ML15078A206

to Engineering Evaluation 14-E16, Dominion Flooding Hazard Reevaluation Report for Millstone, Units 2 and 3, in Response to 50.54(F) Information Request Regarding Near-Term Task Force Recommendation 2.1: Flooding, Pp. 2-58 Through 2-125

ML15078A206
Person / Time
Site:	Millstone
Issue date:	03/12/2015
From:	Zachry Nuclear Engineering
To:	Dominion Nuclear Connecticut, Office of Nuclear Reactor Regulation
Shared Package
ML15078A203	List: ML15078A204 ML15078A205 ML15078A206 ML15078A207 ML15078A208
References
15-106 14-E16, Rev. 1
Download: ML15078A206 (68)

Text

MIACR DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.2.4. Storm SurgeAn evaluation of the Probable Maximum Storm Surge (PMSS) flood hazard at MPS wasperformed in a manner consistent with the HHA approach (NRC, 2011). The evaluation includeddetailed analyses of the Probable Maximum Hurricane (PMH), the resulting deterministic stillwater elevation (i.e., the water surface elevation in the absence of waves, wave set-up andriver flood PMSS) and the probabilistically-derived 1 E-6 Annual Exceedance Probability (AEP)stillwater elevation.

To support the deterministic PMSS analysis, parameters defining the PMH were developed through a review of National Weather Service (NWS) guidance and a Site and Region SpecificHurricane Climatology Study, which included an analysis of a large set of synthetic hurricane data. Hydrodynamic modeling was then used to simulate the parameterized PMH and identifythe resulting PMSS elevation at MPS.The Site and Region Specific Hurricane Climatology Study also supported the evaluation of the1 E-6 AEP stillwater elevation by providing probabilistic definitions of key hurricane parameters.

These definitions were used as input to a method of recovering combined event probabilities toevaluate the relationship between maximum stillwater elevation and AEP at MPS.The methodology and results associated with the storm surge analysis at Surry Power Station(SPS) in Surry, Virginia were the subjects of an independent review performed by Dr. Donald T.Resio, Professor of Ocean Engineering at the University of North Florida.

As the methodology used at SPS aligns with the approach taken with respect to assessing storm surge effects atMPS, Dr. Resio's review (i.e., methodology only) has been extended to be applicable to theapproach described in the following sections (refer to Appendix D for further details with respectto the specific elements of this assessment to which the extended review is applicable).

Thefavorable findings of the SPS review act to support the methods used to assess the PMSSflooding hazard at MPS.2.4.1 Methodology The following sections summarize the methodology used to evaluate the PMH, the deterministic stillwater PMSS elevation (i.e., deterministic storm surge) and the probabilistically-derived 1 E-6AEP stillwater PMSS elevation (i.e., probabilistic storm surge) at MPS.2.4.1.1 Probable Maximum Hurricane A step-wise approach consistent with HHA methodology, as described by NUREG/CR-7046 (NRC, 2011), was used to deterministically evaluate the PMH at MPS. The evaluation includedanalyses of National Hurricane Center (NHC) historical, "Best-Track" hurricane data (i.e.,HURDAT2) and, to supplement the limited historical data, synthetic hurricane datarepresentative of a large set of synthetic tropical cyclone tracks. The synthetic data weredeveloped for the MPS region by Dr. Kerry Emanuel of WindRiskTech, LLC (WRT) usingcoupled intensity and atmospheric models (WRT, 2013). Details of the development methodology, comparisons with historical hurricane data and the application of extant methodsare described in two key peer reviewed references (Emanuel et al, 2004; Emanuel et al 2006).The methodology used to develop the synthetic tropical cyclone tracks and storm parameters includes:

1) storm generation;

2) storm track generation; and 3) deterministic modeling ofEE 14-El 6, REV. 12-58 DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.hurricane intensity.

Using this methodology, a large number (10,013) of synthetic storm tracks(i.e., WRT storm set) was generated and filtered within a radius of 200 kilometers (kin) ofWesthampton, New York to support the evaluation of the PMH at MPS. The WRT storm set wasverified by evaluating the statistical consistency with the HURDAT2 data set and, wherepresent, variance from the historical hurricane data was identified.

Comparisons wereperformed for parameters reflecting storm intensity, direction, size and speed. Based on thesecomparisons, the WRT data set was ultimately validated as a conservative reflection of stormcharacteristics within the MPS vicinity.

Following validation, the WRT storm set was spatially filtered to eliminate data not reflective of conditions within the potential storm surge production region (i.e., filtered to include only over-water points within the 200 km radius zone)The following steps were used to evaluate the historical and synthetic data and characterize thePMH at MPS:1. Identification of controlling event type: hurricanes and extra-tropical storms are ofprimary concern in Atlantic coastal areas located in the regional area of MPS. The firststep is to confirm which type of storm event (i.e. hurricane or extra-tropical storm)controls the PMSS. This evaluation is performed by: 1) reviewing recorded water leveldata from the NOAA Co-operative (Co Op) Stations at the Newport, Rhode Island(Station

8452660, "Newport"),

New London, Connecticut (Station

8461490, "NewLondon"),

Bridgeport, Connecticut (Station

8467150, "Bridgeport")

and Montauk, NewYork (Station

8510560, "Montauk")

stations to identify the events that resulted inhistorical extreme water levels (Figure 2.4-1); 2) examining the extreme water levelspredicted for Category 1 through Category 4 Hurricanes storm surge elevations byNOAA using the Sea, Lakes, and Overland Surges from Hurricanes (SLOSH) softwaremodel and presented in the NOAA SLOSH Display Program at the four NOAA Co-Opstations located near MPS relative to recorded water level data (NOAA 2012a); and 3)reviewing the historical hurricanes in the vicinity of MPS using historical storm datafrom the years 1851-2010 (Blake et al 2011 and NOAA 2013a).2. Determine NWS-23 PMH parameters:

consistent with guidance presented inNUREG/CR-7046 (NRC, 2011), ranges of permissible PMH meteorological parameters are initially determined using NWS-23 (NOAA 1979). These parameters include the following:

1) peripheral

pressure,

2) central pressure,

3) permissible rangefor radius of maximum winds, 4) permissible range of forward speeds, 5) permissible range of track direction, and 6) estimated maximum 10-meter, 10-minute over-water wind speed of the key PMH parameters.

3. Site and Region Specific Hurricane Climatology Study: the Site and Region SpecificHurricane Climatology Study included a statistical analysis of the HURDAT2 databaseand verification and statistical analysis of the synthetically-developed hurricane parameter data set. The analysis focused on data reflecting storm intensity, direction and physical dimensions in the region of MPS. Parameter selection was based on dataavailability within the HURDAT2 database and relevance with respect to comparison toparameter estimates derived from NWS-23. Probability Density Functions (PDFs) wereconstructed for these parameters from Probability Density Histograms (PDHs) using anon-parametric kernel method. To further refine the analysis of the low probability portion of the 1-min, 10-m average wind speed (mxw) distribution, Extreme ValueAnalysis (EVA) was used based on the Peak Over Threshold (POT) method and theEE 14-El 6, REV. 12-59 DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Generalized Pareto Distribution (GPD). A detailed statistical analysis of the filteredWRT storm data was then performed including a univariate storm parameter probability

analysis, an analysis of storm parameter covariance, development of alarge synthetic storm set extension (i.e., the 3,000,000 or 3M data set), and anevaluation of error, uncertainty, and conservatism.

Based on this analysis, adimensionless scaling function was developed to conservatively reflect thedeterministic upper-limit of storm intensity (i.e., maximum wind speed) in the MPSvicinity in consideration of co-variability with storm direction (i.e., storm bearing).

Additional relationships were also developed to characterize additional stormparameters (e.g., forward speed and radius of maximum winds) in consideration ofparameter co-variability.

The above-described methodology provided input to the deterministic and probabilistic stormsurge analyses described below.2.4.1.2 Deterministic Storm SurqeA step-wise approach consistent with HHA methodology, as described by NUREG/CR-7046 (NRC, 2011), was used to deterministically evaluate the PMSS stillwater elevation at MPS. Asdiscussed below, two different hydrodynamic models were applied in a phased approach.

A screening-level assessment was performed using the two-dimensional Sea, Lakes andOverland Surges from Hurricanes (SLOSH) computer model (NOAA 2012a and NOAA 2012b).SLOSH is computationally efficient, allowing many simulations to be performed over a relatively short period of time; however, the SLOSH model has limitations, including its relatively coarse,structured model grid and the inability to represent dynamic tides and external boundary fluxes(e.g., river flow). Therefore, in a second phase of modeling (i.e., refinement-level assessment),

additional simulations were performed using the ADvanced CIRCulation (ADCIRC) model(USACE, 1994). While ADCIRC is not hindered by many of the limitations associated withSLOSH, the high-resolution, finite-element mesh and related high computational demandprevent broad applications (i.e., only a limited number of storm simulations is practicable in thecontext of a given analysis).

Therefore, ADCIRC was applied in a targeted fashion (i.e.,refinement-level assessment) to further evaluate the storms identified during the screening-level assessment as potentially causing large surges at MPS and develop the final PMSS stillwater elevations.

Inclusive of the phased modeling

approach, the methodology included the following steps:1. Generation of the Initial Storm Set: An Initial Storm Set was generated using the MPSPMH results as input. Hypothetical storm tracks were first created by combining 10potential storm bearings (i.e., -600 to +300 in 100 intervals) with five potential landfalllocations between NWS 23 Mile Posts 2550 and 2650 (NOAA 1979). The storm trackswere assigned maximum wind speeds based on the bearing-specific values identified in the MPS PMH. Each potential storm track was then expanded into a set of stormsusing the bearing-specific ranges of forward speed and radius of maximum wind. Eachrange was divided into finite units, and a unique hypothetical (i.e., synthetic) storm wascreated for each combination.

2. Calculation of the Antecedent Water Level: An Antecedent Water Level (AWL) wascalculated using data obtained from the Newport, Rhode Island NOAA tidal gagingEE 14-El 6, REV. 12-60 00111DOMINION FLOODING HAZARD REEVALUATION REPORT FORZACHRY o,,N+oo,,,+o=++-o MILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.station per applicable regulatory guidelines (NRC, 2011 and ANS, 1992). Inaccordance with these guidelines, observed monthly maximum tide data obtained overa continuous 21-year period (i.e., January 1, 1993 through December 31, 2013) wereused to calculate the 10% exceedance high tide. A factor representing attenuation tothe Watch Hill Point, Rhode Island subordinate station at the high tide condition wasthen applied, and cumulative Sea Level Rise (SLR) was then added to obtain theAWL.3. Screening-Level Assessment (SLOSH):

Screening-level storm surge simulations wereperformed using the SLOSH model, the Initial Storm Set and the AWL. Thesesimulations were performed to identify:

1) the sensitivity of storm surge at MPS todifferent storm parameters (i.e., storm track, radius of maximum winds, etc.) asconstrained by the PMH results; and 2) the specific combinations of storm parameters and storm tracks that result in the largest predicted storm surges at MPS, alsoconstrained by the PMH results.

The screening-level simulations performed usingSLOSH assumed steady-state conditions (i.e., storm parameters were not varied fromthe initial specifications).

4. Selection of the Refinement Storm Set: A Refinement Storm Set was selected forADCIRC simulations after processing the results of the screening-level assessment (i.e., SLOSH model predictions).

Maximum simulated stillwater elevations wererecovered from the model output at the MPS intake location.

These simulated elevations were ranked and sorted, and storms within the Initial Storm Set resulting inSLOSH-simulated stillwater elevations exceeding an applied threshold of 20 feetrelative to the North American Vertical Datum of 1988 or NAVD88 were identified andcombined to form the Refinement Storm Set.5. Refinement-Level Assessment (ADCIRC):

Refined storm surge simulations wereperformed using ADCIRC and a representation of dynamic tidal fluctuations for stormsindicated by the screening-level assessment (i.e., Step 2, above) to be representative of the PMH. A preliminary sensitivity analysis was performed to establish storm arrivaltiming such that maximum stillwater elevations were achieved at MPS within the model(i.e., storm surge combined with variations in tide). Simulations were performed assuming steady conditions similar to the screening-level assessment for the purposeof comparing ADCIRC to SLOSH.The above-described methodology produced simulated stillwater elevations at a location withinthe ADCIRC model domain representative of the MPS intake location.

