ML14329B209

Summary of Regulatory Conference to Discuss Safety Significance of Arkansas Nuclear One Flood Protection Deficiencies

ML14329B209
Person / Time
Site:	Arkansas Nuclear
Issue date:	11/25/2014
From:	Ryan Lantz NRC/RGN-IV/DRP/RPB-E
To:	Jeremy G. Browning Entergy Operations
C. Young
References
EA-14-088
Download: ML14329B209 (40)

Text

UNITED STATES NUCLEAR REGULATORY COMMISSION REGION IV 1600 E LAMAR BLVD ARLINGTON, TX 76011-4511 EA 14-088 Jeremy Browning, Site Vice President Arkansas Nuclear One Entergy Operations, Inc.

1448 SR 333 Russellville, AR 72802-0967

SUMMARY

OF REGULATORY CONFERENCE TO DISCUSS SAFETY SIGNIFICANCE OF ARKANSAS NUCLEAR ONE FLOOD PROTECTION DEFICIENCIES

Dear Mr. Browning:

On October 28, 2014, members of the U.S. Nuclear Regulatory Commission staff met with representatives of the Arkansas Nuclear One facility to discuss apparent violations affecting Units 1 and 2 related to flood protection deficiencies as documented in Nuclear Regulatory Commission Inspection Report 05000313; 368/2014009, issued on September 9, 2014 (ML14253A122). The focus of the regulatory conference was a discussion of information important to characterize the safety significance of the flood protection deficiencies. The discussion included methodologies used by Entergy to develop the probable maximum precipitation and probable maximum flood for the Arkansas Nuclear One site, including development of an annual exceedance probability for the probable maximum precipitation. The discussion also included actions that may have been taken in the event of flooding at the site to mitigate the consequences of the flooding performance deficiencies.

The Nuclear Regulatory Commission staff asked questions during this regulatory conference, with some questions requiring additional information that you indicated you would supply to us.

The Nuclear Regulatory Commission will continue to review the information that you provided during the Regulatory Conference and the subsequent information that was requested in order to reach a final significance determination. We will issue a final significance determination letter to you when that review has been completed.

This Category 1 public meeting was attended by two members of the public at the Region IV office, as well as several members of the public on the teleconference bridge that was provided.

A copy of your presentation slides is included as (Enclosure 1). Copies of the Nuclear Regulatory Commission slides (Enclosure 2) and meeting attendance lists (Enclosure 3) are also included.

November 25, 2014

In accordance with 10 CFR 2.390 of the NRCs Rules of Practice, a copy of this letter and its enclosures will be available electronically for public inspection in the NRCs Public Document Room or from the Publicly Available Records (PARS) component of the NRCs ADAMS. ADAMS is accessible from the NRC web site at http://www.nrc.gov/reading-rm/adams.html (The Public Electronic Reading Room).

Sincerely, Ryan E. Lantz, Chief Project Branch E Division of Reactor Projects Docket Nos.: 50-313, 50-368 License Nos.: DPR-51, NPF-6

Enclosures:

1. ANO Presentation Slides

2. NRC Slides

3. Meeting Attendance Forms

/RA/

ARKANSAS NUCLEAR ONE REGULATORY CONFERENCE October 28, 2014 OPENING COMMENTS Jeremy Browning ANO Site Vice-President Agenda

• Opening Comments

• Performance Deficiency

• Risk Significance Assessment

• Overview of Risk Assessment

• Station Response

• PMP and PMF Analysis

• PMF Hydrology and Hydraulics External Peer Review

• PMP Probability Evaluation

• Significance Determinations

• Closing Comments 3

Opening Comments

• ANO and Entergy recognize the significance of the performance deficiency

• Entergy concurs with the performance deficiency

• Root cause evaluations performed

• Effective immediate and long-term corrective actions have been taken

• Fleet actions

• Focus of this meeting is to discuss risk significance of the violation

• Provide best estimate of initiating event frequency

• Site specific flood hazard analysis

• Industry experts

• Up to date techniques

• Credit given to reasonable actions that would be pursued by the station to mitigate effects of flooding

• We appreciate the opportunity to share additional information that improves the accuracy of the estimated risk following the event and supports a qualitative risk assessment 4

PERFORMANCE DEFICIENCY Bryan Daiber ANO Design Engineering Manager Performance Deficiency

• ANO agrees that there were deficiencies in the flood boundary

• Deficiencies can be placed in three categories 1.

Original construction 2.

Preventative Maintenance associated with hatches 3.

Equipment removed from service

• Deficiencies are legacy issues

• Deficiencies have been corrected or compensated for

• Inspection conducted using industry approved guidelines in Nuclear Energy Institute (NEI) 12-07

• Incorporated lessons learned 6

Performance Deficiency

• Plant grade elevation 354

• No mitigating actions required

• The design flood elevation is 361

• Risk significance varies with elevation

• Below 356 - Mitigating actions available

• Unit 1 - 28 deficiencies (~1.0 ft2)

• Unit 2 - 35 deficiencies (~0.72 ft2)

• Emergency Diesel Generator (EDG) Fuel Oil Vault - No deficiencies 7

OVERVIEW OF RISK ASSESSMENT Dale James Regulatory and Performance Improvement Director

Overview of Risk Assessment

• Vulnerability of performance deficiencies

• Site mitigating actions

• Flood protection

• Operator response

• ANO flooding reevaluation performed

• Probable Maximum Precipitation (PMP) Study

• Probable Maximum Flooding (PMF) Evaluation 9

Overview of Risk Assessment

• The World Meteorological Organization defines the PMP as The greatest depth of precipitation for a given duration meteorologically possible over a given size storm area at a particular location and at a particular time of the year, with no allowance made for [future] long term climatic trends.

• The PMF is defined by ANSI/ANS-2.8-1992 (ANS 1992) as the hypothetical flood (peak discharge, volume, and hydrograph shape) that is considered to be the most severe reasonably possible, based on comprehensive hydrometeorological application of PMP and other hydrologic factors favorable for maximum flood runoff such as sequential storms and snowmelt.

10 Overview of Risk Assessment

• PMP/PMF evaluation

• Site-specific study performed for 50.54(f) response to Fukushima flooding reassessment

• Consistent with NRC and other federal agencies guidelines

• Where information was not available conservative assumptions were made

• Results in site PMF level of 353.8 ft.

