ML16272A217

Revision 1 to Flood Hazard Reevaluation Report on Oconee Nuclear Station, Pp. 1-70, Dated March 6, 2015 (Redacted)

ML16272A217
Person / Time
Site:	Oconee
Issue date:	03/06/2015
From:	HDR Engineering
To:	Office of Nuclear Reactor Regulation
Shared Package
ML16070A295	List: ML16070A284 ML16070A287 ML16070A288 ML16070A290 ML16272A216 ML16272A217 ML16272A218 ML16272A220 ONS-2015-028, Supplemental Information Re NRC 2008 and 2012 Request for Information Pursuant to 10 CFR 50.54(f) Pertaining to External Flooding, Dated March 6, 2015 (Redacted)
References
FOIA/PA-2016-0071
Download: ML16272A217 (83)

Text

ML13UfZAlU(

_^__ __ t*... C*

.*---i... h -- l . i* .... ntzn 8!-- .*Lt--I... I. fz . t .. l..lti- ----. 10l CFRIm 3l*l~fl.?OO l)

Enclosure 1 Revision 1 to Flood Hazard Reevaluation Report Oconee Nuclear Station Revision 1 Summary

1. Revisions made as applicable to a revised Jocassee Dam Break flood analysis which applied multiple alternate breach-parameter estimations in accordance with JLD-ISG-201 3-1.

2. Revisions made as applicable to incorporate the Jocassee specific seismic analysis into the report (satisfies RAl 1 from March 20,2014 RAI).

3. Addresses revised responses to RAls 11 - 15 from the March 20, 2014 PAI.

4. Changes to errata in the use of the terms "licensing basis" and/or "design basis" throughout the document.

Sections Affected by Revision 1:

• Executive Summary

• Section 2.3

• Section 2.8

• Section 3.3

• Section 4

• Section 5

• Section 6

• Appendix D

~.uI~uIuU ~~1at up ~wuis;iJvu IIIUJuIutUlIUuI, WuLJII;UHZ uluill JJUUIH.. 5151 IV ~FI~ £..iW~JtUI~ ~I FLOODING HAZARD REEVALUATION REPORT Oconee Nuclear RESPONSE TO REQUEST FOR INFORMATION PURSUANT TO TITLE 10 OF THE CODE OF FEDERAL REGULATIONS 50.54 (F) REGARDING RECOMMENDATION 2.1: FLOODING OF THE NEAR-TERM TASK FORCE REVIEW OF INSIGHTS FROM THE FUKUSHIMA DAI-ICHI ACCIDENT Seneca. South Carolina

E .......................... *lf

,,; ...... U....

................ 1

............. .. I.,I; .... .... 2.nn*;u;; ;*,

Prepared for:

DUKE ENERGY CAROLINAS, LLC Charlotte, North Carolina Prepared by:

HOR ENGINEERING, INC. OF THE CAROLINAS Charlotte, North Carolina Revision 1 January 29, 2015

Duke Energy Carolinas, LLC I ONS FLOODING ~~REPORT HAZARD REEVALUATION REPORT VERIFICATION 'lJ¶L,'

REPORT VERIFICATION PROJECT: ONS FLOODING HAZARD, REEVALUATION REPORT REVISION 1 TITLE: RESPONSE TO REQUEST FOR INFORMATION PURSUANT TO TITLE 10 OF THE CODE OF FEDERAL REGULATIONS 50.54 (F) REGARDING RECOMMENDATION 2.1:

FLOODING OF THE. NEAR-TERM TASK FORCE REVIEW OF INSIGHTS FROM THE FUKUSHIMA DAI-ICHI ACCIDENT This ,document has been reviewed for accuracy and quality commensurate .withthe intended application.

Prepared b-: J. ChritphrE,

- ,P.E.- 7 6.J.AZ1**Dae: 112121 Checked.by: Tim Banta, P.E. 'ot J 4 L. 1 Dae 12901 Qualhty Review by: - Angie Scangas, PE.E *t***A('uk 5 CfaA5:o4.

,*AJ J* Date: 1129/2015 9

Approved by: . J. Christophere P.E. *,*Date: 112912015 This document has been reviewed for accuracy and quality commensurate with the intended application.

Corporate Seal: Professional Engineer Seal:

HDR Engineering, Inc. *of the Carolinas 440 S. Church Street, Suite 900 Charlotte, NC 28202 South Carolina License No. C03i 8

"QI 3°.. "~ .. ....~nv~iI~m~u .n. lu... uH. m Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT

~r Table of Contents Table of Contents*

REPORT VERIFICATION.............................. ............ , ...... , ................ .. :.... . .... ...

Table of Contents.... ....................... ., .. ........  ;...*......... .*............... ........ i List of Figures .... .... ;........................;................,....... ...................... ...... *..........

• List of Acronyms . .......................................................................................... vii E xec utive S u m m a ry . . . ... ..... . . . . . . . . . . . . . . . . . . . . E S-I Section 1 - Site Information Related to the Flood Hazard ....... ....... .................... *......... 1 1.1 Detailed Site Information!.. ... ., ...... ..... ..... .. ,................ ,.......................14 1.2 Current Licensing Basis Flood Elevations .. .. .. ... ......................... . ........*.....

1.2.1 Local Intense Precipitation..................................,. .. ............ 1 1.2:2 Flooding in Reservoirs ... . ....... : .......... ........ .. ..... : .. .. 5 -

1 .2 .3 Da m Fa ilu res . . *..... . . . . . . . .... . . ... . . . . . .. . . . . . . ..... . . ....... .. . .,... .. . . ... . . . . . . . . . . 6 1.2.4 Storm Surge and Seiche .............................................. 6 1.2.5 Tsunami ...... ,............................................................................ 6 1 .21.6 Ice-Induced Flooding.........................................,... ......................... 6 1.2.7 Channel Diversion .................................................... ,..................... 7 1.2.8 Combined Effects ..................................... ,.... ........... 7 1.3 Licensing Basis Flood-Related and Flood Protectioni Changes ..... *.... ,...*.........,..7 1.4 Watershed and Local Area Changes ........... ,............; .... ... . .... .. 8 1.5 Current Licensing Basis Flood Protection and Mitigation Features .... ,;. ........ .9 Section 2 - Flooding H azard Reevaluation ................ ,......... ............... ,...1

. ............... i 2.1 Local Intense Precipitation ................................................................... 11 2.2 Flooding in Reservoirs : ...................... .... .. .................... 17 2.2..1 Probable Maximum Flood - Keowee 1..........................

2..2.2 Probable MaXimum Flood - Jocassee ................................. ,......,..........24 2.3 Dam Failures .................. ,.. ...... *.... ....... , .. .. ,.,.......................... *........ 31 2.3.1 Potential Dam Failure ...... *.............:.. .. . ....... ,..,.................,............. 31 2.3.2 Dam Failure Permutations .... ......... ., .......... ,. .......... ,.............. *........ .32 2.3.2.1 Potential Hydrologic Overtopping .............. ,............,........................... 32 2.3.2.2 Potential Seismic Failure .................  ;.......................................... 32 JI

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Table 0o"Contents 2.3.2.3 Site Geology Characteristics ...... ,.................,.................. ,.................. 33 2.3.2.4 Seismic Hazard Methodology Description ..... ... ,..**,... ... .... ..., .... 36 2.3.2.5 Seismic Hazard for Hard-Rock Conditions ..... .............. ,..... ,................ 36 2.3.2.6 Site Response Analysis .... **.................................,........................... .37 2.3.2.7 Seismic Hazard for Soil Conditions........................... .. ,....... ;......*........37 2.3.2.8 Seismic Performance Evaluation ......... ;....... ......................  ;............. 39 2.3.2.9 Seismic Performance.................i............................................... 40 2.3.2.1.0 Appurtenant Structures ....................... ...... .............................. 41 2.3.3 Unsteady Flow Analysis of Potential Dam Failures ............................ ...... 42 2.3.4 Water Level at the Plant Site..,....,.... ............... :...,  :...................... ,...... 52 2.5 Tsunami ...................... ........................

,.*..... ,....... .................... ...... ... ,....56 2.6 Ice-Induced Flooding ............................... ....... ,................................... 56 2.6.1 Ice.

e Efe ts..... ...... .Effects.., .. . ...... *............ ............ .  ;.5

.... .56.

2.6.2 Ice Jam Events .................................... ........... ,. ........ ........ 56 2.7 Channel. Diversions...................................................................... ........56 2.8 Combined Effects..... ............. .......................................................... 56 Section 3 - Comparison of Current Licensing Basis .and Reevaluated Flood-Causing Mechanisms ... *..... .. ............. . ...... ......................... *.....,...........,,.................. 58 3.1 Local Intense Precipitation. ... ,......... ................................................. ,.....58 3.2 Probable Maximum Flooding ....................................... ......................... 59 3.3 Dam Failures ... .......

• • .................................. *........ ................. 59 3.4 Storm Surge and Seiche ........................ ..... ,. .... . ....... 59 3.5 Tsunami ........................ , ............................................. .,....., ........ ... 59 3.6 Ice-Induced Flooding ............. ,,............. .............................................. 59 3.7 Channel Diversion................ ...........

.......... ,................. .,.........* ........ * .. 59 3.8 Combined Effects .....I............ ...... ;......................*........;....................... .59 Section 4 - Interim Evaluation and Actions Taken or Planned..................................... 60 Section 5 - Additional Actions.............................,........................ ..................... 61

5.1 RAI-14

Hazard input to the integrated assessment: Flood event duration parameters 61 5.1.1 Response to pRAl-14 ...............,...... * ...........

, .... .. ....... ,..:....,,............ 62 5.2 RAI 15: Input to integrated assessment: Flood height and associated effects., ....... 63 ill

L.onlalns oe~uriiy ae,,uiuvu  ;,.,,,.v, w,,,,  ;,~ .~....Lv;;, *... -i. ..... *.Z3O(J)1)

Duke Energy Carolinas,, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Tabre of Contents 5.2.1 Response to RAI 15....................,.................................................. .64

.Section 6-- References.................. ... ......-.............................. :......................... .67 Reference CalcUlations: ..... .*.. .. ..... ........ *:.....;...................... ,........

.... ..... 70 APPENDICES APPENDIX A - FIGURES AND TABLES APPENDIX B - CURRENT LICENSING BASIS, LIP MODEL ANALYSIS INUNDATION LEVELS APPENDIX C - FLOOD HAZARD REEVALUATION BEYOND DESIGN BASIS LIP MODEL ANALYSIS INUNDATION LEVELS APPENDIX D - DAM BREACH MODEL UNSTEADY FLOW MODELING.DETAILS 9

iv¸

3 C,-,

.. ,. ,.uIty.*,aaI;aIvtr .,ai,h,aveu,;ul,, wmu;[~uuu .zfuzJu puUDiw per ,u *,rn* -,,*u(/l ='

Duke Energy Carolinas. LLC I ONS FLOODING HAZARD REEVALUATION REPORT List of Figures List of Figures Figure 1. Probable Maximum Precipitation Depth Duration Curve .... ............................ 13 Figure 2L incremental precipitation during the 6-hour PMP for six different temporal distributions

........................................ '............................................................... 14 Figure 3. Incremental precipitation during the 72-hour PMP for six different temporal distributions........*............ .... .................................................. *......... 15 Figure 4. PMP DAD curves for Keowee watershed PMP depth vs. Duration: .................... 18, Figure 5. PMP DAD curves for Keowee watershed PMP depth vs. Drainage Area .............. 19 Figure 6. Keowee watershed sub-basins ............................... ,...............*........... ..... 22 Figure 7* Table 6.2 PMS from HMR 52 which generates PMF..................................... 23 Figure 8. KeoweePMF hydrographs ................... ,...........;............. ........................... 24 Figure 9. Jocassee PMP DAD curves PMP depth vs. Area....... ................................. 25 Figure 10. Jocassee watershed sub-basins ................................................ .. * ...... 27 Figure 11. Jocassee PMF hydrographs ............................................................. 31 Figure 12. Map of Jocassee Main Dam location showing Latitude and Longitude. North is to the top of the figure.....;.............. .............................. ..... ,...................... 34 Figure 13. Composite profile 1 representing geologic conditions below the control point elevation at Jocassee Dam (LCI, 2014). Profiles JD-1 and JD-5 are individual profiles at Jocassee from AMEC (AMEC, 2014b) that are combined into, composite 1. "Lower" and "Upper" profiles correspond .to the lower-range and upper-range cases, respectively, to capture epistemic uncertainty in mean shear wave velocity per the SPID (EPRI, 2013a)................ .........  :................................................ 35.

Figure 14. Site-specific mean soil hazard curves for seven sPectral frequencies at Jocassee (LCI, 2014) ............. ,........ ..............  ;.............................. ... ........... 38 Figure 15. Horizontal mean UHRS for MAFEs of 1 0 -4,10"s, and 10"e at Jocassee (LCI, 2014). ,39 V

• Duke Energy Carolinas, LIC I ONS FLOODING HAZARD REEVALUATION REPORT List of Tables List of Tables Table 1. Reservoir Flooding Results...................*............................................... 5 Table 2. Calculation Results................. ........................................... ...... ......... 6 Table 3. Probable Maximum Precipitation Depths for the Site Using HMR 51 and 52 ........... 12 Table 4. 5-Minute Interval Duration Depths ....... ................................................... 13 Table 5. Keowee Damn - Normal and probabie-maximum flood freeboard (ft)......................18 Table 6. Jocassee Dam - Normal and probable maximum flood freeboard (It)................... 24 Table 7. Final ,Jocassee Dam Breach Parameters ................................................. 48 Table 8. Final Jocassee Dam Breach Formation Pattern (HEC-RAS Input) ...... .......... 49 Table 9. Comparison of 1-D and 2-D model results - time to breach and maximum water surface elevations (Elevation in ft msl and Time in hours) ... .. , ....................... 54 Table 10. Wind-driven wave run-up results .................. .............. ................. ............ 55 Table 11. 2-year wind velocity results ..................... ........ *.................................. 55 Table 12. Combined effects flood elevations... ..... ,... .................................... ...... ..... 57 Table 13. Current licensing basis and reevaluation flood elevations .............................. 58 vi

-- - - i,- C..:t...._ Lf;......tL,- , .1t'iht.r'l~ frtr .... kri, *r IAl r~Q'2"* Qfll~l4~1 Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT List of Acronyms List of Acronyms BWST Borated Water.Storage Tank CAL Confirmatory Action Letter CAP Corrective Action Program CCW Condenser Circulating Water cfs cubic feet per second CM compensatory measures CN curve number CPE control point elevation DAD depth-area-duration DEM digital elevation model EAP emergency action plan EPRI Electric Power Research Institute FERC Federal Energy Regulatory Commission GIS Geographic Information Systems GMRS Ground Motion Response Spectrum HEC-RAS Hydrologic Engineering Center's - River Analysis System HHA Hierarchical Hazard Assessment HMR Hydrometeorological Report IDF Inflow Design Flood ISFSI Independent Spent Fuel Storage Installation IWCS InfoWorks CS LCl Lettis Consultants International LIP local intense precipitation msl mean sea level NED National Elevation Data Set NGVD National Geodetic Vertical Datum NLSWE non-linear shallow water equations NRC Nuclear Regulatory Commission NTTF Near-Term Task Force PGA peak ground acceleration PMF probable maximum flood PMP probable maximum precipitation PMS probable maximum storm PSHA probabilistic seismic hazard analysis SCS Soil Conservation Service SE Safety Evaluation SPID Screening Prioritization and Implementation Details SRM Staff Requirements Memorandum SSC systems, structures, and components, SSF Standby Shutdown Facility vii

Sa*..it ~.: l..r-::: .LL IJ f- ... pl:.ti : 410 £l'l~

FF 20")tnl,)4 Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT List of Acronyms Sri Supporting Technical Information SWMM storm water management model TIN triangulated irregular network UFSAR Updated Final Safety Analysis Report UHR~S uniform hazard response spectra USACE United States Army Corps of Engineers USGS United States Geological Survey viii

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Executive Summary Executive Summary Following the accident at the Fukushima Dai-ichi nuclear power plant resulting from the 2011 Great Tohoku Earthquake and Tsunami, the Nuclear"Regulatory Commission (NRC) established the Near-Term Task Force (NTTF) and tasked it with conducting a systematic and methodical .review of NRC. processes and regulations to determine whether improvements are necessary.

The resulting NTTF report concludes that continued United States (U.S.) nuclear plant operation does not pose an imminent risk to public health and safety and provides a set of recommendations to the NRC. The. NRC directed its staff to determine which recommendations should be implemented without unnecessary delay (Staff Requirements Memorandum [SRM] on SECY-1 1-0093).

The NRC issued its request for information pursuant to Title 10 of the Code of Federal Regulations, Section 50.54(f) (10 CFR 50.54[f]) on March. 12. 2012, based on the following NTTF flood-related recommendations:

• Recommendation 2.1: Flooding

• Recommendation 2.3: Flooding to the NRC 50.54(f) letter addresses Recommendation 2.1 and requests a written response from licensees to:

1. Gather information with respect to NTTF Recommendation 2.1, as amended by SRM on SECY-1 1-01 24 and SECY-1 1-0137, and the Consolidated Appropriations Act for 2012, Section 402, to reevaluate seismic and flooding hazards at operating reactor sites.

2. Collect information to facilitate NRC's determination of the need to update the safety-related design basis and systems, structures, and components (SSCs) that are important to protect the updated hazards at operating reactor sites.

3. To collect information to address Generic Issue 204 regarding the flooding of nuclear power plant sites following upstream darn failures.

A 'Flood Hazard Reevaluation Report" was prepared for the Oconee Nuclear Station (ONS),

Units 1, 2, and 3 in response to NTTF Recommendation 2.1 and submitted March 12, 20i3.