Because dynamic tideswere simulated based on astronomical/predicted tides (i.e., excluding meteorological effects andSLR), a final adjustment was made to linearly add the difference between the peak predicted tide and the calculated AWL. The results represent maximum predicted stillwater (i.e., withoutwind-wave action) PMSS elevations at MPS.2.4.1.3 Probabilistic Storm SurgeA step-wise approach consistent with HHA methodology, as described by NUREG/CR-7046 (NRC, 2011), was used to probabilistically evaluate the PMSS stillwater storm surge elevations at MPS. Similar to the deterministic analysis described in the previous

section, two different hydrodynamic models (i.e., SLOSH and ADCIRC) were used in a phased approach thatEE 14-E16, REV..12-61 Z EW DOMINION FLOODING HAZARD REEVALUATION REPORT FORZAC HRYMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.involved applications of the Joint Probability Method (JPM) and Joint Probability Methodology with Optimal Sampling (JPM-OS).

The JPM is a statistical method developed in the 1970s that is commonly used toprobabilistically evaluate the coastal surge risk due to tropical cyclones (ESSA, 1970). The JPMutilizes a set of synthetic storms representing local climatology by combining individual stormmeteorological parameters (e.g., intensity,

bearing, forward speed, etc.), with each storm havinga joint probability of occurrence calculated from the combined probability of each of theindividual parameter probabilities.

A frequency response relationship may then be derived fromthese parameter combinations and associated probabilities via a hydrodynamic model.The most significant limitation associated with applying the JPM is the significant computational requirement; many simulations representing a large number of unique parameter combinations are required in order to characterize the JPM integral.

This limitation is exacerbated when usingmore robust, higher-precision numerical hydrodynamic models that require more computational effort (i.e., longer simulation times). To address this limitation and facilitate the use of thesemore robust models, modified forms of the JPM that rely on characterizing storm parameter space with fewer simulations through Optimal Sampling (OS) techniques have been developed (FEMA, 2012).The Joint Probability Method -Optimal Sampling, or JPM-OS, was developed to statistically characterize surge-frequency relationships in the same manner as the traditional JPM but usingfewer storm surge simulations.

By limiting the number of storm surge simulations

required, arobust, computationally intensive storm surge model, such as ADCIRC, can be applied.

Thebasic JPM OS concept is to: 1) define the complete surge-frequency curve using acomputationally efficient surge model (i.e., the NOAA SLOSH model); 2) statistically represent the complete surge-frequency relationship using relatively few storms, based on the appropriate selection of storm parameters and an understanding of surge response to varying stormparameters;

3) perform storm surge simulations for the storm subset using a robust storm surgemodel (i.e., ADCIRC);

and 4) extend the results of these simulations to define the surge-frequency relationship at MPS.The JPM-OS Response Surface technique developed by the United States Army Corps ofEngineers (USACE) and applied by the Federal Emergency Management Agency (FEMA) wasused (FEMA, 2012). In this approach, parameter space is characterized primarily viainterpolation and extrapolation based on a carefully-selected set of reference parameter combinations or synthetic storms, where parameter perturbations and sensitivity testing basedon this reference set are used to evaluate response functions.

In concept, the JPM-OScalculation involves two steps: 1) searching for a reference storm based on the proximity of thegiven subject storm parameter combination to reference storm parameters; and 2) applyingbest-estimated surge responses along the multi-dimensional space.The steps used to evaluate response functions and apply the JPM and JPM-OS are described as follows:1. Creation of the JPM Storm Set: Storm parameters, including storm bearing (i.e.,translational direction, fdir), forward speed (fspd), radius of maximum winds (RMW)and maximum (i.e., 1-minute average at an altitude of 10 meters) wind speed (Vm)were combined to generate hypothetical storms. Each storm track (i.e., storm bearingand landfall location) was expanded into a set of storms by considering ranges ofEE 14-E16, REV. 12-62 EW DOMINION FLOODING HAZARD REEVALUATION REPORT FORZ'ACHRYMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.forward speed, radius of maximum winds and maximum wind speed. Each range wasdiscretized into units of 5 knots (kt), 5 nautical miles (nm) and 10 kt, respectively.

Thisdiscretization process resulted in 8 potential values of forward speed ranging from 15kt to 50 kt, 9 potential values of radius of maximum winds ranging from 15 nm to 55nm, and 8 potential values of maximum wind speed ranging from 70 kt to 140 kt. Aunique hypothetical storm was created for each combination of values, storm trackdirection and landfall location (i.e., 8 x 9 x 8 x 11 x 5 = 31,680 hypothetical storms).Joint probabilities based on simultaneous occurrences of the Fdir, Fspd, RMW and Vmparameters were calculated for each hypothetical storm by querying the 3M data set torecover parameter co-variability.

Hypothetical storms with joint probabilities of zero(i.e., no events within the 3,000,000, or 3M, data set matching or exceeding theparameter combination; a joint probability of less than 3.33E-7 or less than 1 in3,000,000) were eliminated; the remaining 13,485 parameter combinations represented the JPM Storm Set. This method of establishing limits on parameter combinations differs from use of the 3M data set in the evaluation of the PMH, where adimensionless scaling relationship was derived to adjust the NWS 23 maximum windspeed in recognition of co-variability of intensity and storm bearing.2. Addition of tidal condition:

An initial condition (i.e., static starting water level) wasrequired by SLOSH for each simulation in performing the JPM analysis.

Mean HighWater (MHW) and Mean Low Water (MLW) at the Newport, Rhode Island (RI) NOAACO-OPS station attenuated to the Watch Hill Point, RI subordinate station wereselected as being representative of high and low tide conditions at the site,respectively.

Both conditions were conservatively represented as having equaloccurrence probabilities of 0.5 (i.e., equal probabilities of a hypothetical stormoccurring at conditions representative of high and low tides).3. Calculation of Annual Exceedance Probabilities:

Annual Exceedance Probability (AEP)values were calculated for each hypothetical storm based on the joint probabilities calculated during Step 2. To develop surge-frequency relationships reflecting annualized probabilities, two additional factors were considered.

First, the jointprobabilities were multiplied by a normalized (i.e., normalized per unit length ofcoastline within the study area) annual storm occurrence frequency (i.e., 0.20371storms per year divided by 400 kilometers, or the diameter of the capture zone fromthe PMH evaluation).

As a final adjustment, the annualized probabilities weremultiplied by a factor of 0.5 to represent coincidence with a high or low tide condition.

Thus, for each simulated JPM Storm Set event, the maximum stillwater surge eventwas assigned an AEP value representing the storm parameter combination, the annualstorm frequency and the tidal condition.

4. Performance of SLOSH simulations:

A total of 26,970 simulations (i.e., 13,485 JPMStorm Set events, each at a high and a low tide condition) were performed using theSLOSH model. These simulations were performed to: 1) develop a preliminary surge-frequency relationship at MPS based on potential hurricane parameter combinations constrained by the PMH results; and 2) identify the set of storms to be simulated withADCIRC for the purpose of refining this surge-frequency relationship.

Results of theSLOSH simulations were extracted in the form of maximum stillwater surge elevations at several SLOSH model cells including the cell representing MPS.EE 14-E16, REV. 12-63 DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.5. Generation of an initial stillwater surge-frequency relationship:

As a first step indeveloping an initial stillwater surge-frequency relationship at MPS, histograms werecreated for the MPS location using maximum stillwater surge elevations and theassociated AEP values produced during Step 7. Bin sizes were defined as increments (i.e., 0.5 feet [ft]) of stillwater surge elevation based on the results of a sensitivity

analysis, and AEP values associated with the storms producing stillwater surgeelevations falling into each bin were summed. The summed AEP values were thensummed again from highest elevation bin to the lowest elevation bin to produce bin-specific cumulative AEP values. The stillwater surge elevations representing the centerof each bin were then plotted versus their respective cumulative AEP values togenerate an initial stillwater surge-frequency relationship at MPS.6. Identification of the OS Storm Set: As previously noted, perhaps the most significant challenge associated with the JPM-OS (i.e., where OS refers to optimal sampling) technique used in this analysis is selecting the storm parameter combinations used toformulate the basis for evaluating surge response.

To guide this selection process inthe case of this analysis, experiments were performed using the initial surge-frequency relationship developed using SLOSH. The goal of the experiments was to identify theminimum number of storm parameter combinations required to reproduce (i.e., usingJPM-OS) the surge-response relationship at MPS with reasonable accuracy.

Ultimately, the experiments suggested that the very-low probability range of the surge-frequency relationship was accurately reproduced using 55 production runs spanningthe -50° to +500 bearing range (i.e., the reference set) and 16 additional simulations toestablish sensitivities to various parameter perturbations (i.e., the sensitivity set).7. Performance of ADCIRC simulations:

As a first step in refining the surge-frequency relationship at MPS, reference and sensitivity set parameter combinations identified during Step 9 were simulated using ADCIRC (i.e., a total of 71 simulations).

Whereasinitial conditions representative of high and low tide levels were specified as input tothe SLOSH simulations, a mean tide initial condition was used for the ADCIRCsimulations; coincidence between surge and fluctuating tides was addressed duringfinal adjustments (i.e., refer to Step 9). Results of the ADCIRC simulations wereextracted in the form of maximum stillwater surge elevations at a location within theADCIRC mesh representative of the MPS intake.8. Refinement of stillwater surge-frequency relationships:

A refined stillwater surge-frequency relationship was developed for MPS based on the results of the ADCIRCsimulations using the JPM-OS technique.

Maximum stillwater elevations wereestimated via JPM-OS for an expanded set of storms (i.e., including two additional potential forward speeds: 5 and 10 kt) to support assessment of aleatory variability inmaximum wind speeds (i.e., Step 9). AEP values were adjusted to correct for the useof a single initial condition (i.e., mean tide) prior to histogram development.

9. Adjustments to the surge-frequency relationship to reflect uncertainty, error and sealevel rise: Adjustments to account for uncertainty (i.e., epistemic uncertainty andaleatory variability),

error and projected sea level rise (SLR) were required in order toprobabilistically characterize storm surge at MPS. The uncertainty adjustments considered the following factors:

variability in tide occurring coincident with maximumstorm surge; model skill associated with ADCIRC and the applied wind/vortex EE 14-El 6, REV. 12-64 2P DOMINION FLOODING HAZARD REEVALUATION REPORT FORZ'ACHRYMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.formulation; and aleatory variability and error associated with maximum wind speedspecifications.

A uniform adjustment for SLR was added linearly as a final step.The above-described methodology produced an estimated stillwater surge-frequency relationship for MPS. The 1E-6 AEP stillwater elevation was then extracted from thisrelationship.

2.4.2 Results2.4.2.1 Probable Maximum Hurricane The following sections describe the results of the evaluation of the PMH at MPS.2.4.2.1.1 Determination of the Controlling Storm EventAs Table 2.4-1 indicates, both extra-tropical storms and hurricanes have resulted in significant coastal storm surges at the four stations (i.e., Newport, New London, Bridgeport, and Montauk).

The data indicate that three to four of the top five extreme water level events were caused byeither a hurricane or tropical storm for the site vicinity.