11 Overview of Risk Assessment

• Probability of PMP/PMF

• Probability is not zero

• Site-specific study performed to derive probabilistic estimate of PMP

• Multiple methods applied

• Consistent with methodology used by other federal agencies to make risk-based decisions

• Annual Exceedance Probability (AEP) based on PMP studied conservative with respect to PMF 12

Overview of Risk Assessment

• Risk Assessment performed utilizing PMP probability combined with site mitigating actions

• Risk from internal flooding in turbine building combined with external risk to determine overall risk

• PRA Results

• Unit 1

• CDF = 7.99E-07/year

• Unit 2

• CDF = 2.16E-06/year 13 Overview of Risk Assessment

• Manual Chapter 0609 - Appendix M Assessment

• 1.4 E-5 bounding risk - Based on site specific upper 95% for PMP resulting in 354 site flood elevation

• Bounding value should be approximately an order of magnitude lower when considering PMF

• Mitigating actions would reduce risk further 14 STATION RESPONSE Gary Sullins Unit 1 Operations Manager, Shift John Hathcote Unit 2 Operations Manager, Shift

• Describe plant response to a flooding event

• Actions to prepare the plant

• Equipment effects

• Mitigating actions for flooding of Auxiliary Buildings

• Focus on water level between 354 and 356

• Developed in accordance with procedures in effect at the time

• No knowledge of deficiencies 16 Objective

317 U1 EFW SFP Clg LPI/CS/DHR U2 HPSI AB Sump U2 EFW U2 Charging U1 HPI Control Room SFP 362 EDGs AC/DC Switchgear, Batteries, etc.

AUXILIARY BLDG FUEL OIL VAULT 355 378 366 378 INTAKE M

Normal Lake Level 338 404 386 372 354 335 368 329 TURB.

BLDG U2 AFW Ground Level 354 335 General Area 317 General Area SW Pumps Fire Water Pumps Note: 317 and 335 general areas are isolated between units.

Plant Layout Offsite Power 17

• Unprecedented rainfall upstream of ANO

• Flood is forecasted days in advance

• 5-day (120-hour) advance forecast per ANO SARs

• PMP/PMF analysis supports > 5 day advance warning

• Abnormal Operating Procedure (AOP) entry occurs at forecast of > 350

• Corporate Severe Weather procedure 18 Initiating Event 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 0

5 10 15 20 25 30 Precipitation (inch)

Stage (feet, NGVD29)

Time (days)

Warning Time for PMF Precipitation PMF Stage 135 hr 120 hr 19 Warning Time PMF Peak Stage 353.8 ft End of PMP Onset of PMP Antecedent Storm (40% PMP)

• Corporate Severe Weather Procedure

• Establishes severe weather duty roster

• Staffs Corporate Emergency Response Center

• Prompts installation of temporary flood barriers

• Natural Emergencies AOP

• Establishes contact with Corps of Engineers

• Shuts down both units

• Aligns SU2 Transformer for high water conditions

• Protects equipment (i.e., portable pump for steam generator (SG) feed)

• Response for Forecasted Flood > 350

• Well-staffed severe weather response teams (on and off site)

• Timely plant shutdown

• Intensive effort to protect the site 20 Site Preparation

21 Timeline Time (hrs)

Pre-Flood Actions 120 Enter severe weather procedures (forecast > 350) 108 Implement severe weather duty roster Initiate flood protection actions 96 Commence shut down of both units 84 Both units in Mode 3 Hot Standby (Reactor Subcritical) 72 Both units in Mode 5 Cold Shutdown (CSD) (RCS < 200F) 30 Notification of Unusual Event (NUE) Declared at > 345

• Prompted by Corporate Severe Weather Procedure

• Two Options/Examples

• Water Tube System

• Use of local materials 22 Temporary Flood Barriers 23 Water Tube System Entergy Sub-station - New Orleans 24 Flood Protection - Use of Local Materials Unit 1 South Entrance - Elevation 354

• Initial Conditions

• Mode 5 CSD (RCS <200F) with Decay Heat Removal (DHR) in service

• Reactor Coolant System (RCS) intact

• Emergency Feedwater (EFW) available

• Potential Impact to Plant Equipment

• Water accumulates in general areas of Auxiliary Building

• As water rises in Auxiliary Building, water eventually enters DHR vaults and DHR System becomes unavailable

• As water level rises above 335, EFW and High Pressure Injection (HPI) pumps become unavailable 25 Unit 1 - Challenges to Plant Equipment 26 Equipment Effects - Unit 1 Time Estimates (hours)

Event 0

Water begins entering the Auxiliary Building 4.5 Level at 335. Rising at 1.7 ft/hr 4.5 A DH Vault HI LEVEL alarm 5

B DH Vault HI LEVEL alarm 5

B Motor Driven EFW pump not available 6.5 Level at 338.8, lowest elevation of MOV for EFW suction alignment 9

A DHR Pump unavailable 12 B DHR Pump unavailable Based on in-flow analysis of identified barrier deficiencies

• Service Water (SW) to EFW

• Aligned at suction to EFW pumps

• Portable Pump for SG Feed

• Connected via Main Feedwater header 27 Procedure Options to Mitigate Flooding of Unit 1 Auxiliary Building

• Operating Crew and Support Staff recognize loss or impending loss of DHR and EFW

• Field monitoring of water level in Auxiliary Building

• DH Vault flood alarms (2 in room)

• Service Water aligned to suction of EFW Pump(s) per Overheating Emergency Operating Procedure (EOP)

(Control Room action)

• Two handswitches per train from the Control Room

• Injection control remains available from the Control Room 28 Unit 1 Service Water to EFW

• Three portable pumps for SG feed available

• Method of core cooling independent of EFW and DHR

• 1203.048 Security Event - Feedwater Actions

• Operators trained on procedure

• Feed source connected to Main Feedwater piping >372

• Provides 200 gpm flow to Steam Generator

• Available response time > 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> from loss of DHR based on Time to Core Uncovery (TTCU) calculations

• Demonstrated implementation time ~ 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />

• Practical to 356 29 Unit 1 Portable Pump for SG Feed

• Similar initial conditions at time of postulated flood Unit 2

• Mode 5 CSD (RCS < 200F) with Shutdown Cooling (SDC) in service

• RCS intact

• EFW available

• Potential Impact to Plant Equipment

• Water accumulation in general areas of Auxiliary Building, Emergency Safeguard Features (ESF) Vaults and EFW Rooms

• As water rises in the ESF Vaults, the SDC system becomes unavailable

• As water level continues to rise, the EFW pumps become unavailable 30 Unit 2 - Challenges to Plant Equipment 31 Equipment Effects - Unit 2 Time Estimates (hours)

Event 0

External flood level at 354.2

<0.5 Water begins to enter Aux Building General Area, B ESF Vault, 2P-7B EFW Room

<0.5 B ESF Vault and 2P-7B Room HI LVL Alarms 3

2P-7B unavailable 15 SW MOV to 2P-7A Unavailable 16 A SDC Pump unavailable 29 B SDC pump unavailable (Loss of forced DHR)

• Operating crew and plant support staff recognize the impending loss of EFW and SDC

• Room alarms and field reports

• 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> before option is unavailable

