                                                                                  to 2015 from 2 ESMs for RCP2.6 and 6 ESMs for RCP8.5 LARMIP-2; excludes surface mass balance        0.13 (0.07 to 0.24)      0.14 (0.07 to 0.28)      0.17 (0.09 to 0.36)  Median (67% range) [90% range] LARMIP-2 multi-model (SMB) (Levermann et al., 2020)                    [0.04 to 0.37]          [0.05 to 0.44]            [0.06 to 0.58]  dynamic contribution in 2100 relative to 1900 0.08 (0.06 to 0.12)      0.09 (0.07 to 0.11)      0.34 (0.19 to 0.53)
MICI (DeConto et al., 2021)                                                                                            Median (66% range) [90% range]
[0.06 to 0.15]          [0.07 to 0.15]            [0.11 to 0.63]
9 Shared Socio-economic Pathways (SSPs)
Study                            SSP12.6                SSP24.5                  SSP58.5                                Notes Multi-model ensemble projections ISMIP6 CMIP6-forced (Payne et al., 2021);                                                                              Range of ISMIP6 multi-model contributions in 2100 relative
                                                  -0.05 to +0.01                  -                -0.09 to +0.11 excludes historical dynamic response                                                                                  to 2015 from 1 ESM for SSP12.6 and 4 ESMs for SSP58.5 Median (66% range) [90% range] contribution from ISMIP6 all (CMIP5 and CMIP6-forced)          -0.05 (0.04 to 0.08)                                0.04 (0.00 to 0.12)
                                                                                    -                                    ISMIP6 CMIP5 and CMIP5-forced multi-model ensembles, including historical dynamic response              [0.03 to 0.11]                                  [-0.02 to +0.23]
(see caption)
Median (66% range) [90% range] contribution in 2100 Emulated ISMIP6; excludes historical        0.04 (-0.01 to +0.10)    0.04 (-0.02 to +0.10)    0.04 (-0.01 to +0.09) relative to 2015 from emulator of ISMIP6 used with dynamic response (Edwards et al., 2021)        [-0.05 to +0.14]        [-0.06 to +0.14]          [-0.05 to +0.14]
Chapter 7 climate forcing 0.09 (0.03 to 0.14)      0.09 (0.03 to 0.14)      0.08 (0.03 to 0.14)  Emulated ISMIP6, but relative to 1995-2014 and Emulated ISMIP6 total
[-0.01 to +0.19]        [-0.01 to +0.18]            [0.00 to 0.18]    including historical dynamic response (see caption)
                                              -0.02 (-0.03 to -0.01)  -0.03 (-0.04 to -0.02)    -0.05 (-0.07 to -0.03) Median (66% range) [90% range] SMB estimated for the SMB
[-0.04 to -0.01]        [-0.06 to -0.01]          [-0.09 to -0.02]  AR5, used to correct LARMIP-2 below Median (66% range) [90% range] dynamic contribution 0.15 (0.08 to 0.29)      0.17 (0.09 to 0.33)      0.20 (0.10 to 0.39)
LARMIP-2; excludes SMB                                                                                                from LARMIP-2 multi-model method used with Chapter 7
[0.05 to 0.44]          [0.06 to 0.49]            [0.07 to 0.61]
climate forcing 0.14 (0.08 to 0.26)      0.15 (0.08 to 0.29)      0.17 (0.10 to 0.35)  As above, but using only the 13 of 16 ice-sheet models LARMIP-2 subset of models; excludes SMB
[0.05 to 0.39]          [0.05 to 0.45]            [0.06 to 0.54]  common to both ISMIP6 and LARMIP-2 0.11 (0.05 to 0.24)      0.12 (0.05 to 0.26)      0.12 (0.05 to 0.30)
LARMIP-2 subset of models; includes SMB                                                                                As above, but including the SMB estimate
[0.03 to 0.37]          [0.02 to 0.42]            [0.01 to 0.49]
Median (66% range) [90% range] dynamic contribution 0.13 (0.06 to 0.27)      0.14 (0.06 to 0.29)      0.15 (0.05 to 0.34)
LARMIP-2 total                                                                                                        from LARMIP-2 multi-model method used with
[0.03 to 0.41]          [0.02 to 0.46]            [0.01 to 0.57]
Chapter 7 climate forcing, including the SMB estimate This assessment: combination of              0.11 (0.03 to 0.27)      0.11 (0.03 to 0.29)      0.12 (0.03 to 0.34)  Median (66% range) [90% range] assessment emulated ISMIP6 and LARMIP-2                  [-0.01 to +0.41]        [-0.01 to +0.46]            [0.00 to 0.57]    combining emulated ISMIP6 and LARMIP-2 RCP4.5 to RCP8.5 in projections that allow MICI (Section 9.4.2.4;                            The LARMIP-2 median projections are higher than those of the ISMIP6 DeConto et al., 2021), though less so than earlier projections                                emulator (by 0.04-0.07 m), and the 95th percentiles are two to three (DeConto and Pollard, 2016) due to later ice-shelf disintegrations.                          times higher. Two possible reasons for the differences between the A negative or positive scenario dependence of the AIS response this                          emulated ISMIP6 and LARMIP-2 projections are assessed: the set of century cannot be deduced from recent observations, because there                            ice-sheet models (Annex II) and the parameter values determining sub-is still low confidence in attributing the causes of observed mass                            shelf melt sensitivity to ocean temperature (Section 9.4.2.3; Box 9.3).
loss (Section 9.4.2.1), and neither regional mass increases by SMB                            Using only the 13 ice-sheet models common to ISMIP6 and LARMIP-2 nor regional mass losses by ice flow have a linear relationship with                          reduces the LARMIP-2 median projections by 0.02-0.03 m SLE and the global mean temperature (Sections 9.4.2.1, 9.4.2.2, 9.4.2.3). There                          95th percentiles by 0.04-0.08 m SLE (Table 9.3). This approximately is therefore low agreement on the relationship between emissions                              halves the difference in medians, but has a relatively small effect on scenario and AIS response. However, in the longer term, mass loss is                          the upper end. Sub-shelf melt sensitivity has a larger effect, due to the expected to dominate (Section 9.4.2.6).                                                      wide variation of estimates from different regions and methods. Using only the Pine Island Glacier sub-shelf melt distribution (Sections 9.4.2.2 1271
 
Chapter 9                                                                                        Ocean, Cryosphere and Sea Level Change and 9.4.2.3) in the ISMIP6 emulator gives a median Antarctic projection noted that deep uncertainty remained beyond 2100: while solid of about 0.08 m in 2100 in all scenarios before historical correction,  Earth feedbacks could reduce ice loss over multi-century time scales, compared with around 0 m using only the mean Antarctic distribution;    MICI (Section 9.4.2.4) might give contributions higher than the likely the published projections use a joint distribution (Edwards et al.,    ranges. The SROCC also presented structured expert judgement 2021). Reese et al. (2020) find that using the basal melt sensitivities (SEJ) projections for comparison (Bamber et al., 2019), which give of LARMIP-2 yields an order of magnitude greater mass loss under        higher values. Since SROCC, three studies have made projections RCP8.5 than with the ISMIP6 mean Antarctic values. Halving the basal    to 2300: (i) Rodehacke et al. (2020) assessed two methods for melt sensitivity parameter range (i.e., in line with a continental mean implementing precipitation changes (based on repeating 2071-2100 estimate: Section 9.4.2.3) would lead to a halving of the LARMIP-2      forcings beyond 2100), which both gave negative projections at dynamic contribution. This would reconcile the LARMIP-2 and ISMIP6      2300 because the dynamic response was very small (-0.11 to emulator median and 95th percentile projections using the common        -0.01 m SLE for RCP2.6; -0.25 to -0.07 m for RCP8.5 forcing); (ii) In subset of models within about 0.02-0.05 m. There is therefore limited  contrast, simulations forced by 2081-2100 ocean-only projections evidence that the ISMIP6 and LARMIP-2 projections could be reconciled  under RCP8.5/SSP58.5 beyond 2100, using two implementations by using common ice-sheet models and basal melt sensitivity values. of the ISMIP6 non-local basal melt parametrizations (Box 9.3 and 9                                                                        Section 9.4.2.2) and two sliding laws, are all positive (0.08 m to 0.96 m It is not possible to distinguish which of ISMIP6 and LARMIP-2 is      SLE by 2300), though these do not include the negative contribution more realistic, due to limitations in historical simulations (Box 9.3)  from SMB changes (Lipscomb et al., 2021); (iii) Finally, DeConto et al.
and understanding of basal melting (Section 9.4.2.3.2), so the          (2021) update projections for the MICI hypothesis (Section 9.4.2.4) projections are combined using a p-box approach (Section 9.6.3.2). using the extensions of the RCPs to 2300, and obtain far higher The mean of the ISMIP6 emulated and LARMIP-2 medians gives the          contributions: median (17-83%) ranges of 1.09 (0.71-1.35) m SLE assessed median projections, and the outer edges of the 17-83%          under RCP2.6 and 9.60 (6.87-13.54) m SLE under RCP8.5. These ranges give the outer edges of the assessed likely (17-83%) ranges -    are larger than previous estimates (DeConto and Pollard, 2016),
that is, encompassing the structural and parametric uncertainties      particularly at the upper end: 0.68 (0.29-1.13) m SLE for RCP2.6 of both methods, giving medium confidence in their combined            and 8.40 (7.47-9.76) m for RCP8.5 (Edwards et al., 2019), which can projections. The main difference between this assessment and SROCC      largely be explained by the higher maximum ice cliff calving rate.
is to increase the medians of the lower scenarios by 0.05-0.07 m,      LARMIP-2 dynamic projections (Box 9.3) are also estimated under so that all SSPs are similar to SROCC assessment of RCP8.5, and        the extended SSPs and corrected with SMB (as in Section 9.4.2.5),
to substantially increase the upper ends of the likely ranges: by      giving median (17-83%) ranges of 0.40 (0.18-0.78) m SLE at 0.14-0.16 m for RCP2.6/SSP12.6 and RCP4.5/SSP24.5, and 0.06 m        2300 under SSP12.6 and 1.57 (0.68-3.14) m under SSP58.5. The for RCP8.5/SSP58.5. The increase relative to SROCC is partly due to    longer time scale may invalidate the linear response assumption of the increase in LARMIP-2 projections relative to the original LARMIP    LARMIP-2, which neglects any self-dampening or self-amplifying study (Levermann et al., 2014), arising from the larger number of      processes. The ranges of projections for 2300 without MICI (Golledge participating ice-sheet models (Levermann et al., 2020). The historical et al., 2015; Bulthuis et al., 2019; Levermann et al., 2020; Rodehacke dynamic response to pre-2015 climate forcing applied to the ISMIP6      et al., 2020; Lipscomb et al., 2021; assessed ice-sheet contributions emulator could be overestimated, due to the assumption of a constant    in Section 9.6.3.5 are -0.14 to +0.78 m SLE under RCP2.6/SSP12.6, future rate (Box 9.3). This assessment encompasses SROCC and all        and -0.27 to 3.14 m SLE under RCP8.5/SSP58.5). The lower bounds projections since, except the 83rd percentiles of projections that      are the 5th percentile of Bulthuis et al. (2019) and the lowest mean/
allow MICI under RCP8.5 (DeConto et al., 2021) and the Structured      median from Rodehacke et al. (2020), respectively; the upper bounds Expert Judgement (SEJ) under 5&deg;C shown in SROCC (Bamber et al.,        are the 83% percentiles of the LARMIP-2 estimates. These ranges 2019). Both are used in further p-box estimates to give the outer      are wider than SROCC likely ranges, and more consistent with the limits of low confidence assessments (Section 9.6.3.2).                SEJ (Bamber et al., 2019). However, projections in which Antarctica contributes much more than the assessed ranges under sustained In summary, it is likely that the AIS will continue to lose mass        very high greenhouse gas emissions - that is, around 7-14 m to GMSL throughout this century under all emissions scenarios - that is,        by 2300 (DeConto et al., 2021), cannot be ruled out, and are taken dynamic losses driven by ocean warming and ice-shelf disintegration    as a sensitivity case (Section 9.6.3.5; Table 9.11). In summary, there will likely continue to outpace increasing snowfall (medium            is high confidence that Antarctic mass loss will be greater beyond confidence). The upper end of projections is not well constrained,      2100 under high greenhouse gas emissions, but the large range due to different assumptions about the future sensitivity of sub-shelf  of projections mean we have only low confidence in the likely AIS basal melting to ocean warming and the proposed marine ice cliff        contribution to GMSL by 2300 for a given scenario. Deep uncertainty instability triggered by ice-shelf disintegration (Sections 9.4.2.3    remains in the role of AIS instabilities under very high emissions.
and 9.4.2.4; Box 9.4).
The West and East Antarctic ice sheets are considered to be tipping 9.4.2.6    Projections Beyond 2100                                    elements - that is, susceptible to critical thresholds. The SR1.5 (Hoegh-Guldberg et al., 2018) assessed that a threshold for WAIS The SROCC assessed the median and likely range of Antarctic SLE        instability may be close to 1.5&deg;C-2&deg;C (medium confidence), as only contributions at 2300 as 0.16 (0.07-0.37) m under RCP2.6 and            RCP2.6 led to long-term projections of less than 1 m (Golledge 1.46 (0.60-2.89) m under RCP8.5, based on three studies. It was        et al., 2015; DeConto and Pollard, 2016). Based on the agreement 1272
 
Ocean, Cryosphere and Sea Level Change                                                                                              Chapter 9 of a further study (Bulthuis et al., 2019), SROCC confirmed that        sheet reorganizes to a new state, leading to large losses from East low emissions would limit Antarctic ice loss over multi-century          Antarctica (including the Aurora Subglacial Basin) and leading to time scales (high confidence), but it was not possible to determine      a further 10 m sea level contribution per degree of warming; other whether this was sufficient to prevent substantial ice loss (medium      studies also show much higher mass loss per &deg;C at higher levels of confidence). Since SROCC, new studies have revisited this topic          warming (Section 9.6.3.5 and Figure 9.30; Van Breedam et al., 2020; (Garbe et al., 2020; Rodehacke et al., 2020; Van Breedam et al.,        DeConto et al., 2021).
2020; DeConto et al., 2021; Lipscomb et al., 2021), allowing a more complete assessment along with other studies (Feldmann and              The SROCC (Meredith et al., 2019; Oppenheimer et al., 2019) assessed Levermann, 2015; Clark et al., 2016; Golledge et al., 2017a; Edwards    that Antarctic mass losses could be irreversible over decades to et al., 2019) and the extension to LARMIP-2 above. The majority          millennia (low confidence). Garbe et al. (2020) show that the AIS project 0-1.3 m SLE on multi-century time scales under scenarios of      is always volumetrically smaller when regrowing under a given 1&deg;C-2&deg;C warming. Projections can increase up to 2 m SLE under high      warming level than when it retreats under the same forcing. Even if basal melt sensitivity to ocean warming (Section 9.4.2.3; Lipscomb      retreat followed by regrowth results in a net zero change in volume, et al., 2021) or MICI (Section 9.4.2.4). On multi-millennial time scales the spatial distribution of mass may be altered, especially in parts of (2,000 years), many projections remain below 1.6 m SLE under            West Antarctica vulnerable to MISI. Projections that start reducing      9 1&deg;C-2&deg;C warming - that is, less than about half of the WAIS in SLE      CO2 concentrations from 2030 onwards, reaching pre-industrial (see also Section 9.6.3.5 and Figure 9.30). Other studies project        levels around 2300, show sea level contributions exceeding 1 m majority or total loss of WAIS under 1&deg;C-2&deg;C warming, exceeding          by 2500 when including MICI (DeConto et al., 2021). New research 2 m SLE, under the higher end of the warming range (1.5&deg;C), or high    therefore confirms SROCC assessment that mass loss from the AIS is ocean warming (0.5&deg;C) and/or high basal melting around WAIS, or        irreversible on decadal to millennial time scales (low confidence) (FAQ MICI. All but two of these multi-millennial studies use variants of      9.1), and suggests that reducing atmospheric CO2 concentrations or the same ice-sheet model, though different modelling choices mean        temperatures to pre-industrial levels may not be sufficient to prevent they can be considered quasi-independent. Simulations of previous        or reverse substantial Antarctic mass losses (low confidence).
interglacial periods often show near or total WAIS disintegration, with mass loss exceeding 3 m SLE (e.g. Figure 9.18), although limitations of these studies or inferences that can be drawn under different        9.5        Glaciers, Permafrost and Seasonal forcings limit confidence in the robustness of these as quantitative                Snow Cover analogues (Sections 9.4.2.4 and 9.6.2). Overall, increased evidence and agreement on the time scales and drivers of mass loss confirm        9.5.1      Glaciers the SR1.5 assessment that a threshold for WAIS instability may be close to 1.5&deg;C-2&deg;C (medium confidence), and that the probability        9.5.1.1    Observed and Reconstructed Glacier Extent of passing a threshold is larger for 2&deg;C warming than for 1.5&deg;C                      and Mass Changes (medium confidence), particularly under strong ocean warming. New projections agree with previous studies that only part of WAIS would    9.5.1.1.1 Global glacier contribution be lost on multi-century time scales if warming remains less than 2&deg;C (medium confidence). There is limited agreement about whether        The IPCCs Fifth Assessment Report (AR5; Vaughan et al., 2013) complete disintegration would eventually occur at this level of          assessed glacier changes from studies based on the regions defined warming, but medium confidence this would take millennia.                in the Randolph Glacier Inventory (RGI; RGI version 2.0): a satellite observation-based, global inventory of glacier outlines for the Under around 2&deg;C-3&deg;C peak warming, complete or near-complete            year 2000. Following Special Report on the Ocean and Cryosphere loss of the WAIS is projected in most studies after multiple millennia  in a Changing Climate (SROCC; Hock et al., 2019b; Meredith (low confidence), with continent-wide mass losses of around 2-5 m        et al., 2019), we report on studies based on RGI version 6.0 (RGI SLE or more; this could occur on multi-century time scales under very    Consortium, 2017). Increased volume of satellite observations and high basal melting (Lipscomb et al., 2021) or widespread ice-shelf      the inclusion of detailed regional glacier inventories has resulted in loss and/or MICI (low confidence) (Sun et al., 2020; DeConto et al.,    an improved inventory (RGI Consortium, 2017). A new consensus 2021). Mass losses under around 2&deg;C-3&deg;C warming could be less            estimate for the ice thickness distribution of all glaciers in RGI 6.0 than 2 m SLE, particularly for multi-century time scales, low basal      was obtained from an ensemble of five numerical models. However, melting, or less responsive sliding laws. If warming exceeds around      only one out of five models covered all regions (Farinotti et al., 2019),
3&deg;C above pre-industrial, part of the EAIS (typically the Wilkes        and was, where possible, calibrated and validated with the worldwide Subglacial Basin) is projected to be lost on multi-millennial time      Glacier Thickness Database (GlaThiDa 3.0: GlaThiDa Consortium, scales (low confidence), with total AIS mass loss equivalent to around  2019; Welty et al., 2020). The updated inventory shows decreases 6-12 m or more sea level rise; mass loss could be much smaller if the    in estimated glacier volume in the Arctic, High Mountain Asia and dynamic response is small (Bulthuis et al., 2019; Rodehacke et al.,      Southern Andes, partially compensated by increases in Antarctica.
2020), or much faster under widespread ice-shelf loss and/or MICI        15% of the total glacier volume is estimated to be below sea level (Sun et al., 2020; DeConto et al., 2021). A study by Garbe et al. (2020) and would not contribute to sea level rise if melted (Farinotti et al.,
suggests that 6&deg;C sustained warming and associated mass loss of          2019). Supplementary Material Table 9.SM.2 shows the inventory about 12 m SLE may be a critical threshold beyond which the ice          glacier area and mass for each region in the year 2000.
1273
 
Chapter 9                                                                                                                            Ocean, Cryosphere and Sea Level Change Global (all)              Global (excep 5&19)                Alaska (1)                West Canada and U.S. (2) 500                            500                            2000                                2000 1000                                1000 0                              0 0                                  0 1000                            1000 500                            500 2000                            2000 1000                          1000                          3000                            3000 Arctic Canada (N) (3)          Arctic Canada (S) (4)      Greenland periphery (5)                    Iceland (6) 2000                          2000                            2000                                2000 1000                          1000                            1000                                1000 9                                              0                              0                              0                                  0 1000                          1000                          1000                            1000 2000                          2000                          2000                            2000 3000                          3000                          3000                            3000 Svalbard (7)                  Scandinavia (8)              Russian Arctic (9)                  North Asia (10) 2000                          2000                            2000                                2000 Mass change rate (kg m-2 yr-1) 1000                          1000                            1000                                1000 0                              0                              0                                  0 1000                          1000                          1000                            1000 2000                          2000                          2000                            2000 3000                          3000                          3000                            3000 Central Europe (1 1)              Caucasus (12)          High Mountain Asia (1315)              Low Latitudes (16) 2000                          2000                            2000                                2000 1000                          1000                            1000                                1000 0                              0                              0                                  0 1000                          1000                          1000                            1000 2000                          2000                          2000                            2000 3000                          3000                          3000                            3000 1960    1980    2000      2020 Southern Andes (17)              New Zealand (18)          Antarctic periphery . (19)                  Year s 2000                          2000                            2000 1000                          1000                            1000 Zemp et al. (2019/20) 0                              0                              0                                          Ciraci et al. (2020)
SROCC 1000                          1000                          1000 Hugonnet et al. (2021) 2000                          2000                          2000                                        Regional estimates ()
3000                          3000                          3000 1960    1980  2000    2020 1960        1980  2000    2020 1960      1980  2000      2020 Figure 9.20 l Global and regional glacier mass change rate between 1960 and 2019. The time series of annual and decadal mean mass change are based on glaciological and geodetic balances (Zemp et al., 2019, 2020). Superimposed are the 2002-2019 average rates by (Cirac et al., 2020) based on the Gravity Recovery and Climate Experiment (GRACE), 2006-2015 estimated rates as assessed in Special Report on Ocean and Cryosphere in a Changing Climate (SROCC) and the new decadal averages (2000-2009 and 2010-2019) by Hugonnet et al. (2021).
* New regional estimates for the Andes (Dussaillant et al., 2019), High Mountain Asia (Shean et al., 2020),
Iceland (Aalgeirsd&#xf3;ttir et al., 2020), Central Europe (Sommer et al., 2020) and Svalbard (Schuler et al., 2020) are also shown. The uncertainty reported in each study is shown.
See Figure 9.2 for the location of each region. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1274
 
Ocean, Cryosphere and Sea Level Change                                                                                                              Chapter 9 The SROCC found a globally coherent trend of glacier decline in                          9.5.1.1.2 Regional glacier changes the last decades, despite large annual variability and regional differences (very high confidence). Section 2.3.2.3 assesses the                        A major advance since SROCC is the availability of high-accuracy global glacier mass changes for the whole 20th century (see                              mass loss estimates for individual glaciers (Hugonnet et al., 2021).
Table 9.5 for contribution to the sea level budget. Note that the                        These results show that, during the last 20 years, the highest regional peripheral glaciers in Greenland and Antarctica are added to the ice                    mass loss rates (>720 kg m-2 yr -1) were observed in the Southern sheets for the budget). The AR6 assessment is based on Marzeion                          Andes, New Zealand, Alaska, Central Europe, and Iceland. Meanwhile, et al. (2015), using glacier-length reconstructions (Leclercq et al.,                    the lowest regional mass loss rates (<250 kg m-2 yr -1) were observed 2011) and a glacier model forced by gridded climate observations                        in High Mountain Asia, the Russian Arctic, and the periphery of (Marzeion et al., 2012), and not considering the estimated mass                          Antarctica. Glacier mass loss in Alaska (25% of 2000-2019 total loss of uncharted glaciers (100 +/- 50 Gt yr -1; Parkes and Marzeion,                      mass loss), the periphery of Greenland (13%), Arctic Canada North 2018). The time series are assumed independent, resulting in larger                      (11%), Arctic Canada South (10%), the periphery of Antarctica (8%),
uncertainty than presented in SROCC (see also Section 9.6.1). The rate                  the Southern Andes (8%) and High Mountain Asia (8%), represent of global glacier mass loss (excluding the periphery of ice sheets) for                  the majority (83%) of the total glacier mass loss during the last the period 1901-1990 is estimated to be very likely 210 +/- 90 Gt yr -1,                  20 years (2000-2019).                                                  9 representing 16 [28 to 7] % of the glacier mass in 1901, in agreement with SROCC within uncertainty estimates.                                                The glacier mass loss rate from geodetic mass balance assessments in the Southern Andes during 2006-2015 was smaller Since SROCC, new regional estimates for the Andes (Dussaillant                          (720 +/- 70 kg m-2 yr -1; Braun et al., 2019; Dussaillant et al.,
et al., 2019), High Mountain Asia (Shean et al., 2020), Iceland                          2019; Hugonnet et al., 2021) than previously assessed in SROCC (Aalgeirsd&#xf3;ttir et al., 2020), the European Alps (Davaze et al., 2020;                  (860 +/- 160 kg m-2 yr -1), though within uncertainties. In the Central Sommer et al., 2020) and Svalbard (Schuler et al., 2020), two new                        and Desert regions of the Southern Andes, an increase in mass loss global (Cirac et al., 2020; Hugonnet et al., 2021) and an ad hoc                        from 2000-2009 to 2010-2018, and a high loss rate in Patagonia for estimate for the latest glaciological observations (Zemp et al., 2020)                  the whole period, are observed (Dussaillant et al., 2019). Records of have extended the glacier mass change time series up to 2018-2019                        glacier mass loss in Peru (Seehaus et al., 2019a) and Bolivia (Seehaus (Figure 9.21 and Supplementary Material Table 9.SM.3). A reconciled                      et al., 2019b) in the period 2000-2016 show an increase in mass global estimate for the period 1962-2019 has been compiled by Slater                    loss towards the end of the observation period. In western North et al. (2021). However, in contrast to Slater et al. (2021), after 2000                  America, outside of Alaska and western Yukon, there was a fourfold this assessment is based on the first globally complete and consistent                  increase in mass loss for 2009-2018 (860 +/- 320 kg m-2 yr -1) estimate of 21st-century glacier mass change from differencing                          compared to 2000-2009 (203 +/- 214 kg m-2 yr -1; Menounos et al.,
of digital elevation models (Hugonnet et al., 2021) covering                            2019), and in the Canadian Arctic there was a doubling of mass 94.7% of glacier area with glacier mass change for each glacier in                      loss in the last two decades compared with pre-1996 (Nol et al.,
the inventory produced with unprecedented accuracy. The estimates                        2018; Cook et al., 2019). The peripheral glaciers in NE Greenland from Hugonnet et al. (2021) agree within uncertainties with new and                      experienced a 23% increase in mass loss in 1980-2014 compared previous estimates at global (Hock et al., 2019b; Wouters et al., 2019;                  to the period 1910 to 1978-1987 (Carrivick et al., 2019). In Iceland, Zemp et al., 2019; Cirac et al., 2020; Slater et al., 2021) and regional                16 +/- 4% of the around 1890 glacier mass has been lost; about half scale (Dussaillant et al., 2019; Aalgeirsd&#xf3;ttir et al., 2020; Schuler                  of that loss occurred in the period 1994-2019 (Aalgeirsd&#xf3;ttir et al.,
et al., 2020; Shean et al., 2020). Excluding peripheral glaciers of ice                  2020). Glacier records starting in 1960 in Norway show that half of sheets (RGI regions 5 and 19), glacier mass loss rate was very likely                    the observed glaciers advanced in the 1990s but all have retreated 170 +/- 80 Gt yr -1 for the period 1971 to 2019 (8 [4 to 14] % of 1971                    since 2000 (Andreassen et al., 2020). In Svalbard, glaciers have glacier mass), 210 +/- 50 Gt yr -1 over the period 1993-2019 (6 [4 to                      been losing mass since the 1960s, with a tendency towards more 8] % of 1993 glacier mass) and 240 +/- 40 Gt yr -1 over the period                        negative mass balance since 2000 (Deschamps-Berger et al., 2019; 2006-2019 (3 [2 to 4] % of 2006 glacier mass; Sections 2.3.2.3                          Van Pelt et al., 2019; Morris et al., 2020; Nol et al., 2020; Schuler and 9.6.1, Table 9.5,4 and Cross-Chapter Box 9.1). Including the                        et al., 2020). A similar increase in mass loss has been observed for peripheral glaciers of the ice sheets, the global glacier mass loss rate                Franz Josef Land in the Russian Arctic (Zheng et al., 2018). Rapid in the period 2000-2019 is very likely 266 +/- 16 Gt yr -1 (4 [3 to 6] %                  retreat and downwasting throughout the European Alps in the of glacier mass in 2000) with an increase in the mass loss rate from                    early 21st century is reported (Sommer et al., 2020) and long-term 240 +/- 9 Gt yr -1 in 2000-2009 to 290 +/- 10 Gt yr -1 in 2010-2019 (high                    records, although limited, indicate sustained glacier mass loss in confidence). These estimates are in agreement with SROCC estimate                        High Mountain Asia since around 1850, with increased mass loss and extend the period to 2018-2019. In summary, new evidence                            in recent decades (Shean et al., 2020). In summary, although published since SROCC shows that, during the decade 2010-2019,                          interannual variability is high in many regions, glacier mass records glaciers lost more mass than in any other decade since the beginning                    throughout the world show with very high confidence that the of the observational record (very high confidence) (Section 8.3.1.7.1                    loss rate has been increasing in the last two decades (see also and Figure 9.20).                                                                        Section 8.3.1.7.1 and 12.4 for regional glacier assessment).
4    The periods in Table 9.5 end in 2018, leading to a slight difference in the values.
1275
 
Chapter 9                                                                                              Ocean, Cryosphere and Sea Level Change Section 2.3.2.3 assesses that the rate and global character of glacier      last glacial period and the Holocene (e.g., Solomina et al., 2015, 2016; retreat in the latter part of 20th century, and finds that the first decades Eaves et al., 2019; Hall et al., 2019; Marcott et al., 2019; Bohleber et al.,
of the 21st century appear to be unusual in the context of the Holocene      2020; Davies et al., 2020; Palacios et al., 2020) confirm the dominant (medium confidence) and the global glacier recession in the beginning        role of orbital forcing for millennial-scale glacier fluctuations, but of the 21st century to be unprecedented in the last 2000 years              emphasize the role of other forcings - solar and volcanic activity, ocean (medium confidence). These assessments are supported by regional            circulation, sea ice and internal climate variability - in explaining the evidence. New reconstructions of the Patagonian Ice Sheet suggest            regional variability of glacier fluctuations at shorter time scales. Shakun that 20th-century glacial recession occurred faster than at any time        et al. (2015) demonstrated that, during the last deglacial transition during the Holocene (Davies et al., 2020). The reconstructions of glacier    (18-11 ka), the mid-to-low-latitude glacier retreat was driven by an variations show that the glaciers in some regions are now smaller            increase in atmospheric CO2 and global temperature.
than previously recorded: since the mid-16th century in the Mont Blanc and Grindelwald regions of the European Alps (Nussbaumer and          In the Northern Hemisphere, where summer insolation decreased Zumb&#xfc;hl, 2012), since the 9th century in Norway (Nesje et al., 2012),        during the Holocene (Section 2.2.1), glaciers generally waxed (Briner and for the past 1800 years in north-west Iceland (Harning et al.,          et al., 2016; Kaufman et al., 2016; Lecavalier et al., 2017; Zhang et al.,
9 2016, 2018). In Arctic Canada and Svalbard, many glaciers are now            2017; Axford et al., 2019; Geirsd&#xf3;ttir et al., 2019; Larsen et al., 2019; smaller than they have been in at least 4000 years (Lowell et al., 2013;    Luckman et al., 2020). Conversely, in the Southern Hemisphere, where Miller et al., 2013, 2017; Schweinsberg et al., 2017, 2018) and more        summer insolation increased during the Holocene, glaciers generally than 40,000 years in Baffin Island (Pendleton et al., 2019). Although        waned (Solomina et al., 2015; Kaplan et al., 2016; Reynhout et al.,
the millennial glacier length variation records are incomplete and          2019). However, these general global trends were modulated by discontinuous, and glacier fluctuations depend on multiple factors          regional climate variations in temperature and precipitation (Murari (e.g., temperature, precipitation, topography, internal glacial dynamics),  et al., 2014; Kaplan et al., 2016; Batbaatar et al., 2018; Saha et al.,
there is a coherent relationship between rising temperatures, negative      2018) and there are a number of examples of this. A precipitation mass balance and glacier retreat on centennial time scales across most      increase led to a local early Holocene (7-8 ka) glacier maximum in arid of the world. Glaciological and geodetic observations show that the          Mongolia (Gichginii Range). Glacier advances at about 9 ka in south-rates of early 21st-century mass loss are the highest since 1850 (Zemp      west Greenland have been suggested to be a consequence of the et al., 2015). For all regions with long-term observations, glacier mass    freshwater pulse from the Laurentide Ice Sheet, which led to cooling in the decade 2010-2019 was the smallest since at least the beginning        in the Baffin Bay area (Schweinsberg et al., 2018). Lake sediments of the 20th century (medium confidence).                                    indicate that the glaciers in the region were smaller than today, or absent between 8.6 and 1.4 ka (Larocca et al., 2020). Glaciers on In contrast to the global glacier mass decline (Figure 9.21, Table 9.5, and  the Antarctic Peninsula and in Patagonia during the Holocene were Supplementary Material 9.SM.2), a few glaciers have gained mass or          strongly affected by the southern westerly winds, sea ice extent, advanced due to internal glacier dynamics or locally restricted climatic    and ocean circulation (Garc&#xed;a et al., 2020). Recent studies indicate causes. The SROCC discusses the Karakoram anomaly (centred on the          that explosive volcanism can drive glacier advances (Solomina et al.,
western Kunlun range (at about 80&deg;E, 35&deg;N), but also covering part          2015, 2016; Schweinsberg et al., 2018; Brnnimann et al., 2019).
of the Pamir and Karakoram ranges), where glaciers have been close          In summary, on millennial time scales over the Holocene, there is to balance since at least the 1970s, and had a slightly positive mass        high confidence that orbital forcing drove hemispheric-scale glacier balance since the 2000s. Since SROCC, new evidence suggests that this        variations, but new studies provide a nuanced picture of responses to anomaly is related to a combination of low-temperature sensitivity of        a variety of regional-scale forcings.
debris-covered glaciers, a decrease of summer air temperatures (Cross-Chapter Box 10.3), and an increase in snowfall, possibly caused by          Section 3.4.3.1 assesses new attribution studies for glaciers and increases in evapotranspiration from irrigated agriculture (Bonekamp        finds that human influence is very likely the main driver of the global, et al., 2019; de Kok et al., 2020; Farinotti et al., 2020; Shean et al.,    near-universal retreat of glaciers since the 1990s. The SROCC assessed 2020). However, a recent geodetic mass balance estimate suggests            that it is very likely that atmospheric warming is the primary driver substantially increased thinning rates of High Mountain Asian glaciers      for the global glacier recession. Since SROCC, a study of glaciers after about 2010 (Hugonnet et al., 2021). There is limited evidence to      in New Zealand used event attribution to confirm a connection assess whether the Karakoram anomaly will persist in coming decades          between extreme glacier mass loss years and anthropogenic warming but, due to the projected increase in air temperature throughout the        (Vargo et al., 2020).
region, its long-term persistence is unlikely (high confidence) (Cross-Chapter Box 10.3; Kraaijenbrink et al., 2017; de Kok et al., 2020;          The SROCC stated with high confidence that, besides temperature, Farinotti et al., 2020).                                                    other factors, such as precipitation changes or internal glacier dynamics, have modified the temperature-induced glacier response 9.5.1.1.3 Drivers of glacier change                                          in some regions. Deposition of a thin layer (<2 cm) of light-absorbing particles (e.g., black carbon, brown carbon, algae, mineral dust The AR5 (Masson-Delmotte et al., 2013) noted that early-to-mid-              or volcanic ash) can exert an important control on glacier mass Holocene glacier minima could be attributed to high summer insolation        balance, by decreasing surface albedo and thus increasing absorbed (high confidence), unlike the current situation. Since AR5, new and          shortwave radiation and melt (see also Section 7.3.4.3). The SROCC improved chronologies of glacier size variations from the end of the        found limited evidence and low agreement that this process has 1276
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                        Chapter 9 Global (all)            Global (excep 5&19)                      Alaska (1)                W est Canada and U.S. (2) 200 150 100 50 0
Arctic Canada (N) (3)          Arctic Canada (S) (4)          Greenland periphery (5)                      Iceland (6) 200 150                                                                                                                                                9 100 50 0
Svalbard (7)                Scandinavia (8)                  Russian Arctic (9)                  North Asia (10)
Glacier mass relative to year 2015 (%)
200 150 100 50 0
Central Europe (11)              Caucasus (12)              High Mountain Asia (1315)              Low Latitudes (16) 200 150 100 50 0
1950  2000    2050            2100 Southern Andes (17)            New Zealand (18)              Antarctic periphery (19) 200 Reconstruction (1) 150                                                                                                                                  (2)
Observation RCP 2. 6(3) 100 RCP 4. 5(3)
(3 )
50                                                                                                                RCP 8. 5 0
1950    2000    2050  2100    1950    2000  2050    2100      1950    2000      2050  2100 Figure 9.21 l Global and regional glacier mass evolution between 1901 and 2100 relative to glacier mass in 2015.
1277
 