Figure 2.4-3 shows the historical stormtracks intersecting the area of interest, including those storms responsible for many of therecorded high water levels at the stations identified above.Storm surge elevations predicted by NOAA (NOAA 2012a) using the SLOSH model forhurricanes ranging from Category 1 to Category 4 were compared to the historical water leveldata. The purpose of this comparison was to determine whether large hurricanes are expectedto result in storm surges greater than those measured in the historical record resulting fromextra-tropical storms. A comparison of the Maximum of MEOW (MOM, where MEOWrepresents Maximum Envelope Of Water) values presented in Table 2.4-2 to the recorded waterlevels in Table 2.4-1 indicates that historic extra tropical storms have caused storm surgessimilar to those predicted for simulated Category 1 or 2 hurricanes.

Table 2.4-2 also confirms that the recorded water levels resulting from historic hurricanes havecaused storm surges similar to those predicted for simulated Category 2 and 3 hurricanes.

Bydefinition, the PMH is a "hypothetical steady state hurricane having a combination of values ofmeteorological parameters that will give the highest sustained wind speeds that can probablyoccur at a specified coastal location" (NOAA 1979). At higher hurricane category levels (i.e.,Category 3 and above), the potential surge elevations predicted by NOAA significantly exceedhistorical water levels recorded at the CO-OPS stations.

Finally, the frequency of hurricane strikes on the U.S. East Coast was analyzed in the NWSNHC-6 using the data from 1851 to 2010. Of all hurricanes making landfall in the U.S., 0.7percent struck New Jersey, 4.2 percent struck New York, 3.9 percent struck Connecticut and3.2 percent struck Rhode Island. According to NHC-6, the coastline represented by thesestates was impacted by 6 major hurricanes (Category 3 or higher) between 1851 and 2010(Blake et al 2011).Some of these storms occurred prior to 1900, and as a result, many accounts do not includestorm surge, tide values, central pressure, or other specific storm details.

With respect to thestorms that are well characterized by historical

records, many weakened significantly ordramatically changed direction prior to reaching the MPS vicinity; therefore, only some of theseEE 1.4-El 6, REV. 12-65 ZACHIRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.storms produced significant storm surges. For storms that did impact the subject area after1900, available track information is shown in Figure 2.4-4.Based on: 1) a review of historical extreme water level data from the NOAA CO-OPS stations;

2) an examination of the extreme water level events associated with predicted hurricane stormsurge elevations produced by NOAA using the SLOSH model at the four NOAA CO-OPSstations located near MPS; and 3) a review of available historical storm information, it isconcluded that a major hurricane (i.e., the PMH) will be the controlling storm resulting in thePMSS at MPS.2.4.2.1.2 Determination of Hurricane Parameters from NWS-23The location of MPS is shown in Figure 2.4-5 in relation to coastal distance intervals (i.e., mileposts) presented in NWS 23 (NOAA 1979). Storm surge at MPS is caused by surge generated within Long Island Sound and/or coastal storm surge developed to the east of Montauk.

Asindicated on Figure 2.4-5, MPS is located approximately between NWS 23 mile posts 2575 and2650 (more specifically, mile post 2650 approximately represents the opening of Long IslandSound to the Atlantic Ocean), where coastal distance is measured in nautical miles from theGulf of Mexico. Based on the location of MPS the following range of PMH parameters are asfollows:Parameter Lower Limit Upper LimitPeripheral Pressure (millibar) 1020 1020Central Pressure (millibar) 907.4 908.5Radius of Max Winds 13 32(nautical miles)Forward Speed (knots) 35 48Track Direction (degrees) 80 1901-min, 10-meter over water 169.9 169.9wind speed (miles per hour)The methods of parameter development presented in NWS 23 are generally not consistent withthe current state of knowledge for characterizing the PMH affecting the MPS vicinity.

In specificreference to PMH intensity reflected by maximum wind speed, NWS 23 values are recognized as lacking a reflection of the relationship between storm direction and storm magnitude (i.e., co-variability).

Thus, a detailed Site and Region Specific Hurricane Climatology study wasperformed to develop the hurricane meteorological parameters for analysis of flooding due tocombined storm surge and wind-generated waves.EE 14-El 6, REV. 1 2-66EE 14-El 6, REV. 12-66 ZI CW DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.2.4.2.1.3 Site and Region Specific Hurricane Climatology Study2.4.2.1.3.1 Statistical Analysis of Historical DataUnivariate Parameter Probability AnalysisBest track positions of tropical storms and hurricanes are maintained by the NHC in theannually-updated HURDAT2 database (NOAA 2013a). The official HURDAT2 databasecontains data representing cyclones occurring between 1851 and 2012. The analysis of theHURDAT2 database for the MPS study area used all applicable 1851-2012 storm data,including the 1979-2012 subset containing central pressure data.As the focus is on land-falling storms in the vicinity of Southern Long Island (i.e., since thesestorms result in large storm surges at MPS), the HURDAT2 database was filtered to extractstorm parameters associated with three zones of increasing spatial coverage, as shown inFigure 2.4-6, with the Inner Region (IR) representing storms occurring in the MPS site vicinity:

• The IR -200 kilometer radius centered at Westhampton, New York;* The Outer Region (OR) -500 sector region extending 400 kilometers radially to thesouth-southeast of the IR center; and* The Remote Region (RR) -500 sector region extending 800 kilometers radially tothe south-southeast of the IR center.Probability Density Functions (i.e., PDFs) were developed for key hurricane parameters withinthe IR using a non-parametric kernel method. The resulting IR distributions were compared tosimilarly sized sample distributions developed by randomly drawing from the OR and RR. Thedifference between any IR sample distribution and the OR and RR distributions was evaluated to determine statistical consistency for the purpose of maximizing the sample population (i.e.,determine if an expanded spatial filter could improve sample size without bias). The non-parametric estimate of the population PDF for maximum wind speed, mxw, for the IR is shownin Figure 2.4-7 as a line. The 90% confidence

interval, indicated by gray shading, was estimated by randomly drawing sample sizes from the OR and RR equivalent to the IR population (i.e., N=25).Comparisons between the IR PDFs and the corresponding OR and RR sample-based PDFs forthe fdir, fspd, dmxw and cpd parameters are shown in Figure 2.4-8 through Figure 2.4-10. Muchlike the mxw comparison, substantial differences are evident between the IR PDF and the ORand RR populations.

With central pressures routinely recorded in the HURDAT2 database only after 1979, thesample sizes for analyzing central pressure and pressure tendencies are considerably reducedrelative to the full dataset.

For each mid-6-hour position in the three regions, the centralpressure

deficit, cpd, was conservatively calculated using a peripheral pressure of 1020 mb(30.12 in Hg). The resulting PDFs for the IR are compared to the OR and RR samples in Figure2.4-10. Figure 2.4-11 presents the PDF and Probability Density Histogram (PDH) for maximumsustained winds inside the IR.Because the HURDAT2 database contains tropical storm and hurricane data, all primaryEE 14-El 6, REV. 12-67 211111DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.distribution peaks correspond to cpd values of slightly greater than 20 mb (i.e., central pressureequal to 1000 mb). The PDF representing the distribution of cpd within the IR is bi-modal with asecondary peak at approximately 70 mb representing stronger storms. This separation in stormintensity probabilities within the IR represents an artifact of the limited sample size associated with this capture zone. In consideration of this result, the empirical distribution of the mxwparameter represents a preferable metric of storm intensity, as it has a much larger sample size.In summary, these results suggest that the IR cannot be reliably expanded to increase thehistorical data sample size.Hurricane Parameter Co-variability The annual probabilities from the parameter-specific, univariate PDFs can be directly combinedas a product to obtain joint probability estimates of various parameter combinations as long asthe distributions can be demonstrated to be independent.

If significant co-variability existsamong the hurricane parameters, the probabilities of certain combinations may be different fromthe product of their independent probabilities.

A cross-correlation matrix of four hurricane parameters (mxw, fdir, fspd, dmxw) from data withinthe IR, OR and RR is shown in Figure 2.4-12. The statistical significance of each cross-correlation was determined in a manner similar to determining the significance of eachparameter's distribution.

Although paired parameter correlations are quite low in general, many are statistically significant (i.e., highlighted in yellow).

Scatter plots of the paired parameters in the off-diagonal elements ofthe cross-correlation matrix shown in Figure 2.4-12 are presented in Figure 2.4-13 throughFigure 2.4-18 along with a least-squares estimate of each respective regression line. Thefigures are presented in order from least viable linear co-variability to most-viable co-variability, as assessed by visually inspecting the scatter of the parameter plotted on the ordinate axisalong the range of the parameter plotted on the abscissa axis.Figure 2.4-13 shows a complex co-variability of forward storm speed (fspd) and forwarddirection (fdir). The fastest moving storms are those moving toward the north and east (i.e.,bearings between 0 0 and 90 0); whereas, storms with strong eastward or westward motionstypically move at slower speeds. This characteristic is consistent among the three sampledregions.

Similarly, Figure 2.4-14 and Figure 2.4-15 indicate a strong non-linear relationship exists between the storm intensity, as measured by mxw, and forward storm direction (fdir) andforward speed (fspd), respectively.

With the exception of one strong storm moving nearly dueeast, storms moving with strong eastward and westward motions are less intense compared tostorms moving with northward or northeastward components, as indicated by Figure 2.4-14.Figure 2.4-16, through Figure 2.4-18 show scatter plots of storm intensity change, as indicated by dmxw, versus forward direction (fdir), storm intensity (mxw) and forward speed (fspd),respectively.

As indicated by Figure 2.4-16, the probability distributions of intensity changes arenonlinearly related to storm forward direction, at least in terms of the width of the distributions.

Storms moving in north to northeastward directions exhibit broader intensity changedistributions compared to the more westward and eastward moving storms. Within the IR, theEE 14-El 6, REV. 12-68 M;EA C I-RY DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.majority of storms are weakening when moving west of north.Figure 2.4-17 indicates that distributions of intensity changes as functions of intensity bulge inwidth for moderately-strong storms and exhibit narrower spreads at lower and higher intensities.

Figure 2.4-18 shows little variation in the breadth of the intensity change distribution withforward speed, and the small slope associated with the regression line suggests that this pairedparameter set exhibits nearly independent co-variability.

The nonlinearity exhibited by the variation of the spread in one parameter in relation to the valueof the second parameter is evident within all sampled regions.

This finding argues againstcharacterizing the co-variability by linear means. Also, in reference to Figure 2.4-13, whichshows the relationship between fdir and fspd, treating these parameters as independent wouldproduce artificially high probabilities that fast moving storms approach the IR with northeast ornorth-northeast bearings; a condition that is not consistent with the analysis of the HURDAT2dataset.In summary, the statistical analysis of historical data suggests that the IR cannot be reliablyexpanded to increase the historical data sample size. Furthermore, the data within the IRrepresent a small sample size for determining the joint probability of storm parameters, especially considering that many of the storms within the IR have made landfall south of theregion and are passing inland prior to approaching the study area. Fortunately, validation andanalysis of the synthetic WRT storm set, as presented below, indicates that parameter distributions are similar to those developed from historical data. Thus, the WRT storm setrepresents an acceptable basis for characterizing PMH parameters in the vicinity of MPS.2.4.2.1.3.2 Statistical Analysis and Verification of Synthetic Hurricane DataThe synthetic storm set contains over 10,000 storms (i.e., tropical storms and hurricanes) characterized by various angles, translational speeds, intensities, and maximum wind radii. Thestorm parameters are available at 2-hr intervals and represent 10,013 storms pre-screened toimpact the Long Island Sound area (i.e., the source of surge impacts to MPS). The pre-screening (i.e., limiting the synthetic storm set to storms with tracks that approach MPS within200 km of Westhampton, New York) has an effect on the probability density distribution of someparameters; a fact that is considered in the validation of the data set.Figure 2.4-19 through Figure 2.4-22 present the validation results for the hurricane parameters fdir, fspd, dmxw and mxw, respectively.

Each figure contains three panels, showing distribution comparisons for the IR (top), OR (middle),

and RR (lower) domains, respectively.