• Procedural Guidance to seek alternate sources of water for SGs

• Lower Mode Functional Recovery (LMFR) EOP Continuing Actions

• Heightened site focus on heat removal safety function

• Common approach to Unit 1

• Simple Control Room action to line up SW MOVs

• Two handswitches in Control Room 32 Unit 2 SW to EFW Success Path

• Consistent with Unit 1

• RCS Heat Removal independent of SDC

• 1203.048 Security Event Attachment J Feedwater Actions

• Guidance to feed Unit 2 SGs using portable pump

• Operations training provided on specific approach

• Procedural Actions

• Stage portable pump for SG feed near hydrant

• Connect hose from hydrant to flange on Startup and Blowdown (SU/BD) system

• Requires cutting 4 pipe to allow access to flange

• Procedure successfully mocked-up in less than one hour

• Response time estimate >24 hours from loss of SDC 33 Unit 2 Portable Pump for SG Feed

• Flooding conditions forecasted days in advance

• Actions to prepare

• Augment staffing

• Shut down both units

• Install temporary flood barriers

• Advance plant shutdown provides time and resources to support mitigating actions

• Mitigating actions

• Service Water to EFW Pump Suction

• Portable Pump for SG feed connected via Main Feedwater Piping 34 Station Response - Conclusion PMP AND PMF ANALYSIS David M. Leone, P.E.

Associate Principal / Hydraulic Engineer, GZA Overview

• Calculations initiated to respond to NRC 50.54(f) request

• Conforms to NRC guidance and other federal guidelines (NUREG/CR-7046, FERC, USACE & NWS HMRs)

• Addressed large watershed (153,000+ sq.mi.) and data availability through

• Site specific PMP using methods endorsed by FERC, NRCS & USACE, and various state dam safety agencies

• Extensive calibration and verification in lieu of complete USACE dam information

• Incorporation of conservatisms (non-linearity, etc.)

• Unsteady USACE HEC-RAS model used to calculate PMF elevation at ANO of 353.8 ft (NGVD29) 36

=

Background===

• Original Design Basis Flood

• Maximum Probable Flood analysis by USACE in 1956

• Design Discharges for Dardanelle Dam, USACE, June 1956

• Watershed PMP developed based on 1943 historic storms

• Moved to critical locations and maximized

• Occurred in series (back-to-back)

• Storms covered an area of 49,612 square miles

• Infiltration losses for Maximum Probable Flood (MPF) computed based on 1943 stream gage data

• Peak discharge = 1.5 million cfs

• Effects of flood regulation from dams and reservoirs were minimal (at that time) 37 38

Background:

Original Design Basis Flood

• Straight line interpolation used to estimate water surface profile (from tailwater of Ozark L&D to Dardanelle L&D headwater)

• PMF maximum stillwater elevation = 358 ft

• Wind-generated waves add 2.5 ft = 360.5 ft

• Alternative combined effect flood used was PMF (358 ft) plus failure of Ozark Dam (3 ft) = 361 ft

• No waves added to this alternative Ozark tailwater Dardanelle headwater ANO Overview David Leone, GZA, PMF Calculation Preparer B.S/M.S. in Civil Engineering, Licensed P.E. in three states Over 16 years of hydrologic and hydraulic engineering experience focusing on rainfall-runoff and hydraulic computer modeling and assessment and design of hydraulic structures such as dams.

Team member for Post-Fukushima 50.54(f) Flood calculations on 18 NPP sites Peter Baril, GZA, PMF Calculation Reviewer B.S. Biology /M.S. Hydrology, Licensed P.E. in 4 states, Professional Hydrologist (AIH) since 1988 Over 26 years of hydrologic and hydraulic experience focusing on flood control analysis and design.

Reviewer for Post-Fukushima 50.54(f) calculations on more than 10 Nuclear Power Plant (NPP) sites Bill Kappel, Applied Weather Associates, PMP Calculation Meteorologist Preparer B.S. Physical Science / M.S. Broadcast Meteorology Over 17 years of meteorology experience. Probable Maximum Precipitation and extreme storm analysis specialist since 2003.

Applied Weather has completed over 60 site-specific PMP studies, including numerous 50.54(f) nuclear power plants.

Dr. Ed Tomlinson, Applied Weather Associates, PMP Calculation Meteorologist Reviewer B.A. Mathematics / M.S. and Ph.D. in Meteorology Over 40 years of meteorology experience. Peer reviewer for HMR-57.

PMP and extreme storm analysis specialist.

39 Overview

• Background Information and Consistency with Standards and Guidance for Flood Evaluation

• Hydrologic Setting

• Probable Maximum Precipitation (PMP)

• Probable Maximum Flood (PMF) - Hydrology

• Probable Maximum Flood (PMF) - Hydraulics 40

Consistency with Standards and Guidance for Flood Evaluation

• Work performed to support NRC request (50.54(f) letter):

• Uses current state of knowledge and analytical methods

• Methods used in present-day standard engineering practice to develop the flood hazard

• NRC guidance: NUREG/CR-7046

• Federal guidance for standard hydrologic investigations:

• PMF:

• FERC (Engineering Guidelines for the Evaluation of Hydro power Projects Chapter 8 Determination of the Probable Maximum Flood)

• USACE Engineering Manuals (EM1110-2-1417 Flood Runoff Analysis, etc.)

• PMP:

• NWS/USACE HMR Publications (HMRs) 41 42 Hydrologic Setting

• Watershed is large: 153,000+ square miles ANO 43 Hydrologic Setting

• Watershed has changed over the years: Dams and reservoirs constructed between c.1940 and c.1970 for flood control, most after 1956 44 PMF Hydrology - Flood Control Dams Date Completed Number of Dams Percent 1800 - 1939 9

1%

1940 - 1949 25 1%

1950 - 1959 (Design Basis 1956) 102 6%

1960 - 1969 688 39%

1970 - 1979 487 27%

1980 - 1989 213 12%

1990 - 1999 138 8%

2000 - 2011 55 3%

Unknown 55 3%

Total 1,772 100%

45 Hydrologic Setting

• Arkansas River highly regulated since completion of MKARNS

• ANO located 5 miles upstream of Dardanelle Lock & Dam ANO 46 Hydrologic Setting Dardanelle Lock and Dam:

Regulates water level 336 to 338 ft for flows up to 600,000 cfs ANO:

Site Finished Elevation 354 ft PMP Calculation Process Evaluate Applicability of HMR51/52 Watershed Area > 20,000 sq.mi.?

Is Watershed in Stippled Area?

Derive PMP Depth Area Duration with HMR51/52, etc.