Chapter 9                                                                                                                Ocean, Cryosphere and Sea Level Change Figure 9.21 (continued): Reconstructed glacier mass change through the 20th century (Marzeion et al., 2015) and observed during 1961-2016 (Zemp et al., 2019). Projected (2015-2100) glacier mass evolution is based on the median of three RCP emissions scenarios (Marzeion et al., 2020). In all cases, uncertainties are the 90% confidence interval.
For a better comparison between regions, the maximum relative mass change was set to 200%, although for three regions, the volume changes between 1901 and 2015 exceeded that value. For the Low Latitude, New Zealand, and High Mountain Asia glaciers, the changes were larger than 1000%, 350%, and 250%, respectively. See Figure 9.2 for the location of each region. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
had a significant effect on observed long-term glacier changes.                            Kraaijenbrink et al., 2017; Maussion et al., 2019; Zekollari et al., 2019; Several studies have shown melt increases due to the deposition                            Rounce et al., 2020) and simplified energy balance calculations (Sakai of light-absorbing particles (Schmale et al., 2017; Wittmann et al.,                      and Fujita, 2017; Shannon et al., 2019). Compared to simpler, empirical 2017; Sigl et al., 2018; Di Mauro et al., 2019, 2020; Magalhes et al.,                    parametrizations, full energy-balance models are not necessarily the 2019; Constantin et al., 2020). Conversely, increasingly thick debris                      most appropriate choice for simulating future glacier response to cover (>2-5 cm) on retreating glaciers can slow down glacier melt                          climate change, even at the local scale (R&#xe9;veillet et al., 2017, 2018),
(Pratap et al., 2015; Brun et al., 2016). Although debris covers only                      because of parameter and forcing uncertainties. All models account about 4-7% of the total glacier area globally (Scherler et al., 2018;                      for glacier retreat and advance, but only two models (Anderson and 9 Herreid and Pellicciotti, 2020), many glaciers are heavily debris-                        Mackintosh, 2012; Huss and Hock, 2015) include frontal ablation.
covered in their lower reaches, especially in High Mountain Asia, the Caucasus, the European Alps, Southern Andes and Alaska, resulting                          Secondary processes such as debris-cover thickening (e.g., Herreid in different responses to warming than similar clean-ice glaciers. A                      and Pellicciotti, 2020), albedo changes due to light-absorbing shift in regional meteorological conditions, driven by the location and                    particles (e.g., Magalhes et al., 2019; Williamson et al., 2019), trends strength of the upper level zonal wind, has been found to have forced                      of refreezing and water storage in firn (e.g., Ochwat et al., 2021),
recent high mass loss rates in Western North America (Menounos                            dynamic instabilities such as surges (e.g., Th&#xf8;gersen et al., 2019) et al., 2019). High geothermal heat flux areas underneath glaciers and                    or glacier collapse (e.g., Kb et al., 2018), are not represented high energy dissipation in the flow of water and ice causes additional                    in global glacier models, resulting in both underestimated and mass loss of the glaciers in Iceland (J&#xf3;hannesson et al., 2020),                          overestimated sensitivity to warming that is currently not possible to accounting for 20% of the mass loss since 1994 (Aalgeirsd&#xf3;ttir                            quantify. Furthermore, challenges for future projections are caused et al. 2020). Glacier lake volume in front of retreating glaciers, has                    by the low-resolution and high-spatial variability at sub-grid scale increased globally by around 48% between 1990 and 2018 (Shugar                            of the precipitation amount provided by general circulation models et al., 2020), which can increase both subaqueous melt and calving.                        (GCMs), which requires downscaling to the spatial scale of a glacier In summary, there is high confidence that non-climatic drivers have                        (Maussion et al., 2019; Zekollari et al., 2019; Marzeion et al., 2020).
and will continue to modulate the first-order temperature response                        In summary, in agreement with SROCC, progress in global scale of glaciers in some regions.                                                              glacier modelling efforts allows medium confidence in the capability of current-generation glacier models to simulate the magnitude and 9.5.1.2        Model Evaluation                                                            timing of glacier mass changes as a response to climatic forcing.
Since AR5, glacier mass projections have been coordinated by the                          9.5.1.3      Projections Glacier Model Intercomparison Project (GlacierMIP; Hock et al.,
2019a; Marzeion et al., 2020). The SROCC (Hock et al., 2019b) relied                      The AR5 (Vaughan et al., 2013) and SROCC (Hock et al., 2019b) on six global-scale glacier models based on previously published                          stated with high confidence that the worlds glaciers are presently glacier model projections (Hock et al., 2019a). It found with high                        in imbalance due to the warming of recent decades. The observed confidence that glaciers will lose substantial mass by the end of                          retreat of glaciers is only a partial response to the already realized the century, but assigned medium confidence to the magnitude and                          warming (Christian et al., 2018), and they are committed to losing timing of the projected glacier mass loss, because of the simplicity                      considerable mass in the future, even without further change in of the models, the limited observations in some regions to calibrate                      air temperature (Mernild et al., 2013; Tr&#xfc;ssel et al., 2013; Zekollari them, and the diverging initial glacier volumes.                                          and Huybrechts, 2015; Huss and Fischer, 2016; Marzeion et al.,
2018; Jouvet and Huss, 2019). One model estimates that 36 +/- 8 %
Since SROCC, Marzeion et al. (2020) projected 21st century global-                        of global glacier mass is already committed to be lost due to scale glacier mass changes based on seven global-scale and four                            past greenhouse gas emissions (Marzeion et al., 2018). Although regional-scale glacier models (Annex II). All models used the same                        accumulation and ablation instantly determine the SMB, the glacier initial and boundary conditions, forming a more coherent ensemble                          geometries adjust to changed atmospheric conditions over a longer of projections compared to SROCC. Nevertheless, challenges remain                          time (Zekollari et al., 2020). The adjustment time, often referred because of scarcity of glacier thickness, surface mass balance (SMB)                      to as the response time, is variable from one glacier to another, and frontal ablation data for model calibration, but also due to                          depending on the glacier geometry (thickness and steepness), SMB uncertainties in glacier outlines, surface elevations and ice velocities.                  and gradient (e.g., J&#xf3;hannesson et al., 1989; Harrison et al., 2001; The global SMB models are of varying complexity, including mass                            L&#xfc;thi, 2009; Zekollari et al., 2020). Response time is variable: years balance sensitivity approaches (van de Wal and Wild, 2001),                                for smaller and steeper glaciers (Beedle et al., 2009; L&#xfc;thi and temperature-index methods (Anderson and Mackintosh, 2012;                                  Bauder, 2010; Rabatel et al., 2013), up to tens or hundreds of years Marzeion et al., 2012; Radi et al., 2014; Huss and Hock, 2015;                            for larger and gentle-sloped glaciers (e.g., Burgess and Sharp, 2004; 1278
 
Ocean, Cryosphere and Sea Level Change                                                                                                                      Chapter 9 L&#xfc;thi et al., 2010; Zekollari et al., 2020). The models indicate that                  mass loss compared to the RCP forced simulations, although with the disequilibrium between the glaciers and present atmospheric                        fewer global glacier models. To enable the glacier contribution to conditions (1995 to 2014) reduces and then disappears at around                        future sea level rise to be estimated under the full range of SSP year 2070 (Marzeion et al., 2020). There is therefore very high                        scenarios (Section 9.6.3.3), the GlacierMIP results are emulated confidence that the disequilibrium of glaciers will persist as warming                using a Gaussian process model (Box 9.3 and Table 9.4; Edwards continues, and that glaciers will continue to lose mass for at least                  et al., 2021). The emulated projections show a narrower range than several decades because of their lagged response, even if global                      the roughly equivalent RCP projections, which may be explained temperature is stabilized.                                                            by not accounting for covariance in the regional uncertainties (Marzeion et al., 2020) and by the fact that the emulator caps sea The SROCC assessed that global glacier mass loss by 2100, relative                    level contribution for each region at the volume above floatation to 2015 will be 18 [likely range 11 to 25] % for scenario RCP2.6                      estimated by Farinotti et al. (2019) (Table 9.SM.2). Comparison of and 36 [likely range 26 to 47] % for RCP8.5, and that many glaciers                    simulated and emulated regional sea level contributions support this will disappear regardless of the emissions scenario (very high                        explanation. Rates of change and post-2100 sea level projections confidence). Since SROCC, new results from Marzeion et al. (2020)                      are estimated with the AR5 parametric fit (Supplementary have been published (Box 9.3, Figure 9.21 and Table 9.4, including                    Material 9.SM.4.5; Church et al., 2013b) applied to the GlacierMIP                9 peripheral glaciers in Greenland and Antarctica). Glaciers will lose                  results (Marzeion et al., 2020), and these are also shown in Table 9.4 29,000 [9000 to 49,000] Gt and 58,000 [28,000 to 88,000] Gt                            for comparison.
over the period 2015-2100 for RCP2.6 and RCP8.5, respectively (medium confidence), which represents 18 [5 to 31] % and 36 [16 to                    The mass loss rates vary between regions and there are distinctively 56] % of their early 21st century mass, respectively (Table 9.4).                      different patterns between scenarios (Marzeion et al., 2020).
Within uncertainties, these agree with SROCC estimates, although                      The global models agree that regions characterized by relatively with a slightly smaller mass loss due to the inclusion of models with                  little glacier-covered area (Low Latitude, Central Europe, Caucasus, lower sensitivity to changing climate conditions (Marzeion et al.,                    Western Canada and USA, North Asia, Scandinavia and New Zealand) 2020). The greatest source of uncertainty in glacier mass loss until                  will lose nearly all (>80%) glacier mass by 2100 in the RCP8.5 the middle of the 21st century is the disagreement between glacier                    scenario, but their corresponding contribution to sea level rise will models, with emissions scenario becoming the dominant cause of                        be small. A study using detailed ice dynamics for the largest glacier uncertainty by the end of the 21st century (Marzeion et al., 2020).                    of the European Alps, Great Aletsch Glacier, projects 60% of present ice volume will be lost by 2100 in RCP2.6 and an almost complete Although the GlacierMIP projections (Hock et al., 2019a; Marzeion                      wastage of the ice in RCP8.5 (Jouvet and Huss, 2019). Due to their et al., 2020) were forced by RCP scenarios, two global glacier models                  larger mass, the largest contribution to sea level rise comes from (Huss and Hock, 2015; Maussion et al., 2019) were also run with                        glaciers in the Arctic and Antarctic regions (Antarctic, Arctic Canada, 13 GCMs and SSP scenarios (Table 9.4). These results show increased                    Alaska, Greenland, Svalbard and Russian Arctic), in spite of having Table 9.4 l Projected sea level contributions from global glaciers (including peripheral glaciers in Greenland and Antarctica) by 2100 relative to 2015, for selected Representative Concentration Pathway (RCP) and Shared Socio-economic Pathway (SSP) scenarios.
Representative Concentration Pathways (RCPs)
Study                          RCP2.6                RCP4.5                  RCP8.5                              Notes IPCC AR5 and SROCC                                      0.10                  0.12                    0.17      Median and likely (66% range) contributions in (Church et al., 2013b; Oppenheimer et al., 2019)  (0.04-0.16) m        (0.06-0.19) m            (0.09-0.25) m  2100 relative to 1995-2014 GlacierMIP                                            0.094                0.142                    0.200 Mean (+/-1 standard deviation range) contributions Hock et al. (2019a)                              (0.069-0.119) m          (107-177) m          (0.156-0.240) m GlacierMIP                                            0.079                0.119                    0.159 Median [90% range]
Marzeion et al. (2020)                          [0.023-0.135] m      [0.053-0.185] m          [0.073-0.245] m Shared Socio-economic Pathways (SSPs)
Study                          SSP12.6              SSP24.5                SSP58.5                            Notes 0.111                0.136                    0.190 GlacierMIP experimental protocol                                                                                Mean (66% range) [90% range] using 13 GCMs (0.077-0.145)        (0.096-0.176)            (0.133-0.247)
(Marzeion et al., 2020) with CMIP6 forcing                                                                      and 2 glacier modelsa
[0.05-0.167] m        [0.07-0.201] m          [0.09-0.283] m GlacierMIP (Marzeion et al., 2020) with AR5            0.102                0.128                    0.171      Median (66% range) [90% range] contribution from parametric fit: used for rates and post-2100      (0.076-0.134)        (0.095-0.167)            (0.124-0.224)  AR5 parametric fit to GlacierMIP ensemble, relative projections (Supplementary Material 9.SM.4.5)    [0.059-0.154] m      [0.076-0.192] m          [0.098-0.259] m  to 1995-2014 0.080                0.115                    0.170      Median (66% range) [90% range] contribution in Emulated (0.059-0.101)        (0.093-0.137)            (0.144-0.196)  2100 relative to 2015 from emulator of GlacierMIP6 (Marzeion et al., 2020; Edwards et al., 2021)
[0.046-0.116] m      [0.077-0.155] m          [0.124-0.218] m  used with Chapter 7: Climate Forcing a
OGGM (Maussion et al., 2019) and GloGEM (Huss and Hock, 2015).
1279
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change the smallest relative mass loss, and it is expected that they will        reported medium confidence in the estimation that Earths total continue to contribute to sea level rise beyond 2100. The regions          perennial ground ice volume is equivalent to 2-10 cm of global with intermediate glacier mass (Southern Andes, High Mountain              sea level (Zhang et al., 2000). There is no evidence suggesting Asia and Iceland) show decreasing mass loss rates for RCP2.6              that a large part of this volume, if melted, would run off and throughout the 21st century, and increasing rates for RCP8.5 that          contribute to global sea level. Therefore, and because of the modest peak in the mid-to-late 21st century (Figure 9.21). The peak in mass      total volume of mobilizable water, the contribution of permafrost loss rate followed by reduction is due to decreasing glacier volume        thaw to past and future sea level budgets is usually neglected and stabilizing mass balance (Marzeion et al., 2020). Vatnajkull,        (see Section 9.6.3.2).
the largest glacier in Iceland, is projected to lose about 50% of its mass by 2300 in extended RCP4.5 and 80-100% in extended                Permafrost changes mostly refer to changes in extent, temperature RCP8.5 scenarios (Schmidt et al., 2019). In summary, both global          and active layer thickness (ALT). The SROCC (Hock et al., 2019b; and regional studies agree that glacier mass loss will continue in        Meredith et al., 2019) reported with very high confidence that all regions, with larger mass loss for high-emissions scenarios (high      record high permafrost temperatures at the depth of the zero confidence) (see also Section 8.4.1.7.1).                                  annual amplitude (the depth about 10-20 m below the surface 9                                                                            where the seasonal soil temperature cycle vanishes) were attained In AR5 and SROCC, glacier mass loss beyond 2100 was calculated            in recent decades in the Northern circumpolar permafrost region, using a parametric fit to available model simulations. In section 9.6.3.5, high confidence that permafrost has warmed over recent decades in that same parametric fit is applied to Marzeion et al. (2020) projections, many mountain ranges, and overall very high confidence that global resulting in complete glacier mass loss at year 2300 under SSP58.5        warming over the last decades has led to widespread permafrost and 40-100% mass loss under SSP12.6. Clark et al. (2016) simulate        warming. As reported in SROCC, the global (polar and mountain) glacier mass evolution, not including glaciers peripheral to the Antarctic permafrost temperature has increased at 0.29&deg;C +/- 0.12&deg;C near Ice Sheet (AIS), for different warming levels for the next 10,000 years. the depth of zero annual amplitude between 2007 and 2016 There is limited evidence and low confidence that, at sustained            (Biskaborn et al., 2019). Stronger warming has been observed in warming levels between 1.5 and 2&deg;C, about 50-60% of glacier mass          the continuous permafrost zone (0.39&deg;C +/- 0.15&deg;C) compared to will remain, predominantly in the polar regions. At sustained warming      the discontinuous zone (0.20&deg;C +/- 0.10&deg;C), consistent with the fact levels between 2 and 3&deg;C, about 50-60% of glacier mass outside            that, near the melting point, a large amount of energy is required Antarctica will be lost and, at sustained warming levels, between 3 and    for melting the ice (Figure 9.22), and because of the reduced effect 5&deg;C, 60-75% of glacier mass outside Antarctica will disappear. Based      of Arctic amplification in more southerly locations (Romanovsky on Marzeion et al. (2020), there is medium confidence that nearly all      et al., 2017). This is consistent with longer-term Arctic trends glacier mass in low latitudes, Central Europe, the Caucasus, western      from deep boreholes shown in Figure 2.22. Mountain permafrost Canada and the USA, North Asia, Scandinavia and New Zealand will          temperature trends are heterogeneous, reflecting variations in local disappear at this high warming level.                                      conditions such as topography, surface type, soil texture and snow cover, but again, generally weaker warming rates are observed in warmer permafrost at temperatures close to 0&deg;C, particularly 9.5.2      Permafrost                                                    when ice content is high (e.g., Mollaret et al., 2019; Noetzli et al.,
2019; PERMOS, 2019). In summary, strong variability in recent This section focuses on the physical aspects of permafrost (perennially    permafrost temperature trends is linked to local conditions, frozen ground) as an element of the climate system, drawing on            regionally varying temperature trends, and the thermal state of the assessment of observed global permafrost changes provided              permafrost itself. However, as discussed in Section 2.3.2.5, there in Section 2.3.2.5, and more specifically model evaluation and            is overall high confidence in the observed increases in permafrost projections. The permafrost carbon feedback is assessed in Box 5.1.        temperature over the past three to four decades throughout the Section 12.4 of this Report provides permafrost information relevant      permafrost regions.
to impacts and risk on regional scales.
Closer to the surface, the active layer undergoes annual cycles of 9.5.2.1    Observed and Reconstructed Changes                            freeze and thaw. The SROCC reported medium confidence in ALT increase as a pan-Arctic phenomenon. Recent evidence presented The current extent of the global permafrost region is about                in Section 2.3.2.5 shows pervasive ALT increase in the European 22 +/- 3x106 km2 (Gruber, 2012). Permafrost underlies about 15% of          and Russian Arctic in the 21st century, and in high elevation areas Northern Hemisphere land and more than 50% of the unglacierized            in Europe and Asia since the mid-1990s. Emergence of a clearer land north of 60&deg;N (Zhang et al., 1999; Gruber, 2012; Obu et al.,          global picture is hampered by: (i) uneven distribution of observing 2019). It is also found in high-altitude areas of mountain ranges          sites; (ii) substantial variability among the existing sites, strongly in both hemispheres - estimated in SROCC (Hock et al., 2019b) as          influenced by local conditions (soil constituents and moisture, representing about 27-29% of the global permafrost area (medium            snow cover, vegetation); (iii) interannual variability; and (iv) thaw confidence) and most unglacierized areas in Antarctica (Vieira et al.,    settlement in ice-rich terrain (Streletskiy et al., 2017; ONeill et al.,
2010; Obu et al., 2020). Ground ice volume in permafrost is variable,      2019). In summary, in agreement with SROCC and recent evidence reaching up to 90% in syngenetic permafrost deposits (Kanevskiy            presented in Section 2.3.2.5, there is medium confidence that ALT et al., 2013; Gilbert et al., 2016). The SROCC (Meredith et al., 2019)    increase is a pan-Arctic phenomenon.
1280
 
Ocean, Cryosphere and Sea Level Change                                                                                                Chapter 9 There is medium confidence that the observed acceleration and            9.5.2.2    Evaluation of Permafrost in Climate Models destabilization of rock glaciers is related to warming temperatures and increase in water content at the permafrost table in recent          As stated in AR5 (Flato et al., 2013), coupled models contributing to decades (Deline et al., 2015; Cicoira et al., 2019; Marcer et al., 2019;  CMIP5 showed large inter-model variability of permafrost extent due PERMOS, 2019; Kenner et al., 2020). There is also medium confidence      to deficiencies in reproducing surface characteristics and processes that observed increases in size and frequency of rock avalanches          (Koven et al., 2013), particularly thermal properties of the ground are linked to permafrost degradation in rock walls (Ravanel et al.,      and snow. These deficiencies led SROCC (Meredith et al., 2019) to 2017; Patton et al., 2019; Tapia Baldis and Trombotto Liaudat, 2019). express only medium confidence in the models capacity to correctly In summary, there is medium confidence that mountain permafrost          project the magnitude of future permafrost changes, in spite of high degradation at high altitude has increased the instability of mountain    confidence in the models projection of a general thaw depth increase slopes in the past decade.                                                and a substantial loss of shallow permafrost. The SROCC further noted that several types of physical pulse disturbances, in particular fire The SROCC assessed with high confidence that the extent of subsea        and thermokarst formation, are usually not represented in coupled permafrost, formed before submersion on Arctic continental shelves        climate models. This has been discussed in detail in SROCC, which during the last deglaciation, is much reduced compared to older          assessed that there is high confidence that permafrost degradation      9 studies that had estimated the entire formerly exposed Arctic shelf      through fire (Jones et al., 2015; Gibson et al., 2018) is currently area to be underlain by permafrost. This is supported by observations    occurring faster in some well-studied regions than during the first (Shakhova et al., 2017) that show rapid thaw of recently submerged        half of the 20th century, and medium confidence that thermokarst permafrost on the East Siberian Shelf. A modelling study (Overduin        formation, to which about 20% of the northern permafrost region et al., 2019) estimates that 97% of permafrost under Arctic shelves      is vulnerable (Olefeldt et al., 2016), can lead to faster large-scale is currently thinning.                                                    permafrost degradation in response to climate change.
Based on multiple studies, there is medium confidence that                Since SROCC, dedicated modelling of the evolution of ice- and widespread retreat of coastal permafrost is accelerating in the          organic-rich permafrost in the north-east Siberian lowlands (Nitzbon Arctic (G&#xfc;nther et al., 2015; Cunliffe et al., 2019; Isaev et al., 2019). et al., 2020) has shown that not representing thermokarst-inducing There is also consistent evidence of complete permafrost thaw in          processes in ice-rich terrain leads to a systematic underestimation areas of discontinuous and sporadic permafrost since about 1980,          of the rapidity and magnitude of permafrost thaw. Simplified but this evidence is geographically scattered (Camill, 2005; Kirpotin    inventory-based modelling (Turetsky et al., 2020) points towards et al., 2011; James et al., 2013; B.M. Jones et al., 2016; Borge et al.,  similar conclusions. Although these pulse disturbances still need to 2017; Chasmer and Hopkinson, 2017; Gibson et al., 2018). In spite        be represented in CMIP-type models, there have been many new of increasing evidence of landscape changes from site studies and        developments to that type of model since CMIP5 and AR5. Soil remote sensing, quantifying permafrost extent change remains              freezing and its thermal and hydrological effects are now included in challenging because it is a subsurface phenomenon that cannot be          a large number of land-surface modules that are part of the CMIP6 observed directly (Jorgenson and Grosse, 2016; Trofaier et al., 2017). ensemble (S. Chadburn et al., 2015; Hagemann et al., 2016; Cuntz A modelling study for the Qinghai-Tibet Plateau between the 1960s        and Haverd, 2018; Guimberteau et al., 2018; Yokohata et al., 2020),
and the 2000s (Ran et al., 2018) suggests transition from permafrost      sometimes allowing for the effects of excess ice (Lee et al., 2014).
to seasonally frozen ground over an area of more than 400,000 km2.        Improved representation of snow insulation in models has led In summary, there is medium confidence that complete permafrost          to more realistic simulated permafrost extents (e.g., Paquin and thaw in recent decades is a common phenomenon in discontinuous            Sushama, 2015). In a post-CMIP5 ensemble of land-surface models and sporadic permafrost regions. In addition, paleoclimatic evidence      driven by observed meteorological conditions (McGuire et al., 2016),
presented in Section 2.3.2.5 confirms a long-term sensitivity of          inter-model spread was substantially reduced when the ensemble permafrost extent to climatic variations, although an analysis            was restricted to models that appropriately represented the effect of of North American speleothem records over the last two glacial cycles    snow insulation on the underlying soil (W. Wang et al., 2016). More indicates that this apparent high sensitivity could be a consequence      detailed descriptions of high-latitude vegetation characteristics, of regional-scale variability (Batchelor et al., 2019).                  vegetation dynamics, and snow-vegetation interactions have been included in several models since AR5 (S.E. Chadburn et al., 2015; There is a lack of formal studies attributing observed permafrost        Porada et al., 2016; Druel et al., 2017).
changes (thaw depth, thermal state) or associated landscape changes to anthropogenic forcing. However, the observed Arctic warming has        A total soil column depth of at least about 10 m is required to been attributed to anthropogenic forcing (e.g., Najafi et al., 2015) and  adequately represent the dampening effect of seasonal-scale heat an obvious physical link exists between ground temperatures (and thus    exchanges between shallow and deeper ground, and thus to correctly permafrost) and surface air temperatures. Therefore, physically          simulate ALT (Lawrence et al., 2008; Ekici et al., 2015). However, many consistent and convergent lines of evidence lead to medium confidence    CMIP6 models still have shallower total soil columns (Burke et al.,
in anthropogenic forcing being the dominant cause of the observed        2020) and the proportion of models with deeper total soil columns pan-Arctic permafrost changes. Added to this, local permafrost change    has not increased since CMIP5 (Koven et al., 2013). Another recently by soil and ecosystem disturbance is induced by increasing human          identified process, usually not represented in the current (CMIP6) industrial activities in the Arctic (e.g., Raynolds et al., 2014).        generation of climate models (Zhu et al., 2019), is warming-driven 1281
 