The grayshaded region with the central gray line within each panel provides estimates of the WRTpopulation distribution made from HURDAT2-sized sampling (i.e., sampling from the WRT stormset); whereas, the ten superimposed lines show PDFs calculated for the 10 HURDAT2 samples.As noted through inspection of the figures referenced above, the results of the validation indicate that use of the WRT representation of the empirical storm data for estimating independent and joint variability of hurricane parameters will contain a conservative bias. Whilestorm intensity is generally well-represented by the WRT data for major storms, the WRT's biastoward faster forward speeds and more westerly storms is expected, given the sensitivity tostorm surge within the Long Island Sound vicinity, to conservatively predict more frequent andlarger storm surges near MPS.EE 14-El 6, REV. 12-69 I C DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.2.4.2.1.3.3 PMH Parameter Calculations Following validation, the WRT storm set was used to statistically characterize storm parameters within the storm surge production region. The analysis was focused on data located primarily within the over-water area of the IR, referred to as the Offshore IR (OIR), as spatially filtering thedata to include only over-water points will produce a conservative set of storm parameters tosupport storm surge modeling.

Figure 2.4-23 shows the IR, the OIR and the reduced number ofstorms (i.e., 7,957) and 2-hr storm segments (i.e., 23,993) resulting from spatial filtering to thedimensions of the OIR versus the IR.With a reduction in the number of storms relative to the IR, the annual frequency of stormswithin the OIR is also reduced, as are the numbers of hurricanes of various intensities.

Table 2.4-3 lists the numbers of storms and 2-hr storm segments for all storms and for the threemajor storm categories.

Figure 2.4-24 shows the annual frequency of synthetic storms within theIR and OIR by year and as a 31-year average (i.e., the 1980 to 2010 period corresponding tothe synthetic storm simulations).

Univariate PDFs of relevant hurricane parameters from the WRT storm set filtered to the stormsurge production region are presented in Figure 2.4-25 through Figure 2.4-28. The vertical linesrepresent central points for each interval.

The tabulated probability densities apply to all datavalues within the stated intervals centered on the central point values. The middle rows of thetabulated probabilities represent interval-integrated values, and the bottom rows of valuesrepresent the middle row values multiplied by the adjusted annual frequency of occurrence forthe WRT storm set (i.e., 0.20371).

This adjusted annual frequency of occurrence represents theaverage year-by-year frequency of occurrence of all synthetic storms (i.e., intersecting the OIR)during the period between 1980 and 2010 within the storm surge production region.Using an extension of the WRT data set based on sampling from each univariate PDFdeveloped for synthetic data within the OIR (i.e., the 3M data set), parameters and parameter ranges for the PMH at MPS were developed in recognition of parameter co-variability.

PMHparameters were determined by identifying a dimensionless scaling function that recovered variability of the NWS 23 PMH maximum wind speed, as described below. While the process ofidentifying this scaling function involved probability calculations for parameter combinations (i.e.,storm bearing and maximum wind speed), the resulting parameter combinations represent conservative deterministic PMH upper limits, as the NWS 23 PMH maximum wind speed isused as the basis of function development.

In developing the dimensionless scaling function, AEP values were first assigned to maximumintensities for 10° storm bearing increments spanning the potential PMSS-causing sector (i.e.,100 sectors centered at -60°, -50°, -40 0...+300) by querying the 3M data set. Criteria based onupper and lower limits were assigned to reflect a considered bearing sector (e.g., greater thanor equal to -250 and less than -150 for the 100 bearing sector centered at -200). Then, thenumber of events within the 3M data set meeting these criteria (i.e., all parameters falling withinthe pre-defined bounds) was counted.

Finally, the resulting count for each considered bearingsector was divided by the size of the data set (i.e., 3,000,000) to produce the joint probability associated with the parameter combination; the reciprocal of this probability multiplied by theOIR's annual storm frequency (i.e., 0.20371) was calculated to reflect the return period (i.e.,reciprocal of the AEP) for each parameter combination.

EE 14-E16, REV. 12-70 ZI Cw DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.For extracted values exceeding 96 kt, adjustments were performed to scale intensity results toaccount for error introduced by the kernel-based method of PDF generation.

Scaling wasperformed by evaluating differences in intensity over the potential PMSS-causing sector (i.e., -600 to +300) described by differences between the extreme value analyses of the WRT and 3Mdata. For a given data set rank, the intensity predicted by an extreme value analysis fit to the3M data was increased to be consistent with the intensity predicted by the fit to the WRT data.Finally, bearing-specific PMH intensities were calculated using the following steps:1. Storm intensity variation with bearing was evaluated over the potential PMSS-causing sector (i.e., -600 to +300) and the sector calculated from NWS 23 (i.e., -100 0 to 10 0) by arbitrarily specifying data set ranks.2. As indicated by Figure 2.4-29, the 100 bearing interval associated with themaximum intensity was identified as 100 (i.e., storms from the 3M data set withbearings greater than or equal to 50 and less than 150).3. Using the NWS 23 PMH maximum wind speed (i.e., 147.6 kt), the data set rankassociated with the 3M data set intensity most closely matching this value for the100 bearing interval was identified (i.e., the 19th highest intensity for this bearingsector, 148.4 kt). This rank recognizes NWS 23 as being a conservative representation of PMH intensity within the -600 to +300 bearing sector.4. Using the PMH intensity and bearing rank, intensities were extracted from the 3Mdata set for the 100 bearing intervals spanning the potential PMSS-causing bearingsector (i.e. -600to +300).A least-squares regression line (i.e., fifth-order polynomial anchored to 147.6 kt at the 100bearing interval) was then applied, and bearing-specific PMH intensities (i.e., vm as a functionof fdir) were extracted from the regression function (Figure 2.4-29).

The considered bearingrange includes storms with bearings between -600 and 30° to provide a bounding parameter set(i.e., relative to the anticipated PMH) inclusive of more intense, northerly-bound storms.Ranges of the rmw and fspd parameters were developed on a bearing-specific basis using the3M data set and the results of the intensity analysis presented above (Figure 2.4-30).

In thecase of rmw, upper and lower parameter bounds were assigned based on the relationship tointensity, which shows a trend toward smaller radius and tighter range as intensity increases.

For fspd, upper and lower parameter bounds were assigned based on relationships to intensity and bearing, which indicate a general trend toward faster forward speed as bearing shifts from -60 0 to 30 0 and intensity increases.

Recommended PMH-level parameters and parameter ranges are presented in Table 2.4-4 overa range of bearings

(+/- 50, centered on 100 increments) to capture the storm that causes thePMSS on a bearing-specific basis. The parameter ranges for fspd and rmw reflect boundingconditions relative to the likely controlling (i.e., maximum surge-producing) event within eachstorm bearing range. The maximum wind speeds represent conservative PMH intensities inconsideration of applicable uncertainty and error, as supported by the following:

0 Conservative intensity bias of WRT storm set versus HURDAT2 data;EE 14-El 6, REV. 12-71 DOMINION FLOODING HAZARD REEVALUATION REPORT FORZACHR~YMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc." Assessment of potential uncertainty associated with the WRT storm set; and* Assessment of error introduced by non-parametric, kernel-based fit.Thus, the revised parameters and parameter ranges presented represent a conservative assessment of the PMH and provide input to the deterministic PMSS evaluation at MPS.Similarly, the PDFs developed from the WRT data within the OIR represent conservative reflections of parameter occurrence likelihoods and provide input to the probabilistic storm surgeevaluation at MPS.2.4.2.2 Deterministic Storm SurgeThe following sections describe the results of the deterministic evaluation of the PMSS at MPS.2.4.2.2.1 Generation of the Initial Storm SetFifty storm tracks were created based on ten potential bearings and five potential landfalllocations (Figure 2.4-31) to span the potential surge generation region in the MPS vicinity.

Eachtrack was then assigned a maximum intensity based on the bearing-specific maximum windspeed determined as part of the PMH analysis.

Finally, the tracks were assigned to uniquepairings of the remaining, discretized parameters (i.e., radius of maximum winds and forwardspeed) to create the Initial Storm Set of 1,395 hypothetical events, each with a unique stormidentification (STORMID) number.Whereas a single maximum wind speed was determined for each bearing, the radius ofmaximum winds and forward speed parameters were presented as ranges (i.e., with upper andlower bounds varying by bearing).

Thus, in generating the Initial Storm Set, these ranges werefinely discretized by multiples of 5 nm and 5 kt, respectively.

The upper and lower bounds ofthe ranges determined as part of the PMH analysis do not necessarily correspond to multiples of 5 nm or 5 kt; therefore, in generating the Initial Storm Set, the ranges were rounded to thenearest multiple (i.e., to span each range)2.4.2.2.2 Calculation of the Antecedent Water LevelIn accordance with NUREG/CR-7046 (NRC, 2011), the PMSS is required to be evaluated coincidentally with an AWL equal to the ten percent exceedance high tide plus long termchanges in sea level. The ten percent exceedance high tide is defined as the high tide level thatis equaled or exceeded by ten percent of the maximum monthly tides over a continuous 21 yearperiod. In accordance with ANSI/ANS-2.8-1992 (ANS, 1992), this tide can be determined fromrecorded tide data or from predicted astronomical tide tables.The ten percent exceedance high tide was calculated using recorded monthly maximum tideelevations from the Newport station attenuated to the Watch Hill Point subordinate station.Using this approach, a value of 2.796 ft NAVD88 was obtained.

In consideration of Sea LevelRise (SLR), which was projected over 50 years using the annual rate at the Newport, RI station,the AWL was determined to be 3.2 ft NAVD88.2.4.2.2.3 Screening-Level Assessment (SLOSH)As part of performing the screening-level assessment performed using the NOAA SLOSHmodel, results from the 1,395 simulations, in the form of simulated surge elevation time seriesfor each simulation, were extracted at four locations within the pv2 basin, including the modelEE 14-El 6, REV. 12-72 MWA CAN' ny DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.cell representing the MPS intake location (Figure 2.4-32 and Figure 2.4-33).

The time serieswere reduced to peak surge elevations at these locations for each simulated storm. The stormsresponsible for the largest simulated surges at MPS were identified, as discussed in thefollowing section.Figure 2.4-34 and Figure 2.4-35 summarize the results at MPS in the form of three-dimensional surfaces depicting maximum stillwater elevations as functions of storm bearing and forwardspeed or radius of maximum winds, respectively.

Figure 2.4-34 suggests limited sensitivity toforward speed with some notable response differences between northbound storms and stormsbearing west-of-north;

whereas, Figure 2.4-35 indicates significant sensitivity to the radius ofmaximum winds parameter but generally consistent behavior across the storm bearing range.Based on these responses, storm surge at MPS appears to be maximized by large-radius, slow-moving, northbound storms capable of moving significant coastal surge into the eastern LongIsland Sound region. Due to the location of the maximum winds within the idealized

cyclones, simulated storm surge is also maximized by storms making landfall west of the Long IslandSound opening.

These slow-moving storms are able to maintain momentum and route surgefrom the open ocean south and southeast of Long Island through the Long Island Soundopening and toward MPS.2.4.2.2.4 Selection of the Refinement Storm SetThe results of the screening-level assessment were used to target specific storm parameter combinations resulting in the largest stillwater surge elevations at MPS for refinement using theADCIRC model (i.e., refinement-level assessment, which was performed using the Refinement Storm Set). The parameter combinations associated with the Refinement Storm Set events aresummarized in Table 2.4-5. As Table 2.4-5 table indicates, the 17 storms making up theRefinement Storm Set represent 6 different potential storm bearings, 1 potential landfalllocation, 5 potential forward speeds and 3 potential radii of maximum winds.2.4.2.2.5 Refinement-Level Assessment (ADCIRC)Prior to performing dedicated refinement simulations, a sensitivity analysis was performed todetermine the appropriate storm arrival time relative to the tide phase at Newport and MPS thatwould produce the most conservative results (i.e., the highest stillwater elevations at MPS).Time of landfall was used as an indicator for storm arrival time in this analysis.

