ML13233A068

Monticello Nuclear Generating Plant - Response to an Apparent Violation in NRC Inspection Report 05000263/2013008 (EA-13-096)

ML13233A068
Person / Time
Site:	Monticello
Issue date:	07/11/2013
From:	Schimmel M A Northern States Power Co, Xcel Energy
To:	Document Control Desk, NRC/RGN-III
References
EA-13-096, L-MT-13-062 IR-13-008
Download: ML13233A068 (99)

See also: IR 05000263/2013008

Text

Monticello Nuclear Generating PlantXcelEnergy 2807 W County Road 75Monticello, MN 55362July 11, 2013 L-MT-13-062EA-13-096U.S. Nuclear Regulatory CommissionATTN: Document Control DeskWashington, DC 20555-0001Monticello Nuclear Generating PlantDocket 50-263Renewed Facility Operating License No. DPR-22Response to an Apparent Violation in NRC Inspection Report 05000263/2013008(EA-1 3-096)References: 1) Letter from Nuclear Regulatory Commission (NRC) to Mr. Mark A.Schimmel, "Monticello Nuclear Generating Plant, NRC InspectionReport 05000263/2013008; Preliminary Yellow Finding," dated June11, 2013 (Accession Number ML13162A776)2) Letter from Northern States Power -Minnesota to NRC, "Notificationof Intention Regarding NRC Inspection Report 05000263/2013008(EA-13-096)," dated June 19, 2013By the above referenced letter dated June 11, 2013, the NRC transmitted InspectionReport 05000263/2013008 for the Monticello Nuclear Generating Plant (MNGP). In theinspection report, the NRC identified one finding and apparent violation with apreliminary significance of Yellow for MNGP. In the referenced letter, the NRC statedthat the site had failed to maintain a flood procedure, A.6, "Acts of Nature", such that itcould support the timely implementation of flood protection activities within the 12 daytimeframe credited in the design basis, as stated in the updated safety analysis report(USAR).Northern States Power Company -Minnesota (NSPM) reviewed the apparent violationand, pursuant to the provisions of the choice letter, prepared a written response to theapparent violation. NSPM agrees that the failure to maintain a flood plan to protect thesite from external flooding events is a violation of Technical Specification 5.4.1.a.This letter submits additional information for the NRC's consideration in its finaldetermination of the significance of the apparent violation. The enclosures address thefollowing:e-7A

Document Control DeskL-MT-1 3-062Page 2 of 5Response to Apparent Violation (Enclosure 1)NSPM agrees that the failure to maintain an adequate flood plan to protect the site fromexternal flooding events is a violation of Technical Specification 5.4.1 .a. NSPM is takingthis failure to protect the site from external flooding very seriously and has used it toreinforce NSPM's policy and commitment to safety as a top priority in our EmergencyResponse plans, response to acts of nature, and effective corporate governance andoversight. The site and nuclear fleet are taking corrective actions to ensure protection ofthe radiological health and safety of the public in the event of an external flooding worstcase scenario. A summary of the corrective actions to resolve the performancedeficiency is presented in Enclosure 1.As part of those actions, NSPM is performing cultural assessments focusing on decisionmaking, effective communication, and closure follow-through not only at the site levels,but across the nuclear fleet to maximize learning from this situation.Probabilistic Risk Analysis (Enclosure 2)NSPM developed additional information providing further insight into the probability of aProbable Maximum Flood (PMF) at the Monticello site for the NRC's consideration. Thereport provides probabilistic risk analyses to support a best-estimate assessment of thesignificance of this finding as well as bounding analyses to support final significancedetermination prior to corrective actions taken by the site. The best-estimate analysisincorporated the assumptions necessary to support the assessment of a finding relatedto an external flooding event. Results are shown in the table below:Nominal, Best Sensitivity 1: Sensitivity 2:Estimate Bounding Flood SPAR-H HRAFreauencv ProbabilitiesCDF 1.04E-06 3.1 OE-06 1.83E-06ACDP 8.92E-07 2.66E-06 1.57E-06The full results of the event tree quantification are summarized in Enclosure 2.Monticello Nuclear Generating Plant Flood Protection Analysis (Enclosure 3)The postulated PMF for the MNGP is compared to other site Mississippi river conditionsin the table below. The PMF is not an instantaneous event, but rather a slowlydeveloping evolution that allows for plant staff to monitor, predict, prepare, andimplement appropriate actions to provide the required flood protection. Since actionshave been taken to procure the bin wall and levee materials, performance of areasonable simulation demonstrated that the levee and bin wall system can now beinstalled within the available time as defined in the licensing basis.

Document Control DeskL-MT-1 3-062Page 3 of 5Normal and Flooded River Flow Rates and Water ElevationsMississippi River Flow Rate (cfs) Water ElevationCondition (ft. msl)Normal 4,600 905Maximum Recorded 51,000 916(1965)1000 Year Flood -90,000 (1) 921Probable Maximum 364,900 939.2Flood 364,900_ 939.2A report entitled "Monticello Flood Protection," was prepared for Monticello andaddresses the aspects of flood protection for which MNGP was licensed and is includedin Enclosure 3.Annual Exceedance Probability (Enclosure 4)Annual river exceedance probabilities based on annual peak flood estimates at theMonticello site were developed to support the probabilistic risk assessment. Theprobability of a PMF at the site was determined to be extremely low.Enclosure 4 provides a copy of the report entitled, "Annual Exceedance ProbabilityEstimates for Mississippi River Stages at the Monticello Nuclear Generating Plant basedon At-site Data for Spring and Summer Annual Peak Floods", June 28, 2013, developedby RAC Engineers & Economists.Stakeholder Outreach (Enclosure 5)NSPM hosted an open house to share information with its community neighbors on itsoperations and preparedness to handle potential emergencies and how it would respondto flooding, earthquakes and other unforeseen challenges.The key message presented to visitors was that safety and security at the NSPM nucleargenerating plants are top priorities for Xcel Energy. Further, that we understand theindustry, NRC, and public's demand of higher safety standards and flood preparednessat the nation's nuclear power plants in the wake of events such as 9/11 and FukushimaDaiichi. The Monticello Flood Protection Strategy was identified and explained todemonstrate that the site is capable of withstanding a PMF and that the site isincorporating lessons learned from the industry to improve and assure protectionmethods.Safety Culture ReviewNSPM agrees it missed an opportunity within its control to identify challenges to theimplementation of the A.6 procedure, leading to the identified apparent violation. Assuch, NSPM assembled an expert panel to examine the behavioral and cultural aspectsimpacting decision making within the nuclear business unit. This activity was chartered

Document Control DeskL-MT-13-062Page 4 of 5as an immediate and interim measure preceding the extensive root cause evaluationthat will be performed to identify the full magnitude of this issue, associated causes, andcorrective actions to prevent recurrence.This expert panel assembled to examine safety culture within its nuclear organizationwas comprised of five independent consultants and one Xcel representative. The teamreported directly to the Vice President of Nuclear Operations Support. A phasedapproach is being utilized to examine the behavioral and cultural aspects impactingdecision making within the nuclear business unit. Three phases are planned to examinethis subject: Phase (1) is specifically focused on the Monticello flooding issue, Phase (2)more broadly examines Monticello issues and Phase (3) examines Prairie Island issues.While the phases are specific to the individual sites, the scope includes developing anunderstanding of the corporate culture and influence beyond a site-centric examinationof the behavioral and environmental influences. To date Phase (1) has been completedwith scheduling of Phase (2) and (3) expected to commence and complete over the nextfew months. The initial phase identified improvement opportunities in the areas ofdecision making, leadership behaviors, and questioning attitude regarding the station'spreparedness for a PMF. The results of this assessment have been insightful and will beapplied across the nuclear fleet to ensure a healthy safety culture exists. Opportunitieshave been identified to strengthen fleet and Nuclear Oversight accountability forproviding oversight to proactively detect performance gaps.Interim actions are in place for the short term to focus on the areas for improvement, andlonger term actions are in development.SummaryNSPM respectfully requests that the NRC consider the enclosed information in its finaldetermination of the significance of the finding. Notwithstanding our assessment of thesignificance of the finding, NSPM clearly understands our performance shortcomingsconcerning flood protection for the entire spectrum of possible flooding events at theMNGP. Corrective actions have already been completed to address the NRC's identifiedperformance deficiency. Additionally, we unequivocally acknowledge the need for overallperformance improvement at MNGP. Actions are underway to ensure that the lessonslearned from this finding are applied more broadly to overall performance.Summary of CommitmentsThis letter contains no new commitments and no revisions to existing commitments.Mark A. SchimmelSite Vice-PresidentMonticello Nuclear Generating PlantNorthern States Power Company-Minnesota

Document Control DeskL-MT-1 3-062Page 5 of 5Enclosures:Enclosure 1 -Response to Apparent ViolationEnclosure 2 -External Flooding Evaluation for Monticello NuclearGenerating PlantEnclosure 3 -Monticello Flood ProtectionEnclosure 4 -Annual Exceedance Probability EstimatesEnclosure 5 -Stakeholder Outreachcc: Regional Administrator, Region III, USNRCProject Manager, Monticello Nuclear Generating Plant, USNRCResident Inspector, Monticello Nuclear Generating Plant, USNRC

Enclosure IResponse to Apparent ViolationEA-1 3-096NRC Inspection Report 05000263/2013008Monticello Nuclear Generating Plant3 Pages Follow

Northern States Power Company -MinnesotaResponse to Preliminary Yellow FindingNRC Finding SummaryThe inspectors identified a preliminary Yellow finding with substantial safety significance andassociated apparent violation (AV) of Technical Specification 5.4.1 for the licensee's failure tomaintain a flood plan to protect the site from external flooding events. Specifically, the site failedto maintain flood Procedure A.6, "Acts of Nature," such that it could support the timelyimplementation of flood protection activities within the 12 day timeframe credited in the designbasis as stated in the updated safety analysis report (USAR).The inspectors determined that the licensee's failure to maintain an adequate flood planconsistent with the USAR was a performance deficiency, because it was the result of the failure tomeet the requirements of TS 5.4.1 .a, "Procedures;" the cause was reasonably within thelicensee's ability to foresee and correct; and should have been prevented. The inspectorsscreened the performance deficiency per Inspection Manual Chapter (IMC) 0612, "Power ReactorInspection Reports," Appendix B, dated September 7, 2012, and determined that the issue wasmore than minor because it impacted the 'Protection Against External Factors' attribute of theMitigating Systems Cornerstone and affected the cornerstone's objective to ensure theavailability, reliability, and capability of systems that respond to initiating events to preventundesirable consequences (i.e. core damage). Specifically, if the necessary flood actions cannotbe completed in the time required, much of the station's accident mitigation equipment could benegatively impacted by flood waters.NRC Baseline Significance Determination Process ReviewAs part of the process, the Region III Senior Reactor Analyst (SRA) developed an event treemodel to perform a bounding quantitative evaluation. The model presents an external flood eventthat exceeds grade level (930 ft. MSL) and requires implementation of Procedure A.6, "Acts ofNature" Section 5.0.NSPM ResponseNSPM agrees that a performance deficiency exists. Procedure A.6, "Acts of Nature", at the time ofthe violation, did not provide sufficient guidance to execute mitigation strategies for a probablemaximum flood (PMF) event. Adequate management oversight and engagement was notprovided to ensure that the Monticello external flood mitigation procedure and strategies metexpected industry standards and licensing basis requirements.Actions have been completed to reduce the flood mitigation plan timeline by pre-stagingequipment and materials required for bin-wall levee construction, improving the quality of theA.6 "Acts of Nature" procedure and pre-planning work orders necessary to carry out the A.6actions.Summary of Corrective Actions:Acquired materials required for flood mitigation including, but not limited to:" Hardware and components for construction of Bin-Wall" Clay for levee construction (30000 cubic yards)* Rip-Rap stone for levee construction (1700 cubic yards)Page 1 of 3

• Sand for levee construction and filling sandbags (11000 cubic yards)" Sand bagging machine (Capacity 1600 sand bags/hr)* Manual sand bag filling tools (25 on site)" Gas Sump Pumps* Electric Sump Pumps" Crushed concrete for alternate road access (2400 cubic yards)* Preventative maintenance plans are being developed for new flood mitigation equipment* Performance of reasonable simulation of major steps required by procedure A.6 "Acts ofNature" Section 5.0, including building of bin-wall sections, sandbagging, placement ofvarious covers, and relocation of vital equipment.* Extensive procedure revisions to enhance feasibility of actions and reduce overall timerequired to execute the strategy.* Table top exercises of new revisions performed to ensure practicality.* Development of work orders to provide more detail for execution of steps within procedureA.6, "Acts of Nature" Section 5." The existing flood prediction surveillance was revised to occur on a monthly basis insteadof yearly and contains provisions to continually monitor river predictions if certainconditions are met.* Meetings with the National Weather Service were held to develop more robust predictioncapabilities and options.* Enhanced construction drawings of levee and bin-wall to provide more detail* Updated existing contracts and memorandums of understanding with vendors to assureequipment availability.* A modification is also in the design phase to install the base of the bin-wall on the westside of the Intake structure, simplify construction on the east side of the Intake Structure,and also update the steel plate design for protection of the Intake Structure.Review of NRC Significance DeterminationNSPM has developed additional information providing new insight into the probability of a PMF atthe Monticello site for your consideration. A report entitled "External Flooding Evaluation forMonticello Nuclear Generating Plant" was prepared by Hughes Associates, Inc., for NSPM. Thereport provides a best-estimate assessment of the significance of this finding. Two (2)sensitivities were performed to assess the bounding risk, addressing some of the uncertaintyassociated with this assessment.The first sensitivity study provides the risk assessment if a bounding annual exceedanceprobability is assumed. As noted in the Hughes' report, the uncertainty associated with extremeflooding can be addressed by artificially restraining the AEP to a value of no less than 1 E-05/year.When this restraint is assessed, the ACDP is 2.66E-06.Enclosure 4 of this letter provides a copy of the report entitled, "Annual Exceedance ProbabilityEstimates for Mississippi River Stages at the Monticello Nuclear Generating Plant based on At-site Data for Spring and Summer Annual Peak Floods", June 28, 2013, developed by RACEngineers & Economists.The second sensitivity address the different methodologies available for quantifying the HumanError Probability (HEP) associated with the manual operation of RCIC (Reactor Core IsolationCooling) and HPV (Hard Pipe Vent). This sensitivity provides quantification using The SPAR-HPage 2 of 3

Human Reliability Analysis Method, NUREG/CR-6883. When the simplified SPAR-Hmethodology is used, the assessment results in a ACDP of 1.57E-06.The result of the nominal, best-estimate assessment and the two (2) sensitivities performed areshown in the table, below, for ease of reference.Sensitivity 1:Bounding FloodFrequencySensitivity 2: SPAR-HHRA ProbabilitiesNominal Best-EstimateCDF 1.04E-06 3.1OE-06 1.83E-06ACDP 8.92E-07 2.66E-06 1.57E-06Enclosure 2 of this letter provides a copy of a report entitled, Report Number 1SML16012.000-1,"External Flooding Evaluation for Monticello Nuclear Generating Plant," developed by HughesAssociates for consideration.Page 3 of 3

Enclosure 2Monticello Nuclear Generating Plant"External Flooding Evaluation for Monticello Nuclear Generating Plant"ISMLI16012.000-1Hughes Associates62 Pages Follow

.IHUGHESEASSOCIATESENGINEERS CONSULTANTS SCIENTISTSExternal Flooding Evaluation for MonticelloNuclear Generating Plant1SML16012.000-1Prepared for:Xcel EnergyProject Number: 1SML16012.000Project Title: Monticello External Flooding SDPRevision: 1Name DatePreparer: Erin Collins/Paul Amico/Suzanne Loyd 7/8/2013~ ~" Erin P. Collins2013.07.082018:01 -04-00-Reviewer: Pierre Macheret 7/8/2013P.",: 2"13 007 .M0 1-. o U.Review Method Design Review E] Alternate Calculation E]Approved by: Francisco Joglar Francisco Joglar. ,. 7/8/2013N. I l: "013.07-

ISML-16012.000-1 Table of ContentsTABLE OF CONTENTS1.0 INTRODUCTION .................................................................................................. 12.0 REFERENCES ................................................................................................. 23.0 METHODOLOGY and analysis ........................................................................ 43.1 Event Tree Analysis ............................................................................. 43.1.1 Evaluation of Flood Frequency ................................................... 43.1.2 Early W arning Probability ........................................................... 73.1.3 Protection of the Reactor Building ............................................... 73.1.4 Manual Local Operation of RCIC and the Hard Pipe Vent .......... 84.0 CONCLUSIONS ............................................................................................. 15APPENDICES TABLE OF CONTENTS ................................................................... 16Revision I Page ii

1SML16012.000-1Introduction1.0 INTRODUCTIONThis analysis was developed to address the significance of a finding that was received by theMonticello Nuclear Generating Plant (MNGP) associated with External Flooding hazards. ThisSDP is summarized in NRC Letter EA-13-096 (Reference 5). The analysis quantifies the coredamage frequency associated with a flood exceeding 930' at MNGP.Revision 2 Page 1Revision 2Page1I

ISML16012.000-1References2.0 REFERENCES1. Annual Exceedance Probability Estimates for Mississippi River Stages at the MonticelloNuclear Generating Station based on At-Site Data for Spring and Summer Annual PeakFloods, David S. Bowles and Sanjay S. Chauhan, RAC Engineers and Economists, June28, 2013.2. Hydrologic Atlas of Minnesota, Division of Water, Department of Conservation, State ofMinnesota, 1959.3. US Geological Survey, Guidelines for Determining Flood Flow Frequency, Bulletin#17B, Hydrology Subcommittee, Interagency Committee on Water Data, Office of WaterData Coordination, 1982.4. A Framework for Characterization of Extreme Floods for Dam Safety Risk Assessments,Robert E. Swain, David Bowles and Dean Ostenea, Proceedings of the 1998 USCOLDAnnual Lecture, Buffalo, New York, August 1998.5. NRC Letter EA-13-096, Subject: Monticello Nuclear Generating Plant, NRC InspectionReport 05000263/2013008; Preliminary Yellow Finding, United States NuclearRegulatory Commission, Region III, 11 June 2013.6. A Preliminary Approach to Human Reliability Analysis for External Events with a Focuson Seismic, EPRI 1025294, EPRI, December 2012.7. Interim Staff Guidance for Performing the Integrated Assessment for External Flooding,Appendix C: Evaluation of Manual Actions, JLD-ISG-2012-05, Revision 0, U.S. NRCJapan Lessons-Learned Project Directorate, November 30, 2012.8. Job Performance Measure JPM-A.8-05.01-001, Manual Operation of RCIC, Rev. 0, TaskNumber NL217.108 -Operation of RCIC without Electric Power, 3 timed exercisesperformed 17 June 2013.9. Job Performance Measure JPM-A.8-05.08.001, Manually Open Containment Vent Lines,Rev. 0, Task Number FB008.007, 4 timed exercises performed 17 June 2013.10. Hughes Associates Record of Correspondence, RCIC Manual Operation e-mails withXcel Energy during June -July 2013, Hughes Associates, Baltimore, MD, 7 July 2013.11. Hughes Associates Record of Correspondence, Hard Pipe Vent Manual Operation e-mails with Xcel Energy during June 2013, Hughes Associates, Baltimore, MD, 7 July2013.12. Monticello Procedures:" 8900, OPERATION OF RCIC WITHOUT ELECTRIC POWER, Revision 2" C.5-1 100, RPV CONTROL flowchart, Revision 11" C.5-1200, PRIMARY CONTAINMENT CONTROL flowchart, Revision 16" C.5-3505-A, Revision 10* A.6, ACTS OF NATURE, Revision 43* A.8-05.08, Manually Open Containment Vent Lines, Revision 1Revision 2Page 2

ISML16012.000-1References0 A.8-05.01, Manual Operation of RCIC, Revision 213. The EPRI HRA Calculator Software Users Manual, Version 4.21, EPRI, Palo Alto, CA,and Scientech, a Curtiss-Wright Flow Control company, Tukwila, WA.14. NUREG/CR-1278, Handbook of Human Reliability Analysis with Emphasis on NuclearPower Plant Applications, (THERP) Swain, A.D. and Guttman, H.E., August 1983.15. NUREG-1921, EPRI/NRC-RES Fire Human Reliability Analysis Guidelines, DraftReport for Public Review and Comment, November 2009.16. NUREG- 1852, Demonstrating the Feasibility and Reliability of Operator Manual Actionsin Response to Fire, October 2007.17. NUREG/CR-1278, Handbook of Human Reliability Analysis with Emphasis on NuclearPower Plant Applications, (THERP) Swain, A.D. and Guttman, H.E., August 1983.18. Whaley, A.M, Kelly, D.L, Boring, R.L. and Galyean, W.J, "SPAR-H Step-by-StepGuidance", INL/EXT-10-18533, Revision 2, Idaho National Laboratory, Risk,Reliability, and NRC Programs Department, Idaho Falls, Idaho, May 2011.Revision 2 Page 3Revision 2Page 3