Chapter 9                                                                                                                      Ocean, Cryosphere and Sea Level Change decomposition and burning of organic material that provides strong                          This further points to deficiencies in the land modules as the main reason thermal insulation of underlying ground. Decay of the insulating                            for biases, consistent with conclusions drawn from the analysis of CMIP5 organic material can lead to increased permafrost thaw, creating                            output (Koven et al., 2013), as reported in SROCC and AR5.
a positive feedback loop.
In spite of more realistic description of permafrost-related processes in In spite of the aforementioned structural improvements to many                              many coupled climate models, the CMIP6 models still produce a very models, the simulated current permafrost extent from available                              scattered ensemble of estimates of current permafrost extent, and CMIP6 models shows no substantial improvement with respect to                                there is high confidence that this is strongly linked to deficiencies of CMIP5 (see Figure 9.22a). The extent of the region where permafrost                          the representation of soil processes. Furthermore, current-generation is simulated within the top 15 m in the Northern Hemisphere for the                          climate models tend to neglect several physical disturbances that 1979-1998 period is characterized by very large scatter in the coupled                      can lead to faster permafrost thaw. Because of large uncertainties CMIP5 and CMIP6 historical simulations compared to estimates                                in the future evolution of these drivers (see SROCC), there is limited of the present permafrost extent based on multiple observational                            evidence that these shortcomings lead to an underestimate of lines of evidence (Zhang et al., 1999) and models based on satellite                        permafrost degradation rates in response to climate change in the 9 observations and reanalyses (Gruber, 2012; Obu et al., 2019). Outliers                      CMIP6 ensemble. In summary, there is high confidence that coupled with very low simulated permafrost extent are models that have only                          models correctly simulate the sign of future permafrost changes a very shallow soil column (leading to an underestimate of thermal                          linked to surface climate changes, but only medium confidence in the inertia at depth) and do not take into account soil water phase                              amplitude and timing of the transient response.
changes. These inadequacies lead to an overestimate of seasonal thaw depth, exceeding the total thickness of the models soil                                9.5.2.3        Projected Permafrost Changes columns (Burke et al., 2020). Excessive simulated permafrost extent can in several cases be traced to insufficient thermal insulation by the                    The AR5 (Collins et al., 2013) and SROCC (Meredith et al., 2019) winter snow cover (Burke et al., 2020).                                                      (based on available CMIP5 output) both expressed high confidence that future pan-Arctic thaw depth will increase and near-surface Figure 9.22a also shows that the corresponding land-atmosphere                              permafrost extent will decrease under future global warming, and simulations with prescribed observed sea surface temperatures and                            medium confidence in the magnitude of the simulated changes sea ice concentrations, and the land-only simulations with prescribed                        because of model deficiencies and the large spread of the results.
reanalysis-based meteorological forcing, do not provide an improved simulation of the current permafrost extent, although, by construction,                      The equilibrium sensitivity of permafrost extent to stabilized global they can be expected to exhibit lower land surface climate biases.                          mean warming has been inferred (by constraining CMIP5 output Figure 9.22 l Simulated versus observed permafrost extent and volume change by warming level. (a) Diagnosed Northern Hemisphere permafrost extent (area with perennially frozen ground at 15 m depth, or at the deepest model soil level if this is above 15 m) for 1979-1998, for available Coupled Model Intercomparison Project Phase 5 and 6 (CMIP5 and CMIP6) models, from the first ensemble member of the historical coupled run, and for CMIP6 Atmospheric Model Intercomparison Project (AMIP)
(atmosphere+land surface, prescribed ocean) and land-hist (land only, prescribed atmospheric forcing) runs. Estimates of current permafrost extents based on physical evidence and reanalyses are indicated as black symbols - triangle: Obu et al. (2018); star: Zhang et al. (1999); circle: central value and associated range from Gruber (2012). (b) Simulated global permafrost volume change between the surface and 3 m depth as a function of the simulated global surface air temperature (GSAT) change, from the first ensemble members of a selection of scenarios, for available CMIP6 models. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1282
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 with diagnosed relationships between the observed present-day            Terrestrial snow cover is characterized via three variables: (i) areal snow spatial distribution of permafrost and air temperature) to be about      cover extent (SCE); (ii) the time period of continuous snow cover -
4.0x106 km2 &deg;C-1 (Chadburn et al., 2017) for global surface air          snow cover duration (SCD) that reflects snow-on and snow-off dates temperature (GSAT) changes with respect to the present below            (i.e., the first and last days of observed snow cover); and (iii) snow about +3&deg;C. This equilibrium permafrost sensitivity, relevant for        accumulation - expressed either as snow depth (SD) or snow water assessing long-term permafrost changes at a stabilized warming          equivalent (SWE), the depth of water stored by the snowpack.
level, is about 20% higher than the transient centennial-scale near-surface permafrost extent sensitivity (diagnosed from seasonal      Observed large-scale snow cover changes, their attribution to thaw down to 3 m depth) suggested by direct analysis of CMIP5            human activity, and their effects on the hydrological cycle are also output (Slater and Lawrence, 2013). Compared to these and other          discussed in Chapter 2 (Section 2.3.2.2), Chapter 3 (Section 3.4.2) studies reported in AR5 and SROCC (Koven et al., 2013), the recently    and Chapter 8 (Section 8.2.3.1) of this Report. The role of snow in suggested equilibrium extent sensitivity to GSAT changes of about        the global surface albedo feedback is assessed in Section 7.4.2.3.
1.5x106 km2 &deg;C-1 based on idealized ground temperature modelling        The effect of aerosol deposition on snow albedo and associated (Liu et al., 2021) appears unrealistically low.                          climate forcing is assessed in Section 7.3.4.3.
9 A strong transient temperature sensitivity of the volume of              9.5.3.1      Observed Changes of Seasonal Snow Cover perennially frozen soil in the top 3 m below the surface is consistently suggested by the available CMIP6 models (Figure 9.22b). Relative        The AR5 (Vaughan et al., 2013) reported that NH SCE in June to the current volume, the transient sensitivity of the modelled        very likely decreased by 11.7 [8.8 to 14.6] % per decade over the permafrost volume in the top 3 m to GSAT changes (with respect          1967-2012 period, exceeding the absolute and relative reductions to the 1995-2014 average and up to +3&deg;C change, that is, about up        observed in March and April. The AR5 further reported very high to +4&deg;C with respect to pre-industrial levels) is about 25 +/- 5 % &deg;C-1    confidence that NH March and April SCE decreased over the 90 years (Burke et al., 2020), but there is only medium confidence in this        after 1922. The SROCC only assessed snow cover changes for the value and 1 standard deviation uncertainty range because of the          Arctic and mountain areas. For the Arctic (north of 60&deg;N), SROCC model deficiencies discussed in 9.5.2.2. It is important to note that    (Meredith et al., 2019) expressed high confidence in SCE decreases permafrost loss will not be limited to the top 3 m, with delayed        of -3.5 +/- 1.9% per decade in May and -13.4 +/- 5.4% per decade in response of deeper permafrost. The simulated transient temperature      June, based on a combination of multiple datasets (Mudryk et al.,
sensitivity of permafrost volume is slightly stronger in the SSP12.6    2017). Concerning mountain snow cover, SROCC (Hock et al., 2019b) scenario than in other SSPs because subsurface temperature lag          reported with high confidence that mountain snow cover (both in increases with higher atmospheric warming rates, particularly when      terms of SCE and maximum SWE) has generally declined since the ground ice melting induces additional delays.                            middle of the 20th century at lower elevations. At higher elevations, SROCC reported medium confidence in generally insignificant Due to the role of air temperature as a major driver of permafrost      snow cover trends (where these were available). The large-scale change, SROCC (Hock et al., 2019b) expressed very high confidence        assessment provided in Section 2.3.2.2 of this Report reports very that permafrost in high mountain regions is expected to undergo          high confidence in substantial reductions of NH SCE (particularly in increasing thaw and degradation during the 21st century, with            spring) since 1978, and states that there is limited evidence that this stronger consequences expected for higher greenhouse gas emissions      decline extends back to the early 20th century.
scenarios. Recently published studies (e.g., Zhao et al., 2019) support this SROCC assessment.                                                  Since SROCC, progress has been made in characterizing seasonal NH snow cover changes through the combined analysis of datasets In summary, based on high agreement across CMIP6 and older model        from multiple sources (surface observations, remote sensing, land projections, fundamental process understanding, and paleoclimate        surface models and reanalysis products). A recent combined dataset evidence, it is virtually certain that permafrost extent and volume      (Mudryk et al., 2020) identified negative NH SCE trends in all months will shrink as global climate warms.                                    between 1981 and 2018, exceeding -50 x 103 km2 yr -1 in November, December, March and May (Figure 9.23a,b). The loss of spring SCE is also reflected in earlier spring snow melt, derived from surface 9.5.3        Seasonal Snow Cover                                        observations (Bulygina et al., 2011; Brown et al., 2017), satellite observations (Wang et al., 2013; Estilow et al., 2015; Anttila et al.,
Mean snow cover extent in January and February, the usual months        2018), and model-based analyses (Liston and Hiemstra, 2011).
of maximum extent, covers about 45% of the Northern Hemisphere          There is considerable inter-dataset and regional variability, but the (NH) land surface - more than 45 million km2 over the 1967-2014          continental-scale trends of snow-off dates from these datasets are period (Estilow et al., 2015). In contrast, maximum seasonal snow        consistently negative (Brown et al., 2017; Kouki et al., 2019).
cover in South America, the dominant ice-free land mass in the Southern Hemisphere in terms of seasonal snow cover extent,              Satellite-derived estimates of NH SCE compiled within the National remains well below 1 million km2 (Foster et al., 2009) or less than 2%  Oceanic and Atmospheric Administration Climate Data Record of the Southern Hemisphere land surface.                                (NOAA CDR) snow chart extend back to 1967, providing one of the longest environmental data records from spaceborne measurements 1283
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change (Estilow et al., 2015). Continental trends from these coarse resolution      Mudryk et al., 2017). The positive trends from the NOAA CDR are also estimates (about 200 km) show declining snow cover during                    inconsistent with later autumn snow-on dates since 1980 (-0.6 to the spring period, consistent with surface warming (Hern&#xe1;ndez-              -1.4 days per decade), based on historical surface observations, Henr&#xed;quez et al., 2015; Mudryk et al., 2017). Therefore, as assessed in      model-derived analyses and independent satellite datasets (updated Section 2.3.2.2, there is very high confidence that the NH spring SCE        from Derksen et al., 2017). The SCE trend sensitivity to surface has been decreasing since 1978.                                              temperature forcing in the NOAA CDR is anomalous compared to other datasets during October and November (Mudryk et al., 2017).
Hemispheric reconstructions with simple snow models and in situ              There is therefore medium confidence that the NH SCE trend for observations have extended a pre-satellite record to precede the            the 1981-2016 period was also negative during these two months satellite record and extend back to 1922 (Brown and Robinson,                (Mudryk et al., 2020).
2011), putting the satellite era in historical context. This study, also assessed in AR5, suggests an increase in North American spring              In the low-to-mid latitude (18&deg;S-40&deg;S) South American Andes, (March-April) SCE from 1915 to about 1950, followed by a decrease            a dry-season snow cover decrease of about 12% per decade has of the same total magnitude afterwards. In Eurasia, a negative trend        been reported for the 1986-2018 period (Cordero et al., 2019), linked 9 in April is visible over the entire 1922-2010 period of record, while        to El Nino-Southern Oscillation (ENSO) changes dominant in the in March, a step decrease at about 1985 separates two periods with          northern part, and an additional influence of poleward migration of insignificant trends. Overall, combining March and April, consistency        the westerly wind zone in the southern part of the study area. Further between the continental trends since 1950, and agreement in sign            south, long-term warming has been identified as the dominant cause with the NOAA satellite record since 1967, provides high confidence          of observed winter snow cover reduction over the 1972-2016 period in Northern Hemisphere spring snow cover decrease since about                at about 53&deg;S in Brunswick Peninsula (Aguirre et al., 2018).
1950. Analysis of paleoclimate records (Pederson et al., 2011; Belmecheri et al., 2016) suggests that recent snowpack reductions            The AR5 (Hock et al., 2019b) reported on SWE and SD in situ in western North America are exceptional on a millennial time scale          observations mostly from mountain areas, the majority of which (medium confidence).                                                        showed negative trends over their respective observational periods.
However, AR5 did not provide an assessment of large-scale snow Recent remote sensing global-scale studies (Hammond et al., 2018;            mass changes across the Northern Hemisphere. The SROCC attributed Notarnicola, 2020) report that, since 2000, snow cover area and/or          medium confidence to reports of negative SWE trends in the Russian duration decreased in 78% of global mountain areas (Notarnicola,            Arctic between 1966 and 2014, and stated that seasonal maximum 2020). Due to the shortness of these records and high spatial                SD trends in the North American Arctic were mostly insignificant variability, they only provide limited evidence in medium agreement          and inconsistently positive or negative. It further attributed medium that snow cover area and duration changes over that recent period            confidence to gridded products that suggest negative pan-Arctic SWE are more consistently negative at higher (>4000 m) than at lower            trends between 1981 and 2016, and high confidence in a general elevations, and do not alter the high confidence in longer-term              decline of mountain snow mass at lower elevations, albeit with mountain snow cover decrease at lower elevations since the middle            regional variations.
of the 20th century that was already reported in SROCC.
Since AR5, the number of global or hemispheric-scale gridded As assessed in detail in Section 3.4.2, it is very likely that anthropogenic SWE products has substantially increased. A validation and influence contributed to the observed reductions in Northern                intercomparison (Mortimer et al., 2020) of datasets - derived from:
Hemisphere spring snow cover since the mid-20th century. The reasons        (i) reanalysis-based products; (ii) a combined surface observation -
for this assessment are: (i) physical consistency of the observed            passive microwave remote sensing product; and (iii) stand-alone spring snowpack and surface temperature changes in observations              passive microwave products - has led to better understanding of the and models; (ii) the strong observed hemispheric and regional spring        strengths and limitations of each. These gridded products consistently SCE and SWE trends; and (iii) the general attribution of hemispheric        identify negative trends in maximum pre-melt SWE across the temperature changes to human influence. Consistent between multiple          1981-2016 period over Eurasia and North America (Figure 9.23c,d; observational products and historical climate model simulations, the        Mudryk et al., 2020). To further constrain SWE uncertainty, Pulliainen observed NH SCE sensitivity to NH land (>30&deg;N) warming (Mudryk              et al. (2020) implemented a bias correction based on snow course et al., 2017) is approximately -1.9x106 km2 &deg;C-1 (95% confidence            observations which yielded a current best estimate for the average range of +/-0.9x106 km2 &deg;C-1) throughout the snow season.                      1980-2018 March SWE over NH non-alpine land north of 40&deg;N of 2938
[likely range 2846-3062] Gt. Using this method, the bias-corrected Compared to numerous studies on spring SCE changes, less attention          GlobSnow v3.0 dataset suggests a 4.6 Gt yr -1 decrease of March SWE has been paid to changes in NH snow cover during the onset period            over this 39-year period across North America, and a negligible trend in the autumn, a challenging period to retrieve snow information            across Eurasia. These SWE trends are consistent with the continental from optical satellite imagery due to persistent clouds and decreased        SCE trends over this period, as assessed above, but strong regional solar illumination at higher latitudes. Positive trends in October and      and temporal variability only allows medium confidence in the signs November SCE in the NOAA CDR (Hern&#xe1;ndez-Henr&#xed;quez et al., 2015)              and magnitudes of these trends. However, there is high confidence in are not replicated in other surface, satellite, and model datasets          a general decline of NH spring SWE since 1981 (Section 2.3.2.2). In the (Brown and Derksen, 2013; Peng et al., 2013; Hori et al., 2017;              longer term (see also Section 2.3.2.2), annual maximum SD trends 1284
 
Ocean, Cryosphere and Sea Level Change                                                                                                                      Chapter 9 from site measurements confirm mostly negative trends in North                        significant during transitional seasons and at transitional (from America (Kunkel et al., 2016) between 1960-1961 and 2014-2015,                        no snow to snow) altitudes, and exhibit strong regional variations, and strong spatial variability in Eurasia (Zhong et al., 2018) between                consistent with earlier reports for the Swiss and Austrian Alps 1966 and 2012, with spatial patterns bearing some resemblance to                      (Schner et al., 2019) and the Pyrenees (L&#xf3;pezMoreno et al., 2020).
the shorter satellite-based trends reported by Pulliainen et al. (2020).
However, over this longer period, the Eurasian measurements (Zhong                    In summary, since AR5, intercomparison, dataset blending of gridded et al., 2018) exhibit, on average, a positive trend. On the Qinghai-Tibet              products, and bias correction using snow course measurements Plateau, site measurements between 1961 and 2010 (Xu et al., 2017)                    contributed to an improved estimate of the average 1980-2018 suggest a shift from an initial increase of spring SD until about 1980                March SWE over NH non-alpine land north of 40&deg;N of 2938 [likely to a decreasing trend afterwards.                                                      range 2846-3062] Gt, with medium confidence in the magnitudes of continental-scale trends over that period. However, there is Concerning the assessment of SWE trends in mountainous regions,                        high confidence in a general decline of NH spring SWE since 1981 SROCC noted a need for observations spanning several decades                          (Section 2.3.2.2). In mountain areas, in situ observations tend to because of very strong temporal variability. Moreover, determining                    suggest that annual maximum SWE reductions are generally stronger SWE trends in mountain regions is challenging because the coarse                      at elevation bands where shifts from solid to liquid precipitation                  9 resolution (typically 25 to 50 km) of gridded SWE products is                          affected the snow mass.
inadequate in areas of mountainous terrain (Snauffer et al., 2016).
Based on a compilation of a large number of studies of SWE trends                      9.5.3.2      Evaluation of Seasonal Snow in Climate Models in mountain regions, SROCC noted strong regional variations, but a general consistency in greater reductions in SWE at lower elevations                Building on AR5 (Flato et al., 2013) and subsequent published work, associated with shifts from solid to liquid precipitation. A recent                    SROCC (Meredith et al., 2019) stated that CMIP5 models tended synthesis of snow observations in the European Alps (Matiu et al.,                    to underestimate the observed decrease of Northern Hemisphere 2021) shows a 1971-2019 seasonal (November to May) SD trend of                        spring SCE due to inappropriate parametrization of snow processes,
-8.4% per decade, along with negative maximum SD and seasonal                          misrepresentation of the snow-albedo feedback, underestimated snow cover duration trends. The trends are stronger and more                          temperature sensitivity, and biased climatological spring snow cover.
Figure 9.23 l Observed monthly Northern Hemisphere snow cover (a) trends and (b) anomalies, and snow mass (c) trends and (d) anomalies. From the observation-based ensemble discussed in the text (Mudryk et al., 2020). Trends and anomalies are calculated over the 1981-2018 period. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1285
 
Chapter 9                                                                                                            Ocean, Cryosphere and Sea Level Change Since AR5, progress in the observation, description and understanding                  There is high confidence that large inter-model variations in the of snow microstructure (Kinar and Pomeroy, 2015; Calonne et al.,                        snow-cover sensitivity to temperature can largely be explained by 2017) and its links to physical (thermal and radiative) properties (Lwe                inaccuracies in the simulated snow-albedo feedback (Qu and Hall, et al., 2013; Calonne et al., 2014) has prompted efforts to represent                  2014); a multi-model sub-ensemble of CMIP5 models that simulate physical properties as a function of the evolving snow microstructure in                a correct magnitude of this feedback presents a 40% reduced models (Carmagnola et al., 2014; Calonne et al., 2015). However, even                  spread in the projected 21st century Northern Hemisphere land state-of-the-art snow models intended for meteorological and climate                    warming trend (Thackeray and Fletcher, 2016). Errors of the applications still struggle to correctly represent the time evolution of                simulated feedback strength were linked to: (i) systematic positive the snow thermal properties, particularly of cold and dry tundra snow                  albedo biases over the boreal forest belt, mostly due to unrealistic (Domine et al., 2016). Moreover, most, if not all, CMIP6 climate models                treatment of vegetation masking (Thackeray and Fletcher, 2016);
do not explicitly represent the darkening of snow by deposition of black                (ii) inaccurate prescribed tree cover fraction and inappropriate carbon and other light-absorbing aerosol species known to influence                    parametrization of leaf area index in some models (Loranty et al.,
snow melt rates (Section 7.3.4.3). Regardless of these shortcomings,                    2014; L. Wang et al., 2016); and (iii) low spatial resolution leading snow modules of climate models continue to be improved. Recent                          to inaccuracies in the strength of the simulated snow albedo 9 progress includes the incorporation of multiple energy balances                        feedback in mountainous regions (Letcher and Minder, 2015).
within the canopy and between sub-grid tiles with different snow                        Although the representation of snow-albedo feedback improved heights (Aas et al., 2017; Boone et al., 2017) and inclusion of advanced                in many CMIP5 models over CMIP3, some models deteriorated specific snow models in coupled climate models (Niwano et al., 2018;                    (Thackeray et al., 2018).
Voldoire et al., 2019), opening the prospect of future progress in quantifying snow-related feedbacks in a changing climate. Recently                      Analysis of the available CMIP6 historical simulations for the developed multi-physics snow models (Essery, 2015; Lafaysse et al.,                    1981-2014 period shows that, on average, CMIP6 models simulate 2017), which are able to emulate the behaviour of a large number of                    well the observed SCE (Mudryk et al., 2020), except for outliers and models in a broad range of climates, allow model shortcomings and                      a median low bias during the winter months (Figure 9.24a). This is key parameter uncertainties, for example, concerning snow masking by                    an improvement over CMIP5 (Mudryk et al., 2020), where many vegetation or snow thermal conductivity, to be identified. Guidance for                snow-related biases were linked to inadequacies of the vegetation future model improvement can be provided by improved diagnostics,                      masking of snow cover over the boreal forests (Thackeray et al.,
such as a concise metric of snow insulation (A.G. Slater et al., 2017),                2015). A comparison between CMIP5 and CMIP6 results (Mudryk which builds on an observed relation between effective seasonal mean                    et al., 2020) shows that there is no notable progress in the quality SD and the dampening of winter season temperature decrease within                      of the representation of the observed 1981-2014 monthly snow the soil, and allows an efficient quantification of inaccuracies in the                cover trends.
simulated snow insulation effect.
Figure 9.24 l Simulated Coupled Model Intercomparison Project Phase 6 (CMIP6) and observed snow cover extent (SCE). (a) Simulated CMIP6 and observed (Mudryk et al., 2020) SCE (in millions of km2) for 1981-2014. Boxes and whiskers with outliers represent monthly mean values for the individual CMIP6 models averaged over 1981-2014, with the red bar indicating the median of the CMIP6 multi-model ensemble for that period. The observed interannual distribution over the period is represented in green, with the yellow bar indicating the median. (b) Spring (March to May) Northern Hemisphere SCE against global surface air temperature (GSAT) (relative to the 1995-2014 average) for the CMIP6 Tier 1 scenarios (SSP12.6, SSP24.5, SSP37.0 and SSP58.5), with linear regressions. Each data point is the mean for one CMIP6 simulation (first ensemble member for each available model) in the corresponding temperature bin. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1286
 
Ocean, Cryosphere and Sea Level Change                                                                                              Chapter 9 9.5.3.3    Projected Snow Cover Changes                                a transition from seasonal to persistent snow cover due to a strong snow-albedo feedback, are a typical feature of glacial inceptions The AR5 (Collins et al., 2013) stated that substantial NH spring snow  (e.g., Baum and Crowley, 2003; Calov et al., 2005), and these can cover reductions at the end of the 21st century were very likely under  be irreversible on centennial or longer time scales because of this strong emissions scenarios, and expressed medium confidence in the      feedback. In summary, based on physical understanding and the projected geographic patterns of annual maximum SWE changes.            absence of occurrence of such events in climate model projections, Based on studies using downscaled CMIP5 or regional climate model      abrupt future changes of seasonal snow cover on large scales in output, either directly or via snowpack models driven by such output,  the absence of concomitant abrupt atmospheric change as a driver SROCC (Hock et al., 2019b) reported likely SD or mass decreases at      appear very unlikely in the context of current and projected warming.
lower elevations in many mountain ranges over the 21st century and high confidence in smaller future changes at higher elevations.
9.6        Sea Level Change Since AR5, one study (Brown et al., 2017), applying a method developed by de El&#xed;a et al. (2013) to a CMIP5 sub-ensemble,            9.6.1      Global and Regional Sea Level Change suggested that over most of the Northern Hemisphere, the projected                  in the Instrumental Era                                    9 decrease of SCD will exceed natural variability before this will be the case for annual maximum SWE. The same study reports that, over          9.6.1.1    Global Mean Sea Level Change Budget large parts of Eastern and Western North America and Europe, forced                in the Pre-satellite Era SCD changes are projected to exceed natural variability in the 2020s in spring and autumn, while the signals tend to emerge later in the    The SROCC (Oppenheimer et al., 2019) discussed the development Arctic regions and particularly late, after 2060, in Eastern Siberia    and application of new statistical methodologies for reconstructing under the RCP8.5 scenario. Thackeray and Fletcher (2016) have          global mean sea level (GMSL) from tide gauge data over the shown that inter-model spread in projected spring SCE trends could      20th century (Box 9.1). Based on an ensemble of tide gauge be reduced through improved simulation of spring season warming        reconstructions, SROCC assessed an average rate of GMSL rise because of the tight coupling between temperature and SCE linked        of 1.38 [0.81 to 1.95, very likely range] mm yr -1 for the period to the snow-albedo feedback (Qu and Hall, 2014; Thackeray and          1901-1990. Since SROCC, two new GMSL reconstructions have Fletcher, 2016).                                                        been published (Dangendorf et al., 2019; Frederikse et al., 2020b) and are included in an updated ensemble estimate of GMSL change Across all emissions scenarios, and with negligible scenario            (Section 2.3.3.3; Palmer et al., 2021). Based on these updated data and dependence (Figure 9.24b), CMIP6 models consistently (all models        methods, the GMSL change over the (pre-satellite) period 1901-1990 and all months) simulate Northern Hemisphere snow cover decrease        is assessed to be 0.12 [0.07 to 0.17, very likely range] m with an in response to future GSAT change over the 21st century (Mudryk        average rate of 1.35 [0.78 to 1.92, very likely range] mm yr -1 (high et al., 2020). The simulated SCE decrease is close to a linear function confidence) (Table 9.5; Section 2.3.3.3) in agreement with SROCC of global temperature change for all months (shown in Figure 9.24b      assessment. Both this assessment and SROCC have substantially for spring, with medium confidence in an average sensitivity of about  larger uncertainties than the AR5 assessment, which was based on
-8% per &deg;C of GSAT increase), except when snow cover vanishes.          a single tide gauge reconstruction and did not account for structural This occurs at about +2&deg;C of GSAT change above the 1995-2014            uncertainty (see Palmer et al., 2021 for a discussion).
level (that is, about +3&deg;C above the pre-industrial level) for the months of July and August, and at about +3&deg;C above the 1995-2014        The SROCC found that four of the five available tide gauge level for June and September. Possible effects of such changes on      reconstructions that extend back to at least 1902 showed a robust the hydrological cycle are assessed in Section 8.2.3.1.                acceleration (high confidence) of GMSL rise over the 20th century, with estimates for the period 1902-2010 (-0.002 to +0.019 mm yr -2)
In summary, consistent projections from all generations of global      that were consistent with AR5. New tide gauge reconstructions climate models, elementary process understanding and strong            published since SROCC (Dangendorf et al., 2019; Frederikse et al.,
covariance between snow cover and temperature on several time          2020b) support this assessment and suggest that increased ocean scales make it virtually certain that future Northern Hemisphere snow  heat uptake related to changes in Southern Hemisphere winds cover extent and duration will continue to decrease as global climate  and increased mass loss from Greenland are the primary physical continues to warm, and process understanding strongly suggests          mechanisms for the acceleration (Section 2.3.3.3). Therefore, that this also applies to Southern Hemisphere seasonal snow cover      the SROCC assessment on the acceleration of GMSL rise over the (high confidence).                                                      20th century is maintained.
Seasonal snow cover, by definition, has a clear annual cycle            The evaluation of the sea level budget presented here, and in with usually complete disappearance in spring and summer and            Section 9.6.1.2, draws on assessments of the individual components re-formation in autumn or winter. Therefore, there is very high        (Sections 2.3.3.1 and 9.2.4.1 for global-mean thermosteric and confidence that the current and projected changes to seasonal snow      Sections 9.5.1.1, 9.4.1.1 and 9.4.2.1 for ice mass loss contributions cover are reversible (Verfaillie et al., 2018). In the case of global  to GMSL change from glaciers and ice sheets). Following SROCC or regional cooling, abrupt large-scale snow-cover changes, with        approach, the mass loss from ice sheet peripheral glaciers is 1287
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change included in the ice-sheet contributions to GMSL change (glacier mass      9.6.1.2    Global Mean Sea Level Change Budget loss from regions 5 and 19 of the Randolph Glacier Inventory 6.0                      in the Satellite Era (RGI Consortium, 2017) are added to ice-sheet mass loss where applicable, with uncertainties added in quadrature). The total change      The SROCC (Oppenheimer et al., 2019) concluded that GMSL in GMSL for each component, and their sum, is summarized in                increased at a rate of 3.16 [2.79 to 3.53, very likely range] mm yr -1 Table 9.5 (uncertainties added in quadrature). For consistency across      in the period 1993-2015 (the satellite altimetry era), and a rate the report, and to simplify the treatment of uncertainties, all budget    of 3.58 [3.10 to 4.06, very likely range] mm yr -1 in the period calculations are based on the difference between the first and last        2006-2015 - the Gravity Recovery and Climate Experiment year in each period (Palmer et al., 2021), rather than a linear fit to the (GRACE)/Argo data era (high confidence). An updated assessment underlying time series as used in SROCC and AR5.                          for the periods 1993-2018 and 2006-2018 yields values of 3.25 [2.88 to 3.61] and 3.69 [3.21 to 4.17] mm yr -1 (high The sea level budget in SROCC included the anthropogenic                  confidence) (Table 9.5), with the slightly larger central estimates contribution of land-water storage (LWS; Box 9.1) change from              consistent with the observed acceleration in GMSL rise since the a single estimate (Wada, 2016). Since SROCC, two studies have              late 1960s (Dangendorf et al., 2019), given the longer assessment 9 combined estimates of natural LWS change with anthropogenic LWS            periods. Based on the GMSL assessed time series presented in changes from reservoir impoundment and groundwater depletion              Section 2.3.3.3, GMSL acceleration is estimated as 0.075 [0.066 to (C&#xe1;ceres et al., 2020; Frederikse et al., 2020b). For C&#xe1;ceres et al.      0.080] mm yr -2 for 1971-2018 and 0.094 [0.082-0.115] mm yr -2 (2020), zero change is assumed for the period 1901-1948, since            for 1993-2018 (high confidence). For the common period of their LWS change estimates are not available before 1948. Given the        1993-2010, the assessed rate of GMSL rise based on tide gauge large year-to-year changes associated with hydrological variability,      reconstructions (3.19 [1.18 to 5.20] mm yr -1) is consistent the assessed changes in LWS (Table 9.5) are based on linear trends        with the assessment based on satellite altimetry (2.77 [2.26 to for each period, following Palmer et al. (2021). Structural uncertainty    3.28] mm yr -1), within the estimated uncertainties.
is estimated from the standard deviation of the trends across the two studies, and parametric uncertainty is estimated based on the          Since SROCC, two new estimates of the LWS contribution have Monte Carlo simulations of Frederikse et al. (2020b). These two            been published (Section 9.6.1.1; C&#xe1;ceres et al., 2020; Frederikse sources of uncertainty are combined in quadrature, and the assessed        et al., 2020b). For the early 21st century (the periods 1993-2018 central estimate is taken as the average of the ensemble mean              and 2006-2018) both publications find a positive LWS contribution trends. Compared to SROCC-assessed LWS trend of -0.12 mm yr -1            (Table 9.5), based on the most recent GRACE-derived estimates.
for the period 1901-1990, the updated assessment leads to a more          This contrasts with the negative LWS contribution presented for the negative trend of -0.16 [-0.35 to 0.04] mm yr -1, although the two        same periods in SROCC based on World Climate Research Programme are consistent within the estimated uncertainties. Previous studies        (WCRP) Global Sea Level Budget Group (2018), and reinforces the and SROCC have highlighted the large uncertainty in estimates              low confidence assessment of the LWS contribution.
of LWS change over the 20th century (Gregory et al., 2013), and therefore SROCC assessment of low confidence in the estimated LWS          For both periods in the satellite era - that is, 1993-2018 and contribution to GMSL change is maintained.                                2006-2018 - the sum of contributions is consistent with the total observed GMSL change (high confidence) (Table 9.5). However, the Since SROCC, a new ocean heat content reconstruction                      latter period, which is characterized by improved data quality and (Section 2.3.3.1; Zanna et al., 2019) has allowed global thermosteric      coverage associated with satellite and Argo observations, shows sea level change to be estimated over the 20th century. As a result,      much closer agreement in the central estimates. The marginal the sea level budget for the 20th century can now be assessed for the      sea level budget closure for the period 1993-2018 may indicate first time. For the periods 1901-1990 and 1901-2018, the assessed          underestimated uncertainty, which may be structural as well as very likely range for the sum of components is found to be consistent      parametric. The sea level budget assessments across the various with the assessed very likely range of observed GMSL change                periods in Table 9.5 demonstrate that the acceleration in GMSL rise (medium confidence), in agreement with Frederikse et al. (2020b;          (Section 2.3.3.3) since the late 1960s is mostly the result of increased Table 9.5). This represents a major step forward in the understanding      ice-sheet mass loss. However, all contributions to GMSL rise show of observed GMSL change over the 20th century, which is dominated          their largest rate during 2006-2018, with the ice sheets accounting by glacier (52%) and Greenland Ice Sheet mass loss (29%) and              for 27% of the total change during this period. Because of the the effect of ocean thermal expansion (32%), with a negative              increased ice-sheet mass loss, the total loss of land ice (glaciers and contribution from the LWS change (-14%). While the combined mass          ice sheets) was the largest contributor to GMSL rise over the period loss for Greenland and glaciers is consistent with SROCC, updates          2006-2018 (high confidence).
in the underlying datasets lead to differences in partitioning of the mass loss.
1288
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                  Chapter 9 Table 9.5 l Observed contributions to global mean sea level (GMSL) change for five different periods. Values are expressed as the total change () in the annual mean or year mid-point value over each period (mm) along with the equivalent rate (mm yr -1). The very likely ranges appear in brackets based on the various section assessments as indicated. Uncertainties for the sum of contributions are added in quadrature, assuming independence. Percentages are based on central estimate contributions compared to the central estimate of the sum of contributions.
Observed contribution                              1901-1990                1971-2018              1993-2018              2006-2018            1901-2018 to GMSL change                                    {9.6.1.1}            {CCBox 9.1}              {9.6.1.2}              {9.6.1.2}            {9.6.1.1}
31.6                    47.5                    32.7                    16.7                  63.2 (mm)    [14.7 to 48.5]          [34.3 to 60.7]        [23.8 to 41.6]          [8.9 to 24.6]        [47.0 to 79.4]
Thermal expansion                                                  (31.9%)                (50.4%)                (45.9%)                (38.6%)              (38.4%)
(Section 2.3.3.1; Table 2.7) 0.36                    1.01                    1.31                    1.39                  0.54 mm yr -1
[0.17 to 0.54]          [0.73 to 1.29]        [0.95 to 1.66]          [0.74 to 2.05]        [0.40 to 0.68]
51.8                    20.9                    13.8                    7.5                  67.2 (mm)    [30.4 to 73.2]          [10.0 to 31.7]        [10.0 to 17.6]            [6.8 to 8.2]        [41.8 to 92.6]
Glaciers (excluding peripheral glaciers)                          (52.3%)                (22.2%)                (19.4%)                (17.3%)              (40.8%)
(Sections 2.3.2.3, 9.5.1.1) 0.58                    0.44                    0.55                    0.62                  0.57 mm yr -1
[0.34 to 0.82]          [0.21 to 0.67]        [0.40 to 0.70]          [0.57 to 0.68]        [0.36 to 0.79]    9 29.0                    11.9                    10.8                    7.5                  40.4 (mm)    [16.3 to 41.7]            [7.7 to 16.1]          [8.9 to 12.7]            [6.2 to 8.9]        [27.2 to 53.5]
Greenland Ice Sheet (including peripheral glaciers)                (29.3%)                (12.6%)                (15.2%)                (17.3%)              (24.5%)
(Sections 2.3.2.4.1, 9.4.1.1) 0.33                    0.25                    0.43                    0.63                  0.35 mm yr -1
[0.18 to 0.47]          [0.16 to 0.34]        [0.36 to 0.51]          [0.51 to 0.74]        [0.23 to 0.46]
0.4                    6.7                    6.1                    4.4                  6.7 (mm)      [-8.8 to 9.6]          [-4.0 to 17.3]          [4.0 to 8.3]            [2.9 to 6.0]        [-4.0 to 17.4]
Antarctic Ice Sheet (including peripheral glaciers)                (0.4%)                  (7.1%)                  (8.6%)                (10.2%)              (4.1%)
(Sections 2.3.2.4.2, 9.4.2.1) 0.00                    0.14                    0.25                    0.37                  0.06 mm yr -1
[-0.10 to 0.11]          [-0.09 to 0.37]        [0.16 to 0.33]          [0.24 to 0.50]      [-0.03 to 0.15]
                                                                          -13.8                    7.3                    7.8                    7.2                -12.9 (mm)    [-31.4 to 3.8]          [-2.4 to 16.9]          [3.3 to 12.2]          [3.8 to 10.6]      [-45.8 to 20.0]
Land-water storagea                                              (-13.9%)                  (7.7%)                (10.9%)                (16.6%)              (-7.8%)
(Section 9.6.1.1)
                                                                          -0.15                  0.15                    0.31                    0.60                -0.11 mm yr -1
[-0.35 to 0.04]          [-0.05 to 0.36]        [0.13 to 0.49]          [0.32 to 0.88]      [-0.39 to 0.17]
99.0                    94.2                    71.2                    43.4                164.6 (mm)
[63.0 to 135.1]          [71.5 to 117.0]        [60.2 to 82.3]          [34.5 to 52.2]      [116.9 to 212.4]
Sum of observed contributions 1.11                    2.00                    2.85                    3.61                1.41 mm yr -1
[0.71 to 1.52]          [1.52 to 2.49]        [2.41 to 3.29]          [2.88 to 4.35]        [1.00 to 1.82]
120.1T                109.6T&A                  81.2A                  44.3A              201.9T&A (mm)
Observed GMSL change                                          [69.3 to 170.8]          [72.8 to 146.4]        [72.1 to 90.2]          [38.6 to 50.0]      [150.3 to 253.5]
(Section 2.3.3.3)                                                    1.35T                  2.33T&A                  3.25A                  3.69A                1.73T&A mm yr -1
[0.78 to 1.92]          [1.55 to 3.12]        [2.88 to 3.61]          [3.21 to 4.17]        [1.28 to 2.17]
T, A and T&A indicate assessments based on tide gauge reconstructions (T), satellite altimetry (A), or a combination of both (T&A). The assessment uses tide gauge reconstructions before 1993 and satellite altimetry after 1993.
a For the periods 1971-2018, 1993-2018, 2006-2018 and 1901-2018 the C&#xe1;ceres et al. (2020) linear trends are based on the period up to 2016.
9.6.1.3            Regional Sea Level Change in the Satellite Era                              in the North Atlantic, and 85% of the ocean surface experiencing significant sea level acceleration or deceleration, above instrumental Regional sea level changes are resolved by both tide gauge and                                and post-processing noise. Longer records are available from tide satellite altimetry observations (Hamlington et al., 2020a). Altimeters                        gauges, albeit with variable coverage by basin. Regional departures have the advantage of quasi-global coverage but are limited to                                from GMSL rise are primarily driven by ocean transport divergences a period (1993-present) in which the forced trend response is just                            that result from wind stress anomalies and spatial variability emerging on regional scales (Section 9.6.1.4). An analysis of the                              in atmospheric heat and freshwater fluxes (Section 9.2.4).
local altimetry error budget to estimate 90% confidence intervals on regional sea level trends and accelerations reports that 98% of                            The SROCC (Oppenheimer et al., 2019) noted the occurrence of the ocean surface has experienced significant sea level rise over                              large multiannual sea level variations in the Pacific, associated with the satellite era (Prandi et al., 2021). The same study finds that sea                        the Pacific Decadal Oscillation (PDO) in particular, and involving the level accelerations display a less uniform pattern, with an east-west                          El Nino Southern Oscillation (ENSO), North Pacific Gyre Oscillation dipole in the Pacific, a north-south dipole in the Southern Ocean and                          (NPGO) and Indian Ocean Dipole (IOD; Annex IV; Royston et al., 2018; 1289
 