By E-mail dated September 15, 2014, the NRC transmitted a Request for Additional Information (RAI) regarding Oconee's flood analysis of a postulated upstream dam failure. The September 2014 RAI requests that the dam failure be re-analyzed by applying alternate breach methodologies.

This report is the revised "FloodHazard Reevaluation Report"for Oconee Nuclear Station.

There are two major changes encompassed by the revision. The first is that it incorporates the results of a revised analysis which utilizes alternate breach methodologies, as requested by the September 15, 2014 RAt. Secondly, it incorporated a recently completed seismic analysis, ES-I

• .*.,L .:,oL- ,,,,.*u, w,.,,i~

• purntu per mu .. rr* L.*U~ua)t l)

Duke Energy Carolinas, LLC IONS FLOODING HAZARD REEVALUATION REPORT Executive Summary specific to the Jocassee Dam. By incorporating this seismic analysis, the revised FHRR now contains the information being sought by PAL 1, from the March 20, 2014, RAI.

When preparing Duke Energy's October 15, 2014, response, RAls 11 through 15 were identified as possibly being affected by the revision. The answers to RAls 11 and 12 are now contained in the appropriate sections of the report body. RAl 13 requested the dam breach model's electronic input and output files. This information is provided on electronic media discs as an attachment to the submittal letter for this report. The new responses to RAls 14 and 15 are included in Section 5 of the revised FHRR.

The revised report addresses the request of the September 15, 2014 RAl and satisfies the "Flood Hazard Reevaluation Report~' response of NTTF Recommendation 2.1 for the Oconee Nuclear Station (ONS) Generating Plant Units 1, 2, and 3 (ONS).

ES-2

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the Flood Hazard Section 1 - Site Information Related to the Flood Hazard 1.1 Detailed Site Information Oconee Nuclear Station (ONS) is located in eastern Oconee County, South Carolina, approximately 8 miles northeast of Seneca, South Carolina. Duke Energy Carolinas, LLC's (Duke Energy) Lake Keowee occupies the area immediately north and west of the site. The United States Army Corps Of Engineers' (USACE) Hartwell Reservoir is south (downstream) of the site. Duke Energy's Lake Jocassee lies approximately 11 miles to the north.

The location and description of ONS presented in the Updated Final Safety Analysis Report (UFSAR) Chapter 1 and Chapter 2 include reference to figures showing the general arrangement, layout, and relevant elevations of the station. The original design of the ONS Yard (Yard) grade was nominally 796.ft mean sea level (msl) (all elevations referenced in this report are based on National Geodetic Vertical Datum [NGVD] 1929). The mezzanine floor elevation in the Turbine, Auxiliary, and Service Buildings is 796.5 ft msl. ExteriOr accesses to these buildings are at an elevation of 796.5 ft msl. Section 4.0 provides further discussion of modifications to the Yard grade.

All of the man-made dikes and dams forming the Keowee Reservoir rise to a minimum elevation of 815 ft msl including the ONS Intake Canal Dike.

The ONS site, as per the original licensing basis, is a "flood-dry site," which is not subject to flooding from the nearby streams and Lake Keowee River (excluding postulated dam break scenarios). The full pond water elevation of Lake Keowee is 800 ft msl.

1.2 Current Licensing Basis Flood Elevations 1.2.1 Local Intense Precipitation The flood water elevation due to a maximum local intense precipitation (LIP) (also known as the probable maximum precipitation [PMP]) was reevaluated by HDR Engineering, Inc. of the Carolinas (HDR) in the last quarter of 2012 using the current licensing basis case PMP and state-of-the-practice engineering software. The analysis evaluated the maximum water surface elevation within the ONS power block area resulting from the occurrence of'the PMP and

• assuming three different site drainage scenarios with* an updated site topogr'aphy and building layout. Figures and tables referenced in this section are presented in Appendices A andB.

Rainfall precipitation for the ONS site was based on the UFSAR 2.4.2.2 stated depth and duration of 26.6 inches in 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />. As prescribed by the UFSAR 2.4.2.2, the temporal distribution applied to the rainfall was derived from the normalization Of mass rainfall curves located at Clemson College from the historical precipitation event of mid-August 1940. Figures A-I and A-2 illustrate the cumulative and incremental precipitation, as well as rainfall mass curve used in the model analysis. The maximum 1-hour rainfall intensity for the normalized

" ""~.,........... r....i.,..l..L . ...a LLC Duke Energy Carolinas, ...;~

IONS ,. . uui.,

....LFLOODING v'.r' £.U REPORT HAZARD REEVALUATION "flI Section*1 - Site information Related to the Flood Hazard rainfall precipitation event was approximately 3.5 inches per hour. Figure A-3 is an output from the rainfall/runoff simulation model showing the rainfall distribution as input in the analysis.

The coupled one-dimensional (l-D) and two-dimensional (2-0) hydraulic model was used for the basis of this study. This model encompasses the main Yard and extends to include the 230 kilovolt (kV) switchyard east of the main Yard and the Independent Spent Fuel Storage Installation (ISFSI) south of the main Yard (Figures A-4 and A-5-A). A separate model was created for the Keowee Hydro Powerhouse sub-basin (Figure A-5-B).

The topography terrain data for the Yard and ISFSI was supplied by Duke Energy and was used as the basis for generating the modeling surface for the main Yard. The 3-inch topography data was extended using the 1-foot topography contours for areas outside the coverage of the 3-inch topography. The areas using the 1-foot topography contours are the 230-kV switchyard, extension to the ISFSI area, and the Keowee Hydro Powerhouse sub-basin. The extension to the ISFSI area is mainly between the main Yard and !SFSI where there is no 3-inch topography coverage, extending towards the inlet channel and extending to the drainage basin boundary.

The supplied data was processed in ESRI ArcGIS software to create a triangulated irregular network (TIN) file, which waS imported into the. hydraulic mOdeling software, Inf0Works CS, Version 11.5 (IWCS).

The modeling of the Yard reflects the up-to-date configuration of the Yard, based on 2010 surveyed data. There were some locations in the. terrain model that had to be modified due to temporary construction (excavation) that was taking place at the time the site was surveyed.

These areas were identified and the mesh for the 2-D simulation was modified to either raise the area locally or lower the area locally. The approach used was to raise an area by specifying a minimum elevation to use for that zone. If a terrain value was lower than this defined minimum elevation, this minimum elevation was substituted for the actual elevation for determining the 2-D element elevation. Similarly for lowering a zone, a maximum elevation was chosen and any value in the terrain data would be capped by this value for determining the 2-D mesh element elevation. The final mesh elevations used for the simulation are illustrated in Figures A-6-A and A-6-B.

Soil and terrain surface characteristics for the Yard and off-site sub-basins were used to develop the hydrology characteristics Using the Soil Conservation Service (SCS) curve number (ON) method. Figure A-i-A shows the various off-site boundary areas by color designation.

The flow coming off these sub-basins was further subdivided to be applied where it would enter the Yard as shown in Figure A-i-B. In addition, model node l~s are shown in Figure A-7-B.

Flow hydrographs were calculated for the downstream loCation of these sub-basins and typically applied directly to the 2-0 mesh at locations where it had been determined by contOurs that the flow would naturally concentrate. For off-site Basins 1, 2, and 5 shown in Figure A-i-A that are on the western side of the site, the calculated flows were routed along overland flow paths, through culverts, or over weirs that represent roadways to adjacent overland flow paths that connect to the 2-D mesh of the Yard. The flow coming off the sub-basins was further subdivided to be applied where it would enter the Yard.

2

I~,,. *.. * .. ,* ...... .... ..... ........ --t ... . .... iu 2.-. (~l Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the Flood Hazard The Keowee Hydro Powerhouse area has no additional modeled routed drainage. Rainfall is directly applied to the 2-D mesh. This assumption was used because the subject drainage area is fairly small (at less than 10 acres) and with the pervious areas being grassY and steep, initial and continual losses would be minimal and thus the approach taken would yield a more conservative result.

Physiographic characteristics were derived from the following resources:

• Soil Characteristics (UFSAR 2.5.4.2; 2.5.6.3) (Contractor determined) (Natural Resources Conservation Service)

• Surface Characteristics (Contractor determined by aerial photographs provided by Duke

.Energy)

The Yard drainage was created from drawings and walkdown information of the current site layout. This data was used to build the existing conditions model (Figure A-7-A) where color shading represents differ'ent sub-basins in the runoff model. The layout of the site, including the location of building and drainage information, was taken from the Duke Energy-supplied MicroStation drawing files. These files were also used to input pipe size and elevation of the elements within the system along with the associated supplied Spreadsheet of the catch basin data. Duke Energy catalogued the site drainage catch basins. This information is represented in the model as an equivalent-sized orifice for each catch basin.

The roof drainage (Figure A-8) for the Reactor, Auxiliary, and Turbine buildings was created from drawing and photograph files. Each roof section was modeled as a conceptual volume defined by an elevation-area relationship. An orifice was used to drain the roof area to the respective segments of the roof drainage system. Overflow of the roof gutter or parapet was accounted for by use of a conceptual weir of appropriate elevation and length that discharges to a location based on determined flow paths. These locations are typically directly applied to the 2-D mesh, at downspout locations, where appropriate, or other roof sections depending on roof geometry. Details for parts of the Auxiliary Building and the other buildings on site were reviewed by the modeler; and where extensive detail was not deemed necessary (based on experience), a more simplified roof catchment and orifice drainage system was defined to simulate the roof drainage that connected directly to the Yard drainage system.

The roofs that drain to the Yard directly (no downspouts) and those with downspouts were modeled by defining a catchment area for each roof area and connecting this to the 2-D Yard Surface (Figure A-9). An orifice was. used to represent each downspout to discharge to the Yard at the appropriate location. A weir was also used at each downspout location to allow overtopping of the gutter to add flow to the same location the downspouts discharge to. Roofs with a direct response to the Yard and overflow from gutters were represented by the use of a weir to direct the flow to the 2-D surface.

The modeling software used for this effort was IWCS with 2-0 - a Coupled 1-D and 2-D simulation model produced by Innovyze (formerly Wallingford Software). This modeling package allows the hydrology and hydraulics for 1 -D pipe flow and 2-D overland flow to be modeled-within one software environment. The hydrology was divided into three areas: off-site 3

Duke Energy Carolinas, LLC IONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the Flood Hazard drainage, which is modeled as SCS basins and derived flows loaded onto the Yard 2-D mesh at appropriate locations; the Yard, which is the 2-D surface directly around the Reactor and Turbine buildings where the rainfall is applied directly to the mesh; and roof drainage. The roof surfaces use the storm water management model (SWMM) routing and direct flow to the roof drains, Yard drains, or directly to the Yard as appropriate. Two models were developed: one for the main Yard to the 230-ky switchyard and ISFSI area, and one for the Keowee Hydro Powerhouse sub-basin.

Figure A-i0-A shows catch basins and other nodes located within the Yard study area. The link between the 1-D and 2-D simulations can be an important factor in the model results along with the simulation time step used. The 1-D and 2-D simulation engines are separate with a link between them which, in this IWCS model, represents the catch basin nodes. The simulation engines Calculate depth and flow in the pipes and across the surface at every point simultaneously for each "major" time step. If a result cannot be determined at a particular time step, then the time step is halved to a "minor" time step and the simulation engines try to calculate again until a result is achieved. The interaction between the 1-D and 2-D Simulations takes place at the major time steps at the catch basins. Thus, it is important to use an appropriate time step for the simulation to ensure proper exchange of flow between the 1-D pipes and 2-D surface. A 1-second time step has been used for these simulations to prevent excess flow from building up on the suJrface over the catch basins between major time steps, which would artificially increase the water surface elevations on the 2-0 surface.

The 2-D mesh is comprised of an irregular array of triangles known as computational mesh.

These triangles have a defined maximum size. The mesh also has a minimum element size defined for simulation. If mesh triangles have an area less than the minimum mesh element size, then triangles are merged together to create a 2-D element greater than the minimum mesh element size. This is to maintain simulation stability. The main Yard has a maximum triangle size of 250 square ft and a minimum mesh element size of 30 squareft.

The average mesh element size in highly detailed areas around buildings is approximately 80 square ft with an average size of about 150 Square ft across the open part of the Yard. This mesh is tighter around features of hydraulic significance such as buildings, swales, and catch basins; and the mesh is looser in areas that are open and reasonably flat. The Keowee Hydro Powerhouse sub-basin mesh has a maximum triangle size of 90 square ft and a minimum mesh element size of 30 square ft.. The average mesh element size is about 60 square ft.

Three scenarios were investigated and are summarized below.

All three scenarios use the 26.6 inches in 48-hours rainfall event with temporal distribution modeled after the August 11-14, 1940, storm event for the area.

• Scenario 1: Complete system with fully functioning roof and Yard drainage.

• Scenario 2: Yard and roof drainage considering the Yard drainage catch basins to be blocked (not functioning).

• Scenario 3: Yard drainage (surface) only considering both the roof inlets and Yard drainage (sub-surface) catch basins to be blocked (not functioning).

4

"c:C h :. ,C ~ ..... v..... ,,.,i wv~pIJIU *ab puu I si, p*,tel iu !.rtK -'.,su~a)'I, Duke Energy Carolinas. LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the Flood Hazard The simulations were run for a period of 50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br /> to allow enough time for all of the water from the rainfall event to flow off or out of the system. Site inundation elevations under current licensing basis PMP and existing site layout are provided in Appendix B.

Table B-I quantifies the increase in flood depth around the Reactor, Auxiliary, Turbine, and Administrative buildings due to modeling the Yard by simulating the Yard drainage catch basins being blocked. The average increase in water depth due to the catch basins being blocked is 0.67 ft. The average increase in flood depth east of the Turbine Building is 0.77 ft. The average increase west of the Auxiliary and Reactor buildings is 0.83 ft.

The Yard flood depths are the same for Scenario 2 and Scenario 3 where only the Yard drains are modeled as blocked and where the Yard drains and roof drains are both modeled as blocked.

Figures B-2 and B-4 show the location of maximum water surface elevations and durations for points in the Yard around the Standby Shutdown Facility (SSF), Protected Service Water (PSW), and Essential Siphon Vacuum (ESV) structures and in the CT-5 area when Yard drainage is effective.

The maximum depth of water for the current licensing basis case around the Keowee Hydro Powerhouse sub-basin is shown in Figures B-3 and B-5. The 2-D terrain in this area is showing local low areas in paving next the powerhouse but the base map contours around the powerhouse are 1 foot while the modeled water depths are around 0.5 ft.

The efficiency of the Yard drainage and roof drainage systems was evaluated using the IWCS model but for this review, the drainage systems were assumed to be blocked, producing a more conservative LiP inundation condition at the site.

1.2.2 Flooding in Reservoirs Since ONS is located near the ridgeline between the Keowee and Little River valleys, or more than 100 ft above the maximum known flood in either valley, the records of past floods are not directly applicable to siting considerations.

Original ProJect studies were conducted to evaluate effects on reservoirs and spillways of maximum hypothetical precipitation (PMP) occurrin~g over the entire Lake Keowee drainage area. This rainfall was estimated to be 26.6 inches within a 48-hour period. Unit hydrographs were prepared based on a distribution in time of the storms of October 4-6, 1964, for Jocassee and August 13-15, 1940, for Keowee. Reservoir flooding results are summarized as follows:

Table I. Reservoir Flooding Results Keowee Jocassee Maximum spillway discharge 147,800 cfs 70,500 cfs Maximum reservoir elevation 808.0 ft 1,114.6 ft Freeboard below top of dam 7.0 ft 10.4 ft Note: Data above from UFSAR.

S

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the Flood Hazard While spiliway capacities at Keowee and Jocassee have been designed to pass the inflow design flood (IDF) with no surcharge on full pond, the damns and other hydraulic structures have been designed with adequate freeboard and structural safety factors to safely accommodate the effects of PMP. Due to the time-lag characteristics of the runoff hydrograph after a storm, it is extremely unlikely the maximum reservoir elevation due to the PMP would occur simultaneously with winds causing maximum wave heights and run-ups.

Considering this assessment, the ONS site can be characterized as a "flood-dry site," as described in Section 5.1.3 of the American National Standard Report, "Determining Design Basis Flooding at Power Reactor Sites," because the safety-related structures of the existing ONS are above spillway discharge flooding elevations. The Yard is nominally 796 ft msl and during the passing of a probable maximum flood (PMF) through the Keowee spillway, discharged water (approximately 145,000 cubic feet per second [cfs]) is not expected to backup significantly over the river elevation of approximately 686 ft msl. This meets the intent of the definition of a "flood-dry site."

1.2.3 Dam Failures There were no dam failures postulated in the original licensing or design basis of the plant. A dam failure event was postulated more recently and is described below in Section 1.3.

1.2.4 Storm Surge and Seiche Storm surge and seiche events were never postulated to affect the site, and no flood elevation is given in the original licensing basis of the plant. The maximum wave height and wave run-up were calculated for Lake Keowee and Lake Jocassee by the Sverdrup-Munk formulae. The results of these calculations are as follows:

Table 2. Calculation Results Lake - -Wave Height' Wave Run-Up -Maximum Fetch, Keowee' (Keowee.River Arm) 3.70 ft 7.85 ft 8 miles Jocassee 3.02 ft 6.42 ft 4 miles Keowee (Little River Ar~m)' 3.02 ft 6.42 ft 4 miles Note: Data above from UFSAR The wave height and wave run-up figures are vertical measurements above normal full pond elevations as tabulated above. The design freeboard at each dam is 15 ft which is adequate to prevent overtopping of wind-driven waves.

1.2.5 Tsunami Tsunamis were never postulated to affect the site, and no flood elevation is given in the original licensing or design basis of the plant.

1.2.6 Ice-Induced Flooding Ice-induced flooding were never postulated to affect the site, and no flood elevation is given in the original licensing or design basis of the plant.

6

~ri J ......