1SML16012.000-1Methodology and Analysis3.0 METHODOLOGY AND ANALYSIS3.1 Event Tree AnalysisEvent trees were developed to calculate the core damage frequency (CDF) associated withExternal Floods. Event trees were developed for both a best estimate case as well as sensitivitycases. Floods were evaluated at three particular heights -917' but less than 930', 930' but lessthan 935', and greater than or equal to 935'. The first flood height range (917' to 930') wasevaluated for frequency but was not deemed feasible to cause core damage since much of theplant's critical safety equipment is not threatened unless flood levels exceed the 930' elevation.The following sections describe the development of the flood frequency and conditionalprobabilities of the other events that may lead to core damage events.Additionally, it is agreed that, as stated in the NRC letter EA- 13-96 (Reference 5), the evaluationof large early release frequency (LERF) risk is assumed to be no more significant than CDF-based risk; therefore, no evaluation of LERF was performed for this analysis.3.1.1 Evaluation of Flood FrequencyMNGP teamed with industry experts to evaluate the frequency of major floods in detail. Dr.David Bowles was the lead author for the flood frequency analysis. To ensure that the floodfrequency of exceedance analysis was sound, an independent review of the analysis wasperformed as part of this analysis (Reference 1). The review resulted in some recommendationsfor improvement that were incorporated by Dr. Bowles and his team prior to the development ofthe event trees that are included in this report. This section documents the results of that finalflood frequency analysis.The frequencies of Mississippi River floods at MNGP for flood heights of 917', 930' and 935'were estimated using stream-flow data from 1970 -2012. This estimation is described in"Annual Exceedance Probability Estimates for Mississippi River Stages at the MNGP based onAt-Site Data for Spring and Summer Annual Peak Floods" (Reference 1).Separate flood frequency relationships were developed for spring and summer annual peakfloods based on the Hydrologic Atlas of Minnesota (Reference 2) and an examination of flowrecords.a) Spring annual peak floods generally peaked in the period March to May, but if it wasclear from examination of the hydrograph that a snow melt flood event peaked in Junethen that peak was used.b) Summer annual peak floods generally peaked in June to October, but flood peaks in Juneassociated with snow melts were excluded since they were assigned to spring floods.Frequencies were estimated based on extrapolation of flood frequency relationships developedfrom at-site flow data. The mean daily annual peak stages were obtained for spring and summerannual peak floods. The daily annual peak stages were converted to daily annual peak flows forspring and summer floods using a combined rating curve described in Reference 1. TheExpected Moments Algorithm (PeakfqSA) was applied using methodology described in thecurrent draft of the upcoming revision to USGS Bulletin 17B (Reference 3). The EMA softwareprovided annual exceedance probability (AEP) estimates down to 1 in 10,000/yr. Annual peakdischarge was plotted on a Log scale and AEPs on a z-variate scale (corresponding to a NormalRevision 2Page 4

ISML16012.000-1Methodology and Analysisprobability distribution) to allow AEPs lower than 1 in 10,000/yr to be estimated by linearextrapolation.The resulting median estimates for spring and summer floods at 917', 930' and 935' river heightswere as follows:Table 3-1 Median Flood Frequency Estimates at MNGP Site (Ref. 1)Elevation 917' Elevation 930' Elevation 935'Spring Floods 6.3E-3/yr <lE-9/yr <1E-9/yrSummer Floods 1.7E-3/yr <1E-9/yr <1E-9/yrA full family of flood hazard curves is provided in the Bowles report. There is a substantialamount of aleatory uncertainty in the development of these curves, resulting in very wideconfidence bounds as can be seen in the figures and tables in the report, especially for the floodlevels of interest. For example, the elevation 917' 95th percentile estimate is 3.4E-2/yr and the5th percentile estimate is 5.9E-7/yr for spring floods. This is indicative of the limited dataavailable in the 1970 -2012 year period. In addition, not all potential flood influences were seenin the 42 years of experience. Moreover, there is additional epistemic uncertainty due tosimplifications used in the modeling process, For example, regional flood impacts were notconsidered, nor were the potential effect of ice blockage or changes in flood protective featuressuch as dikes along the river. To address these as well as other potential flood contributors, themean hazard for the purposes of quantification will be represented by the 84th percentile, whichis the median plus one standard deviation for a lognormal form distribution. This is a commonpractice to provide margin to account for the uncertainties.As a further consideration, there is a general consensus that the practical limit on AEPextrapolation is no better than 1.OE-5/yr no matter how much information, including informationon paleofloods, is available (see for example, "A Framework for Characterization of ExtremeFloods for Dam Safety Risk Assessments" (Ref. 4). It is believed that all of these factors wouldtend to increase the estimated frequency of floodings. Some of these uncertainties could beaddressed in a more detailed analysis, such as a Monte Carlo rainfall-runoff approach, but others,such as the potential effect of ice blockage, would remain. In the case of this evaluation, thereadditional factors will be addressed by way of a sensitivity study, discussed at the end of thissection.The 84th percentile is the median plus one standard deviation, which is a common "marginvalue" that addresses uncertainty without being overly conservative. The margin is to cover thewide statistical uncertainty in the distribution plus the modeling uncertainty that comes fromsome of the issues we discussed, such as not considering ice blockage or other downstream flowbottlenecks, only considering site data, and not actually modeling the flows and performing asimulation. This results in the following flood frequency estimates:The results presented in the Bowles report support the use of the 84th %-tile as a reasonableconservative, but not overly conservative, representation of the mean. A manually generatedapproximation of the mean (described in Ref. 1 -the code used cannot itself generate a mean) isshown to be in the range of the 70th percentile, plus or minus about 10 percentile. This is theresult of the aleatory uncertainty. The use of the 84th percentile provides some additionalRevision 2Page 5

1SM L16012.000-1Methodology and Analysismargin to account for epistemic uncertainty and give confidence that the "true mean" is notlikely to exceed this value.Treating the 84th percentile as the mean, the process of developing the initiating eventfrequencies is straightforward. It is typical in external hazard PRA to create initiating events bydiscretizing the hazard curve. Because for external flooding there are a clear series of floodlevels of concern, the selection of the initiating events is clear. For MNGP, the levels of concernare 917', 930', and 935', with the exceedance of each level causing the same impact on plantsystems until the flood exceeds the next level of concern. Therefore the initiating events can bedefined as follows:" IE1 -level >917' and <930'" IE2 -level >930' and <935'" IE3 -level >935'Using the 84th percentile values to represent the mean, we get exceedance probabilities for eachlevel of concern as follows:Table 3-2 84th Percentile Exceedance Frequencies (Used as Means) (Ref. 1)Elevation 917' Elevation 930' Elevation 935'Spring Floods. 2.0E-02/yr 7.5E-08/yr <1E-09/yrSummer Floods 8.8E-03/yr 8.8E-06/yr 9.7E-07/yrTotal 2.9E-021yr 8.9E-061yr 9.7E-071yrThe values for spring and summer can be summed because they are each calculated on a per-calendar-year basis and because we are using these values to represent the "true mean" of thedistribution, albeit conservatively.Because these values are exceedance (i.e., the frequency that a flood exceeds a specified level),the way to determine the frequency of a flood in a given range is to subtract the frequency ofexceedance of the upper flood in the range from the frequency of exceedance of the lower floodin the range.One event tree was developed to account for all floods that were greater than 930' (includingthose greater than 935'). The frequency of exceedance for the 930' floods was used and aconditional probability was applied to account for the floods within the 930' to 935' range andfloods greater than 935'. The frequency shown as IE2 in Table 3-3 was used in the event tree.The conditional probability that the flood was indeed greater than 935' was then applied in asubsequent branch to determine the CDF for sequences specific to floods greater than 935'(value of 0.109). Table 3-3 below shows the initiating event frequencies used for thisassessment.Revision 2 Page 6Revision 2Page 6

1 SML16012.000-1Methodology and AnalysisI SMLI 6012.000-1 Methodology and AnalysisTable 3-3 Best Estimate Initiating Event FrequenciesInitiating Event Initiating Event FrequencyIEl1 2.9E-02/yrIE22 8.90E-06/yrNot deemed feasible for core damage sequences. Not evaluated for this analysis.2This frequency Is represented in the event tree along with a conditional probability that the flood will be >935' toevaluate the correct range of flood heights.As discussed earlier, there is the question of the practical limit of extrapolation of flooding data.Therefore, in addition to the initiating event frequencies presented above, a sensitivity analysiswas performed limiting the flood frequency to no less than 1.OE-5/yr, consistent with theconsensus reached in Ref. 4 concerning the limit of credible extrapolation for annual floodexceedance probability.Table 3-4 Sensitivity Case Initiating Event FrequenciesInitiating Event Initiating Event FrequencyIE1 2.9E-02/yrIE22 2.OE-05/yr'Not deemed feasible for core damage sequences. Not evaluated for this analysis.2This frequency is represented In the event tree along with a conditional probability that the flood will be >935' toevaluate the correct range of flood heights.It could be argued that the limit should apply across the 930' events (i.e., that the hazard curvegoes flat at 1E-5/yr) and that it is sufficient to ignore the 935' event (because F(IE2930<=1v1<935) =fexceed(930) -fcxceed(935) = 0) and simply evaluate the 930' flood at a frequency of lE-5/yr. Sincethis is a sensitivity case and not the best estimate, it was decided to assign 2E-5/yr to IE2 forpurposes of providing the sensitivity insights and use a conditional probability of 0.5 to accountfor floods that are >935'.In the event trees (shown in Appendix A), the flood frequencies are represented by the headings"EXTERNAL FLOOD >930"' and "<935.' for IE2 events. In addition, event trees used forsensitivities have similar headings.3.1.2 Early Warning ProbabilityAlthough it was considered qualitatively, there was no basis determined for giving credit to thepotential for early warning of a flood >930'. The event tree model shows a failure probability of1.0 for this node.3.1.3 Protection of the Reactor BuildingIn the event of a major flood, such as those evaluated in this analysis, MNGP plans to build a binwall barrier that will protect the site from the high flood waters. If the bin wall levee issuccessful, the safety equipment needed to prevent core damage will be protected, providingdefense in depth for each required critical safety function. Simple flood protection measuresmay be taken to protect the reactor building from floodwater, even if the bin wall levee fails.These measures to protect the reactor building do not need to be in place until the flood heightapproaches the 935' elevation. A conservative value of 0.11 is assigned, consistent with the ES-13-096 NRC letter (Reference 5), to the failure probability of protecting the reactor building.Revision 2 Page 7Revision 2Page 7

1SML16012.000-1Methodology and Analysis3.1.4 Manual Local Operation of RCIC and the Hard Pipe VentMNGP requested Hughes Associates to conduct a detailed human reliability analysis (HRA) toestimate human error probabilities for operator manual actions to prevent core damage throughthe use of RCIC as a high pressure injection source and the Hard Pipe Vent (HPV) to removedecay heat from containment. This detailed HRA was based upon thermal/hydraulics analyses,battery depletion calculations, several Job Performance Measure (JPM) exercises for the specificprocedure A8.05.01 and A8.05.01 actions (References 8 and 9) considering flood and StationBlackout (SBO) conditions, and discussions with Operations and PRA staff (References 10 and11). This section documents the initial conditions assumed, the analytical process, and theresults of the detailed HRA. A comparison between the detailed analysis results and thoseobtained using SPAR-H is also provided as a sensitivity evaluation.3.1.4.1 Initial Conditions for the Operator Manual ActionsOperations staff at MNGP provided the following information on the conditions that would beevolving leading up to the need for the postulated operator manual actions evaluated in thisHRA.The A.6 Procedure, "Acts of Nature" (Reference 12) that addresses External Flooding directs de-energizing the 115 KV, 230 KV and 345 KV substations for flood levels in excess of the 930'elevation. In the case where the flood level is above 930' this would lead to a loss of offsitepower and reliance on the EDGs. Since the normal long term fuel oil storage (Tank T-44) for theEDG's is not evaluated to survive flood levels above 932' elevation, and alternate fuel oil makeupmethods are not pre-prescribed to support the EDG's, long term EDG operation is not easilydefensible utilizing existing procedures. Additionally, there are several penetrations in the PlantAdministration Building that would make positive flood proofing of the building difficult,leaving three of the four station batteries vulnerable to flooding.The flooding engineer provides daily updates to the station on high river water levels includingpotentials to rise above any trigger points from the A.6 procedure. At this point, heightenedawareness of the potential for flooding is implemented.When river level exceeds 921 feet an evaluation of EALs would be performed. If visible damagehas occurred due to flood water rising greater than 921 feet, then an Alert per EAL HA1.6 wouldbe declared.Prior to the river reaching these levels, operators would be walking down the procedures foralternate methods to vent primary containment and operate RCIC remotely. This would involvestaging of equipment in the torus area to open the Hard Pipe Vent and verification thatequipment is properly staged to operate RCIC remotely.As water level reached the 930' elevations, Operations would prepare for isolation of off-sitepower and loading essential loads onto the emergency diesel generators as needed to conservefuel. Only a single EDG is required for shutdown cooling and inventory makeup. Operatorswould be in the EDG rooms, intake and other critical areas ensuring no water intrusion andwould be pumping water out of the room as needed. Also the battery rooms in the PAB wouldbe of concern due to the high flood levels. At this point operators would be briefed to be readyto operate RCIC without electrical power as necessary.Revision 2Page 8

1 SML16012.000-1Methodology and AnalysisThe portable diesel fire pumps would also be staged at higher locations with hoses stagedthrough higher elevations of the buildings to support alternate RPV makeup.Operators would be dispatched with I&C technicians to the 962' elevation of the reactor buildingto install the temporary level indication per procedure A.8-05.01. This would allow for alternatelevel indication to be available in the event that the batteries are lost.In the response scenario postulated for the performance of the operator manual actions evaluatedin this HRA, EDGs and batteries are not available. Shutdown cooling, HPCI, and RCIC are notavailable from normal electrical means. RCIC is available for manual operation.Operators would have temporary level indication set up in the reactor building. Pressureindication is available in the direct area of the level transmitters. The building is dark and mostlikely there is water in the basement of the reactor building. Additional portable lights areavailable to assist with lighting and boots staged for higher water. The operators would utilizeprocedure A.8-05.01 to un-latch the governor from the remote servo linkage and throttle steamflow to RCIC to start the turbine rolling while coordinating with operators monitoring waterlevel and reactor pressure. Upon reaching the high end of the level band the operators wouldthrottle closed the steam admission valve and await direction to re-start RCIC. Local operationof RCIC is demonstrated each refueling outage during the over speed test. Operation of acoupled turbine run is less complex because the turbine is easier to control with a load.3.1.4.2 Detailed HRAThe human error probabilities (HEPs) for manual local operation of RCIC and the Hard PipeVent during an extreme flooding and SBO event have been developed in detail utilizing theEPRI HRA Calculator (Reference 13). Event RCICSBOFLOOD (Fail to manually operateRCIC during SBO and extreme flooding conditions), and event HPVSBOFLOOD (Fail tooperate the HPV using N2 bottles to provide containment heat removal during SBO/Flood) havevalues of 9.3E-02 and 1.3 E-02 respectively, for a combined value of 1.06E-01.The HRA Calculator reports for these events, which provide a detailed basis for these HEPestimates, are presented in Appendix A. Section A. 1 documents RCICSBOFLOOD andSection A.2 documents HPVSBOFLOOD.The cognitive portion of these HEPs was developed using the Cause Based Decision TreeMethod (CBDTM) (Reference 13) and the execution portion utilized the Technique for HumanError Rate Prediction (THERP) (Reference 14). These methods are commonly used in internalevents PRA, have been cited in the EPRI/NRC-RES Fire HRA Guidelines, NUREG-1921(Reference 15) and applied in many fire PRAs, and are also cited in the following reference:* A Preliminary Approach to Human Reliability Analysis for External Events with a Focuson Seismic, EPRI 1025294, EPRI, December 2012.NUREG-1921 and THERP are also listed as references to the following document:* Interim Staff Guidance for Performing the Integrated Assessment for External Flooding,Appendix C: Evaluation of Manual Actions, JLD-ISG-2012-05, Revision 0, U.S. NRCJapan Lessons-Learned Project Directorate, November 30, 2012.Revision 2 Page 9Revision 2Page 9

ISML16012.000-1Methodology and AnalysisAppendix C of the ISG on External Flooding concentrates primarily on demonstrating thefeasibility and reliability of the manual actions consistent with NUREG-1852 (Reference 16),"Demonstrating the Feasibility and Reliability of Operator Manual Actions in Response to Fire."One of the key variables in determining feasibility of operator manual actions is timing. Asstated in the ISG, "For an action to be feasible, the time available must be greater than the timerequired when using bounding values that account for estimation uncertainty and humanperformance variability." In order to assess this feasibility, the following process isrecommended in section C.3.2.4 Calculate Time Margin:"The licensee should calculate the time margin available for the action using the values for timeavailable and time required that have been developed for the analysis."The time margin formula provided is:[(Tsw-Tdelay)--(Tcog+Texe))Time Margin = (Tcog.Texe) X 100%The terms of the equation are defined as follows:Tdelay = time delay, or the duration of time it takes for the cue to become available that indicatesthat the action will be necessary (assumes that action will not be taken in the absence of a cue);Tsw = the time window within which the action must be performed to achieve its objective;Tcog = cognition time, consisting of detection, diagnosis, and decision-making; andTexe = execution time including travel, collection of tools, donning of PPE, and manipulation ofrelevant equipment.The HEP calculations performed for the RCIC and HPV manual actions involved the estimationof the time parameters cited in the Time Margin formula. These times are based uponthermal/hydraulics analyses, battery depletion calculations, several Job Performance Measure(JPM) exercises for the specific proceduralized actions and considering flood and SBOconditions, and discussions with Operations and PRA staff.Using this information, the time margin for these two events is calculated as:Table 3-5 Time Margin for HFEsTimes (in minutes) RCIC HPVTsw 478.2 900Tdelay 345 330Tcog 10 10Texe 80 45Time Margin 48% 936%The HEP timing estimates therefore meet the criteria that "using the calculation under C.3.2.4,the margin must be a positive percent value for an action to be deemed feasible."Revision 2Page 10

1SML16012.000-1Methodology and AnalysisThe External Flooding ISG also recommends that estimates of time available and time requiredshould account for sources of uncertainty and human performance variability. The timingestimates shown above are already believed to be conservative. For example, the Tdelay of 345min for the RCIC event is based on 5.75 hours8.680556e-4 days <br />0.0208 hours <br />1.240079e-4 weeks <br />2.85375e-5 months <br /> until RCIC battery depletion. This can beconsidered conservative since, in reality, it is likely that an action would be taken before waitingfor battery depletion. For the Texe values, the highest observed time in the JPM trials was usedfor the RCIC case and further time was added for transit time for actions in various locations. Forthe HPV case, the Texe value was also based on JPM trial data and the results above show thatuncertainties are covered by a significant time margin.In addition to feasibility, the reliability of the actions was evaluated through the detailed HRACalculator analysis used to quantify HEPs.Section C4 of the ISG says that for an action to be deemed reliable, "sufficient margin shouldexist between the time available for the action and the time required to complete it. This marginshould account for: (1) limitations of the analysis (e.g., failure to identify factors that may delayor complicate performance of the manual action); and (2) the potential for workload, timepressure and stress conditions to create a non-negligible likelihood for errors in taskcompletion... A simplified alternative criterion for determining if the margin is adequate to deeman action as reliable is to establish that the margin is not less than 100%. Such a margin may bejustified when recovery from an error in performing the action could be accomplished byrestarting the task from the beginning."As shown above, the time margin for the HPV manual action is greater than 100% and the timemargin for the RCIC manual action is estimated at around 50%. The evaluation of PRA andOperations staff in performing the JPMs for these actions was that there would be 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> ormore to perform the procedures, allowing several opportunities to troubleshoot and/or re-performsteps if necessary. In addition, based on the HEP quantification, the combined probability for theRCIC and HPV actions is estimated to have a success rate of 89%. A sensitivity study wasperformed using SPAR-H estimates to evaluate the effect of reliability uncertainty.Regarding the evaluation of factors such as workload and stress that could complicate taskperformance, performance Shaping Factors (PSFs) were considered for feasibility and addressedin the assessment of reliability of the operator manual actions.The following table discusses each of the PSF categories cited in Appendix C of the ISG, theissues related to the RCIC and HPV manual actions and how they were addressed in thequalitative and quantitative analysis of these human failure events.Revision 2 Page 11Revision 2Page 11