Chapter 9                                                                                                Ocean, Cryosphere and Sea Level Change Hamlington et al., 2020b). There was intensified sea level rise during        9.6.1.4    Attribution and Time of Emergence the 1990s and 2000s, with 10-year trends exceeding 20 mm yr -1 in the                    of Regional Sea Level Change western tropical Pacific Ocean, while sea level trends were negative on the North American west coast. During the 2010s, the situation            The SROCC (Oppenheimer et al., 2019) attributed anthropogenic reversed, with western Pacific sea level falling at more than 10 mm yr -1    forcing to be the dominant cause of GMSL rise since 1970 (see also (Hamlington et al., 2020b). For the Atlantic Ocean, SROCC described          Section 3.5.3.2), but detection and attribution (Cross-Working regional sea level variability as being driven primarily by wind and heat    Group Box: Attribution in Chapter 1) of 20th century externally flux variations associated with the North Atlantic Oscillation (NAO) and      forced regional sea level changes is more challenging, as regional heat transport changes associated with Atlantic Meridional Overturning        variability is larger (Section 9.6.1.3), and therefore the signal-to-Circulation (AMOC) variability. During periods of subpolar North              noise ratio is smaller (Richter and Marzeion, 2014; Monselesan Atlantic warming, winds along the European coast are predominantly            et al., 2015; Palanisamy et al., 2015). Whereas SROCC assessed with from the south and may communicate steric anomalies onto the                  high confidence that GMSL rise is attributable to anthropogenic continental shelf, driving regional sea level rise, with the reverse during  greenhouse gas emissions, they assessed with medium confidence periods of cooling (Chafik et al., 2019). High rates of sea level rise in the that the regional anomalies in ocean basins are a combination of 9 North Indian Ocean are accompanied by a weakening summer South                the response to anthropogenic greenhouse gas emissions and Asian monsoon circulation (Swapna et al., 2017).                              internal variability.
The Arctic ocean is typically excluded from global sea level studies,        The simulated ocean dynamic and thermosteric response to external owing to the uncertainties associated with resolving sea level in            forcings during 1861-2005 is only larger than simulated internal ice-covered regions, strong variations in gravitational, rotational,          variability in the Southern Ocean and North Pacific on a 1&deg; grid and deformational (GRD) effects, and uncertain glacial isostatic              (Slangen et al., 2015). However, on spatial scales exceeding 2000 km, adjustment (GIA) estimates (Box 9.1). Spanning 1991-2018, a very              a detectable signal is revealed in the last 45 years in 63% of the likely sea level rise of 1.16-1.81 mm yr -1 is observed (Rose et al.,        global ocean area (Richter et al., 2017). The thermosteric change in 2019). Since SROCC, the forced response in regional sea level                the upper 700 m in the period 1970-2005 shows similar observed varies in time with the relative influence of different forcing agents        and simulated forced geographical patterns, and anthropogenic (Fasullo et al., 2020).                                                      forcing accounts for part (North Atlantic, 65%) or all (tropical Pacific, Southern Ocean) of the observed regional mean (Marcos and Amores, The SROCC estimated regional sea level changes from combinations              2014). The influences of greenhouse gases and anthropogenic of the various contributions to sea level change from CMIP5 climate          aerosols can be partially distinguished by considering geographical model outputs, allowing comparison with satellite altimeter and              or vertical ocean temperature variations (Slangen et al., 2015; Bilbao tide gauge observations. Closure of the regional sea level budget is          et al., 2019; Fasullo et al., 2020). Zonal-mean forced ocean dynamic complicated by the fact that regional sea level variability is larger        sea level change alone is not detectable but, using spatial correlation, than GMSL variability. Also, there are more processes that need to            the global geographical pattern during the altimeter period is be considered, such as vertical land movement and ocean dynamical            detectable in sea level trends (Fasullo and Nerem, 2018). This pattern changes (Box 9.1). A number of observation-based studies have                may already or will soon be detectable in individual years, based focused on specific areas, such as the Mediterranean (Garc&#xed;a et al.,          on an analysis of CMIP5 climate model simulations (Bilbao et al.,
2006), the South China Sea (Feng et al., 2012), the east coast of the        2015). Anthropogenic forcing, dominated by greenhouse gases, USA (Frederikse et al., 2017; Piecuch et al., 2018), the North Atlantic      has strengthened the meridional sea level gradient in the Southern basin (Kleinherenbrink et al., 2016) and the north-western European          Ocean since the 1960s (Slangen et al., 2015; Bilbao et al., 2019; continental shelf seas (Frederikse et al., 2016). Studies using tide          Fasullo et al., 2020). New evidence finds that observed zonal-mean gauge data and observation-based estimates of the contributions              total sea level trends during 1993-2018 in all basins are inconsistent find that, while local agreement is not yet possible, the observational      with unforced variability alone, but are consistent with the modelled sea level budget can be closed on a basin scale (Slangen et al.,              response to external forcing (Richter et al., 2020).
2014b; Frederikse et al., 2016, 2018, 2020b). A budget analysis for the GRACE era found that the budget closes in some, but not all,              A region that has been studied intensely in the context of sea coastal regions: substantial parts of the sea level change signal in          level detection and attribution is the tropical Pacific. Observed the North Atlantic could not be explained by steric or barystatic            sea level trends in the tropical Pacific show a PDO-like (Annex IV) changes (Rietbroek et al., 2016). This is in agreement with other            east-west dipole (with a greater rate of rise in the west, see work comparing climate model estimates to 20th-century tide                  Section 9.6.1.3). This dipole does not occur in CMIP5 simulations gauge observations (Meyssignac et al., 2017), where the majority              with the magnitude and duration that was observed in the 1990s of local spatial variability is determined by the ocean dynamic              and 2000s, neither in response to historical forcing, nor as internal component. Vertical land movement is another major cause of local            variability after removing the variability associated with the PDO spatial variability in sea level change and, for instance, relevant for      (Bilbao et al., 2015). Hamlington et al. (2014) did obtain a residual oceanic islands (Forbes et al., 2013; Mart&#xed;nez-Asensio et al., 2019).        trend pattern for 1993-2010 in the tropical Pacific that may link to In summary, the regional sea level budget, using either observations          anthropogenic warming of the tropical Indian Ocean. Allowing for or models, can currently only be closed on basin scales (medium              PDO and ENSO variations, (Royston et al., 2018) describe patches confidence), with large uncertainties remaining on smaller scales.            of the Pacific Ocean where the sea level trend for 1993-2015 is 1290
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                Chapter 9 distinguishable from temporally correlated noise. The acceleration in                        will not emerge before late in the century. Adding the projected sea eastern Pacific sea level rise is largely accounted for by variations                        level change from land ice mass loss and groundwater extraction resembling PDO and ENSO (Hamlington et al., 2020a).                                          strengthens and modifies the forced signal, leading to times of emergence 10 to 20 years earlier in most parts of the ocean, except In the future, the anthropogenic signal in regional sea level change                        in regions close to sources of mass loss, with emergence over 50% of from ocean density and dynamics is projected to emerge first in                              the ocean area by 2020, and nearly everywhere by 2100 (medium regions with relatively small internal variability, such as the tropical                    confidence) (Lyu et al., 2014; Richter et al., 2017).
Atlantic Ocean and the tropical Indian Ocean (Jord, 2014; Lyu et al.,
2014; Richter and Marzeion, 2014; Bilbao et al., 2015). The signal                          In summary, detection of forced regional changes for some ocean is projected to emerge over 50% of the ocean area by the 2040s                              areas in recent decades is possible (medium confidence), but (Lyu et al., 2014), but in regions where variability is large and                            attribution of regional sea level change to forcings over longer projected changes are small, such as the Southern Ocean, the signal                          periods (20th century) and for all ocean basins is not yet possible.
9 Cross-Chapter Box 9.1 l Global Energy Inventory and Sea Level Budget Coordinators: Matthew D. Palmer (United Kingdom), Aim&#xe9;e B.A. Slangen (The Netherlands)
Contributors: Gufinna Aalgeirsd&#xf3;ttir (Iceland), F&#xe1;bio Boeira Dias (Finland/Brazil), Catia M. Domingues (Australia, United Kingdom/
Brazil), Gerhard Krinner (France/Germany, France), Johannes Quaas (Germany), Lucas Ruiz (Argentina)
Increased atmospheric greenhouse gas emissions since the 19th century have led to a net positive radiative forcing of Earths climate (Sections 2.2 and 7.3) and a corresponding accumulation of energy in the Earth system. Quantification of this energy gain is essential to our understanding of observed climate change, and for estimates of climate sensitivity (Section 7.5). The global energy inventory is closely linked to our understanding of observed global sea level change, through the energy associated with loss of land-based ice and the effect of thermal expansion associated with ocean warming (Box 9.1, Sections 2.3.3.1 and 9.6.1; Table 9.5).
Cross-Chapter 9.1, Figure 1 l Global Energy Inventory and Sea Level Budget. (a) Observed changes in the global energy inventory for 1971-2018 (shaded time series) with component contributions as indicated in the figure legend. Earth System Heating for the whole period and associated uncertainty is indicated to the right of the plot (red bar = central estimate; shading = very likely range); (b) Observed changes in components of global mean sea level for 1971-2018 (shaded time series) as indicated in the figure legend. Observed global mean sea level change from tide gauge reconstructions (1971-1993) and satellite altimeter measurements (1993-2018) is shown for comparison (dashed line) as a three-year running mean to reduce sampling noise. Closure of the global sea level budget for the whole period is indicated to the right of the plot (red bar = component sum central estimate; red shading = very likely range; black bar = total sea level central estimate; grey shading = very likely range). Full details of the datasets and methods used are available in Annex I. Further details on energy and sea level components are reported in Table 7.1 and Table 9.5.
1291
 
Chapter 9                                                                                                            Ocean, Cryosphere and Sea Level Change Cross-Chapter Box 9.1 (continued)
The Earth system gained substantial energy over the period 1971-2018 (high confidence), with an assessed very likely range of 325-546 ZJ or 0.43-0.72 W m-2 expressed per unit area of the Earths surface (Cross-Chapter Box 9.1, Figure 1a; Section 7.2, Box 7.2).
Ocean warming dominates the energy inventory change (high confidence), accounting for 91% of the observed energy increase for the period 1971-2018, with upper-ocean warming (0-700 m) accounting for 56% (Section 7.2). Much smaller amounts went into melting of ice (3%) and heating of the land (5%) and atmosphere (1%). Overall, the percentage contributions are similar to those reported in IPCCs Fifth Assessment Report (AR5) for the period 1971-2010 (Rhein et al., 2013).
The observed global mean sea level (GMSL) budget is assessed through comparison of the sum of individual components of GMSL change with independent observations of total GMSL change from tide gauge and satellite altimeter observations (Cross-Chapter Box 9.1, Figure 1b; Sections 2.3.3 and 9.6.1 and Table 9.5). The assessed sum of the observed components indicates that GMSL very likely increased by 72 mm to 117 mm over the period 1971-2018 (Table 9.5), with the largest contributions from ocean thermal 9      expansion (50%) and melting of ice sheets and glaciers (42%). The assessed total GMSL change (Section 2.3.3) for the period 1971-2018 has a very likely range of 73-146 mm and, as a result, the sea level budget is closed for this period (Cross-Chapter Box 9.1, Figure 1b; Section 9.6.1, Table 9.5).
The sea level budget closure demonstrates improved quantification of the processes of observed GMSL change for this period relative to previous IPCC assessments (Church et al., 2013b; Oppenheimer et al., 2019). A related assessment presented in Chapter 7 demonstrates closure of the global energy budget (high confidence) (Box 7.2) and strengthens the confidence in scientific understanding of both of these key aspects of climate change.
9.6.2        Paleo Context of Global and Regional                                        Proxy constraints on GMSL and global ice volume are assessed in Sea Level Change                                                            Sections 2.3.2.4. and 2.3.3.3 (see also FAQ 9.1). This section updates prior assessments of drivers of past GMSL changes and climatically As SROCC (Oppenheimer et al., 2019) noted, paleo sea level records                        coherent areas of relative sea level (RSL) variability. GMSL changes are provide information on past ice-sheet changes, and process-based                          framed in terms of global mean surface temperature (GMST) but noting ice-sheet models of past warm periods inform equilibrium responses.                        that amplified high-latitude warming is a robust equilibrium response However, given uncertainties in paleo sea level and polar paleoclimate,                    to elevated CO2 (Masson-Delmotte et al., 2013): polar air temperatures and limited temporal resolution of paleo sea level records, there is low                  during past warm periods were up to twice the GMST changes shown confidence in the utility of paleo sea level records for quantitatively                    in Table 9.6. The SROCC assessment that past multi-metre sea level informing near-term GMSL change. Nonetheless, the paleorecord does                        changes have resulted from significant ice-sheet changes beyond those contextualize sea level and can test projection models (see also FAQ 1.3).                presently observed is confirmed (very high confidence).
Table 9.6 l Reference ranges of age, global mean surface temperature, atmospheric carbon dioxide (CO2) concentration, and global mean sea level (GMSL) for the paleo periods discussed in this chapter.
GMST relative to                    CO2                  Global Mean Sea Years Paleo Period                                                            1850-1900                  Sections 2.2.3.1                Level (GMSL)
Cross-Chapter Box 2.1 Section 2.3.1.1                and 2.2.3.2                Section 2.3.3.3 Early Eocene Climatic Optimum (EECO)                    53-49 Ma                  +10&deg;C to +18&deg;C              1150 to 2500 ppm                +70 to +76 m Mid-Pliocene Warm Period (MPWP)                        3.3-3.0 Ma                  +2.5&deg;C to +4&deg;C                360 to 420 ppm                  +5 to +25 m Marine Isotope Stage (MIS) 11                        about 424-395 ka                0.5&deg;C +/- 1.6&deg;Ca              265 to 286 ppm                  +6 to +13 m Last Interglacial (LIG)                              about 129-116 ka              +0.5&deg;C to +1.5&deg;C              266 to 282 ppm                  +5 to +10 m Last Glacial Maximum (LGM)                              21-19 ka                    -5&deg;C to -7&deg;C                188 to 194 ppm                -125 to -134 m Last Deglacial Transition                                18-11 ka                          n/a                    193 to 271 ppm                -120 to -50 m Early Holocene                                        11.65-6.5 ka                        n/a                    250 to 268 ppm                -50 to -3.5 m Mid-Holocene                                            6.5-5.5 ka                +0.2&deg;C to +1.0&deg;C              260 to 268 ppm                -3.5 to +0.5 m Last Millennium                                        850-1850 CE                -0.14&deg;C to +0.24&deg;C              278 to 285 ppm              -0.05 to +0.03 m a
Based on one study (Irval et al., 2020) relative to SST values around year 2000.
1292
 
Ocean, Cryosphere and Sea Level Change                                                                                                  Chapter 9 9.6.2.1    Mid-Pliocene Warm Period                                      concur that MIS 11 was an extremely long interglacial that exhibited positive annual at 0.5&deg;C +/- 1.6 &deg;C (Irval et al., 2020) and summer During the mid-Pliocene Warm Period (MPWP), GMST was                      at 2.1&deg;C-3.4 &deg;C (Robinson et al., 2017) temperature anomalies 2.5&deg;C-4&deg;C warmer than 1850-1900 (medium confidence) and GMSL              (de Wet et al., 2016). The GMSL was 6-13 m above present (medium was between 5 and 25 m higher than today (medium confidence)              confidence) (Section 2.3.3.3). The Greenland Ice Sheet lost 4.5-6 m (Table 9.6 and Section 2.3.3.3). The AR5 (Masson-Delmotte et al.,        (Reyes et al., 2014) or about 6.1 m (3.9-7 m, 95% confidence) sea 2013) concluded that ice-sheet models consistently produce                level equivalent (SLE) by about 7 kyr after peak summer warmth near-complete deglaciation of the Greenland and West Antarctic            (Robinson et al., 2017), with marine-based ice from AIS (Blackburn ice sheets, and multi-meter loss of the East Antarctic Ice Sheet          et al., 2020) contributing 6.4-8.8 m SLE at this time (Mas e Braga (EAIS) in response to MPWP climate conditions. Studies since AR5          et al., 2021). Agreement between GMSL and ice-sheet reconstructions have yielded a consistent but broader range, due in part to larger        gives high confidence in identifying a high sensitivity of both ice ensembles exploring more parameters (DeConto and Pollard, 2016;          sheets to the protracted duration of thermal forcing, even at low Yan et al., 2016; DeConto et al., 2021). Partly on the basis of these    warming levels (Reyes et al., 2014; Robinson et al., 2017; Irval et al.,
studies, SROCC proposed a plausible upper bound on GMSL of 25 m        2020; Mas e Braga et al., 2021). Modelled mean mass loss rates for (low confidence) with evidence suggesting an Antarctic contribution      the Greenland Ice Sheet of 0.4 m kyr -1 during MIS 11 (Robinson          9 of anywhere between 5.4 and 17.8 m.                                      et al., 2017) are indistinguishable from recent mass loss rates averaged over 1992-2018 (Section 9.4.1.1). In summary, geological The MPWP climate had substantial polar amplification, up to 8&deg;C          reconstructions and numerical simulations consistently show that above pre-industrial levels in Arctic Russia (Section 7.4.4.1; Fischer    past warming levels of <2&deg;C (GMST) are sufficient to trigger multi-et al., 2018). Ice-sheet model simulations indicate that Northern        metre mass loss from both the Greenland and Antarctic ice sheets Hemisphere glaciation was limited to high-elevation regions in            if maintained for millennia (high confidence), in agreement with eastern and southern Greenland (medium confidence) (Figure 9.17;          SROCC findings for comparable warming levels during MIS 5e, the De Schepper et al., 2014; Yan et al., 2014; Koenig et al., 2015;          Last Interglacial.
Dowsett et al., 2016; Berends et al., 2019) with Northern Hemisphere glaciation only becoming more widespread from the (cooler) late          9.6.2.3    Last Interglacial Pliocene (Bachem et al., 2017; Blake-Mizen et al., 2019; Knutz et al.,
2019; S&#xe1;nchez-Montes et al., 2020). Southern Hemisphere glaciation        The AR5 found that the Last Interglacial (LIG) GMSL was >5 m was characterized by an Antarctic Ice Sheet (AIS) reduced in volume      (very high confidence) but <10 m (high confidence). Their best from the present (medium confidence) (Figure 9.18; Dowsett et al.,        estimate of 6 m was based on two studies (Kopp et al., 2009; 2016; Berends et al., 2019; Grant et al., 2019; Miller et al., 2020) with Dutton and Lambeck, 2012). The SROCC concluded that, during the mountain ice fields in the Andes of South America (De Schepper et al.,    LIG, Greenlands contribution to the GMSL highstand (the highest 2014). Ice-sheet models are inconsistent in the magnitude of the sea      sea levels during the LIG) of 6-9 m increased gradually, whereas level contribution from Antarctica (DeConto and Pollard, 2016; Yan        the Antarctic contribution occurred early, from about 129 ka. Due et al., 2016; Golledge et al., 2017b; Berends et al., 2019; DeConto      to widely varying reconstructions from model studies (Greenland) et al., 2021) but near-field sedimentological reconstructions support    and the paucity of direct evidence of ice-sheet change (Antarctic),
precessionally modulated and eccentricity-paced multi-metre sea          the magnitude of sea level contributions from both ice sheets was level contributions from the Wilkes Subglacial Basin over 3-5 kyr        assigned low confidence.
(Patterson et al., 2014; Bertram et al., 2018). In summary, under a past warming level of around 2.5&deg;C-4&deg;C, ice sheets in both hemispheres        Since AR5, information has improved about the LIG, when GMST were reduced in extent compared to present (high confidence).            was about 0.5&deg;C-1.5&deg;C above 1850-1900 (medium confidence)
Proxy-based evidence (Section 2.3.3.3) combined with numerical            (Section 2.3.1.1). The LIG had higher summer insolation than present modelling indicates that, on millennial time scales, the GMSL            and polar amplified sea surface and surface air temperatures that contribution arising from ice sheets was >5 m (high confidence)          reached >1&deg;C-4&deg;C and >3&deg;C-11 &deg;C in the Arctic respectively (Landais or >10 m (medium confidence) (Figures 9.17 and 9.18; Moucha and          et al., 2016; Capron et al., 2017; Fischer et al., 2018). Mean annual and Ruetenik, 2017; Berends et al., 2019; Dumitru et al., 2019).              maximum summer ocean temperatures peaked early (129-125 ka) in the interglacial period, reaching 1.1 +/- 0.3 &deg;C above the modern 9.6.2.2    Marine Isotope Stage 11                                      global mean (Shackleton et al., 2020) with summer anomalies of 2.5&deg;C-3.5 &deg;C in the Southern Ocean (Bianchi and Gersonde, 2002)
The SROCC (Meredith et al., 2019) noted that Greenland may                and spatially variable timing (Chadwick et al., 2020). It is virtually have been ice-free for extensive periods during Pleistocene              certain that GMSL was higher than today, likely by 5-10 m (medium interglaciations, implying a high sensitivity of the Greenland Ice        confidence) (Section 2.3.3.3). Global mean thermal expansion Sheet to warming levels close to present day. The AR5 (Church et al.,    peaked at about 0.9 +/- 0.3 m early in the LIG (about 129 ka), declining 2013b) assigned medium confidence to a Marine Isotope Stage              to modern levels by about 127 ka (Shackleton et al., 2020). With no 11 (MIS 11) GMSL of 6-15 m above present, requiring a loss of            more than 0.3 +/- 0.1 m of GMSL rise from glaciers (Section 9.5.1), at much of the Greenland and West Antarctic ice sheets, and a possible      most 1.0 +/- 0.3 m of the GMSL rise originated from sources other than contribution from East Antarctica. High-resolution multi-proxy sea        the polar ice sheets.
surface temperature reconstructions and climate model simulations 1293
 
Chapter 9                                                                                              Ocean, Cryosphere and Sea Level Change Recent LIG ice-sheet simulations agree that peak loss from the            9.6.2.5      Last Deglacial Transition: Meltwater pulse 1A Greenland Ice Sheet occurred late (125-120 ka; Goelzer et al., 2016; Tabone et al., 2018; Plach et al., 2019) when Northern Hemisphere          During Meltwater pulse 1A (MWP-1A), GMSL very likely (medium insolation was greater than at present (medium confidence) (Capron        confidence) rose by 8-15 m (Liu et al., 2016). Consistent with AR5, the et al., 2017), consistent with inferences from marine sediment records    drivers of this rapid rise remain ambiguous. The spatial patterns of RSL (Hatfield et al., 2016; Irval et al., 2020) and far-field GMSL indicators change over this interval are inadequately observed to constrain the (Rohling et al., 2019). Best estimates of the GMSL contribution from      relative contributions of the North American and Antarctic ice sheets (Liu Greenland (Figure 9.17) differ between models: 1 m (Albrecht et al.,      et al., 2016). Modelling studies of the North American Ice Sheet permit 2020; Clark et al., 2020), 1-2 m (Calov et al., 2015; Goelzer              a 3-6 m (Gregoire et al., 2016) or 6-9 m contribution over the duration et al., 2016; Bradley et al., 2018), up to 3 m (Tabone et al., 2018; Plach of MWP-1A (Tarasov et al., 2012). Sedimentological evidence (Weber et al., 2019), and >5 m (Yau et al., 2016). There is high confidence      et al., 2014; Bart et al., 2018) provides near-field evidence for an Antarctic that the response time of the Greenland Ice Sheet to LIG warming          contribution, consistent with modelling studies (Golledge et al., 2014; was multi-millennial, and high confidence that it contributed to LIG      Stuhne and Peltier, 2015), but does not constrain the magnitude of the GMSL change, but low agreement in the contribution magnitude.              contribution. A recent statistical analysis of Norwegian Sea and Arctic 9                                                                            Ocean sediments suggests a 3-7 m contribution from the Eurasian Ice Far-field GMSL records suggest that the AIS contributed to LIG sea        Sheet (Brendryen et al., 2020), a possibility not considered in AR5 or level from 129.5-125 ka (Figure 9.18) but direct evidence is sparse.      the meta-analysis of Liu et al. (2016). In summary, MWP-1A appears Thinning of part of the WAIS is interpreted from a 130-80 ka hiatus in    to have been driven by a combination of melt in North America (high the Patriot Hills horizontal ice core record (Turney et al., 2020). Marine confidence), Eurasia (low confidence), and Antarctica (low confidence),
sediment records suggest a dynamic response of the Wilkes Subglacial      but the budget is not closed.
Basin (WSB) of the EAIS during this period, indicating a response time scale of 1000-2500 yr (Wilson et al., 2018), consistent with modelling    9.6.2.6      Holocene studies (Mengel and Levermann, 2014; Golledge et al., 2017b; Sutter et al., 2020). Isotopic changes in the Talos Dome ice core are      Around half (50-60 m) of the GMSL rise since the LGM occurred inconsistent with local surface lowering, limiting retreat to 0.4-0.8 m    during the early Holocene at a sustained rate of about 15 m kyr -1 from SLE from this sector (Sutter et al., 2020). Ice-sheet models forced with  around 11.4-8.2 ka (Lambeck et al., 2014), possibly punctuated by unmodified atmosphere-ocean models (Goelzer et al., 2016; Clark            abrupt meltwater pulses (Smith et al., 2011; Carlson and Clark, 2012; et al., 2020) simulate 3-4.4 m SLE mass loss, primarily from the WAIS,    Trnqvist and Hijma, 2012; Harrison et al., 2019). An abrupt (about with no retreat in WSB (e.g., Figure 9.18). Models forced with proxy-      1.1 m) sea level rise around 8.2 ka was associated with drainage of the based or ad hoc LIG ocean temperature anomalies (DeConto and              pro-glacial Agassiz and Ojibway lakes, attributed to accelerated melt Pollard, 2016; Sutter et al., 2016) indicate collapse of West Antarctica  from collapsing Laurentide Ice Sheet ice saddles (Matero et al., 2017).
under 2&deg;C-3&deg;C ocean forcing yielding 3-7.5 m sea level contribution,      The Laurentide Ice Sheet provided the greatest contribution (27 m) to but modest or no retreat in the WSB. Based on limited evidence and        early Holocene GMSL (Peltier et al., 2015; Roy and Peltier, 2017), the limited agreement between models, there is low confidence in both          Scandinavian Ice Sheet contributed about 2 m from the beginning of the magnitude and timing of LIG mass loss from the AIS.                    the Holocene until its demise by around 10.5 ka, (Cuzzone et al., 2016),
while the Barents Sea Ice Sheet contributed a small but unknown In summary, paleo-environmental and modelling studies indicate            amount (Patton et al., 2015, 2017; Auriac et al., 2016). The Greenland that, under past warming of the level achieved during the LIG              Ice Sheet contributed about 4 m, consistent with ice thinning rates (ca. 0.5&deg;C-1.5&deg;C), it is likely that both the Greenland and Antarctic ice  inferred from the Camp Century ice core (Lecavalier et al., 2017; sheets responded dynamically over multiple millennia (high confidence). McFarlin et al., 2018). Recent estimates of Antarctic contributions during the early Holocene vary considerably from about 1.2 m to 9.6.2.4    Last Glacial Maximum                                          8.5 m (Whitehouse et al., 2012; Ivins et al., 2013; Argus et al., 2014; Briggs et al., 2014; Golledge et al., 2014; Pollard et al., 2016; Roy and At the Last Glacial Maximum (LGM) geological proxies and GIA models        Peltier, 2017; Albrecht et al., 2020). In summary, the early Holocene was indicate that GMSL was 125-134 m below present (Section 2.3.3.3            characterized by steadily rising GMSL as global ice sheets continued and Figures 9.17 and 9.18). New studies have not changed AR5s            to retreat from their LGM extents. This steady rise was punctuated by conclusions regarding the size or timing of the LGM and last glacial      abrupt pulses during episodes of rapid meltwater discharge.
termination, but have further examined the LGM sea level budget.
Based on a synthesis of multiple prior studies, (Simms et al., 2019)      In the middle Holocene, GMST peaked at 0.2&deg;C-1.0&deg;C higher than estimated central 67% probability contributions to the LGM lowstand        1850-1900 temperature between 7 and 6 ka (Section 2.3.1.1.2).
(i.e., lowest levels during the LGM) of 76 +/- 7 m from the North            GMSL rise slowed coincidently with final melting of the Laurentide ice American Laurentide Ice Sheet, 18 +/- 5 m from the Eurasian Ice Sheet,      sheet by 6.7 +/- 0.4 ka (Ullman et al., 2016), after which only Greenland 10 +/- 2 m from Antarctica, 4 +/- 1 m from Greenland, 5.5 +/- 0.5 m from        and Antarctic ice sheets could have contributed significantly. At 6 ka, glaciers, and 2.4 +/- 0.3 m due to an increase in ocean density. Of the      GMSL was -3.5 to +0.5 m (medium confidence) (Section 2.3.3.3).
residual, up to about 1.4 m may be ascribed to groundwater, leaving        Simulations of the Holocene Thermal Maximum give a Greenland Ice a shortfall of 16 +/- 10 m yet to be allocated among land ice reservoirs    Sheet broadly consistent with geological reconstructions so, despite or lakes.                                                                  uncertainties regarding the timing of minimum ice-sheet volume 1294
 
Ocean, Cryosphere and Sea Level Change                                                                                                Chapter 9 and extent, there is medium confidence that minima were reached        9.6.3      Future Sea Level Changes at different times in different areas during the period 8-3 ka BP (Larsen et al., 2015; Young and Briner, 2015; Briner et al., 2016). This section first assesses sea level projections since AR5 (Church Geochronological and numerical modelling studies indicate that it      et al., 2013b) and including SROCC (Oppenheimer et al., 2019) based is likely (medium confidence) that the period of smaller-than-present  on Representative Concentration Pathways (RCPs; Section 9.6.3.1).
ice extent in all sectors of Greenland persisted for at least 2000      Process-level assessments in sections 9.2.4, 9.4.1.3, 9.4.1.4, 9.4.2.5, to 3000 years (Larsen et al., 2015; Young and Briner, 2015; Briner      9.4.2.6 and 9.5.1.3 are synthesized (Section 9.6.3.2) to produce new et al., 2016; Nielsen et al., 2018). Based on ice-sheet modelling and  global mean and regional sea level projections based on the Shared carbon-14 (14C) dating (Kingslake et al., 2018) suggested that West    Socio-economic Pathways up to 2150 (Section 9.6.3.3) and on global Antarctic grounding lines retreated prior to around 10 ka BP, followed  warming levels up to 2100 (Section 9.6.3.4). Long-term global mean by a readvance. Other studies from the same region conclude            sea level (GMSL) projections, both at 2300 and on multimillennial that retreat was fastest from 9-8 ka BP (Spector et al., 2017), or      time scales, are also assessed (Section 9.6.3.5).
from 7.5-4.8 ka BP (Venturelli et al., 2020). Marine geological evidence indicates open marine conditions east of Ross Island by        Sections 9.6.3.3 and 9.6.3.4 present likely ranges of the new global 8.6 +/- 0.2 ka BP (McKay et al., 2016). In the western Weddell Sea,      mean sea levels, incorporating only processes in whose projections        9 Johnson et al. (2019) reported rapid glacier thinning from 7.5-6 ka BP. there is at least medium confidence, consistent with headline Hein et al. (2016) concluded that the fastest thinning further south    projections in AR5 and SROCC. As emphasized by SROCC, there is took place from 6.5-3.5 ka BP, potentially contributing 1.4-2 m to      a substantial likelihood that sea level rise will be outside the likely GMSL. Geophysical data indicate stabilization or readvance in this      range. As described in Box 1.1, since the definition of likely refers to area around 6 +/- 2 ka BP (Wearing and Kingslake, 2019). In coastal      at least 66% probability, there may be as much as a 34% probability Dronning Maud Land (East Antarctica) rapid thinning occurred            that the processes in which there is at least medium confidence will 9-5 ka BP (Kawamata et al., 2020), whereas glaciers in the Northern    generate outcomes outside the likely range. Furthermore, additional Antarctic Peninsula receded during the period 11-8 ka BP and            processes in which there is low confidence (Section 9.4.2.4; Box 9.4) readvanced to their maximal extents by 7-4 ka BP (Kaplan et al.,        may also contribute to sea level change. The presentation of likely sea 2020). In summary, higher-than-pre-industrial GMST during the          level change (Tables 9.8-9.9 and in Figures 9.27, 9.29) is therefore mid-Holocene coincided with recession of the Greenland Ice Sheet        accompanied by a low confidence range intended to reflect potential to a smaller-than-present extent (high confidence). Multiple lines      contributions from additional processes under high-emissions of evidence give high confidence that thinning or retreat in parts      scenarios. The low confidence range incorporates ice-sheet projections of Antarctica during the Holocene took place at different times in      based on Structured Expert Judgement (SEJ) - that is, a formal, different places. However, limited data means there is only low        calibrated method of combining quantified expert assessments that confidence in whether or not the ice sheet as a whole was smaller      incorporates all potential processes - and projections from an AIS than present during the mid-Holocene.                                  model that includes the marine ice cliff instability (a specific uncertain process not generally included in ice-sheet models; Section 9.4.2.4).
In summary, both proxies and model simulations indicate that GMSL changes during the early to mid-Holocene were the result of episodic    9.6.3.1    Global Mean Sea Level Projections Based on pulses, due to drainage of meltwater lakes, superimposed on a trend                the Representative Concentration Pathways of steady rise due to continued ice-sheet retreat (high confidence).
The AR5 (Church et al., 2013b) generated GMSL projections for the RCPs The combination of tide gauge observations and geological              by combining information from CMIP5 climate models with glacier reconstructions indicates that a sustained increase of GMSL began      and ice-sheet surface mass balance (SMB) models and assessments between 1820-1860 and led to a 20th-century GMSL rise that was          of projected ice-sheet dynamic and land-water storage contributions very likely (high confidence) faster than in any preceding century in  (Section 9.6.3.2). The SROCC (Oppenheimer et al., 2019) updated AR5 the last 3000 years (Section 2.3.3.3). At a regional level, tide gauge  projections based on a revised assessment of the AIS contribution to and geological data from the North Atlantic and Australasia show        GMSL rise. The AR5 and SROCC employ a baseline period of 1986 inflections in RSL trends between 1895-1935, with an increase of        to 2005, which is updated in this Report to a baseline period of 1995 0.8 to 2.5 mm yr -1 across the inflection (Gehrels and Woodworth,      to 2014 (Section 1.4.1). Between these two periods, GMSL rose by 2013). A statistical meta-analysis of globally distributed geological  3 cm, and this correction is applied to projections from previous reports and tide gauge data (Kopp et al., 2016) found that, in all 20 examined  to allow comparison (Table 9.8). Accounting for this shift, SROCC regions with geological records stretching back at least 2000 years,    concludes that, with medium confidence, GMSL will rise between the rate of RSL rise in the 20th century was greater than the local    0.40 (0.26-0.56, likely range) m (RCP2.6) and 0.81 (0.58-1.07 m, likely average over 0-1700 CE. In four of the 20 regions, all in the North    range) m (RCP8.5) by 2100 relative to 1995-2014. The AR5 and SROCC Atlantic (Connecticut, New Jersey, North Carolina, and Iceland), the    GMSL projections for the 2007-2018 period have been shown to be 19th century rate was also greater than the 0-1700 CE average          consistent with observed trends in GMSL and regional weighted mean (90% confidence interval). In summary, rates of RSL rise exceeding      tide gauges (J. Wang et al., 2021).
the pre-industrial background rate of rise are apparent in parts of the North Atlantic in the 19th century (medium confidence), and in most    Since AR5, a number of projections of GMSL rise have been published of the world in the 20th century (high confidence).                    based on the RCPs (Kopp et al., 2014, 2017; Slangen et al., 2014b; 1295
 