C .nbr fm..... ......... . * ....... - I ......

.:t ~vIrT-- ir .... .... o( J(

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the Flood! Hazard 1.2.7 Channel [Diversion Channel diversions were never postulated to, affect the site and no flood elevation -is given in the original licensing or design basis of the plant.

1.2.8 Combined Effects Combined flooding effects (PMP, PMF, dam failure and/or wind-driven waves) were never

• postulated to affect the site and no flood elevation is given in the original licensing ordesign basis of the plant.

1.3 Licensing Basis Flood-Related and Flood Protection Changes ONS has made the following flood-related and flood protection changes described below.

Thereis one exception to the Keowee ReServoir Retention dikes and dams minimum elevation of 815 ft msl. This .exception is described as two reinforced concrete trenches extending through the ONS Intake Canal Dike with a minimUm elevation Of 810 ft msl. This exception only occurs during very rare occasions when the covers would be removed for maintenance. These trenches are protected from wave action by the Condenser Circulating water (CCW) Intake Structure and the Causeway at the west end of the CCW Intake Structure. Therefore only the maximum reservoir elevation of 808 ft msl is applicable with regard to flooding through the reinforced concrete trenches.

An additional licensing caveat exists in ONS's Updated Final Safety Analysis Report (UFSAR).

UFSAR Section 9.6.3.1 describes the design basis loads for the SSF. The original design basis.

for the SSF stated, "Flood studies show that Lake Keowee and Lake Jocassee are designed with adequate margins to contain and cOntrol floods... The Final Safety Analysis Report (FSAR) addresses Oconee's location as on a ridgeline 100' above maximum known floods. Therefore, external flooding due to rainfall affecting rivers and reservoirs is not a problem. The SSF is

,within the site boundary and, therefore, is not subject to flooding from lake waters" (UFSAR Section 9.6.3.1). In 1983, a JocaSsee Dam failure studY was completed to determine the maximum water surface elevation around the SSF if a postulated fair-weather failure of the Jocassee Dam is considered. The results of the study estimated a peak flood elevation of 817.45 ft msl at the Keowee dam, and a Yard flood level of 4.71 ft. In resPonse to this study, ONS erected several walls approximately 5 ft tall, around the entrances to the SSF as a risk reduction measure. This was not part of the design basis. ,Based on O NS's September 26, 2008. response to a Title 10 of the Code of Federal Regulations, Section 50*54(f) (10 CFR 50.54[f]) request for information, ONS increased the height of the flood wall around theS5SF by 2.5 ft as a risk reduction measure with a resulting elevation of 803.5 ft ins!.

The UFSAR Section 9.6.3.1 was updated to say:"As a PRA enhancement the SSF is provided with a five foot external flood wall which is equipped with a water tight door near the south entrance of the SSF. A stairway Over the wall provides accessto 'the north entrance. The yard

.elevation at both the north and the south entrances to the SSF is 796.0 ft msl. Based on the as-7

Duke Energy Carolinas, .LLC j ONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the Flood Hazard built configuration of the 5' flood wall provided at the north entrance and the flood wall at the south entrance to the SSF, SSF external flood protection is provided for flooding *that does not exceed 801 ft above mean sea level." In its current as-built condition, the 8SF has flood protection features UP to an elevation 'of 803.5 ft msl.

A separate external flood hazard is the postulated fair-weather failure of the Jocassee Dam."

This is derived from a NRC Safety Evaluation (SE) dated January. 28, 2011, addressing commitments made in regards to the June 22, 2010, Confirmatory Action Letter (CAL) to address external flooding concerns.

The following description is a synopsis of the SE.

In April 2006, the U.S. NRC staff questioned the flood protection barrier for the SSF. The NRC' identified that the licensee had incorrectly calculated the Jocassee Dam Failure frequency and had not adequately addressed the potential consequences of flood heights predicted at ONS.

Based on concerns raised by the NRC, by letter dated August 15, 2008, (ML081640244) the NRC requested information in a 10 CFR 50.54(f) letter. Duke Energy responded to the licensee's request On September 26, 2008 (ML082750106).

After further correspondence, the NRC staff issued a CAL to the licensee on June*22, 2010, requesting the following: 1) submit to the NRC all documentation necessary to demonstrate that the inundation of the ONS site, from the postulated fair-weather failure of Jocassee Dam, has been bounded; 2) by November 30, 20"10, submit a list of all modifications necessary to mitigate the inundation* and 3) make all necessary modifications by November 30, 2011. Subsequent correspondence with the NRC has deferred these dates.

The staff also requested that the compensatory measures (CMs) listed in the CAL remain in place until they can be superseded by regulatory, action related to the Fukushimaresponses.

The safety evaluation (SE) describes the methods and parameters chosen for the analysis. It concludes by.stating: "The unmitigated Case 2 dam breach parameters that were used in the flooding models, provided by* Duke 'Energy for ONS site, demonstrated that the licensee has included conservatisms of the parameters utilized in the dam breach scenario. These conservatisms provide the staff with additional assurance that the above Case 2 scenario-will bound the inundation at ONS, therefore providing reasonable assurance for the overall flooding scenario at the site. This new flooding scenario is based on a random fair-weather failure of the Jocassee Dam. This Case 2 scenario will be the new flooding, basis for the site"' (SE dated January 28, 2011 ). The SE also states that the licensee has committed to keep the CMs in place Until final resolution has been agreed upon between the licensee and the NRC staff.

1.4 Watershed -and Local Area Changes Changes to the local site topography and support~buildings have taken place since original construction. Changes* in local area conditions have been captured in the modeling performed to support the flooding assessment due to LIP and dam failure 'inundation using recent aerial 8

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 1- Site Information Related to the Flood Hazard and ground survey data (2010) along with updated drainage, utility trench location and building geometry.

There has been Construction of housing and support facilities directly around Lake Keowee in the watershed since 1971, but the overall percentage of land use has not significantly, changed since the construction of the reservoirs and ONS. There is no significant change in land use around Lake Jocassee. Most of the Lake Jocassee watershed is comprised of protected forest lands.

1.5 Current Licensing Basis Flood Protection and Mitigation Features Based on the ONS flood hazards, a list of flood protection, mitigation, and earlywarning indicator features has been compiled. This list of features was created to fulfill the NRC-issued information request on March 12, 2012, in accordance with 10 CFR 50.54(f). Enclosure 4. of the 50.54(f) letter was directed toward addressing the NTTF Recommendation 2.3 for Flooding and requested the results of a flooding design basis walkdown. Below is a high level description of the different features presented in the 2.3 Flooding Walkdown report. Many of the features are actually groups of smaller features and for convenience have been grouped together.

The first group consists of Features I1through 9. These features are included because of the PMP initiating flood as described above. They are all either at Yard elevation (-796 ft .msl) or below grade and are listed below:

Flood Protectionand Mitiqation Features:

1. Auxiliary Building exterior subsurface walls and seals

2. Radwaste trench covers and seals from Radwaste Facility to Turbine building

3. Radwaste trench covers and Seals from Radwaste Facility to Auxiliary building

4. Interim Radwaste trench covers and seals

5. Manhole 7 Cover, Technical Support building vault and seals

6. Borated Water Storage Tank (BWST) trench covers

7. Yard drainage system

8. CT-5 trench covers in Auxiliary building

9. PMP rainfall event flood barrier sandbags and Gryffolyn coverings A second group consists of Features 10 through 16. These features are included because they are part of the Keowee Reservoir Retention items. Listed below are the seven features:

Flood Protectionand Mitigqation Features:

10. Keowee River Dam

11. ONS Intake Canal Dike

12. Little River Dam

13. Little River Dikes A, B, C, D

14. Keowee Intake

15. Keowee Hydro Powerhouse

16. Keowee Spillway 9

Duke Energy Carolinas, LLC JONS FLOODING HAZARD REEVALUATION REPORT Section 1 - Site Information Related to the FlOOd Hazard The third group consists of Features 17 through 31. Their purpose is to mitigate flood conditions based on the January 28, 2011, SE postulated fair-weather failure of the Jocassee Dam. This list was derived from the fifteen CAL CMs, and the actions taken by ONS in order to satisfy the CAL. These features include plant flood protection procedures, early flood detection warning equipment housed at Jocassee Hydro, and emergency equipment housed at ONS and operated under emergency procedures. The list of features is provided below:

Flood Protection. MitigationFeatures, and Measures:

17. EM 5.3 Procedure (External Flood Procedure)

18. APiO/AN1700/047 Procedure (External Flood Abnormal Procedure)

19. Jocassee Flood mitigating plans and procedures

20. Duke Energy Hydro generation guidance document

21. Dam safety inspection program

22. Maintained monitoring program

23. Keowee Spillway enhancements

24. Jocassee Forebay and tailrace alarms

25. Jocassee storage building with backup spillway operating equipment

26. Portable generator and electric drive motor near the spillway

27. Documentation of table top exercises

28. Instrumentation and alarm selected 'seepage mnonitoring locations

29. Video monitoring of Jocassee Dam

30. Second set of B.5.b-like equipment

31. Jocassee Dam-ONS response drill documentation The fourth group consists of Features 32 through 34. These features are included because they are the exterior flood protectiOn features for the SSF. Listed below are the three features:

Flood Protectionand Mitiqation Features:"

32. SSF flood barriers including exterior walls

33. SSF steel plate with CO2 refill access

34. SSF external wall penetrations

35. Feature #35 is the last feature on the Flood Walkdown list and corresponds to the Site Elevation and ToPography.

More info'rmation on these features and walkdown conclusions can be found in the Flood Walkdown Report (NRC 50.54 (f) NTTF Recommendation 2.3).

10

,.onan oeuuyOfb.uUuu uiu.,

Duke Energy Carolinas,

,;,, .. =,, ~Ll. . i r .z~)1 LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 -- Flooding Hazard Reevaluation Section 2 - Flooding Hazard Reevaluation The reevaluations in Section 2.0 are not part of the ONS licensing basis as documented in the UFSAR. The following sections describe additional reevaluation analysis for assessing appropriate external potential flooding hazard events including the effects from LIP on the site, PMF on reservoirs and dam failure that havebeen performed post-design basis to meet the Hierarchical Hazard Assessment (HHA) procedure described in NUREG/CR-7046 for assessment of flooding hazard at safety-related SSCs.

ONS is located on the Keowee reservoir which is a Federal Energy Regulatory Commission (FERC) regulated hydroelectric development that is impounded by two embankments adjacent to ONS and five additional embankments along the Little River Arm of the reservoir away from the site. As FERC-regulated dams, the structures have been designed and verified with

• hydrologic and hydraulic analysis and dam stability analyses in accordance with "FERC Engineering Guidelines for the Evaluation of Hydropower Projects." These structures are monitored and inspected on regular intervals by Duke Energy, FERCI, and independent dam safety experts.

2.1 Local Intense Precipitation The flood water elevation due to a maximum LIP was reevaluated using current practice Hydrometeorological Report (HMR) PMP and state-of-the-practice engineering software. The analysis evaluated the maximum water surface elevation within the ONS power block area resulting from the occurrence of the PMP and assuming three different site drainage scenarios with an updated site topography and building layout. This reevaruation differs from the licensing basis analysis in that it utilizes HMR 5.1 PMP and different rainfall distribution` patterns in accordance with guidance in NUREG/CR-7046. The modeling software used in this evaluation exceeds procedures outlined in NUREG/CR-7046, !-D channelized flow using the USACE's Hydrologic Engineering Center's-River An~alysis System (HEC-RAS), as it analyzes overland flow on relatively flat surfaces using 2-D flow equations. Due to the complex site geometry including building layout, roof drainage, subsurface drainage, and lack of defined 1-D flow channel in the currently Configured Yard, a model using 2-D flow equations was employed as described in the following sections. Figures and tables referenced in this section are presented in Appendices A and C.

This section presents a summary of the model developed to reView local flooding at the ONS yard due to an assumed rainfall event based on HMR 51 for a PMP rainfall event.

This work consists of the ONS site flood inundation evaluation and is based on an expanded IWCS 2-D model based on updated site data from Duke Energy regarding the Yard drainage and building and roof drainage transmitted over the period of September 2011 to September 2012. The roof drainage and Yard drainage performance was evaluated with the subsequent flow routing (sub-surface and overland flow) through the ONS site utilizing 1-D modeling for roof and sub-surface Yard drainage and 2-D modeling for Surface drainage (overland). This analysis 11

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation includes the maximum depths and durations for the Yard and roof inundation as well as the flow patterns around the Yard. This section provides a summary of the IWOS model construction.

The Beyond design basis case utilizes the HMR 51/52 methodology for an area less than 10 square miles. The PMP event that produces the study case site flood event was determined by trial procedures using the model to route the runoff. The Beyond design basis case analysis was performed for three scenarios:

• Scenario 1 is the complete system with fully functioning roof and Yard drainage.

• Scenario 2 is the Yard and roof drainage considering the Yard drainage catch basins to be blocked (not functioning).

• Scenario 3 is the Yard drainage (surface) only considering both the roof inlets and Yard drainage (sub-surface) catch basins to be blocked (not functioning).

Rainfall precipitation for the Yard was determined using HMR 5 1/52 as they apply to the site located in the U.S. east of the 105th Meridian. HMR 51 defines the depth in inches of PMP for durations of 6, 12, 24, 48, and 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> for watersheds from 10 mi2 to 20,000 mi2. HMR 52 defines the depth in inches of PMP for durations of 5, 15, 30, and 60 minutes for watersheds from 1 mi2 to 200 mi2 . Both HMRs provide isohyetal charts to determine the PMP values based on location. The location used for the site is approximated by Latitude 34.80 N and Longitude 82.90 W. It should be noted that the PMP charts are not sensitive to a small deviation in Latitude and Longitude. Table 3 contains the PMP depths for the site using HMR 51 and 52.

Table 3. Probable Maximum Precipitation Depths for the Site Using HMR 51 and 52 Depth-Duration All-Seasonal PMP values, by Duration (hrs) 1-mi 2 Point Rainfall 10-mi 2

_____5-min I5-min 30-min 1-hr 6-hr 12-hr" 24-hr '48-hr. 72-hr PMP (Ice) 6.2 9.7 14.0 18.95 30 35.8 40.2 44.3 46.7 As directed by procedures outlined in HMR 51 and 52, intermediate points are determined from applying a smooth curve through the derived points. Intermediate points were determined in this manner for a duration interval of 5 minutes up to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />, and a duration interval of 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> beyond 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> up to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. The complete PMP depth-duration curve is provided below in Figure 1.

Taking the first portion of the curve, up to 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, the 5-minute interval duration depths are provided below in Table 4. Bold values correspond to the values derived directly from HMR 51 and 52 PMP charts.

The process outlined in this report is based on NRC regulations for the next generation nuclear stations to be constructed in the U.S. (NRC Standard Review Plan, NUREG-0800, 2.4.2). The PMP rainfall applied to the ONS watershed for this analysis was an all-season, 6-hour event and a 72-hour event.

12

£ ~ .::........Ll .i r. ..........

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation Figure 1. Probable Maximum Precipitation Depth Duration Curve 50 . .. .

450 -- --

• "I I I ,

05 6 . .. . 18....3.36.4.... ... 6 .. ... 6 .. 72, Durtin hors Table 4. 5-Minute Interval Duration Dep*ths Duration Cumulative PMP (minutes), Depth (inches) 5 6.18 10 8.02 15 9.72 20 11.29 25 12.72 30 14.02 35 15.14 40 16.05 45 16.84 50 17.54 55 18.23 60 18.95 13

t. ...... ... . . .... , . ..... ... . . ...... ..... ! . .. .. .. .. .. r . .... r i --- in ( (1 .

OueEnergy Carolinas, LLC I ONS FLOOOING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation As described in HMR 52, PMP estimates were obtained from HMR 51 for 6-, 12-, 24-, 48-. and 72-hour storm durations (HMR 51 Figures 18 through 22) to estimate the largest volume PMP event shown in Table 3. Additionally, a 6-hour local intense event was used for analysis based on discussion found in NUREG7046, Section 3. The 6-hour PMP LIP was developed for 5-minute increments following prceures outlined in Setin 6 of HMR 52. After a review of the 6-hour and 72-hour precipitation modeled results, it was determined that a longer duration event could have a significant impact on site flooding. The maximum 5-minute rainfall intensity for the 6-hour event was 72 inches per hour while the maximum 5-minute rainfall intensity for the 72-hour event was 19 inches per hour. Therefore, the 72-hour precipitation was modified to include the peak intensity indicated by using the 6-hour event based on the peak 1-hour distribution for the 6-hour event shown in Table 4. The IWCS model was re-run for each storm duration using six temporal distributions applied to the rainfall: front, one-third, center, two-thirds, and end-loaded storm as well as the distribution recommended in HMR 52. Figures 2 and 3 compare the six temporal distributions.

The 72-hour PMP event was identified as the storm duration which produced the most conservative flo Yard elevations through simulations using the IWCS model. This event was determined to be the HMR 51, modified 72-hour event, with 46.6 inches of total rainfall, and two-thirds loaded distribution.

The coupled 1-D and 2-1) hydraulic model was developed to be used for the basis of the study.

The model encompasses the currently configured main Yard and extends to include the 230-kV switchyard east of the main Yard and the ISFSI south of the main Yard. A separate model was created for the Keowee Hydro Powerhouse sub-basin (Figures A-4, A-S-A and A-5-B). Refer to Section 1.2 for an additional description of the model configuration and inputs in addition to the reevaluation of the PMP defined in this section.