1SML16012.000-1Methodology and AnalysisTable 3-6 Performance Shaping Factors for Human Failure EventsPerformance Shaping Factors Specific Considerations for RCIC and HPV Events under SBO and(PSFs) Flooding ConditionsCues Cues not only were provided by the Emergency Operating Procedureflowcharts for RPV Level and Containment Pressure and supportingprocedure, but would be expected to be provided by the EmergencyResponse Organization staff and STA. It is expected that daily meetingswould be held to assess the plant situation considering the long termeffects of flooding and SBO and plans would be made to implement theRCIC and HPV actions. The decision to perform these actions wouldtherefore be made by ERO and STA and the Shift Supervisor would givethe direction to operations staff.Indications Due to the Station Blackout, impacts to the normal set of indications wereexpected and these effects were implemented in the CBDTM module ofthe HRA Calculator consistent with the methods recommended inNUREG-1921, Appendix B as well as in the Execution PSFs and Stressmodule. In addition, the "Degree of Clarity of Cues & Indications" wasdegraded from "Very Good" to "Poor".Complexity of the Required Action The response is considered to be Complex due to the flooding and SBOimpacts to lighting and accessibility. The procedure steps of the actionsthat are required and that were evaluated in the timed Job PerformanceMeasures (JPMs) specifically performed for these tasks are listed in theExecution Unrecovered module of HRA Calculator.Special Equipment Flashlights, headlamps and boots were considered necessary by Trainingwhen the JPMs were performed for these tasks, and are reflected in thetiming estimates for Tm and in the Execution PSFs Special Requirementsby indicating that Tools, Parts and Clothing were required and available.Human-System Interfaces The interfaces with the system will be similar, but with impacts due to theSBO and flooding in terms of indications and lighting. For the RCIC task,a hand-held reactor level monitor is installed and used by I&C techniciansto monitor the parameter. These steps of the procedure were specificallyincluded in the Execution portion of the RCIC action quantification. TheHPV task involves the use of air cylinders and these steps were alsoincluded in the Execution portion of the quantification.Revision 2 Page 12Revision 2Page 12

ISMLI16012.000-1 Methodology and AnalysisTable 3-6 Performance Shaping Factors for Human Failure EventsPerformance Shaping Factors Specific Considerations for RCIC and HPV Events under SBO and(PSFs) Flooding ConditionsProcedures Multiple procedures provide for operation of the RCIC System without theavailability of AC or DC power. They address the lighting and ventilationlimitations that accompany a SBO, and provide for alternate means ofmonitoring reactor water level/pressure, controlling turbine speed,assuring adequate water source is available and accommodating thecondensate from the turbine condenser.* Procedure 8900 (Operation of RCIC without Electric Power)* Procedure C.4-L; Part F (Response to Security Threats; Initiate Injectionto the RPV with RCIC)* Procedure A.8-05.01 (Manual Operation of RCIC)* Procedure B.08.09-05.H.4 (Condensate Storage System -FillingCondensate Storage Tanks from Alternate Source)-Procedure A.8-05.05 (Makeup to CST)Any of the following procedures provide for operation of the HPV withoutthe need for normal support systems including electric power andpneumatic supplies. All necessary equipment to perform this function isspecifically manufactured and pre-staged to allow opening the HPVvalves.-Procedure B.04.01-05.H.2 (Primary Containment System Operation -Alternate N2 Supply for Operating AO-4539 and AO-4540)-Procedure C.5-3505; Part A (Venting Primary Containment; VentThrough the Hard Pipe Vent)-Procedure A.8-05.08 (Manually Open Containment Vent Lines)The relevant procedures were reviewed, cues and operator action stepswere itemized in the quantification and were evaluated using multipleiterations of the timed JPMs.Training and Experience The RCIC and HPV capabilities are included in various aspects ofperiodic operations training. Training materials and mockup trainingdevices related to these activities include:" JPM -B.02.03-005 (Reset RCIC Overspeed Trip)" Training mockup of the RCIC Turbine Trip Throttle Valve (MO-2080)" Lesson Plan MT-ILT-EOP-002L (RPV Control)" Lesson Plan MT-NLO-12C-002L (Emergency Operating ProceduresOverview)* Lesson Plan MT-NLO-EOP-001 L (EOPs for NLOs (Turbine Building))* Lesson Plan MT-OPS-FB-004L (Level 4 -Extensive Damage MitigationGuidelines)-Lesson Plan M-8107L-003 (RCIC)Additionally, control of RCIC, using the RCIC MO-2080 Turbine TripThrottle Valve is actually performed during the startup from each refuelingoutage, in the performance of procedure 1056 (RCICI Turbine OverspeedTrip Test).Operator training, procedure adequacy, and equipment readinessregarding use of these mitigation measures has been reviewed by theNRC as part of the B.5.b and Fukushima Flex response inspections andbeen determined to be acceptable.* ML1 1235A897 (NRC Fire/B.5.b Inspection Report; 8/23/11)* ML1 11320400 (NRC Temporary Instruction 2515/183; 5/13/11).Perceived Workload, Pressure Under the quantification Execution PSFs, Workload has been assessedand Stress as High and PSFs of Negative, so the overall stress is assessed as High.Revision 2 Page 13Revision 2Page 13

ISMLI16012.000-1 Methodology and AnalysisTable 3-6 Performance Shaping Factors for Human Failure EventsPerformance Shaping Factors Specific Considerations for RCIC and HPV Events under SBO and(PSFs) Flooding ConditionsEnvironmental Factors The environment would be consistent with SBO (hot, dark, damp) andsome Reactor Building flooding requiring boots. These were theconditions evaluated during the timed JPMs and are addressed in thequantification under Execution PSFs for Lighting, Heat/Humidity andAtmosphere.Special Fitness Issues No special fitness issues were identified although the performance ofmultiple trials of the JPMs is considered to address the variability inpersonnel fitness.Staffing Operations staffing to perform the procedures would be optimal (severaloperators assigned as desired to each procedure). Discussions were heldto evaluate whether there would be dependencies in staffing between theRCIC and HPV actions and it was considered that due to the differenttimeframes, that the actions and staffing would be separate.Communications Under SBO conditions the use of radio and walkie talkies would beexpected to allow communications to be maintained between the MainControl Room and the I&C Technicians and the staff performing the keyactions.Accessibility The Equipment Accessibility is evaluated as "With Difficulty" due toreactor building lighting and flooding issues. The quantification ExecutionPSFs indicates this and these issues were addressed during the JPMtiming sessions.A sensitivity analysis was also performed by developing human error probabilities (HEPs) formanual local operation of RCIC and the Hard Pipe Vent during an extreme flooding and SBOevent utilizing the SPAR-H module of the EPRI HRA Calculator and recommended practicesfrom the Idaho National Laboratory step-by-step SPAR-H guidance (Reference 17). UsingSPAR-H, event RCICSBOFLOOD (Fail to manually operate RCIC during SBO and extremeflooding conditions), and event HPVSBO FLOOD (Fail to operate the HPV using N2 bottles toprovide containment heat removal during SBO/Flood) have values of 1.4E-01 and 5.5E-02respectively, for a combined value of 1.95E-01.The HRA calculator reports associated with these HEPs can be found in sections A.3 and A.4 ofAppendix A.Revision 2 Page 14Revision 2Page 14

ISML16012.000-1Conclusions4.0 CONCLUSIONSThe results of the analysis are shown in Table 4-1. This table includes the results of both the bestestimate quantification and the sensitivities that were performed in support of this analysis.The ACDF represents the difference between the best-estimate CDF and the baseline CDF. Thebaseline CDF is the CDF without the performance deficiency. The best-estimate CDF is theresult of the event tree calculations shown in Appendix A. The exposure time represents theperiod of time the plant was exposed to the performance deficiency. This time period was fromFebruary 29, 2012 to February, 15, 2013 resulting in an exposure time of 352 days or 0.964years. The failure to build construct a levee is assumed to be 0.11 (consistent with theprobability assumed by the NRC). Therefore, the baseline CDF is calculated by multiplying thebest-estimate CDF by 0.11. ACDF is equal to the best-estimate minus the baseline, thus:= best-estimate -0.11 x best-estimate= (1-0.11) x best-estimate= 0.89 x best-estimate.ACDP = exposure x ACDF = 0.964 x 0.89 x best-estimate.Table 4-1 Results of Event Tree QuantificationSensitivity 1: Sensitivity 2:NRC Case Nominal, Bounding Flood SPAR-H HRAFrequency ProbabilitiesCD Seq 1 8.41 E-07 1.06E-06 1.55E-06CD Seq 2 9.15E-08 9.43E-07 1.68E-07CD Seq 3 1.07E-07 1.10E-06 1.07E-07CDF 4.20E-05 1.04E-06 3.10E-06 1.83E-06ACDF 3.60E-05 8.92E-07 2.66E-06 1.57E-06Significance Yellow Green White WhiteThe results of this analysis show that by a best-estimate analysis, the significance of the floodingevent is Green. Two sensitivity studies performed to evaluate the effects of sources ofuncertainty show that the significance of the flooding could be characterized as low to moderatesafety or security significance or 'White' using bounding assumptions.Revision 2 Page 15Revision 2Page 15

I SML16012.000-1AppendicesAPPENDICESTABLE OF CONTENTSA. APPENDIX A -HRA CALCULATOR REPORTS ........................................ A-1A.1. RCICSBOFLOOD, Fail to manually operate RCIC during SBO andextrem e flooding conditions ............................................................... A-1A.2. HPVSBOFLOOD, Fail to operate the HPV using N2 bottles to providecontainment heat removal during SBO/Flood ....................................... A-19A.3. RCICSBOFLOOD, Fail to manually operate RCIC during SBO andextreme flooding conditions (SPAR-H) ................................................. A-31A.4. HPVSBOFLOOD, Fail to operate the HPV using N2 bottles to providecontainment heat removal during SBO/Flood (SPAR-H) ...................... A-36B. APPENDIX B -EVENT TREES ................................................................... B-1Revision 2 Page 16Revision 2Page 16

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSA. APPENDIX A -HRA CALCULATOR REPORTSThe following sections document the HRA Calculator reports that support this analysis.A.1. RCIC_SBOFLOOD, Fail to manually operate RCIC during SBO and extremeflooding conditionsBasic Event Summary.%Plant Data File: ':FiledSize' File Dated. IR1ecord. DateMonticello Ext 913408 07/02/13 07/02/13Flooding SDPHRAJune2013.HRA':Nam ;-DateAnalyst Erin P. Collins, Hughes 07/02/2013AssociatesReviewer John Spaargaren & Pierre 07/02/2013Macheret, Hug hes AssociatesTable 41: RCIC_SBOFLOOD SUMMARY______________ ~HEP,-Sumay __Pcog Pexe Total HEP ErrorFactorMethod CBDTM THERP CBDTM + THERPWithout Recovery 2.9e-02 5.2e-01With Recovery 9.2e-04 9.2e-02 9.3e-02 5Initial Cue:RPV Water Level Below 9 in.Recovery Cue:Before RPV level drops to -149 inchesCue Comments:The cue for action is that the TSC and the Emergency Director have determined that RCIC operation isneeded. The Control Room Supervisor (CRS) directs operator to initiate RCIC and inject into the RPVusing procedure A.8-05.01, Manual Operation of RCIC, Part A, Placing RCIC in Service.Due to the SBO, it is assumed that there will be multiple impacts to indications, so the degree of clarityhas been set at "Poor".Deqree of Clarity of Cues & Indications:PoorProcedures:Cognitive: C.5-1 100 (RPV CONTROL flowchart (Monticello)) Revision: 11Execution: A.8-05.01 (Manual Operation of RCIC) Revision: 2Other: A.6 (ACTS OF NATURE (Monticello)) Revision: 43Other: 8900 (OPERATION OF RCIC WITHOUT ELECTRIC POWER (Monticello)) Revision: 2Revision 2 Page A-IRevision 2Page A-1

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSCognitive Procedure:Step: LEVELInstruction: Restore and maintain RPV water level 9 to 48 in. using Preferred Injection SystemsProcedure and Training Notes:Three JPM trials were performed emulating the specific external flooding conditions of this scenario on 18June 2013. Observations were factored into this analysis.Procedure A.8-05.01 -2.0 ENTRY CONDITIONSThis strategy is entered when one or more of the following conditions are met:o As directed from A.8-01.01 (Extensive Damage Mitigation Strategy Overview)o As directed from A.8-03.01 (Initial Response Actions)o As directed from 5790-110-01 (Monticello Emergency Management Guideline) -presume this is therelevant case since ERF is staffed and will make the call on implementing the procedureTraining:Classroom, Frequency: 0.5 per yearSimulator, Frequency: 0.5 per yearJPM Procedure:JPM-A.8-05-01-001 (Manual Operation of RCIC) Revision: 0Identification and Definition:1. Letter to Region III SRA write-up, 8 April 2013Section G -HEP Associated with Protecting Plant Buildings from 930' Elevation FloodFor the case where the site is not protected by a ring levee, but individual buildings / equipment areprotected by flood barriers as called out in the A.6 (Acts of Nature) procedure, several redundant optionsremain available for protecting critical safety functions related to injecting water to the reactor, maintainingdesired reactor pressure, and removing decay heat from containment.The A.6 Procedure [Acts of Nature that addresses External Flooding] directs de-energizing the 115 KV,230 KV and 345 KV substations for flood levels in excess of the 930' elevation. In the case where theflood level is above 930' this would lead to a loss of offsite power and reliance on the EDGs. Since thenormal long term fuel oil storage (Tank T-44) for the EDG's is not evaluated to survive flood levels above932' elevation, and alternate fuel oil makeup methods are not pre-prescribed to support the EDG's, longterm EDG operation is not easily defensible. Additionally, there are several penetrations in the PlantAdministration Building that would make positive flood proofing of the building difficult, leaving three of thefour station batteries vulnerable to flooding.2. PRA-MT-SY-RCIC, Reactor Core Isolation Cooling System Notebook, Revision 3.0, December2012Table 3 -IE_LOOP (Loss Of Offsite Power Initiating Event) Impact on RCIC System: A loss-of-offsitepower does not affect RCIC, provided AC power remains available to the Division 1 battery chargers. Aloss of Feedwater would result, and RCIC operation would be automatically actuated when Low-LowLevel in the RPV is reached.Key Assumptions:It is assumed that there will be sufficient water and fuel supply for the equipment needed in this scenariodespite the flooding conditions and considering the long term nature of the flood, which may take as manyas 12 days to recede (from the Monticello Design Basis documentation).This means that CST level is assumed to be maintained via a step in this HEP and is not quantifiedseparately.Another key assumption is that because the staffing needed for this event is separate from that used forthe Hard Pipe Vent human failure event, and the timeframe for HPV is much longer, no dependencybetween these events was evaluated (in other words, the timing and staffing were considered as totallyseparate events).Revision 2Page A-2

1SMIL160112.000-11Appendix A -HRA CALCULATOR REPORTSISM LI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSJPM-A.8-05.01-001 (Manual Operation of RCIC) Rev. 0INITIAL CONDITIONS:o Extreme flooding has led to a Station Blackout that has existed at Monticello for the last 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />.o Div. 1 250 VDC battery system has been depleted and is not available.o The plant was in Shutdown Cooling until the station blackout and has since been slowlyrepressurizing due to heating up.o Current RPV pressure is 75 psig-and slowly rising.o Current RPV water level is -40" and very slowly lowering.o The TSC and the Emergency Director have determined that RCIC operation is needed.o HPCI is inoperable.o RCIC suction is from the CSTs.o RCIC operation is required to maintain RPV level above TAF.o Approximately 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> ago Radiation Protection reported -2" of water on the Rx Bldg basementfloor.o A second operator will be performing Part B, Set Up and Monitor of Rx Vessel with Fluke 707.o A third operator will be maintaining CST level using A.8-05.05, Makeup to the CST, and verifyingvalve status in the steam chase.INITIATING CUES (IF APPLICABLE):o The CRS directs you to initiate RCIC and inject into the RPV using A.8-05.01, Manual OperationofRCIC, Part A, Placing RCIC in Service.o Inform the CRS when RCIC is in service with discharge pressure at least 76 psig greater than reactorpressure..RCIC local manual operation Job Performance Measure entry condition assumptions, Documented in andexcerpted from Hughes Associates Record of Correspondence, RCIC Manual Operation e-mails with XcelEnergy during June -July 2013, Hughes Associates, Baltimore, MD, 7 July 2013:Conditions anticipated following a SBO resulting from an external flooding event:o ERO has been manned for the past several days, with these procedures predicted and plannedto be implemented ahead of timeo Plant is in cold shutdown condition (mode 4)o I&C would perform the reactor level monitoring portion of the RCIC procedureo Operations staffing to perform the procedures would be optimal (several operators assigned asdesired to each procedure)o Environment would be consistent with SBO (hot, dark, damp)o There would be 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> to perform the procedures, allowing several opportunities to troubleshootand/or re-perform steps if necessaryo The ERO would place maximum priority on maximizing chances of successful performance ofthese proceduresOperator Interview Insights:Documented in and excerpted from Hughes Associates Record of Correspondence, RCIC ManualOperation e-mails with Xcel Energy during June -July 2013, Hughes Associates, Baltimore, MD, 7 July2013:From Xcel Operations: The series of events would progress during a flooding event such that the need forthe alternate instrumentation would be known before the flood completely resulted in a station blackoutwith loss of DC. The alternate instrumentation (Fluke) would be connected to the selected locations. Ifthis instrumentation did not agree with the actual permanent instrumentation (i.e. prior to failure),assistance would be obtained to figure out the discrepancy. It is a relatively simple solution to get thetemporary instrumentation to work by either pulling a fuse or lifting a lead. This action would be requiredRevision 2Page A-3

I SMLI16012,000-1Appendix A -HRA CALCULATOR REPORTSfor the control room, cable spreading room and EFT locations (again not proceduralized). The otheroption would be to use the transmitter in the reactor building to get the readings which does not requirethe lifted lead or fuse pulled.The other issue that is of concern is the need to density compensate the fluke readings to get an actuallevel. The indicated level can be drastically different from the actual level depending on the calibrationconditions of the instrument (hot or cold calibration conditions) and the actual pressure/temperature at thetime of the reading. None of this information is contained in the A.8 procedures. There are densitycompensation tables in the B.1.1 operations manual figures section six and they are also posted in thecontrol room. It would take additional action for the on-shift team/technical staff to put this all together todetermine what actual level was from the readings that came off the fluke.FD "nniramantael=~ :!.n cluded. Total Available Reuired for PNotes________.......___.... ..__ ... ... __ .... .. .._____ Execuition _________________iReactor operators Yes 2 1Plant operators Yes 2 0Mechanics Yes 2 0Electricians Yes 2 0I&C Technicians Yes 2 0Health Physics Technicians Yes 2 0Chemistry Technicians Yes 1 0Execution Performance Shapina Factors:Environment: Lighting PortableHeat/Humidity Hot / HumidRadiation BackgroundAtmosphere Steam (although steam will notbe present, this PSF was usedto indicate an off-nominalcondition, such as would bepresent for flood and SBO)Special Requirements: Tools RequiredAdequateAvailableParts RequiredAdequateClothing RequiredAdequateComplexity of Response: Cognitive ComplexExecution ComplexEquipment Accessibility Main Control Room Accessible(Cognitive):Equipment Accessibility Reactor Building With Difficulty(Execution):Stress: HighPlant Response As Expected: YesWorkload: HighI Performance Shaping Factors: NegativeRevision 2Page A-4

ISML16012.000-1 Appendix A -HRA CALCULATOR REPORTSPerformance Shaping Factor Notes:The response is considered to be Complex due to the flooding and SBO impacts to lighting andaccessibility. Flashlights, headlamps and boots were considered necessary by Training when the JPMswere performed for these tasks.The Equipment Accessibility is evaluated as With Difficulty due to Rx building lighting and flooding issues.Despite preparations and training, the flooding scenario is considered to be a high stress situation.Key Assumptions (see that section) regarding the conditions provided to Training for performing the JPMfor this task said that the "Environment would be consistent with SBO (hot, dark, damp)". The Traininginsights from the JPM performance (see Operator Interview Insights) stated that the operatorsrecommended to "Stage additional flashlights and headlamps in RCIC room", so it is clear that portablelighting is used.Revision 2 Page A-5Revision 2Page A-5

1 SML16012.000-1Appendix A -HRA CALCULATOR REPORTSTiming:T S 7.97 HoursT dly5.75 Hours T / 10.00 Minutes T M 80.00 MinutesIrreversibleCue DamageStatet=oTimingi Analysis:TO = Station BlackoutTsw = Time from Station Blackout to the time by which RCIC must be restored.Per Monticello MAAP Calculations, case "SBOCase3-R1", 27 June 2013:Time to TAF = 7.17 hrsTime to -149" = 7.2 hrsTime to 1800 F = 7.97 hrsDamage is assumed to occur if the temperature exceeds 1800 F or 7.97 hrs, so this was used for Tsw asthe time by which RCIC restoration is required.Tdelay = PRA battery calc (PRA-CALC-1 1-002) indicates that there are 5.75 hrs until RCIC batterydepletion. Also, the RCIC Water Flow (column BC) of the d41 tabs in the "SBOCase3-R1" MAAPanalysis spreadsheet shows that RCIC injection stops at approximately the same time (5.74 hrs), so theMAAP runs agree with the calc. This Tdelay can be considered somewhat conservative, since in reality, itis likely that an action would be taken before waiting for battery depletion.T1/2 = The cue for action is that the TSC and the Emergency Response Director have determined thatRCIC operation is needed. Daily planning meetings will have been held to discuss actions to be taken assoon as the diesels are lost, so the 10 minutes is simply an estimate of the meeting time between TSCand ERF personnel to make the actual decision to manually operate RCIC. The Control Room Supervisor(CRS) directs operator to initiate RCIC and inject into the RPV using procedure A.8-05.01, ManualOperation of RCIC, Part A, Placing RCIC in Service.Tm = Results of RCIC local manual operation Job Performance Measure performed 18 June 2013. Theprocedure was performed three times, taking 49 minutes, 37 minutes and 50 minutes to complete for anaverage time of 45 minutes. 50 minutes was used as the conservative value for JPM performance.The JPM did not include the performance of Part B for installation and use of the Fluke level monitoringdevice; this was estimated to require 30 minutes, so the total time for Tm was estimated as 50 min + 30min = 80 min.Time available for cognition and recovery: 53.20 MinutesTime available for recovery: 43.20 MinutesSPAR-H Available time (cognitive): 53.20 MinutesRevision 2 Page A-6Revision 2Page A-6