Five simulations were performed, each with a different time of landfall relative to high tide atNew London and MPS. The results indicated that when a storm made landfall one hour later(02:00) than the time for peak tide at Newport (01:00),

maximum water level was obtained atMPS. Therefore, the storm tide elevation was considered to be maximized when a storm madelandfall one hour later than the peak tide at Newport.Guided by the results of the tidal phasing sensitivity

analysis, ADCIRC simulations wereperformed using input defining the 17 storms within the Refinement Storm Set. Based on thesesimulations, the following combination of storm parameters was identified as being responsible for the deterministic PMSS at MPS:STORMID = 8180 Track Direction (0) = -30';EE 14-E16, REV. 12-73 ZI CW DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.* Landfall Mile Post = 2600 (Latitude 40.880, Longitude

-72.35°);

• Radius of Maximum Winds (Rmax) = 35 nm;* Forward Speed (Vf) = 20 kt* Maximum 1 -min, 1 0-m Overwater Wind Speed (Vm) = 125.6 kt; and* Central Pressure Deficit (CPD) = 107 mb.The ADCIRC simulation representing this combination of parameters (i.e., STORMID 818)resulted in a maximum stillwater elevation of 23.3 ft MSL (i.e., reflecting linear adjustment to theAWL) at the MPS intake location.

This deterministically-derived maximum stillwater elevation represents a conservative result in consideration of the following:

• Conservatisms associated with the deterministic PMH inputs, as previously discussed;

• Consideration of the sensitivity to tidal phasing and coincidence between peaksimulated tide and maximum storm surge; and* The conservative value of the AWL applied in this analysis.

2.4.2.3 Probabilistic Storm SurgeThe following sections describe the results of the probabilistic evaluation of the PMSS at MPS.2.4.2.3.1 Creation of the JPM Storm SetAs a first step in generating the JPM Storm Set, potential storm bearings were assessed toidentify combinations likely to contribute to the low-probability range of the surge-frequency relationship at MPS. As low-probability surge responses (i.e., relatively high maximum stillwater surge elevations) were anticipated for storms with approximately northerly bearings based onstorm surge sensitivities observed as part of the deterministic evaluation, the range ofconsidered storm bearings was limited to -500 to +500.Storm bearings between -500 to +500 (i.e., in 103 intervals) were combined with landfalllocations to create a set of 55 storm tracks (Figure 2.4-36).

Each potential storm track wasexpanded into a set of storms by considering ranges of forward speed, radius of maximumwinds and maximum wind speed. This discretization process resulted in 8 potential values offorward speed ranging from 15 kt to 50 kt, 9 potential values of radius of maximum windsranging from 15 nm to 55 nm, and 8 potential values of maximum wind speed ranging from 70 ktto 140 kt. A unique synthetic storm, each identified with a unique STORMID number, wascreated for each combination of parameters including storm track direction and landfall location(i.e., 8 x 9 x 8 x 11 x 5 = 31,680 synthetic storms).Joint probabilities were calculated for each synthetic storm in a manner that recovered parameter co-variability, as reflected within the 3M data set. Based on these calculations, 13,485 parameter combinations were identified as having non-zero joint probabilities (i.e., atEE 14-El 6, REV. 12-74 DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.least one record within the 3M data set falling within the upper and lower parameter bounds oneach parameter).

These parameter combinations were isolated to create the JPM Storm Set.2.4.2.3.2 Addition of Tidal Condition Given the computational efficiency of the NOAA SLOSH model (i.e., as compared to the morerobust but computationally cumbersome ADCIRC model), two bounding tidal conditions couldbe practically simulated.

Mean High Water (MHW) and Mean Low Water (MLW) at theNewport, RI NOAA CO-OPS station attenuated to the Watch Hill Point, RI subordinate stationwere selected as being representative of bounding tidal conditions at MPS.Each storm in the JPM Storm Set was split into two conditions:

one version of the stormoccurring coincidentally with the high tide (i.e., MHW) condition, and another version of thesame storm occurring coincidentally with the low tide (i.e., MLW) condition.

This processdoubled the size of the JPM Storm Set (i.e., 13,485 to 26,970 unique parameter combinations).

2.4.2.3.3 Calculation of Annual Exceedance Probabilities In order to convert joint probabilities to annual exceedance probabilities (AEPs), two additional factors were considered:

the probability associated with the simulated tidal condition and theomni-directional annual storm occurrence rate. Each joint probability was first multiplied by 0.5to represent the probability of occurrence for the associated simulation's tidal condition (i.e.,high tide or low tide). Each modified value was then multiplied by an omni-directional annualstorm occurrence rate, which considered annual storm frequency (i.e., determined from theanalysis of the WRT storm set), the approximate length of evaluated coastline and storm trackspacing.2.4.2.3.4 SLOSH Simulations A total of 26,970 SLOSH simulations were performed using input representing the reduced JPMStorm Set. In accordance with applicable guidance (e.g., NRC, 2013), storms were simulated as steady-state events (i.e., input parameters, including storm bearing, were not varied from theinitial values prior to landfall).

Time series extracted for each storm in the JPM Storm Set werereduced to peak simulated surge elevations at several locations within the pv2 basin, including the location representative of the MPS intake. These maximum simulated stillwater elevations were used to develop a preliminary stillwater surge-frequency relationship at MPS, as described below.2.4.2.3.5 Generation of an Initial Stillwater Surge-Frequency Relationship Using the SLOSH model results and the calculated AEP values, a preliminary stillwater surgefrequency relationship was calculated for the MPS intake using the standard JPM (i.e., non-OS).The calculations included the following steps which align with FEMA methodology for surgefrequency determination (FEMA, 2012). This step established the basis for reducing the JPMStorm Set to the OS Storm Set (i.e., required for implementation of JPM-OS with ADCIRC),

asdescribed below.2.4.2.3.6 Identification of the OS Storm SetAs indicated by Figure 2.4-37, experiments performed using the SLOSH model suggested thatthe very-low probability ranges of the surge-frequency relationships were well-defined (i.e.,lower than an AEP value of approximately 1 E-6) by an OS Storm Set comprised of model-simulated surges produced by storms with bearings between -500 and +500. Furthermore, theEE 14-El 6, REV. 12-75 DoMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.experiments suggested that the stillwater surge-frequency relationship at MPS represented bythe remaining parameter combinations could be accurately reproduced using 55 production runsspanning the -500 to +500 bearing range (i.e., the reference set) and 16 additional simulations toestablish sensitivities (i.e., calculate derivative terms) to various parameter perturbations (i.e.,the sensitivity set).2.4.2.3.7 Performance of ADCIRC Simulations Results of the ADCIRC simulations (i.e., OS Storm Set) were evaluated at several locations within the model mesh including the node representative of the MPS intake location.

ADCIRCsimulations were performed using a static initial condition (i.e., 0 ft NAVD88) approximately representative of a mean tide condition at MPS; whereas, SLOSH simulations for the entire JPMStorm Set were performed using two static initial conditions representative of high and low tideconditions.

This difference was addressed during final adjustments to account for uncertainty.

Simulated maximum stillwater elevations at the MPS intake location for the OS Storm Set -reference set are shown in Table 2.4-6; the corresponding OS Storm Set -sensitivity set resultsare shown in Table 2.4-7.For the storms simulated using ADCIRC, the ADCIRC wind field profiles generally comparedfavorably to the SLOSH predictions.

Where slight differences were evident (e.g., large distances from the storm centers),

the comparisons indicated conservatism in the ADCIRC representation (i.e., ADCIRC is predicting higher wind speeds compared to SLOSH). These favorable comparisons support the utility of applying SLOSH as screening-level assessment tool.2.4.2.3.8 Refinement of Stillwater Surge-Frequency Relationships Based on the ADCIRC results described above, a refined stillwater surge-frequency relationship was developed for MPS. AEP values were revised to exclude the tidal adjustment factor (i.e.,0.5) in recognition of the representation of a mean tide condition versus high and low tideconditions.

In addition to deriving the stillwater surge-frequency relationship from ADCIRCresults as opposed to SLOSH results, two other methodological modifications were made at thisstage:1. Potential forward speeds of 5 and 10 kt were considered by extrapolating the fspdsurge response at 15 kt. This was done to fully span the low range of the forwardspeed parameter given the relatively high independent probability of a slow-moving storm and the previously-identified inverse correlation between forward speed andpeak surge for some storm bearings at MPS.2. Stillwater surge elevations were estimated for every storm parameter combination without consideration of non-zero joint probability (i.e., expanded to include forwardspeeds of 5 and 10 kt). These elevation estimates were necessary in order tocharacterize aleatory variability, as discussed in the following section.The surge-frequency relationship at MPS was finalized by assessing uncertainty and error andlinearly adding projected SLR, as described in the following section2.4.2.3.9 Adiustments to the Surgqe-Frequency Relationship to Reflect Uncertainty, Error andSea Level RiseTwo forms of uncertainty were considered in this analysis:

epistemic uncertainty and aleatoryvariability.

The former form of uncertainty generally represents a "lack of understanding" of theEE 14-El 6, REV. 12-76 DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.physics within the system (i.e., measurement uncertainty, model skill, etc.); whereas, the latterform of uncertainty is attributed to sample size limitations associated with empirical data and/orthe existence of unresolved or unpredictable variations in system behavior (NRC, 2012). Errorassociated with characterizing maximum wind speeds was also considered by evaluating thedifference between EVA fits to the WRT and 3M data sets (i.e., similar to the comparison madeas part of the PMH evaluation).

The sources of significant uncertainty considered in this analysis included (note that the first twopotential sources are examples of epistemic uncertainty, the third potential source is anexample of aleatory variability and the final potential source an example of error):1. uncertainty in representing tide occurring coincidentally with surge:The effect of this source of uncertainty was quantified based on datum analysisresults at the Newport, RI NOAA CO-OPS station (NOAA, 2013b). This uncertainty accounts for tide variation from the simulated initial condition (i.e., 0 ft NAVD88).This difference was calculated as the attenuated difference between the Mean HighWater (MHW) elevation and the simulated initial condition (i.e., NAVD88 datum).2. bias or uncertainty in numerical surge and wind field models (i.e., ADCIRC):The effect of this source of uncertainty was quantified based on the results ofADCIRC verifications performed by GZA. To estimate the 95% confidence

interval, the maximum absolute error calculated based on observed and simulated peakwater levels was conservatively multiplied by a factor of 2.3. uncertainty due to samplinq (i.e., aleatory variability associated with maximum windspeed):This source of uncertainty is variable as a function of maximum wind speed. Theeffect was quantified based on random sampling (i.e., "bootstrapping")

performed using maximum wind speed values from the 3M dataset.4. error within the 3M data set (i.e., deviation from the WRT storm set) associated withmaximum wind speeds above 96 kt:This source of error is also variable as a function of maximum wind speed. Theeffect was quantified based on a comparison of EVA fits to the 3M and WRT datasets.Consistent with FEMA methodology (FEMA, 2012), this analysis considered only uncertainty which resulted in higher or more-probable surge results (i.e., added conservatism);

therefore, alluncertainty terms were considered to be positive such that they increased calculated surgeelevations The final uncertainty-adjusted stillwater relationship at MPS is shown in Figure 2.4-38. Inaddition to the uncertainty adjustments described above, this relationship also reflects anEE 14-E16, REV. 12-77 DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.additional linear adjustment to account for the 50-year SLR projection at MPS. The samerelationship, converted to the MSL vertical datum, is shown in Figure 2.4-39. Similar to thedeterministic PMSS evaluation, the probabilistic storm surge results described above areconservative in consideration of the following:

• Conservatisms associated with the deterministic PMH inputs, as previously discussed;

• Conservatism added through SLOSH assessment to establish appropriate reference and sensitivity storm sets for use with JPM-OS; and* Conservatism added through consideration of applicable uncertainty, error andSLR.2.4.3 Conclusions Deterministic Probable Maximum Storm SurgeThe ADCIRC simulation representing this combination of parameters (i.e., STORMID 818)resulted in a maximum stillwater elevation of 23.3 ft MSL (i.e., reflecting linear adjustment to theAWL) at the MPS intake location.