Perform Site-Specific PMP Meteorology Study NO YES Derive PMP Depth Area Duration Using Study Results Apply Depth Area Duration to Watershed (Calculate Hyetographs)

OR

• Site specific PMP studies are standard hydrologic practice using Hydrometeorological Reports and World Meteorological Organization procedures

• Site specific PMP case studies have been generally accepted on an individual basis by FERC, NRCS &

USACE, various state dam safety agencies Site Specific PMP - Consistency with Standards and Guidance

• Storm Maximization, transposition and elevation-adjusted to watershed

• Increase observed extreme rainfall by increasing available moisture

• Dates: 1895 to 2013

• Dew points increased to a climatological maximum

• Vary time of occurrence +/-2 weeks toward warm/wet season

• Depth-Area-Duration up to 72-hours for area sizes of up to 100,000-square miles (meteorological limits) 49 Computing PMP Values

• Technical conservatisms used from study:

• Storm rainfall patterns positioned within HMR-52 guidelines to preserve full rainfall value of the PMP

• PMP conservatively simulates the movement of a storm, covering as much of the watershed as practicable (Not included in HMR-51/52)

• PMP speed and storm track for moving storms selected to maximize precipitation in the watershed 50 Conservatisms - PMP 51 Results: PMP Values

• Storm orientation optimized for maximum rainfall per NUREG/CR-7046, consistent with FERC guidelines

• PMP temporal distribution per NUREG/CR-7046 and HMR-52 recommendations

• 40% of the PMP used as an antecedent storm per NUREG/CR-7046 Antecedent Storm Probable Maximum Precipitation Source Storm Area (square miles)

Estimated Volume (acre-feet)

Estimated Average Depth (inches)

Storm Area (square miles)

Estimated Volume (acre-feet)

Estimated Average Depth (inches)

FSAR / Original PMP 49,612 19 million 7.5 (5 day break) 49,612 27 million 10.0 Reevaluation PMP (moving at 80 miles per day)

(about 50,000)

N/A 2.5 (about 50,000)

N/A 10.7 116,747 12 million 2.3 (3 day break) 119,103 41 million 8.7

• Analyzed three stationary and two moving PMP candidates

• PMP moving at 80 miles per day results in the most precipitation within the watershed.

52 PMP Application Example of moving PMP with different storm center at each 24-hr increment.

Heaviest rainfall is closest to the storm center with Ellipses of equal rainfall offset from center.

ANO

53 ANO PMF Calculation Process Delineate Watershed and Subwatersheds Calibrate and Verify HEC-HMS Rainfall-Runoff Model (20 stream gages, 6 floods used for each)

Develop PMF Flows with HEC-HMS using PMP Inputs Develop PMF Elevations with HEC-RAS using HEC-HMS Inputs Apply Site Specific PMP Results to Calculate PMP Input

• PMF development performed per NRC, USACE, and FERC guidance

• Use of USACE HEC-HMS software as per NUREG/CR-7046 and USACE standard procedure

• Use of Unit Hydrograph method as per NUREG/CR-7046, USACE, and FERC guidance

• HEC-HMS model was calibrated/verified to observed floods per NRC, USACE and FERC guidance

• Initial losses are zero for the PMF (conservative) per NUREG/CR-7046

• Non-linearity adjustment per NUREG/CR-7046 and USACE guidance 54 PMF Hydrology - Consistency with Standards and Guidance

• 5,200+ dams located in watershed

• 1,772 dams list flood control as a purpose

• Conservative assumptions based on publically available information about watershed dams were used

• No operator actions to create or preserve flood storage:

• The starting pool elevation was conservatively modeled at the upper limit of normal pool elevation during the PMF

• Low level outlet works were not modeled in HEC-HMS

• Gate operations were not modeled in HEC-HMS

• Six large dams were modeled based on publically available info

• Other flood control dams were represented by river routing parameters and rainfall-runoff transformation parameters, as calculated through calibration / verification 55 PMF Hydrology - Dams 56 PMF Hydrology - Dams

• Six flood control dams explicitly modeled (of 1,772)

57 PMF Hydrology - Calibrate and Verify

• This process addresses uncertainty due to lack of dam information

• Demonstrates conservatism

• 20 USGS/USACE stream gages used

• Three calibration and three verification floods used for each gage

• Verified parameters bounded by calibrated parameters

• Conservatism added to account for the PMF significantly exceeding the magnitude of the calibration and verification floods:

• Non-linearity: NUREG/CR-7046 recommends increasing the peak discharge of the unit hydrograph by one-fifth and decreasing the time-to-peak by one third.

• HEC-HMS model was designed to over-predict peak flow vs observed flood data and minimize flood volume differences

• Initial losses of rainfall set to zero for PMF simulations 58 PMF Hydrology - Calibrate & Verify 59 PMF Hydrology - Nonlinearity 0

20,000 40,000 60,000 80,000 100,000 120,000 0

20 40 60 80 100 120 140 160 Unit Hydrograph Discharge (CFS)

Time (hours)

Non-linearity Response Adjustment to Unit Hydrograph Original Nonlinear This adjustment was done AFTER the calibration and verification process was completed

• Initial losses during the PMF were zero

• Constant losses initially estimated from the minimum published typical infiltration rate for each hydrologic soil group

• Soil group with the higher runoff potential was used in constant loss calculation for dual class soils

• Model typically slightly over-predicted peak flows and flood volumes for calibration / verification floods

• The starting pool elevation was conservatively modeled at the upper limit of normal pool elevation during the PMF

• Spillway weir coefficient for ogee-shaped crests were conservatively assumed to be 3.2

• Low level outlets and gates were not modeled in HEC-HMS

• Non-linearity adjustments were made 60 Summary of Hydrology Conservatisms

0 250,000 500,000 750,000 1,000,000 1,250,000 0

5 10 15 20 25 30 Flow (cfs)

Time (days)

PMF Hydrograph - at Dardanelle 61 PMF Hydrology Post-MKARNS Flood of Record: 433,000 cfs (1990)

Pre-MKARNS Flood of Record: 683,000 cfs (1943)

Maximum PMF Peak Flow Rate (Re-evaluated) =

1,226,000 cfs FSAR PMF Flow Rate =

1,500,000 cfs (18% change likely attributable to the effects of watershed regulation/dams and use of more accurate modern day methods)

• PMF Hydraulics used standard-of-practice USACE computer models and conformed to NRC and USACE guidance

• Use of USACE HEC-RAS per NUREG/CR-7046, Appendix C

• Unsteady flow routines (more accurate than steady flow) used to fully account for dynamic and momentum effects

• Roughness for Arkansas River and floodplain were selected based on published FEMA values 62 PMF Hydraulics - Consistency with Standards and Guidance

• No floods after MKARNS exceed the 600,000 cfs threshold (about 338 ft)

• HEC-RAS: Over 500 cross-sections over 117 miles, including four Locks & Dams and two bridges PMF - Hydraulics Conservative assumptions for bathymetry (underwater areas)

Bathymetry and channel geometry was based on data from USGS topographic maps in certain areas Channel invert elevations and dimensions based on USACE information

• Nine-foot by 250-foot rectangular channel

• Channel is a small part of overall conveyance and below water area in lake areas

• Channel shape does not artificially lower entire lake bed

• Conservative because sections of the channel may be wider and deeper, yielding increased channel capacity 64 Conservatisms - PMF Hydraulics

PMF - Hydraulics 290 310 330 350 370 390 410 430 0

2,000 4,000 6,000 8,000 10,000 12,000 14,000 Elevation (feet, NGVD29)