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSSPAR-H Available time (execution) ratio: 1.54Minimum level of dependence for recovery: LDRevision 2 Page A-7Revision 2Page A-7

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSI SMII 6012.000-1 Appendix A -HRA CALCULATOR REPORTSCognitive UnrecoveredRCICSBOFLOODTable 42: RCICSBOFLOOD COGNITIVE UNRECOVEREDPc Failure Mechanism;. Birainch 'HEPPca: Availability of Information d 1.5e-03PCb: Failure of Attention m 1.5e-02Pcc: Misread/miscommunicate data e 3.0e-03Pcd: Information misleading b 3.0e-03Pce: Skip a step in procedure 9 6.0e-03Pcf: Misinterpret instruction a neg.Pcg: Misinterpret decision logic k neg.PCh: Deliberate violation a neg.Sum of Pca through PCh = Initial Pc = 2.9e-02Notes:Normal RPV water level indication is not available and must be monitored with a hand held deviceinstalled by I&C. This is clearly proceduralized in Part B of A.8-05.01.Presumed that SBO causes issues with normal alarms and indications so pc-a through -d were adjustedconsistent with insights from EPRI 1025294, A Preliminary Approach to Human Reliability Analysis forExternal Events with a Focus on Seismic, October 2012.Revision 2 Page A-8Revision 2Page A-8

I SM L16012.000-11Appendix A -HRA CALCULATOR REPORTSISM LI 6012.000-1 Appendix A -NRA CALCULATOR REPORTSpca: Avalabilty of infonnationu ,cation Aal in CR Inicalion WaIingAlternate 'l'kaining onCR Accurate W Procedure InfelatorsYes(a) neg.(b) neg.(c) neg.3.-e-03 -(d) 1.5e-0311.0e+001.0e-01 (e) 5.0e-021.0e+00 (f) 5.0e-011.0e400(g) I.0e+O01LOO1MMCR indications may not be accurate due to the Station Blackout, however, either procedural or informalcrew information on alternate indications and training should provide operator input to decision-making.pcb: Falum of attentionLow vs. M Check vs. Monitor Front vs. Back Alaued vmNotWorkload Pan el Alnnedek IO O(a) neg..e*i Back I 0(b) .5e-043.0e-0 (c) 3.0e-031.8e+.00 Front 15.0e-02 (d) I.5e-04io.00M oI (e) 3.0e-033e-03 Back 15.0e-02 (f) 3.0e-041. Cb ie 3.0e-03 (g) 6.Oe-032. Cb ice Front 15.0e-02 ((h) neg.Check O.e00 e4 (i)neg.o.oeo Back 5.0e,.2 (i) 7.Se-04lh 3.0e-03 (k) 15e-02Front -(I)7.e-04Monito r O.O+- WL -- (m) I.5e-023.e35 I50e-02 (n) 1.5e-033.0e-3 I (o) 3.Oe-021.0e4.0Revision 2 Page 9Revision 2Page 9

ISML16012.000-1Appendix A -HRA CALCULATOR REPORTSISM LI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSpcc nicate dataIhucatom Easy to Good&Bad Inicator FoanalLocateI CoImunKicaUonsS O.e 4.0 0 (a) neg.S3.0e-033.e-030.Oe+00 I (c) I.Oe-03YeIot.0e-0 A(d) 4.0e-033.0e-03No ---o (e) 3.0e03o.oe*Coo (6e33F)03 6.e-03L 3Oe-033.OeO3 (g) 4.0e-031.0e-03 (3.Oe-03(h) 7.Oe-033.0e-03For this scenario, normal reactor water level indication is not available and a hand-held level monitor willbe jumpered in. Although procedural guidance on reading the monitor is clear, "not Easy" is selected toreflect the additional challenges to the task posed by this alternative source for level indication.pcd: Infoonation mismemingM Cues as Stated mring of Specicr ,Rang I General Ta-anugI ~ ~Ifferences IIo.oesoo (a) neg.No---- -- -- ----- --- --- ----- --- ----- --- (b) 3.e-031.0e+O0 (c) 1.0e-021.0e-02MCR indications may not be accurate due to the Station Blackout so cues may not be as stated inprocedures.pce: Sip a step in procedureObqgiousv& Singl vs. NfUN*l Graphicaly Placekeqing Ad(a) 3.0e-033.3"1 (b) 3.0e-03()e-02o.Oe+O 13.0e-03 (c) 3.0e-03130e0eO0I (e) 2.0e-03(h) 1.3e-02---------------------------(g) .oe-o31.0e-021.0e-4D 1 01) t.Oe -01Revision 2 Page 10Revision 2Page 10

ISML16012.000-1Appendix A -HRA CALCULATOR REPORTSISM LI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSpd' Misinterpret instructionStandard or AN Required Training on StepAmbiguous wording Information I--- ------ ------- ------- ------- (a) neg.(b) 3.0e-033.0e-02 (:c) 3.0e-02Yes 1.0e4.00NO1.0e,4 (d) 3.0e-030.0e+00 (e) 3.0e-023.0e-02 1(f) 6.0e-033.0e-02 (g) 6.0e-02pcg: Misinterprt decision ogic"NOr Statement 'ANUD or 'Or Both "AND" & practiced ScenarioStatement "OW3-le0M (a) 1.6e-023.0e.02 I(b) 4.9e-021.2e-02 (c) 6.0e-030o.oe+oo l(d) 1.9-02(e) 2.0e-03O.Oe.+O0 (t) 6.0e-03Yes 3~e-. (g) 1.0e-02NO 3.0e..02 -(h) 3.1e-02.0e4.001.0e-03 (I) 3.0e-04---O.OeO Fi) 1.0e-030.0e+00 l .0e44)3.----- (k) neg.O.Oe.F (1) neg-pch: Delbe ate WiolationBEleinAdequacy Advere Reasonable Polcy ofof Instruction Consequenceff Alternatives "Vembatin"-.---- ------- --- ------- ------- ------- --- --(a)neg.Ye15.0e- (b) 5.oe-o1NO.e00 I.OeO (c) I.Oe-Oo1.0e+00 I O.Oe+O0 (d)neg.0.0e44)0 (e) neg.Revision 2 Page 11Revision 2Page11

1SML-16012.000-1Appendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSCognitive RecoveryRCICSBOFLOODTable 43: RCICSBOFLOOD COGNITIVE RECOVERYU- ', a_ , .E -FinalInitial HEP FinalS0) W C ValueLU OOr "(LU >S o n- -ý.'PCa: 1.5e-03 X -5.0e-01 7.5e-04Pcb: 1.5e-02 X -X X MD 3.8e-03 5.7e-05Pcc:: 3.0e-03 --X X -1.0e-02 3.0e-05Pcd! 3.0e-03 -X X X -5.0e-03 1.5e-05Pce: 6.0e-03 X X X MD 1. le-02 6.6e-05Pof, nell. -1.0e+00Pc-:. neg.- -1.0e+00Pch: : ne .-1.0e+00Sumof Pch = Inita Pc-= 9.2e-04Notes:Due to long timeframe and severity of scenario, STA and Emergency Response Facility will be available.Operations staffing to perform the procedures was assessed by Xcel as optimal (several operatorsassigned as desired to each procedure) so Extra Crew was credited.Used Moderate Dependency due to high stress.Revision 2 Page 12Revision 2Page 12

ISML16012.000-1Appendix A -HRA CALCULATOR REPORTSISM LI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSExecution UnrecoveredRCICSBOFLOODTable 44: RCICSBOFLOOD EXECUTION UNRECOVEREDProcedure: A.8-05.01, Manual Operation of RCIC .Comment. Stress- Over Ride%* .- ,.FactorStep No. instrUction/Comment Error ..THERP -HEP... ..* .....YPe.. .. 1. Table .. I .*Item * __._ * ._.In RCIC Room, verify open valves MO-2106 and MO-2096 Depress declutch lever and turn handwheel.A.8-05.01, Step 8 Location: Reactor Building EOM 20-7b 2 1.3e-03 5EOC 20-13 1 1.3E-3Total Step HEP 1.3e-02Remove the pin securing the slip link to the governor lever to preventinterference from the hydraulic governor (See Attachment 1) 5A.8-05.01, Step 9 Location: Reactor Building EOM 20-7b 2 1.3e-03EOC 99 1 1.OE-2Total Step HEP 5.7e-02Uncap and throttle open RCIC-27 condenser cooling water starting YSA.8-05.01, Step 20 6082 drain. Close when steam is present. 5-22 Location: Reactor Building EOM 20-7b 2 1.5e-03_5-_22_EOC 20-12 5 1.3E-3Total Step HEP 1.3e-02Throttle MO-2080 and control as necessaryA.8-05.01 Steps Location: Reactor Building EOM 20-7b 2 1.3e-03 517,18,25,26 EOC 20-12 5 1.3E-3Total Step HEP 1.3e-02At D31, Access Control -SET UP for Monitoring RX Vessel level byA.8-05-01, Step 2, opening circuits and kick out to Part B Mat D31 Location: Reactor Building EOM 20-7b 3 1.3e-03_5atD31_EOC 20-12 3 1.3E-3Total Step HEP 1.3e-02At D100, 1st Floor EFT -SET UP for Monitoring RX Vessel level byA.8-05-01, Step 2, opening circuits and kick out to Part BAt ST Location: EFT EOM 20-7b 3 1 .3e-03At EFT EOC 20-12 3 1.3E-3Total Step HEP 1.3e-02SET UP for Monitoring RX Vessel level by selecting level instrument and PART B SET UP AND MONITOR OF RX VESSELattaching Fluke 707 monitoring device -Control Room WITH FLUKE 707Part B, Steps 29 & 29. SELECT a level instrument from list below,30P -MCR (preference should be given to accessibility and 5environmental conditions such as radiation,temperature, lighting, etc.):In Control Room;Revision 2Page A-1 3

ISML16012.000-1Appendix A -HRA CALCULATOR REPORTSSýProcedure:'A.8.45.01; Manual Operationof.RCIC Comment Stress. Over RideFactorStep No. Instruction/Commetn:t-: Error THERP HEPi. " .i : .... ..... ..-" ' i , " : i :+ Tabley. ; I;. , tem ... .I. .LI-2-3-85A, Reactor Vessel Water Level, Panel C-03, Term Strip TT-62 to TT-61.LI-2-3-85B, Reactor Vessel Water Level, Panel C-03, Term Strip PP-70 to PP-71.LI-2-3-91A, Fuel Zone, Panel C-03, Term Strip TT-59 to TT-58.30.If Level Indicator selected for use,Then ATTACH one lead from Fluke 707 to eachterminal point listed.Location: Main Control Room EOM 20-7b 3 1.3e-03EOC 20-12 13 1.3E-2Total Step HEP 7.2e-02SET UP for Monitoring RX Vessel level by selecting level instrument and PART B SET UP AND MONITOR OF RX VESSELattaching Fluke 707 monitoring device -Cable Spreading Room WITH FLUKE 70729. SELECT a level instrument from list below,(preference should be given to accessibility andenvironmental conditions such as radiation,temperature, lighting, etc.):In Cable Spreading Room;Part B, Steps 29.& LI-2-3-91A, Fuel Zone, Panel C-18, Term Strip BB- 530 -CSR 57 to BB-59.If Level Indicator selected for use,Then ATTACH one lead from Fluke 707 to eachterminal point listed.Location: Cable Spreading Room EOM 20-7b 3 1.3e-03EOC 20-12 13 1.3E-2Total Step HEP 7.2e-02SET UP for Monitoring RX Vessel level by selecting level instrument and PART B SET UP AND MONITOR OF RX VESSELattaching Fluke 707 monitoring device -EFT WITH FLUKE 70729. SELECT a level instrument from list below,(preference should be given to accessibility andenvironmental conditions such as radiation,temperature, lighting, etc.):At 3rd Floor EFT, ASDS Panel;LI-2-3-86, Reactor Flooding Level, Panel C-292,Part B, Steps 29 & Term Strip (EFT 3) HH-4 to HH-5. 530 -EFT LI-2-3-91 B, Fuel Zone, Panel C-292, Term Strip(EFT 3) HH-1 to HH-2.If Level Indicator selected for use,Then ATTACH one lead from Fluke 707 to eachterminal point listed.Location: EFT EOM 20-7b 3 1.3e-03EOC 20-12 13 1.3E-2 IRevision 2Page A-14

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSProcedure: A.8-05.01, ManUal Operation of R(Step No.instruction/CotSET UP for Monitoring RX Vessel level by selecting level instrument andattaching Fluke 707 monitoring device -Reactor BuildingPART B SET UP AND MONITOR OF RX VESSELWITH FLUKE 70729. SELECT a level instrument from list below,(preference should be given to accessibility andenvironmental conditions such as radiation,temperature, lighting, etc.):At 962', Rx Bldg;LT-2-3-85A, Reactor Vessel Water Level, Panel C-56.LT-2-3-85B, Reactor Vessel Water Level, Panel C-55.LT-2-3-61, Reactor Flooding Level, Panel C-55.At 935', Rx Bldg;LT-2-3-112A, Fuel Zone, Rx Bldg 935' West, PanelC-122.LT-2-3-112B, Fuel Zone, Rx Bldg 935' East, PanelC-121.31. If level transmitter selected for use,Then PERFORM the following:a. REMOVE the cover (see Attachment 5 & 6).b. LIFT positive lead from its conductor (seeAttachment 7).c. Slightly ENGAGE screw threads into transmitter.d. CLAMP Fluke 707 leads in series with liftedpositive wire and positive terminal point ontransmitter.Part B, Steps 29 &31 -RxB5Location: Reactor Building EOM 20-7b 3 1 .3e-03EOC 1 20-12 1 13 1 1.3E-2-t --Total Step HEP 7.2e-02MONITOR OF RX VESSEL WITH FLUKE 707 (Conducted separately by DETERMINE Rx water level by performing theI&C) following: (see Attachment 8 -diagram of FLUKE 707CALIBRATOR pointing out buttons and displays)a. PRESS green button to START Fluke 707.b. PRESS MODE button until display readsMEASURE mA and Loop Power.c. USE Attachment 9 to obtain vessel level. [readPart B, Step 32 table of RCIC -mA VS. RPV WATER LEVEL]33. If connected to LT-2-3-61 or LI-2-3-86,Then OPERATE RCIC turbine to maintain steadyindication as close to 8 mA aspossible.34. If NOT connected to LT-2-3-61 or LI-2-3-86,Then OPERATE RCIC turbine to maintain steadyindication as close to 20 mA aspossible.Revision 2 Page A-15Revision 2Page A-1 5

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSProcedure:-A.8-05.01, Manual Operation of RCIC Comment-w .'Stress Over Ride,FactorStep INo .Instrudtion/comnment. Error -~ THERP :F HEP.Typeý. Table,. ItemLocation: Reactor Building EOM 20-7b 3 1.3e-03EOC 20-11 1 1.3E-3Total Step HEP 1.3e-02Interpret Fluke monitor readings using density compensation tables in E-mail from Xcel Operations toB.1.1 operations manual, section 6 (posted in Control Room) Xcel PRA Manager, 2 July 2013The indicated level can be drastically different fromthe actual level depending on the calibrationconditions of the instrument (hot or cold calibrationconditions) and the actual pressure/temperature atthe time of the reading. None of this information iscontained in the A.8 procedures. There are density 5Part B, Step 32c compensation tables in the B.1.1 operations manualfigures section six and they are also posted in thecontrol room. It would take additional action for theon-shift team/technical staff to put this all together todetermine what actual level was from the readingsthat came off the fluke.Location: Main Control Room EOM I 20-7b 3 1.3e-03EOC 120-10 10 1.3E-2Total Step HEP 7.2e-02Maintain CST level and venfy valve status in the steam chase ILocation: Reactor Building EOM 20-7b 4 4.3e-03 5EOC 20-12 5 1.3E-3Total Step HEP 2.8e-02Feedback from I&C Level Monitoring i 5Recovery 1 Location: Reactor Building EOM 99 1 1.0e-02Total Step HEP 5.0e-02Feedback from Control Room 1 5Recovery 2 Location: Main Control Room EOM 20-7b .3 1 1.3e-03Total Step HEP 6.5e-03Revision 2Page A-1 6

1SML-16012,000-1Appendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSExecution RecoveryRCICSBOFLOODTable 4-5: RCICSBOFLOOD EXECUTION RECOVERYCritical Step No.Recovery Step No.ActionHEP (Crit)HEP (Rec).Dep. Cond. HEP Total forDe. (Recl Stan,A-8-05.01, Step 8 In RCIC Room, verify open valves MO-2106 and MO-2096 1.3e-02 7.3e-04Recovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02A-8-05.01, Step 20 -Uncap and throttle open RCIC-27 condenser cooling water 1.3e-02 7.3e-0422 starting YS 6082 drain. Close when steam is present.Recovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02A.8-05.01 Steps 17, Throttle MO-2080 and control as necessary18,25,26 13e02 1.4e04Recovery 1 Feedback from I&C Level Monitoring 5.0e-02 MD 1.9e-01Recovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02A.8-05-01, Step 2, At D31, Access Control -SET UP for Monitoring RX Vessel 1.3e-02 7.3e-04at D31 level by opening circuits and kick out to Part BRecovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02A.8-05-01, Step 2, At D100, 1st Floor EFT -SET UP for Monitoring RX Vessel level 1.3e-02 7.3e-04At EFT by opening circuits and kick out to Part BRecovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02Part B, Steps 29 & SET UP for Monitoring RX Vessel level by selecting level30 -MCR instrument and attaching Fluke 707 monitoring device -7.2e-02 1.1e-02Control RoomRecovery 2 Feedback from Control Room 6.5e-03 MD 1.5e-01Part B, Steps 29 & SET UP for Monitoring RX Vessel level by selecting level30 -CSR Instrument and attaching Fluke 707 monitoring device -Cable 7.2e-02 4.0e-03Spreading Room__Recovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02Part B, Steps 29 & SET UP for Monitoring RX Vessel level by selecting level 7.2e-02 4.0e-0330 -EFT instrument and attaching Fluke 707 monitoring device -EFTRecovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02Part B, Steps 29 & SET UP for Monitoring RX Vessel level by selecting level31 -RxB Instrument and attaching Fluke 707 monitoring device -7.2e-02 4.0e-03Reactor BuildingRecovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02Part B, Step 32 MONITOR OF RX VESSEL WITH FLUKE 707 (Conducted 1.3e-02 7.3e-04separately by I&C)Recovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02Part B, Step 32c Interpret Fluke monitor readings using density compensationtables In B.1.1 operations manual, section 6 (posted In Control 7.2e-02 7.0e-03I Room)Recovery I Feedback from I&C Level Monitoring 5.0e-02 LD 9.8e-02Revision 2 Page A-17Revision 2Page A-17

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSCritical StA.8-05.05A.8-05.01,ep No. Recovery Step No. .ý.Action. HEP (Crit) HEP (Rec) Dep. Cond. HEP sTotapfor(Rec) : *stepMaintain CST level and verify valve status in the steam chase 2.8e-02 1.6e-03Recovery 2 Feedback from Control Room 6.5e-03 LD 5.6e-02Step 9 Remove the pin securing the slip link to the governor lever toprevent interference from the hydraulic governor (See 5.7e-02 5.7e-02Attachment 1)Total Unrecovered:. 5.2e-01 Total Recovered: 9.2e-02Revision 2 Page A-18Revision 2Page A-1 8