Figure 2. Incemntal precipitation during them 6-hour PMP for six different temporal distributions 7.(00 I .00 3.00 3.00

• 1 II Inrementi Precilpitation Crente

• ncremental UIcrermental Preciptataion Ton-Thirds Precipetateon CeRter U Incremental IPrecipitation End 0aq 0 Minute 14

Duke Energy Carolinas, LU.C I ONS FLOOOING* HAZARD REEVALUATION REPORT Sectbon 2 - Flooding Hazard Reevaluaon Figure 3. Incremental precipitation during the 72-hour PMP for six different temporal distribtion 2000 14.00 uncremetld Precuptato Fron 12.00 1aO0 U Incremental Preciuaron One-Thid 8.00 U Incemnt~al Preapitaton C:ente 3 .0IJ sincremental Precr~ation Two-Thrd

- 2.00 I Incremental Predipitation End U

0.00 1 5 9 131721252933374145495357616569 Hour Several tests were simulated with the model using the 6-hr and 72-hr storms to select the storm that produced the greatest flooding impact at critical areas around the ONS site. Both storm durations were reviewed using the six different temporal distributions shown in Figures 2 and 3 (six 6-hour cases plus six 72-hour cases : 12 trial cases). From this sensitvt analysis, the critical temporal distribution was selected as the HMR 51152 72-hour, two-thirds loaded PMP event.

The results provided in Table C-i are based on model results using the HMR 51 /52 PMP with two-thirds loaded temporal distribution. Locations referenced in this section are based on Figure C-1.

Figures C-2 through C-4 show the Beyond design basis case maximum flood depths in the Yard with Yard drainage and roof drainage, with roof drainage and no Yard drainage, with no Yard drainage or roof drainage, and specific locations of maximum flood depth in the Yard. The slope and configuration of the Yard facilitate rapid removal of the intense rainfall of the HMR PMP event.

Table C-i quantifies the inrese in flood dept around the Administrative, Turbine, Auxiliary, and Reactor buildings due to modeling the Yard by simulating the Yard drainage catch basins being blocked. The average increase in flood depth due to the catch basins being blocked is 0.19 ff, which is a 7 percent increase. The average increase in flood depth east of the turbine building is 0. 1 if, a difference of 7 percent. The average increase in flood depth west of the Auxiliary and Reactor buildings is 0.29 ft, a difference of 8 percent.

The Yard flood depths are the same for Scenarios 2 and 3 where only the Yard drains are modeled as blocked and where the Yard drains and roof drains are both modeled as blocked.

15

Duke Energy Carolinas, LLC j ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation The maximum flood water surface elevations during the Beyond design basis case around the Keowee Hydro Powerhouse sub-basin structure are shown in Table C-1. There is effectively no rise in water surface elevation around the Keowee Hydro Powerhouse due to the catch basins being blocked. However, assuming the catch basins and the culvert under the roadway that conveys-the hillside runoff are blocked increases the depth of water adjacent to the western side of the powerhouse and produces a nominal increase on the northern side. The increase in modeled depth of water is due to the flow that runs off the hillside and into the drainage ditches conveying the flow to the culvert results in flow overtopping the culvert inlet, and the water then flows over the roadway and up against the western side of the structure.

The Yard drainage is surcharged in numerous locations when trying to handle the flows associated with the HMR PMP event and excess water flows onto the surface. From testing model response for just the roof connections to the Yard drainage, it was noted that the Yard drains overflowed onto the surface in a number of locations with just the flows coming off the roof system. Both the east and west Yard drainage lines show heavy surcharging of the system. This produces flooding at the surface in some locations; and under the strain of the intense rainfall, the drain system fills to an extent that prevents additional flow from entering from the surface.

The roof drainage is undersized for the HMR PMP event. With the exception of the Reactor Building and the lower level roof of the central Auxiliary Building, the water does not overtop the roof parapets with the roof and Yard drainage functioning. Comparing the total inflow for the roof section to the maximum volume stored on the roof, the lower section of the central Auxiliary Building, B8075-8078-8082-B1 (Figure C-12), has flow leaving the roof drainage system under pressure and up onto the roof. The other areas where flow appears to overtop the parapets are likely due to the simplified way that they are modeled or the level of detail that was available to build the model. Most of the roof areas of the Reactor, Auxiliary, and Turbine buildings have a large amount of available storage until the rainwater overflows the parapets.

The Reactor Building roof was not fully developed in the model and would require additional refinement to allow the excess rainfall that cannot get into the roof drain system to overflow to the appropriate areas. This refinement was not performed due to the small percentage of contribution of this water to the total site inundation flow and resulting ground level inundation estimates.

The inundation of rainwater on the Turbine Building averages 4.06 ft in depth over the central span of the Turbine Building. The outer edges range from approximately 1.37 to 3.87 ft in depth and the lower roof on the eastern edge is 6.49 ft in depth. The rainwater in general stays on the roof for 30 minutes and for 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> for the central span.

The range of depths of inundated rainwater on the Auxiliary Building roof is between 0.33 and 3.10 ft, correcting for parapet height. The average is less than 1 ft taking into account the excessive depth on the lower central roof. The duration of inundation is generally 30 minutes to 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> with the exception of the lower levels of the central and southern sections of the Auxiliary Building that have durations closer to 2 to 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />. This is due to the heavy surcharging of the 16

Duke Energy Carolinas, LLC [ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation system. This results in some extended pooling of water on the roofs which can be seen in Table C-2. Table C-2 lists a summary of the Beyond design basis case roof drainage.

characteristics.

The buildings with the greatest depth above the roof deck are those with the smallest surface area and corresponding small volumes with relation to the entire roofed area. The excess volume on the central span of the Turbine Building will overflow onto the east and west sections of the roof where there is available volume. Whether the roof can hold this much weight of water is beyond the scope of this study.

The modeled scenario with the Yard catch basins blocked off relieves the lower roof sections somewhat, but has no real effect on the upper roof sections, as can be seen in Table C-3.

2.2 Flooding in Reservoirs Section 5.5.1 of American Nuclear Society (ANS) 2.8, under "Hydrologic Dam Failures," states that "critical dams should be subjected analytically to the probable maximum flood from their contributing watershed. If a dam can sustain this flood, no further hydrologic analysis shall be required." The two significant reservoirs upstream of ONS are Jocassee and Keowee; both are licensed by Duke Energy and were constructed and are maintained in accordance with FERC guidelines. Dam-specific PMP and PMF documentation was reviewed, demonstrating that both dams can safely pass their PMFs.

Per ANS 2.8, Section 5.5.4, "if no overtopping is demonstrated, the evaluation may be terminated and the embankment may be declared safe from hydrologic failure." Overtopping should be investigated for either of these two conditions:

,, PMF surcharge level plus maximum (1 percent) average height resulting from sustained 2-year wind speed applied in the critical direction; or

• Normal operating level plus maximum (1 percent) wave height based on the probable maximum gradient wind.

2.2.1 Probable Maximum Flood - Keowee Flooding hazard reevaluation for a PMF on the Lake Keowee watershed was performed by reviewing the updated analysis developed for the FERC-Iicensed and regulated hydroelectric development (#2503). The peak reservoir elevation from that analysis is 808.9 ft msl based on hydrologic and hydraulic modeling performed in accordance with Chapters II and VIII of the FERO "Engineering Guidelines for the Evaluation of Hydropower Projects" using HMR 51/52 for the PMP development and HEC-1 for the hydrologic and routing model (0.9 ft higher than the UFSAR study,). The spillway and dams at the Keowee development have been designed to pass the PMF with adequate freeboard and structural stability factors to safely accommodate the effects of the PMP. Table 5 summarizes the normal and PMF freeboard for the dam structures that impound Lake Keowee.

17

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT' Section 2 - Flooding Hazard Reevaluation Table 5. Keowee Dam - Normal and probable maximum flood freeboard (ft)

~~Reservoir' Elevation (ft ms!)

... . Top. of Dam Elevation' Full Pond. PMF Dam Structure .. (ft msl) (800.0). (808,.9)

Keowee Dam 815.0 15.0 6.1 ONS Intake Canal Dike 815.0 15.0 6.1 Saddle Dikes A, B, C, and D 815.0 15.0 6.1 Little River Dam 815.0 15.0 6.1 The PMP used in the evaluation of Lake Keowee was estimated from HMR No. 51, Probable Maximum PrecipitationEstimates, United States East of the 105th Meridian (HMR-51). The Probable Maximum Storm (PMS) which generates the PMF was estimated using computer techniques outlined in HMR No. 52, Application of Probable Maximum PrecipitationEstimates, United States East of the 105th Meridian.

A set of depth-area-duration (DAD) curves (Figures 4 and 5) was prepared for the Keowee drainage basin using the all-season PMP charts published in HMR-51 ,and the basin centroid location. The PMP depths in inches were for storm durations ranging from 6 to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and for storms smaller and larger in area than the drainage basin.

Figure 4. PMP DAD curves for Keowee watershed PMP depth vs. Duration 450, 40.0

_ -- o 2>00 mn12

35.0 ont A* I 1.000 mL.2

~25.0

-.--- o 5,00O nil.,?

20.0 ._,,._.*---* 10,000 fTiI2

..---- 20,000 rnL2 10.0 0.0 I--

0 0 12 18 24 30 38 42 48 54 60 68 72 78 84 Dura~tn (?u.)

18

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation Figure 5. PMP DAD curves for Keowee watershed PMP depth vs. Drainage Area AO.0 q "*.. ..

• 35.0 3.0.0 25.0 ----

0"N 01 o 25.0 N. ho'0 10 100 1.o00 10.000 100,000 Dwing. area (rrt.)

The drainage basin for the Keowee Development covers approximately 435 square miles of the southeastern slope of the Blue Ridge escarpment. The drainage basin terrain transitions from rounded mountains through foothills to Piedmont. The land is characterized by dense forests -

some original growth, but most of the forest is mature second growth. Except for the tops of ridges, the vegetation is generally dense and the soils are permeable and absorbent. The small amount of farming in the region is confined to the wider portions of valley floors and the Piedmont portion of the drainage basin. In the southern part of the basin, there is more development than in the rest of the basin. Development is generally characterized by scattered rural housing and small towns.

The Keowee drainage basin does not fulfill all of the terms of the definition of a gauged basin as described in the FERC guidelines. To the extent that the basin does not meet the definition of a gauged basin, it must be analyzed as an ungauged basin. For this reason, a regional study was utilized to develop unit hydrographs for the sub-basins in the drainage area of the Keowee Development.

The Jocassee Development is the largest upstream development with substantial storage capacity. The PMF was routed through the Jocassee Development by level pool routing.

Theoretically, the routing parameters at Jocassee could be set to help reduce the peak reservoir elevation at Keowee by holding back additional runoff to reduce the peak reservoir elevation at Keowee. However, Jocassee was modeled to operate as if it were experiencing a PMF.

Therefore, to be conservative, the same model input parameters that were specified for the 19

8,.,,J

,I E,,,,,,;II.7- .. i..r"t t;, ... .;.........

I'fZ.... ... ~..., ... *,., 0 ur Duke Energy Carolinas, LLC I O*NS FLOODING HAZARD Section REEVALUATION 2-* Flooding REPORT Hazard Reevaluation .w, Jocassee Development PMF were used in the Keowee model. During a PMF event in the vicinity of either of these adjacent developments, both Jocassee and Keowee would be operated as if each of them were expecting a PMF event.

There are a number of ponds and small lakes in the Keowee drainage basin. These are characterized by passive sp~illways and negligible storage capacity. These structures have very little effect on the stOrm hydrograph as it passes through their ponds, and the total storage volumes are small so that even a breach would not have any measurable effect on reservoir elevations at the Keowee D evelopment.

The SCS runoff CNs were used to model abstractions and compute excess rainfall. The CN method 'relates cumulative rainfall to cumulative excess rainfall based upon the single parameter CN which ranges in value from 0 to 100 depending on the combination of soil type and land cover/use. The CN 'method is widely used for PMF studies as it is incorporated directly into the HEC-1 Computer model. The Geographic Information Systems (GIS) facilitated these computations.

To determine the sCS curve number for a sub-basin, both land use and hydrologic soil group must be known. This information is combined using the published tables (SCS 1972) to establish the curve number value for th~e sub-basin.

Soil types were obtained from the soil surveys of Oconee, Pickens, Transylvania, and Jackson counties. Land use information was obtained from the U.S. Geological Survey (USGS) for North Carolina and the Land Resources Commission for South Carolina. The resolution on this data was 60 meters so that every 60-by-60 meter cell was characterized by one of the following six land use categories: urban, agricultural/grass, scrub/shrub, forest, water, and barren/transitional land. A determination of the set of CNs for the "water" land use category is not relevant because water is accounted for in the determination of percent impervious. SOS CNs for each combination of hydrologic soil group and landluse were then area-weighted to determine the average CN for each sub-basin.

The drainage basin was divided into 26 sub-basins as shown in Figure 6. Sub-basins I through 25 were delineated on USGS quad Sheets (map scale 1:24,000) based on similarity of topography, geology, and land use. In order to use *the sub-basin outlines in GIS, the outlines were digitized from the quad sheets. As a matter of interest, the sub-basin outlines were digitized into MicroStation files. The MicroStation files were then imported into the Arclnfo GIS which produced an ASCII listing of the )(-Y coordinates for each of the*25 sub-basins. For sub-basin 26 (Jocassee basin), coordinates from the Jocassee HMR-52 input bekw re iought into GIS, transformed to the origin used for the Keowee sub-basins, converted to ASCii code, and added to the ASCII coordinate files.

The PMP DAD curves (Figures 4 and 5) and digitized sub-basin coordinates were used as inputs into the USACE HMR-52 computer pr'ogram. The HMR-52 program computed the storm size, location, and orientation that produced the maximum precipitation volume for the drainage basin (PMS). A 700-square-mile storm with centroid at X=10.3 miles and Y=I13.0 miles and 20

Duke Energy Carolinas, LLC IONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation orientation angle of 187° produced the greatest total rainfall volume for the drainage basin (Figure 7).

The centroid of the PMS obtained from HMR-52 was moved in 2-mile in~crements to obtain the PMF which produces the maximum reservoir elevation. The reservoir elevation was calculated using the level pool routing option in HEC-1 (dated February 1, 1985) assuming an initial reservoir elevation at 800.0 ft msl (full pond). The results indicated that the PMS with centroid at X=1 0.3 miles and Y=9.0 miles is the PMF.

It was not necerssary to perform a separate regional unit hydrograph analysis for the PMF study because of the availability of a state-wide study performed by the USGS (Bohman, 1989). The PMF study utilized available hydrologic data from gauges in the vicinity of the study area and derived regional regression equations for unit hydrograph lag time. Dimensionless unit hydrographs were developed for the two major physiographic regions in the study area (ie.,,

Blue Ridge and Piedmont). An independent review of the unit hydrograph _analysis and verification study was performed by Clayton Engineering.

In review of the existing FERC PMF analysis, the effectiveness of the gated spillway structure and two hydro turbines at discharging the estimated PMF inflow was considered. Lake Keowee historically has not experienced significant floating debris field problems over the 40+ years since impoundment. The shoreline is stable and~debris is managed to facilitate boating activity on the reservoir. Due to the layout of the spillway structure and the relatively low velocity profile in the two reservoir arms during a natural flood~event like the PMF, floating debris is not anticipated to become a problem that could significantly impact the effectiveness of the four spillway gates.

The PMF outflow hydrographs and corresponding reservoir levels are shown on Figure. 8. The outflow hydrograph is a product of the inflow hydrograph, reservoir storage, and dam operation.

The PMF produced a peak inflow of approximately 332,721 cfs, The estimated PMF peak discharge is 139,961 cfs. The PMF peak reservoir (headwater) is approximately 808.9 ft msl resulting in 6 ft of margin between the peak flood elevation and the top of dam at 81 5*ft msl.

21

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation Figure 6. Keowee watershed sub-basins 0 ___ £ I 6 60 IMkI 22

Duke Energy Carolinas, LIC [ ONS FLOODING HAZARD REEVALUATION REPORT I-y~

Section 2 - Flooding Hazard Reevaluation Figure 7. Table 6.2 PMS from HMR 52 which generates PMF mesas 44010 9105 tOt 4001-2.

mw~o 4145

• 144 43. rj.. 03w20flm . 381.. 00190490 082WTsflfrJ

• 122.

27340 1126400t4313410, A

• 24.). T* 0.4 22201 iii f 0 3 2)W£f 9+

2 10 41o~* 32 9- 03. 34- 0.14 1,1 2,00 44I1 4.15 9.15 9.93 4.53 0.1 041+

• 35 3, 24.10 4.03 5.11 0,40 0.12 5911 9,42 034 13 71 1.24 0.31 1.14 8.14 o so* 0,49 0 04 t IOU. lIDOL 0,99 4.99 4.'. 3.37 0.1 0.4 coO.4 9 190 144 5.24 4.41 to'l 4.11 420S 0.04 003*

.1.4 oakI O',

• 32 3.12 343t a.ti 9,1$ *.44 (.9, 39.41. 4.42 8.0 0.39

o. iii. 211. 4,01 0.44 4 14

• 3') 90 404 02%*

24.4 2.21 l.AS 0.04

£ 414*. 416. 4.43 8.41 9.81 2.21 4 40  !.0 29 ,11 5,10 9.00 8.12 8.11 6,24 4.41 430 9.21 0.31 9.l1 210410. 421. 1444 8.46 0.44 *.4@4 4.4, U.nS 8.21 0 .44 3.20 4.44 6.00 441 3.08 4344 i4.01 4.1 945i 3.,93 3.44 4.2 4.14 4.04 9.50 4-44T 980 4)00 000 0.03 0.00 9,31 r:I 110mmO.:. 9 34. m.JmS 34 01.3 941 09 201411

• 1 07 1121 60 9* o0r~ 260 Il ,iI'i *l 1I 4*I 1111 il 7I 1

£ .315 3.1 0.1 4.4 6.23 31.46ll)I..13 *0,11.0 2 1 2.1 18.14) 13.04 ) 1 3.41),lL),7 1.4 )1.0 94 34.46137 209.

4.

