ML22182A405

All Creatures Great & Small: a Brief Survey of the Impact of Flora/Fauna on Nuclear Power Plants

ML22182A405
Person / Time
Issue date:	07/01/2022
From:	John Hanna NRC/RGN-III
To:
References
Download: ML22182A405 (6)

Text

All Creatures Great & Small (1): A Brief Survey of the Impact of Flora/Fauna on Nuclear Power Plants

John David Hanna Region III Office, US Nuclear Regulatory Commission, USA. E-mail: john.hanna@nrc.gov

The US Nuclear Regulatory Commission licenses and regulates the nations civilian use of radioactive materials to provide reasonable assurance of adequate protection of public health an d safety, promote the common defense and security, and protect the environment. The impacts of nuclear power plants on the environ ment and specifically on neighboring flora and fauna are considered in the design and licensing processes for these faci lities. Some of these impacts have been analyzed in scientific articles, (e.g., service water cooling systems affecting fish populations, seaweed, etc.) But the vector/threat also goes in the opposi te direction and the environment can pose a threat to the safety o f nuclear power plants. Flora and fauna have caused a number o f safety significant events and/or conditions at these facilities. This paper surveys the wide variety of biological challenges and describes, where possible, the r isk significance of those event s and/or conditions. The current state-of-the-art of probabili stic risk assessment modeling is briefly described and potential modeling improvements are broached. Potential operational and design enhancements that may mitigate these risk impacts---which are described in other scientific papers---are referenced.

Keywords: nuclear power, external event, PRA, flora, fauna, environment.

1. Introduction that the insights presented in this paper may be useful As analysts, regulators, and operators of nuclear power outside of the US.

plants (NPPs) we need to be sensitive to the impact of The reason we as analysts, regulators, and operators our facilities on the environment. That focus is written should consider these vectors/threats is because they are:

i nto t h e m i s si on s t a t em en t of th e U S N u c l ea r R eg ul at o r y 1) creating actual events and demanding front-line safety Commission (NRC) and is a goal the agency takes very systems, and hence 2) challenging the stability, seriously. (2) However, we also need to be similarly reliability, and sustainability of the electric grid. Given concerned when the direction of the arrow is reversed, the need for a resilient, reliable, and stable electrical grid namely when flora and fauna are the threat vector to the and the potential for increasing frequency/severity of safety of NPPs. These environmental threats have been natural phenomenon we need to understand and manage evaluated through the licensing process and have been these risks.

considered in various scientific articles, however this 2. Survey of Biological Impacts on NPPs paper will attempt to advance additional risk-informed perspectives. Biological impacts, whether from flora or fauna, are a The NRC has evaluated the impact of biota through subset of all potential environmental hazards that could safety-system functional inspections, analyzed the affect NPPs. The greater population of environmental generic safety implications, (3) required our licensees to events includes tornadoes/high winds, extreme take actions or evaluate specific failure mechanisms, heat/cold, external flooding, and seismic threats. Types e.g., via bulletins or generic letters. (4) The NRC has of vectors/threats that are environmental and arguably engaged the industry on biological impacts in the past, within the definition of biota, but are not characterized but these interactions heretofore typically were: in this paper include:

• confined to reviews of a specific system or

• Sedimentation deposition (silt, mud, and clay) in biological threat, cooling systems,

• evaluated the risk of an individual event, outside of

• Non-living biological fouling (e.g., detritus such as a greater context, and twigs, leaves deposited in cooling water ultimate

• rarely evaluated the state-of-the-art of probabilistic heat-sinks, or larger branches, trees, etc. thrown by modeling the biological/environmental impact. tornadoes or high wind events), and Disclaimer: the events/conditions described in this

• Microbiologically induced corrosion. While this paper occurred at NPPs in the United States, and hence last vector is a living one-celled organism that has may not be representative of all the different potential posed challenges to NPPs, it is not included impacts from biological sources worldwide. However, because the threat is a slowly evolving condition, given the size of the US, the number of operating as opposed to an event, and has been effectively nuclear units (93 as of the time this paper was mitigated by the nuclear industry. (5) submitted), and the diversity in wildlife and the design/manufacture of NPPs in the US, it is believed

2 John David Hanna

Table 1. Sample of events that have occurred at NPPs in the US, including the flora/fauna that created the event, a (very) brief description of the impact and the calculated risk.