ISML16012.000-1 Appendix A -HRA CALCULATOR REPORTSA.2. HPV_SBO_FLOOD, Fail to operate the HPV using N2 bottles to providecontainment heat removal during SBO/FloodBasic Event SummaryPlant DataFile ...File;Size. File Datei ."Re&cordDateMonticello Ext 901120 06/28/13 06/28/13Flooding SDPHRA June2013.HRAJohn Spaargaren & PierreMacheret, Hughes AssociatesTable 46: HPVSBOFLOOD SUMMARY... .._*__"___._."_ __.___ '._____HEP Sumrnhiaty :. ..: .., .-. -.. ..Pcog Pexe Total HEP ErrorFactorMethod CBDTM THERP CBDTM + THERPWithout Recovery 3.le-02 2.3e-01With Recovery 6.1e-04 1.3e-02 1.3e-02 5Initial Cue:Drywell pressure above 2 psigCue Comments:The cue for action is that the TSC has recommended venting the DW by using the Hard Pipe Vent usingprocedure A.8-05.08, Manually Open Containment Vent Lines.Initial procedure entered on high drywell pressure is C.5-1200. The DW/Torus Pressure leg directsoperators to C.5-3505. For the limiting PRA case, it is assumed that normal and alternate nitrogen andpower via Y-80 is not available and operators must therefore use A.8-05.08 to install pre-staged nitrogenbottles directly to the inboard and outboard HPV isolation valves to open them.Due to the SBO, it is assumed that there will be multiple impacts to indications, so the degree of clarityhas been set at "Poor".Degree of Clarity of Cues & Indications:PoorProcedures:Cognitive: C.5-1200 (PRIMARY CONTAINMENT CONTROL flowchart (Monticello)) Revision: 16Execution: A.8-05.08 (Manually Open Containment Vent Lines) Revision: 1Other: A.6 (ACTS OF NATURE (Monticello)) Revision: 43Other: C.5-3505-A 0 Revision: 10Cognitive Procedure:Step: DW/TORUS PRESSURERevision 2 Page A-19Revision 2Page A-1 9

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSInstruction: BEFORE DW pressure reaches Fig. D, DW Pressure Limit (56 psig) vent to stay below Fig. D,DW Pressure Limit per C.5-3505Procedure and Training Notes:Three JPM trials were performed emulating the specific external flooding conditions of this scenario on 18June 2013. Observations were factored into this analysis.Training:Classroom, Frequency: 0.5 per yearSimulator, Frequency: 0.5 per yearJPM Procedure:JPM-A.8-05.08-001 (Manually Open Containment Vent Lines) Revision: 0Identification and Definition:This HFE is for the external flooding model for venting prior to core damage.1. Initial Conditions: SBO due to external flooding.2. Initiating Events: External flooding causes station blackout.3. Accident sequence (preceding functional failures and successes):No containment venting or heat removalNeed to vent to maintain containment integrity prior to ultimate containment pressure for RCIC injection4. Preceding operator error or success in sequence: None.5. Operator action success criterion: Align pre-staged alternate nitrogen bottles directly to AO-4539 andAO-4540 (located on the torus catwalk) to open the hard pipe vent.6: Consequence of failure: Failure to vent containment leads to containment failure. Any subsequentrelease will likely be through an unscrubbed and unfiltered release path.Key Assumptions:JPM A.8-05.08-001, Rev. 0, Manually Open Containment Vent LinesINITIAL CONDITIONS:o Extreme flooding has led to a Station Blackout that has. existed at Monticello for the last 10hours.o Div. 1 and Div 2 250 VDC battery systems have been depleted and are not available.o The plant was in Shutdown Cooling until the station blackout and has since been slowlyrepressurizing due to heating up.o Current RPV pressure is 75 psig-and slowly rising.o H SRV has failed to reseat and indication of the tailpipe vacuum breaker sticking open have led toa Drywell pressure of 45 psig and rising about 1 psig every 30 minutes.o Efforts to align the diesel fire pump to DW sprays have been unsuccessful due to the flooding.o The TSC has recommended venting the DW by using the Hard Pipe Vent using procedure A.8-05.08, Manually Open Containment Vent LinesINITIATING CUES (IF APPLICABLE):o The CRS directs operator to initiate DW venting through the Hard Pipe Vent lAW procedure A.8-05.08, Manually Open Containment Vent Lines, Parts A and B.Revision 2Page A-20

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSHard Pipe Vent local manual operation Job Performance Measure entry condition assumptions,Information excerpted from Hughes Associates Record of Correspondence, Hard Pipe Vent ManualOperation e-mails with Xcel Energy during June 2013, Hughes Associates, Baltimore, MD, 7 July 2013:Conditions anticipated following a SBO resulting from an external flooding event:o ERO has been manned for the past several days, with these procedures predicted and plannedto be implemented ahead of timeo Plant is in cold shutdown condition (mode 4)o Operations staffing to perform the procedures would be optimal (several operators assigned asdesired to each procedure)o Environment would be consistent with SBO (hot, dark, damp)o There would be more than 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> to perform the procedures, allowing several opportunities totroubleshoot and/or re-perform steps if necessaryo The ERO would place maximum priority on maximizing chances of successful performance ofthese proceduresOperator Interview Insights:The JPM that was completed by Xcel with 2 ROs and 2 NLOs was JPM A.8-05.08-001, Rev. 0, ManuallyOpen Containment Vent Lines. The average time to complete the JPM was 30 minutes given theinformation that the N2 bottles were staged on the CRD catwalk. There were no problems or issues thatrequired any of the operators to stop and get clarifying information, it was identified that removing fittingswas not the best idea it would be better if the lines had tee's installed where the caps could be removedand the appropriate lines connected. This way the capped connection could be labeled to furtherminimize connecting to the wrong fitting. The operators stated that strips of non-skid should be placed onthe areas around the site for safety reasons. They also mentioned using the LED headlights versusflashlights to allow both hands to be free.Manpower Requirements: ..._....._.____"_Ci~Wei~ier hclu~de~d:: TotaAIIVailale 4KRqiiiredfr 'Noe____________ Ex'c~itip'"in _ _ _ __ _Reactor operators Yes 2 1Plant operators Yes 2 0Mechanics Yes 2 0Electricians Yes 2 0I&C Technicians Yes 2 0Health Physics Technicians Yes 2 0Chemistry Technicians Yes 1 0Execution Performance Shapina Factors:Environment: Lighting PortableHeat/Humidity Hot / HumidRadiation BackgroundAtmosphere Steam (although steam will notbe present, this PSF was usedto indicate an off-normalcondition, such as would bepresent for flood and SBO)Special Requirements: Tools RequiredAdequateAvailableParts RequiredI I_ AdequateRevision 2Page A-21

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSClothing RequiredAdequateComplexity of Response: Cognitive ComplexExecution ComplexEquipment Accessibility Main Control Room Accessible(Cognitive):Equipment Accessibility Reactor Building With Difficulty(Execution):Stress: HighPlant Response As Expected: YesWorkload: HighI Performance Shaping Factors: NegativePerformance Shaping Factor Notes:The response is considered to be Complex due to the flooding and SBO impacts to lighting andaccessibility. Flashlights, headlamps and boots were considered necessary by Training when the JPMswere performed for these tasks.The Equipment Accessibility is evaluated as With Difficulty due to Rx building lighting and flooding issues.Key Assumptions (see that section) regarding the conditions provided to Training for performing the JPMfor this task said that the "Environment would be consistent with SBO (hot, dark, damp)". The Traininginsights from the JPM performance stated that the operators recommended the use of "LED headlightsversus flashlights", so it is clear that portable lighting is used.Despite preparations and training, the flooding scenario is considered to be a high stress situation.The steps identified as Critical in JPM-A.8-05.08-001 were used for the Execution quantification.Timing:T SW15.00 HoursT 5.50 HoursdelayT1/2 10.00 MinutesTM 45.00 MinutesM ~ I1CueIIrreversibleDamageStateI-I.t=0Timing Analysis:TO = Station Blackout.Tsw = Per MAAP run Rcic-dgl 3-cts-ABS performed in support of an external flooding SDP, containmentpressure reaches 56 psig at 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br /> following a SBO (flooding >930'). Core temperature reaches 1800degrees F at 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> due to CST depletion and no transfer of RCIC to the torus. This is conservativetiming as refilling of the CST is very likely.Td = 5.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> -Based on an interview conducted in a prior analysis with a senior Shift Manager, theorder to begin the procedure to manually operate the hard pipe vent would be given at approximately 27psig containment pressure. This is due to the step in C.5-1200 (DW/Torus Pressure leg) that says if youcannot restore and maintain drywell pressure within Figure 0 (27psig for 0 ft torus level), then maintaindrywell pressure less than Figure D (56 psig).The 5.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> is based on MAAP run Rcic-dg13-cts-ABS as the time when drywell pressure reaches 42psia (27 psig) [Worksheet d43-1, column AC Drywell Pressure]T1/2 = According to the initial conditions assumed by Training for the Job Performance Measureperformed for this task, the ERO will have been manned for the past several days, with these proceduresRevision 2Page A-22

1 SML16012.000-1Appendix A -HRA CALCULATOR REPORTSpredicted and planned to be implemented ahead of time. Daily planning meetings will have been held todiscuss actions to be taken, so the 10 minutes is simply an estimate of the meeting time between TSCand ERF personnel to make the actual decision to vent the DW by using the Hard Pipe Vent. The controlroom supervisor (CRS) will then direct operators to initiate the process.Tm = Results of HPV local manual operation Job Performance Measure A.8-05.08-001 performed 18June 2013. The procedure was performed four times, taking an average of 30 minutes. Additional 15minutes for C.5-3505-A steps 3 and 4.Time available for cognition and recovery: 525.00 MinutesTime available for recovery: 515.00 MinutesSPAR-H Available time (cognitive): 525.00 MinutesSPAR-H Available time (execution) ratio: 12.44Minimum level of dependence for recovery: ZDRevision 2Page A-23

1SMLII16012.000-1Appendix A -HRA CALCULATOR REPORTSCognitive UnrecoveredHPVSBOFLOODTable 47: HPVSBOFLOOD COGNITIVE UNRECOVERED:Pc Failure .Mecdhanism ... -". 1 .1 ." 1 1 .Branch : .HEP.::..Pca: Availability of Information d 1.5e-03PCb: Failure of Attention m 1.5e-02Pcc: Misread/miscommunicate data e 3.0e-03Pcd: Information misleading b 3.0e-03Pce: Skip a step in procedure e 2.0e-03Pcf: Misinterpret instruction a neg.Pcg: Misinterpret decision logic c 6.0e-03PCh: Deliberate violation a neg.Sum of Pca through PCh = Initial Pc = 3.1e-02Notes:Presumed that SBO causes issues with normal alarms and indications so pc-a through -d were adjustedconsistent with insights from EPRI 1025294, A Preliminary Approach to Human Reliability Analysis forExtemal Events with a Focus on Seismic, October 2012.Revision 2 Page A-24Revision 2Page A-24

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSpca: Avlabhty of infonnationIndiation ail in CR Indication bminglAlftmate Training onCR Accurate in Procedure IndcatorsI.0e-01 (a) neg..Oe.00 1.0e+00 (b) neg.O.0e4.00 I .oe-o (c) neg-7- e0(d) 1-pe-03Yes 5.0e-0 (e) 5.0e-02No1.e+00 f) 5.oe-o01.0e+001.0e-+O (g) .0" WOMCR indications may not be accurate due to the Station Blackout, however, either procedural or informalcrew information on alternate indications and training should provide operator input to decision-making.pcrb: Failure of attentionLow vs. Hi Check vs. Monitor Front vs. Back Aarmed vs.NotWokdload Panel AlarmedEvout(a) neg-O.Oe0 Back .0(b) 1.5e-4LOW 3.0e-03 (c) 3.0e-031.0e+001.0e+00 Front 5e-2(d) 1.-%-045.0e-02OMonitor I0-0e+00 l(e) 3.0e-033.0e-03 Back 5.0e-02 (f) 3.0e-041. Nice 3.0e-03 1(g) 6.0e-032. ic Fro,,t 15.,10 (h eg-5(b0n-02Check O.Oe+0 18 0(i) neg.0.0e*WBack (.0e.so5.0e-02e-04WIgh 3.0e-03 1Ae(k) 1.e-02Front ()7.5e-04Monitor --- Oe --0- (m) I.5e-023 -Back 50e-02 (n) 1.5e-033.Oe-03 I1.0e0 (o) 3.0e-02Revision 2 Page A-25Revision 2Page A-25

I SM L16012.000-1Appendix A -HRA CALCULATOR REPORTSISM LI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSpcc: Misreadftniscmmnnicate dataInhicators Easy to Good--ad Indicator Fom- --Locate cminunicafionsI O.Oe.H) (a) neg.O.Oe.)O 3i 3 (b) 3.0e-03, O-Oe+0 (c) 1.0e-030(g) 4.0e-031.0e-03 (h) 4.0e-033.0e-03o.oe-4oo (g) 3.Oe-0313.0e-033.oe-o3 (h) 7.Oe-O3pcd: Infonmtion fidNilngM Cuesas Stated Warnng SpecoTc Tbn,,ig e TiningDire IIo.oe+0o (a) meg.No-------------------- -------------- --------(b) 3Je-03.0-L-02 (c) 1.0e-021..eOe0 1..od-o (d) 1.Oe-O11.Oe+OO (e) 1.0e" 0MCR indications may not be accurate due to the Station Blackout so cues may not be as stated inprocedures.pce: Sli a step in procedureObvious Ms Eing s Mlle V& MUPleawaf cekeqing A&d13.Oe-O3(a) 1 .0e-033.3e-01 (b) 3.Oe-031.0e.-02O.e.O0 3(c) 3.0e-03.Oe.001e-0 (d) 1.8e-02-(e) 4-e-033.3e-o------.----Yes3.0e-03 13.e-3 (g) 6.0e-031.Oe4-OI (h) I.3e-021..0e-021.0e-01 (i) t.0e -0tRevision 2 Page A-26Revision 2Page A-26

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSpf: Misinterpret instructionStandard or AM Required 7haning on StepAmbiguous wonling Information--- ------- ------- --------- --- (a) neg.(b) 3.oe-033.0e-02 (c) 3.0e-02NO .0e4H (d) 3.0e-03o.e-,.o I(e) 3.0e-023.0e-02 1.0e-01 (f) 6.Oe-033.0e..02 I 0 (g) 6.0e-02pcg: Mi§sinrmpt decision ogicNMor statemmut -ir*N or -ow- Both AND' & IPracticed ScenailStatement "OFr3.3e41 (a) 1.Ge-023.0e-02 (b) 4.9--021.2-02 -3(c) 6.0e-0333.ie-Ol-.Oe. (d) .9e-026.0e-03 (1.0e.9.423.3e-1 (e) 2.0e-03O.Oe-Oe-0Yes OO + 1.00,6 0O (f) 6.0e-03(g) t.Oe-02No 3.0e-02 --(h) 3.1e-021.0e30.-0I.0e-03 30"1 3.0e-0M13.3e-01 (k) neg.O.OedO l- (1) neg.pch: Deliberate violationBelef in Adequacy Adverse Reasonable Poky ofof Instructon Consequence if Alternatis "'Verbatic"oew------- --- ------- ------- --------- ------- ------ (a) neg.Yes 5.0eM (b) 5.0"-01oI. (c) 1.0e4WI.Oe4O110e00 O.Oe,.O0 (d) neg.Io.oe.,o (e) neg.Revision 2 Page A-27Revision 2Page A-27

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSCognitive RecoveryHPVSBOFLOODTable 48: HPVSBOFLOOD COGNITIVE RECOVERY< a) (" I- LL '.D FinalInitiaIIl. HE > Z 1 -'5: -r-.O Value0CO W nOa C -O) Lu ValuePc. 1.5e-03 X X -2.5e-01 3.8e-04Pcb: 1.5e-02 X X X MD 3.8e-03 5.7e-05Pc': 3.0e-03 X X MD 2.1e-02 6.3e-05PCd: 3.0e-03 X X X MD 7.3e-03 2.2e-05<PCe: 2.0e-03 X X X MD 1.0e-02 2.0e-05, :; neg. -1.0e+00P. 6.e-03 X X X MD 1.1 e-02 6.6e-05-I--neg. 1.0e+00P 6 SuoP dt 6.1e-04Notes:Due to long timeframe and severity of scenario, STA and Emergency Response Facility will be available.Operations staffing to perform the procedures was assessed by Xcel as optimal (several operatorsassigned as desired to each procedure) so Extra Crew was credited.Used Moderate Dependency due to high stress.Revision 2 Page A-28Revision 2Page A-28

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSISM LI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSExecution UnrecoveredHPVSBOFLOODTable 49: HPVSBOFLOOD EXECUTION UNRECOVEREDProcedurYe C:A8-5.08,eManuallyOpen Containm ent CvehtLines ., omment Stress Over RideI IStep.No.* jnstructionlComnent,-, Error' -,. HTHERP HEP Factor.,Type' Table Item -..___- _Connect and apply pressure from AH-1 cylinder to rupture Rupture DiskPSD-4543A.8-05.08, Step 4 Location: Reactor Building EOM 20-7b 2 1.3e-03EOC 20-12 5 1.3E-3Total Step HEP 1.3e-02Connect and adjust AH-1 regulator to less than 100 psig and slowly openAH-1 discharge valve to open valve AO-4539 5A.8-05.08, Step 5 Location: Reactor Building EOM 20-7b 4 4.3e-03EOC 20-13 5 1.3E-2Total Step HEP 8.7e-02Connect and adjust AH-2 regulator to less than 100 psig and slowly openAH-2 discharge valve to fully open valve AO-4540 5A.8-05.08, Step 6 Location: Reactor Building EOM 20-7b 4 4.3e-03EOC 20-13 5 1.3E-2 ITotal Step HEP 8.7e-02Open and Close the HPV isolation valves as directed by shift supervisorC.5-3505 Part A, Location: Reactor Building EOM 20-7b 1 4.3e-04 5Step 3 1 EOC 20-13 2 3.8E-3 ITotal Step HEP 2.1e-02Monitor Containment Pressure and Radiation Levels in the Hard PipeC.5-3505 Part A Vent. 5Step 4 Location: Reactor Building EOM 20-7b 1 4.3e-04Step 4 EOC 20-10 1 3.8E-3Total Step HEP 2.1e-02Feedback from Control Room 5Recovery Location: Main Control Room EOM 20-7b 3 1.3e-03 ITotal Step HEP 6.5e-03Revision 2 Page A-29Revision 2Page A-29

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSExecution RecoveryHPV_SBOFLOODTable 4-10: HPVSBOFLOOD EXECUTION RECOVERYc it.Recovery Step No.. -i HEP (crit)i HEP'Rec .Dep. P"Cond. :.Totalffor.~Criticalý 'te k o .__________ _______________________________________________________....___.... ..._(Rec) StepA.8-05.08, Step 4 Connect and apply pressure from AH-1 cylinder to rupture 1.3e-02 7.3e-04Rupture Disk PSD-4543Recovery Feedback from Control Room 6.5e-03 LD 5.6e-02A.8-05.08, Step 5 Connect and adjust AH-1 regulator to less than 100 psig andslowly open AH-1 discharge valve to open valve AO-4539 8.7e-02 4.9e-03Recovery Feedback from Control Room 6.5e-03 LD 5.6e-02A.8-05.08, Step 6 Connect and adjust AH-2 regulator to less than 100 pslg and 8.7e-02 4.9e-03slowly open AH-2 discharge valve to fully open valve AO-4540Recovery Feedback from Control Room 6.5e-03 LD 5.6e-02C.5-3505 Part A, Open and Close the HPV isolation valves as directed by shift 2.1 e-02 1.2e-03Step 3 supervisorRecovery Feedback from Control Room 6.5e-03 LD 5.6e-02C.5-3505 Part A, Monitor Containment Pressure and Radiation Levels in the 2.1 e-02 1.2e-03Step 4 1 Hard Pipe Vent.I Recovery Feedback from Control Room 6.5e-03 LD 5.6e-02-* .: .Total ....I.- Unrecoered:- 2.3e1- .Total R 2ovIered: 1.3e2Revision 2 Page A-30Revision 2Page A-30

ISML16012.000-1Appendix A -HRA CALCULATOR REPORTSA.3. RCICSBOFLOOD, Fail to manually operate RCIC during SBO and extremeflooding conditions (SPAR-H)Basic Event Summary'Planlt;:".. i. Data :File Dati e ;:: Rebo d ::".Monticello Ext 909312 07/02/13 07/02/13Flooding SDPHRAJune2013_SPAR Hquant forsensitivity.HRATable 11: RCIC_SBOFLOOD SUMMARY[Ana l, i Resu!ts:.-I4 Cognitive Execution[Failor 3.2e.. 9.1 e-021 .4e-01Plant:MonticelloInitiating Event:External Flood + SBOBasic Event Context:The flooding engineer provides daily updates to the station on high river water levels including potentialsto rise above any A.6 trigger points. At this point, heightened awareness of the potential for flooding isimplemented.When river level exceeds 921 feet an evaluation of EALs would be performed. If visible damage hasoccurred due to flood water rising greater than 921 feet, then an Alert per EAL HA1.6 would be declared.Prior to river levels reaching these levels, operators would be walking down the A.8 procedures foralternate methods to vent primary containment and operate RCIC remotely. This would involve staging ofequipment in the torus area to open the Hard Pipe Vent and verification that equipment is properly stagedto operate RCIC remotely.EDGs and batteries are not available. Shutdown cooling, HPCI, and RCIC are not available from normalelectrical means. RCIC is available for manual operation.Operators have temporary level indication setup in the reactor building. Pressure indication is available inthe direct area of the level transmitters. The building is dark and most likely water in the basement of thereactor building. Additional portable lights are available to assist with lighting and boots staged for higherwater. The operators would utilize A.8-05.01 to un-latch the governor from the remote servo linkage andthrottle steam flow to RCIC to start the turbine rolling while coordinating with operators monitoring waterlevel and reactor pressure. Upon reaching the high end of the level band the operators would throttleclosed the steam admission valve and await direction to re-start RCIC. Local operation of RCIC isdemonstrated each refueling outage during the over speed test. Operation of a coupled turbine run isless complex because the turbine is easier to control with a load.Revision 2Page A-31