4 ,1 1.&4* -141 145 7,+43 II11320.11 210.21 19.16 01.3) 1t.21 14 IS 14,4' 2142l 28.1 2.41 27 12} 31-.5 C .04 1.431.2 4.21 7.23 2.1 4 20.l1,il30.42.4 4 33,4 *lOP 22.72 33 0322 30,10834 43 3)12 34.j3 35.94*,*

0 0.0 3. 21 340 3.42 4.7) I14.0 22.4 16.3) 24.1 11.2 03.40 13.41 23.0 12.47 04*1M+314923 11 19326 34.1-

1. 1.10 1.11N 1.49 .6 0.20 4.1 4 32.6',114.41 22.4 0.7 11,12212 24 2'I~l53.30 20.lrI14 I*I71 21T0 20p24 28.12 9 .15 3.21 ,13 0.12 0.140 I.0 00,81 13.44 11,44 24.31.0 04.2l1421.13 004.1 29*, IIS2,4A1 2.I 22- 31 2.01

• .56Il Oe1.96+1 3.393 .13.84 35.47 29l*l.4122.4 09112 I*11 ll.4 2.1 21).2121 442 41. 148019 03 21.42

.11

• 9.41 4.9 1.8 2.43 4.32 7.48 4i .07 14,41 14,1' .00.20.10Jl l 3I*

1.0 *1.34 *1,49 2)?J51 49 ,11 099 04.84 2 4.39 318*]r 0 ,2 4 l *

.5 a sQ40 lll 081.13 I 10.18 22.08 1.1 03.11 2. 10 1 9.91 11.42J 31.80 07?Ii.714-44 2 4.24 3.14 0.42 0.85 3.34 2.01 4.61 9.81.21.14 34.11 dq 1.44 14 1.83l 0,4 02 1*.4t 3i 45.414 + a et.

U i,.3) 3.94 0.05 0.11 4.32 4.81 4.11 3.72 2.11 2.24 2.41 1.34 1.24 4.34 4.14 4.03 4.31l 4.1 1.41.

O .90 8.42 0.0) 0.4' 4.i) 4,94 4.03 4.44 2.84 1.49 1.44 3.14 2 44 2,44 14 014Ii 033 30 25-19 P 4,440.8 04*

O 40,9 4.04 3.4 0.0 8.96 8.44 on *..1 4.30 o,l3 3.4 0.94 0 .,34 303 0.08 0.01 0 1.4 94(4 0.04 0.08 6.40 4.40 8,90 4.241 0.00 4.l 4-H.0 5.0 4.00 8l.92 9.30 00 04 4 0.14 4.49

• ,41<.00 <0.44 .4 0.03 3,30 0.40 4.40 4.414 4.00 4.4 0.49 3.90 0.88 4.X04 00 094 34s 004s 4.40 0419 4.12 0.132 821 3.11 8.44 3.01 23.31 11.07 22123. 27 34.50 43.34 20.01 29.8  !,40 21 20 82 32 32 38.1 32.24,1 947' 1 7190 91W1319477190 Yl P*)Its 1010.'*

uo iwroo 4100 0.01 4.47 Ifl0O 4O 23 .2.

4226 0.83 4.38 0900 C.CI 041. 0134 81) 114*

4230 4.51 4.22 110- 41.34 1.97 4423 8.31 0.26 2.049 5.04 I 5.193 1020 0.20 I. 40 47.03 3 51 4,74 0320 0,03 4.43 8494 03 317l' 4520 3.97 4.43 1142 434t1.* 235+?

4.44 305.AJ. 043 2202 4.14 2I.21 61073 4t*.01. 4. 2201 4.00 II $

4296 00~891421040204 010 10143 824£G0.03971 .110 4212 6.12 2 10 3330 1.42 24.41 4230 0 11 3.41 1*119 4.14 2/4.4 0112 628 2.01 4421- 4.24 2 44 1424 1.4.2 41..24 0546 0.33 1 12 4444 1.11 445 4,4' 2+00 II 1.1 1,4 4.40 70714. 1.32

$110- I.2,4 29.1 310 30, 21 9301 0342 4.10 21.31 be I 13910 9'1t213lTMTrh 323* 4.21 2904Z 7190 9t291t&7202 203 4 ..

o 1. 14.4 12014 12274L 829 4, 0 4 4214 0.23 11.44 3130 4.02 20.10 ebQ 4.44 1-0.4) 3410 4.02] 20.47 234.,

, 9 70744+

4200 8.14 20.44 4410 0.10 23.4) 3000 4.21 3* 92 021 .11 211 4420 9.39 24.21 2300 4.12 23i0 4129 8.19 z:.ua 0000 4.01 33.31 4.4 10211. 3.30 4.14 23

. .... u .... . ..... rit . . .. t.I*.~. . . r ..... -fPE f( * #J(

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation Figure 8. Keowee PMF hydrographs 24 3G 38 42 46 54 60 66 12 78 84 80 86 Th~o 0u.)

2.2.2 Probable Maximum Flood - Jocassee The PMF has been evaluated in accordance with current engineering guidelines for Jocassee; a FERC-licensed and regulated hydroelectric development (#2503). The peak reservoir elevation from that analysis is 1+122 ft msl based on hydrologic and hydraulic modeling performed in accordance with Chapters II and VII! of the FERC guidelines using l-MR51i52 for the PMP development and HEC-1 for the hydrologic and routing model. The spillway, pump-turbines, and dams at the Jocassee Development have been designed to pass the PMF with adequate freeboard and structural stability factors to safely accommodate the effects of the PMP. Table 6 summarizes the normal and PMF freeboard for the dam structures that impound Lake Jocassee.

Table 6. Jocassee Dam - Normal and probable maximum flood freeboard (ft)

Reservoir Elevation (ft ras!)

Top of Dam Elevation ' .. Full Pond PMF Dam Structure (ft msl) (1,110.0) -(1,122.0)

Jocassee. Dam 1,125.0 15.0 3.0 Dike 1 1,125.0 15.0 3.0 Dike 2 1.,125.0 15.0 3.0 Generalized all-season estimates of PMP for the Jocassee watershed were obtained from HMR

51. Following procedures given in HMR 52, a set of DAD curves (Figure 9) was prepared for 24

LNmroll E*

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Scin2 - Floodng Hazard Reevaluation the Jocassee drainage basin using th all-season PMP charts published in HMR 51 and the basin centroid location. The PMP depths in inches ware for storm durations ranging from 6 to 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and for storms smaller and larger in area than the drainage basin. The HMR 52 computer program was used to calculate the PMS in each sub-basin in the watershed. The average PMS depth over the entire watershed was 36.41 inches in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. Average sub-basin depths ranged from 38.32 inches near the centrold to 25.91 inches near the periphery.

The USACE computer program HEC-1 was used to calculate the inflow hydrograph to Jocassee reservoir. For the computer model the Jocassee Lake watershed was conceptually subdivided into 42 separate basins (Figure 10). The lag-times calculated for each sub-basin and the SCS dimensionless unit hydrograph were used in HEC-1 to compute synthetic unit hydrographs. The direct runoff hydrographs for each sub-basin were calculated by computing rainfall excess depths at each time step using the SOS ON Method and then applying these depths through the unit hydrograph procedure. The series of hydrograph combinations and stream routings was continued until a total inflow hydrograph was computed for the Jocassee reservoir. Channel routing was performed using the HEC-1 normal depth routing option. Stream channel cross-sections were obtained from 7.5-minute series USGS topographic maps for the watershed.

Figure 9. Jocesses PMP DAD curves PMP depth vs. Area 4a 4a U1S **"*'- ~w' I...- III jI

,,,,,,,,,, I

• 1 S I

'0 st - 1tUGS Aies* a*I The SOS runoff ON was used to estimate infiltration rates during design rainfall events. The hydrologic soil group for the watershed was estimated based on studies of soil types using generalized state soil maps for South Carolina and North Carolina. The soils in the watershed generally fall into the "B" hydrologic soil classification. The watershed soil cover is basically forest in good hydrologic condition. These factors lead to the selection of a runoff CN equal to 55.

25

Duke Energy Carolinas, ILLC ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation In 1993, Duke Energy conducted a sensitivity analysis of the watershed by assuming an equivalent uniform infiltration rate of 0.093 inch/hour and back calculating a CN of 60 using the SCS CN formula. This parametric study indicated that varying the CN did not result in any significant changes in the peak stage.

There are no continuous recording rainfall or stream gauges in the Jocassee watershed that would allow computation of a unit hydrograph for the basin. This fact required the use of synthetic unit hydrographs to model the watershed rainfall/runoff response. The Kirpich method was used to define the hydrograph lag-time in all of the Jocassee watershed sub-basins.

26

..... .iuw u~..

.. ... ......... ..... u~uI~pwr1U I.#K Duke Energy Carolinas, LLC IONS FLOODING HAZARD REEVALUATION REPORT 1-m,'

Section 2 - Flooding Hazard Reevaluation Figure 10. Jocass.. watershed sub-basins There are four reservoirs within the watershed upstream of Lake Jocassee: the Bad Creek pumped storage reservoir, Fairfield Lake reservoir, Sapphire Lake reservoir, and Lake Toxaway reservoir. The following setins describe the assumptions that were included in the HEC-1 model for reservoirs upstream of Jocassee.

Bad Creek Pumoed Storaoe Reservoir Reservoir storage and elevation vary considerably over the course of a week. The iniJtial conditions assumed to apply to this lake at the time of occurrence of the design storm were based upon an average storage of 16,565 acre-feet (ac-if) at a lake elevation of 2,269 ft msl.

The surface area of the lake at this elevation is 0.29 square miles and the drainage into the 27

Duke Energy Carolinas, lIC I oNs FLOODING HAZ*_ARD REEV'ALUATION R*EPORT Section 2 - Flooding Hazard Reevaluation reservoir is very small at 1.39 square miles. The principal means of outflow from the reservoir is through four pump-turbines, which have a combined discharge capacity of 8,000 cfs. The East Dike was designed 2 ft lower than the Main and West Dams as an emergency spillway. The rock-fill design of the East Dike would provide protection from erosion due to overtopping. The 2-ft design height difference, storage, and 4-pump-turbine discharge provide adequate margin for a site PMP. The reservoir operation is-modeled by assuming no outflow up to an elevation of 2,310 ft msl at which point the four pump-turbines start discharging at a constant rate of 8.000 cfs. The HEC-1 model also provides for dam overtopping using the East Dike ifthe water surface elevation rises to the crest elevation of 2,313 ft msl; however, this elevation was not reached in the simulation. Bad Creek is capable of safely passing the sub.'basin specific inflow associated with the Jocassee PMF by a combination of storage and discharge.

FairfieldLake Reservoir The Fairfield Lake Dam was built in 1895. Under present state and federal criteria, the appropriate spillway design criteria are one-half PMF and full PMF, respectively. The USAGE Phase I inspection report indicates the dam will be overtopped for both the one-half and full PMF (USAGE PMF= 14,800 cfs) by 3.4 ft and 5.2 ft, respectively. The normal pool and spillway are both at an elevation of 3,155 ft msl. The dam crest is at an elevation of 3,159.6 ft msl. The only discharge from the dam is through a double arch culvert spillway with a pipe or box of 6-ft nominal radius. The total storage capacity of the reservoir at the top of the dam is approximately 1,040 ac-ft. Breaching of this reservoir at 0.5 ft of overtopping was simulated in the HEC-1 model.

Sapphire Lake Reservoir The normal pool elevation is 3,100 ft msl and the lake at this elevation covers 46 acres. Two 36-inch-diameter concrete pipes are employed for discharging low flows. In the Jocassee watershed model, the Sapphire Lake Reservoir is modeled with discharges through both the low-flow concrete pipes and dam overtopping along the concrete masonry crest. No dam breach formation was assumed. Sapphire Lake Reservoir is capable of safely passing the sub-basin specific inflow associated with the Jocassee PMF by a combination of storage. and discharge.

Lake Toxawav Reservoir The normal pool elevation is 3,000 ft msl with a lake covering 514 acres. The dam has a side-channel spillway that is 115 ft wide. One 48-inch-diameter gated pipe with a crest elevation of 5 ft below normal pool is used for low flow discharging. The Lake Toxaway Reservoir was modeled by considering the discharge through the side-channel spillway only. Lake Toxaway is capable of safely passing the sub-basin specific inflow associated with the Jocassee PMF by a Combination of storage and discharge.

Lake Jocassee Reservoir The normal pool elevation varies between 1,108 and 1,110 ft msl within the week, but the maximum power-pool during the hurricane season is 1 ,108 ft ms!. All four pump-turbines were assumed to be available to pass flood flows during the PMF. The constant four pump-turbine 28

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation discharge is approximately 28,000 cfs. BOth Tainter gates were assumed to be fully functional during 'the PMF.

The spillway rating curve for the ogee was computed using standard methods,, accounting for abutment and pier energy losses. These discharges were then combined to produce the reservoir outflow rating curve. The 1,110 ft msl initial water level elevation selected for modeling is somewhat higher than the maximum power pool level of 1,108 ft that historically occurs during the hurricane season. Storage-elevation values are calculated by HEC-1 using the conic method based upon the elevation versus water surface area relationship.

In review of the existing FERC PMF analysis, .the effectiveness of the ability of the Jocassee Development to pass the PMF was evaluated by reviewing the installed equipment to pass flow, lake level monitoring equipment, procedures to anticipate and respond to storm events, and maintenance practices to ensure that equipment will be reliable.

The H EC-1 model was used to test the sensitivity of availability of the two spillway gates combined with four pump-turbines where different combinations of pump-turbine availability were examined. These simulations show that adequate margin to pass the PMF is proVided by two gates' without pump-turbines. The two 30-ft by 38-ft-wide radial gates have a proven record of reliability throughout the Duke Energy fleet due to the simple and effective design. There are four different ways to operate the gates including: the normal installed electric motor, backup air driven motor operation, backup electrical generator driven motor operation, and manual operation with a hand wheel. All equipment for the four diverse methods of operation is located at the spillway gates. Personnel are trained and procedures are used to direct the required actions. Reliability is also assured through a comPrehensive periodic maintenance program as required by FERC including an annual test to open the gates and 5-year full open testing. Amp readings for the motors provide readings to verify that the spillway gates are functioning as expected.

In addition to having reliable spillway gates, the four pump-turbines provide significant additional margin for protection against overtopping. A conservative assumption was made that based on maintenance and operational procedures only two of the four hydro units will be available to pass flow. This, in addition to the spillway gates, yields additional margin.

Spillway gate operation and unit operation were evaluated for the impact of debris. Lake Jocassee historically does not have significant floating debris due to the stable Shoreline, and a debris management program is in place to facilitate boating activity on the reservoir. The spillways and intake structures are also protected by the submerged openings of the spillway and intake structures. The PMF event in an impounded reservoir results in a low velocity profile in the reservoir during a natural flood event which limits the generation and transport of significant debris.

Monitoring and control of the reservoirs is assured by redundant level measurement in the reservoir and tailrace. Multiple video cameras provide redundant means to vedify and monitor levels. All level monitoring including level alarms is provided to both the control room op site and to a central hydro control station that is staffed 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> per day.

29

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT

_Section 2 - Flooding Hazard Reevaluation inventory control for storm events is managed through a combination of 15 ft of minimum reserve freeboard and procedures to anticipate major rain events. Weekly rain forecasts are prepared by Duke Energy and system operations of Bad Creek, Jocassee, and Keowee are coordinated based on the forecasted rainfall for thie basin. This planning provides additional time to manage reservoir levels to anticipate large precipitation events like the PMF. Hydra Fleet Central Operations Center procedures require consideration for lowering reservoir levels in anticipation of significant storms. For normal operation, the weekly projection and dailylhourly precipitation monitoring has resulted in no precipitation-based spillway operations at Jocassee over 40+ years of operation.

The HEC-1 Jocassee watershed PMF modeling results were developed for each sub-basin including the five reservoirs. All reservoirs except Fairfield Lake were capable of safely passing the sub-basin specific inflow associated with the Jocassee PMF by a combination of storage and discharge. The modeling showed Fairfield Dam overtopped during the PMF; but due to the limited storage behind the dam (1,040 ac-ft at top of dam), breach inflow to Jocassee was managed through storage and discharge.

The PMF Outflow hydrographs and corresponding reservoir levels are shown on Figure 11. The outflow hydrograph is a product of the inflow hydrograph, reservoir storage, and dam operation.

The PMF produced a peak, inflow of approximately 522,734 cfs. The estimated PMF peak discharge is 85,405 cfs. The PMF peak reservoir (headwater) is approximately 1,122.0 ft msl.

Overtopping of the Jocassee Dam is not considered a credible event due to the multiple and diverse means of passing water and the large margin of storage capacity associated with the reservoir.!

30

Duke Eneg C.anroinasm, LLC I ON FLOODING HAZARD REEVALUATION REPORT Scon2 - Floodng Hazard Reevaluation Figure 11. Jocasseoo IPMF hyclrogrphs 44004 II14

,,* -m 4-es.

3111 1124 4*6166 -

340004l 1110 i 4 o(  %

Illi I'.i

'IDO ila4 qu 4*i00 .u

'~ -

"at'l 2.3 Dam Failures 2.3.1 Potential Dam Failure The Keowee Hydroelectric Development is owned and operated by Duke Energy, and is part of the Keowee-Toxaway Project licensed by the FERC. The Keowee-Toxaway Project includes the Jocassee Pumped Storage Development and the Keowee Development. The FERC regulates the Project and their role includes regulatory oversight of all water-retaining structures at both developments. The ONS is located on Lake Keowee adjacent to the Keowee Dam and tailrace.