This elevation exceeds the existing design basis maximumstillwater PMSS elevations of 18.2 ft MSL for MPS2 and 19.7 ft MSL for MPS3; however, floodprotection at MPS2 (i.e., via flood wall containment) extends to 22.0 ft MSL, and MPS3 isprotected from storm surge by a site grade elevation of 24.0 ft MSL (Dominion 2014b andDominion 2014a).Probabilistic Storm SurqeAt an AEP level of approximately 1 E-6, the stillwater elevation at MPS is calculated to be 19.7 ftNAVD88. This elevation translates to 20.7 ft MSL. This elevation may be compared to theexisting design basis stillwater elevations of 18.2 ft MSL for MPS2 and 19.7 ft MSL for MPS3;however, it is important to note that flood protection at MPS2 (i.e., via flood wall containment) extends to 22.0 ft MSL, and MPS3 is protected from storm surge by a site grade elevation of24.0 ft MSL (Dominion 2014b and Dominion 2014a).EE 14-El 6, REv. 1 2-78EE 14-El 6, REV. 12-78 Z NW DOMINION FLOODING HAZARD REEVALUATION REPORT FORZ'ACHRY r+MILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.2.4.4 References 2.4.4-1 ANS 1992. "ANS/ANS-2.8-1992

-Determining Design Basis Flooding at PowerReactor Sites", American National Standards Institute/American Nuclear Society, 19922.4.4-2 Blake et al. 2011. "The Deadliest, Costliest, and Most Intense United States TropicalCyclones from 1851 to 2010 (and Other Frequently Requested Hurricane Facts)",Blake, E.S., Landsea, C.W. and Gibney, E.J., National Hurricane Center, NationalOceanic and Atmospheric Administration Technical Report NWS NHC-6, August 2011.2.4.4-3 Dominion 2014a. Millstone Power Station Unit-3 Final Safety Analysis Report (MPS-3FSAR), Revision 25.2.2.4.4-4 Dominion 2014b. Millstone Power Station Unit-2 Final Safety Analysis Report (MPS-2FSAR), Revision 30.2.2.4.4-5 Emanuel et al. 2004. "Environmental Control of Tropical Cyclone Intensity",

Emanuel,K., Des Autels, C., Holloway, C. and Korty, R., Journal of the Atmospheric

Sciences, Vol. 61, 843-858, April 2004.2.4.4-6 Emanuel et al. 2006. "A statistical-deterministic approach to hurricane riskassessment",

Bull. Amer. Meteor. Soc., 19, 299-314, K. Emanuel, A., S. Ravela, E.Vivant, and C. Risi, 2006.2.4.4-7 ESSA, 1970. "Joint probability of tide frequency analysis applied to Atlantic City andLong Beach Island, NJ", U.S.Department of Commerce, Environmental ScienceService Administration, Weather Bureau, Myers, V.A., April 1970.2.4.4-8 FEMA, 2012. "Operating Guidance No. 8-12, Joint Probability

-Optimal SamplingMethod for Tropical Storm Surge Frequency Analysis",

U.S. Department of HomelandSecurity, Federal Emergency Management Agency, March, 2012.2.4.4-9 NOAA 1979. "Meteorological Criteria for Standard Project Hurricane and ProbableMaximum Hurricane Wind Fields, Gulf and East Coast of the United States",

NationalOceanic and Atmospheric Administration Technical Report NWS 23, September 1979.2.4.4-10 NOAA 2012a. "SLOSH Display Program (1.65b)",

National Oceanic and Atmospheric Administration, Evaluation Branch, Meteorological Development Lab, National WeatherService, January 2012.2.4.4-11 NOAA 2012b. "SLOSH Model v3.97" National Oceanic and Atmospheric Administration, Evaluation Branch, Meteorological Development Lab, National WeatherService, January 2012.2.4.4-12 NOAA 2013a. "Revised Atlantic Hurricane Database (HURDAT 2)", National Oceanicand Atmospheric Administration, National Hurricane Center,http://www.aoml.noaa.gov/hrd/hurdat/2011

.html, Date accessed December andJanuary, 2013, Date updated June 10, 2013.2.4.4-13 NOAA 2013b. "Tides and Currents Datum Information:

Newport, RI Station 8452660",

http://tidesandcurrents.noaa.gov/datums.html?id=8452660, Date accessed September, 2014. Date updated October, 2013.2.4.4-14 NRC 2011. "NUREG / CR-7046:

Design Basis Flood Estimation for SiteCharacterization at Nuclear Power Plants",

U.S. Nuclear Regulatory Commission, November 2011.EE 14-El 6, REV. 12-79 DOMINION FLOODING HAZARD REEVALUATION REPORT FORZACHRYMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.2.4.4-15 NRC 2012. "NUREG/CR-7134

-The Estimation of Very-Low Probability Hurricane Storm Surges for Design and Licensing of Nuclear Power Plants in Coastal Areas",U.S. Nuclear Regulatory Commission, October 2012.2.4.4-16 NRC 2013. "JLD-ISG-2012-06:

Guidance for Performing a Tsunami, Surge, or SeicheHazard Assessment",

U.S. Nuclear Regulatory Commission, Revision 0, January 2013.2.4.4-17 USACE 1994. "ADCIRC:

an advanced three-dimensional circulation model for shelvescoasts and estuaries, report 2: user's manual for ADCIRC-2DDI",

Westerink, J.J., C.A.Blain, R.A. Luettich, Jr. and N.W. Scheffner, 1994, Dredging Research ProgramTechnical Report DRP-92-6, U.S. Army Engineers Waterways Experiment Station,Vicksburg, MS., 156p.2.4.4-18 WRT 2013. Synthetic Hurricane Event Set, WindRiskTech, LLC,Chesapeakenceprenanalcal.csv, Chesapeake

_ncep-reanalcal2.csv andChesapeake

_ncepreanalcal freq.csv, September, 2013.EE 14-El 6, REV. 1 2-80EE 14-El 6, REV. 12-80 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Table 2.4-1: Top 10 Extreme Water Levels.(a) Newport Station 8452660)Highest EventRank Year Date WL (feet, Typ Event NameNAVD88)1 1938 9/21/1938 11.27 H2 New England Hurricane of 19382 1954 8/31/1954 8.57 H2 Carol 19543 2012 10/29/2012 6.02 ET Sandy 20124 1991 8/19/1991 5.79 H2 Bob 19915 1944 9/14/1944 5.77 H1 Not Named6 1978 1/9/1978 5.41 N/A N/A7 1991 10/31/1991 5.07 TS Not Named8 1978 2/7/1978 5.06 N/A N/A9 1963 11/30/1963 5.06 N/A N/A10 1974 12/2/1974 5.05 N/A N/A(b) New London (Sration 8461490)RankYearDateHighestWL (feet,NAVD88)EventTypeEvent Name1 1938 9/21/1938 8.74 H2 New England Hurricane of 19382 1954 8/31/1954 7.74 H2 Carol 19543 2012 10/30/2012 6.1 ET Sandy 20124 1950 11/25/1950 5.74 N/A Not Named5 1944 9/14/1944 5.24 H1 Not Named 19446 1960 9/12/1960 5.04 H2 Donna 19607 1953 11/7/1953 4.94 N/A Not Named8 1991 10/31/1991 4.63 TS Not Named 19919 2011 8/28/2011 4.6 TS Irene 201110 1968 11/12/1968 4.54 N/A Not NamedEE 14-El 6, REV. 1 2-81EE 14-E16, REV. 12-81 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Table 2.4-1 continued:

(c) Bridgeport (Station 8467150)fHighest EventRank Year Date WL (feet, Typ Event NameNAVD88)1 2012 10/30/2012 9.2 ET Sandy 20122 2011 8/28/2011 8.2 TS Irene 20113 1992 12/11/1992 8.2 N/A Nor'Easter 19924 1991 10/31/1991 7.54 TS Not Named 19915 1980 10/25/1980 7.15 N/A Not Named6 1984 3/29/1984 6.77 N/A Not Named7 1985 9/27/1985 6.75 TS Henri 19968 1996 10/19/1996 6.69 N/A Not Named9 1968 11/12/1968 6.68 N/A Not Named10 2007 4/16/2007 6.67 N/A Nor'Easter 2007Montauk (Station 8510560)Highest EventRank Year Date WWL (feet, Typ Event NameNAVD88)1 1954 8/31/1954 6.87 ET Carol 19542 2012 10/29/2012 5.49 TS Sandy 20123 1978 2/6/1978 5.18 N/A Blizzard of '784 1991 10/31/1991 4.76 TS Not Named 19915 1950 11/25/1950 4.67 N/A Great Appalachian Storm 19506 1953 11/7/1953 4.37 N/A Not Named7 1968 11/12/1968 4.27 TS N/A8 1972 2/19/1972 4.20 N/A N/A9 1992 12/11/1992 4.17 N/A N/A10 2010 12/27/2010 4.03 N/A Not Named(d)Notes:1. H1, H2 indicate Category 1 and Category 2 Hurricane, respectively.

2. TS indicates Tropical Storm; ET indicates Extra-tropical Storm.Table 2.4-2: NOAA SLOSH MOM Water Levels at Selected Gage Locations.

SLOSH Grid CAT 1 CAT2 CAT3 CAT4NOAA CO-OP Station (No.) Cell (pv2) (feet, NAVD88)Newport (8452660) 73 -74 4.4 7.8 11.5 15.0New London (8461490) 102-28 5.6 9.9 14.2 18.3Bridgeport (8467150) 143-5 6.9 11.7 16.7 22.5Montauk (8510560) 114 -46 3.9 6.8 9.6 12.3Note: 1. MOM elevations reflect a mean tide initial condition.

EE 14-El 6, REV. 12-82 Z'ACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Table 2.4-3: Numbers of WRT storms and storm segments within the OIR, including classifications by Saffir-Simpson category.

I.018 RegionstormsstormsegmentsAll 7957 23993>= Cat 3 (96) 106 164>= Cat 4 (113) 11 16>=Ca: 5(137) 1 1Table 2.4-4: Recommended PMH-level parameters and parameter rangesRadius ofStorm Maximum Wind Forward Speed, Maximum Wind,Bearing Speed, vm fspd rmw-600 107.1 kt 16.1 -37.0 kt 10.6 -37.7 nm-500 111.9 kt 17.4- 40.4 kt 10.9- 35.8 nm-400 118.3 kt 19.2-43.9 kt 11.2- 33.3 nm-30o 125.6 kt 21.2 -44.2 kt 11.7 -30.5 nm-200 133.2 kt 23.3 -42.0 kt 12.1 -27.5 nm-100 140.1 kt 25.3 -40.0 kt 12.5-24.9 nm00 145.2 kt 26.7 -38.4 kt 12.8 -22.9 nm100 147.6 kt 27.3 -37.7 kt 12.9 -21.9 nm200 146.6 kt 27.1 -38.0 kt 12.8-22.3 nm300 141.6 kt 25.7 -39.5 kt 12.6- 24.2 nmEE 14-El 6, REv. 1 2-83EE 14-El 6, REV. 12-83 ZACH-RYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Table 2.4-5: Refinement Storm Set parameters.