Distance (feet)

Typical Cross Section Info Sources DEM Data USGS Bathymetric Data USACE Channel Data DEM Data 334 336 338 340 342 344 346 348 350 352 354 0

5 10 15 20 25 30 Stage (feet, NGVD29)

Time (days)

Stage Hydrograph at ANO 66 PMF - Hydraulics 353.8 ft Candidate PMPs / PMFs Candidate PMP Storm PMF Peak Discharge Adjusted for Nonlinearity (cfs)

Total Outflow Volume (acre-feet)

Peak Elevation at ANO (feet, NGVD29)

Moving PMP (80 miles per day) 1,226,000 20,970,202 353.8 Watershed Centroid 1,222,000 16,195,091 353.3 John Martin Subwatershed Centroid 90,000 4,487,452

<338.0 Robert Kerr Subwatershed Centroid 1,260,000 18,273,796 351.7 Moving PMP (160 miles per day) 1,015,000 16,559,493 351.6 67 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 0

5 10 15 20 25 30 Precipitation (inch)

Stage (feet, NGVD29)

Time (days)

Warning Time for PMF Precipitation PMF Stage 135 hr 205 hr 120 hr 68 Warning Time PMF Peak Stage 353.8 ft End of PMP Onset of PMP Antecedent Storm (40% PMP)

69 Waves

• Wave action minimal shallow flood depths generally less than one foot Maximum possible wave is 0.78 x depth from crest to trough Wave crest heights are thus even lower

• Waves would be broken by barriers prior to reaching important buildings

• Waves at peak flood are independent from PMP storm due to 5+ day lag time Calculation Current Licensing Basis Flood Re-evaluation Notable improvements PMP Used May 6-12, 1943 storm followed by transposed May 6-12, 1943 storm, increased by 34%

Study performed in the 1950s by the U.S.

Weather Bureau Comprehensive site-specific PMP study using the 1943 storms and dozens of others per HMR and WMO procedures.

Incorporates several decades of additional data collection and numerous analyses improvements. Use of a moving PMP, which would not have been easily done in the 1950s and was not included in HMR-51/52.

70 Flood Re-evaluation Comparison Calculation Current Licensing Basis Flood Re-evaluation Notable improvements PMF-Hydrology Study performed in 1956.

Flows from manual hydrograph manipulation based on 1943 flood, adjusting for increased rain (PMP) & estimated rainfall losses. Manual channel routing with minimal reservoir effects. Essentially unregulated watershed.

HEC-HMS computer model using unit hydrograph method and Muskingum channel routing.

Modeled 22 subwatersheds. Six reservoirs explicitly modeled with calibration/verification used to provide reality check. Applied non-linearity response adjustments to PMF.

Comprehensive HEC-HMS computer model used. Calibration and verification per NRC, USACE, and FERC guidance.

Demonstrably /

conservatively captures response of the post-MKARNS watershed.

71 Flood Re-evaluation Comparison Calculation Current Licensing Basis Flood Re-evaluation Notable improvements 358.0 ft (stillwater) 353.8 ft (stillwater)

PMF-Hydraulics Study performed in 1950s.

Estimated flood elevation at ANO using a straight-line interpolation from USACE information at the tailwater of Ozark Lock and Dam and crest of Dardanelle Dam HEC-RAS computer model using unsteady (dynamic) flow module.

Hydraulic routing of hydrograph through Ozark Lock and Dam and Dardanelle Lock and Dam.

Unsteady (flow varying) dynamic analysis based on actual floodplain geometry and lock & dam configuration.

HEC-RAS model captures complexities of water surface profile variation through Lake Dardanelle instead of simplified straight-line method.

72 Flood Re-evaluation Comparison

PMF HYDROLOGY AND HYDRAULICS EXTERNAL PEER REVIEW Jeff Harris Senior Engineer, WEST Consultants, Inc Jeff Harris, P.H.

• 36-year career in Hydrology and Hydraulics at USACE

• Retired May, 2013

• Last 12 Years at Hydrologic Engineering Center

• HEC is USACE CX

• Eight years Chief of Hydrology and Hydraulics Technology Division

• Chair of USACE Committee on Hydrology

• USACE Surface Water Hydrology Subject Matter Expert

• H&H Lead for USACE Katrina IPET 74 Hydrology and Hydraulics Analysis

• Challenge

• Develop Hydrologic Modeling System (HMS) and River Analysis System (RAS) models of Arkansas River Basin for Probable Maximum Flood computation

• Obstacle

• Lack of available data for modeling multiple reservoirs and routing flows. Only public data available

• Solution

• Since PMF elevation in Lake Dardanelle is the goal, model parameters were used in the hydrologic models to generate the maximum PMF flow and stage in the lake and stay consistent with USACE methods.

75 HEC-HMS Model Parameters Selected Comparison to USACE PMF Practices

• Maximum Runoff

• Initial Loss

• Set to zero.

• Typical USACE assumption. (In accordance with NUREG/CR-7046)

• Constant Loss

• STATSGO

• Loss rates used conform to USACE Guidance EM 1110-2-1417

• Used soil type with highest runoff potential

• Hydrologic Routing

• Muskingum method

• Storage coefficient set low as possible and still maintain stability

• Decrease channel storage, more water moves downstream.

• Described in HEC-HMS User Manual and textbooks (i.e. Chow) 76

HEC-HMS Dam Modeling Decisions Comparison to USACE PMF Practices

• Maximize PMF Runoff

• Used HMS Reservoir Element

• Six Large Dams Modeled

• Storage at normal pool which is above spillway crest

• Three large Dams not included

• 4.1M ac-ft. storage

• Small dams not included

• Increased runoff

• Dardanelle

• Low weir coefficient (USACE HEC-RAS Hydraulic Reference Manual)

• Model Calibration 77 Upstream Dam Modeling

• No Failures Assumed

• PMP generated for specific location

• Arkansas a very large basin

• PMP does not occur at all locations

• In general, spillways designed for PMF

• Including failures would make PMF a more rare event 78 HEC-RAS Model Parameters

• USACE PMF Computation

• Spillway Adequacy

• Dardanelle Weir Coefficient

• Low weir coefficients. (USACE HEC-RAS Hydraulic Reference Manual)

• Topographical Data

• Assumptions made to minimize storage

• USACE model may differ 79 USACE Guidance ER 1110-8-2 (FR)

• Antecedent Conditions

• Reservoir at full flood control pool elevation

• Elevation five days after an event equal to 1/2 PMF

• Gate Operation

• Reservoir regulating outlets should not be assumed operable during PMF unless designed for PMF

• Dardanelle manned 24/7/365

• Long warning time

• Unit Hydrograph

• Peaked 25-50%

• NRC guidance NUREG/CR-7046 80

Summary

• Modeling methods maximized PMF runoff

• Losses

• Routing

• Dam Modeling

• USACE Software Methods and Parameters (HMS, RAS)

• Applied correctly

• Applied in accordance with USACE guidance

• USACE PMF modeling will follow similar steps

• Methodology GZA applied is conservative but consistent with USACE guidance 81 PMP PROBABILITY EVALUATION Jason Caldwell, Ph.D.