Site & Unit Event Date Flora/Faunaa Impact Risk Affected (CCDP)b Oyster Creek 15 April Sea lettuce Sea lettuce caused decreasing levels in the intake Minimal Nuclear 1981 (Genus = ulva) structure, and one loop of containment spray was change Generating declared inoperable. above Station (7) baseline

Brunswick 19 April American oysters Loss of shutdown cooling due oyster shell buildup 7E-3 Steam Electric 1981 (Crassostrea in the residual heat removal heat exchanger.

Plant, Unit 1 (8) virginica)

Oyster Creek 8 June 1981 Sea grass (Order Two emergency service water pumps inoperable Minimal Nuclear = Alismatales) and potentially non-function; possible common change Generating cause failure (CCF) for other pumps. above Station (9) baseline

Salem Nuclear 11 August American oysters Various single systems and component adversely 1.2E-4 Generating 1983 (Crassostrea affected (8 events in 1980s). The August 1983 Station, Unit 1 virginica) event was a loss of condenser heat sink with a (10) consequential loss of offsite power (LOOP) &

failure of an auxiliary feedwater pump.

Pilgrim Nuclear 28 August Blue mussels Bivalves at heat exchangers created blockage & 1.3E-4 Power Station 1981 (Mytilus edulis) high differential pressure resulting in flow bypass.

(11) Potential loss of service water at system level.

La Crosse 16 July 1984 Mayflies (Order = LOOP, failure of an emergency diesel generator 9E-4 Boiling Water Ephemeroptera) (EDG) breaker to close & unavailability of both Reactor (12) trains of high-pressure core spray.

Catawba 9 March Asiatic clams Asiatic clams caused CCF failure to two of four 3E-4 Nuclear Station, 1988 (Corbicula auxiliary feedwater flow control valves and Unit 2 (13) fluminea) subsequent reactor trip.

Zion Nuclear 7 March Zebra mussels With plant in hot shutdown, an auxiliary feedwater 2.3E-5 Power Station, 1994 (Dreissena pump failed with an accompanying EDG trip (heat Unit 2 (14) polymorpha) exchangers blocked by zebra mussels).

Salem Nuclear 7 April 1994 Marsh grass Automatic reactor trip and loss of condenser heat 2.6E-6 Generating (Spartina sink, multiple safety injections, multiple cycles of Station, Unit 1 alterniflora) primary power operated reliefs with damage and an (15) alert declaration.

Wolf Creek 4 September Squirrel (Genus = Automatic reactor trip with a fire in a unit auxiliary 7.2E-5 Generating 2000 Sciurus) power transformer.

Station (16)

Donald C. Cook 29 August Zebra mussels Bivalves overwhelmed the service water (SW) 1E-5 Nuclear Plant, 2001 (Dreissena strainers and led to the inoperability of all four Units 1 & 2 (17) polymorpha) EDGs, auxiliary feedwater room coolers, and a component cooling water (CCW) train in each unit.

Donald C. Cook 24 April Alewife fish Manual reactor trip of both units, all EDGs 4E-4 Nuclear Plant, 2003 (Alosa inoperable (but ultimately determined to be Units 1 & 2 (18) pseudoharengus) functional), multiple heat exchangers and traveling screens damaged resulting in an alert declaration.

a The common and scientific name of the flora/fauna is provided. When the exact genus and species is uncertain, the known order, suborder or family is shown.

b For the calculated risk, when th ere was a multi-unit impact, the higher of the calculated risk values is shown. Per the guida nce for evaluating the risk significance of a given event, the cond itional core damage probability (CCDP) is shown. (6) When the risk values are not already available from a historical source (e.g., the Accident Sequence Precursor analysis), the CCDP values we re calculated based on the description in the event report.

All Creatures Great & Small: A Brief Survey of the Impact of Flora/Fauna on Nuclear Power Plants 3

Table 1. (Continued)

Palo Verde 14 June Scientific name = Bird caused a ground fault on a transmission line 4E-5 Nuclear 2004 Aves (animal and failure in protective relaying which then caused Generating class) excreta a three-unit NPP trip with six additional generation Station Units 1, units lost; several components failed on Unit-2; 2 & 3 (19) short duration LOOP but extensive regional impact up to Canada.