ISML16012.000-1Appendix A -HRA CALCULATOR REPORTSTiming:T 7.97 HoursT delay 5.75 Hours i T1/2 10.00 Minutes TM 80.00 Minutes1IrreversibleCue DamageStatet=oTiming Analysis: TO = Station BlackoutTsw = Time from Station Blackout to the time by which RCIC must be restored.Per Monticello MAAP Calculations, case "SBOCase3-RI", 27 June 2013:Time to TAF = 7.17 hrsTime to -149" = 7.2 hrsTime to 1800 F = 7.97 hrsDamage is assumed to occur if the temperature exceeds 1800 F or 7.97 hrs, so this was used as the timeby which RCIC restoration is required.Tdelay = PRA battery calc (PRA-CALC-1 1-002) indicates that there are 5.75 hrs until RCIC batterydepletion. Also, the RCIC Water Flow (column BC) of the d41 tabs in the "SBOCase3-RI" MAAPanalysis spreadsheet shows that RCIC injection stops at approximately the same time (5.74 hrs), so theMAAP runs agree with the calc. This Tdelay can be considered somewhat conservative, since in reality, itis likely that an action would be taken before waiting for battery depletion.T1/2 = The cue for action is that the TSC and the Emergency Response Director have determined thatRCIC operation is needed. Daily planning meetings will have been held to discuss actions to be taken assoon as the diesels are lost, so the 10 minutes is simply an estimate of the meeting time between TSCand ERF personnel to make the actual decision to manually operate RCIC. The Control Room Supervisor(CRS) directs operator to initiate RCIC and inject into the RPV using procedure A.8-05.01, ManualOperation of RCIC, Part A, Placing RCIC in Service.Tm = Results of RCIC local manual operation Job Performance Measure performed 18 June 2013. Theprocedure was performed three times, taking 49 minutes, 37 minutes and 50 minutes to complete for anaverage time of 45 minutes. 50 minutes was used as the conservative value for JPM performance.The JPM did not include the performance of Part B for installation and use of the Fluke level monitoringdevice; this was estimated to require 30 minutes, so the total time for Tm was estimated as 50 min + 30min = 80 min.Time available for recovery: 43.20 MinutesSPAR-H Available time (cognitive): 53.20 MinutesSPAR-H Available time (execution) ratio: 1.54Minimum level of dependence for recovery: LDRevision 2 Page A-32Revision 2Page A-32

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSISMLI6OI2.000.1 Appendix A -HRA CALCULATOR REPORTSPART I. DIAGNOSISPSFss PSF ,ve;s9 %fMultlper~or<.Available Time Inadequate Time P(failure) = 1.0(recommended choice Barely adequate time (~ 2/3 x nominal) 10based on timing Nominal time 1information in bold) Extra time (between 1 and 2 x nominal 0.1and > 30 min)Expansive time (> 2 x nominal and > 30 X 0.01min)Insufficient Information IStress Extreme 5High X 2Nominal 1Insufficient Information IComplexity Highly complex 5Moderately complex X 2Nominal 1Obvious diagnosis 0.1Insufficient Information 1Experience/Training Low 10Nominal X 1High 0.5Insufficient Information 1Procedures Not available 50Incomplete 20Available, but poor 5Nominal X 1Diagnostic/symptom oriented 0.5Insufficient Information 1ErgonomicslHMI Missing/MisleadingJ 50Poor 10Nominal X IGood 0.5Insufficient Information 1Fitness for Duty Unfit P(failure) = 1.0Degraded Fitness 5Nominal X 1Insufficient Information 1Work Processes Poor 2Nominal 1Good X 0.8Insufficient Information 1Revision 2 Page A-33Revision 2Page A-33

ISM LI6012.000-1ApndxA-HACLUTORERSAppendix A -HRA CALCULATOR REPORTSDiagnosis HEP:3.2e-04PART I1. ACTIONPSFs PSF IUVeis. ~ Multi 11ýr for.__________________ griagn sisAvailable Time Inadequate Time P(failure) = 1.0(recommended choice Time available is -the time required 10based on timing Nominal time X 1information in bold) Time available >= 5x the time required 0.1Time available >= 50x the time required 0.01Insufficient Information 1Stress/Stressors Extreme 5High X 2Nominal 1Insufficient Information 1Complexity Highly complex 5Moderately complex X 2Nominal 1Insufficient Information 1Experience/Training Low 3Nominal X 1High 0.5Insufficient Information 1Procedures Not available 50Incomplete 20Available, but poor X 5Nominal 1Insufficient Information 1Ergonomics/HMI Missing/Misleading 50Poor X 10Nominal 1Good 0.5Insufficient Information 1Fitness for Duty Unfit P(failure) = 1.0Degraded Fitness 5Nominal X 1Insufficient Information 1Work Processes Poor 5Nominal 1Good X 0.5Insufficient Information 0.5Revision 2 Page A-34Revision 2Page A-34

ISML16012.000-1 Appendix A -HRA CALCULATOR REPORTSAction Probability:9.1e-02 [Adjustment applied: 1.0e-3 * 1.0e+02 / (1.0e-3 * (1.0e+02 -1) + 1)]PART Ill. DEPENDENCYcaeI Im In~ mC..-mhwamccwiTask Failure WITHOUT Formal Dependence:9.1le-02Task Failure WITH Formal Dependence:1 .4e-01Revision 2 Page A-35Revision 2Page A-35

ISML16012.000-1Appendix A -HRA CALCULATOR REPORTSA.4. HPVSBOFLOOD, Fail to operate the HPV using N2 bottles to providecontainment heat removal during SBO/Flood (SPAR-H)Basic Event Summary.:Plant , :Datale FileSize".. * FileDate : Rerd DateMonticello Ext 901120 06/28/13 06/28/13Flooding SDPHRAJune2013-SPAR Hquant for1 sensitivity.HRAJohn Spaargaren & PierreMacheret, Hughes AssociatesTable 412: HPVSBOFLOOD SUMMARYAnalyssResults-,* Cognitive ExecutionFe ! rN r 1b6 1i.ii 3.2e-04 5.0e-03Total :HEP. I 5.5e-02Plant:MonticelloInitiating Event:External Flood + SBOBasic Event Context:The flooding engineer provides daily updates to the station on high river water levels including potentialsto rise above any A.6 trigger points. At this point, heightened awareness of the potential for flooding isimplemented.When river level exceeds 921 feet an evaluation of EALs would be performed. If visible damage hasoccurred due to flood water rising greater than 921 feet, then an Alert per EAL HA1.6 would be declared.Prior to river levels reaching these levels, operators would be walking down the A.8 procedures foralternate methods to vent primary containment and operate RCIC remotely. This would involve staging ofequipment in the torus area to open the Hard Pipe Vent and verification that equipment is properly stagedto operate RCIC remotely.EDGs and batteries are not available. Shutdown cooling, HPCI, and RCIC are not available from normalelectrical means. RCIC is available for manual operation.Revision 2 Page A-36Revision 2Page A-36

1SML16012.000-1 Appendix A -HRA CALCULATOR REPORTSTimina:T S 15.00 HoursTdelay 5.50 Hours T1/2 10.00 Minutes TM 45.00 MinutesIreversibleCue DamageStatet=oAnalysis: TO = Station Blackout.Tsw = Per MAAP run Rcic-dg13-cts-ABS performed in support of an external flooding SDP, containmentpressure reaches 56 psig at 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br /> following a SBO (flooding >930'). Core temperature reaches 1800degrees F at 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> due to CST depletion and no transfer of RCIC to the torus. This is conservativetiming as refilling of the CST is very likely.Td = 5.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> -Based on an interview conducted in a prior analysis with a senior Shift Manager, theorder to begin the procedure to manually operate the hard pipe vent would be given at approximately 27psig containment pressure. This is due to the step in C.5-1200 (DW/Torus Pressure leg) that says if youcannot restore and maintain drywell pressure within Figure 0 (27psig for 0 ft torus level), then maintaindrywell pressure less than Figure D (56 psig).The 5.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> is based on MAAP run Rcic-dg13-cts-ABS as the time when drywell pressure reaches 42psia (27 psig) [Worksheet d43-1, column AC Drywell Pressure]T1/2 = According to the initial conditions assumed by Training for the Job Performance Measureperformed for this task, the ERO will have been manned for the past several days, with these procedurespredicted and planned to be implemented ahead of time. Daily planning meetings will have been held todiscuss actions to be taken, so the 10 minutes is simply an estimate of the meeting time between TSCand ERF personnel to make the actual decision to vent the DW by using the Hard Pipe Vent. The controlroom supervisor (CRS) will then direct operators to initiate the process.Tm = Results of HPV local manual operation Job Performance Measure A.8-05.08-001 performed 18June 2013. The procedure was performed four times, taking an average of 30 minutes. Additional 15minutes for C.5-3505-A steps 3 and 4.Time available for recovery: 515.00 MinutesSPAR-H Available time (cognitive): 525.00 MinutesSPAR-H Available time (execution) ratio: 12.44Minimum level of dependence for recovery: ZDRevision 2 Page A-37Revision 2Page A-37

1SML16012.000-1Appendix A -HRA CALCULATOR REPORTSPART I. DIAGNOSISP PSFs,6V , AfPSF vels u-ti'for-Dagposis,Available Time Inadequate Time P(failure) = 1.0(recommended choice Barely adequate time (~ 2/3 x nominal) 10based on timing Nominal time 1information in bold) Extra time (between 1 and 2 x nominal 0.1and > 30 min)Expansive time (> 2 x nominal and > 30 X 0.01min)Insufficient InformationStress Extreme 5High X 2Nominal 1Insufficient Information 1Complexity Highly complex 5Moderately complex X 2Nominal 1Obvious diagnosis 0.1Insufficient Information 1Experience/Training Low 10Nominal X 1High 0.5Insufficient Information 1Procedures Not available 50Incomplete 20Available, but poor 5Nominal X 1Diagnostic/symptom oriented 0.5Insufficient Information 1Ergonomics/HMI Missing/Misleading 50Poor 10Nominal X 1Good 0.5Insufficient Information 1Fitness for Duty Unfit P(failure) = 1.0Degraded Fitness 5Nominal X 1Insufficient Information 1Work Processes Poor 2Nominal 1Good X 0.8Insufficient Information 1Revision 2 Page A-38Revision 2Page A-38

1 SML-16012.000-IAppendix A -HRA CALCULATOR REPORTSI SMLI 6012.000-1 Appendix A -HRA CALCULATOR REPORTSDiagnosis HEP:3.2e-04PART II. ACTION.PSFs,, PSFLevels' s Multiplier forS~ DiagflosisAvailable Time Inadequate Time P(failure) = 1.0(recommended choice Time available is -the time required 10based on timing Nominal time X 1information in bold) Time available >= 5x the time required 0.1Time available >= 50x the time required 0.01Insufficient Information 1Stress/Stressors Extreme 5High X 2Nominal 1Insufficient Information 1Complexity Highly complex 5Moderately complex 2Nominal X IInsufficient Information 1Experience/Training Low 3Nominal 1High X 0.5Insufficient Information 1Procedures Not available 50Incomplete 20Available, but poor 5Nominal X 1Insufficient Information 1Ergonomics/HMI Missing/Misleading 50Poor X 10Nominal 1Good 0.5Insufficient Information IFitness for Duty Unfit P =failure) 1.0Degraded Fitness 5Nominal X 1Insufficient Information 1Work Processes Poor 5Nominal 1Good X 0.5Insufficient Information 0.5Action Probability:5.0e-03Revision 2 Page A-39Revision 2Page A-39

ISMLI16012.000-1 Appendix A -HRA CALCULATOR REPORTSPART II1. DEPENDENCYC,=, TD I E I I C ,F.-- -Cb ntin i...,=, ,w M&IMAB ~ F, ibTask. Failre WTHOUForal D endncesocb addiiorW khedlafr-I.-ks. &k*T Faaulia u HD5emTask Failure WITHOUT Formal Dependence:5.3e-03Task Failure WITH Formal Dependence:5.5e-02Revision 2 Page A-40Revision 2Page A-40'

ISML16012.000-1B. APPENDIX B -EVENT TREESAppendix B -EVENT TREESRevision 2 Page B-IRevision 2Page B-1

EXTERNAL FLOOD >930' < 935' EARLY WARNING REACTOR BUILDING PROTECTED RCIC/RPV & HARD PIPE VENT Prob NameSUCCESSFUL EARLY WARNINGO.OOE+00FLOOD <935'0.891RCIC SUCCESS0.894EXTERNAL FLOOD >930'[8.90E-06][1]SUCCESSFUL EARLY WARNING[0.106]O.00E+007.09E-068.41 E-070.OOE+007.72E-070.15E-081 07F-07DKOK'D Seq 1:)KDK,D Seq 2MD Seq 3O.00E+00FLOOD > 935'[0.109]RCIC SUCCESSREACTOR BLDG PROTECTED0.8940.89FAILURE TO PROTECT RB[0.106]1lI[0.11]IMonticello, Flood SDP 930-935.eta17/3/2013 1 Page 1IMonticello Flood SDP 930-935.eta 7/3/2013 Page 1

EXTERNAL FLOOD >930' < 935' EARLY WARNING REACTOR BUILDING PROTECTED RCIC/RPV & HARD PIPE VENT Prob I NameSUCCESSFUL EARLY WARNINGO.OOE+00FLOOD <935'0.5in flflF4flfRCIC SUCCESS1 a; mai--fnr,[1]0.894'I .uhlt--ubEXTERNAL FLOOD >930'[2.00E-05][0.106]SUCCESSFUL EARLY WARNING:)K::)K,D Seq 1:)K,D Seq 2'D Seq 3O.00E÷00FLOOD > 935'[0.5]RCIC SUCCESSREACTOR BLDG PROTECTED0.8940.89FAILURE TO PROTECT RB7.96E-06-9.43E-071.10E-06[0.106][1][0.11]Monticello Flood SDP 930-935 Sens Freq.eta17/3/2013 1 Page 1

EXTERNAL FLOOD >930' < 935' EARLY WARNING REACTOR BUILDING PROTECTED RCIC/RPV & HARD PIPE VENT Prob NameSUCCESSFUL EARLY WARNINGIO.OOE+00FLOOD <935'0.891 1RCIC SUCCESS0.805D.O0E+005.38E-061.55E-06EXTERNAL FLOOD >930'[8.90E-06][1]SUCCESSFUL EARLY WARNING[0.195]ILl=t =IPI Jt I:)K:)K,D Seq 1:)K:)K'D Seq 2'D Seq 30.OOE+00RCIC SUCCESSFLOOD > 935'[0.109]REACTOR BLDG PROTECTED0.805[1]0.89] FAILURE TO PROTECT RB[0.195]5.95E-071 .68E-071.07E-07[0.11]Monticello Flood SDP 930-935 Sens SPAR-H.eta 7/3/2013 Page 1

Enclosure 3Monticello Nuclear Generating Plant"Monticello Flood Protection"11 Pages Follow

Monticello Flood Protection1.0 PURPOSEThe purpose of this document is to evaluate the flood protection provided at MonticelloNuclear Generating Plant (MNGP).Nuclear power plants are designed to meet robust design criteria, referred to as GeneralDesign Criteria (GDC); which are now codified as part of NRC regulations in 10 CFRPart 50. The GDC have existed in various forms prior to being codified in part 50 andplant commitments to meet the GDC (or pre-existing requirements) depend on the ageof the plant.MNGP was designed before the publishing of the 70 General Design Criteria (GDC) forNuclear Power Plant Construction Permits proposed by the Atomic Energy Commission(AEC) for public comment in July 1967, and constructed prior to the 1971 publication ofthe 10 CFR 50, Appendix A, GDC. As such, MNGP was not licensed to 10 CFR 50Appendix A, GDC. The MNGP USAR, Section 1.2, lists the Principal Design Criteria(PDC) for the design, construction and operation of the plant. MNGP USAR Appendix Eprovides a plant comparative evaluation to the 70 proposed AEC design criteria. It wasconcluded in the USAR that the plant conforms to the intent of the GDC. A listing of thePDC and AEC GDC (by number and title) pertaining to external flooding is providedbelow:PDC 1.2.1 .c "General Criteria""The design of those components which are important to the safety of the plantincludes allowances for the appropriate environmental phenomena at the site.Those components important to safety and required to operate during accidentconditions are designed to operate in the post accident environment."AEC Criterion 2 -Performance Standards (Category A)"Those systems and components of reactor facilities which are essential toprevention of accidents which could affect the public health and safety or tomitigation to their consequences shall be designed, fabricated, and erected toperformance standards that will enable the facility to withstand, without loss ofthe capability to protect the public, the additional forces that might be imposed bynatural phenomena such as earthquakes, tornadoes, flooding conditions, winds,ice, and other local site effects. The design bases so established shall reflect: (a)appropriate consideration of the most severe of these natural phenomena thathave been recorded for the site and surrounding area and (b) an appropriatemargin for withstanding forces greater than those recorded to reflectuncertainties about the historical data and their suitability as a basis for design."This evaluation addresses the following aspects of flood protection that are provided forthe MNGP to meet AEC Criterion 2:* Flood Analyses -this discussion describes the site location, hydrology anddetermination of the maximum predicted flood water elevations and timing.Page 1 of 11

• Flood Mitigation Strategy -this discussion describes the aspects provided topreclude the design bases flood from adversely impacting the site. Thisprotection is provided by structural design and procedural actions.* Flood Protection Implementation -this discussion describes the actions taken atthe site to provide reasonable assurance that flood protection strategy can beeffectively implemented in a design bases flood scenario.2.0 FLOOD ANALYSISThis section describes the site location and hydrology, and a summary of current designbasis flood elevations. Information in this section is based on information in the MNGPUpdated Safety Analysis Report (USAR) (Reference 1); specific sections are identifiedbelow.2.1 Site Location and DescriptionThe plant is located within the city limits of Monticello, Minnesota on the right (west)bank of the Mississippi River. The topography of the MNGP site is characterized byrelatively level bluffs which rise sharply above the river. Three distinct bluffs exist at theplant site at elevations 920, 930, and 940 ft. above msl. The finished plant grade isapproximately 930 ft. msl. The plant grade surrounding Class I and Class II structureshousing Class I equipment varies between 935 ft. msl and 930 ft. msl. The sitedescription and topography is described in detail in the MNGP USAR, Section 2.2.HydrologyThe Mississippi River is the major hydrologic feature for the site. The river poses thesignificant flooding source for the site. Table 1, below, summarizes normal and floodedriver flow rates and water elevations.Table 1Normal and Flooded River Flow Rates and Water ElevationsMississippi River Flow Rate (cfs) Water ElevationCondition (ft. msl)Normal 4,600 905Maximum Recorded(1965) 51,000 9161000 Year Flood -90,000 (1) 921Probable Maximum 364,900 939.2Flood 364,900 _ 939.2(1) Estimated using USAR Appendix G, Exhibit 8, for a waterelevation of 921 ft.Normal river level at the MNGP site is about 905 ft. msl at a distance 1.5 milesupstream, the normal river elevation is about 910 ft. msl and at an equal distancedownstream, the river is at 900 ft. msl. The following flow statistics are estimated for theMississippi River at the MNGP site:Page 2 of 11