Vm is maximum sustained wind speed;CPD is central pressure deficit; Rmax is radius of maximum wind.STORMIDvmn CPDNOt (mb)ForwardBearing Speed(deg. from N) (kts)Rmax Landfall Mile Post(kt) (via NWS 23)Landfall Latitude Landfall Longitude (Dec. Degrees)

(Dec. Degrees)8188238289689739789831058106311381143114811531218122312281298125.6125.6125.6133.2133.2133.2133.2140.1140.1145.2145.2145.2145.2147.6147.6147.6146.61071039911210810499112107120117112107125120116124-30-30-30-20-20-20-20-10-10000010101020202530202530352530253035402530352535353530303030252525252525252525252600260026002600260026002600260026002600260026002600260026002600260040.8840.8840,8840.8840.8840.8840.8840.8840.8840.8840.8840.8840.8840.8840.8840.8840.88-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35-72.35EE 14-El 6, REV. 1 2-84EE 14-El 6, REV. 12-84 ZCHRiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Table 2.4-6: Maximum simulated stillwater surge elevations associated with the OS Storm Set -reference set at the MPSintake location (SLOSH and ADCIRC results shown).SLOSH -MPS SLOSH -MPS ADCIRC -MPS ADCIRC -MPSSTORMID (ft NAVD88) (ft MSL) (ft NAVDOS) (ft MSL)137613771378137913804256425742584259426071367137713871397140100161001710018100191002012896128978.610.612.19.74.47.910.212.29.74.87.510.012.09.75.77.310.011.69.66.47.510.09.611.613.110.75.48.911.213.210.75.88.511.013.010.76.78.311.012.610.67.48.511.08.410.312.28.94.07.59.812.39.03.76.79.512.08.74.66.29.211.48.75.46.28.99.411.313.29.95.08.510.813.310.04.77.710.513.09.75.67.210.212.39.76.47.29.912898 11.1 12.1 10.4 11.412899 9.5 10.5 8.4 9.312900 6.9 7.9 5.7 6.715776 7.8 8.8 6.3 7.315777 10.0 11.0 8.4 9.415778 10.6 11.6 9.3 10.315779 9.1 10.1 7.9 8.915780 7.4 8.4 6.0 7.018656 8.1 9.1 6.4 7.418657 9.9 10.9 7.9 8.918658 10.0 11.0 8.2 9.218659 8.9 9.9 7.3 8.318660 7.7 8.7 5.9 6.921536 8.4 9.4 6.4 7.421537 9.5 10.5 7.1 8.121538 9.6 10.6 7.3 8.321539 8.7 9.7 6.7 7.721540 8.0 9.0 5.9 6.824416 8.6 9.6 6.1 7.124417 9.0 10.0 6.0 7.024418 9.1 10.1 6.4 7.424419 8.5 9.5 5.9 6.924420 8.1 9.1 5.6 6.527296 8.3 9.3 5.1 6.127297 8.6 9.6 5.4 6.427298 8.6 9.6 5.7 6.727299 8.2 9.2 5.2 6.227300 8.1 9.1 5.1 6.130176 8.0 9.0 4.5 5.530177 8.2 9.2 4.8 5.830178 8.1 9.1 4.8 5.830179 7.8 8.8 4.6 5.630180 7.9 8.9 4.7 5.7EE 14-E16, REV. 12-85 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Table 2.4-7: Maximum simulated stillwater surge elevations associated with the OS StormSet -sensitivity set at the MPS intake location (SLOSH and ADCIRC results shown). Notethat STORMID = 12898 is used to evaluate parameter sensitivities (i.e., results associated withSTORMID = 12898 are highlighted)

STORMID12883128881289312903129081281812858SLOSH -MPS(ft NAVD88)5.97.39.013.515.810.811.0SLOSH -MPS(ft MSL)6.98.310.014.516.811.812.0ADCIRC -MPS(ft NAVD88)3.85.67.813.115.910.910.6ADCIRC -MPS(ft MSL)4.86.68.814.116.911.911.612898 11.1 12.1 10.4 11.412938 10.9 11.9 10.0 11.012978 10.9 11.9 9.6 10.613018 10.6 11.6 9.0 10.011618 6.8 7.8 6.2 7.211938 8.5 9.5 7.8 8.812258 9.4 10.4 8.9 9.912578 10.2 11.2 9.9 10.913218 11.7 12.7 10.7 11.713538 12.5 13.5 10.9 11.9EE 14-El 6, REV. 1 2-86EE 14-El 6, REV. 12-86 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-1: Site locus and NOAA tide gage locations.

EE 14-E16, REV. 1 2-87EE 14-E16, REV. 12-87 ZACHRiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-2: Illustration of several key PMH parameters.

EE 14-El 6, REV. 1 2-88EE 14-E16, REV. 12-88 Z ACHIVYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-3: Historical hurricane tracks intersecting the study area (200 km radius fromWesthampton, New York).EE 14-E16, REV. 1 2-89EE 14-E16, REV. 12-89 Z'ACHli-YDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-4: Selected historical hurricane tracks impacting the MPS vicinity.

EE 14-E16, REV. 1 2-90EE 14-E16, REV. 12-90 Z'ACHRiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-5: MPS mile post location (NWS 23, Figure 1.1). Adapted from NOAA 1979(NOAA 1979).Figwv. I. I.-Looator map with oast4Z~ chietanoe i~nterval*

marked int nautioalottlGs and kitometa'o.

EE 14-E16, REV. 12-91 ZACH-IKYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-6: Hurricane parameter sampling regions.Analysis RegionsEE 14-El 6, REV. 1 2-92EE 14-El 6, REV. 12-92 ZACH-lRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-7: The PDF, CDF, tabulated probabilities and annual frequencies for the mxwparameter within the IR. The envelope PDFs pertain to the OR and RR.IR i~i~~ ato-to IIIiM IR, I' *uol'ai Rf3271z" 3 2 54! 42 *~ ~ ~* '2r... .... L ........

..(,1 1 I \-A I ,I'4 *0 4!'5,I0os 1iivo fl\wA (MC iliel'NYJI of j10 kt)2 7:~'-o4 *I4 '4 j -CA,'1' ~ ~ "2 ~ *4444~~ ~*~EE 14-El 6, REV. 1 2-93EE 14-El 6, REV. 12-93 ZACHRI-anY DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-8: The PDF, CDF, tabulated probabilities and annual frequencies for the fdirparameter within the IR. The envelope PDFs pertain to the OR and RR.ffda I IR 11,II I O I~.... .. I. *. .,CDF~fdoi):

IRI it414I I ,ý54 4,4Pg&4'.,bjltn of rdis 4wi ij na of 1I dcg),:i lo f i~44 I,-~~44 4,'4 I5 i ,4t,4*'.,

4~, ~ 44 45 ~ 444EE 14-El 6, REV. 1 2-94EE 14-El 6, REV. 12-94 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-9: The PDF, CDF, tabulated probabilities and annual frequencies for the fspdparameter within the IR. The envelope PDFs pertain to the OR and RR.U22'~p1 15 Ni: f15n It '0ti1~1~ h 1K~1 jhobablhity of fspd %per inlteia of 5r kIPJ 1Iz ý~l 1wEE 14-El 6, REV. 1 2-95EE 14-E16, REV. 12-95 ZEACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-10: The PDF, CDF, tabulated probabilities and annual frequencies for the cpd(central pressure deficit) parameter within the IR. The envelope PDFs pertain to the OR andRR. Only HURDAT2 data from 1979 through 2012 were used to produce these distributions.

H ~ )1 11R POPUAI lit (ORI 4I 4I. II ')4 1 N§ # I I111 miof pd~ (1tj~EE 14-El 6, REV. 1 2-96EE 14-El 6, REV. 12-96 ZACHI-IYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-11: A Probability Density Histogram (PDH) and non-parametric Probability Density Function (PDF) for maximum sustained winds (mxw) within the IR. The insetshows the Gaussian kernel function.

Maximum Wind PDF0.14 Kemcl function01.12o 0.10 W0;'0.08-~9 4 10" K11 120Il 4 1 4(o.0% Wind Speed10.020.02)0.0)0 140401 60 80 101) 120 140Maximum Sustained Wind fkt)EE 14-El 6, REV. 1 2-97EE 14-El 6, REV. 12-97 ZACHRiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-12: Hurricane parameter (i.e, mxw, fspd, fdir and dmxw) cross-correlations forthe three analytical regions (IR, OR and RR) based on HURDAT2 dataset.

Shadingindicates statistical significance at the 95% level.Cross -Corrclation (Top=IR; Mid=OR: Bottom=RR) mxwfpafdirdmxwSI. -0.02 -0.41 -0.42I. 0.1 -0.15 -0.18I. 0.09 -0.16 -0.11-0.02 1. 0.02 0.070.1 I. 0.1 0.080.09 1. 0.16 0-0.41 0.02 1. 0.16-0.15 0.1 1. 0.21-0.16 0.16 1. 0.05-0.42 0.07 0.16 I.-0.18 0.08 0.21 1.-0.11 0 0.05 1.EEE 14-El 6, REv. 1 2-98EE 14-El 6, REV. 12-98 ZACHI;YDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-13: Scatter plots of fdlr versus fspd data within the three analytical regions forthe 162-year HURDAT2 record.Scatter:

fdir v. rspd (HURDAT2.

IR sample)60)40-1)S20*S00*S-attet:

rdir v. fspd (HURDAT2; OR sample)40%4)%NI)lo,%).2II10 * * * * ~,0~0100)FkAnnr 10=%mth, Pc"Alam)Scatter:

fdir v. fspd (HURDAT2:

RR sample)60:0.2S20IO)VSSb. -LS-194) -I100~ -to4 0 V4) 100) 1['EE 14-El 6, REV. 1 2-99EE 14-El 6, REV. 12-99 SACHI-IYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zuchry Nuclear Engineering, Inc.Figure 2.4-14: Scatter plots of fdir versus mxw data within the three analytical regions forthe 162-year HURDAT2 record.Scattcr:

fiir v. mxw (HURDAT2:

IR simple) Sc.attcr:

Mdir v. mxw (HURDAT2:

OR sdmplc)"et4AI 40I 2(1"'IWI402(1II* 9~ ** 99 9999 9 99 9999.9* 9 99 9.9 9 9 9.mm 0*9 ~9 90.9 ~. -..-I I40 00~WI 90 99it)

• 0 0 a0~:*.. ~I L 4~O4A0~-5(1 I) 9) IMKI Y)Scamter:

fifir v. mxw (HURIDAT2 RRK simple)4AI4'010400 a *ini 9 09 0 a"-~I-- tE) -1+4) I) (W4 11%) OilEE 14-E16, REV. 1 2-100EE 14-E16, REV.12-100 ZACHKYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-15: Scatter plots of mxw versus fspd data within the three analytical regionsfor the 162-year HURDAT2 record.Sia~tcr:

mxw v. fpsd (HURDAT2:

IR sample) Scattcc:

mxw v. fpsd (HURDAT2:

OR sample)0) 60)2'A )fl-3~20I000S.9 9* 99 99* * *!:u' :N.1A~inxmu Windgo) 90 1900 :) 4 v) 0 2 4402* .IME*.2d.'A-3'4'3 20 405 #0 W4) 100 1 2N) 140)MStamamm Winid (M )EE 14-El 6, REV. 1 2-10 1EE 14-El 6, REV.12-101 ZIACHKYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-16: Scatter plots of fdir versus dmxw data within the three analytical regionsfor the 162-year HURDAT2 record.

Wdir v. dmxw (HURDAT2:

IR sample)Y~)Scautter:

r4ir %,. dmxw (HURDAT2:

OR saImple)£3.IA2N0-,oto.0-0 0*-toa99-20' ~ *~0* 40%0-** a"401.0 0starinr (o=.Nt4nh:

Po4.aIIIS IkM~nn t 0)= Mvih. Pot z EAi0Scatter:

rdir v. dmxw (HURDAT2.:

RR samnple)I.IAato-o r%, r 41.40 -100 -504 0 Y) 100I 1 V)EE 14-El 6, REV. 1 2-102EE 14-El 6, REV.12-102 ZACHIRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-17: Scatter plots of mxw versus dmxw data within the three analytical regionsfor the 162-year HURDAT2 record.Scatter:

mxw v. dmxw (liURDAT2:

IR simple)V)Scatter:

mxw v. dmxw (IIURDAT2:

OR sample)%()E 2'"100-2O* * *CI.'44AIS2100 O0 1 2( 40 ky) go 100 1120Mtaimum Wind (kll* *I, 20 401 kr ) SOn 100,Mba~mUI?