Dam failures at the Jocassee and Keowee Developments have been examined for potential impact at ONS. The loss of Lake Keowee initiated by a failure of the Keowee Dam was postulated as a design criteria event that was focused on cooling water supply and not on a dam failure initiating event. Based on a review of the UFSAR, modeling performed between 2008-2011 to respond to the August 15, 2008 10CFR 50.54(f) information request and the January 28, 2011, NRC Staff Assessment of Duke's response to the Confirmatory Action Letter regarding Duke's commitment to address external flooding concerns, it was determined that a hypothtical fair-weather failure of the Jocassee Dam and a resulting cascading overtopping failure of Keowee Dam would be the enveloping critical dam failure event for ONS.

31

n - "v....... , ......... f.. .A  ; t.. ...... >.....

.... L1 - ..... Z m ..Z. ZZ~v j*i.

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Srn~ian 2 - Flooding Hazard Reevaluation 2.3.2 Dam Failure Permutations FERC regulations require that dams for which failures pose a risk to human life be designed to survive a combination of loading conditions including forces from water and seismic events without risk of failure. Both the Jocassee and Keowee Dams were designed, constructed, and are maintained to high standards and have performed in accordance with these stringent standards for over 40 years.

Consideration of upstream dam failure initiating events criteria as described in ANS 2.8 will be discussed in the following sections including hydrologic event dam overtopping failure, potential seismic loading induced dam and water-retaining appurtenances failure, fair-weather piping dam failure, surge and seiche, ice-induced flooding, tsunami and channel diversion.

2.3. 2. 1 PotentialHydrologqic Overtopping As a result of reviewing the dam structures at Jocassee and Keowee for criteria outlined in ANS 2.8 for design basis floods and comparing to the design basis requirements required by the regulating agency, FERC, it was determined that the Jocassee Development and Keowee Development have adequate margin to meet the criteria for overtopping and stability as required by NTTF Recommendation 2.1. Both Jocassee and Keowee can safely pass their respective PMFs without overtopping as described in Section 2.2.

2.3.2.2 PotentialSeismic Failure The Jocassee Dam is a robust rock-fill dam that is designed to be stable under site-specific seismic loading. Deformation along the respective slip planes at the Maximum Section and Abutment Section is insignificant and would not result in a breach failure of the dam. The postulated seismic-induced failure of any of the water-retaining appurtenant structures at the Jocassee Dam site was compared to the bounding case fair-weather failure of the Jocassee Main Dam presented in the Flood Hazard Reevaluation Report (FHRR). The appurtenant structure peak discharge and reservoir volume release, due to the postulated seismic-induced failure, is significantly less than the fair-weather failure of Jocassee Dam. Therefore, the Jocassee Dam fair-weather failure and resulting flood inundation at ONS presented in the ONS-2.1 FHRR is affirmed as the bounding case for external flooding at ONS.

Duke Energy performed new seismic analyses for the Jocassee Dam as requested by a March 20, 2014, Request for Additional Information to update the previous analysis initially submitted in the FH-RR. Duke Energy performed a new combination deterministic/probabilistic analysis per the guidance provided in JLD-ISG-2013-01 to determine the dam factor of safety and displacements at the ground motion response spectrum (GMRS) level. A series of slope stability and dynamic response analyses were performed for representative slip surfaces (sliding masses) of the maximum and abutment sections of Jocassee Main Dam. For the post-earthquake loading condition, the calculated downstream and upstream slope stability factors of safety are above 1.5 and 1.6, respectively for both sections. The calculated earthquake-induced displacements for representative sliding masses of both sections are insignificant (less than 0.1 cm or 0.04 inch). Since the calculated post-earthquake slope stability factors of safety are well above 1.0 and the estimated earthquake induced displacements are relatively small, 32

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation the Jocassee Main Dam is expected to perform satisfactorily following the design earthquake ground motions considered. No fragility level was calculated.

2.3. 2.3 Site Geoloqy Characteristics The .Jocassee Dam site is underlain by gneiss bedrock. The weathering profile of the rock consists of residual soil (saprolite) transitioning with depth to less weathered materials and finally to the unweathered rock itself. The foundation of the maximum section of the dam was excavated to expose a zone (layer) of weathered rock overlying sound rock. The shear wave velocity of these materials was estimated from field measurements in the gneissic rock at the Oconee Nuclear site.

Figure 12 shows the coordinates used to perform a site-specific probabilistic seismic hazard analysis (PSHA) for Jocassee. Geologic information in this area is provided for two individual soil profiles, "JD-I" and "JD-5". These profiles are combined into a single composite profile, "composite profile 1", for the site-specific PSHA that was performed. Figure 13 shows shear wave velocities of JD-1, JD-5, and composite profile 1 versus depth relative to the control point elevation (CPE) of each profile. The CPE of JD-1 is 690 ft msl and JD-5 is 840 ft msl.

33

Duke Energy Carollna.. LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation Figure 12. Map of Jocassee Main Dam location showing Latitude and Longitude. North is to the top of the figure

.. ... t C* t ihi

• R.CTNO* l Om~.ONO Cw~Un~

34

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation Figure 13. Composite profile 1 representing geologic conditions below the control point elevation at .Jcse Dam (LCl, 2014). Profiles JD-1 and JD-5 are Individual profiles at Jocse from AMEC (AMEC, 2014b) that are combined into composite 1. "Lower" and "Upper" profiles correspond to the lower-range and upper-range cases, respectively, to capture epistemic uncertainty In mean shear wave velocity per the SPID (EPRI, 2013a).

Vs (ft/sec 0 2000 4000 60 8000 10000 o II 20 60~ I

~'80I I1Q0I S160 1 pe 180 35

Duke Energy Carolinas, LLC IONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation 2.3.2.4 Seismic Hazard Methodoloqy Description The methodology used to calculate site-specific probabilistic seismic hazard analysis (PSHA) at Jocassee is documented in Lettis Consultants International (LCI) (LCI, 2014) and follows Seismic Evaluation Guidance: Screening Prioritization and Implementation Details (SPID; EPRI, 2013a) and Regulatory Guide 1.208 (USNRC, 2007). The process involves:

1. Perform PSHA for hard-rock conditions at Jocassee using:

• Central and eastern United States (CEUS) seismic source characterization, CEUS-SSC (2012);

• EPRI (2013b) CEUS ground motion prediction equations (GMPEs);

• A revised seismicity catalog for select sources that removes several earthquakes identified as reservoir-induced seismicity and aftershocks related to the Charleston event (Youngs, 2014);

2. Determine the frequency- and amplitude-dependent site response (median values and standard deviations) for all spectral frequencies of interest using the results of step 1;

3. Perform PSHA for soil conditions at Jocassee using:

• CEUS-SSC (2012);

• EPRI (201 3b) CEUS GMPEs;

• A revised seismicity catalog for select sources that removes several earthquakes identified as reservoir-induced seismicity and aftershocks related to the Charleston event (Youngs, 2014);

• Determine the site response.

2.3.2.5 Seismic Hazard for Hard-Rock Conditions The procedure to develop probabilistic site-specific control point hazard curves for hard-rock conditions follows the methodology described in Section 2 of the SPID (EPRI, 2013a). This procedure computes the site-specific bedrock hazard curve for each of the seven spectral frequencies for which ground motion equations are available.

CEUS-SSC background seismic sources within a 400-mile (640 kin) radius of Jocassee were included. This distance exceeds the 200-mile (320 kmn) recommendation of USNRC (2007) and was chosen for completeness. Background sources included in this site analysis are the following:

• Atlantic Highly Extended Crust (AHEX)

• Extended Continental Crust--Atlantic Margin (EGGAM)

• Extended Continental Crust--Gulf Coast (ECCGO)

• Illinois Basin Extended Basement (IBEB)

• Mesozoic and younger extended prior - narrow (MESE-N)

• Mesozoic and younger extended prior - wide (MESE-W)

• Midcontinent-Craton alternative A (MIDCA)

• Midcontinent-Craton alternative B (MIDCB)

• Midcontinent-Craton alternative C (MIDCC)

• Mldcontinent-Craton alternative D (MIDCD) 36

. °Duke Energy .. ....

Carolinas, LLC I ONS

~.... . .. . ... FF.. .. .......~

FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation

• Non-Mesozoic and younger extended prior - narrow (NMESE-N)

• Non-Mesozoic and younger extended prior - wide (NMESE-W)

• ] Paleozoic Extended Crust narrow (PEZN)

[] Paleozoic Extended Crust wide (PEZW)

[] Reelfoot Rift (RR)

• Reelfoot Rift including the Rough Creek Graben (RR-RCG)

• ] Study region (STUDY R)

For sources of large magnitude earthquakes, designated Repeated Large Magnitude Earthquake (RLME) sources in CEUS-SSC (2012), the following sources lie within 1,000 km of the site and were included in the analysis:

[] Charleston

[] Commerce

• ] Eastern Rift Margin Fault northern segment (ERM-N)

• ] Eastern Rift Margin Fault southern segment (ERM-S)

[] Marianna

• ] New Madrid Fault System (NMFS)

• ] Wabash Valley For each of the above background and RLME sources, the mid-continent versions of the updated CEUS EPRI GMPEs were used as well as a lower-bound moment magnitude of 5.0.

2.3.2.6 Site Response Analysis Site effects are accounted for using 1-D equivalent-linear analysis of a Soil profile representative of the geologic conditions at Jocassee Dam between bedrock and the CPE. Two borings at Jocassee are combined into 'composite profile 1', (Figure 13). This profile is randomized according to the recommendations of Appendix B of the SPID (EPRI, 2013a) to account for aleatory and epistemic variability as they relate to shear wave velocities, layer thicknesses, equivalent-linear material properties (shear modulus and damping), and depth to bedrock. The dynamic response of each randomized profile is then determined using random vibration theory (RVTI) for a number of input control motions developed from the hard-rock hazard calculated in step 1 at a range of PGA amplitudes. The re'sults of this analysis yield site amplification functions (median and standard deviation) for a range of frequencies and amplitudes which are incorporated into the following site-specific PSHA.

2.3.2. 7 Seismic Hazardfor Soil Conditions The procedure to develop probabilistic site-specific control point hazard curves used in the present analysis follows the methodology described in Section B-6.0 of the SPID (EPRI, 2013a).

This procedure (referred to as Method 3) computes a site-specific control point hazard curve for a broad range of spectral accelerations given the site-specific bedrock hazard curve and site-specific estimates of soil, soft-rock, or firm-rock response and associated uncertainties. The same seismic sources (background and RLME) are incorporated as specified above for hard-rock hazard. This process is repeated for each of the seven spectral frequencies for which ground motion equations are available. The dynamic response of the materials below the 37

m L i Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Sedion 2 - Flooding Hazar Revuation control point is represented by the amplification functions described above. The resulting control point mean hazard curves for Jocassee are shown in Figure 14 and the corresponding uniform hazard response spectra (UHRS) are shown in Figure 15. Figure 14 (LCI. 2014) shows mean seismic hazard calculated for a range of spectral frequencies and amplitudes at the Jocassee dam site (Jocassee).

Figure 14. Site-specifIc mean soil hazard curves for seven spectral frequencies at Jocassee (LCl, 2014).

1E-3 I 1E-4 1E-S

-25

-10

-PGA

-25 Hz Hz S5Hz Hz

-1 Hz i 1E-6

,,,-0,5 Ha 1E-7 0.01 0.11 10 Spectral acceleration (g)

Note: g - the acceleration due to Earth's gravity. feet per second squared Hz - counts per unit of time, the SI unit for frequency is hertz Figure 15 (LCI, 2014) shows horizontal mean UHRS for mean annual frequencies of exceedance (MAFE) of 10.4, 105 and 104 at Jocassee using control point elevation (CPE) information provided in AMEC (AMEC, 201 4b).

38

Duke Energy Carolinas. LLC IONS FLOODING HAZARD REEVALUATION REPORT Stoi2 - Flodng Hazard Revaluation Figure 15. Horizontal mean UHRS for MAFEI of 104, 10"', and 104 at Jocassee (CId, 2014).

10.*

-1E-6 i -IE-5 M -1E-4 0.1 (u.

0.01 o01 S0~ 100 Frequency, Hz 2.3.2.8 Seismic PerformanceEvaluation Seismic performance evaluations were completed on two representative sections of Jocassee Main Dam including the right abutment section (looking downstream) located at Station 8+00 and a maximum section located at Station 12+00. Site-specific seismic ground motion time histories were used to evaluate the selected representative cross sections. In accordance with the combined events flood hazard evaluation guidelines in ANS 2.8 and NUREG7046, the site-specific SSE used for the analyses is consistent with the UHRS defined at a mean annual frequency of exceedance of 10-4 . This spectrum was calculated from a PSHA conducted by LCl (LCI, 2014), and further subdivided into high-frequency (HF) and low-frequency (LF) components based on hazard deaggregation per USNRC Regulatory Guide 1.208 and NUREG/CR-6769. Three sets of three-component time histories for the LF spectrum were used to analyze the dam. In addition, the Jocassee appurtenant water-retaining structures (Saddle Dikes 1 and 2. gated spillway, and intake tower/powerhouse water conveyance system) were evaluated to determine their respective potential seismic-induced failure discharge and reservoir volume release and associated inundation potential at ONS for comparison with the FHRR bounding case Jocassee Main Dam fair weather breach. The appurtenant stutres are located at various distances away from the Main Dam and their assumed failure discharges would not contribute to a failure of the Main Dam. The design of the appurtenant structures considered seismic loadings; however for this evaluation, failure of the structures was assumed.

The results and conclusions are summarized in the respective seismic performance and appurtenant structures sections.

39

........... ,S nut alnoru .. .:~ i. ... I. ,.. ~ *z Duke Energy Carolinas, LIC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevatuation 2.3.2.9 Seismic Performance Two specified rivers and stream combination loading-conditions: were considered in the seismic dam performance .evaluation consistent with ANSI/ANS-2.8-1 992, Section 9.2.1 .2 (Seismic Dam

-Failures).. The two loading alternatives are defined as: ...

Alternative 1 - Combination of

, ., A 25-year flood

• A flood caused by damnfailure resulting .from a safe shutdown earthquake (SSE), and Coincident with the peak Of the 25-year flood, and

• Waves induced by 2-year wind speed applied along the critical direction.

Alternative 2- Combination of

• The lesser: ofone-half of the PMF or the 500-year flood

• A flood caused by dam failure resulting from an operating basis earthquake (OBE), and coincident with the peak of the flood selected above, where the OBE is defined as half the SSE defined above, and

• Waves induced by 2-year wind speed applied along the critical direction.

Alternatives 1 and 2 require the combination of a hydrologic event and a triggering seismic event for the evaluation of the water-retaining structure. The hydrologic event is. assumed to increase the loading on the structure. A hydrologic and hydraulic (H&H) evaluation was performed to determine the peak Jocassee reservoir elevation associated with a 25-year (24-hour) flood event, 500-year (24-hr) flood event, and one-half of the PMF. The results of the H&H evaluationwere added to the dam stability evaluation to support the SSE .and OBE seismic performance review of JocasSee Dam. The results of the H&H evaluation indicate the 25-yr flood does not result in an increase in reservoir elevation and the 500-yr inflow is less than the half PMF inflow. The 25-year flood event Peak elevation is the-normal maximum reservoir elevation at 1,110.0 ft msl and the corresponding 500-year flood event peak elevation is 1,1-11 .5 ft msl. These results assume operator action to control the level by running the Hydra units to pass water downstream utilizing current Hydro procedures.

Seismic performance for the respectiVe Jocassee: Dam sections (Maximum and Abutment) was evaluated by performing a suite of geotechnical analyses that included a static stress, slope stability and dynamic stress/deformation analysis. These analyses represent current geotechnical practice for dam seismic stability evaluations and include:

• The static stress, analyses were performed using SIlGMAIW, a finite element software program for stress ,and deformation analyses of earth structures,

• The dynamic response analYSes were performed using QUAD4MU, a computer program for evaluating the seismic response of soil structures using finite element procedures,

• The slope stability analysis was performed using' software UTEXAS4, and

• Deformation analysis is performed using s~ftware TNMN to calculate the displacement of potential sliding mass using the Newmark method. The Newmark method assumes that the potential, sliding mass underlain by a well-defined slip .surface is rigid and the yield acceleration is constant. The Newmark method utilizes double-integration of the 40

Cu.iiaiue, *ezurILy SIbiuvu-flIurmaUaf; wi~nno~u Tr-m pUDIC per iu *,rt* v..;oJa~tJUiOJ(1 Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT

', ~Section 2 - Flooding' Hazard Reevaluation difference between the horizontal acceleration and the horizontal yield acceleration of the sliding mass.

Eleven slip surfaces were identified for each section of Jocassee Dam, five slip surfaces for the upstream slope and Six slip surfaces for the downstream slope. The slip surfaces evaluated include variable slip surface thickness (shallow, mid-depth, and deep) and heights of the dam, and were selected to represent various, sliding modes based .on engineering experience with similar structures. Acceleration time histories were developed specific to the Jocassee Dam site to match the 1 0 -4horizontal and :vertical low-frequency UHRS. The acceleration time histories (design earthquake ground motions) correspond to the Safe Shutdown Earthquake (SSE). The dynamic response analyses are performed for six combinations, of horizontal component (Hi, H2) and vertical (V) acceleration time histories. The resulting seismic performance evalUation identified insignificant deformation (less than i ft)of the dam structure that would not result in a condition that would lead to the breach of the dam.