Sr. Project Engineer/Modeler Leonard Rice Engineers, Inc.

Overview

• Jason Caldwell, Ph.D., P.H., LRE, Lead - Probability of PMP

• B.S./M.S. Meteorology; Ph.D. Civil Engineering (Stochastic methods/uncertainty)

• Over ten years of atmospheric, hydrologic, and climate modeling experience focusing on extreme precipitation events

• General and site-specific PMP and regional frequency for hydrologic hazards analysis

• Research projects for NRC on evaluation of existing PMP estimates in the Carolinas and Tennessee River Valley

• LRE Team:

• Monica Bortolini: Group Manager - Civil Engineering

• Todd Street: Project Engineer - Civil Engineering

• Others: Katy Kaproth-Gerecht, James Tobler, Jessica DiToro

• Peer Reviewer and Technical Support:

• Mel Schaefer, MGS: Stochastic hydrologic modeling, precipitation frequency, Watts Bar/Sequoyah probability study

• Tye Parzybok and Debbie Martin, MetStat: Co-authors on NOAA Atlas 14, WMO extreme precipitation evaluation committee member 83 Organization

• Review of the multiple methods approach to hydrologic hazards

• Calculations to evaluate probability of PMP consistent with procedures at federal level (NOAA, Reclamation, NRC)

• Address limitations of large watershed and AEP information in national products through regional analysis

• Review of conservatisms associated with the PMP and relationship to the PMF 84

Background:

Inspection Report Notes

• Extreme events vary between 1 x 10-3 and 1 x 10-6 AEP based on Harris and Brunner conference paper

• Similar estimates available from literature review

• Typically based on generalized PMP at 20,000 mi2 or smaller

• Flood frequency extrapolations limited to approximately twice the length of record (Interagency Advisory Committee on Water Data 1982)

• Regional methods allow statistical extension beyond individual site period of record

• USNRC NUREG/CP-0302: PFHA Workshop Proceeding identifies Bulletin 17B as not intended for extending estimates to 1-in-10,000 year events or for identifying outliers

• Multiple methods and approaches utilized at federal level to statistically extend beyond 10,000 years 85

Background:

AEP of PMP

• Schaefer (1994)

• Probability of PMP 0

• Difference between theoretical and operational upper limits

• AEP of PMP ranges from 10-5 to 10-9 AEP

• Varies by location, duration, and storm area

• Based on generalized PMP at small area sizes 86

Background:

AEP of PMP

• Why AEP estimates of PMP and PMF are conservative (i.e., too frequent)

• Generalized PMP values only valid to 20,000 mi2 - large rainfall amounts over larger area sizes less likely to occur due to physical limitations of meteorology

• Generalized PMP studies do not typically include antecedent rainfall prior to the PMP event to prime the hydrologic response -

joint probabilities lower the likelihood of the event

• Generalized PMP used in PMF studies use stationary events -

stationary events for ANO watershed reduced PMF

• PMF studies include additional conservatisms independent of the precipitation falling on the watershed 87

Background:

Site-Specific Studies

• PMP Estimates: (Applied Weather Associates)

• 80-mile per day storm hyetographs

• Basin Average Precipitation:

• 3-Day Antecedent Storm (40% PMP) = 1.72

• 3-Day Full PMP Storm (100%PMP) = 6.76

• 9-Day Storm Total = 8.48

• PMF Estimates: (GZA)

• Peak Flow ~ 1,226,000 cfs (9-Day PMP of 8.48) 88

Background:

Hydrologic Hazards Analysis

• Key Concepts of HHA: (Merz and Bloschl, WRR 2008)

• Integrated Teams

• Geologists, hydrologists, meteorologists, engineers

• Multiple Methods Approach

• Regional Analyses

• Hydrologic Modeling

• Expansion of Data (temporal, spatial, causal)

• Quantification of Uncertainty

• NRC PFHA Workshop (Jan 2013)

• Focus on state-of-the-science on probabilistic methods

• Reclamation Dam Safety Program

• Issue Evaluations and Corrective Action Studies

• Best Practices Course in HHA National Research Council(1988) Estimating Probabilities of Extreme Floods 89

Background:

Federal Products/Methods

• Precipitation Frequency:

• NOAA Atlas 14

• Regional frequency analysis using L-Moments

• State-level studies

• WA, OR, TX, IL/MO, etc.

• Reclamation Dam Safety Studies

• Friant, Altus, El Vado, others

• NRC Pilot Project

• Carolinas NOAA14 Extension

• Evaluation of Recent PMP Events

• TVA Watts Bar and Sequoyah (Schaefer)

• NOAA14 Extension 90

Background:

NOAA Atlas 14

• Available in most US locations

• Point-specific

• 5-min to 60-day durations

• Up to AEP of 10-3

• Median and 90%

confidence bounds

• Underlying annual maximum precipitation data

• Regions not specific to watersheds 91

Background:

L-Moments

• Homogeneous region of similar climatological characteristics

• Numerous stations large #

station-years of record (EIRL)

• Reduces sampling variability and enhances reliability of regional probability distribution and estimation of magnitude-frequency relationship

• Allows uncertainty analysis (Monte Carlo and Latin Hypercube) based on variability of parameters from the underlying distribution (typically GEV - use Kappa for GLO to GPA)

Hosking and Wallis (1997) 92

Multiple Methods Employed:

NOAA Atlas 14 Statistical Extension Identifies a key gauge representative of the watershed mean L-Moments Regional Frequency Analysis (precipitation)

Uses NOAA14 annual maximum time series Stochastic Storm Transposition Uses DAD information from the AWA PMP study and HMRs 93 NOAA14 Statistical Extension 94

• Steps in statistical extension:

• Identify a site representative of the region (Coldwater, KS)

• Reduction of Point to Area

• Fit polynomial regression using EV1 as explanatory variable

• Simulate from EV1 to generate annual maximum time series

• Use Kappa distribution parameters to statistically extend NOAA14 Statistical Extension 95 Factors Affecting AEP:

72-hour precipitation (not 9-day)

AEP of 72-hour PMP:

Ranges from 10-5 to < 10-7 L-moments

• ANO Watershed L-moments

• 454 Stations amounting to 20,000+ years of station record

• Equivalent independent record length = 3,433 years

• 3-day analysis for comparison to Coldwater and Stochastic Storm Transposition

• 10-day analysis for comparison to 9-day PMP 96

L-moments (3-Day)

• Conservative estimate compared to 9-Day

• AEP of PMP from <10-5 to < 10-7 97 Stochastic Storm Transposition

• Stochastic Storm Transposition

• 92 storms available from the AWA PMP study

• Compute ratios of storm magnitude to PMP at the location of occurrence

• Transpose storms into the basin through moisture adjustment

• Calculate the probability using Cunnane 98 Stochastic Storm Transposition

• Stochastic Storm Transposition

• Product of :