Average Flow -4,600 cubic feet per second (cfs)Minimum Flow -240 cfsMaximum Flow -51,000 cfsThe maximum reported high water level at the MNGP site was about 916 ft. msl whichwas recorded during the spring flood of 1965 with an estimated river flow of 51,000 cfs.The results of flood frequency study for the 1000 year flood estimated a peak stage of921 ft. msl (USAR Section 2.4)2.2 Design Basis Flood HazardThe following flood scenarios are evaluated as part of the MNGP licensing basis [USARAppendix G]:* Flooding in Streams and Rivers* Flooding due to Downstream Ice Dam Build-UpA summary of the results as described in Reference 2 for each of these floodingscenarios is provided below; specific sections from Reference 2 are identified with theassociated discussion.2.2.1 Flooding in Streams and RiversThe probable maximum discharge was determined to be 364,900 cfs and acorresponding peak stage of elevation 939.2 ft. msl. The flood would result frommeteorological conditions which could occur in the spring and would reach maximumriver level in about 12 days. It was estimated the flood stage would remain aboveelevation 930.0 ft. msl for approximately 11 days.The most critical sequence of events leading to a major flood would be to have anunusually heavy spring snowfall and low temperatures after a period of intermittent warmspells and sub-freezing temperatures has formed an impervious ground surface andthen a period of extremely high temperatures followed by a major storm. The snowmeltand rainfall excesses were then routed to the plant site by computer modeling. A stagedischarge rating curve was then constructed. The probable maximum discharge wasdetermined to be 364,900 cfs with a corresponding peak stage elevation of 939.2 ft. mslfrom the discharge rating curve.A probable maximum summer storm over the project area was also studied in detail andthe resulting flood at the project site determined. Although the summer storm was muchlarger than the spring storm, the initial retention rate of zero for spring conditions, andthe snowmelt contribution to runoff, resulted in the spring storm producing the morecritical flood.Key Assumptions Used to Determine Design Basis Flood HazardThe PMF evaluation for the spring storm conservatively maximizes the potential snowcover and precipitation. A limiting temperature sequence that results in an imperviousground surface due to subfreezing temperatures is assumed. This is followed byextreme high temperatures, and a subsequent major spring storm. The snowmelt andPage 3 of 11

rainfall maximizes the runoff to the river basin. This sequence of events is postulated toproduce a PMF. Additional details regarding key assumptions used in the analyses aredescribed in USAR Appendix G.Methodology Used to Develop Design Basis Flood HazardThe predicted flood discharge flow and PMF level at the MNGP site was defined usingDepartment of the Army, Office of the Chief of Engineers, the U.S. Army Corps ofEngineers, Engineer Circular No. 1110-2-27, Enclosure 2, "Policies and ProceduresPertaining to Determination of Spillway Capacities and Freeboard Allowances for Dams,"dated August 1, 1966 (Reference 2).The PMF at the MNGP site was determined by transposing an actual critical springstorm to the drainage basin and maximizing the precipitation for potential moisture.Potential snow cover and a critical temperature sequence were developed fordetermining snowmelt contribution to flood runoff.The study area was divided into four major sub-basins and synthetic unit hydrographswere developed for each, using Snyder's method, which is derived from the variousphysical basin characteristics. Unit hydrograph peaks were also increased by 25 percentand basin lag decreased by one-sixth, in accordance with standard Corps of Engineerpractice.Snowmelt and rainfall excesses were applied to unit hydrographs and the resultinghydrographs determined for each sub-basin. Sub-basin hydrographs were then routed tothe project site by computer program using the modified Wilson method. Travel times forflood routing were taken from Corps of Engineers recorded travel times for large floods.Base flow was determined from long-term records of stream flow for nearby stations.Base flow was then added to the total of the routed flood hydrographs.The stage-discharge curve at the MNGP Site was extended above the range of historicalexperience by means of hydraulic computations based on the river channel downstream.This was done by a series of backwater computations based on a range of discharges.Backwater computations were made using water surface elevations and theircorresponding discharges as determined from the rating curve downstream fromMonticello. Using the discharges and the resulting water surface elevations, a stagedischarge curve was constructed for the site.ResultsThe detailed analysis results are presented in USAR Appendix G. To summarize, theanalysis predicts a probable maximum discharge of 364,900 cfs and a correspondingpeak stage of elevation 939.2 ft. msl. The flood would reach maximum river level inabout 12 days after the beginning of high temperatures, and it was estimated the floodstage would remain above elevation 930.0 ft. msl for approximately 11 days.It is noted that the 12 day time period is for the river elevation to reach the peak level.Other important levels are the elevation of the Intake Structure (919 ft.) and Plant Grade(930 ft.). Based on USAR Appendix G Exhibits 8 and 9, water elevation of 919 ft. couldbe exceeded at about the fourth day and water elevation of 930 ft. could be exceeded atthe eighth day.Page 4 of 11

2.2.2 Floods due to Ice Dam Build-UpFlooding due to backwater, usually caused by ice jams, was considered. USAR,Appendix G, Chapter II, Page G.2-5 states that two types of flooding occur in the basin --open-water flooding and backwater flooding. Flooding while open-water conditionsprevail is caused by runoff producing rains, or by melting snow, or by a combination ofthe two. Flooding because of backwater is usually caused by ice jams. The most seriousflooding throughout the basin has been associated with excessive snowmelt and rainfall.Thus, the open-water flooding was considered to be more limiting that the backwaterflooding, and was analyzed in detail in the USAR.3.0 FLOOD MITIGATION STRATEGYFlood protection features and flood mitigation procedures are described below. The PMFevent is applicable to all modes of operation (i.e., power operation, startup, hotshutdown, cold shutdown, and refueling). Flood Protection requirements necessary toprevent external flooding or flood damage to Class I Structures or Class II structureshousing Class I equipment, are identified in USAR Section 12.2.1.7.1. Flood protectionfeatures utilized at MNGP in the event of a PMF include both incorporated (installed) andtemporary active and passive barriers. MNGP does not rely upon any flood protectionfeatures external to the immediate plant area as part of the current licensing basis thatprotect safety related systems, structures and components from inundation andstatic/dynamic effects of external floods.Incorporated engineered passive or active flood protection features are features that arepermanently installed in the plant that protect safety related systems, structures, andcomponents from inundation and static/dynamic effects of external flooding. Examplesinclude external walls and penetration seals that are permanently incorporated into aplant structure.Temporary passive or active flood protection features at MNGP include portable pumps,sandbags, plastic sheeting, steel plates, levees, etc., that protect safety related systems,structures and components from the effects of external flooding.These features aretemporary in nature, i.e., they are installed prior to design basis external flood levelsattaining specific levels.The following Class I and II structures are protected from flooding up to 939.2 ft. msl:1. Reactor Building (including High Pressure Coolant Injection (HPCI) structure)2. Turbine Building3. Intake Structure (including access tunnel)4. Off-gas Stack and Compressed Gas Storage Building5. Radwaste Building6. Diesel Generator Building7. Plant Control and Cable Spreading Structure8. Emergency Filtration Train (EFT) Building9. Diesel Fuel Oil Pump House10. Diesel Oil Storage TankPage 5 of 11

Flood preparations at the site begin with a flood surveillance procedure (Reference 6).During the time period of interest the surveillance was initiated by procedure annually inthe late winter. The procedure is currently performed monthly for river level predictionsand an annual performance includes inventory and inspection in addition to the riverlevel prediction. The purpose of this procedure is to determine if the potential for plantflood exist prior to and during the spring flooding season to ensure adequate steps aretaken to protect the plant if the potential for flooding exists. The actions taken inReference 6 are summarized as follows:* Based on the nature of the design basis flood (heavy snow pack,thawing/freezing cycle, coupled with heavy rain) the flood scenario is slowdeveloping and flood levels are generally predictable. Reference 6 determinesthe potential for flooding based on forecast information from the NationalWeather Service and river level monitoring. Procedure A.6 (Reference 7), Noteto Step 5.2.1, indicates that the National Weather Service Flow ExceedanceProbability Forecast on internet http://www.crh.noaa.qov is used to forecast riverelevations. The information for the St. Cloud and Anoka measurement stations isprovided on a weekly basis in terms of the probability that the river flow willexceed a given flow rate. The prediction information at the website is for the next90 days based on current conditions. A flow discharge curve in Reference 7 isused to determine predicted river water elevation based on the predicted flowrate. Given the conditions that precede the PMF; i.e., snowpack with thawing andrefreezing, it is reasonable to expect that the responsible individuals at the plant(engineering, operations, management) would be keenly aware of the need tomonitor river water elevations for predicted flood conditions. Increasedmonitoring and use of the predictive National Weather Service tools wouldincrease the time available to implement flood protective actions." Flood preparation measures are taken as part of Reference 6 to ensure that floodprotection materials such as sandbags, steel plates, covers and gaskets, andplugs are available. Contact information for vendors that would be used as part offlood preparation activities are confirmed to still be valid. This contact informationincludes vendors that would be involved with construction of the bin wall andearthen levee. These actions are implemented even if flood conditions are notpredicted. A memorandum of understanding is in place with VeitVeit & Company,a local construction firm, to provide construction related services in the event of asite emergency, and would cover activities such as construction of the earthenlevee." In the event that the potential for flood conditions, dump trucks and excavatorsare ensured to be available for installation of the levee, and a detailed flood planis developed.MNGP Procedure A.6 (Reference 7), "Acts of Nature," (Part 5 -.External Flooding)stipulates the actions to be taken in the event flood waters are predicted to exceedelevation 918 ft. Revision 41 through Revision 45 of Procedure A.6 (Reference 7) werein effect during the period of time from February 29, 2012 through February 15, 2013.Revision 41 was issued on February 28, 2012 and Revision 45 was issued on February14, 2013.Page 6 of 11

The following summarize the actions in A.6 based on the different predicted flood waterelevations.* Step 5.2.8, river level is predicted to exceed elevation 918 ft. Notification ofUnusual Event is declared. Actions are taken to protect equipment such as thedischarge structure substation.* Step 5.2.9, river level is predicted to exceed elevation 919 ft. Actions are taken toprotect the Intake Structure from flooding. As noted above the Intake Structure isat elevation 919 feet.* Step 5.2.10, river level is predicted to exceed elevation 921 ft. An Alert isdeclared and the plant is shutdown and cooled down to cold shutdownconditions. Actions are taken to ensure a supply of service water is available.* Step 5.2.11, river level is predicted to exceed elevation 930 ft. The bin walls andearthen levee are built. Steel plates are installed on the outside roof areas of theIntake Structure. Yard drains and other paths that could result in a water pathwaythat bypasses the levee are closed. An alternate access route to the plant isprovided from higher ground in the event that the normal access road is flooded.The levee is designed to provide flood protection up to a river elevation of 941 ft.Backup flood protection to the levee can be provided by closing up the variousbuildings using steel plates, installing sand bags, etc. It is noted that the levee isidentified in the procedure as the preferred option but, per the procedure, thebackup flood protection can be used in lieu of constructing the levee. This isdiscussed in more detail below." The remaining Steps 5.2.12 and 5.2.13 provide additional backup floodprotection for predicted river elevations above 930 feet. These are backup floodprotection measures to the levee.As described in A.6, Step 5.2.11, Note 2, the preferred flood protection measure isconstruction of a levee around the plant. The decision to use the levee as the preferredflood protection is based on a recommendation from the US Army Corps of Engineers(USACE), letter dated November 8, 2001. This USACE letter is referred to in the Basesdiscussion for Part 5 of Reference 7. However, Reference 7 includes an option forproviding flood protection in lieu of construction of the levee. This optional floodprotection means involves installing barriers (steel plates, etc.), sandbags, and sealingpenetrations. Resource loaded schedules developed in support of the A.6 proceduredemonstrate that the activities were achievable in the time required. Recent simulationsand demonstrations confirm the construction time for the bin wall, steel plate installationand sand bagging.As shown on Figure 13.10 of Reference 7, construction of the levee includesconstruction of a bin wall to the immediate east and west of the Intake Structure. The binwall was added as part of Revision 41 to A.6 on February 28, 2012. Prior to Revision41, the levee was made entirely of earthen material. The decision to use the bin wall wasbased on an analysis performed by Short Elliot Hendrickson, Inc., (SEH) (Reference 8).As part of this same change, the configuration of the levee was modified from a ringlevee entirely around the plant to a horseshoe design that ties into areas of the site thatPage 7 of 11

are above the peak PMF water elevation. The recommendation to use the bin wall wasmade as part of Reference 8 after considering various options for the tie to the IntakeStructure. Reference 8 included the following recommendations:* Secure a borrow source of levee fill within 15 minutes of the site or purchase andstore on site.* Purchase bin wall materials, assemble in modules to reduce installation timeframe, and store on site.The deficiencies identified in Reference 5 have subsequently been addressed. Inaddition to the noted deficiencies, other areas were also identified for improvement tothe plant and procedures. All of these areas for improvement were entered into the plantcorrective action system.Additional actions have been implemented to further improve the flood protection at thesite. These additional actions are summarized below:* Bin wall materials have been procured and are now stored on site. Theprocurement of the bin walls took approximately eight weeks; however, this wastreated as a normal procurement. The bin walls were supplied by ContechEngineered Solutions. Based on discussions with Contech Engineered Solutionsit is estimated that the bin wall sections could be provided in approximately 14days in an emergency situation. As discussed above, in the event that the binwalls cannot be constructed due to unavailability of materials, flood protectioncould still be provided as stipulated in the procedure using the sandbag and floodbarrier option. This option is independent of the levee and bin walls.* Levee materials have been procured and are now stored on site. Levee materialswere delivered to the site within four 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> shifts.* External flood surveillance procedure, 1478, (Reference 6) has been improved toincrease the frequency of river level monitoring during potential floodingconditions. The additional river level monitoring ensures timely plant preparationfor a potential flood. The increased river level monitoring serves to provide earlierwarning of predicted flood levels and increases available time to implementprotective actions.* Procedure A.6 (Reference 7) has been revised to improve the procedure clarity,remove unnecessary steps, and ensure completeness of protective actions." Detailed work instructions have been developed to implement actions in A.6.The work instructions provide the technical detail necessary to implement therequired action. Pre-staging the work instructions prior to the event reduces therequired time frames to implement the required actions in A.6.* Monticello conducted a self-assessment of the site flood protection response totake an additional critical review. The self assessment was performed by a teamof Xcel and contract professionals experienced in areas of flood protection.Page 8 of 11

Specific areas for improvement were identified during the self assessment andwere entered in the corrective action program and are being actively addressed.4.0 FLOOD MITIGATION STRATEGY -FURTHER DEMONSTRATIONSTable top walkthroughs of procedure A.6 have been performed to demonstrate feasibilityof performance of the required actions. A detailed schedule is developed for the actionsin A.6 using input from the site departments who would execute the actions. Theschedule shows actions to be performed, time frames, and sequencing, anddemonstrates that the actions can be completed within the available time period.Detailed work instructions have been developed to implement the actions in A.6. Pre-staging of the work instructions reduces the overall time to perform the tasks byremoving the time associated with work planning, identifies that materials that may beneeded to accomplish the work, and identifies any potential interferences orimpediments to completing the required task ahead of time.As described in Section 3.0, above, the materials to construct the bin wall sections andthe earthen levee have been procured and are stored on site. As previously discussed, amemorandum of understanding (MOU) is in place with VeitVeit & Company, a localconstruction firm, to provide equipment and services for construction of the bin wall andearthen levee.Reasonable simulation of construction of several of the actions believed to be more timeconsuming was performed in order to demonstrate that the actions could be performedwithin the available time frame. Specific actions examined were construction and fillingof the bin walls, construction of the steel plates around the roof of the Intake Structure,and filling of sandbags. The results from these reasonable simulations are summarizedbelow.* Construction of bin walls. For the reasonable simulation, approximately 5% of thetotal bin wall sections were constructed and filled. The simulation was contractedto Veit to add realism per our MOU and exercise the mobilization of personnel.The reasonable simulation took 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> for one crew to fully construct and fill.Based on using six crews, available per our MOU, to construct and fill the binwalls during implementation of procedure A.6, this would indicate that the entirebin wall sections could be fully constructed within 1.4 days. Accounting for issuessuch as excavation, inclement weather, security concerns, coordination, totalconstruction time of four days is reasonable.* Steel plates around roof of Intake Structure. As part of procedure A.6 steel platesare attached to the wall of the Intake Structure with anchors and the seamsbetween the plates welded to form part of the flood protection barrier. For thereasonable simulation, approximately 20% of the plates were installed on amock-up. The reasonable simulation took 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> and 11 minutes. Based on onewelder all of the plates could be installed within 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br />. Using two welderswould reduce this time to 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />. Additional welders would reduce this timeeven more. This time period is much less than the available time and providesmargin for working in inclement weather conditions.Page 9 of 11

Sandbagging. Per procedure A.6, approximately 100,000 sandbags are filled.Sandbags are used in several steps in A.6 to seal openings, provide backupflood protection. Sandbags are critical in the event that the sandbag and floodbarrier option were implemented in A.6 in lieu of the levee option. Reasonablesimulation indicates that 600 sandbags can be filled per hour using one machineand 8 people. Go-Baggers are a manual bagging apparatus with which anindividual can fill 55 bags an hour. With one machine and 20 people workingaround the clock, the 100,000 sandbags can be filled in 2 1/2 days. Reasonablesimulation also showed that a steel double door can be sandbagged by fivepersonnel in 34 minutes. Furthermore, it was shown that laying lumber andsandbagging 1 EDG room can be accomplished by 14 personnel in four hours.In the three cases discussed above, the reasonable simulation concluded that therequired actions can be accomplished within the available time frame.5.0 CONCLUSIONSThe following conclusions are drawn from the above discussion:* The postulated flood scenario for the MNGP is considered to be veryconservative. The methodology employed provides conservative results. Thiscan be seen from the comparison of river flow rates and water elevations inTable 1 in Section 2.1, above.* The flood is a relatively slow developing evolution that allows time for plant staffto monitor, predict and implement appropriate actions to provide the requiredflood protection.* The flood mitigation procedure clearly identifies actions for plant staff toimplement to provide the required flood protection.* In the event that the levee were not able to be constructed due to not having thebin wall materials available, the procedure provides an optional approach toimplement flood protection without relying on the levee and bin wall system.Table top review of the steps to implement this optional means of floodprotection demonstrated that the protection could be provided within theavailable time.* Subsequently actions have been taken to procure the bin wall and leveematerials. Reasonable simulation has demonstrated that the levee and bin wallsystem can be installed well within the available time.Page 10 of 11

6.0 REFERENCES1. Monticello Nuclear Generating Plant, Updated Safety Analysis Report (USAR), Revision 29.2. Department of the Army, Office of the Chief of Engineers, the U.S. Army Corps ofEngineers, Engineer Circular No. 1110-2-27, Enclosure 2, "Policies and ProceduresPertaining to Determination of Spillway Capacities and Freeboard Allowances for Dams,"dated August 1, 1966.3. NRC Letter to Licensees, dated March 12,2012, "Request for Information Pursuant to Title10 of the Code of Federal Regulations 50.54(f) Regarding Recommendations 2.1,2.3, and9.3 of the Near Term Task Force Review of Insights from the Fukushima Daiichi Accident"(ADAMS Accession No. ML 12053A340).4. NEI 12-07, Revision O-A, "Guidelines for Performing Verification Walkdowns of Plant FloodProtection Features," dated May 2012 (ADAMS Accession No. ML 12173A215).5. Xcel Energy Letter L-MT-1 2-097, "MNGP Final Response to NRC Request for InformationPursuant to 10 CFR 50.54(f) Regarding the Flooding Aspects of Recommendation 2.3 of theNear-Term Task Force Review of Insights from the Fukushima Dai-ichi Accident," datedNovember 27, 2012.6. Procedure 1478, "External Flood Surveillance," Revision 7. [Revision 7 is the procedurerevision currently in effect. During 2012, Revisions 4 through 6 was in effect and theprocedure was titled "Annual Flood Surveillance."]7. Procedure A.6, "Acts of Nature," Revision 46. [Revision 46 is the procedure revisioncurrently in effect. When used, previous revision numbers are identified in the text.]8. Xcel Contract No. 38398, SEH No. MONNE 117980, "Monticello Nuclear Generation Plant,External Flooding Plan Update: Alternative Analysis and Final Design Report," datedJanuary 5, 2012.Page 11 of 11

Enclosure 4Monticello Nuclear Generating Plant"Annual Exceedance Probability Estimates for Mississippi River Stages at theMonticello Nuclear Generating Plant based on At-site Data for Spring and SummerAnnual Peak Floods"12 Pages Follow

Annual Exceedance Probability Estimates for Mississippi River Stages at the Monticello NuclearGenerating Plant based on At-site Data for Spring and Summer Annual Peak FloodsDavid S. Bowles and Sanjay S. ChauhanRAC Engineers & EconomistsJune 28, 2013Purpose:To estimate the annual exceedance probabilities (AEPs) for Mississippi River Stages 917, 930 and 935 ft.NGVD 29 at the Monticello Nuclear Generating Plant (MNGP) using at-site data for spring and summerfloods.These estimates are intended to improve on the previous annual peak flood estimates that weresubmitted on April 8, 2013. The previous estimates were was based on an at-site flood frequency curveconstructed using a) a conservatively assigned AEP to the Harza spring PMF, and b) at-site floodfrequency estimates obtained from a drainage-area weighted interpolation between provisional USGSannual peak flood frequency estimates for the upstream and downstream Mississippi River gages at St.Cloud and Elk River, respectively.Given more time, we recommend that a Monte Carlo rainfall-runoff approach should be used to developestimates of extreme flood frequencies to make use of regional precipitation data and a morephysically-based transformation of rainfall to runoff, including snow melt and explicit consideration ofuncertainties.Available Information:Observed mean daily flows at the following locations:1) Station Number 05270700 Mississippi River at St. Cloud, MN with a period of record from 1989to 2012.2) Station Number 05275500 Mississippi River at Elk River, MN with a period of record from 1916to 1969.3) Monticello Nuclear Generating Plant (at-site) with a period of record from 1970 to 2012.Flood frequency estimates of annual peak discharges:4) Provisional 2013 USGS annual peak discharge flood frequency analyses with estimates of AEPsranging from 1 in 1.005 to 1 in 500 based on maximum daily flow rates for the annual peakflows:1

a. Station Number 05270700 Mississippi River at St. Cloud, MN with drainage area of13,320 sq. miles.b. Station Number 05275500 Mississippi River at Elk River, MN with drainage area of14,500 sq. miles.Probable maximum flood (PMF) peak discharge and stage estimates:5) Harza 1969 (spring) PMF peak discharge and river stage at the Monticello Nuclear GeneratingPlant -1912 Datum.6) Bechtel 20121 spring and summer PMF peak discharges and river stages at the MonticelloNuclear Generating Plant -NAVD88 Datum.Discharge rating relationships:7) Harza 1969 relationship between river stage (1912 Datum) and river discharge (cubic feet persecond, cfs) over the range 26,000 to 437,000 cfs.8) Ops manual equation between river discharge (cubic feet per second, cfs) up to 4,000 cfs andriver stage (NGVD 29 Datum): Q = 122(Stage -901)2.2.Datum conversions for river stages:9) NGVD 29 Datum = 1912 Datum -0.36 ft.10) NGVD 29 Datum = NAVD 88 Datum -0.4 ft.Procedure:The following two approaches were examined for developing improved at-site estimates of annualexceedance probabilities (AEPs) for the spring and summer annual floods in the Mississippi River at theMNGP:1) Drainage-area weighted interpolation of flood frequency estimates for the upstream anddownstream USGS gages: Similar to the April 8, 2013 approach, an at-site flood frequency curvewas obtained from a drainage-area weighted interpolation between flood frequency estimatesfor the upstream and downstream Mississippi River gages at St. Cloud and Elk River,respectively. However, this revised approach was conducted separately for spring and summerannual peak floods and it did not conservatively assign an AEP to the Harza PMF as was done inthe April 8, 2013 approach. Instead the flood frequency curve was extrapolated to extremefloods thus providing estimate of the AEPs for the spring and summer PMF peak flow estimatesand for the three elevations of interest.1 The report, Bechtel 2012 spring and summer PMF peak discharges and river stages at the Monticello NuclearGenerating Plant -NAVD88 Datum, has been provided to NSPM. This study provides bounding estimates to sitepeak flood elevations applicable to the development of annual exceedance probabilities at the Monticello NuclearGenerating Plant.2