Chapter 9                                                                                                                  Ocean, Cryosphere and Sea Level Change Grinsted et al., 2015; Jackson et al., 2016; Mengel et al., 2016; Bakker                  MICI because there is low confidence in the current ability to quantify et al., 2017; Bittermann et al., 2017; Le Bars et al., 2017; Nauels                      MICI (Section 9.4.2.4). Low confidence is also ascribed to projections et al., 2017; Wong et al., 2017; Goodwin et al., 2018; Nicholls et al.,                  based on SEJ, because individual experts participating in the SEJ study 2018; Le Cozannet et al., 2019; Palmer et al., 2020). See Garner                          may have incorporated processes in whose quantification there is et al. (2018) or a database (Tables 9.SM.5, 9.SM.6). Some studies                        low confidence, and the experts reasoning has not been examined in also produced associated global sets of regional projections (Kopp                        detail. In general, the range of GMSL projections based on ice-sheet et al., 2014, 2017; Slangen et al., 2014b; Le Cozannet et al., 2019;                      models not incorporating MICI overlaps with, but is lower than, Palmer et al., 2020). Since SROCC (Le Cozannet et al., 2019) focused                      projections incorporating MICI or employing SEJ (Figure 9.25).
on the low end of the probability distribution of GMSL rise, Palmer et al. (2020) extended projections beyond 2100 using a climate                            There is high agreement across published GMSL projections for model emulator (Cross-Chapter Box 7.1), and Horton et al. (2020)                          2050, and there is little sensitivity to emissions scenario (Figure 9.25, conducted a survey of 106 sea level experts, providing additional                        left panel). Up to 2050, projections are broadly consistent context for interpreting sea level rise projections for 2100 and 2300.                    with extrapolation of the observed acceleration of GMSL rise (Sections 2.3.3.3, 9.6.1.1 and 9.6.1.2). Considering only projections 9 As noted by SROCC, the largest differences between projections of                        incorporating ice-sheet processes in whose quantification there is at GMSL in 2100 are due to the ice-sheet projection method, which                            least medium confidence, the GMSL projections for 2050, across all generally fall into one of three categories: (i) projections from                        emissions scenarios, fall between 0.1 and 0.4 m (5th-95th percentile ice-sheet models that represent processes where there is at least                        range). Projections incorporating MICI or SEJ do not extend this range medium confidence (Sections 9.4.1.2 and 9.4.2.2); (ii) projections                        under RCP2.6 or RCP4.5 but do extend the upper part of the range to from an Antarctic ice-sheet model that incorporates the marine                            0.6 m under RCP8.5. On the basis of these studies, we therefore ice cliff instability (MICI; Section 9.4.2.4; DeConto and Pollard,                        have high confidence that GMSL in 2050 will be between 0.1 and 2016); or (iii) projections based on SEJ (Sections 9.4.1.3, 9.4.1.4,                      0.4 m higher than in 1995-2014 under low- and moderate-emissions 9.4.2.5 and 9.4.2.6; Bamber and Aspinall, 2013; Bamber et al.,                            scenarios, and between 0.1 and 0.6 m under high-emissions scenarios.
2019). Low confidence is ascribed to projections incorporating 2050 GMSL Projections                                                            2100 GMSL Projections AR6 SSP5-8.5 SROCC RCP 8.5 AR5 Survey SEJ MICI MED AR6 SSP2-4.5 SROCC RCP 4.5 AR5 MICI MED AR6 SSP1-2.6 SROCC RCP 2.6                                                                                                          17 th -83 rd percentile AR5                                                                                                              5th -95 th percentile Survey                                                                                                                17 th -83 rd (low confidence; see caption)
MICI                                                                                                              5th -95 th (low confidence)
MED 0      0.1      0.2    0.3      0.4    0.5      0.6 0                    0.5                  1                  1.5                2              2.5 m                                                                                m Figure 9.25 l Literature global mean sea level (GMSL) projections (m) for 2050 (left) and 2100 (right) since 1995-2014, for RCP8.5/SSP58.5 (top set),
RCP4.5/SSP24.5 (middle set), and RCP2.6/SSP12.6 (bottom set). Projections are standardized to account for minor differences in time periods. Thick bars span from the 17th-83rd percentile projections, and thin bars span the 5th-95th percentile projections. The different assessments of ice-sheet contributions are indicated by MED (ice-sheet projections include only processes in whose quantification there is medium confidence), MICI (ice-sheet projections which incorporate marine ice cliff instability),
and SEJ (structured expert judgement) to assess the central range of the ice-sheet projection distributions. Survey indicates the results of a 2020 survey of sea level experts on global mean sea level (GMSL) rise from all sources (Horton et al., 2020). Projection categories incorporating processes in which there is low confidence (MICI and SEJ) are lightly shaded. Dispersion among the different projections represents deep uncertainty, which arises as a result of low agreement regarding appropriate conceptual models describing ice-sheet behaviour and low agreement regarding probability distributions used to represent key uncertainties. Individual studies are shown in Tables 9.SM.5 and 9.SM.6. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1296
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                              Chapter 9 Conversely, there is low agreement across published GMSL projections                            RCP4.5, and 0.4 and 2.4 m under RCP8.5. In summary, RCP-based for 2100, particularly for higher-emissions scenarios, as well as a higher                      projections published since AR5 show high agreement for 2050, degree of sensitivity to the choice of emissions scenario (Figure 9.25,                          but exhibit broad ranges and low agreement for 2100, particularly right panel). Considering only projections representing processes in                            under RCP8.5.
whose quantification there is at least medium confidence, the GMSL projections for 2100 fall between 0.2 and 1.0 m (5th-95th percentile                            9.6.3.2        Drivers of Projected Sea Level Change range) under RCP2.6 and RCP4.5, and between 0.3 and 1.6 m under RCP8.5. Considering also projections incorporating MICI or SEJ                                  This section describes the choices made for the contributions (low confidence), the projections for 2100 fall between 0.2 and 1.0 m                            to the updated global mean and regional sea level projections (5th-95th percentile range) under RCP2.6, 0.2, and 1.6 m under                                  (Section 9.6.3.3) based on assessments in this Report and compares Table 9.7 l Methods used to project the drivers of global mean sea level (GMSL) and relative sea level (RSL) change in the Shared Socio-economic Pathway (SSP) and warming-level-based projections of GMSL, RSL and extreme sea level (ESL) change. Section numbers indicate location of primary assessment text.
Driver of Global Mean or Regional Sea Level change SROCC Projection Method                                                          AR6 Projection method                                      9 CMIP5 ensemble drift-corrected zostoga,                      Two-layer emulator with climate sensitivity calibrated to AR6 assessment Thermal expansion with surrogates derived from climate system heat              (Supplementary Material 7.SM.2) and expansion coefficients calibrated to emulate (Section 9.2.4.1) content where not available                                  CMIP6 models (Supplementary Material 9.SM.4.2 and 9.SM.4.3)
Surface mass balance:                                        Medium confidence processes up to 2100: Emulated Ice Sheet Model Intercomparison scaled cubic polynomial fit to global mean surface            Project for CMIP6 (ISMIP6) simulations (Box 9.3; Edwards et al., 2021)
Greenland Ice Sheet                temperature (GMST)                                            Medium confidence processes after 2100: Parametric model fit to ISMIP6 simulations (excluding peripheral glaciers)
Dynamics:                                                    up to 2100, extrapolated based on either constant post-2100 rates or a quadratic (Sections 9.4.1.3 and 9.4.1.4)
Quadratic function of time, calibrated based on multi-        interpolation to the multi-model assessed 2300 range (Supplementary Material 9.SM.4.4) model assessment                                              Low confidence processes: Structured expert judgement (Bamber et al., 2019)
Medium confidence processes up to 2100: p-box including: (i) Emulated ISMIP6 simulations (Edwards et al., 2021); and (ii) Linear Antarctic Response Model Intercomparison Project (LARMIP-2) simulations (Levermann et al., 2020) augmented by AR5 surface mass balance model (Box 9.3)
Medium confidence processes after 2100: p-box including: (i) AR5 parametric AIS Antarctic Ice Sheet model; and (ii) LARMIP-2 simulations augmented by AR5 surface mass balance model (excluding peripheral glaciersa)  Multi-model assessment applied to CMIP6 models, with both methods extrapolated based on either constant (Sections 9.4.2.5 and 9.4.2.6) post-2100 rates or a quadratic interpolation to the multi-model assessed 2300 range (Section 9.6.3.2)
Low confidence processes: (i) Single-ice-sheet-model ensemble simulations incorporating marine ice cliff instability (DeConto et al., 2021); and (ii) structured expert judgement (Bamber et al., 2019)
Up to 2100: Emulated GlacierMIP (Marzeion et al., 2020; Edwards et al., 2021)
Glaciers (including                                                                              simulations (Box 9.3)
Power law function of integrated GMST fit to peripheral glaciers) glacier models                                                Beyond 2100: AR5 parametric model re-fit to GlacierMIP (Supplementary Material (Section 9.5.1.3) 9.SM.4.5; Marzeion et al., 2020)
Groundwater depletion:
Groundwater depletion:
combination of: (i) continuation of early 21st-century Population/groundwater depletion relationship calibrated based on Konikow (2011) trends; and (ii) land-surface hydrology models Land-water storage                                                                              and Wada et al. (2012, 2016)
(Wada et al., 2012)
(Section 9.6.3.2)                                                                                Water impoundment:
Water impoundment:
Population/dam impoundment relationship calibrated based on Chao et al. (2008),
combination of: (i) continuation of historical rate; and adjusted for new construction following Hawley et al. (2020) for 2020 to 2040 (ii) assumption of no net impoundment after 2010 Ocean dynamic sea level                                                                          Distribution derived from CMIP6 ensemble zos field after linear drift removal CMIP5 ensemble zos field after polynomial drift removal (Section 9.2.4.2)                                                                                (Supplementary Material 9.SM.4.2 and 9.SM.4.3)
Gravitational, rotational, and deformational effects          Sea level equation solver (Slangen et al., 2014b) driven by projections of ice-sheet, glacier, and land-water storage changes (Section 9.6.3.2)
Glacial isostatic adjustment      Glacial Isostatic Adjustment model, with ice history from Spatio-temporal statistical model of tide gauge data (updated from Kopp et al., 2014) and other drivers of vertical      mean of the Australian National University (ANU) and (Supplementary Material 9.SM.4.6) land motion (Section 9.6.3.2)      ICE-5G reconstructions a
Ice-sheet models include some of the larger islands in the Antarctic periphery, so there is some overlap in the projected glacier contribution and the projected Antarctic contribution, but the effect is estimated to be on the order of 0.5-1 cm or less (Edwards et al., 2021).
1297
 
Chapter 9                                                                                                                            Ocean, Cryosphere and Sea Level Change the updated projections to AR5 (Church et al., 2013b) and SROCC                                        changes on mass loss), and the dynamic contribution was estimated (Oppenheimer et al., 2019) (Tables 9.7 and 9.8). Since there is no                                    based on a multi-model assessment interpolated as a quadratic single model that can directly compute all of the contributions                                        function of time.
to sea level change (Box 9.1), the contributions to sea level are computed separately and then combined (Tables 9.8 and 9.9). For                                        For processes whose projections we have at least medium consistency with global surface air temperature (GSAT) projections                                    confidence in, the updated projections use emulated Ice Sheet (Section 4.3.1.1), and assessment of equilibrium climate sensitivity                                  Model Intercomparison Project for CMIP6 (ISMIP6) projections of the (ECS) and transient climate response (TCR; Section 7.5), temperature-                                  Greenland Ice Sheet (Section 9.4.1.3; Figure 9.17; Tables 9.2 and 9.7; dependent projections (thermal expansion, ice sheets, glaciers) are                                    Box 9.3). Since the ISMIP6 emulator does not account for temporal forced by GSAT projections from a two-layer energy budget emulator                                    correlation, a parametric fit to the ISMIP6 results is used to calculate (Smith et al., 2018) that is calibrated to be consistent with the                                      rates of change (Supplementary Material 9.SM.4.4). For projections assessment of ECS and TCR (Box 7.1, Supplementary Material 7.SM.2).                                    beyond 2100 (when the ISMIP6 simulations end), the polynomial fit Throughout, likely ranges are assessed based on the combination                                        is extrapolated based on two alternate approaches: (i) an assumption of uncertainty in the GSAT distribution and uncertainty in the                                        of constant rates of mass change after 2100; and (ii) for SSP12.6 9 relationships between GSAT and changes to individual components.                                      and SSP58.5, a quadratic function of time extending to 2300 based In general, 17th-83rd percentile results, incorporating both GSAT                                      on the multi-model assessment of contributions under RCP2.6 and and sea level process uncertainty, are interpreted as likely ranges.                                  RCP8.5 at 2300 (Section 9.4.1.4). Differences between the two This is distinct from the approach used by AR5, which interpreted                                      approaches are small up to 2150, and since the latter approach is not the 5th-95th percentile range of CMIP5 projections, and therefore                                      available for all scenarios, only the former (constant rates) is used of GMSL projections driven by them, as likely ranges. The shift in                                    for time series projections up to 2150. Both approaches are used for interpretation is consistent with the use of the emulator for GSAT                                    examining uncertainty in the timing of different levels of GMSL rise (Box 4.1, Cross-Chapter Box 7.1). Very likely ranges are not assessed                                  and to inform projections for the year 2300 (Section 9.4.1.4). For because of the potential for processes in whose projections there                                      2100, the ISMIP6 emulator yields the likely contribution from the is currently low confidence to substantially augment total projected                                  Greenland Ice Sheet shown in Table 9.2 and Figure 9.17, representing GMSL change.                                                                                          a slight narrowing from AR5 projections.
9.6.3.2.1 Global mean thermosteric sea level rise                                                      9.6.3.2.3 Antarctic Ice Sheet In AR5 and SROCC, global mean thermosteric sea level rise                                              For the Antarctic Ice Sheet (AIS), AR5 applied a temperature-based was derived from the 21 members of the CMIP5 ensemble that                                            scaling approach for SMB and a quadratic function of time, calibrated provided the required variables (Section 9.2.4.1). The AR5 and                                        to a multi-model assessment, for dynamic contributions. The SROCC SROCC removed drift estimated based on a pointwise polynomial                                          used a new assessment based on the results of five process-based fit to pre-industrial control simulations. They extended projections                                  studies (Section 9.4.2.5). For processes in whose projections we to scenarios not provided by the models by calculating the heat                                        have at least medium confidence, the likely range projections content of the climate system from GMST and net radiative flux,                                        for the AIS are based on: (i) the emulated ISMIP6 ensemble; and and converting this to global mean thermosteric sea level rise                                        (ii) the LARMIP-2 ensemble, augmented with AR5 parametric using each models diagnosed expansion efficiency coefficient. The                                    Antarctic SMB model. The GMSL projections are produced with both AR5 and SROCC derived the associated uncertainties by assuming                                        distributions and combined in a p-box (Kriegler and Held, 2005; a normal distribution, with the 5th-95th percentile CMIP5 ensemble                                    Le Cozannet et al., 2017), which represents the upper and lower interpreted as the likely range. In this Report, global mean thermosteric                              bounds of the distribution (Section 9.4.2.5, Box 9.3 and Table 9.3).
sea level rise is derived from a two-layer energy budget emulator                                      A likely range is then identified, spanning the lower of the two consistent with the assessment of ECS and TCR (Section 9.2.4.1;                                        17th percentile projections and the higher of the two 83rd percentile Supplementary Material 9.SM.4.2 and 9.SM.4.3). Despite the change                                      projections,5 with the median taken as the mean of the medians of in methodology, this leads to a likely global mean thermosteric                                        the two projections. Since the ISMIP6 emulator does not account contribution (17th-83rd percentile) between 1995-2014 and 2100                                        for temporal correlation, the AR5 parametric AIS model is substituted that represents a minimal change from AR5 and SROCC (Table 9.8).                                      for the emulator in the p-box for rates of change. As AR5 projections are modestly lower than those from the ISMIP6 emulator, this 9.6.3.2.2 Greenland Ice Sheet                                                                          substitution modestly broadens the likely range at the low end for projections of rate and changes beyond 2100. For projections beyond The AR5 and SROCC projected the Greenland surface-mass balance                                        2100 (when the ISMIP6 and LARMIP-2 simulations end), the AIS using a cubic polynomial fit to a regional climate model as a function                                simulations are extrapolated using the same two approaches as the of global mean surface temperature (with a log-normal scaling factor                                  Greenland Ice Sheet (GrIS) projections (Section 9.4.1.4). The likely reflecting uncertainty in surface-mass balance models, and another                                    ranges to 2100 are consistent with SROCC (Table 9.8).
scaling factor reflecting the positive feedback of ice-sheet elevation 5    Note that the use of this approach implies that the likely ranges are likely in the use of the term to mean 66-100% probable; this is distinct from usage in SROCC, where the likely range was defined to have a precise 66% probability.
1298
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                  Chapter 9 9.6.3.2.4 Low confidence ice-sheet projections                                                9.6.3.2.5 Glaciers To test the possible effect of additional ice-sheet processes for which                      In AR5 and SROCC, global glacier mass changes were derived from there is low confidence (Sections 9.4.1.3, 9.4.1.4, 9.4.2.5, 9.4.2.6 and                      a power law of integrated global mean surface temperature change fit 9.6.3.1, and Box 9.4), two additional approaches are considered. For                          to results from four different glacier models. The updated projections both the Greenland and Antarctic ice sheets, we produce sensitivity                          use emulated GlacierMIP projections (Section 9.5.1.3; Box 9.3). Since cases employing the SEJ projections of Bamber et al. (2019), mapping                          the GlacierMIP emulator does not account for temporal correlation 2&deg;C and 5&deg;C stabilization scenarios to SSP12.6 and SSP58.5,                                and terminates, along with the GlacierMIP simulations, in 2100, respectively. For the AIS, we produce an additional sensitivity case                          we employ a parametric fit to the GlacierMIP simulations, with using projections, which incorporate MICI (DeConto et al., 2021),                            a functional form similar to that employed by AR5, to calculate mapping projections for RCP2.6 and RCP8.5 to SSP12.6 and SSP58.5.                          rates of change and extrapolate changes beyond 2100 (up to For the Greenland Ice Sheet, the SEJ projections indicate the potential                      a maximum potential contribution of 0.32 m; see Supplementary for outcomes outside the corresponding likely ranges (Table 9.8).                            Material 9.SM.4.5). This approach leads to a median glacier For the AIS, there is no evidence from these studies to suggest an                            contribution that is a minimal change (Table 9.8) from AR5 and important role under lower-emissions scenarios for processes in                              SROCC and a modest narrowing of likely ranges (Section 9.5.1.3). For                  9 whose projections we have low confidence. By contrast, for SSP58.5,                          RCP2.6, AR5 projected 0.10 (0.04 to 0.16, likely range) m, compared the SEJ and MICI projections exhibit 17th-83rd percentile ranges of                          to 0.09 (0.07 to 0.11) m projected for SSP12.6. For RCP8.5, AR5 0.02-0.56 m and 0.19-0.53 m by 2100, consistent with one another                              projected a likely contribution of 0.17 (0.09 to 0.25) m, compared to but considerably broader than the likely contribution for medium                              0.18 (0.15 to 0.21) m projected here.
confidence processes of 0.03-0.34 m. This lower level of agreement for higher-emissions scenarios reflects the deep uncertainty in the                          9.6.3.2.6 Land-water storage AIS contribution to GMSL change under higher-emissions scenarios (Box 9.4). This deep uncertainty grows after 2100: by 2150, under                            In AR5 and SROCC, the groundwater depletion contribution to SSP58.5, medium confidence processes likely lead to a -0.1-0.7                              GMSL rise was based on combining results from two approaches:
m AIS contribution, while SEJ- and MICI-based projections indicate                            one assuming a continuation of early 21st-century trends 0.0-1.1 m and 1.4-3.7 m, respectively.                                                        (Konikow, 2011); and the other using land-surface hydrology Table 9.8 l Global mean sea level projections between 1995-2014 and 2100 for total change and individual contributions, median values, (likely) ranges of the process-based model ensemble for RCP 2.6 (from AR5 (Church et al., 2013a) and SROCC (Oppenheimer et al., 2019)) and SSP12.6 (this Report), and for RCP8.5 (from AR5 (Church et al., 2013a) and SROCC (Oppenheimer et al., 2019)) and SSP58.5 (this Report). Values for AR5 (Church et al., 2013a) and SROCC (Oppenheimer et al.,
2019) are adjusted from the 1986-2005 baseline used in past reports. Only the Antarctic contribution changed between AR5 (Church et al., 2013a) and SROCC (Oppenheimer et al., 2019). Unshaded cells represent processes in which there is medium confidence; shading indicates the inclusion of processes in which there is low confidence. For the MICI- and SEJ-based projections, parenthetical numbers represent the 17th-83rd percentile of the associated probability distributions, not assessed likely ranges.
RCP2.6                                                              SSP12.6 Medium confidence m relative to 1995-2014                      AR5                        SROCC                                                  MICI                        SEJ processes Thermal expansion (Section 9.2.4.1)                      0.14 (0.10-0.19) m                                                    0.14 (0.11-0.18) m Greenland (Section 9.4.1.3)                              0.07 (0.03-0.11) m                                      0.06 (0.01-0.10) m                        0.13 (0.07-0.30) m Antarctica (Section 9.4.2.5)              0.06 (-0.04 to +0.16) m        0.04 (0.01-0.11) m        0.11 (0.03-0.27) m          0.08 (0.06-0.12) m      0.09 (-0.01 to +0.25) m Glaciers (Section 9.5.1.3)                                0.10 (0.04-0.16) m                                                    0.09 (0.07-0.11) m Land-water storage (Section 9.6.3.2)                    0.05 (-0.01 to +0.11) m                                                0.03 (0.01-0.04) m Total (2100)                                0.41 (0.25-0.58) m          0.40 (0.26-0.56) m        0.44 (0.33-0.62) m          0.41 (0.35-0.48) m          0.53 (0.38-0.79) m Total (2150)                                    0.29-0.63 m              0.56 (0.40-0.73) m        0.68 (0.46-0.99) m          0.74 (0.62-0.91) m          0.84 (0.56-1.34) m GMSL rate, 2080-2100 (mm yr ) -1 4.4 (2.0-6.8) mm yr -1 4 (2-6) mm yr -1 5.2 (3.2-8.0) mm yr -1 5.1 (4.3-6.2) mm yr -1 5.9 (2.8-11.0) mm yr -1 RCP8.5                                                              SSP58.5 Medium confidence m relative to 1995-2014                      AR5                        SROCC                                                  MICI                        SEJ processes Thermal expansion (Section 9.2.4.1)                      0.31 (0.24-0.38) m                                                    0.30 (0.24-0.36) m Greenland (Section 9.4.1.3)                              0.14 (0.08-0.27) m                                      0.13 (0.09-0.18) m                        0.23 (0.10-0.59) m Antarctica (Section 9.4.2.5)              0.04 (-0.08 to +0.14) m        0.12 (0.03-0.28) m        0.12 (0.03-0.34) m          0.34 (0.19-0.53) m          0.21 (0.02-0.56) m Glaciers (Section 9.5.1.3)                                0.17 (0.09-0.25) m                                                    0.18 (0.15-0.20) m Land-water storage (Section 9.6.3.2)                    0.05 (-0.01 to +0.11) m                                                0.03 (0.01-0.04) m Total (2100)                                0.71 (0.49-0.95) m          0.81 (0.58-1.07) m        0.77 (0.63-1.01) m          0.99 (0.82-1.19) m          1.00 (0.70-1.60) m Total (2150)                                    0.34-1.35 m              1.27 (0.80-1.79) m        1.32 (0.98-1.88) m          3.48 (2.57-4.82) m          1.79 (1.22-2.94) m GMSL rate, 2080-2100 (mm yr ) -1 11.2 (7.5-15.7) mm yr  -1 15 (10-20) mm yr    -1 12.1 (8.6-17.6) mm yr  -1 23.1 (17.5-30.1) mm yr  -1 16.0 (9.8-28.9) mm yr -1 1299
 
Chapter 9                                                                                            Ocean, Cryosphere and Sea Level Change models (Wada et al., 2012). Together, these yielded a range of            ocean dynamic sea level change along the coast (Section 9.2.3.6) and about 0.02-0.09 m of GMSL rise by 2080-2099. The rate of water            in semi-enclosed basins, such as the Mediterranean, where coarse impoundment in reservoirs was likewise based on two approaches:            models can misrepresent key dynamic processes. Regional high-one assuming the continuation of the average rate over 1971-2010          resolution models can improve projections of coastal ocean dynamic (and thus -0.01 to -0.03 m by 2080-2099; Chao et al., 2008); and          sea level change (Section 12.4; Hermans et al., 2020), but have not the other assuming no net impoundment after 2010 (Lettenmaier              been implemented at a global scale.
and Milly, 2009). Together, these yield a GMSL contribution from groundwater impoundment of -0.03 to 0 m. Combining groundwater            9.6.3.2.8 Gravitational, rotational and deformational effects depletion and water impoundment led AR5 and SROCC to infer a projected range of -0.01 to +0.11 m by 2100.                            Gravitational, rotational, and deformational (GRD) effects (Box 9.1) lead to distinct variations in the RSL change pattern, which are similar In the updated projections, a statistical relationship is applied, linking across a range of benchmarked GRD solvers (Martinec et al., 2018; historical and future SSP global population to dam impoundment and        Palmer et al., 2020). There is high confidence in the understanding groundwater extraction (Rahmstorf et al., 2012; Kopp et al., 2014).        of GRD processes. RSL rise associated with GRD is very likely to be 9 The population/groundwater depletion relationship is calibrated            largest in the Pacific, due to the combined effects of projected GrIS, based on the same studies used in AR5 (Konikow, 2011; Wada et al.,        AIS and glacier mass loss (high confidence) (e.g., Kopp et al., 2014; 2012), reduced by about 20% to account for water retained on land          Slangen et al., 2014b; Larour et al., 2017; Mitrovica et al., 2018).
(Wada et al., 2016). The population/dam impoundment relationship          The GRD effect associated with mass loss from an ice sheet is sensitive is calibrated based on Chao et al. (2008). However, while historically    to the spatial distribution of that mass loss. For example, the GRD dam impoundment has been declining with population, recent                contribution to RSL rise in Australia will be larger for Antarctic mass literature shows that planned dam construction considerably                loss sourced fromthe Antarctic Peninsula than for Antarctic mass loss exceeds the historical trend (Zarfl et al., 2015; Hawley et al., 2020). sourced fromThwaites Glacier. In parts of north-eastern North Over 2020-2040, the impoundment contribution to GMSL rise                  America and north-western Europe, GRD effects associated with based on past trends would be about -0.1 mm yr -1, compared to            mass loss from southern Greenland will lead to an RSL fall, whereas about -0.5 mm yr -1 if all currently planned dams are built (Hawley        mass loss from northern Greenland will lead to an RSL rise (high et al., 2020) and the statistical projection is therefore augmented        confidence) (Figure 9.26; Larour et al., 2017; Mitrovica et al., 2018).
by an additional -0.4 to 0.0 mm yr -1 over 2020-2040 to account            The AR5 and SROCC computed RSL patterns using a gravitationally for the possible effects of planned dam construction. As in AR5 and        self-consistent GRD solver given the amounts, locations and timing SROCC, climatically driven changes to land-water storage (LWS)            of the projected barystatic sea level changes driven by glaciers, ice have not been included in published sea level projections, as they        sheets and LWS (Church et al., 2013b). A similar GRD solver is used are not well quantified (e.g., Jensen et al., 2019) or are considered      in the updated projections (following Slangen et al., 2014b). The negligible (e.g., permafrost, Section 9.5.2). This approach yields        Earth model used is based on the Preliminary reference Earth model a likely global-mean land-water storage contribution (Figure 9.27,        (PREM: Dziewonski and Anderson, 1981), and is elastic, compressible Table 9.8) that is slightly lower and narrower than the AR5 and            and radially stratified.
SROCC likely ranges. Since the projections are explicitly population driven, these projections also exhibit a weak scenario dependence,        9.6.3.2.9 Glacial isostatic adjustment and other drivers with a contribution around 0.01 m higher under SSP3 than under                        of vertical land motion other scenarios.
Glacial Isostatic Adjustment (GIA) leads to vertical land motion 9.6.3.2.7 Ocean dynamic sea level                                          (VLM; see Box 9.1) and changes in sea surface height, both of which contribute to RSL change. GIA uncertainty is caused by uncertainty In AR5 and SROCC, the ocean dynamic sea level contribution to RSL          in the rheological structure of the solid Earth, which drives the projections was derived from the CMIP5 ensemble, after removing the        longer-term viscous Earth deformation, as well as uncertainty in drift estimate based on pre-industrial control simulations. This Report    the modelled global ice history (e.g., Whitehouse, 2018). In AR5 and uses updated simulations from the CMIP6 ensemble (Section 9.2.4.2;        SROCC, GIA contributions to RSL change were calculated using a sea Supplementary Material 9.SM.4.2) to project the ocean dynamic              level equation solver with an ice-sheet history taken as the mean sea level contribution to RSL change (Section 9.2.4.2; Figure 9.26).      of the ICE5G (Peltier et al., 2015) and ANU (Lambeck et al., 2014)
To produce ocean dynamic sea level projections consistent with the        ice-sheet models. Since AR5, new global models are emerging that global mean thermosteric projections from the two-layer energy            more rigorously treat ice and Earth structure uncertainty (Caron budget emulator, we follow the approach of Kopp et al. (2014),            et al., 2018). However, there is also a growing recognition that lateral employing a correlation between global-mean thermosteric sea              variations in Earth structure limit the utility of global models that treat level change and ocean dynamic sea level derived from the CMIP6            the solid Earth as though it were laterally uniform (Love et al., 2016; ensemble (Supplementary Material 9.SM.4.3). Since CMIP6 models            Huang et al., 2019; T. Li et al., 2020).
are of fairly coarse resolution (typically about 100 km), and even the models participating in HighResMIP (near 10 km resolution)            As noted by SROCC, VLM from sources other than GIA - including do not capture all the phenomena that contribute to coastal ocean          tectonics and mantle dynamic topography, volcanism, compaction, dynamic sea level change, there is low confidence in the details of        and anthropogenic subsidence - can be locally important, producing 1300
 
Ocean, Cryosphere and Sea Level Change                                                                                                                    Chapter 9 VLM rates comparable to or greater than rates of GMSL change.                          constant over the projected period. In areas where rapid subsidence Complete global projections of these processes are not available                      occurs in a cluster of tide gauges (e.g., the western Gulf of Mexico),
because of the small spatial scales, the sensitivity of subsidence to                  the associated rates are interpolated between the tide gauges.
local human activities, and the stochasticity of tectonics (Wppelmann                In areas where the available tide gauges exhibit large, tectonically and Marcos, 2016; Oppenheimer et al., 2019). Therefore, integrated                    driven VLM that changes considerably in rate over short distances RSL projections to date have either included only the component of                    (e.g., Alaska and the Bering Strait), a sizable uncertainty propagates VLM associated with GIA (as in AR5 and SROCC), or used a constant                      into the RSL projections (Figure 9.26). Rates of RSL rise are likely to long-term background rate of change (including both GIA and other                      be underestimated due to subsidence in shallow strata that are not long-term drivers of VLM) estimated from historical tide gauge trends                  recorded by tide gauges (Keogh and Trnqvist, 2019) and in some (e.g., Kopp et al., 2014). The updated projections use the second                      locations may therefore be minimum values, especially if anomalously approach and extrapolate the field of long-term background rates                      high subsidence rates associated with fluid extraction are also of RSL change, including long-term VLM derived from tide gauges,                      considered (e.g., Minderhoud et al., 2017). Therefore, depending to global coverage using a spatio-temporal statistical approach                        on location, there is low to medium confidence in the GIA and VLM (Supplementary Material 9.SM.4.6; Kopp et al., 2014). The combined                    projections employed in this Report. In many regions, higher-fidelity GIA and long-term VLM is assumed to be scenario independent and                        projections would require more detailed regional analysis.                      9 SSP1-2.6                    SSP5-8.5 SSP1-2.6                                                                          SSP5-8.5 Figure 9.26 l Median global mean and regional relative sea level projections (m) by contribution for the SSP12.6 and SSP58.5 scenarios. Upper time series: Global mean contributions to sea level change as a function of time, relative to 1995-2014. Lower maps: Regional projections of the sea level contributions in 2100 relative to 1995-2014 for SSP58.5 and SSP12.6. Vertical land motion is common to both Shared Socio-economic Pathways (SSPs). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1301
 