(1) probability of precipitation occurrence (2) probability of storm size within watershed (3) probability of watershed size storm from transposition region where:

• P(1) = Cunnane with Gringorten = (r-0.44)/N or

= Estimate from # storms vs % PMP

• P(2) = 20,000/153,366 = 0.13

• P(3) = 153,366/1,744,188 = 8.8 x 10-2 99 Foufoula-Georgiou 1989; Franchini et al. 1996; Schaefer 2013)

Stochastic Storm Transposition

• Historical storms indicate the 72-hour, 20,000 mi2 precipitation has AEP of 4.3 x 10-5

• Conservative by at least one order of magnitude due to short duration (only 3 days) and small area size (Collier et al., 2011)

• For 153,366 mi2 basin, estimated AEP of 5.6 x 10-6 100

Stochastic Storm Transposition

• Reference to 3-day L-Moments curves (aid in selection)

• Historical storms indicate the 72-hour, 153,366 mi2 precipitation value of 4.05 has return period of ~2500 years (AEP = 4 x 10-4)

• Provides cross-check to regional frequency analysis and estimates of AEP of PMP (multiple methods) 101 Summary of 3-Day Precip Frequency 102

• Multiple methods:

- PMP and estimate of from SST adjusted for 153,366 mi2 Summary of 3-Day Precip Frequency 103

• Multiple methods:

- PMP and estimate of from SST adjusted for 153,366 mi2

- Statistical extension using NOAA14 (site at Coldwater, KS)

Summary of 3-Day Precip Frequency 104

• Multiple methods:

- PMP and estimate of from SST adjusted for 153,366 mi2

- Statistical extension using NOAA14 (site at Coldwater, KS)

- L-moments watershed analysis shows similar results

Summary of 3-Day Precip Frequency

• Multiple methods:

- PMP and estimate of from SST adjusted for 153,366 mi2

- Statistical extension using NOAA14 (site at Coldwater, KS)

- L-moments watershed analysis shows similar results

- SST Results fall within uncertainty bounds 105 Summary of 3-Day Precip Frequency 106

• Multiple methods:

- PMP and estimate of from SST adjusted for 153,366 mi2

- Statistical extension using NOAA14 (site at Coldwater, KS)

- L-moments watershed analysis shows similar results

- SST Results fall within uncertainty bounds

- Used most frequent AEP from each for best estimate 9-Day more conservative due to additional antecedent rainfall L-moments (10-Day) 10-Day Results:

• Upper estimate for AEP of PMP is < 10-5

• No 9-Day available from NOAA, used 10-Day =

Conservative Estimate

SUMMARY

Day results support the AEP of PMP for the watershed is conservatively around 10-6 107 Probability of PMP (and PMF)

• Large Precipitation is a requirement for a PMF

• PMP probabilities affected by

• Large domain of the watershed

• Spatial variability of rainfall (location and orientation)

• Joint probability of sequenced events

• Sustained, conducive atmospheric conditions

• In addition, other factors must be present

• Critical soil moisture conditions

• Limited storage capacity

• Season of occurrence

• Reservoir operations

• Hydrologic parameters 108

Summary and Conclusions

• Existing estimates of the AEP of PMP and PMF are based on generalized PMP estimates for watershed sizes much smaller than Arkansas River Basin

• Joint/conditional probabilities for PMP storm and conducive hydrologic conditions result in reduced AEP of the PMF

• Multiple methods approach provides justification for selection of a best estimate and upper/lower confidence bounds

• In comparison to the PMF, the estimates are likely conservative due to the addition of factors from the hydrologic/hydraulic modeling by an order of magnitude 109 Focus on precipitation magnitude associated with:

354 (1.025*72hPMP = 6.93 inches) and 356 (1.075*72hPMP = 7.27 inches)

(as AEP)

Lower Best Estimate Upper 3-day PMP 8.26E-10 1.39E-06 1.69E-05 associated with 353.1' (3-day only) 9-day PMP 1.71E-10 2.04E-08 2.63E-06 associated with 353.8' (Antecedent + PMP) 1.025*3dayPMP 5.78E-10 1.15E-06 1.44E-05 associated with 354' (3-day only) 1.075*3dayPMP 2.99E-10 7.94E-07 1.05E-05 associated with 356' (3-day only)

SIGNIFICANCE DETERMINATION Richard Harris Emergency Planning Manager Key Objectives

• Realistic estimate of risk utilizing best available information

• Quantitatively characterize the risk significance

• Present ANO risk results to include

• External Flood

• Internal Flood 111 Methodology

• The risk analysis is fairly straightforward

• Event Tree is used to understand the sequences and determine the probabilities

• Initiating Event Frequencies as discussed in previous presentation were used 112

Methodology

• Recovery Factors

• Operator actions to mitigate impact of flooding were credited

• SHARP Decision Tree used in conjunction with HRA Calculator 113 External Flood Analysis Assumptions Plant is in CSD (RCS < 200F)

Analysis performed by partitioning the flood level External Flooding 354-356 Elevation SDC or DH, HPI, RBS, and EFW potentially failed from deficiencies Recovery Credited Portable Pump for Steam Generator Feed Service Water Flood Protection AC power available External Flooding > 356 Elevation SDC or DH, HPI, RBS, and EFW potentially failed from deficiencies Potential Recovery actions available (None Credited) 114 External Flooding Analysis

• Recovery Actions

• Three diverse actions are available

• Portable Pump for Steam Generator Feed

• Actions directed by OP-1203.048

• Previously trained operator actions

• Mockup performed to install from staging area (approximately one hour to implement)

• Service Water Pump

• Used for feeding the inventory in the Steam Generators

• OP-1202.004 (Unit 1) or OP-2202.011 (Unit 2)

• Control Room action to perform < five minutes

• Flood Protection

• EN-FAP-EP-010 - Fleet Severe Weather Response

• Barrier assumed not effective over 356 elevation

• Available time at least five days 115 External Flooding Analysis

• EVENT TREE 116

External Flooding Analysis

• Inputs

• Flood Frequency

• 354 to 356 - 1.15E-06/yr

• >356 - 7.94E-07/yr

• Recovery Actions (354-356)

• Steam Generator Makeup - Failure Probability = 0.05

• Portable Pump for SG feed

• Service Water

• Flood Protection - Failure Probability = 0.3 117 External Events

• Results

• Unit 1

• 354 to 356 External Flood = 5.34E-09/year

• >356 External Flood = 7.94E-07/year

• Total CDF External Flood from deficiencies = 7.99E-07/year

• Unit 2

• 354 to 356 External Flood = 5.34E-09/year

• > 356 External Flood = 7.94E-07/year

• Total CDF External Flood from deficiencies = 7.99E-07/year 118 Internal Flooding Analysis