Edwin I. Hatch 1 August Snake (Suborder Snake ascended power lines, shorted and required a Minimal Nuclear Plant, 2007 = Serpentes) power reduction due to loss of cooling towers; fire change Unit 2 (20) caused an unusual event declaration. above baseline St. Lucie Plant, 22 August Jellyfish (Class = Manual reactor trip of Unit 1 and power reduction 8E-7 Units 1 & 2 (21) 2011 scyphozoa) on Unit 2 due to jellyfish & associated fish kill causing loss of condenser backpressure.

R.E. Ginna 3 June 2012 Raccoon Partial LOOP, loss of two safety-related electrical Minimal Nuclear Power (Procyon lotor) buses and automatic EDG start; unit stayed online. change Plant (22) above baseline

Surry Power 29 Brown Pelican Bird contact with power lines resulted in a partial Minimal Station, Unit 2 December (Pelecanus LOOP, loss of an electrical bus & auto-start of an change (23) 2012 occidentalis) EDG; unit stayed online. above baseline

Fermi, Unit 2 1 July 2020 Mayflies (Order = During shutdown conditions a partial LOOP 3.3E-7 (24) Ephemeroptera) occurred with a valid actuation of an EDG.

Browns Ferry 20 July 2020 Eel grass (Zostera Units 1 and 2 manually scrammed, with Unit 3 3E-6 Nuclear Plant, marina) down-powered due to loss of condenser vacuum.

Units 1, 2 & 3 (25)

The events from Table 1 above are merely a sample result of climate change, exacerbating the frequency of actual events that have occurred in the US nuclear and/or intensity of these events and, hence affecting a industry. A study covering the time period from 1980- sustainable electric grid. (27) 1987 identified 980 operational events involving service 3. Risk Modeling of Biological Impacts water, of which 276 were determined to have generic applicability. Of those 276 events twenty-eight involved 3.1. Description of the Current State of Risk biofouling, and while the number of these events has Modeling decreased significantly over the following three decades, The NRCs Standard Plant Analysis Risk (SPAR) they are still occurring with some periodicity (26). models are plant-specific probabilistic risk assessments However, despite the limited sample size, the (PRAs) maintained, frequently exercised by analysts following insights can be drawn from Table 1. within the agency, and are used to inform regulatory (i) A wide variety of plant designs in various locales decisions. The NRCs 1995 PRA policy statement have been susceptible to the impact of flora/fauna. specified that PRA evaluations supporting regulatory (ii) The diversity of biological impacts on NPPs (and decisions should be as realistic as practicable. (28) in all likelihood non-nuclear sources of electrical Consistent with this realism principle, the impacts of biota can be further developed in the SPAR models and power, such as gas and coal plants) is large, thus captured in subsequent risk assessments.

ranging from single cell organisms which foul a PRAs, including the SPAR models already account heat exchanger to medium/large birds, reptiles, or for some biological impacts implicitly via IE frequencies mammals which challenge electrical systems and and component failure probabilities. For example, the cause reactor trips and LOOPs. IEs for partial or complete LOOPs, losses of service (iii) The risk impact of the flora/fauna can range from water, and losses of CCW are included in the calculation of initiating event frequencies used in PRA models negligible to substantially high (i.e., two to three regardless if those events were caused by say a jellyfish, orders distant from core damage). or due to a low water level in the ultimate heat sink.

As will be discussed in the next section of this paper, However, this data-driven approach does have the risk impacts are larger when an initiating event (IE) limitations in that it requires actual events (or near-and degradation/loss of mitigating system(s) occur events) to occur before being factored into a PRA.

simultaneously. Additionally, it is important to note that Probabilistic risk assessments also explicitly address the magnitude of these events may be increasing as a biological impacts through the support system initiating 4 John David Hanna

eventc (SSIE) modeling as described in the industry Annual Expected Chance of recommendations. (29) The guidance states: Frequency of Number of observing one Initiator Occurrences in or more events Any historical loss of support systems that (Per reactor 2000 reactor in 2000 reactor resulted in a plant trip or significant transient year) critical years years should be considered as a potential support system 1E-4 0.2 18%

initiating event. As an example, consider the 5E-4 1 63%

potential for seasonal influx of materials causing 1E-3 2 86%

blockage of the plants service water intake 5E-3 10 99.9%

structure. This has been a historical occurrence at 1E-2 20 99.9%

several plants and can be caused by debris from

• Seasonal Aspects - Biological events typically occur flooding or storms, or environmental sources such at a specific time of year, especially when as mussels, marsh grass, or frazzle ice. flora/fauna populations fluctuate with growth and/or The SPAR models typically include loss of service breeding.