Chapter 9                                                                                                              Ocean, Cryosphere and Sea Level Change 9.6.3.3      Sea Level Projections to 2150 Based on Shared                                (Table 9.5 and Section 2.3.3.3), which would imply a likely GMSL Socio-economic Pathway Scenarios                                            rise of 0.24 m (0.23-0.25 m) by 2050 and 0.73 m (0.69-0.77 m) by 2100. This extrapolation would also imply a likely rate of GMSL rise of Up to 2050, consistent with AR5 and SROCC, GMSL projections                                7.5 (7.4-7.6) mm yr -1over 2040-2060 and 11.2 (10.6-11.8) mm yr -1 exhibit little scenario dependence (high confidence) (Figure 9.27 and                      over 2080-2100. Over the satellite period, the observed acceleration Table 9.9) with likely (medium confidence) sea level rise between the                      has been driven primarily by ice-sheet contributions (Section 9.6.1.2 baseline period (1995-2014) and 2050 of 0.19 (0.16-0.25) m under                          and Table 9.5); in the median projections for SSP37.0 and SSP58.5, SSP12.6 and 0.23 (0.20-0.30) m under SSP58.5. These projections                          these accelerations are projected to continue at a slightly lower fall centrally within the range of published projections for RCP2.6                        level, while the GMSL acceleration is augmented by an acceleration and RCP8.5 (Section 9.6.3.1).                                                              of thermal expansion and glacier loss associated with rising global temperature. Overall, these extrapolations imply that, under Beyond 2050, the scenarios increasingly diverge. Between the baseline                      SSP11.9, SSP12.6, and SSP24.5, the GMSL acceleration is projected period (1995-2014) and 2100, processes in whose projection there                          to decrease from its current level.
is medium confidence drive likely GMSL rise of 0.44 (0.32-0.62) m 9 and 0.77 (0.63-1.01) m under SSP12.6 and SSP58.5, respectively                          While ice-sheet processes in whose projection there is low confidence (Tables 9.8, 9.9). While derived using substantially updated methods,                      have little influence up to 2100 on projections under SSP11.9 and these projections are broadly consistent with SROCC, which projected                      SSP12.6 (Table 9.9), this is not the case under higher emissions likely GMSL rise of 0.41 (0.26-0.56) m and 0.81 (0.58-1.07) m                              scenarios, where they could lead to GMSL rise well above the likely under RCP2.6 and RCP8.5, respectively, over this period. They are                          range. In particular, under SSP58.5, low-confidence processes modestly higher than those of AR5, which projected likely GMSL rise                        could lead to a total GMSL rise of 0.6-1.6 m over this time period of 0.41 (0.25-0.58) m under RCP2.6 and 0.71 (0.49-0.95) m under                            (17th-83rd percentile range of p-box, including SEJ- and MICI-based RCP8.5 (Figure 9.25, Table 9.8). They are also broadly consistent with                    projections), with 5th-95th percentile projections extending to projections produced by driving AR5 methods with CMIP6 temperature                        0.5-2.3 m (low confidence). The assessed low confidence range is and thermal expansion projections, which leads to 0.44 (0.27-0.61) m                      slightly narrower than, but broadly consistent with, the full 0.4-2.4 m under SSP12.6 and 0.73 (0.49-1.02) m under SSP58.5 (Hermans                              range of published 5th-95th percentile projections for RCP8.5 since et al., 2021). The SSP12.6 and SSP58.5 projections are consistent                        AR5 (Section 9.6.3.1) - including those based on SEJ or incorporating with the ranges of published projections for RCP2.6 and RCP8.5 that                        MICI - and highlights the deep uncertainty in GMSL rise under the do not incorporate MICI or SEJ (Section 9.6.3.1).                                          highest emissions scenarios (Box 9.4). The assessment of the potential contribution of processes in which there is low confidence to GMSL The likely GMSL projections for SSP37.0 and SSP58.5 are                                  rise by 2100 is broadly consistent with the AR5s assessment (Church consistent with a continuation of the GMSL satellite-observed                              et al., 2013b), which concluded that collapse of marine-based sectors rate (very likely 3.25 [2.88-3.61] mm yr -1) and acceleration (very                        of the AIS could cause several tenths of a metre of GMSL rise above likely 0.094 [0.082-0.115] mm yr -2) of GMSL rise over 1993-2018                          the likely range.
Table 9.9 l Global mean sea level projections for five Shared Socio-economic Pathway (SSP) scenarios, relative to a baseline of 1995-2014, in metres.
Individual contributions are shown for the year 2100. Median values (likely ranges) are shown. Average rates for total sea level change are shown in mm yr -1. Unshaded cells represent processes in whose projections there is medium confidence. Shaded cells incorporate a representation of processes in which there is low confidence; in particular, the SSP58.5 low confidence column shows the 17th-83rd percentile range from a p-box including SEJ- and MICI-based projections rather than an assessed likely range. Methods are described in 9.6.3.2.
SSP58.5 SSP11.9              SSP12.6                SSP24.5                SSP37.0                SSP58.5 Low Confidence Thermal expansion            0.12 (0.09-0.15)      0.14 (0.11-0.18)        0.20 (0.16-0.24)        0.25 (0.21-0.30)        0.30 (0.24-0.36)        0.30 (0.24-0.36)
Greenland                    0.05 (0.00-0.09)      0.06 (0.01-0.10)        0.08 (0.04-0.13)        0.11 (0.07-0.16)        0.13 (0.09-0.18)        0.18 (0.09-0.59)
Antarctica                    0.10 (0.03-0.25)      0.11 (0.03-0.27)        0.11 (0.03-0.29)        0.11 (0.03-0.31)        0.12 (0.03-0.34)        0.19 (0.02-0.56)
Glaciers                      0.08 (0.06-0.10)      0.09 (0.07-0.11)        0.12 (0.10-0.15)        0.16 (0.13-0.18)        0.18 (0.15-0.21)        0.17 (0.11-0.21)
Land-water Storage            0.03 (0.01-0.04)      0.03 (0.01-0.04)        0.03 (0.01-0.04)        0.03 (0.02-0.04)        0.03 (0.01-0.04)        0.03 (0.01-0.04)
Total (2030)                  0.09 (0.08-0.12)      0.09 (0.08-0.12)        0.09 (0.08-0.12)        0.10 (0.08-0.12)        0.10 (0.09-0.12)        0.10 (0.09-0.15)
Total (2050)                  0.18 (0.15-0.23)      0.19 (0.16-0.25)        0.20 (0.17-0.26)        0.22 (0.18-0.27)        0.23 (0.20-0.29)        0.24 (0.20-0.40)
Total (2090)                  0.35 (0.26-0.49)      0.39 (0.30-0.54)        0.48 (0.38-0.65)        0.56 (0.46-0.74)        0.63 (0.52-0.83)        0.71 (0.52-1.30)
Total (2100)                  0.38 (0.28-0.55)      0.44 (0.32-0.62)        0.56 (0.44-0.76)        0.68 (0.55-0.90)        0.77 (0.63-1.01)          0.88 (0.63-1.60)
Total (2150)                  0.57 (0.37-0.86)      0.68 (0.46-0.99)        0.92 (0.66-1.33)        1.19 (0.89-1.65)        1.32 (0.98-1.88)        1.98 (0.98-4.82)
Rate (2040-2060)                4.1 (2.8-6.0)          4.8 (3.5-6.8)          5.8 (4.4-8.0)          6.4 (5.0-8.7)            7.2 (5.6-9.7)            7.9 (5.6-16.1)
Rate (2080-2100)                4.2 (2.4-6.6)          5.2 (3.2-8.0)          7.7 (5.2-11.6)          10.4 (7.4-14.8)          12.1 (8.6-17.6)          15.8 (8.6-30.1) 1302
 
Ocean, Cryosphere and Sea Level Change                                                                                                                            Chapter 9 Projected global mean sea level rise under different SSP scenarios 2.5 Median (medium confidence)
Likely range (medium confidence) 2                Satellite extrapolation (see caption)
SSP5-8.5 Likely range of extrapolation                                                            SSP3-7.0 1.5                SSP5-8.5 Low confidence 83rd percentile m                          SSP5-8.5 Low confidence 95th percentile                                    SSP2-4.5 1
0.5 Historical                                                                        SSP1-1.9          SSP1-2.6                2150 medium
                                                                                                                                                          & low confidence 0                                                                                                                                                  projections 1950                              2000                              2050                              2100                        2150            (see caption)        9 Figure 9.27 l Projected global mean sea level rise under different Shared Socio-economic Pathway (SSP) scenarios. Likely global mean sea level (GMSL) change for SSP scenarios resulting from processes in whose projection there is medium confidence. Projections and likely ranges at 2150 are shown on right. Lightly shaded ranges and thinner lightly shaded ranges on the right show the 17th-83rd and 5th-95th percentile ranges for projections including low confidence processes for SSP12.6 and SSP58.5 only, derived from a p-box including structured expert judgement and marine ice-cliff instability projections. Black lines show historical GMSL change, and thick solid and dash-dotted black lines show the mean and likely range extrapolating the 1993-2018 satellite altimeter trend and acceleration. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
SSP1-1.9                                                SSP2-4.5                                                  SSP5-8.5 SSP3-7.0 SSP1-2.6                                                  SSP3-7.0 Figure 9.28 l Regional sea level change at 2100 for different scenarios (with respect to 1995-2014). Median regional relative sea level change from 1995-2014 up to 2100 for: (a) SSP11.9; (b) SSP12.6; (c) SSP24.5; (d) SSP37.0; (e) SSP58.5; and (f) width of the likely range for SSP37.0. The high uncertainty in projections around Alaska and the Aleutian Islands arises from the tectonic contribution to vertical land motion, which varies greatly over short distances in this region. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
While prior assessment reports, starting with the First Assessment                        1.0-1.9 m under SSP58.5. Processes in which there is low confidence Report (Warrick et al., 1990), have focused on projecting GMSL up                        could drive GMSL rise under SSP58.5 to 1.0-4.8 m (17th-83rd to the year 2100, time has progressed, and the year 2100 is now                          percentile) or even 0.9-5.4 m (5th-95th percentile).
within the time frame of some long-term infrastructure decisions.
For this reason, projections up to the year 2150 are also highlighted                    Median projected RSL changes are shown in Figure 9.28, with driving (Table 9.9). Over this time period, assuming no acceleration in                          factors highlighted in Figure 9.26. Approximately 60% (SSP11.9) ice-sheet mass fluxes after 2100, processes in which there is medium                      to 70% (SSP58.5) of the global coastline has a projected median confidence lead to GMSL rise of 0.5-1.0 m under SSP12.6 and                              21st century regional RSL rise within +/-20% of the global mean 1303
 
Chapter 9                                                                                                                    Ocean, Cryosphere and Sea Level Change increase (medium confidence). Consistent with AR5, loss of land                              do so between about 2160 and 2300 under SSP58.5. However, ice mass will be an important contributor to spatial patterns in                              processes in whose projections there is low confidence could lead to RSL change (high confidence), with ocean dynamic sea level being                              substantially earlier exceedances under higher emissions scenarios:
particularly important as a dipolar contributor in the north-west                            under SSP58.5, 1.0 m could be exceeded by about 2080 and Atlantic, a positive contributor in the Arctic Ocean, and a negative                          2.0 m could be exceeded by about 2110 (17th percentile of p-box, contributor in the Southern Ocean south of the Antarctic Circumpolar                          incorporating projections based on SEJ and MICI), with 5th percentile Current (medium confidence) (Section 9.2.4.2). As today, VLM                                  projections as early as about 2070 for 1.0 m and 2090 for 2.0 m.
will remain a major driver of RSL change (high confidence).
Uncertainty in RSL projections is greatest in tectonically active                            9.6.3.4        Sea Level Projections up to 2100 Based areas in which VLM varies over short distances (e.g., Alaska) and in                                        on Global Warming Levels areas potentially subject to large ocean dynamic sea level change (e.g., the north-western Atlantic) (high confidence).                                        Global warming levels represent a new dimension of integration in the AR6 cycle (Section 1.6.2, Cross-Chapter Box 11.1). The SR1.5 An alternative perspective on uncertainty in future sea level rise is                        (Hoegh-Guldberg et al., 2018) concluded that, based on an assessment 9 provided by looking at uncertainty in time rather than elevation;                            of GMSL projections published for 1.5&deg;C and 2.0&deg;C scenarios, there is that is, looking at the range of dates when specific thresholds of                            medium agreement that GMSL in 2100 would be 0.04-0.16 m higher sea level rise are projected to be crossed (Figure 9.29). Considering                        in a 2&deg;C warmer world, compared to a 1.5&deg;C warmer world based only medium confidence processes, GMSL rise is likely to exceed                              on 17-84% confidence interval projections (0.00-0.24 m based 0.5 m between about 2080 and 2170 under SSP12.6 and between                                  on 5-95% confidence interval projections) with a central value of about 2070 and 2090 under SSP58.5. It is likely to exceed 1.0 m                              around 0.1 m. The SR1.5 did not attempt to standardize the definition between about 2150 and some point after 2300 under SSP12.6,                                  of warming-level scenarios, or to examine additional warming levels.
and between about 2100 and 2150 under SSP58.5. It is unlikely                                No new integrated GMSL projections for 1.5&deg;C or 2.0&deg;C scenarios to exceed 2.0 m until after 2300 under SSP12.6, while it is likely to                        have been published since SR1.5.
Projected timing of sealevel rise milestones Under different forcing scenarios and workflow assumptions SSP12.6                                                                SSP58.5 2.0 meters since 1995-2014                                              2.0 m 1.5 m                                                                  1.5 m 1.0 m                                                                  1.0 m SEJ MICI Assessed ice sheets 0.5 m                                                                  0.5 m              No acceleration 2000              2100                2200              2300+        2000                2100                2200              2300+
Figure 9.29 l Timing of when global mean sea level (GMSL) thresholds of 0.5, 1.0, 1.5 and 2.0 m are exceeded, based on four different ice-sheet projection methods informing post-2100 projections. Methods are labelled based on their treatment of ice sheets. No acceleration assumes constant rates of mass change after 2100.
Assessed ice sheet models post-2100 ice-sheet losses using a parametric fit (Supplementary Material 9.SM.4) extending to 2300 based on a multi-model assessment of contributions under RCP2.6 and RCP8.5 at 2300. Structured expert judgement (SEJ) employs ice-sheet projections from Bamber et al. (2019). Marine ice-cliff instability (MICI) combines the parametric fit (Supplementary Material 9.SM3.4) for Greenland with Antarctic projections based on DeConto et al. (2021). Circles, thick bars and thin bars represent the 50th, 17th-83rd and 5th-95th percentiles of the exceedance timing for the indicated projection method. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1304
 
Ocean, Cryosphere and Sea Level Change                                                                                                                            Chapter 9 Table 9.10 l Global mean sea level (GMSL) projections and commitments for exceedance of five global warming levels, defined by sorting GSAT change in 2081-2100 with respect to 1850-1900. Median values and (likely) ranges are in metres relative to a 1995-2014 baseline. Rates are in mm yr -1. Unshaded cells represent processes in whose projections there is medium confidence. Shaded cells incorporate a representation of processes in which there is low confidence; in particular, the SSP58.5 low confidence column shows the 17th-83rd percentile range from a p-box, including projections based on structured expert judgement (SEJ) and marine ice cliff instability (MICI) rather than an assessed likely range. Methods are described in 9.6.3.2.
SSP58.5 1.5&deg;C                    2.0&deg;C                    3.0&deg;C                    4.0&deg;C                  5.0&deg;C Low Confidence Closest SSPs                SSP12.6            SSP12.6/SSP24.5        SSP24.5/SSP37.0              SSP37.0                SSP58.5 Total (2050)            0.18 (0.16-0.24) m      0.20 (0.17-0.26) m      0.21 (0.18-0.27) m        0.22 (0.19-0.28) m      0.25 (0.22-0.31) m        0.24 (0.20-0.40) m Total (2100)          0.44 (0.34-0.59) m      0.51 (0.40-0.69) m      0.61 (0.50-0.81) m        0.70 (0.58-0.92) m      0.81 (0.69-1.05) m        0.88 (0.63-1.60) m Rate (2040-2060)      4.1 (2.9-5.7) mm yr -1 5.0 (3.7-7.0) mm yr -1 6.0 (4.6-8.1) mm yr -1 6.4 (5.0-8.6) mm yr -1 7.2 (5.7-9.8) mm yr -1 7.9 (5.6-16.1) mm yr -1 Rate (2080-2100)      4.3 (2.6-6.4) mm yr -1  5.5 (3.4-8.4) mm yr -1  7.8 (5.3--11.6) mm yr -1  9.9 (7.1-14.3) mm yr -1 11.7 (8.5-17.0) mm yr -1  15.8 (8.6-30.1) mm yr -1 2000-yr commitment 2 to 3 m                2 to 6 m                  4 to 10 m                12 to 16 m              19 to 22 m                                  9 10,000-yr 6 to 7 m                8 to 13 m                10 to 24 m                19 to 33 m              28 to 37 m commitment Most of the contributors to GMSL are more closely tied to time-                            9.6.3.5      Multi-century and Multi-millennial Sea Level Rise integrated GSAT than instantaneous GSAT (Hermans et al., 2021),
which means that sea level projections by warming level can only                          Neither AR5 nor SROCC discussed the sea level commitment be interpreted if the warming levels are linked to a specific time                        associated with historical emissions. Since AR5, new evidence has frame. Here, the warming level projections are defined based on                            suggested that historical emissions up to 2016 will lead to a likely the 2081-2100 GSAT anomaly (Supplementary Material 9.SM.4.7).                              committed sea level rise (i.e., the rise that would occur in the absence Different pathways in GSAT can be followed to reach a certain                              of additional emissions) of 0.7-1.1 m up to 2300, while pledged temperature level, which affects the temporal evolution of the                            emissions through 2030 increase the committed rise to 0.8-1.4 m different contributors to sea level change. For instance, there will                      (Nauels et al., 2019).
be different ice-sheet and glacier responses to a fast increase to a peak warming of 2&deg;C in 2050, followed by a plateau or a decrease,                        Between the baseline period (1995-2014) and 2300, AR5 projected compared to a gradual increase to the same level of warming in                            a GMSL rise of 0.38-0.82 m under a non-specific low-emissions 2100. The sea level projections presented might include different                          scenario and 0.9-3.6 m under a non-specific high-emissions scenario pathways to the same warming level in 2100, which is reflected                            (Table 9.11). The SROCC projected 0.6-1.0 m under RCP2.6 and in the uncertainty ranges, and should therefore be interpreted as                          2.3-5.3 m under RCP8.5 (low confidence). RCP-based projections illustrative of sea level scenarios under a certain warming level.                        for 2300 published since AR5 span a broader range, even excluding studies employing SEJ or MICI, with 17th-83rd percentile projections Projections of likely 21st-century GMSL rise along climate trajectories                    ranging from 0.3-2.9 m for RCP2.6 and 1.7-6.8 m for RCP8.5 leading to different increases in GSAT between 1850-1900 and                              (Table 9.SM.8; Kopp et al., 2014, 2017; Nauels et al., 2017, 2019; 2081-2100 are shown in Table 9.10, along with the SSPs for                                Bamber et al., 2019; Palmer et al., 2020). Conservatively extending the which the temperature-level projections are most closely aligned.                          ISMIP6- and LARMIP-2-based projections beyond 2100 by assuming For example, considering only processes in which there is medium                          no subsequent change in ice-sheet mass flux rates (an approach confidence, from the baseline period (1995-2014) up to 2100, GMSL                          similar to that adopted by Palmer et al. (2020) for the Greenland in a 2&deg;C scenario is likely to rise by 0.40-0.69, which is intermediate                    Ice Sheet and for the Antarctic Ice Sheet dynamics) leads to a GMSL between the projections for SSP12.6 and SSP24.5. GMSL in a 4&deg;C                          change up to 2300 of 0.8-2.0 m under SSP12.6 and 1.9-4.1 m under scenario is likely to rise by 0.58-0.92 m, similar to the projection                      SSP58.5 (17th-83rd percentile), while incorporating the ice-sheet for SSP37.0. Consistent with the discussion in Section 9.6.3.3, there                    contributions for 2300 assessed in Section 9.4.1.4 and Section 9.4.2.6 is deep uncertainty in the projections for temperature levels above                        leads to 0.6-1.5 m and 2.2-5.9 m, respectively. Incorporating Antarctic 3&deg;C, and alternative approaches to projecting ice-sheet changes may                        results from a model with MICI (Section 9.4.2.4), using RCP forcing yield substantially different projections in 4&deg;C and 5&deg;C futures. For                      to inform SSP-based projections, leads to 1.4-2.1 m for SSP12.6 example, employing SEJ ice-sheet projections (Bamber et al., 2019)                        and 9.5-16.2 m for SSP58.5 (DeConto et al., 2021). Incorporating instead of the projections for medium confidence processes only                            the SEJ-based ice-sheet projections of Bamber et al. (2019) for 2&deg;C leads to a 17th-83rd percentile rise between the baseline period                          and 5&deg;C stabilization scenarios yields 1.0-3.1 m for SSP12.6, and (1995-2014) and 2100 of 0.7-1.6 m, rather than 0.7-1.1 m in                                2.4-6.3 m for SSP58.5, although because of the differences in a 5&deg;C scenario.                                                                            scenarios, the SSP12.6 estimates may be overestimated and the SSP58.5 may be underestimated. The eightfold uncertainty range across projection methods under SSP58.5 reflects deep uncertainty in the multi-century response of ice sheets to strong climate forcing.
1305
 
Chapter 9                                                                                                              Ocean, Cryosphere and Sea Level Change Taking into account all these approaches, including published                              suggests a 2000-year commitment at 1.5&deg;C of about 2.3-3.1 m, projections for RCP2.6, under SSP12.6 GMSL will rise between                              with approximately an additional 1.4-2.3 m commitment between 0.3 and 3.1 m by 2300 (low confidence). This projection range                              1.5&deg;C and 2.0&deg;C (i.e., about 3 to 5 m &deg;C-1). Taken together, both indicates that, while SROCC projections under low emissions to 2300                        studies show a 2000-year GMSL commitment of about 2-6 m for are consistent with no ice-sheet acceleration after 2100, there is the                      peak warming of about 2&deg;C, 4-10 m for 3&deg;C, 12-16 m for 4&deg;C, and possibility of a much broader range of outcomes at the high end,                            19-22 m for 5&deg;C (medium agreement, limited evidence) (Table 9.10).
reflected in the range of published GMSL projections. Under SSP58.5,                      GMSL rise continues after 2000 years, leading to a 10,000-year GMSL will rise between 1.7 and 6.8 m by 2300 in the absence of                              commitment of about 6-7 m for 1.5&deg;C of peak warming (based on MICI and by up to 16 m considering MICI, a wider range than AR5                            Clark et al., 2016), and based on both studies of about 8-13 m for or SROCC assessments, but consistent with published projections                            2.0&deg;C, 10-24 m for 3.0&deg;C, 19-33 m for 4.0&deg;C, and 28-37 m for 5&deg;C (low confidence).                                                                          (medium agreement, limited evidence) (Table 9.10).
On still longer time scales, AR5 concluded with low confidence that                        An indicative metric for the equilibrium sea level response can the multi-millennial GMSL commitment sensitivity to warming was                            be provided by comparing paleo GSAT and GMSL during past 9 about 1-3 m &deg;C-1 GSAT increase. Two process-model studies since                            multimillennial warm periods (Sections 2.3.1.1, 2.3.3.3 and 9.6.2; AR5 (Clark et al., 2016; Van Breedam et al., 2020) indicate higher                          Figure 9.9). However, caution is needed as the present and past warm commitments (Figure 9.30). Ice sheets dominate the multi-millennial                        periods differ in astronomical and other forcings (Cross-chapter sea level commitment (Sections 9.4.1.4 and 9.4.2.6), but the two                            Box 2.1) and in terms of polar amplification. The Last Interglacial studies disagree on the relative contribution of the Greenland and                          (likely 5-10 m higher GMSL than today and 0.5&deg;C-1.5&deg;C warmer Antarctic ice sheets. Notably, processes such as MICI (Section 9.4.2.4)                    than 1850-1900; Section 9.6.2; Table 9.6) is consistent with the that are a major factor behind the deep uncertainty in century-scale                        Clark et al. (2016) projections for the 10,000-year commitment AIS response do not appear to have a substantial effect on the                              associated with 1.5&deg;C of warming. Similarly, the Mid-Pliocene Warm multi-millennial magnitude (DeConto and Pollard, 2016). Only one of                        Period (very likely 5-25 m higher GMSL than today and very likely the studies of multimillennial GMSL commitments includes scenarios                          2.5&deg;C-4&deg;C warmer) (Section 9.6.2; Table 9.6) is consistent with consistent with 1.5&deg;C of peak warming (Clark et al., 2016); this study                      the range of 10,000-year commitments associated with 2.5-4&deg;C Table 9.11 l Global mean sea level (GMSL) projections between 1995-2014 and 2300 for total change and individual contributions. Low emissions projections from: AR5 (Church et al., 2013b); RCP2.6 from SROCC (Oppenheimer et al., 2019) and published projections (Table 9.SM.8); and SSP12.6 (from this Report). High emissions projections from: AR5 (Church et al., 2013b); RCP8.5 from SROCC (Oppenheimer et al., 2019) and published projections (Table 9.SM.8); and SSP58.5 (this Report). Values for AR5 (Church et al., 2013b) and SROCC (Oppenheimer et al., 2019) are adjusted from the 1986-2005 baseline used in past reports. Only total values are shown for published ranges. Only the Antarctic contribution changed between AR5 (Church et al., 2013b) and SROCC (Oppenheimer et al.,
2019). If a range is given, it is the 17th-83rd percentile range.
Low                          RCP2.6                                                        SSP12.6 No Ice-sheet        Assessed m relative to                                                        Post-AR5 AR5              SROCC                                    Acceleration        Ice-sheet              MICI                SEJ 1995-2014                                                    Published Range After 2100        Contribution Thermal expansion                      0.07-0.46 m                                                                        0.19-0.35 m Greenland                                  0.14 m                                            0.22-0.39 m                  0.11-0.25 m                  0.28-1.28 m Antarctica                              0.21-0.25 m                                        -0.05 to +1.14 m  -0.14 to +0.78 m        0.71-1.35 m    -0.11 to +1.56 m Glaciers                                    n/a                                                                            0.12-0.29 m Land-water storage              -0.03 m          0.07-0.37 m                                                              0.05-0.10 m Total (2300)                0.38-0.82 m          0.57-1.04 m          0.3-2.9 m              0.8-2.0 m          0.6-1.5 m            1.4-2.1 m        1.0-3.1 m High                        RCP8.5                                                        SSP58.5 Post-AR5 No Ice-Sheet        Assessed m relative to                                                  Published Range AR5              SROCC                                    Acceleration        Ice-sheet              MICI                SEJ 1995-2014                                                      Without (with) after 2100        Contribution MICI Thermal expansion                      0.28-1.80 m                                                                        0.92-1.51 m Greenland                              0.30-1.18 m                                          0.53-0.88 m                  0.32-1.75 m                  0.40-2.23 m Antarctica                    0.02-0.19 m        0.60-2.89 m                              -0.39 to +1.55 m  -0.28 to +3.13 m      6.87-13.54 m      0.03-3.05 m Glaciers                                0.29-0.39 m                                                                            0.32 m Land-water storage                          n/a                                                                            0.05-0.10 m Total (2300)                0.89-3.56 m          2.25-5.34 m    1.7-6.8 (up to 14.1) m      1.7-4.0 m          2.2-5.9 m            9.5-16.2 m        2.4-6.3 m 1306
 
Ocean, Cryosphere and Sea Level Change                                                                                                                              Chapter 9 9
Figure 9.30 l Global mean sea level (GMSL) commitment as a function of peak global surface air temperature. From models (Clark et al., 2016; DeConto and Pollard, 2016; Garbe et al., 2020; Van Breedam et al., 2020) and paleo data on 2000-year (lower row) and 10,000 year (upper row) time scales. Columns indicate different contributors to GMSL rise (from left to right: total GMSL change, Antarctic Ice Sheet, Greenland Ice Sheet, global mean thermosteric sea level rise, and glaciers). Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
of warming, but GMSL reconstructions provide only a weak,                                  Since AR5, a small number of modelling studies have examined the broad constraint on model-based projections. An additional paleo                          reversibility of the multimillennial sea level commitment under carbon constraint comes from the Early Eocene Climatic Optimum, which                            dioxide (CO2) removal, solar radiation modification or local ice shelf indicates that 10-18&deg;C of warming is associated with ice-free                              engineering. The slow response of the deep ocean to forcing leads conditions and a likely GMSL rise of 70-76 m (Sections 2.3.3 and                          to global-mean thermosteric sea level fall occurring long afterward, 9.6.2). Together with model-based projections (Clark et al., 2016;                        even if CO2 levels are restored after a transient increase: global mean Van Breedam et al., 2020), this period suggests that commitment                            thermosteric sea level rise takes more than a millennium to reverse to ice-free conditions would occur for peak warming of about                              (Ehlert and Zickfeld, 2018). Rapid reversion to pre-industrial CO2 7&deg;C-13&deg;C (medium agreement, limited evidence).                                            concentrations has been found to be ineffective at fostering regrowth of the AIS (DeConto et al., 2021) but may reduce the multimillennial On the basis of modelling studies, paleo constraints, single-ice-sheet                    sea level commitment (DeConto and Pollard, 2016). Altering studies finding multimillennial nonlinear responses from both the                          sub-ice-shelf bathymetry (Wolovick and Moore, 2018) or triggering Greenland and Antarctic ice sheets (Sections 9.4.1.4 and 9.4.2.6),                        ice shelf advance through massive snow deposition (Feldmann et al.,
and the underlying physics, we conclude that GMSL commitment                              2019) might interrupt marine ice sheet instability (Section 9.4.2.4) is nonlinear in peak warming on time scales of both 2,000 and                              and thus reduce sea level commitment. A reversion to pre-industrial 10,000 years (medium confidence) and exceeds the AR5 assessment                            Greenland Ice Sheet temperatures with solar radiation modification of 1-3 m &deg;C-1 (medium agreement, limited evidence) (Table 9.9).                            is projected to stop mass loss in Greenland but leads to minimal Although thermosteric sea level will start to decline slowly about                        regrowth (Applegate and Keller, 2015). Based on limited evidence, 2,000 years after emissions cease, the slower responses from the                          carbon dioxide removal, solar radiation modification, and local Greenland and Antarctic ice sheets mean that GMSL will continue                            ice-shelf engineering may be effective at reducing the yet-to-be-to rise for 10,000 years under most scenarios (medium confidence).                        realized sea level commitment, but ineffective at reversing GMSL rise (low confidence).
1307
 
Chapter 9                                                                                                Ocean, Cryosphere and Sea Level Change Box 9.4 l High-end Storyline of 21st-century Sea Level Rise In this box, we outline a storyline (Glossary, Box 10.2; Shepherd et al., 2018) for high-end sea level projections for 2100. This storyline considers processes whose quantification is highly uncertain regarding the timing of their possible onset and/or their potential to accelerate sea level rise. These processes are therefore not considered for the assessed upper bound of likely sea level rise by 2100 in section 9.6.3.3, as the likely range includes only processes that can be projected skilfully with at least medium confidence (based on agreement and evidence).
As noted by SROCC, stakeholders with a low risk tolerance (e.g., those planning for coastal safety in cities and long-term investment in critical infrastructure) may wish to consider global-mean sea level rise above the assessed likely range by the year 2100, because likely implies an assessed likelihood of up to 16% that sea level rise by 2100 will be higher (see also Siegert et al., 2020). Because of our limited understanding of the rate at which some of the governing processes contribute to long-term sea level rise, we cannot currently robustly quantify the likelihood with which they can cause higher sea level rise before 2100 (Stammer et al., 2019).
9 In light of such deep uncertainty, we employ a storyline approach in examining the potential for, and early warning signals of a high-end sea level scenario unfolding within this century. In doing so, we note upfront that the main uncertainty related to high-end sea level rise is when rather than if it arises: the upper limit of 1.01 m of likely sea level range by 2100 for the SSP5-8.5 scenario will be exceeded in any future warming scenario on time scales of centuries to millennia (high confidence), but it is uncertain how quickly the long-term committed sea level will be reached (Section 9.6.3.5). Hence, global mean sea level might rise well above the likely range before 2100, which is reflected by assessments of ice-sheet contributions based on structured expert judgement (Bamber et al.,
2019) leading to a 95th percentile of projected future sea level rise as high as 2.3 m in 2100 (Section 9.6.3.3).
A plausible storyline for such high-end sea level rise in 2100 assumes a strong warming scenario (Section 4.8). The storyline considers faster-than-projected disintegration of marine ice shelves and the abrupt, widespread onset of marine ice cliff instability (MICI) and marine ice sheet instability (MISI) in Antarctica (Section 9.4.2.4), and faster-than-projected changes in both the surface mass balance and dynamical ice loss in Greenland. While conceptual studies provide medium evidence of these processes, substantial uncertainties and low agreement in quantifying their future evolution arise from limited process understanding, limited availability of evaluation data, missing or crude representation in model simulations, their high sensitivity to uncertain boundary conditions and parameters, and/or uncertain atmosphere and ocean forcing (Sections 9.4.1.2; 9.4.2.2).
In Antarctica, high warming might lead to floating ice shelves starting to break up earlier than expected due to processes not yet accounted for in ice-sheet models or in current climate models used to force ice-sheet projections. Such processes include hydrofracturing driven by surface meltwater, and increase in ocean thermal forcing driven by ocean circulation changes (Sections 9.2.2.3, 9.2.3.2 and 9.4.2.3; Hellmer et al., 2012, 2017; Silvano et al., 2018; Hazel and Stewart, 2020). In particular, the Thwaites and Pine Island Glacier ice shelves could potentially disintegrate this century, which might trigger MICI before 2100 (DeConto and Pollard, 2016; DeConto et al., 2021). MISI could potentially develop earlier and faster than simulated by the majority of models if fast flowing ice streams follow plastic, instead of currently assumed more viscous, sliding laws (Sun et al., 2020). Oceanic feedbacks could drive high-end sea level rise by changes in the meltwater-driven overturning circulation in ice cavities that cause additional melting (Jeong et al., 2020);
by a warming of the ocean water in contact with the ice shelves due to increased stratification and thus reduced vertical mixing (Sections 9.2.2.3 and 9.2.3.2; Golledge et al., 2019; Moorman et al., 2020; Sadai et al., 2020); or by an increase in sea ice cover due to increased ocean stratification (Section 9.3.2.1), which could reduce the amount of warm, moist air that reaches the continent, and limit the mass gain from snowfall over the ice sheet (Sadai et al., 2020).
In Greenland, stronger mass loss than currently projected might also occur (Aschwanden et al., 2019; Khan et al., 2020; T. Slater et al., 2020). For example, warming-induced dynamical changes in atmospheric circulation could enhance summer blocking and produce more frequent extreme melt events over Greenland similar to the record mass loss of more than 500 Gt in summer 2019 (Section 9.4.1.1; Delhasse et al., 2018; Sasgen et al., 2020). Cloud processes in polar areas that are not well represented in models could further enhance surface melt (Hofer et al., 2019), as could feedbacks between surface melt and the increasing albedo from meltwater, detritus and pigmented algae (Section 9.4.1.1; Cook et al., 2020). The same ice dynamical processes associated with basal melt and MISI discussed for Antarctica could also occur in Greenland, as long as the ice sheet is in contact with the ocean.
The strength of all these processes is currently understood to depend strongly on global mean temperature and polar amplification, with additional linkages through feedback from global mean sea level (Gomez et al., 2020). These dependencies on a joint forcing imply that processes are strongly correlated. Hence, both their uncertainties and their possible cascading contribution to high-end sea level rise are expected to combine. Therefore, high-end sea level rise can occur if one or two processes related to ice-sheet collapse 1308
 