• Assumptions

• Plants is operating at 100% power

• Circulating Water Piping failure is the most severe flood mechanism

• Full Range of break sizes considered (Unit 2)

• Large Break >19 diameter (~2 ft2 area) - No Operator action credited to stop the pump in first 30 minutes after break

• Mid Range 8 - 19 diameter - Operator failure to stop pump = 0.1

• Low range <8 diameter (~0.35 ft2 area) - 2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> isolation time (improbable that Operator action to stop the pump would not occur)

• Differences between Unit 1 and Unit 2, resulting in different risk results

• Differences in flood deficiencies

• Differences in plant design and equipment vulnerabilities 119 Internal Flood Analysis

• Event Tree 120

• Unit 1

• Based upon the following

• The volume of water required

• Location of deficiencies

• Due to plant design the break will flow back to the lake

• The time available for operations to stop the Circulating Water Pumps

• It is improbable that Operator actions would not be taken to stop the leak (>2 hours)

• Risk increase is negligible 121 Internal Flood Analysis Internal Flood Analysis

• Unit 2

• Flood Frequency (aggregate frequency for the range of break sizes)

• 9.03E-05/yr

• Recovery Actions

• Isolation of CW rupture

• Large breaks - no credit

• Medium breaks - 0.1 failure probability

• Small breaks - negligible risk

• Alternate portable pump for SG feed - Failure Probability = 0.05

• Service Water feed to SG - Failure Probability = 0.3

• Results

• 1.36E-06/yr 122 Overall Results

• Unit 1

• External Flooding + Internal Flooding

• CDF = 7.99E-07/year

• Unit 2

• External Flooding + Internal Flooding

• CDF = 2.16E-06/year These results reflect a risk as defined in the Significance Determination Process (SDP) as very low safety significance for Unit 1 and low to moderate safety significance for Unit 2 123 RISK ASSESSMENT

SUMMARY

Dale James Regulatory and Performance Improvement Director

Risk Assessment Summary

• Flooding evaluation performed based on current state-of-knowledge techniques endorsed by NRC and other federal agencies responsible for dam safety

• Conservative assumptions used when information was not publically available

• Probability of PMP calculated using statically robust techniques consistent with methods used by other federal agencies to make risk base decisions

• Mitigating actions available

• Utilizing these inputs provide the best available information for a accurate assessment of risk

• Unit 1 CDF = 7.99E-07/year Green

• Unit 2 CDF = 2.16E-6/year White 125 Manual Chapter 0609 -Appendix M Assessment NRC Entergy Bounding Risk 1E-4 1.44 E-5 Based on site specific upper 95% AEP for PMP resulting in 354 Defense-In-Depth No credit PMF below site grade Operator action available to 356 and beyond Flood protection adds additional protection Safety Margin No credit Current PMF below site grade - required level of protection maintained From 354 to 356 margin is threatened Above 356 margin further eroded Extent of PD Degraded condition existed on both units Extent of PD is known and corrected Degree of Degradation Equipment unavailable Degradation not a factor for current PMF Service water feed to SGs and portable pump for SG feed available to maintain core cooling 126 Manual Chapter 0609 -Appendix M Assessment Continued 127 NRC Entergy Exposure Time 1 year 1 year Mitigating recovery actions Operator recovery actions feasible Flood protection measures provide additional defense Additional qualitative circumstances Current PMF below elevation were degraded conditions are a factor ANO specific PMP/PMF calculations has considerable conservatism factored into them and would reduce the bounding risk approximately a order of magnitude Results Yellow White CLOSING COMMENTS Jeremy Browning ANO Site Vice-President

1 Arkansas Nuclear One Regulatory Conference Nuclear Regulatory Commission - Region IV Arlington, TX October 28, 2014 2

Agenda

• Introduction of Participants

• NRC Opening Remarks

• Licensee Presentation

• NRC Caucus

• Final Questions

• Closing Remarks

• Conference Adjournment

• Questions and Comments from Members of the Public

In accordance with 10 CFR 2.390 of the NRCs Rules of Practice, a copy of this letter and its enclosures will be available electronically for public inspection in the NRCs Public Document Room or from the Publicly Available Records (PARS) component of the NRCs ADAMS. ADAMS is accessible from the NRC web site at http://www.nrc.gov/reading-rm/adams.html (The Public Electronic Reading Room).

Sincerely,

/RA/

Ryan E. Lantz, Chief Project Branch E Division of Reactor Projects Docket Nos.: 50-313, 50-368 License Nos.: DPR-51, NPF-6

Enclosures:

1. ANO Presentation Slides

2. NRC Slides

3. Meeting Attendance Forms Electronic Distribution to Arkansas Nuclear One SUNSI Rev Compl.

Yes No ADAMS Yes No Reviewer Initials CHY Publicly Avail.

Yes No Sensitive Yes No Sens. Type Initials CHY SPE:DRP/E BC:DRP/E CYoung RLantz OFFICIAL RECORD COPY

/RA/

/RA/

11/24/14 11/25/14 ML14329B209

Letter to Jeremy Browning from Ryan Lantz dated November

SUMMARY

OF REGULATORY CONFERENCE TO DISCUSS SAFETY SIGNIFICANCE OF ARKANSAS NUCLEAR ONE FLOOD PROTECTION DEFICIENCIES Electronic distribution by RIV:

Regional Administrator (Marc.Dapas@nrc.gov)

Deputy Regional Administrator (Kriss.Kennedy@nrc.gov)

DRP Acting Director (Troy.Pruett@nrc.gov)

DRP Acting Deputy Director (Jason.Kozal@nrc.gov)

DRS Director (Anton.Vegel@nrc.gov)

DRS Deputy Director (Jeff.Clark@nrc.gov)

Senior Resident Inspector (Brian.Tindell@nrc.gov)

Resident Inspector (Matt.Young@nrc.gov)

Resident Inspector (Abin.Fairbanks@nrc.gov)

Branch Chief, DRP/E (Ryan.Lantz@nrc.gov)

Senior Project Engineer, DRP/E (Cale.Young@nrc.gov)

Project Engineer, DRP/E (Jim.Melfi@nrc.gov)

ANO Administrative Assistant (Gloria.Hatfield@nrc.gov)

Public Affairs Officer (Victor.Dricks@nrc.gov)

Public Affairs Officer (Lara.Uselding@nrc.gov)

Project Manager (Andrea.George@nrc.gov)

Branch Chief, DRS/TSB (Geoff.Miller@nrc.gov)

ACES (R4Enforcement.Resource@nrc.gov)

RITS Coordinator (Marisa.Herrera@nrc.gov)

Regional Counsel (Karla.Fuller@nrc.gov)

Technical Support Assistant (Loretta.Williams@nrc.gov)

Congressional Affairs Officer (Angel.Moreno@nrc.gov)

RIV/ETA: OEDO (Cayetano.Santos@nrc.gov)

ROPreports 25, 2014