Wind (koIScatter:

mxw ,. dmxw (IIURDAT2:

RR sample)A1010-10)I* -'g~.-~-

• • . Io) 20 ,4) 40 ) 60 8 1(m) 1210 !4)4&.inmum WiJ (kAl)EE 14-El 6, REV. 1 2-103EE 14-El16, REV.12-103 ZAClHIYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-18: Scatter plots of fspd versus dmxw data within the three analytical regionsfor the 162-year HURDAT2 record.Scatucr:

fspd v. dmxw (HURDAT2:

IR samplc) S.atter:

fslN v. dmxw (HURDAT2:

OR sampIc)i40141030*-~ .0CI.44A2100310:.:::. .* * * *** ..SM) 20 V4uztwwart Sipecd i40 MI 00I 10 1-1) %()Wi NIVA 1*d SPCVJ AS IScattecr:

fspd v. drnxw I HURDAT2:

OR saimple)40 f4)A-10010 240) I0 24) U)41 U)EE 14-El 6, REV. 1 2-104EE 14-El 6, REV.12-104 ZACHFIYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-19: Comparisons between HURDAT2 distributions of storm bearing (fdir),shown by multiple lines, and the WRT population

estimate, shown by gray linesurrounded by central 98% uncertainty bounds.PDF(fdir):

IR (HURDAT v WRT)0 I0OW )%-ISO90 -100 -.94) t)o' 100 ISO)PDFrfdir):

OR (HURDAT v WRT)4l I,7A0 40.00%-1.94) -RX) --94) 0 .110 im 19)tkarinr 1()=Ntvth:

No-= "-soF 10PIDIF(fdir).

RR (HURIDAT v WRT)1,94 -1411 -S 1 ) 9) 1(N) 114Beasnng (0-=Nofth:

N%- "-soI0.00EE 14-El 6, REV. 1 2-105EE 14-El 6, REV.12-105 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-20: As in Figure 2.4-19, except pertaining to the storms' translation speed(fspd).PDF(f6pl).

OR (HURDAT v WRT)0.I1O.120i060)04002000,II M01) N11 NolFtw~ar Spvcd tkoPDF(fspd).

RR (HURIDATv WRT)0.14W02-~0.04)00)0 40JA i UNoNvid pedftiEE 14-El 6, REV. 1 2-106EE 14-E16, REV.12-106 ZACHRIYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-21: As in Figure 2.4-19 except pertaining to the storms' change in intensity, asindicated by the 6-hourly change in 1-min maximum sustained winds (dmxw).PDF(dmxw):

IR (HURDAT v WRT)0.10001)PDF(dmxw):

OR (HURDAT v WRT)7000~-OW) -4' --If) If) 44(Cbjnte in Maik sumun %uslAnud Waginii )kti- htWPDF(dmxw).

RR (HURDAT v WRT)0.1F ,10~0.00.00N~i) 2% N 20) 41Clujnrc in ',Ijgmum SuiamnwJ Wmn&i okt*-ho-60 2N 0 2) 4))Clnein Maihi~mum Suttarnw Win&i ikc*-hroEE 14-El 6, REV. 1 2-107EE 14-El 6, REV.12-107 Z'ACHWI-Y DOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-22: As In Figure 2.4-19 except pertaining to the 1-minute maximum sustained winds. (as Indicated by vm in the WRT data set). The right panels show magnifications ofCategory 1 at higher hurricanes (upper) and Category 3 and higher hurricanes (lower).ifPDF(vm).

IR (HURDAT v WRT) PDF(vm).

IR (HURDAT v WRT)0,1 0140.10'0.1)60070A 00.020,00% 04:0 4in 0) sol 10() 320 3140Mi~imum Srtowned WinJ Ai I110 IIy) I N 140M~A'XMUM slouit&ned Wind AkllPDFfvm):

OR (HURDAT v WRT) PDF(vm):

IR (HURDAT v WRT)015 (M u))00(2%C'001'00% 4. (1(110(111%0(1 fflo OOMl%03 0 00 13(0 120) 1U) 140 l4Mfa~imum Nu%1mncd Wind kik,MA% $mum NIuLtWncd WId (Kl IPDF(vm):.

RR (HURDAT v WRT)000%1101I %oNI~mmum !Sustwned Wind (ILI 1EE 14-El 6, REV. 1 2-108EE 14-E16, REV.12-108 ZACHRiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-23: MPS Circular Region (IR), Offshore Portion and Storm Segment DataStorm segment locations within the OIREE 14-El 6, REV. 1 2-109EE 14-E16, REV.12-109 ZACHKYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-24: Annual Frequency of Synthetic Storms by Year and 31-year Averages31 years (190- 2010) of Annual Synthetic Storm Frequencies 0.50.40.3__- ianuAllfreq4)257775 1OIR Aamwi rq42~IV..0.10.0................................0510 13 20 2.1 30EE 14-El 6, REV. 1 2-110EE 14-E16, REV.12-110 ZAC-HYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-25: The PDF, CDF, tabulated probabilities and annual frequencies for the vmparameter within the OIR based on the WRT dataset.WRT: PDF(y:w):

-ORnv 3927710 fWRT, Ct*(Nin I, 01KI I0I "I ý).I .... .. ..hott ~Itfili tv o dfVil ((,,: 1w 1% f I lbkroPC M 1;K *1An. f If', t K' A FýjEE 14-El 6, REV. 1 2-111EE 14-El 6, REV.12-111 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-26: The PDF, CDF, tabulated probabilities and annual frequencies for the fdirparameter within the OIR based on the WRT dataset.WRT:-PPF(tWr):

0IRAnz 23993N\RT CDF-ttdii 01RI #fI 1 14 0 ";'0412....... .... -An 1, 1 "tEE 14-El 6, REV. 1 2-112EE 14-E16, REV.12-112 ZACHRYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-27: The PDF, CDF, tabulated probabilities and annual frequencies for the fspdparameter within the OIR based on the WRT dataset.NVRT. CDfli1%Wi OIRF.~III of ~I Pd kpo1 illicU'4I o 5 Lulit 14t4 III!,, oI 1f1* it 91"'1EE 14-El 6, REV. 1 2-113EE 14-El 6, REV.12-113 DOMINION FLOODING HAZARD REEVALUATION REPORT FORC H RY MILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-28: The PDF, CDF, tabulated probabilities and annual frequencies for the rmwparameter within the OIR based on the WRT dataset.WRT:-PDFuOW):-OIR nn, 3994 , ' ..... ...WRTAPmx CD i1v0'II"4R ~'t a V. tNO'kalilli of I ..ci uiiciviiI of 5 InihEE 14-El 6, REV. 1 2-114EE 14-E16, REV.12-114 ZACHRiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-29: Storm Intensity (vm) extracted from the 3M data set as a function of stormbearing based on the calculated PMH intensity data set rank. The regression linerepresents a fifth-order polynomial function developed from a least-squares fit to the interval-specific intensity values.17.0r150.0140.013ALI120.0110.0D Data Set Rank = 19thXNWS23Maximumn

---I000S009A05 --------------------------

---

---

..-I.60I -4.4,-40 .0 -W -1.*w Sea (÷/. s deg',s)10 20 soEE 14-El 6, REV. 1 2-115EE 14-E16, REV.12-115

'ACHWIYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-30: Scatter plot pairs representing WRT (left panels) and 3M (right panels)data. PMH parameter bounds for storm forward speed (fspd) and radius of maximum winds(rmw) are depicted in red. The limits depicted in the right panels are identical to the limitsdepicted in the left panels.CScatter:

fdir v. fspd (WRT: OIR sample)70-605o4O-150 -100 -50 0 50 100 ISOIIScatter:

(dir v. tipd (Synah5 OIR sample)70,605040to21010-150 -100-i0 0IO 100 150Bearins:

(0=Nwrth:

Pos=E~as)

Scannj (0-N-6h. 9.Scatter:

vm v. rmw (WRT: OIR sample)Scatter vm V. row (Synih5:

OIR sample)8D,~60C- 4010~*4o210oso to0 ISO)M'Aimom Wftu 1*X#'$rAA1 .a W: 11;,4170'604i0~30~2010toStaatter vm v. rpsi (WIKTý (hR sample)1.402010EE 14-E16, REV. 1 2-116EE 14-E16, REV.12-116 "WASA M1y DOMINION FLOODING HAZARD REEVALUATION REPORT FORzACHKY 1MILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-31: Simulated storm tracks- bearings ranging from -600 to +30°Legend* NWS 23 Milepost* MPS LocationBearing = 30 Deg.Bearing = 20 Dog.Bearing = 10 Dog.Bearing = 0 Deg.Bearing = -10 Deg.Bearing = -20 Dog.-Bearing = -30 Dog.Bearing = -40 Deg.Bearing = -50 Deg.Bearing = -80 Dog.EE 14-El 6, REV. 1 2-117EE 14-E16, REV. I2-117I ZACHAYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-32: SLOSH pv2 model basin -MPS vicinity.

EE 14-El 6, REV. 1 2-118EE 14-E16, REV.12-118 Z'ACHKYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-33: SLOSH pv2 model basin -MPS region. Cell Identifications (i.e., I,J) shownfor proximal NOAA tidal gaging and subordinate stations.

EI~tahnd HARartI rd'Neat Hurtford

?irIJLEE 14-El 6, REV. 1 2-119EE 14-E16, REV.12-119 Z'ACJIiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-34: Screening results -SLOSH-simulated stillwater elevation as a function ofstorm bearing and forward speed.EE 14-E16, REV. 1 2-120EE 14-E16, REV. 12-120M ZA HEIYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-35: Screening results -SLOSH-simulated stillwater elevation as a function ofstorm bearing and radius to maximum winds.EE 14-El 6, REV. 1 2-121EE 14-El 6, REV.12-121 ZACHTiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-36: Simulated storm tracks in the MPS vicinity-bearings ranging from -500 to+50W.EE 14-E16, REV.12-122 ZA MHI7YDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-37: Comparison of stillwater surge-frequency relationships at MPS developed using the JPM and JPM-OS based on SLOSH results.I a ISLOSH JPM-SLOSH JPM-OSI£IIIIA-15-10MPSNote: assumed P(tlde) = 0.51,E-031.E-041.E-O5Event Probability 1.E-061.E-7EE 14-E16, REV. 1 2-123EE 14-E16, REV. 12-123I ZACHRiYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-38: Refined stillwater surge-frequency relationship at MPS calculated usingJPM-OS and ADCIRC Including adjustments for uncertainty, error and SLR. The intialJPM / SLOSH and JPM-OS / ADCIRC (no uncertainty or SLR adjustments) relationships areprovided for reference.

The stillwater surge elevation at an AEP of 1.OE-6 is approximately 19.7ft NAVD88 with uncertainty, error and SLR considered.

-30SLOSH JPM..... ADCIRC JPM-OS-25-ADCIRC JPM-OS + Uncertainty

+ SLRI201510.. ... ..MPSIUI-Ii1.E-031.E-0411E-05Event Probability 1.E-061.E-07EE 14-El 6, REV. 1 2-124EE 14-E16, REV.12-124 ZACHRIYDOMINION FLOODING HAZARD REEVALUATION REPORT FORMILLSTONE POWER STATION UNITS 2 AND 3Zachry Nuclear Engineering, Inc.Figure 2.4-39: Refined stillwater surge-frequency relationship (converted to MSL verticaldatum) at MPS calculated using JPM-OS and ADCIRC including adjustments foruncertainty, error and SLR. The stillwater surge elevation at an AEP of 1.OE-6 isapproximately 20.7 ft MSL with uncertainty, error and SLR considered.

I r I25 -ADCIRC JPM-OS + Uncertainty

+ SLRI*1II-is-105MIPSLH1.E-031.E-041.E-O5Event Probability 1.E-061.E-07EE 14-El 6, REV. 1 2-125EE 14-El 6, REV. I2-125