Ocean, Cryosphere and Sea Level Change                                                                                                    Chapter 9 Box 9.4 (continued) in Antarctica result in an additional sea level rise at the maximum of their plausible ranges (Sections 9.4.2.5 and 9.6.3.3; Table 9.7) or if several of the processes described in this box result in individual contributions to additional sea level rise at moderate levels. In both cases, global-mean sea level rise by 2100 would be substantially higher than the assessed likely range, as indicated by the projections including low confidence processes reaching in 2100 as high as 1.6 m at the 83rd percentile and 2.3 m at the 95th percentile (Section 9.6.3.3).
Identifying the potential drivers of a high-end sea level rise allows identification of sites and observables that can provide early warnings of a much faster sea level rise than the likely range of this and previous reports. One potential site for such monitoring is Thwaites Glacier, which is melting faster in some places and slower in others than models simulate. At this glacier, the effect of tides and channelling of warm water flows on the melting is evident (Milillo et al., 2019), making the floating ice shelf potentially vulnerable to breakup from hydrofracturing, driven by surface meltwater, much earlier than expected. In addition, the glacier is retreating towards a zone with deeper bedrock, which at its present rate of retreat would be reached in 30 years (Yu et al., 2019). Thwaites Glacier is            9 therefore a strong candidate to experience large-scale MISI and/or MICI (Golledge et al., 2019; DeConto et al., 2021), making it the ideal site for monitoring early warning signals of accelerated sea level rise from Antarctica. Such signals could possibly be observed within the next few decades (Scambos et al., 2017).
9.6.4        Extreme Sea Levels: Tides, Surges and Waves                      can result from changes (either positive or negative) in the surge or tidal components, and can include non-linear interactions between An extreme sea level (ESL) refers to an occurrence of exceptionally          tide, surge, and RSL (Arns et al., 2015; Schindelegger et al., 2018).
high or low local sea surface height (Box 9.1). This section focuses on      The positive phase of the 18.6-year nodal cycle of the astronomical oceanographic-driven changes in ESL (Box 9.1).                                tide is a further consideration, contributing to an increased flood hazard relative to the long-term average (Talke et al., 2018; Peng 9.6.4.1      Past Changes                                                    et al., 2019; Baranes et al., 2020). Failing to consider the non-linear interactions between tide, surge and RSL may overestimate trends in The AR5 (Church et al., 2013b) concluded that changes in extreme still        ESWL (low confidence) (Arns et al., 2020). In some regions, changes water levels (ESWL), combining RSL, tide and surge as observed by tide        in ESWL depend more on changes in surge or tide than on sea gauges (Box 9.1) are very likely to be caused by observed increases in        level trends.
RSL, but noted low confidence in region-specific results owing to the limited number of studies considering localized contributions from            Ongoing development of the Global Extreme Sea Level Analysis storm surge, tide or wave effects. Influences from dominant modes            (GESLA) tide gauge database (Woodworth et al., 2016) along with of climate variability, particularly ENSO and NAO (Annex IV), were            data archaeology (Talke and Jay, 2013) extends availability of tide also noted. Climate modes affect sea level extremes in many regions,          gauge records back to the mid 19th century (or earlier). Dynamical as a result of both sea level anomalies (Sections 9.2.4.2 and 9.6.1.3)        datasets used to assess trends in ESL at global or regional scales -
and changes in storminess (Section 11.7). The SROCC (Oppenheimer              for example, tide and surge contributions from the Global Tide and et al., 2019) concluded with high confidence that inclusion of local          Surge Reanalysis (GTSR; Muis et al., 2016, 2020), or wave setup/
processes (wave effects, storm surges, tides plus other regional              swash contributions from available wave hindcasts/reanalyses morphology changes due to erosion, sedimentation and compaction)              (Melet et al., 2018) - have model biases introduced with resolution is essential for estimation of changes in ESL events.                        and parametrization limitations, incomplete atmospheric data, and currently span only a few decades, so they are not yet long or As in AR5 and SROCC, tide gauge observations show that RSL rise              accurate enough to assess long-term trends in ESLs. Therefore, there (Section 9.6.1.3) is the primary driver of changes in ESWL at most            is medium confidence in observed trends in ESWL, but only low locations and, across tide gauges, has led to a median 165% increase          confidence in modelled ESL trends.
in high-tide flooding over 1995-2014 relative to those over 1960-1980 (high confidence) (Figure 9.31). Some locations exhibit            The AR5 indicated that the amplitude and phase of major tidal substantial differences between long-term RSL trends and ESWL                constituents have exhibited long-term change, but that their effects (high confidence), particularly given decadal to multi-decadal                on ESL were not well understood. The SROCC (Bindoff et al., 2019) variations of other ESWL contributors (Rashid and Wahl, 2020). Since          reported changes in tides (amplification and dampening) at some SROCC, RSL rise has been shown to be the dominant contributor                locations to be of comparable importance to changes in mean sea to ESWL rise at most gauge sites along the Chinese coast, but, at            level for explaining changes in high water levels, with the sign of some locations, the surge contribution dominates (Feng et al.,                change being dependent on stability of shoreline position. RSL rise 2019). Trends in the difference between ESWL and mean RSL rise                causes water depth-based alterations to the resonant characteristics 1309
 
Chapter 9                                                                                          Ocean, Cryosphere and Sea Level Change of the basin, changes the bottom friction and increases the wave          deployments of in situ measurements in the very dynamic surf speed (Pickering et al., 2012) and remains the primary hypothesis        zone means that long-term records of ETWL or ECWL are limited to for observed tidal changes. Other contributing processes include          a few sites; tidal gauges are typically located in sheltered locations strong localized anthropogenic drivers (e.g., port development,          (e.g., harbours) where wave contributions are absent (Lambert dredging, flood defences, land reclamation), changes in stratification    et al., 2020). Consequently, trends in wave contributions to ESL are associated with ocean warming (Section 9.2.1.3), and changes in          typically derived from trends in wave conditions observed offshore.
seabed roughness associated with ecological change (e.g., Haigh          On the basis of satellite altimeter observations, SROCC reported et al., 2019). Tide gauge data show that, although principal tidal        increasing extreme wave heights in the Southern and North Atlantic components have varied in amplitude on the order of 2% to                oceans of around 1.0 and 0.8 cm yr -1, respectively, over the period 10% per century (Jay, 2009; Ray, 2009), identifying direct causality      1985-2018 (medium confidence). The SROCC (Collins et al., 2019) remains challenging (Haigh et al., 2019). Combined, observations          also identified sea ice loss in the Arctic as leading to increased wave and models indicate RSL rise and direct anthropogenic factors are        heights over the period 1992-2014 (medium confidence). Since the primary drivers of observed tidal changes at tide gauge stations      SROCC, the satellite wave record has been shown to be sensitive (medium confidence).                                                      to alternate processing techniques, leading to important differences 9                                                                          in reported trends (Timmermans et al., 2020). The most common The SROCC (Oppenheimer et al., 2019) reported variations in storm        observation platforms for surface waves over the past 30 years are surge not related to changes in RSL, and concluded with high              in situ buoys. However, evolving biases associated with changing confidence that consideration of localized storm surge processes was      instrument type, configuration and sampling methodology introduce essential to monitor trends in ESL. SL events driven by storm surge      artificial trends (e.g., Gemmrich et al., 2011; Timmermans et al.,
are a response to tropical and extratropical cyclones. While historical  2020). Accurate metadata is required to address these issues, and, trends in extra-tropical cyclones are less clear (Section 11.7.2.1),      while available locally, are only beginning to be globally coordinated there is mounting evidence for an increasing proportion of stronger      (Centurioni et al., 2019). Wave reanalysis and hindcast products tropical cyclones globally, with an associated poleward migration        have also been used to investigate total water level at global scale (Section 11.7.1.2). These changes are captured in the ESL record, for    (Melet et al., 2018; Reguero et al., 2019). Their applicability for example, via increasing intensity and poleward shift in the location of  trend analysis is limited by inhomogeneous data for assimilation typhoon-driven storm surges reported across 64 years (1950-2013)          (Stopa et al., 2019), but they inform relationships between seasonal, in the western North Pacific (Oey and Chou, 2016). Along the              interannual to inter-decadal variability of climate indices and wind-east coast of the USA, there has been an increase in frequency of        wave characteristics (A.G. Marshall et al., 2015, 2018; Kumar et al.,
ESL events due to tropical cyclone changes since 1923 that can            2016; Stopa et al., 2016). To summarize, satellite era trends in wave be statistically linked to changes in global average temperature          heights of order 0.5 cm yr -1 have been reported, most pronounced in (Grinsted et al., 2013), and the signal is projected to emerge around    the Southern Ocean. However, sensitivity of processing techniques, 2030 (Lee et al., 2017). At century and longer time scales, geological    inadequate spatial distribution of observations, and homogeneity proxies such as overwash deposits in coastal lagoons or sinkholes can    issues in available records limit confidence in reported trends be used to reconstruct past changes in storm activity (e.g., Brandon      (medium confidence).
et al., 2013; Lin et al., 2014) and put recent events into historical perspective (e.g., Brandon et al., 2015). However, there is low          Only a few studies have attempted to quantify the role of confidence in the current ability to quantitatively compare geological    anthropogenic climate change in ESL events (e.g., Mori et al., 2014; proxies with gauge data. Historical storm surge activity is being        Takayabu et al., 2015; Turki et al., 2019). Detection and attribution of increasingly assessed with use of hydrodynamic model simulations          the human influence on climatic changes in surges, and waves remains and data-driven global reconstructions to supplement tide gauge          a challenge (Ceres et al., 2017), with limited evidence to suggest observations to investigate historical changes at centennial to          in some instances - for example, poleward migration of tropical millennial time scales (e.g., Ji et al., 2020; Muis et al., 2020; Tadesse cyclones in the Western North Pacific (Section 11.7.1.2), changes et al., 2020). Large regional variations and limited observational data  in surges and waves can be attributed to anthropogenic climate lead to low confidence in observed trends in the surge contribution      change (low confidence). With RSL change being considered the to increasing ESL.                                                        primary driver of observed tidal changes, there is medium confidence that these changes can be attributed to human influence. The close Waves contribute to ESL via wave setup, infra-gravity waves and          relationship between local ESL and long-term RSL change, combined swash processes (Dodet et al., 2019), with Extreme Total Water            with the robust attribution of GMSL change (Section 9.6.1.4), implies Level (ETWL; Box 9.1) used to represent ESWL with addition of            that observed global changes in ESL can be attributed, at least in wave setup, and Extreme Coastal Water Level (ECWL; Box 9.1) also          part, to human-caused climate change (medium confidence), but including contributions from swash. The SROCC (Oppenheimer et al.,        reconciling regional variation in these changes is not yet possible 2019) reported the dependency of these processes on nearshore            (Section 9.6.1.4).
geomorphology and deep-water wave climate, and thus sensitivity to internal climate variability and climate change. Few long-term 1310
 
Ocean, Cryosphere and Sea Level Change                                                                                                                        Chapter 9 9
Figure 9.31 l Historical occurrences of minor extreme still water levels. Defined as the 99th percentile of daily observed water levels over 1995-2014. (a) Percent change in occurrences over 1995-2014 relative to those over 1960-1980. (b-g) Annual mean sea level (blue) and annual occurrences of extreme still water levels over the 1995-2014 99th percentile daily maximum (yellow) at six selected tide gauge locations. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1311
 
Chapter 9                                                                                          Ocean, Cryosphere and Sea Level Change 9.6.4.2      Future Changes                                              at 19-31% of the 634 stations by 2050, consistent with SROCC.
By 2100, the median frequency amplification factor is projected to There are two distinct methods used to project future ESL changes:        be 163 for SSP12.6, 325 for SSP24.5, and 532 for SSP58.5, with (i) The static, or mean sea level offset, approach employs historical    respectively 60%, 71%, and 82% of the stations experiencing distributions of tidal, surge and wave components and adjusts future      a currently 1% annual probability event at least yearly (medium ESL distributions for mean RSL rise; (ii) The dynamic approach employs    confidence) (Figure 9.32).
hydrodynamic and/or wave models forced with atmospheric fields derived from general circulation models (GCMs) to project changes        In the dynamic approach, the low resolution of the forcing fields in tidal, storm surge and wave distributions, which are then combined    arising from GCMs limits the ability to resolve historical and future with RSL projections to project future ESLs; and (iii) The dynamic        changes in tropical and extra-tropical storm frequency and intensity, approach is computationally expensive. Use of the dynamic approach        and resolution of local geography and morphology limit ability to on large spatial or global scales has only recently been successful      represent ECWL (Box 9.1). Not all relevant processes - such as river to project 21st-century changes in ETWL (Vousdoukas et al., 2017,        discharge - are included in the dynamic models, and ESL events are 2018) and ECWL (Melet et al., 2020). Kirezci et al. (2020) assume        typically a combination of multiple contributing processes, which 9 stationarity in global wave and storm surge simulations to assess        are often not independent (Jevrejeva et al., 2019). In both static projected 21st-century changes in episodic coastal ETWL-driven            and dynamical approaches, global assessment of the performance flooding under global sea level rise scenarios.                          of modelled storm surge and wave contributions to ESL is limited by poor coverage of observations (limited to tide gauges for The SROCC (Oppenheimer et al., 2019) presents projections of ESL          ESWL, Muis et al., 2020), and unavailable for the wave dependent derived using a static approach. Such projections often quantify          ETWL and ECWL estimates (Vitousek et al., 2017; Vousdoukas et al.,
changes in ESL event frequency, expressed as frequency amplification    2018; Kirezci et al., 2020; Lambert et al., 2020; Melet et al., 2020).
factors (Hunter, 2010, 2012). Like RSL projections, frequency            In studies to date, individual models are used to simulate different amplification factors increase under higher-emissions scenarios,          contributions to ESL, non-linear interactions are not well captured, and differences between scenarios increase over time. The SROCC          and uncertainties associated with downscaling methodology are concludes that even small to moderate changes in mean RSL can            poorly resolved, leading to low confidence in available ESL projections lead to hundred- to thousand-fold increases in the frequencies            that include these modelled wave and surge contributions.
with which certain thresholds are exceeded - for example, what is currently a 1-in-100-year ESL height (1% annual probability or        Assessment of dynamic ETWL changes for regions is presented in 0.01 expected annual events) will be expected once or even multiple      Chapter 12, following the methods of Vousdoukas et al. (2018) times per year in future at many locations (Figure 9.32). The SROCC      and Kirezci et al. (2020). Consistent with studies using the static showed that currently rare ESL events (e.g., with an average return      approach, Vousdoukas et al. (2018) finds that by 2050 the historical period of 100 years) will occur annually or more frequently at most      1% average annual probability ETWL will have increased to a 2-50%
available locations for RCP4.5 by the end of the century (high            average annual probability for most high latitude regions, and more confidence). Results from these assessments are sensitive to the type    often (up to multiple times a year, >100% annual probability) in of ESL probability distribution assumed (Buchanan et al., 2016; Wahl      the tropics, under both RCP4.5 and RCP8.5. For 2100, present-day et al., 2017), as well as the magnitude and uncertainty of projected      1% average annual probability extreme sea levels will be exceeded RSL change (Slangen et al., 2017; Wahl et al., 2017; Frederikse et al.,  multiple times each year almost everywhere. In summary, despite 2020a). Frequency amplification factors tend to be largest in tropical    waves and surges being non-negligible contributors to projected regions due in part to higher RSL rise projections, but primarily to      ETWL and ECWL changes (Vousdoukas et al., 2018; Melet et al.,
the relative rarity of high ESLs in areas with little historical exposure 2020), RSL change is expected to be the main driver in changes in to tropical or extratropical cyclones. Alternative representation of      future ESL return periods in most areas (medium confidence).
changes in ESL, such as presenting changes in exceedances per year (Sweet and Park, 2014), are subject to similar sensitivities, and lead    The SROCC (Bindoff et al., 2019) concluded that the majority of to medium confidence in projected changes of event frequency using        coastal regions will experience statistically significant changes these methods.                                                            in tidal amplitudes through the 21st century. Comprehensive high-resolution (of the order 10 km) numerical modelling studies Employing a similar static approach - fitting a Gumbel distribution      provide evidence for spatially coherent changes in tidal amplitudes in between Mean Higher High Water (average of higher high water              shelf seas as a result of RSL rise (Haigh et al., 2019, and references height of each tidal day) and a threshold following Buchanan              therein). There is high confidence that GMSL rise will be the primary et al. (2016) - this Report updates SROCC projections of ESL with        driver of global tidal amplitude increases and decreases over the next the RSL projections from Section 9.6.3.3 (see also Supplementary          100-200 years, changing the baseline tide that ESLs are imposed Material 9.SM.4). By 2050, the median increase in frequency              on. At local and regional scales, anthropogenic factors such as major amplification factor at 634 tide gauge stations is 19 for SSP12.6,      land reclamation efforts, as in the East China Sea (Song et al., 2013) 22 for SSP24.5, and 30 for SSP58.5 (Figure 9.32). This means that,      or differing national coastal management strategies (maintaining the by 2050, a historical (1995-2014) 1% annual probability ESL will          present coastline position or managed retreat) will locally modulate have increased to an 19-30% annual probability. The 1% historical        the influence of GMSL rise on tidal amplitude (medium confidence).
annual probability event is expected to become an annual event 1312
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                Chapter 9 The SROCC (Oppenheimer et al., 2019) concluded that the intensity                          different locations by effects of changes in tracks (Section 11.7.1; of severe tropical cyclones will increase in a warmer climate                              Garner et al., 2017). There is low confidence in projected changes in (Section 11.7.1), but low confidence remains in the future frequency                      ESL driven by changes in tropical cyclone climatology.
of tropical cyclones. Changes in tropical cyclone climatology will contribute to variations in frequency and magnitude of future ESL                          Changes in surface wave conditions occur in response to changes in surge events, although estimates of this contribution range widely                        frequency; intensity and position of forcing winds and storms (Morim (Lin et al., 2012; McInnes et al., 2014, 2016; Little et al., 2015; Garner                et al., 2018, 2019); reduction in sea ice and associated changes et al., 2017; Mori et al., 2019; Muis et al., 2020). In the Gulf of Mexico,                in fetch conditions (Thomson and Rogers, 2014; Casas-Prat and changes in ESL due to tropical cyclone activity may be as important                        Wang, 2020); and changes in coastal morphology associated with as SLR in enhancing future flood hazards (Marsooli et al., 2019).                          RSL rise (Wandres et al., 2017; Storlazzi et al., 2018). A few studies For the Korean Peninsula, a maximum change in 100-year return                              considering the contribution of a non-stationary wave climate on height associated with typhoon-induced storm surges of 10% under                          future changes in ESL infer a small but non-negligible contribution 4&deg;C warming is found (Yang et al., 2018). The effects of projected                        (Vousdoukas et al., 2018; Melet et al., 2020). The SROCC presented changes in tropical cyclone intensity may be enhanced or offset in                        qualitative assessments of projected changes in wave conditions.
9 SSP5-8.5 SSP2-4.5 SSP1-2.6 Figure 9.32 l Projected median frequency amplification factors for the 1% average annual probability extreme still water level in 2050 (a, c, e) and 2100 (b, d, f). Based on a peak-over-threshold (99.7%) method applied to the historical extreme still water levels of Global Extreme Sea Level Analysis version 2 (GESLA2) following Special Report on Ocean and Cryosphere in a Changing Climate (SROCC) and additionally fitting a Gumbel distribution between Mean Higher High Water (MHHW) and the threshold following Buchanan et al. (2016), using the regional sea level projections of Section 9.6.3.3 for (a, b) SSP58.5, (c, d) SSP24.5 and (e, f) SSP12.6. Further details on data sources and processing are available in the chapter data table (Table 9.SM.9).
1313
 
Chapter 9                                                                                          Ocean, Cryosphere and Sea Level Change Since SROCC, a quantitative assessment of a community ensemble            catchments are steep and cause high rainfall near the coast, such as of global wind-wave projections (Morim et al., 2019) found robust        in south-west UK (Svensson and Jones, 2004). The compound effect projected changes of around 5-10% (positive or negative, depending        of storm surge and rainfall contributes greater projected flood risk on region) in annual mean significant wave height, mean wave              than climate-induced amplification (Hsiao et al., 2021). However, at period, and/or mean wave directions along about 52% of the worlds        other locations, co-occurrence was unimportant because streamflow coastline that exceed internal climate variability under RCP8.5 by        timing did not coincide with the coastal peak storm surge (Hudson 2100. Continued retreat of sea ice cover in the Arctic will lead to      River, Orton et al., 2012; Rhine delta, Klerk et al., 2015). The SROCC more energetic wind-wave conditions (Casas-Prat and Wang, 2020).          (Oppenheimer et al., 2019) detailed the complexity of interactions Wave climate modelling methods introduce up to around 50% of              in deltaic environments. Direct increases in flooding driven by the ensemble variance in mean wave climate projections (Morim            increasing RSL and storm surge, rain, or correlations between these et al., 2019). GCMs do not typically resolve the higher-resolution        flood-drivers (e.g., Moftakhari et al., 2017; Orton et al., 2020) are tropical and extratropical storm features required to accurately          expected to be further accompanied by increases in flooding due to determine the contribution of extreme waves to ESLs and individual        subsidence (vertical land movement) and sedimentation (RSL-driven studies have sought to improve resolution to address these issues        blockage of river flows). The probability of concurrent surge, wave 9 (e.g., Timmermans et al., 2017). To date, projections of wave height      and precipitation events has been projected to increase by more extremes have been constrained to single wave model configurations        than 25% by 2100 compared to present, with high northern latitudes (e.g., Timmermans et al., 2017; Meucci et al., 2020). In summary,        displaying compound flooding becoming more than 2.5 times as there is medium confidence in projections of changes in mean wave        frequent, and weakening in the subtropics (Bevacqua et al., 2020).
climate but low confidence in the projected changes in extreme            However, the number of studies on compound events is still limited wave conditions due to limited evidence.                                  and so there is low confidence in understanding the extent by which compound events of surge with rain will change in response to RSL Correlations between changes in sea level-forced (mean sea level          rise and climate change.
and tidal) and atmospherically-forced drivers (ocean surface waves and surges) of ESLs have only been considered in a few studies, although high surge and high waves co-occur along a majority of          9.7        Final Remarks the worlds coastlines (Marcos et al., 2019). Along the east coast fo the USA, ocean dynamic sea level change and change in power            The process-based assessment of observed and projected change in dissipation index (a proxy for North Atlantic tropical cyclone activity)  the ocean, cryosphere and sea level undertaken here reveals advances are correlated across CMIP5 GCMs, resulting in an increase in ESLs        and gaps in reconstructions, observations, models and process relative to analyses assuming independence of these changes              understanding. Revisiting the updated assessments since AR5 and (Little et al., 2015). In the Irish Sea, dynamically coupled wave-tide    SROCC helps to gauge the robustness of understanding and quantitative modelling results in high water wave heights up to 20% higher than        assessments. The CMIP6 family of models builds on the experience in an uncoupled analysis (Lewis et al., 2019). In the German Bight,      of the CMIP5 models, and the projections of ISMIP6, LARMIP-2 and RSL rise relaxes the breaking criterion of nearshore waves (assuming      GlacierMIP strengthen understanding. Taken together with emulators no geomorphological response), allowing larger waves to propagate        of these simulations (Box 9.3) and transparent statistical approaches closer to shore, leading to increased wave runup (Arns et al., 2017).    (Section 9.6.3), this chapter provides projections that are consistent In south-western Australia, the influence of projected SLR was found      with the assessment of equilibrium climate sensitivity in this Report to exceed the influence of projected changes in forcing winds on wave    and that have improved estimates of uncertainty.
characteristics at the coast (Wandres et al., 2017). Thus, projections of ESL that do not consider correlations between and among sea level        The largest uncertainties in future sea level and cryosphere change are forced and atmospherically forced drivers can differ strongly from        related to the Greenland and Antarctic ice sheets (Sections 9.4.1.3, coupled projections (medium confidence).                                  9.4.1.4, 9.4.2.5 and 9.4.2.6). While the ISMIP6 and LARMIP-2 protocols provide simulations permitting uncertainty estimation The SROCC (Collins et al., 2019) highlighted compound events, or          and probabilistic inferences, remaining deep uncertainty relates to coincident occurrence of multiple hazards, as an example of deep          ice-sheet processes and the atmospheric and oceanic conditions uncertainty, and noted that failing to account for multiple factors      simulated by CMIP models in polar regions (Sections 9.4.2.3 and contributing to extreme events will lead to underestimation of the        9.4.2.4). ISMIP6 and LARMIP-2 have not been simulated beyond probabilities of occurrence (high confidence). Statistical studies        2100, which greatly reduces the amount and variety of state-of-the-have shown that high rain or streamflow often co-occurs with storm        art projections available to make ice-sheet and sea level projections surge as examples of compound surge-rain or surge-discharge            beyond 2150. After 2150, limited agreement causes us to consider all events (Sections 11.8.1 and 12.4.5.6; Wahl and Chambers, 2015;            projections as low confidence. Critically, the uncertainty in ice-sheet Moftakhari et al., 2017; Ward et al., 2018; Wu et al., 2018; Couasnon    projections is the leading uncertainty in projections of future global et al., 2020). Dynamical modelling studies show that co-occurrence        sea level for the second half of this century and beyond (Section 9.6.3).
of flood drivers raises ESLs at some locations in estuaries, such as the Rhine Delta (Zhong et al., 2013), the Netherlands (van den        Glacier inventory and projection uncertainty has been a significant Hurk et al., 2015), Taiwan, China (Chen and Liu, 2016), and the          source of past sea level budget uncertainty and remains a dominant Hudson River, USA (Orton et al., 2020), particularly when hydrologic      uncertainty until mid-century. Emissions scenario becomes the largest 1314
 
Ocean, Cryosphere and Sea Level Change                                                                                          Chapter 9 source of glacier change uncertainty by 2100, just as the relative      Acknowledgements importance of glacier loss is projected to decrease (Section 9.5.1).
We acknowledge the contribution of invited expert reviewers and the Ice Sheet Mass Balance Intercomparison Exercise (IMBIE) Team.
New high-resolution climate models show that sea surface Their valuable input and advice have significantly improved the temperature, overturning circulation, ocean heat content change chapter. We thank colleagues, institutions and, in particular, our and sea ice cover are considerably improved in most models when families for their support. Thanks to the Technical Support Unit and compared to the coarser resolution models. Change in the Southern especially Sophie Berger for her support.
Ocean and adjacent shelves (Section 9.2.3.2) is intimately linked to the future of the Antarctic Ice Sheet (Section 9.4.2.3), and projection of the Southern Ocean depends on oceanic and atmospheric drivers affecting heat (and carbon) uptake and sea ice. However, resolution remains a factor, as most CMIP6 models are far from resolutions that directly represent coastal and regional shallow-water processes, such as those beneath Antarctic ice shelves, in Greenland fjords and the eddying convection found by the Overturning in the Subpolar North                                                                            9 Atlantic Program.
Processes that change on long time scales - particularly Atlantic Meridional Overturning Circulation, ocean heat content, and ice sheets - require additional projections beyond the CMIP scenarios to explore longer-term commitment, post-forcing recovery measured in centuries rather than years or decades, and potential tipping points and thresholds. Only a few new studies focused on longer time scales, and none based on CMIP6 models.
1315
 
Chapter 9                                                                              Ocean, Cryosphere and Sea Level Change Frequently Asked Questions FAQ 9.1 l Can Continued Melting of the Greenland and Antarctic Ice Sheets Be Reversed? How Long Would It Take for Them to Grow Back?
Evidence from the distant past shows that some parts of the Earth system might take hundreds to thousands of years to fully adjust to changes in climate. This means that some of the consequences of human-induced climate change will continue for a very long time, even if atmospheric heat-trapping gas levels and global temperatures are stabilized or reduced in the future. This is especially true for the Greenland and Antarctic ice sheets, which grow much more slowly than they retreat. If the current melting of these ice sheets continues for long enough, it becomes effectively irreversible on human time scales, as does the sea level rise caused by that melting.
Humans are changing the climate and there are mechanisms that amplify the warming in the polar regions (Arctic and Antarctic). The Arctic is already warming faster than anywhere else (see FAQ 4.3). This is significant 9
because these colder high latitudes are home to our two remaining ice sheets: Antarctica and Greenland. Ice sheets are huge reservoirs of frozen freshwater, built up by tens of thousands of years of snowfall. If they were to completely melt, the water released would raise global sea level by about 65 m. Understanding how these ice sheets are affected by warming of nearby ocean and atmosphere is therefore critically important. The Greenland and Antarctic ice sheets are already slowly responding to recent changes in climate, but it takes a long time for these huge masses of ice to adjust to changes in global temperature. That means that the full effects of a warming climate may take hundreds or thousands of years to play out. An important question is whether these changes can eventually be reversed, once levels of greenhouse gases in the atmosphere are stabilized or reduced by humans and natural processes. Records from the past can help us answer this question.
For at least the last 800,000 years, the Earth has followed cycles of gradual cooling followed by rapid warming caused by natural processes. During cooling phases, more and more ocean water is gradually deposited as snowfall, causing ice sheets to grow and sea level to slowly decrease. During warming phases, the ice sheets melt more quickly, resulting in more rapid rises in sea level (FAQ 9.1, Figure 1). Ice sheets build up very slowly because growth relies on the steady accumulation of falling snow that eventually compacts into ice. As the climate cools, areas that can accumulate snow expand, reflecting back more sunlight that otherwise would keep the Earth warmer. This means that, once started, glacial climates develop rapidly. However, as the climate cools, the amount of moisture that the air can hold tends to decrease. As a result, even though glaciations begin quite quickly, it takes tens of thousands of years for ice sheets to grow to a point where they are in balance with the colder climate.
Ice sheets retreat more quickly than they grow because of processes that, once triggered, drive self-reinforcing ice loss. For ice sheets that are mostly resting on bedrock above sea level - like the Greenland Ice Sheet - the main self-reinforcing loop that affects them is the elevation-mass balance feedback (FAQ 9.1, Figure 1, right). In this situation, the altitude of the ice-sheet surface decreases as it melts, exposing the sheet to warmer air. The lowered surface then melts even more, lowering it faster still, until eventually the whole ice sheet disappears. In places where the ice sheet rests instead on bedrock that is below sea level, and which also deepens inland, including many parts of the Antarctic Ice Sheet, an important process called marine ice sheet instability is thought to drive rapid retreat (FAQ 9.1, Figure 1, left). This happens when the part of the ice sheet that is surrounded by sea water melts. That leads to additional thinning, which in turn accelerates the motion of the glaciers that feed into these areas. As the ice sheet flows more quickly into the ocean, more melting takes place, leading to more thinning and even faster flow that brings ever-more glacier ice into the ocean, ultimately driving rapid deglaciation of whole ice-sheet drainage basins.
These (and other) self-reinforcing processes explain why relatively small increases in temperature in the past led to very substantial sea level rise over centuries to millennia, compared to the many tens of thousands of years it takes to grow the ice sheets that lowered the sea level in the first place. These insights from the past imply that, if human-induced changes to the Greenland and Antarctic ice sheets continue for the rest of this century, it will take thousands of years to reverse that melting, even if global air temperatures decrease within this or the next century. In this sense, these changes are therefore irreversible, since the ice sheets would take much longer to regrow than the decades or centuries for which modern society is able to plan.
1316
 
Ocean, Cryosphere and Sea Level Change                                                                                                                                            Chapter 9 FAQ 9.1 (continued)
FAQ 9.1: Can melting of the ice sheets be reversed?
Once ice sheets are destabilised, it takes them tens of thousands of years to re-grow.
These changes strongly affect sea level.
10 0
Slow ice sheet growth Global mean sea level (m)
                                      -20 Rapid ice sheet retreat
                                      -40
                                      -60
                                      -80
                                      -100                                                                                                                                                      9
                                      -120
                                      -140 140,000          120,000          100,000        80,000          60,000            40,000              20,000              Today Years before present Melting driven by ocean temperature                                              Melting driven by air temperature Ice sheet Ocean Bedrock                                                          Bedrock                                Ice sheet When bedrock dips seaward or is flat, the retreat                            The ice sheet is very thick therefore its surface is stops when warming stops. When ice sheet retreats,                            very high and the air at high altitude is very cold less ice is released into ocean Ice sheet Ocean Bedrock                                                          Bedrock                                Ice sheet When bedrock dips landward the retreat is quick and                          As the ice sheet melts, its surface goes down self-sustained. When ice sheet retreats, more ice                            until it reaches a threshold, where the surrounding is released into ocean - ice sheet retreats further                          air is warmer and melts the ice even more quickly FAQ 9.1, Figure 1 l Ice sheets growth and decay. (Top) Changes in ice-sheet volume modulate sea level variations. The grey line depicts data from a range of physical environmental sea level recorders such as coral reefs while the blue line is a smoothed version of it. (Bottom left) Example of destabilization mechanism in Antarctica. (Bottom right) Example of destabilization mechanism in Greenland.
1317
 
Chapter 9                                                                              Ocean, Cryosphere and Sea Level Change Frequently Asked Questions FAQ 9.2 l How Much Will Sea Level Rise in the Next Few Decades?
As of 2018, global average sea level was about 15-25 cm higher than in 1900, and 7-15 cm higher than in 1971.
Sea level will continue to rise by an additional 10-25 cm by 2050. The major reasons for this ongoing rise in sea level are the thermal expansion of seawater as its temperature increases, and the melting of glaciers and ice sheets. Local sea level changes can be larger or smaller than the global average, with the smallest changes in formerly glaciated areas, and the largest changes in low-lying river delta regions.
Across the globe, sea level is rising, and the rate of increase has accelerated. Sea level increased by about 4 mm per year from 2006 to 2018, which was more than double the average rate over the 20th century. Rise during the early 1900s was due to natural factors, such as glaciers catching up to warming that occurred in the Northern Hemisphere during the 1800s. However, since at least 1970, human activities have been the dominant cause of 9
global average sea level rise, and they will continue to be for centuries into the future.
Sea level rises either through warming of ocean waters or the addition of water from melting ice and bodies of water on land. Expansion due to warming caused about 50% of the rise observed from 1971 to 2018. Melting glaciers contributed about 22% over the same period. Melting of the two large ice sheets in Greenland and Antarctica has contributed about 13% and 7%, respectively, during 1971 to 2018, but melting has accelerated in the recent decades, increasing their contribution to 22% and 14% since 2016. Another source is changes in land-water storage: reservoirs and aquifers on land have reduced, which contributed about an 8% increase in sea level.
By 2050, sea level is expected to rise an additional 10-25 cm whether or not greenhouse gas emissions are reduced (FAQ 9.2, Figure 1). Beyond 2050, the amount by which sea level will rise is more uncertain. The accumulated total emissions of greenhouse gases over the upcoming decades will play a big role beyond 2050, especially in determining where sea level rise and ice-sheet changes eventually level off.
Even if net zero emissions are reached, sea level rise will continue because the deep ocean will continue to warm and ice sheets will take time to catch up to the warming caused by past and present emissions: ocean and ice sheets are slow to respond to environmental changes (see FAQ 5.3). Some projections under low emissions show sea level rise continuing as net zero is approached at a rate comparable to today (3-8 mm per year by 2100 versus 3-4 mm per year in 2015), while others show substantial acceleration to more than five times the present rate by 2100, especially if emissions continue to be high and processes that accelerate retreat of the Antarctic Ice Sheet occur widely (FAQ 9.1).
Sea level rise will increase the frequency and severity of extreme sea level events at coasts (see FAQ 8.2), such as storm surges, wave inundation and tidal floods: risk can be increased by even small changes in global average sea level. Scientists project that, in some regions, extreme sea level events that were recently expected once in 100 years will occur annually at 20-25% of locations by 2050 regardless of emissions, but by 2100 emissions choice will matter: annually at 60% of locations for low emissions, and at 80% of locations under strong emissions.
In many places, local sea level change will be larger or smaller than the global average. From year to year and place to place, changes in ocean circulation and wind can lead to local sea level change. In regions where large ice sheets, such as the Fennoscandian in Eurasia and the Laurentide and Cordilleran in North America, covered the land during the last ice age, the land is still slowly rising up now that the extra weight of the ice sheets is gone. This local recovery is compensating for global sea level rise in these regions and can even lead to local decrease in sea level. In regions just beyond where the former ice sheets reached and the Earth bulged upwards, the land is now falling and, as a result, local sea level rise is faster than the global rate. In many regions within low-lying delta regions (such as New Orleans and the Ganges-Brahmaputra delta), the land is rapidly subsiding (sinking) because of human activities such as building dams or groundwater and fossil fuel extraction. Further, when an ice sheet melts, it has less gravitational pull on the ocean water nearby. This reduction in gravitational attraction causes sea level to fall close to the (now less-massive) ice sheet while causing sea level to rise farther away. Melt from a polar ice sheet therefore raises sea level most in the opposite hemisphere or in low latitudes -
amounting to tens of centimetres difference in rise between regions by 2100.
1318
 