water, loss of component cooling water or other support

• Plant (or Location) Specific - As described in cooling water system (e.g., reactor building cooling reference 32: Degraded water quality is plant-water system) via the IE fault trees. (30) The SSIE specific. Reports of system failure due to degraded modeling explicitly describes the support system design water quality are included in the generic data bases aspects of the plant (e.g., number of pumps, strainers, but are commonly averaged in with other nominal etc.) while incorporating actual plant-specific events that operating data. Service water degradation has been have occurred. The goal of this modeling is to more shown to occur due to ice, seaweed, sea grass, and accurately reflect the potential impact of environmental fish runs. Because these phenomena are plant-effects (of which biologics are a subset) than could be specific, they should be quantified on a plant-achieved with a single probability distribution specific basis. Because of their short duration and representing the IE frequency. severe effects, they should be modeled with separate common cause factors and split fractions for yearly exposure times.

3.2. Characteristics of Biota Threats

• Potential Effects from Climate Change - Variations There are several aspects to the flora/fauna threat to in biological blooms driven by random consider when developing and maintaining PRAs. fluctuations or due to climate change can

• Design Aspects - Characteristics such as overwhelm the defenses put in place at NPPs which independence, physical separation (e.g., closed had/have historically been effective, (e.g., the mayfly event at Fermi listed in Table 1).

cooling water loops) and defense-in-depth (e.g., 3.3. Areas for Possible PRA Development number of offsite power lines or service water pumps) can significantly affect the PRA modelling. (i) If the PRA and the larger scientific community

• Initiating Event Frequencies - The frequencies of believes that there has been a potential shift in service water system failures and degradations as external hazards on biota (e.g., due to climate observed in the operating experience, are relatively impacts) then the PRA may not be as static as high: 1.2E-2 per reactor year for system failure and previously thought, and events not seen in some 4.1E-1 per reactor year for system degradation. (31) locations may become more common. Hence, Also, as can be seen in Table 2 below, an SSIE with prior experience may not predict future hazards, an expected frequency of 5E-3 per year is likely to which could be an area worthy of investigation.

have occurred in the US nuclear industry, even if the (ii) In general practice, the SSIEs included in PRAs entire plant population is not susceptible to the have been limited to those types of events and SSIE. systems described above, and this modeling has a technical basis due to relying on actual events or near misses. However, there are other potential impacts of biological (or more broadly environmental) vectors. These might include losses of instrument air or ventilation systems, (e.g., EDG room cooling or main control room Table 2: Expected Occurrence Rate and Probability of habitability).

Observation of SSIEs (iii) The potential failure of service water intake structures would likely result in the complete failure of the service water system and potentially lead to loss of the ultimate heat sink, but the frequency of failures may be affected more by

c SSIEs are defined as: Any ev ent such as a component,

• Leads to a reactor trip, and also train, or complete system failure (or causing the failure of a

• Fails a train or complete front-line system normally component, train, or system) that: available to respond to the reactor trip or reactor shutdown

• Challenges a reactor safety function, then and successfully mitigate the loss of the critical safety function.

All Creatures Great & Small: A Brief Survey of the Impact of Flora/Fauna on Nuclear Power Plants 5

environmental conditions such as detritus or important breakers, transformers, frazzle ice than by hardware failures. The disconnects, etc. such that the biota, e.g.,

collection of data (i.e., separating out the insects, will not cause an event. (40) biological impacts and considering them apart o Routine inspection and cleaning, especially from other CCFs) is an open issue for for animal excreta.

consideration in future research. o Ensuring electrical equipment is in fact well (iv) Additionally, modeling the impact of biota in a sealed, i.e., maintenance checks, to ensure seasonal manner in that IE frequencies are not small animals cannot get within.

static throughout the year, could improve licensee o Noise or visual cues to repel birds or decision-making on scheduling maintenance mammals from important equipment e.g.,

activities. sounds of predatory birds to scare smaller (v) Development of Dynamic PRAs may better birds, and prevent them from capture biological impacts through explicitly nesting/congregating.

modeling timing dependencies. Flora/fauna 5. Conclusions events happen quickly, and the recovery probabilities change with time, and thus are the A wide variety of plant designs in numerous locales types of events that Dynamic PRAs capture more have been susceptible to the impacts posed by various effectively than traditional PRAs. (33) types of flora/fauna. The diversity of biological It is in these interstitial spaces where one finds rich impacts on NPPs (and in all likelihood non-nuclear terrain for PRA model development. sources of electrical power) is large, ranging from single cell organisms which foul a heat exchanger to

4. Mitigation Processes or Techniques large birds, reptiles, or mammals which challenge e l e c t r i c a l s y s t e m s a n d c a u s e r e a c t o r t r i p s a n d L O O P s.