2.3.2.10O Appurtenant Structures The methodology employed to evaluate the combined effects flooding impacts at ONS due to the seismic loading failure of a water-retaining appurtenance structure at the Jocassee Development is based on estimates of the peak discharge and volume release for each appurtenant structure. Four water-retaining appurtenances were identified by reviewing project drawings and the available FERC Supporting Technical Information Report including the regulating agency probable failure modes analysis.* These appurtenant structures are all independent structures from the main dam. The two saddle dikes, the gated spillway and the power production water conveyance sYstem were identified for assumed failure and discharge analysis. Estimates of the peak discharge~and volume release were. developed and compared to the fair-weather failure of the Jocassee Main Dam presented in the March 2013 FHRR. Review of the H&H analysis indicated there was not a significant difference between the 25- and 500-yr flood loading conditions and reservoir volume; therefore, the analysis focused on review of the potential discharge from a postulated failure of each of the appurtenance structures. A comparison of the size of the four appurtenant structures to the Jocassee Main Dam was made and resulted in the estimation that the peak failure release from an appurtenance was approximately 6 percent of the fair-weather dam failure discharge presented in .the FHRR. The percentage of released Jocassee reservoir volume, following a potential seismic event for the individual appurtenant structure failures is in the range of 6.5 percent to 39.1 percent of the Jocassee reservoir and is dependent on the particular invert elevation of the appurtenant structure. The-observance of an uncontrolled release from any of the appurtenant structures would result in an operator declaration of a condition "B". This, is procedurally mitigated in the same manner as a declaration of a Condition "B" for the main dam. Condition "B" is procedurally defined as a situation where failure may develop, but preplanned actions taken during certain events (i.e., major floods; earthquakes, etc.) may prevent or mitigate failure. The potentially hazardous situation may allow days or weeks for response and time to take remedial action.

The following volumetric comparisons of the appurtenant structures and the Keowee Reservoir

=provide additional assurance and conservatism that the *appurtenant structures are not the controlling case for external flooding at ONS. Each of the two spillway gates at Jocassee are 38 ft wide and 33 ft high. The available area for release through these structures is less than that of 41

~nta.....:lrit; :t:. v: l pll~giJ

.i~.*tiv r.,,

I V.I;Nr i  ;

Duke ,Energy Carolirnis, LLC I ONS FLOODING HAZARD REEVAL*UATION REPORT Section 2 - Flooding Hazard Reevaluation the four spiliway gates at Keowee which measure 38 ft wide and 35 ft high each. (Reference STI No. 2503-JO-01 Jocassee Development and STI No. 2503-KW-01 Keowee Development)

Additionally the uncontrolled flow that could potentially pass through the power production water conveyance system at Jocassee, approximately 28,000 cfs, is significantly less than the capacity of the four spillway gates at Keowee. At full pond Keowee reservoir elevation (800 ft msl), the spillway discharge of a single gate fully open is nearly 28,000 cfs. The available volume of water that could potentially be released by a coincident failure of both saddle dikes at Jocassee, assuming erosion scours down to the partially weathered rock (PWR) layer, is approximately 282;063 ac-ft. This breach volume is less than the available storage volume of Lake Keowee between elevation 800 ft msl and top of dam elevation 815 ft msl, approximately 303,738 ac-ft, taking no discharge capacity credit for the spillway gates or flow through the Keowee power prodUction Water conveyance system.

2.3.3 Unsteady Flow Analysis of Potential Dam Failures Since a seismic failure of the Jocassee Dam is not credible, a postulated non-seismic failure of the Jocassee Dam is assumed in order to demonstrate the corresponding flood hazard postulated at the ONS site. This section summarizes the resulting flood wave and corresponding flood elevations at the ONS site. To simulate a potential fair-weather failure of the Jocassee Dam, an unsteady flow model of the Jocassee/KeoweelHartwell reservoir system was developed using the USAGE HEC-RAS (Version 4.1) program. HEC-RAS is a 1-0 dynamic flow model with unsteady flow model components used in estimating inundation due to hypothetical dam failures.

The HEC-.RAS Jocassee-Keowee Dam Breach Model (ONS 1-D Model) was developed to include additional Lake Hartwell detail and cross-section refinement of Lake Keowee,. including the Keowee River, Little River,"and the connecting canal sections of the reservoir. The ONS 1-0 Model includes:

• Reservoir storage volume;

• Manning's n value sensitivity;

• Routing of water flow at Keowee Dam (breach and overtopping) to the Keowee tailrace;

• Tributary and storage area flood routing;

• Additional dam failures (ONS Intake Canal Dike, Little River Dam); and.

• River routing to tie the ONS Intake Canal Dike and Little River Dam outflow back into Keowee River.

HEC-GeoRAS Version 4.2.93 was used to develop model geometry independent of HEC-RAS using available GIS data from the State of South Carolina (with some overlap into Georgia) to create Digital Elevation .Model (DEM) electronic files for Jocassee, Keowee, and Haftwell reservoir~systems. A small portion of the lower southwestern corner Of Lake Hartwell was modeled using the USGS National Elevation Data Set (NED) to develop the storage volume.

The completed electronic geometry files were then imported to HEC-RAS, The ONS 1-0 Model consists of the Jocassee, Keowee, .and Hartwell reservoir systems with. an approximate length of 44 miles. The hydraulic responses of three, reservoirs comprising 17 42

~. oi=uou u.iy eauuv i,,.u,,,,ouur,, wi,,,uuiu ,,w,, =JUusI. i .m,  ;,~£.g'FU~I SDuke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation river/tributaries are incorporated in the ONS 1-0 Model. An independent peer review was performed on the geometry of the ONS 1-P Model, unsteady flow boundary and initial conditions, and results. The ONS 1'D Model's complexity and detail required numerous model sensitivity runs in order to optimize the ONS 1-D Model's performance and stability.

Additionally, a TUFLOW FV 2-0 hydraulic model was developed to simulate the complex flow paths moving between the Keowee and Little River arms of Lake Keowee including the ONS Intake Canal and Dike that abuts the ONS site at the south forming an island during damn breach simulations. The 2-D analysis was performed to add detail *to the HEC-RAS analysis and model the potential inundation in critical areas identified around ONS. A 2-D computational model was constructed with a domain of the area immediately surrounding ONS with a 2-0 mesh. The mesh size'was selected to model the desired area while keeping the size of the computational array at a manageable size and facilitate the analysis within the time frame of the study and within the computing power of current, publically available micro-computers. Tables and graphs referenced in this section can be found in Appendix 0.

Using an uncoupled 1-D model to inform both the upstream and downstream boundaries of the 2-D model requires that the results of the 2-D model at the boundaries be compared with the 1-D model and differences resolved. This process~was described .in the independent peer review (Wilson-201 3). The 1-D (or the 2-0) model cannot be independently calibrated and validated as the modeled case is hypothetical and has an extremely low probability of occurring.

Additionally, since the Jocassee and Keowee reservoirs were impounded in the early 1970s, Jocassee Damn has never discharged water through the gated spillway and Keowee has rarely operated the Spillway gates to discharge precipitation runoff. The maximum spillway discharge from Keowee occurred in August 1994 as 'a result of Tropical Storm Beryl. The total discharge from the spil!way was approximately 54,000 cfs which was contained within the downstream riverbanks.

The 2-0 software has been validated against several test cases including cases with unsteady and complex hydraulic phenomena. The inputs into the current 2-0 model do not fall outside of the acceptable range, and the grid is considered fine enough ensuring losses are calculated as accurately as the model will allow. The 2-0 model was used to inform user-defined coefficients within the 1-D mOdel in areas that the 1-0 model assumption "is stretched". This was done in areas with rapidly expanding or contracting flow. 'The 2-0 model does not have user-defined expansion or contraction coefficients that are included in the calculation of :losses like the 1*-0 model does. The 2-0 equations rely more on solving flow physics and less on, user-defined coefficients. As the model development process unfolded, the 1-D model was brought into alignment with the 2-0 model at the boundaries by better informing user-defined coeffcients in areas where the 1-0 model assumption was ustretched".

The flood wave associated with the fair-weather piping breach failure of Jocassee Dam has four potential routes to the Yard* Three of the routes are associated with high water surface elevations within Lake Keowee, while the fourth route is associated with the Keowee River below Keowee Dam. The first route is located at the ONS discharge outlet that adloins the right

.abutment of Keowee Dam. Water could flow over the parking lot adjoining the ONS discharge structure and through the road leading to the Yard. Thle second route is associated with the 43

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section'2- Flooding Hazard Reevaluation Energy Center swalethat adjoins the SC 130 bridge crossing of the Keowee Lake-Connecting Canal. The third route is associated with the overtopping of the north face of the ONS intake Canal Dike. The fourth route is, associated with the rising Keowee Dam-Keowee River tailwater due to a cascading failure of Keowee Dam.

Unsteady flow requirements in HEC-RAS require detailed inputs that describe boundary conditions and initial conditions at the first upstream cross-sections and ONS 1-D Model endpoint. In addition, boundary conditions are presented at internalmodel locations such as spillway'gates and hydroelectric turbines at the respective *hydrauliC.structures.

The initial base flow conditions Utilized for the ONS 1-D Model are presented in Table D-1.

These flows were developed to represent a "normal" (i.e., non-flood) condition that would allow the ONS 1-0 Model to maintain normal conditi~ons at each of the modeled facilities, prior tO the routing Of the dam failure hydrographs.

Total spillway discharge capacities for the Jocassee and Keowee Developments are provided in the FERC Supporting Technical Information documents. The total spillway discharge capacity of the Hartwell Dam was provided by the USACE. The family of curves for various gate openings and heads utilized in the model were developed by HDR based on the provided total spillway discharge capacities for each development. The discharge from all hydro units utilizes rating curves at the respective developments with open/close operation based on reservoir elevation settings.

The gated Spillway at Keowee Dam uses reservoir elevation settings for their respective operations. The spillway gates at Jocassee Dam were included in the model setup; however, operation is not simulated during the ONS external flooding event (breach of JocaSsee Main Dam) and the gates remain clOSed.

Jocassee Development ONS 1-D Model Operating rules for the two spillway Tainter gates are included in the model based on:

• Open gates at a rate of 1 minute per foot; all gates fully opened in approximately 33 minutes.

• Maximum gate opening of 33 ft.

Note: The gate operating rules were not initiated for the simulation of the fair-weather Jocassee* Main Dam breach.

Keowee Development ONS 1-D Model The four spiliway Tainter gates are operated based on:

• Flood operating procedure (non-preemptive operations) - initiate gate operations once the reservoir elevation reaches an elevation of 801 ftrnsl.

• During Jocassee Development failure scenarios, gates start opening as soon as a change in reservoir elevation at the Keowee Dam is detected.

• Open gates at a rate of 45 seconds per foot; all gates fully opened in approximately 23 minutes, if needed.

44¸

S. t .. ... .. ,..............  ; . i..... r.,

1,,,,,,

• ,,,;,* .. 1; 77 ;.3rR ,*

• Duke Energy. Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2-* Flooding Hazard Reevaluation

• Maximum gate opening of 35 ft.

Hartwell Development ONS 1-D Model The 12 spillway Tainter gates are' operated based on:

• Flood operating procedure (non-preemptive operations) - initiate gate oPerations once the reservoir elevation reaches an elevation of '665 ft ins!,

• During preemptive gate operations, gates begin opening 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> after breach of the dam

• of interest is initiated.

• Open gates at a rateof 10 minutes per foot; all gates fully opened in approximately 6.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />, if needed.

• Maximum gate opening of 35.5 ft.

The noted starting reservoir elevations at the Duke Energy and USACE developments (Jocassee, Keowee, and Hartwell) correspond to each facility's respective normal maximum pool elevation, and are provided in Table D-2.

Based on prior Jocassee-Keowee Dam breach modeling, the hypothetical fair-weather failure, of Jocassee Dam produces a flood wave that creates conditions for a cascading failure of Keowee Dam and, potentially, the cascading failures of the ONS Intake Canal Dike and Little River Dam.

Model Breach Parameters The breach geometry and reservoir storage volume along with the type of breach failure are the

• primary parameters used to determine time-to-failure and peak breach discharge .values from typical embankment dam~breach r'egression methodologies. For the reevaluation of potential inundation at~the ONS due to' dam failure, the assumed initiating mechanism is the uncontrolled .

internal piping of embankmentrmaterials. Overtopping as the result ofa basin-specific precipitation event is not considered credible due to the design .of the structures and the available hydrologic and hydraulic-information referenced in section-2.2.

Section '7 of the NRC's Interim Dam failure Staff GUidance (JLD-ISG-1 300-01) document references' several regression methodologies that have been developed to estimate embankment dam breach parameters' and potential peak discharge from the developed breach. i The regression methodologies investigated to support-the development of breach parameters in analyzing the downstream impacts at the Duke Energy ONS include: Froehlich 1995a, 1995b, 2008; Waider and O'Connor 1997; MacDonald and Langridge-Monopolis 1984 and Von Thun

• and Gillette 1990.

Dam breach' parameters are required-at a given dam structure to simulate flood routing from the reservOir into the downstream reach beloW the structure. The contributing outflow from the breached structure is typically called the breach outflow hydrograph. The breach parameters for an earthen structure are typically defined by a trapezoidal~channel having a prescribed bottom base width and side slope. The breach development phase is'different~ for piping failures and overtopping failures. Piping failures are defined: by the establishment of flow patterns through embankments, abutments, or foundations that create an orifice channel that continues to collapse until the final crest portion of the embankment section collapses and forms .an open channel. Overtopping'failures considered in the flood hazard reevaluation are attributed to 45*

Duke Energy Carolinas, LLC IONS FLOODING HAZARD REEVALUATION REPORT

-. Section 2 - Flooding Hazard Reevaluation embankment crest, slope, or embankment toe Scouring due to high overtopping velocities based on a cascading darn failure scenario where the downstream dam is overtopped from an upstream dam failure. Change in embankment slopes (intermediate bench or abrupt changes at the toe of the slope) create ideal locations for relative energy dissipation that can result in scouring. Scouring leads to the phenomena called embankment head cutting. Sustained overtopping from rapidly rdsing reservoir levels cap lead to further UPstream embankment head-Cutting migration until the crest and upstream embankment Slope are compromised and~an open channel develops. Outflow from the overtopping of the dam continues to increase rapidly during the breach development phase. The breach development time ends When the breach reaches its final size, Breach parameter development for the Jocassee Dam andKeowee Dam structureS followed two different patterns due to dam design and construction. The zoned rock-fill construction and size of the Jocassee Dam is :significantly different than the dams that were,,.used to develop the referenced dam failure regression equations. The Keowee Dam. is a well-constructed, homogenous earth-fill dam that is more representative of the type of'dam referenced in the development of the referenced regreSsion equations. Duke Energy Performed dam breach parameter research specific for application at the Jocassee Dam and Keowee Dam and commissioned independent assessments by dam engineering professionals to provide a professional opinion on the application of the PrOPOsed breach parameters for Jocassee and Keowee Dams. Due to the complex nature of these breach analyses, several modeling methods (regression, physical, and hydraulic) were considered to arrive at a conservative, yet plausible, solution. The empirical methods used were compared to actual dam breaches that fall within the range of each dam's physical characteristics. Where needed, parameter constraints were applied to the chosen regression parameters in Order to arrive at a reasonable estimate of the breach characteristics.. Considering regression and physical analyseS methodologies, plausible and conservative values for breach width and formation time were chosen..These

/

Values were used in HEC-RAS to develop the outflowhydrograph from the dams. Multiple breach patterns were analyzed using HEC-RAS in order to determine the sensitivity of the breach Outflow hYdrograph to the various breach parameters.

Keowee Development Breach Parameters Duke Energy provided the Keowee Dam breach parameters to Ehasz and Bowles (2013) for a professional assessment and opinion on their applicability at Keowee Dam. The modeled breach parameters and failure time were determined by Ehasz and Bowles (2013) to be, "realistic but conservative" for the cascading failure of Keowee Dam., An additional review of the Keowee Development breach parameters was conducted by an independent FERC board of consultants (BOC) (three geotechnical engineering experts and one hydraulics expert) to review geotechnical, hydraulic aspectsof breach consequences, and flood mitigation modifications at the ONS site. The BOO concluded the dam breach parameters Were conservative but realistic for the overtopping event. The modeled Keowee breach parameters are presented in Table D-3.

The Keowee Dam is approximately 3,500 ft long land is comprised of the Main Dam and West Saddle Dam '(WSD). The WSD is approximately,2,100 ft long and varies in height from 20 ft to t55 ft with a minimum creSt elevation of 815 ft msl, The WSD is exposed to the Same 48

Duke Energy Carolinas, LLCI ONS FLOODING HAZARD REEVALUATION REPORT Secton 2- Flooding Hazard Reevaluation overtopping characteristics and failure potential as the Keowee Main Dam. HEC-RAS is limited to a single dam breach designation. Therefore, to overcome this model limitation, a breach of the WSD was simulated using a 1,880-foot by 20-foot crest gate in the Joc~assee-Keowee Dam Breach Model. The ONS 1-D Model is assigned a full crest gate opening time frame of 30 minutes to simulate the WSD breach development. The crest gate does not begin to open until the reservoir elevation exceeds 817 ft msl.

The default sinusoidal wave breach progression provided in HEC-RAS, Figure D-l, was determined to be reasonable by Ehasz and Bowles (2013) and approved for application at the three Keowee Development dam structures by Duke Energy.

The time varying topography feature of TUFLOW FV was used to implement the breaches of the Keowee Dam, West Saddle Dam, ONS intake dike, and the Little River Dam. A sinusoidal breach progression which was Consistent with the HEC-RAS analysis was defined for the breach geometry in Table D-3. The bottom width, bottom elevation, and the geometry of the right and left side slopes are defined every 5 percent of the breach time fraction (resulting in 21 inflection points on the sinusoidal breach curve). Using the breach progression curve, a time series of modeled elevations for each cell within the breach was then defined. The breach event is triggered when the water surface at a designated point for each embankment is exceeded.

Then a new bed elevation is calculated by the model and updated. A linear interpolation is performed between the 21 points of the sinusoidal breach curve.