Ocean, Cryosphere and Sea Level Change                                                                                                                              Chapter 9 FAQ 9.2 (continued)
FAQ 9.2: How much will sea level rise in the next few decades?
Emissions scenarios influence little sea level rise of the coming decades but has a huge effect on sea level at the end of the century.
In 2018                      In 2050                                      In 2100 Very high-emission scenario - SSP5-8.5 Contributions cm of sea level rise (since 1971)
Observations Antarctic ice sheet                                                                                                                                                    9 Greenland ice sheet                                                      Low-emission scenario - SSP1-2.6 Glaciers Ocean thermal expansion Water stored on land Sea level has risen            The scenario does                              The scenario faster in the recent years    not matter too much                            matters a lot Global sea level rise (cm) 100
                                                                                                                  -8.5 P5 SS 50 SSP1-2.6 0
1980            2000              2020              2040            2060              2080                2100 FAQ 9.2, Figure 1 l Observed and projected global mean sea level rise and the contributions from its major constituents.
1319
 
Chapter 9                                                                            Ocean, Cryosphere and Sea Level Change Frequently Asked Questions FAQ 9.3 l Will the Gulf Stream Shut Down?
The Gulf Stream is part of two circulation patterns in the North Atlantic: the Atlantic Meridional Overturning Circulation (AMOC) and the North Atlantic subtropical gyre. Based on models and theory, scientific studies indicate that, while the AMOC is expected to slow in a warming climate, the Gulf Stream will not change much and would not shut down totally, even if the AMOC did. Most climate models project that the AMOC slows in the later 21st century under most emissions scenarios, with some models showing it slowing even sooner. The Gulf Stream affects the weather and sea level, so if it slows, North America will see higher sea levels and Europes weather and rate of relative warming will be affected.
The Gulf Stream is the biggest current in the North Atlantic Ocean. It transports about 30 billion kilograms of water per second northward past points on the east coast of North America. It is a warm current, with 9
temperatures 5&deg;C to 15&deg;C warmer than surrounding waters, so it carries warmer water (thermal energy) from its southern origins and releases warmth to the atmosphere and surrounding water.
The Gulf Stream is part of two major circulation patterns, the Atlantic Meridional Overturning Circulation (AMOC) and the North Atlantic Subtropical Gyre (FAQ 9.3, Figure 1). The rotation of the Earth causes the big currents in both circulations to stay on the western side of their basin, which in the Atlantic means the circulations combine to form the Gulf Stream. Other large currents contribute to gyres, such as the Kuroshio in the North Pacific and the East Australian Current in the South Pacific, but the Gulf Stream is special in its dual role. There is no comparable deep overturning circulation in the North Pacific to the AMOC, so the Kuroshio plays only one role as part of a gyre.
The gyres circulate surface waters and result primarily from winds driving the circulation. These winds are not expected to change much and so neither will the gyres, which means the gyre portion of the Gulf Stream and the Kuroshio will continue to transport thermal energy poleward from the equator much as they do now. The gyre contribution to the Gulf Stream is 2 to 10 times larger than the AMOC contribution.
The Gulf Streams role in the AMOC is supplying surface source water that cools, becomes denser and sinks to form cold, deep waters that travel back equatorward, spilling over features on the ocean floor and mixing with other deep Atlantic waters to form a southward current at a depth of about 1500 metres beneath the Gulf Stream. This overturning flow is the AMOC, with the Gulf Stream in the upper kilometre flowing northward, and the colder deep water flowing southward.
The AMOC is expected to slow over the coming centuries. One reason why is freshening of the ocean waters:
by meltwater from Greenland, changing Arctic sea ice, and increased precipitation over warmer northern seas.
An array of moorings across the Atlantic has been monitoring the AMOC since 2004, with recently expanded capabilities. The monitoring of the AMOC has not been long enough for a trend to emerge from variability and detect long-term changes that may be underway (see FAQ 1.2). Other indirect signs may indicate slowing overturning - for example, slower warming where the Gulf Streams surface waters sink. Climate models show that this cold spot of slower-than-average warming occurs as the AMOC weakens, and they project that this will continue. Paleoclimate evidence indicates that the AMOC changed significantly in the past, especially during transitions from colder climates to warmer ones, but that it has been stable for 8000 years.
What happens if the AMOC slows in a warming world? The atmosphere adjusts somewhat by carrying more heat, compensating partly for the decreases in heat carried by AMOC. But the cold spot makes parts of Europe warm more slowly. Models indicate that weather patterns in Greenland and around the Atlantic will be affected, with reduced precipitation in the mid-latitudes, changing strong precipitation patterns in the tropics and Europe, and stronger storms in the North Atlantic storm track. The slowing of this current combined with the rotation of the Earth means that sea level along North America rises as the AMOC contribution to the Gulf Stream slows.
The North Atlantic is not the only site of sensitive meridional overturning. Around Antarctica, the worlds densest seawater is formed by freezing into sea ice, leaving behind salty, cold water that sinks to the bottom and spreads northward. Recent studies show that melting of the Antarctic Ice Sheet and changing winds over the Southern Ocean can affect this southern meridional overturning, affecting regional weather.
1320
 
Ocean, Cryosphere and Sea Level Change                                                                                                        Chapter 9 FAQ 9.3 (continued)
FAQ 9.3: Will the Gulf Stream shut down?
The Gulf Stream, a warm current, is expected to weaken but not cease. This slowdown will affect regional weather and sea level.
Today                                                                    In a warmer world The Gulf Stream is part of both the horizontal, subtropical              Climate change weakens the AMOC, which slows gyre and the vertical, Atlantic Meridional Overturning                  the Gulf Stream down Circulation (AMOC)
Weakened AMOC                                                      3              AMOC Close to the poles,                                      1 water cools, becomes heavier                                      Water has become fresher and and sinks to the bottom                                    lighter and therefore sinks less 9
1 Warm surface current                                                        2 Much less 2
driven by winds and                                                        heat and water replenishing sinking                          Water and heat transferred                are transferred from the tropics to GULF                                  northern latitudes GULF STREAM                                                                  STREAM GYRE                                                                GYRE 3
The Gulf Stream weakens but the 4                                                  portion pushed by winds remains The cold deep water is exported southward FAQ 9.3, Figure 1 l Horizontal (gyre) and vertical (Atlantic Meridional Overturning Circulation, AMOC) circulations in the Atlantic today (left) and in a warmer world (right). The Gulf Stream is a warm current composed of both circulations.
1321
 
Chapter 9                                                                                                      Ocean, Cryosphere and Sea Level Change References Aas, K.S., K. Gisns, S. Westermann, and T.K. Berntsen, 2017: A Tiling Approach  An, L., E. Rignot, R. Millan, K. Tinto, and J. Willis, 2019a: Bathymetry of to Represent Subgrid Snow Variability in Coupled Land Surface-Atmosphere        Northwest Greenland Using Ocean Melting Greenland (OMG)
Models. Journal of Hydrometeorology, 18(1), 49-63, doi:10.1175/                High-Resolution Airborne Gravity and Other Data. Remote Sensing, 11(2),
jhm-d-16-0026.1.                                                                131, doi:10.3390/rs11020131.
Abich, K. et al., 2019: In-Orbit Performance of the GRACE Follow-on Laser        An, L. et al., 2019b: Bathymetry of Southeast Greenland From Oceans Ranging Interferometer. Physical Review Letters, 123(3), 031101,                Melting Greenland (OMG) Data. Geophysical Research Letters, 46(20),
doi:10.1103/physrevlett.123.031101.                                            11197-11205, doi:10.1029/2019gl083953.
Abrahamsen, E.P. et al., 2019: Stabilization of dense Antarctic water supply to  An, L. et al., 2021: Ocean melting of the Zachariae Isstr&#xf8;m and the Atlantic Ocean overturning circulation. Nature Climate Change, 9(10),      Nioghalvfjerdsfjorden glaciers, northeast Greenland. Proceedings of the 742-746, doi:10.1038/s41558-019-0561-2.                                        National Academy of Sciences, 118(2), e2015483118, doi:10.1073/pnas.
Abram, N. et al., 2019: Framing and Context of the Report. In: IPCC Special        2015483118.
Report on the Ocean and Cryosphere in a Changing Climate [Prtner, H.-O.,    Anderson, B. and A. Mackintosh, 2012: Controls on mass balance sensitivity D.C. Roberts, V. Masson-Delmotte, P. Zhai, M. Tignor, E. Poloczanska,          of maritime glaciers in the Southern Alps, New Zealand: The role of debris 9    K. Mintenbeck, A. Alegr&#xed;a, M. Nicolai, A. Okem, J. Petzold, B. Rama, and        cover. Journal of Geophysical Research: Earth Surface, 117(F1), F01003, N.M. Weyer (eds.)]. In Press, pp. 73-129, www.ipcc.ch/srocc/chapter/            doi:10.1029/2011jf002064.
chapter-1-framing-and-context-of-the-report.                                  Anderson, J.B., S.S. Shipp, A.L. Lowe, J.S. Wellner, and A.B. Mosola, 2002:
Aalgeirsd&#xf3;ttir, G. et al., 2020: Glacier changes in Iceland from ~1890 to 2019. The Antarctic Ice Sheet during the Last Glacial Maximum and its subsequent Frontiers in Earth Science, 8, 520, doi:10.3389/feart.2020.523646.              retreat history: A review. Quaternary Science Reviews, 21(1-3), 49-70, Adler, R.F. et al., 2003: The Version-2 Global Precipitation Climatology          doi:10.1016/s0277-3791(01)00083-x.
Project (GPCP) Monthly Precipitation Analysis (1979-Present). Journal        Andreassen, L.M., H. Elveh&#xf8;y, B. Kj&#xf8;llmoen, and J.M.C. Belart, 2020: Glacier of Hydrometeorology, 4(6), 1147-1167, doi:10.1175/1525-7541(2003)              change in Norway since the 1960s - an overview of mass balance, area, 004<1147:tvgpcp>2.0.co;2.                                                      length and surface elevation changes. Journal of Glaciology, 66(256),
Adusumilli, S., H.A. Fricker, B. Medley, L. Padman, and M.R. Siegfried, 2020:      313-328, doi:10.1017/jog.2020.10.
Interannual variations in meltwater input to the Southern Ocean from          Andresen, C.S. et al., 2017: Exceptional 20th century glaciological regime Antarctic ice shelves. Nature Geoscience, 13, 616-620, doi:10.1038/            of a major SE Greenland outlet glacier. Scientific Reports, 7(1), 13626, s41561-020-0616-z.                                                              doi:10.1038/s41598-017-13246-x.
Aguirre, C., M. Rojas, R.D. Garreaud, and D.A. Rahn, 2019: Role of synoptic      Anttila, K., T. Manninen, E. Jskelinen, A. Riihel, and P. Lahtinen, 2018:
activity on projected changes in upwelling-favourable winds at the oceans      The role of climate and land use in the changes in surface albedo prior to eastern boundaries. npj Climate and Atmospheric Science, 2(1), 44,              snow melt and the timing of melt season of seasonal snow in northern doi:10.1038/s41612-019-0101-9.                                                  land areas of 40&deg;N-80&deg;N during 1982-2015. Remote Sensing, 10(10),
Aguirre, F. et al., 2018: Snow Cover Change as a Climate Indicator in Brunswick    doi:10.3390/rs10101619.
Peninsula, Patagonia. Frontiers in Earth Science, 6, 130, doi:10.3389/        Aoki, S. et al., 2020: Freshening of Antarctic Bottom Water off Cape feart.2018.00130.                                                              Darnley, East Antarctica. Journal of Geophysical Research: Oceans, 125, Aksenov, Y. et al., 2016: Arctic pathways of Pacific Water: Arctic Ocean Model    e2020JC016374, doi:10.1029/2020jc016374.
Intercomparison experiments. Journal of Geophysical Research: Oceans,        Applegate, P.J. and K. Keller, 2015: How effective is albedo modification 121(1), 27-59, doi:10.1002/2015jc011299.                                        (solar radiation management geoengineering) in preventing sea-level rise Alberello, A. et al., 2020: Drift of Pancake Ice Floes in the Winter Antarctic    from the Greenland Ice Sheet? Environmental Research Letters, 10(8),
Marginal Ice Zone During Polar Cyclones. Journal of Geophysical Research:      84018, doi:10.1088/1748-9326/10/8/084018.
Oceans, 125(3), e2019JC015418, doi:10.1029/2019jc015418.                      Argus, D.F. and W.R. Peltier, 2010: Constraining models of postglacial rebound Albrecht, T., R. Winkelmann, and A. Levermann, 2020: Glacial-cycle simulations    using space geodesy: A detailed assessment of model ICE-5G (VM2) and its of the Antarctic Ice Sheet with the Parallel Ice Sheet Model (PISM) - Part 1:  relatives. Geophysical Journal International, 181(2), 697-723, doi:10.1111/
Boundary conditions and climatic forcing. Cryosphere, 14(2), 599-632,          j.1365-246x.2010.04562.x.
doi:10.5194/tc-14-599-2020.                                                  Argus, D.F., W.R. Peltier, R. Drummond, and A.W. Moore, 2014: The Antarctica Alekseeva, T. et al., 2019: Comparison of Arctic Sea Ice Concentrations from      component of postglacial rebound model ICE-6G_C (VM5a) based on GPS the NASA Team, ASI, and VASIA2 Algorithms with Summer and Winter Ship          positioning, exposure age dating of ice thicknesses, and relative sea level Data. Remote Sensing, 11(21), 2481, doi:10.3390/rs11212481.                    histories. Geophysical Journal International, 198(1), 537-563, doi:10.1093/
Alexander, P.M. et al., 2019: Simulated Greenland Surface Mass Balance in the      gji/ggu140.
GISS ModelE2 GCM: Role of the Ice Sheet Surface. Journal of Geophysical      Armitage, T.W.K., R. Kwok, A.F. Thompson, and G. Cunningham, 2018: Dynamic Research: Earth Surface, 124(3), 750-765, doi:10.1029/2018jf004772.            Topography and Sea Level Anomalies of the Southern Ocean: Variability Allen, S.E. and X. Durrieu de Madron, 2009: A review of the role of submarine      and Teleconnections. Journal of Geophysical Research: Oceans, 123(1),
canyons in deep-ocean exchange with the shelf. Ocean Science, 5(4),            613-630, doi:10.1002/2017jc013534.
607-620, doi:10.5194/os-5-607-2009.                                          Armitage, T.W.K. et al., 2017: Arctic Ocean surface geostrophic circulation Alley, K.E., T.A. Scambos, M.R. Siegfried, and H.A. Fricker, 2016: Impacts of      2003-2014. The Cryosphere, 11(4), 1767-1780, doi:10.5194/tc-11-1767-warm water on Antarctic ice shelf stability through basal channel formation. 2017.
Nature Geoscience, 9(4), 290-293, doi:10.1038/ngeo2675.                      Armour, K.C., I. Eisenman, E. Blanchard-Wrigglesworth, K.E. McCusker, Alvarez-Solas, J., R. Banderas, A. Robinson, and M. Montoya, 2019:                and C.M. Bitz, 2011: The reversibility of sea ice loss in a state-of-the-art Ocean-driven millennial-scale variability of the Eurasian ice sheet during      climate model. Geophysical Research Letters, 38(16), L16705, the last glacial period simulated with a hybrid ice-sheet-shelf model.          doi:10.1029/2011gl048739.
Climate of the Past, 15(3), 957-979, doi:10.5194/cp-15-957-2019.
1322
 
Ocean, Cryosphere and Sea Level Change                                                                                                                      Chapter 9 Armour, K.C., J. Marshall, J.R. Scott, A. Donohoe, and E.R. Newsom, 2016:          Bamber, J.L. and G.J. Dawson, 2020: Complex evolving patterns of mass loss Southern Ocean warming delayed by circumpolar upwelling and                        from Antarcticas largest glacier. Nature Geoscience, 13(2), 127-131, equatorward transport. Nature Geoscience, 9, 549, doi:10.1038/ngeo2731.            doi:10.1038/s41561-019-0527-z.
Arndt, S., M. Hoppmann, H. Schmith&#xfc;sen, A.D. Fraser, and M. Nicolaus, 2020:        Bamber, J.L., R.E.M. Riva, B.L.A. Vermeersen, and A.M. LeBrocq, 2009:
Seasonal and interannual variability of landfast sea ice in Atka Bay,              Reassessment of the Potential Sea-Level Rise from a Collapse of the Weddell Sea, Antarctica. The Cryosphere, 14(9), 2775-2793, doi:10.5194/            West Antarctic Ice Sheet. Science, 324(5929), 901-903, doi:10.1126/
tc-14-2775-2020.                                                                    science.1169335.
Arns, A., T. Wahl, I.D. Haigh, and J. Jensen, 2015: Determining return water        Bamber, J.L., R.M. Westaway, B. Marzeion, and B. Wouters, 2018a: A new levels at ungauged coastal sites: A case study for northern Germany.                synthesis of annual land ice mass trends 1992 to 2016. PANGAEA. Retrieved Ocean Dynamics, 65(4), 539-554, doi:10.1007/s10236-015-0814-1.                      from: https://doi.org/10.1594/pangaea.890030.
Arns, A. et al., 2017: Sea-level rise induced amplification of coastal protection  Bamber, J.L., R.M. Westaway, B. Marzeion, and B. Wouters, 2018b: The land ice design heights. Scientific reports, 7(1), 40171, doi:10.1038/srep40171.            contribution to sea level during the satellite era. Environmental Research Arns, A. et al., 2020: Non-linear interaction modulates global extreme sea            Letters, 13, 063008, doi:10.1088/1748-9326/aac2f0.
levels, coastal flood exposure, and impacts. Nature Communications, 11(1),        Bamber, J.L., M. Oppenheimer, R.E. Kopp, W.P. Aspinall, and R.M. Cooke, 2019:
1-9, doi:10.1038/s41467-020-15752-5.                                                Ice sheet contributions to future sea-level rise from structured expert Asay-Davis, X.S., N.C. Jourdain, and Y. Nakayama, 2017: Developments                  judgment. Proceedings of the National Academy of Sciences, 116(23),
in Simulating and Parameterizing Interactions Between the Southern                  11195-11200, doi:10.1073/pnas.1817205116.                                          9 Ocean and the Antarctic Ice Sheet. Current Climate Change Reports, 3(4),          Banas, N.S., P. MacCready, and B.M. Hickey, 2009: The Columbia River plume 316-329, doi:10.1007/s40641-017-0071-0.                                            as cross-shelf exporter and along-coast barrier. Continental Shelf Research, Asay-Davis, X.S. et al., 2016: Experimental design for three interrelated marine      29(1), 292-301, doi:10.1016/j.csr.2008.03.011.
ice sheet and ocean model intercomparison projects: MISMIP v. 3 (MISMIP +),      Baranes, H.E. et al., 2020: Tidally Driven Interannual Variation in Extreme Sea ISOMIP v. 2 (ISOMIP +) and MISOMIP v. 1 (MISOMIP1). Geoscientific Model            Level Frequencies in the Gulf of Maine. Journal of Geophysical Research:
Development, 9(7), 2471-2497, doi:10.5194/gmd-9-2471-2016.                          Oceans, 125(10), e2020JC016291, doi:10.1029/2020jc016291.
Aschwanden, A., M.A. Fahnestock, and M. Truffer, 2016: Complex                      Barletta, V.R. et al., 2018: Observed rapid bedrock uplift in Amundsen Sea Greenland outlet glacier flow captured. Nature Communications, 7, 10524,            Embayment promotes ice-sheet stability. Science, 360(6395), 1335-1339, doi:10.1038/ncomms10524.                                                            doi:10.1126/science.aao1447.
Aschwanden, A. et al., 2019: Contribution of the Greenland Ice Sheet to            Bart, P.J., M. DeCesare, B.E. Rosenheim, W. Majewski, and A. McGlannan, 2018:
sea level over the next millennium. Science Advances, 5(6), eaav9396,              A centuries-long delay between a paleo-ice-shelf collapse and grounding-doi:10.1126/sciadv.aav9396.                                                        line retreat in the Whales Deep Basin, eastern Ross Sea, Antarctica.
Atwood, A.R., E. Wu, D.M.W. Frierson, D.S. Battisti, and J.P. Sachs, 2016:            Scientific Reports, 12392, doi:10.1038/s41598-018-29911-8.
Quantifying Climate Forcings and Feedbacks over the Last Millennium in the        Barthel, A. et al., 2020: CMIP5 model selection for ISMIP6 ice sheet model CMIP5-PMIP3 Models. Journal of Climate, 29(3), 1161-1178, doi:10.1175/              forcing: Greenland and Antarctica. The Cryosphere, 14, 855-879, jcli-d-15-0063.1.                                                                  doi:10.5194/tc-14-855-2020.
Auger, M., R. Morrow, E. Kestenare, J.-B. Sall&#xe9;e, and R. Cowley, 2021: Southern    Bassis, J.N. and L. Ultee, 2019:A Thin Film Viscoplastic Theory for Calving Glaciers:
Ocean in-situ temperature trends over 25 years emerge from interannual              Toward a Bound on the Calving Rate of Glaciers. Journal of Geophysical variability. Nature Communications, 12(1), 514, doi:10.1038/s41467-020-            Research: Earth Surface, 124(8), 2036-2055, doi:10.1029/2019jf005160.
20781-1.                                                                          Batbaatar, J.,A. Gillespie, D. Fink,A. Matmon, and T. Fujioka, 2018:Asynchronous Auriac, A. et al., 2016: Glacial isostatic adjustment associated with the Barents    glaciations in arid continental climate. Quaternary Science Reviews, 182, Sea ice sheet: A modelling inter-comparison. Quaternary Science Reviews,            1-19, doi:10.1016/j.quascirev.2017.12.001.
147, 122-135, doi:10.1016/j.quascirev.2016.02.011.                                Batchelor, C.J. et al., 2019: Distinct Permafrost Conditions Across the Last Two Axford, Y. et al., 2019: Holocene temperature history of northwest                    Glacial Periods in Midlatitude North America. Geophysical Research Letters, Greenland - With new ice cap constraints and chironomid assemblages                46, 13318-13326, doi:10.1029/2019gl083951.
from Deltas&#xf8;. Quaternary Science Reviews, 215, 160-172, doi:10.1016/j.            Bates, S.C. et al., 2012: Mean Biases, Variability, and Trends in Air-Sea Fluxes quascirev.2019.05.011.                                                              and Sea Surface Temperature in the CCSM4. Journal of Climate, 25(22),
Bachem, P.E., B. Risebrobakken, S. De Schepper, and E.L. Mcclymont, 2017:            7781-7801, doi:10.1175/jcli-d-11-00442.1.
Highly variable Pliocene sea surface conditions in the Norwegian Sea.            Bathiany, S., J. Hidding, and M. Scheffer, 2020: Edge Detection Reveals Abrupt Climate of the Past, 13(9), 1153-1168, doi:10.5194/cp-13-1153-2017.                and Extreme Climate Events. Journal of Climate, 33(15), 6399-6421, Baggenstos, D. et al., 2019: Earths radiative imbalance from the Last Glacial        doi:10.1175/jcli-d-19-0449.1.
Maximum to the present. Proceedings of the National Academy of Sciences,          Bathiany, S., D. Notz, T. Mauritsen, G. Raedel, and V. Brovkin, 2016: On the 116(30), 14881-14886, doi:10.1073/pnas.1905447116.                                  Potential for Abrupt Arctic Winter Sea Ice Loss. Journal of Climate, 29(7),
Bakker, A.M.R., T.E. Wong, K.L. Ruckert, and K. Keller, 2017: Sea-level projections  2703-2719, doi:10.1175/jcli-d-15-0466.1.
representing the deeply uncertain contribution of the West Antarctic ice          Baum, S.K. and T.J. Crowley, 2003: The snow/ice instability as a mechanism for sheet. Scientific Reports, 7(1), 3880, doi:10.1038/s41598-017-04134-5.              rapid climate change: A Neoproterozoic Snowball Earth model example.
Bakun, A., 1990: Global climate change and intensification of coastal ocean          Geophysical Research Letters, 30(20), 2030, doi:10.1029/2003gl017333.
upwelling. Science, 247(4939), 198-201, www.jstor.org/stable/2873492.            Bauska, T.K. et al., 2016: Carbon isotopes characterize rapid changes in Bakun, A., D.B. Field, A. Redondo-Rodriguez, and S.J. Weeks, 2010: Greenhouse        atmospheric carbon dioxide during the last deglaciation. Proceedings of gas, upwelling-favorable winds, and the future of coastal ocean upwelling          the National Academy of Sciences, 113(13), 3465-3470, doi:10.1073/
ecosystems. Global Change Biology, 16(4), 1213-1228, doi:10.1111/                  pnas.1513868113.
j.1365-2486.2009.02094.x.                                                        Beadling, R.L., J.L. Russell, R.J. Stouffer, P.J. Goodman, and M. Mazloff, 2019:
Bamber, J.L. and W.P. Aspinall, 2013: An expert judgement assessment of future        Assessing the quality of Southern Ocean circulation in CMIP5 AOGCM sea level rise from the ice sheets. Nature Climate Change, 3, 424-427,              and Earth System Model simulations. Journal of Climate, 32, 5915-5940, doi:10.1038/nclimate1778.                                                          doi:10.1175/jcli-d-19-0263.1.
1323
 
Chapter 9                                                                                                        Ocean, Cryosphere and Sea Level Change Beadling, R.L. et al., 2020: Representation of Southern Ocean properties across  Berben, S.M.P., K. Husum, A. Navarro-Rodriguez, S.T. Belt, and S. Aagaard-Coupled Model Intercomparison Project generations: CMIP3 to CMIP6.                S&#xf8;rensen, 2017: Semi-quantitative reconstruction of early to late Holocene Journal of Climate, 33, 6555-6581, doi:10.1175/jcli-d-19-0970.1.                  spring and summer sea ice conditions in the northern Barents Sea.
Beal, L.M. and S. Elipot, 2016: Broadening not strengthening of the Agulhas          Journal of Quaternary Science, 32(5), 587-603, doi:10.1002/jqs.2953.
Current since the early 1990s. Nature, 540(7634), 570-573, doi:10.1038/        Bereiter, B., S. Shackleton, D. Baggenstos, K. Kawamura, and J. Severinghaus, nature19853.                                                                      2018: Mean global ocean temperatures during the last glacial transition.
Beckmann, J. et al., 2019: Modeling the response of Greenland outlet glaciers        Nature, 553(7686), 39-44, doi:10.1038/nature25152.
to global warming using a coupled flowline-plume model. The Cryosphere,        Berends, C.J., B. de Boer, A.M. Dolan, D.J. Hill, and R.S.W. van de Wal, 2019:
13, 2281-2301, doi:10.5194/tc-2018-89.                                            Modelling ice sheet evolution and atmospheric CO2 during the Late Pliocene.
Beedle, M.J., B. Menounos, B.H. Luckman, and R. Wheate, 2009: Annual push            Climate of the Past, 15, 1603-1619, doi:10.5194/cp-2019-34.
moraines as climate proxy. Geophysical Research Letters, 36(20), L20501,        Berloff, P. et al., 2007: The Turbulent Oscillator: A Mechanism of Low-doi:10.1029/2009gl039533.                                                          Frequency Variability of the Wind-Driven Ocean Gyres. Journal of Physical Belcher, S.E. et al., 2012: A global perspective on Langmuir turbulence in          Oceanography, 37(9), 2363-2386, doi:10.1175/jpo3118.1.
the ocean surface boundary layer. Geophysical Research Letters, 39(18),        Bertram, R.A. et al., 2018: Pliocene deglacial event timelines and the L18605, doi:10.1029/2012gl052932.                                                  biogeochemical response offshore Wilkes Subglacial Basin, East Antarctica.
Bell, R.E., A.F. Banwell, L.D. Trusel, and J. Kingslake, 2018: Antarctic surface    Earth and Planetary Science Letters, 494, 109-116, doi:10.1016/j.
9  hydrology and impacts on ice-sheet mass balance. Nature Climate Change,            epsl.2018.04.054.
8(12), 1044-1052, doi:10.1038/s41558-018-0326-3.                                Bettencourt, J.H. et al., 2015: Boundaries of the Peruvian oxygen minimum Bellomo, K., L.N. Murphy, M.A. Cane, A.C. Clement, and L.M. Polvani, 2018:          zone shaped by coherent mesoscale dynamics. Nature Geoscience, 8(12),
Historical forcings as main drivers of the Atlantic multidecadal variability      937-940, doi:10.1038/ngeo2570.
in the CESM large ensemble. Climate Dynamics, 50(9-10), 3687-3698,              Bevacqua, E. et al., 2020: More meteorological events that drive compound doi:10.1007/s00382-017-3834-3.                                                    coastal flooding are projected under climate change. Communications Belmadani, A., V. Echevin, F. Codron, K. Takahashi, and C. Junquas, 2014:            Earth & Environment, 1(1), 47, doi:10.1038/s43247-020-00044-z.
What dynamics drive future wind scenarios for coastal upwelling off Peru        Bevan, S.L., A.J. Luckman, D.I. Benn, T. Cowton, and J. Todd, 2019: Impact of and Chile? Climate dynamics, 43(7-8), 1893-1914.                                  warming shelf waters on ice m&#xe9;lange and terminus retreat at a large SE Belmecheri, S., F. Babst, E.R. Wahl, D.W. Stahle, and V. Trouet, 2016: Multi-        Greenland glacier. Cryosphere, 13(9), 2303-2315, doi:10.5194/tc-13-2303-century evaluation of Sierra Nevada snowpack. Nature Climate Change, 6,            2019.
2-3, doi:10.1038/nclimate2809.                                                  Bevis, M. et al., 2019: Accelerating changes in ice mass within Greenland, and Belt, S.T., 2018: Source-specific biomarkers as proxies for Arctic and              the ice sheets sensitivity to atmospheric forcing. Proceedings of the National Antarctic sea ice. Organic Geochemistry, 125, 277-298, doi:10.1016/j.              Academy of Sciences, 116(6), 1934-1939, doi:10.1073/pnas.1806562116.
orggeochem.2018.10.002.                                                        Bianchi, C. and R. Gersonde, 2002: The Southern Ocean surface between Belt, S.T., 2019: What do IP25 and related biomarkers really reveal about sea        Marine Isotope Stages 6 and 5d: Shape and timing of climate changes.
ice change? Quaternary Science Reviews, 204, 216-219, doi:10.1016/j.              Palaeogeography, Palaeoclimatology, Palaeoecology, 187(1-2), 151-177, quascirev.2018.11.025.                                                            doi:10.1016/s0031-0182(02)00516-3.
Belt, S.T. et al., 2007: A novel chemical fossil of palaeo sea ice: IP25. Organic Biastoch, A. et al., 2015: Atlantic multi-decadal oscillation covaries with Geochemistry, 38(1), 16-27, doi:10.1016/j.orggeochem.2006.09.013.                  Agulhas leakage. Nature Communications, 6(1), 10082, doi:10.1038/
Belt, S.T. et al., 2015: Identification of paleo Arctic winter sea ice limits and    ncomms10082.
the marginal ice zone: Optimised biomarker-based reconstructions of            Bilbao, R.A.F., J.M. Gregory, and N. Bouttes, 2015: Analysis of the regional late Quaternary Arctic sea ice. Earth and Planetary Science Letters, 431,          pattern of sea level change due to ocean dynamics and density change 127-139, doi:10.1016/j.epsl.2015.09.020.                                          for 1993-2099 in observations and CMIP5 AOGCMs. Climate Dynamics, Bendtsen, J. et al., 2017: Sea ice breakup and marine melt of a retreating          45(9-10), 2647-2666, doi:10.1007/s00382-015-2499-z.
tidewater outlet glacier in northeast Greenland (81&deg;N). Scientific Reports,    Bilbao, R.A.F., J.M. Gregory, N. Bouttes, M.D. Palmer, and P. Stott, 2019:
7(1), 1-11, doi:10.1038/s41598-017-05089-3.                                        Attribution of ocean temperature change to anthropogenic and natural Benn, D.I., T. Cowton, J. Todd, and A. Luckman, 2017: Glacier Calving in            forcings using the temporal, vertical and geographical structure. Climate Greenland. Current Climate Change Reports, 3(4), 282-290, doi:10.1007/            Dynamics, 53(9), 5389-5413, doi:10.1007/s00382-019-04910-1.
s40641-017-0070-1.                                                              Bindoff, N.L. and T.J. McDougall, 2000: Decadal Changes along an Indian Ocean Bennartz, R. et al., 2013: July 2012 Greenland melt extent enhanced by              Section at 32&deg;S and Their Interpretation. Journal of Physical Oceanography, low-level liquid clouds. Nature, 496, 83-86, doi:10.1038/nature12002.              30(6), 1207-1222, doi:10.1175/1520-0485(2000)030<1207:dcaaio>2.0.
Bennetts, L.G., S. OFarrell, and P. Uotila, 2017: Brief communication: Impacts      co;2.
of ocean-wave-induced breakup of Antarctic sea ice via thermodynamics          Bindoff, N.L. et al., 2013: Detection and Attribution of Climate Change:
in a stand-alone Version of the CICE sea-ice model. The Cryosphere, 11(3),        from Global to Regional. In: Climate Change 2013: The Physical Science 1035-1040, doi:10.5194/tc-11-1035-2017.                                            Basis. Contribution of Working Group I to the Fifth Assessment Report Bentamy, A. et al., 2017: Review and assessment of latent and sensible heat          of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, flux accuracy over the global oceans. Remote Sensing of Environment, 201,          G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, and 196-218, doi:10.1016/j.rse.2017.08.016.                                            P.M. Midgley (eds.)]. Cambridge University Press, pp. 867-952, doi:10.1017/
Bentley, M.J. et al., 2014: A community-based geological reconstruction of          cbo9781107415324.022.
Antarctic Ice Sheet deglaciation since the Last Glacial Maximum. Quaternary    Bindoff, N.L. et al., 2019: Changing Ocean, Marine Ecosystems, and Dependent Science Reviews, 100, 1-9, doi:10.1016/j.quascirev.2014.06.025.                    Communities. In: IPCC Special Report on the Ocean and Cryosphere in Benz, V., O. Esper, R. Gersonde, F. Lamy, and R. Tiedemann, 2016: Last Glacial      a Changing Climate [Prtner, H.-O., D.C. Roberts, V. Masson-Delmotte, Maximum sea surface temperature and sea-ice extent in the Pacific sector          P. Zhai, M. Tignor, E. Poloczanska, K. Mintenbeck, M. Nicolai, A. Okem, of the Southern Ocean. Quaternary Science Reviews, 146, 216-237,                  J. Petzold, B. Rama, and N. Weyer (eds.)]. In Press, pp. 447-588, www.ipcc.
doi:10.1016/j.quascirev.2016.06.006.                                              ch/srocc/chapter/chapter-5.
1324
 
Ocean, Cryosphere and Sea Lev}}

Latest revision as of 07:01, 13 November 2024

Exhibit 6 - Rebert E. Kopp Declaration
ML23331A986
Person / Time
Site: Turkey Point  NextEra Energy icon.png
Issue date: 11/27/2023
From:
Miami Waterkeeper
To:
NRC/SECY/RAS
SECY RAS
References
RAS 56848, 50-250-SLR-2, 50-251-SLR-2
Download: ML23331A986 (0)
v • d • e


Text

Retrieved from "https://kanterella.com/w/index.php?title=ML23331A986&oldid=3692200"