The following is a summary list of engineering, The risk impact of the flora/fauna have ranged from maintenance and operational practices that can help negligible to substantially high (i.e., three orders mitigate the threat from flora and fauna: distant from core damage.) And in terms of PRA modeling, the nuclear risk community has tackled

• For various CCW and SW systems: issues of CCF, fire modeling, incorporation of o Well-designed service water strainers (e.g., mitigating strategies equipment (commonly known duplex, triplex, self-cleaning), traveling as FLEX) and others. Though progress has been screens, trash racks, interception nets in front made in modeling biological (and the larger of intake structure. (34) environmental) impacts at NPPs, there is room for o Thermal backwashing for bivalves. (35) improvement. Are environmental/biological threats o Periodic checking of flow rates and/or flow the next horizon for PRA?

balancing. (36) The industry needs to recognize the near misses o Periodic flushing of system heat exchangers. and remember the lessons-learned from past events o Visual examinations and routine testing of where biological (or other environmental) threats to heat exchangers for heat transfer capabilities. the NPPs posed a challenge to a resilient, reliable, o Chlorination of systems, timed to coincide stable, and sustainable electrical grid.

with when systems are laid up.

o Routine maintenance (e.g., upkeep for References traveling screens) to prepare systems prior to an event. (37) 1. Alexander, Cecil Frances. 1848. Lyrics to the hymn, All o Fish sonar or other detection methods. (38) Things Bright and Beautiful o Seasonal inspections of service 2. NUREG-1350, 2021-2022 Information Digest, water/circulating water intake bays. Volume 33, October 2021 o Operational responses during an actual event: 3. NUREG-1275, Volume 3 Operating Experience Staggering trains of traveling screens or Feedback Report - Service Water System Failures and SW systems to improve survivability for Degradations, November 1988 the plant for a longer period. 4. US NRC Generic Letter 89-13, Service Water System Considering use of Mitigating Strategies Problems Affecting Safety-Related Equipment, equipment (commonly known as FLEX) Accession number 9003300128 during an event; this equipment was 5. US NRC Information Notice No. 85-30, designed, licensed & built for extended Microbiologically Induced Corrosion of Containment loss-of-AC-power coincident with a loss Service Water System, Accession number C126416

6. US NRC Management Directive 8.3, NRC Incident of the ultimate heat sink. Investigation Program, ML031250592 Carefully consider the cross train/unit 7. US NRC Licensee Event Report for Oyster Creek operation when there is biofouling Nuclear Generating Station 05000219/1981-017, Sea present due to the loss of defense-in-Grass Impact on Intake Structure, dated 14 May 1981, depth & potential common cause Accession number 8105220287 impacts created. (39) 8. Final Accident Sequence Precursor Analysis, Damaged

• For electrical buses, transformers, offsite power RHR Heat Exchangers at Brunswick 1, 19 May 1981, lines, and other electrical equipment: ML20147A359

9. US NRC Licensee Event Report for Oyster Creek o Using systems to draw biologics away from Nuclear Generating Station 05000219/1981-024, A important equipment e.g., lighting away from Slightly Clogged Pump Suction Bells, and Inadequate 6 John David Hanna

Administrative Controls, 8 July 1981, Accession 29. D. Hance and K. Canavan, 2008, Technical Report-number 8108060271 1016741, Support System Initiating Events, Electric

10. Final Accident Sequence Precursor Analysis for Salem Power Research Institute, Palo Alto.

Nuclear Generating Station, Loss of Offsite Power with 30. Risk Assessment of Operational Events Handbook, Auxiliary Feedwater Pump 13 Failed, 11 August 1983, Volume 1, Internal Events, Revision 2.01, Section 11, ML20156A208 Support Systems Initiating Events

11. US NRC Licensee Event Report for Pilgrim Nuclear 31. See reference 29 above.

Power Station 05000293/1981-049, RBCCW Heat 32. See reference 29 above.