2) Flood frequency analysis based on at-site flow data: A flood frequency analysis was conductedon the available at-site streamflow data for the period 1970 to 2012 with extrapolation toextreme floods. This provided estimate of the AEPs for the three elevations of interest and forthe PMF peak flow estimates for spring and summer annual peak floods.Both of the above approaches included estimating separate flood frequency relationships for spring andsummer annual peak floods. Data for estimating these relationships were obtained using the followingdefinitions of spring and summer floods based on discussions in the Hydrologic Atlas of Minnesota (Stateof Minnesota 1959) and examination of the flow records:1) Spring annual peak floods generally peaked in the period March to May, but if it was clear fromexamination of the hydrograph that a snow melt flood event peaked in June then that peak wasused.2) Summer annual peak floods generally peaked in the June to early October period, but floodpeaks occurring in June, which were clearly associated with snow melt events, were excluded asmentioned in 1). Since the recession limb of the annual snow melt hydrograph extends throughthe summer, the peak flow rates for summer floods, which are associated with convectivestorms, are dependent to some degree on the magnitude of flow on this recession limb at thetime of the summer flood.The two approaches are discussed in more detail below.Approach 1): Drainage-area weighted interpolation of flood frequency estimates for the upstream anddownstream USGS gagesThe provisional USGS annual peak flow flood frequency estimates for the Mississippi River gages at St.Cloud and Elk River were verified using USGS flood frequency software following Bulletin #17B floodfrequency analysis procedures (USGS 1982). The following softwares were applied to the maximumdaily annual peak streamflow data assembled by the USGS to verify their results:1) PeakFQ: Bulletin #17B procedure based on method of moments parameter estimation for a LogPearson Type 3 probability distribution (Flynn et al 2006)2) PeakfqSA: The more efficient Expected Moments Algorithm (EMA) applied to the Bulletin #17Bmethodology (Cohn 2012)Following the USGS provisional analysis, the Bulletin #17B (PeakFQ) software was applied to the St Cloudgage and the EMA (PeakfqSA) software was applied to the Elk River gage for the verification step.Separate flood frequency analyses were then conducted for spring and summer annual peak flow dataat the St. Cloud and Elk River USGS gages. The maximum daily annual peak streamflow data wereassembled by the USGS and provided with their provisional flood frequency analyses. These datacomprised a mixture of spring and summer floods. These data were separated into spring and summerfloods and mean daily peak flow data were obtained from USGS flow records at both gages for thosecases that were not covered by the maximum daily annual peak streamflow data assembled by the3

USGS. An additional year (2012) of data was added for the St Cloud gage. Maximum daily annual peakflows were estimated from mean daily annual peak flows for those data not included in the USGSprovisional analyses using regression relationships established between maximum daily annual peakflow data and mean daily annual peak flow data for spring and summer flows for each gage.The PeakfqSA software containing improved EMA parameter estimation (Stedinger 2013) was applied toboth spring and summer flood data for both gages. No outliers were identified. The EMA softwareprovided AEP estimates from I in 1.0001 to I in 10,000. Estimates of annual peak discharge on theMississippi River at the MNGP were then obtained based on linear interpolations between variousfrequency (AEP) estimates developed for the St Cloud and Elk River gages as a function of drainage areawith the drainage area at the MNGP being 14,071 sq. miles. This interpolation the procedure is thesame as developed for the April 2013 AEP estimates. The at-site annual peak discharge estimates wereconverted to at-site river stages using a combination of the Harza and Ops Manual equation ratingcurves shown in Figure 1.Examination of the relationship between flood estimates for various AEPs and drainage area shown inFigure 2 showed an inconsistent relationship that was increasing or decreasing with drainage area. As aresult we have not relied on these estimates in favor of using the flood frequency estimates obtainedfrom analysis of at-site data in the second approach.947942937* 9320JS927 _______.5922 __________0JS917912907 1 1 _________9021100 1,000 10,000 100,000 1,000,000Discharge In cfs-Ops Man Eqn, Q= 122(Stage-901)^2.2 -Harza Discharge Rating -- Transition -.Final curveFigure 1. Combined Harza and Ops Manual Equation stage-discharge rating curve4

80,00070,00060,00050,0000 40,000E.30,00020,00010,000Interpolation of at-site quantiles-TFU-------- ----- ----*0.995-0.99-0.95-0.9.--0.6667ý0.5--*-0.4292-o-0.2-0.04-0.02-.--0.01-0-0.005-4-0.002-MNGS5,000U13,00013,50014,000Drainage Area (sq. miles)14,600IFigure 2. Relationships between flood estimates for various AEPs and drainage area for Approach 1Approach 2): Flood frequency analysis based on at-site flow dataThe second approach is based on extrapolation of flood frequency relationships developed from at-siteflow data. The mean daily annual peak stages were obtained for spring and summer annual peak floodsfollowing the process summarized above. Since only single observations have been recorded for eachday it was not possible to obtain maximum daily annual peak stages. The daily annual peak stages wereconverted to daily annual peak flows for spring and summer floods using the combined rating curveshown in Figure 1. The EMA (PeakfqSA) software was applied to estimate the flood frequencyrelationships for spring and summer annual peak floods. No outliers were identified for the springseason, but one low outlier (2,416 cuffs) was identified for the summer season using the MultipleGrubbs-Beck Test) low outlier identification method. The EMA software provided AEP estimates for therange 1 in 1.0001 to 1 in 10,000.Annual Exceedance Probability Estimates:Figures 3 and 4 show the resulting flood frequency estimates for the spring and summer annual peakfloods obtained from the second approach. The annual peak discharge is plotted on a Log scale andAEPs are plotted on a z-variate scale (corresponding to a Normal probability distribution). In addition to5

Monticello NGS -Spring1,000,000B el 2012 PMF estim tean--- ----a--T9. nbrPes-Tma --e .--., e.0l vafion 93SEl, -atlon 930100,000 ...... ...-, -..-'":! ... ""h...=-- -- ~ ~... :...........10,000 v-Exce dance Probability1E I 1E-1 1E-2 E-3 1E-4 iE-5 1E-6 1E-7 1E-8 1 91,000 .I I I" I 1- 1-4.0 -3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0z-variateFigure 3. Spring Annual Peak Flood Frequency (approximate mean shown by black dashed line)6

Monticello NGS -Summer1,000,000N024)4)4)a.100,00010,0001,000-4.0 -310 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0z-variateFigure 4. Summer Annual Peak Flood Frequency (approximate mean shown by black dashed line)7

providing median (50th percentile) and approximate mean2 estimates, the 20% (40th and 60thpercentiles), 40% (30th and 70th percentiles), 60% (20th and 80th percentiles), 80% (10th and 90thpercentiles), and 90% (5th and 95th percentiles) confidence interval estimates are provided with linearextrapolation to smaller AEPs beyond 1 in 10,000. The site elevations of 917, 930 and 935 are shown byhorizontal lines based on the NGVD 29 datum matching the datum used on the site drawings. Also theHarza and Bechtel PMF estimates are shown by horizontal lines corresponding to peak elevations forthese events.It is noted above that only single daily river stage observations have been recorded at site and thereforeit was not possible to obtain maximum daily annual peak stages. Since the difference between meandaily and maximum daily peak stages decreases with smaller AEPs, it is likely that the flood frequencyrelationships are slightly steeper than shown. This effect would tend to make AEP estimates forextreme flows slightly conservative (i.e. slightly larger) than of this effect were removed.Tables 1 and 2 contain numerical AEP estimates for spring and summer annual peak floods, respectively,for river stages 917, 930 and 935 ft. NGVD 29 for the median (50th percentile) and approximate mean,and for various confidence percentiles.Table 1. Spring Annual Peak Flood Frequency EstimatesElevation 91795th I ge 1 84th 1 0th I 70th I 60 I Median 50e 5th Approx MeanAEP Estimates 3.4E-02 2.5E-02 2.0E-02 1.7E-02 1.2E-02 8.9E-03 6.3E-03 5.9E-07 7.9E-03i in Tyears 29 40 51 59 82 112 158 1,680,000 127Elevation 93095t' 90 84th 80' 70 60h Median 50 5t Approx MeanAEP Estimates 7.OE-06 7.6E-07 7.5E-08 1.7E-08 < 1E-9 < 1E-9 < 1E-9 < 1E-9 < 1E-91inTyears 143,0001 1,320,0001 13,400,0001 60,500,0001 >1E+9 I >1E+9 [ >1E+9 I >1E+9 [ >1E+9I _Elevation 935F-9 90 o 84th 806 70& 60' Median 506' 51 Approx MeanAEP Estimates 4.3E-07 2.OE-08 < 1E-9 < 1E-9 < 1E-9 < 1E-9 < 1E-9 < 1E-9 < 1E-91 in T years 2,350,000 51,100,000 >1E+9 >1E+9 >1E+9 >1E+9 >1E+9 >1E+9 >1E+9Table 2. Summer Annual Peak Flood Frequency Estimates___Elevation 9179Sth 90t 84th 80 th 70 th 60d Median 50th 5th Approx MeanAEP Estimates 1.9E-02 1.3E-02 8.8E-03 7.2E-03 4.4E-03 2.8E-03 1.7E-03 8.5E-07 3.OE-031 in Tyears 52 79 113 140 225 358 590 1,180,000 328_ _Elevation 930_95_ 90go 84th 80eh 70th 60e Median 506h 5th Approx MeanAEP Estimates 2.1E-04 4.3E-05 8.8E-06 3.1E-06 2.4E-07 1.5E-08 < 1E-9 < 1E-9 1.6E-061in Tyears 4,7201 23,4001 11,00 324,000 4,160,000 65,100,000 >1E+9 >1E+9 641,000I-_ Elevation 93595h 90 o 841 80h 70 60'h Median 50' 5 h Approx MeanAEP Estimates 6.1E-05 7.9E-06 9.7E-07 2.5E-07 8.9E-09 < 1E-9 < 1E-9 < 1E-9 2.OE-07I in T years 16,500 127,000 1,030,000 4,000,000 112,000,000 >1E+9 >1E+9 >1E+9 1 4,930,0002 The approximate mean estimates were obtained by weighting the various percentile estimates by theirrespective intervals of probability that each represents. For example, the 60th percentile represents the intervalbetween the mid-points of the 50th -60th and 60th -70th percetile intervals and hence is weighted by thedifference between the percentiles associated with the mid-points of these two intervals, i.e. 0.65-0.55 = 0.10.8

The following is a summary of the median estimates and ranges (Upper -95th percentile and Lower -5thpercentile) of the AEP estimates for the river stages of 917, 930 and 935 ft. NGVD 29 at the MonticelloNuclear Generating Plant for spring and summer annual peak floods and for the Harza and Bechtel PMFestimates. The April 8, 2013 spring estimates are shown in italics for comparison. The comparisonshows that these estimate were conservative relative to those obtained using the second approach.Spring Floods:Elevation 917 ft. NGVD 29:* Upper (95th): 3.4E-02 (1 in 29/year) 4.OE-02 (1 in 25/year)" Median (50th): 6.3E-03 (1 in 158 /year) 7.2E-03 (1 in 140/year)* Lower (5th): 5.9E-07 (1 in 1,680,000 /year) 9.5E-04 (1 in 1,100/year)Elevation 930 ft. NGVD 29:* Upper (95th): 7.OE-06 (1 in 143,000 /year) 1.6E-04 (1 in 6,300/year)* Median (50th): < 1E-9 (1 in >1E+9 /year) 1.6E-05 (1 in 61,000/year)* Lower (5th): < 1E-9 (1 in >1E+9 /year) 2.2E-06 (1 in 460,000/year)Elevation 935 ft. NGVD 29:* Upper (95th): 4.3E-07 (1 in 2,350,000 /year) 3.OE-05 (1 in 33,000/year)* Median (50th): <1E-9 (1 in >1E+9 /year) 3.1E-06 (1 in 330,000/year)* Lower (5th): <1E-9 (1 in >1E+9 /year) 3.4E-07 (1 in 2,900,000/year)The Harza Spring PMF AEP estimates are shown below with the April 8, 2013 AEPs assigned to the Harza(spring) PMF shown in italics for comparison.* Upper (95th): 5.9E-08 (1 in 16,900,000) 1 in 10,000,000* Median (50th): < 1E-9 (1 in >1E+9 /year) 1 in 1,000,000" Lower (5 h): <1E-9 (1 in >1E+9 /year) 1 in 100,000The estimates of AEPs assigned to the Harza PMF in the April 8, 2013 work are therefore confirmed tobe conservative (i.e. larger than now estimated).The AEP Bechtel spring PMF estimates are shown below:" Upper (95th): 1.8E-08 (1 in 54,500,000)* Median (50th): < 1E-9 (1 in >1E+9 /year)* Lower (5th): < 1E-9 (1 in >1E+9 /year)9

Summer Floods:Elevation 917 ft. NGVD 29:* Upper (95th):* Median (50th):* Lower (5th):1.9.4E-02 (1 in 52 /year)1.7E-03 (1 in 590 /year)8.5E-07 (1 in 1,180,000 /year)Elevation 930 ft. NGVD 29:* Upper (95th):* Median (50th):" Lower (5th):2.1E-04 (1 in 4,720 /year)< 1E-9 (1 in >1E+9 /year)<1E-9 (in >1E+9/year)Elevation 935 ft. NGVD 29:000Upper (95th):Median (50th):Lower (5th):6.1E-05 (1 in 16,500 /year)< 1E-9 (1 in >1E+9 /year)< 1E-9 (1 in >1E+9/year)The AEP Bechtel summer PMF estimates are shown below:* Upper (95th):* Median (50th):* Lower (5th):5.7E-04 (1 in 1,750/year)3.3E-08 (1 in 29,900,000 /year)< 1E-9 (1 in >1E+9/year)The second approach used to develop these revised AEP estimates is preferred to the initial approachused to develop our April 2013 estimates for the following reasons:1) It separates the spring and summer flood events.2) It relies on at-site date rather than a drainage-area weighted interpolation of AEP estimates atupstream and downstream Mississippi River USGS gages.3) It does not rely on an assignment of an AEPtothe PMF.A graphical comparison of the April 2013 and the current estimates is presented in Figure 5. It indicatesthat the April 2013 AEP estimates are likely overly conservative as a result of the assignments of theAEPs to the Harza PMF.Figure 5 is similar to Figure 3 but includes the USNRC (2013) AEP estimates: 9.37E-05 for Elevation 930and 2.72E-05 for Elevation 935 (based on 6.65E-05/year for Elevation 930-935). The NRC estimatesexceed our current 95th percentile estimates but are very similar to our April 2013 95th percentileestimates, which were based on assigning an AEP of 1E-5 to the Harza PMF. According to USNRC (2013)their estimates are based on flood frequency estimates from the Monticello USAR and IPEEE but we arenot clear about the origin of those estimates or the curve fitting approach that was used by the USNRC(2013). Therefore it is not possible to make a more informed comparison with our estimates; although it10

Monticello NGS -Spring1,000,000z99LaJtU.X100,00010,0001,000-4.0-3.0 -2.0 -1.0 0.0 1.0 2.0 3.0 4.0 5.0z-variate-April 2013 Best Estimate --April 2013 Upper Estimate --April 2013 Lower Estimate * NRC Estimates -Approx Mean6.0Figure 5. Comparison of the current Spring Annual Peak Flood Frequency with the April 2013 estimates and the NRC (2013) estimates11

would appear that our estimates rely on more recent site-specific data than the NRC had available. Inaddition they use the Bulletin #17B flood frequency approach, which is the standard for flood frequencyanalysis in the US. We also used the improved EMA parameter estimation approach that will beincluded in the revision of the Bulletin #17B procedure that has been drafted.However, we recommend that a Monte Carlo rainfall-runoff approach be considered to developestimates of extreme flood frequencies with explicit consideration of uncertainties in the future. Thisapproach can be expected to provide improved estimates based on the use of regional rainfall analysis,and a more physics-based representation of the rainfall-runoff (including snow melt) processes forextreme floods than is associated with the current extrapolation approach.ReferencesCohn, T. 2012. User Manual for Program PeakfqSA Flood-Frequency Analysis with the ExpectedMoments Algorithm DRAFT. September.Flynn, K.M., Kirby, W.H., and Hummel, P.R., 2006, User's Manual for Program PeakFQ Annual Flood-Frequency Analysis Using Bulletin 17B Guidelines: U.S. Geological Survey, Techniques andMethods Book 4, Chapter B4; 42 pgs.State of Minnesota. 1959. Hydrologic Atlas of Minnesota. Davison of Water, Department ofConservation, State of Minnesota.Stedinger, J.R. V. Griffis, A. Veilleux, E. Martins, and T. Cohn. 2013. Extreme Flood Frequency Analysis:Concepts, Philosophy and Strategies. Proceedings of the "Workshop on Probabilistic FloodHazard Assessment (PFHA)" sponsored by the U.S. Nuclear Regulatory Commission's Offices ofNuclear Regulatory Research, Nuclear Reactor Regulation and New Reactors in cooperation withU.S. Department of Energy, Federal Energy Regulatory Commission, U.S. Army Corps ofEngineers, Bureau of Reclamation and U.S. Geological Survey organized. Rockville, Maryland.January 29 -31.USGS (US Geological Survey). 1982. Guidelines for Determining Flood Flow Frequency. Bulletin #17B,Hydrology Subcommittee, Interagency Advisory Committee on Water Data, Office of Water dataCoordination.USNRC (U.S. Nuclear Regulatory Commission). 2013. Monticello Nuclear Generating Plant, NRCInspection Report 05000263/2013008; Preliminary Yellow Finding.12

Enclosure 5Monticello Nuclear Generating Plant"Stakeholder Outreach"I Page Follows

Stakeholder OutreachNSPM hosted an open house on Thursday, June 6, from 4 p.m.-8 p.m. to shareinformation with its community neighbors on operations and preparedness to handlepotential emergencies and how we would respond to flooding, earthquakes and otherunforeseen challenges. The Site employed numerous methods to publicize the event:personal, direct invitations to community leaders, a full page ad was purchased inweekly newspapers, a news release was distributed to local media and 14,000postcards were mailed to neighbors in surrounding communities. The outreach eventhad full corporate support and the Xcel Energy Chairman, President and CEO, and theChief Nuclear Officer attended, as well as numerous senior members of the corporatenuclear staff. The Monticello Site Vice President and Plant Manager were also joined bythe site's senior leadership team at the event.A total of 515 persons from Monticello and surrounding communities attended the eventat the Monticello Training Center.The key message presented to visitors was that safety and security at the NSPM nucleargenerating plants are top priorities for Xcel Energy. Further, that we understand theNRC's increased scrutiny of safety and flood preparedness at the nation's nuclear powerplants in the wake of events such as 9/11 and Fukushima Daiichi. The Monticello FloodProtection Strategy was identified and explained to demonstrate that the site is designedto withstand a hypothetical flood beyond anything reported in the Monticello area. Thebroad underlying key messages were reinforced and manifest in specific subject itemssuch as: B.5.b Pump/Electrical Generator/Trailer, Portable Emergency ResponseEquipment, Backup Power Sources including description of backups to the backup(Battery Systems) and the continual focus on improving emergency preparednesscapabilities.Page 1 of 1