Exchanger, 8 October 1981, Accession number 33. Kevin Coyne, A Predictive Model of Nuclear Power 8110200569 Plant Crew Decision-Making and Performance in a

12. Final Accident Sequence Precursor Analysis for La Dynamic Simulation Environment (Doctor of Crosse Boiling Water Reactor, LOOP with DG Failure Philosophy dissertation, University of Maryland, 2009),

to Load and Associated and HPCS Unavailability, 16 https://drum.lib.umd.edu/handle/1903/9888 July 1984, ML20149K391 34. Fu, Xiaocheng et al. 2020. Analysis on Critical Factors

13. Final Accident Sequence Precursor Analysis for of Marine Organism Impacts on Water Intake Safety at Catawba Nuclear Station, Asiatic Clams Degrade Nuclear Power Plants, Journal of Nuclear Engineering Auxiliary Feedwater System, 9 March 1988, ML and Radiation Science, Volume 6, 041101-1 thru 6.

20147A500 35. See reference 3 above.

14. Final Accident Sequence Precursor Analysis for Zion 36. See reference 4 above.

Nuclear Power Station, Unavailability of Turbine-37. See reference 18 above.

Driven Auxiliary Feedwater Pump and Emergency 38. Zhang, Jianfei et al. 2021. Research on the Detection Diesel Generator, 7 March 1994, ML20140A230 and Early Warning Technology of Harmful Marine

15. US NRC Information Notice No. 94-90: Transient Organisms in the Water Intake of Nuclear Power Plant[s]

Resulting in a Reactor Trip and Multiple Safety Injection by 3D Image Sonar, Fourteenth International System Actuations at Salem, 30 December 1994, Conference on Genetic and Evolutionary Computing, 1-ML031060383 4.

16. US NRC Event Notification 37287 for Wolf Creek 39. See reference 17 above.

Generating Station, Reactor Trip due to Fire in Unit 40. See reference 24 above.

Auxiliary Transformer, 5 September 2000

17. Final Accident Sequence Precursor Analysis for Donald C. Cook Nuclear Plant, Degraded ESW Flow Renders Both Unit 2 Emergency Diesel Generators Inoperable, 23 December 2003, ML20112F482

18. US NRC Inspection Report 05000315/316/2003-008, DC Cook Nuclear Power Plant, Units 1 and 2, Special Inspection Report, 3 July 2003, ML031880726

19. US NRC Inspection Report 05000528/529/530, Palo Verde Nuclear Generating Station Augmented Inspection Team, 16 July 2004, ML042020061

20. US NRC Event Notification 43541 for Edwin I. Hatch Nuclear Plant, Unusual Event Declared due to Fire in Protected Area Greater than Ten Minutes, 1 August 2007

21. US NRC Event Notification 47178 for St. Lucie Plant, Manual Reactor Trip due to Rising Condenser Backpressure, 23 August 2007

22. US NRC Licensee Event Report for R. E. Ginna Nuclear Power Plant 05000244/2012-001, Automatic Start of B Emergency Diesel Generator Caused by Loss of Offsite Circuit 767 due to Wildlife, 26 July 2012, ML12212A214

23. US NRC Event Notification 48638 for Surry Power Station, Emergency Diesel Generator Auto-Start and Load Upon Trip of Reserve Station Transformer, 31 December 2012

24. US NRC Licensee Event Report for Fermi 05000341/2020-002-00, Valid Actuation of Emergency Diesel Generators During Partial Loss of Offsite Power, 31 August 2020, ML20244A365

25. US NRC Licensee Event Report for Browns Ferry Nuclear Plant 05000259/2020-001-00 Manual Reactor Shutdown of Two Units on Decreasing Condenser Vacuum due to Eel Grass Intrusion, 16 September 2020, ML20260H103

26. See reference 3 above.

27. Ahmad, Ali. 2021. Increase in Frequency of Nuclear Power Outages due to Changing Climate. Nature Energy, Volume 6, 755-762

28. US NRC, Use of Probabilistic Risk Assessment Methods in Nuclear Activities: Final Policy Statement, Federal Register, Vol. 60, p. 42622 (60 FR 42622), 16 August 1995