JocasseeDamn Breach Parameters The breach parameter selection process for JocaSsee Darn is outlined in detail in AMEC's Jocassee Dam Breach Parameter Analysis report dated December 2014 (AMEC, 2014a) which developed a refined set of dam breach parameters applicable to Jocassee Darn using guidance provided in the JLD-ISG-201 3-01. A summary of the breach parameter development process from the AMEC analysis states the following:

The Von Thun and Gillette breach equations were found to be most suited for the breach parameter analysis Of Jocassee Dam based on a linear trendline of average breach widths vs. reservoir Volume.

Comparing the HEC-RAS Sensitivity analYsis to the empirical breach flow predictors, direct use of the dam breach parameters produced by the Von Thun and Gillette equations resulted in an over-prediction of peak outflow leaving Jocassee Dam during a failure. The sensitivity analysis and~research shows that there are several justifiable ways to adjust the peak outflow produced by HEC-RAS for the particular breach event. The key to a realistic peak outflow lies in the breach progressiOn pattern used in the ONS I1-D Model.

The breach progression pattern developed uses the NWS BREACH progression pattern as a base with different aspects of the pattern modified to account for the physical characteristics of Jocassee Dam and Reservoir. Figure D-3 presents the original NWS BREACH progression and the modified progression.

47t

• ,,,d,, usauy oe,,siu.ve Inrormau:on; wimnoid from pulblic per 10 CFR 2.390(d)(1)  :'

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation The breach initiation phase was not altered from the NWS BREACH model because this is a subjective area in the development. For a large volume reservoir such as Jocassee the pool level will not drop significantly duiqthe breach initiation phase. The breach development time was extended tID()F ZZbased on the Von Thun anr4Gilleitt* -*uations for failure time. The time to empty reservoir was extend]ed [o to account for the continuation of breach growth until the reservoir is fully drained. The rate of breach formation after the development time is generally proportional to the depth of water in the reservoir at each time step in relation to the final pool level.

Comparing the sensitivity runs to the outflows a breach formed to nearly 100 percent during the development phase over-predicts outflows by as much as 200 percent. Considering the size and low erodibility of, n..--Dam, the breach

b)(,)(F) .. .................... progression~owas-adjusted-.os o p re tat th*rI mark (after the breach initiation and development phases). As a result of sntvianalysis on breach development time, the breach development was limited tclijpercent-...................

during the formation which allows the peak outflow to align with historical data.

Because the breach is assumed to occur against one of the abutments (based on geologic features) it was assumed that the breach growth would be limited by the abutment in one direction. As previously mentioned, in the case of Teton Dam where the breach occurred against the abutment, the chosen regression method over predicted the breach widt*1b* a minimum of 152 percent. For this analysis

t)(7)(F) ........-..... the breach widthi-was-~imited-tcJLlpercent of the total width predicted by the Von Thun and Gillette equations which is considered conservative in this case.

The final Jocassee Dam breach parameters utilized in the ONS 1-0 Model are presented in Table 7.

Table 7. Final Jocassee Dam Breach Parameters Final JOcassee barn Breach Parameters Top Width 959 ft Bottom Width 634 ft Side Slopes 0.5H: 1V Final Bottom Elevation 800 ft Breach Development Time to Peak Outflow io)(7)IF)

Full Formation Time Time to Empty Reservoir Computed Peak Outflow at Jocassee Dam _________

The breach progression for the Jocassee Dam required the development of a site-specific modified sine wave. Table 8 provides the recommended breach progression ordinates for the ONS 1-D Model.

48

sta,,,,*;,.Gm.u, iy .t1,bIL~v* ;IiUIIIldLI~II, wiI.hIOuu 11,01,, puu~II pe ivAU r*I L.3ouuJ1)* I) ""

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Bll 1 Section 2 - Flooding Hazard Reevaluation Table 8. Final Jocassee Dam Breach Formation Pattern (HEC-RAS Input)

Time Time* Breach Time; Time, Breach (h) Fraction Fractionl JJhr Fraction 'Fraction-(b)(7)(F) 0.0000 0.000o b(7(F 0.2236 0.7964 0.0028 0.0032 0.2486 0.81 33 0.0047 0.0063 0.2736 0.8289 0.0069 0.0102 0.2986 0.8433 0.0091 0.0144 0.3236 0.8568 0.01 16 0.0194 0.3486 0.8695 0.0139 0.0243 0.3736 0.8814 0.1234 0.7000 0.6036 0.9958 0.1256 0.7077 0.7599 0.9985 0.1486 0.7343 0.8816 0.9996 0.1736 0.7574 0.9536 0.9999 0.1986 0.7779 1.0000 1.0000 The Jocassee Dam stage-discharge hydrographs, breach discharge, and modified sine wave breach progression are shown in Figure D-2. The detailed site-specific, modified breach progression and breach discharge are shown in Figure D-3.

2-D Modeling The 2-0 model domain includes the area immediately surrounding the station including the Little River arm and the Little River Dam, shown in Figure 0-4. The results of the ONS 1-D Model analysis were extracted and utilized as boundary conditions for the 2-0 analysis. Version 2014.01.007 of the computational hydraulic solver TUFLOW-FV was utilized to define the boundary conditions for the 2-D analysis based on the results of the ONS 1-0 Model analysis (BMT WBM 2014). TUFLOW FV solves the Non-linear Shallow Water Equations (NLSWE), also known as the St. Venant equations, using a finite-volume numerical scheme. The code can process wetting and drying of mesh elements, steady and unsteady flows, and sub-critical and super-critical flows. It also includes hydraulic structures, time-varying bed elevations, and logic controls. It is parallelized to take advantage of running simulations on multiple central processing unit (CPU) cores. User-defined logic controls allow for both time-varying bed elevations (i.e., breach development) and breach initiation to occur within the model.

An explicit time discretization is used and an appropriate timestep is determined dynamically based on the Courant-Fredrichs-Lewy (CFL) condition, commonly known as the Courant number. A second-order, limited central-difference scheme was used for the spatial discretization for this analysis. Additionally, the code allows for a flexible, or "hybrid,"

computational mesh, which may include both triangles and quadrilaterals. TUFLOW FV has undergone extensive validation and benchmarking (BMT WBM 2013, N6elz and Pender 2013).

The NLSWE are a system of equations describing the conservation of momentum and mass in an incompressible fluid under the hydrostatic pressure and Boussinesq assumptions. The standard form of the NLSWE is:

49

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation au

-+ v.F(u) = s(U) where F(U) represents the flux terms in the x and y directions and S(U) represents the source terms such as the gravitational force-and bed shear stress (BMT WBM 2013). The x and y components of the inviscid flux (F') and viscous flux (F'> terms in the NLSWE are:

hv [ -hK. o*

where Kv is the horizontal eddy-viscosity term (BMT WBM 2013)., For this analysis, thte Srnagorinsky formula is used to calculate the eddy-viscosity. The Smagorinsky formula is:

Ky .2 ,()2 where C5 is the Smagordnsky coefficient and I! is the Smagorinsky length scale, which is related to the local mesh size (BMT WBM 2013).

Boundary condition locations and types are shown in Figure D-4. A flow hydrograph taken from the Keowee River 3-38889 HEC-RAS cross-seCtion .was used ?for the 2-0 upstream boundary condition. The boundary condition is approximately 1.6 miles uPstream of, the Keowee Dam in Lake Keowee. A stage .hydrograph taken from the Keowee River 6-23089 HEC-RAS cross-section was used for the Keowee River at the downstream boundary of the 2-0 model which is approximately 6.6 miles downstream of the Keowee .Dam..The Keowee powerhouse and Spiliway discharges were imposed as internal boundary conditions to ensure consistency with the 0NS 1-D Model which simulates gate flow and weir flow directly. Figure 0-5 shows the flow

.hydrographs for the Upstream boundary and the internal powerhousetspillway boundary.

Additional detail pertaining to the development of the internal Keowee powerhouse and spillway discharge boundary condition is provided below. Stage hydrographs for the downstream

.boundary are shown in Figure D-6.

The 2-D analysis was performed to add detail to the 1-D HEC-RAS analysis and simulate potential :inundation in critical .areas identified, around ONS. The upstream flow boundary for the 2-0 model was based on review of the 1-D Jocassee Darn breach flows routed from Jocassee Dam to the Keowee Dam through the Keowee Arm of the Keowee Reservoir. The selection of the boundary, approximately 1.6 river miles upstream of the Keowee, was determined by examination of the I1-D model output and through testing using the 2-D model and comparison

• with 1-0 results. Using an uncoupled I1-D model to inform both the upstream and downstreamr boundaries of the 2-0 model requires the results of the 2-D model at the boundaries to be.

compared with the 1-D model .and differences resolved. Grid refinement increases the ability of the. model to capture velocity gradients in the flow, and the grid Size was decreased to the level that was required to capture velocity gradients. The inflow boundary location Was selected 50O

i.vn~ia~lmi* *euruiy .*¢IlbuI;VU ;uu~nu..;u,.l, w;u;,;uuk; stu., pill.pu,, u *,ari *.QavulI m.= p)"

Duke Energy Carolinas, LL.C !ONS FLOODING HAZARD REEVALUATION REPORT' Section 2 - Floodtng Hazard Reevaluation based on a location that was far enough upstream to have a fully developed velocity distribution in the flow as it approached the Keowee Dam. The boundary lOcations and the grid size were also selected based on balancing computational detail and reporting multiple cases within a reasonable timeframe.

Prior to beginning the unsteady simulation, the 2-0 model was run to a steady-state initial condition consisting of a Keowee reservoir level of .800 ft msl and a discharge of 1,650 cfs flowing from the. Keowee Upstream inflow boundary to~the spillway outflow boundary. Below the dam, an inflow boundary, of 1,650 cfs at the sp~llway and a downstream water surface elevation of 660 ft msl at the downstream boundary, were used. The 800 ft msl represents the full reservoir elevation of Lake Keowee and the 660 ft msl represents the full normal reservoir*

elevation of Hartwell Lake downstream of Lake Keowee..

Time accurate conveyance of flow through Keowee Dam is important for the determination-of proper headwater and. tailwater elevations prior to initiation of the dam breach,. The 2-D model is not able to explicitly model the 1-0 flows through the ,powerhouse or sPillway gates in the same way as .the ONS 1-D Model. As a result, a single discharge boundary was used to represent the powerhouse flows, gated spillway flows, and the spillway overtopping flows (thatoccur when water elevations in the reservoir are generally higher than 815 ft msl). The initial overtopping elevation at Keowee Dam is 815 ft msl. The powerhouse and gated spillway flows were determined from the time series data output by the ONS 1-D Model and the ,spillway overtopping flows were calculated with the weir equation. Spillway flows in the 2-D model did not include the ",gate flow"~ used to represent flows through the West Saddle Dam (WSD) breach in the ONS 1-0 Model. These flows were represented in the 2-D model using the same breaching method as the main Keowee Dam breach.

Flows through the spillway and .powerhouse were modeled by removing mesh cells from the.

model domain at the outflow structure-location and assigning a subcritical outflow and inflow boundary-condition upstream and downstream of the outflow structure. The flow hydrograPh for the Keowee Spillway boundary is included on Figure D-5. The TUFLOW FV software supports a continuous update of the dam topography for simulating the breach of the Keowee Dam. This enhanced code feature facilitated a close breach formation simulation relationship between the 2-0 and 1-D models that resulted ingood agreement of the dam breach hYdrographs between the two models. .

in order to improve model stability near the boundaries representing the spillway 'flow, the elevation of nodes approaching and adjoining the spillway were manually adjusted. Upstream of the spillwvay, mesh elevations, were lowered to 745 ft msl to increase the' cross-sectional area for the outflow boundary ,upstream of the spillway. This prevented the flow upstream of the spillway from becoming supercritical. Similarly, mesh node elevations immediately downstream of the inflow boundary representing flow entering the tailrace were adjusted to prevent, supercritical flOW near the model boundary. These modifications had no effect on conditions near ONS.

The channel bathymetry and land topography assigned to the computational mesh were generated Using a DEM created from digital data sources Using Geographic Information Systems (GIS) software. Topographic data is from the USGS-NED and site survey data of the 51

• *oniarns oe~uriiy penszilve .nrormanion; wirnowo from puW~iG per 1u Id-K z.isuua,ll)j' "

Duke Energy carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazai'd Reevialuation ONS vicinity collected in 2010. Bathymetric data is from pre-impoundment 1 :24,000 contours for Hartwell Lake and surveyed data collected, by Duke Energy in 2010 for Lake Keowee. Updated bathymetric-survey information of the ONS intake canal gathered in 201.3 was used where available. Topographic breaklines were enforced for the Keowee Dam, West Saddle Dam, ONS Intake Dike, and the Little River Dam. Additional breaklines were enforced for roadway embankmnents and bridge approaches in the vicinity Of ONS.

In the 2-0 model, Manning's roughness coefficients (Manning's n-values) were used to represent the frictional resistance and energy loss in the flowing water. For this analysis, to the extent possible, the Manning~s n-values match the companion ONS 1-D Model, where Manning's n-values were selected for the river cross-sections based on a review of topographic maps and reference materials in Chow's "Open-Channel Hydraulics" (Chow 1988). Additional sensitivity and assessment of Manning's n-values was performed and documented during the ONS 1-0 Model development. Manning's n-values were primarily limited to two designations in the model - floodplain and overbank (n = 0.08) and main channel (n = 0.025). The riprap installed for erosion protection along-the east side of the ONS yard was designated with a Manning's n-value of 0.04. Figure D-7 depiCts the spatial extent of the Manning's n-values.

Both the 1-0 and 2-D models were improved by utilizing appropriate results from one another thus bringing the models into close alignment. A series of model runs was performed to test performance between the models; and adjustments were made that produced excellent comparison and agreement and a high degree of confidence for using both models in this evaluation. This is specifically true for the 2-0 model in areas of flow complexity and site-

• specific details, such as the depth of inundation in and around QNS.

2.3.4 Water Level at the plant Site A comparison of model output results is presented in Table 9. The comparison details the high level of agreement between the 1-0 and 2%D models as well as the overall hydraulic impact on ONS resulting from a poStulated fair-weather piping breach failure of Jocassee Dam coupled

-with the cascading failure at Keowee Dam.

As can be inferred from Table 9, the comparison of the 2-0 model results with the ONS 1-D SModel results indicates that the timing of breach initiation (K~ 7 (F oir reaching 817 ft msl) occurs slightly earlier in the 2-D model

• wee DamJ Jafter the Jocassee Dam breach, which can be compared to hou~i"' Ifor the ONS 1-D Model. The maximum water

• RF~nr.-*Ilv~tinn in I *k*..Keowee peaks neal(x() ,

-in the 2-D model and (b)(7XF /in the oNs 1-0 Model. The ONS intake dike is not breached during The analysis as the intake canal water surface does, not overtop the structure. The maximum

.water surfac eeainithitkeanal reache (fl)(7)(F) Iin the 2-0 model compared td(b)(7)(F La= QL*~ Model. The maximum watersud, ...... r.

" , bl(7(FI " i(b)(7)(F) i

• to in 'the Keowee tailrace area iq"... un the 2-D model :compared tq, J L*~n the ONS 1-0 Model.' While this differece ofT. feet~is greater than th~e difference at other points, it is still a reasonable comparison considering the complex hydraulics;of the Keowee River south of ONS, location where the 1-0 and 2-D elevations were measured and 52

• .oniawns OuuuIILy *UfIlbIUVU Illlaulnlmluull, WiueuuluU ui.. Vuul;n,;aa 1 .ZII.08vti

,,

• 1)

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation more specif ically the complex hydraulics that exist as the river channel meanders through the oxbow area known as Old Picken's Knob just east of the SCi 83 highway bridge.

Dam breach inundation levels reached elevatior(b7')F) Iresulting in approximnately~F)!!F of inundation above the nominal Yard elevation of 796 ft around the SSF.

The comparison indicates that between the ONS 1-D Model results and the 2-D model results there are no significant differences in the model results in terms of maximum stage reached and timing in the reservoir. While the models do not produce the Same results, the trends in the results between the models are generally, preserved. This agreement provides confidence in the results of the simulations.

Figures D-8A through D-10B illustrate the general flow patterns, depthadvlct~iesiaur

' A.*~r~o

• H .- f~ ,,,f~e elevations (Figure B-Series) at time (t) =1brz)(F) . I, I~bX!)(! Iforthe modeled case over the 2-D model domain. The time (t) =1 ()(7)(F)

• b()()occurs nhasher breach is *rig~i Keowee Reservoir reaches the peak elevation just upstream of Keowee Dam, (t) * .. Icusas the peak tailwater and ONS Yard inundation reaches the peak depth and (t) =1b()F Ioccurs after the breach and the peak floWS have been conveyed through out the Study area and peak water surface elevations have receded.

The purpose of the 2-D model analysis was to provide estimates for the water depth and flow, velocity in the vicinity of ONS resulting from a postulated fair-weather failure at Jocassee Dam.

Figure D-1 1 shows the location of monitoring locations that provide the basis for Figures D-12 through D-16, which show the water depths and elevations upstream of Keowee Dam, Keowe&

Dam tailwater, Visitor's Center Swale wall, ONS Intake Canal Dike, ~and the SSF, respectively.

The Visitor's Center Swale wall is actually located Closer to the reservoir, since this area :is dry most of the time. The Yard. is potentially exposed to four inundation paths. The primary source as modeled is the Keowee Dam tailwater, while the Visitor's Center Swale wall, ONS discharge outlet and ONS Intake Canal Dike (because of their proximity to the Yard) could serve as secondary sources and are dependent on maximum post-Jocassee Dam breach reservoir elevations.

Flooding at the ONS site is primarily caused b ib)(7X(F) Asa result* the predominant flow direction during initial inundation of the si (bx)(7F)

I()7

)(F) *When water surface elevations in Lake Koepaknear--~*in the simulation, a small flood wave resulting.. in water

... depths of approximatel j 1b(7(F) *,vertops-the West Saddle Dam (ONS discharge outlet path) and flows around the north side of the Administration building.

The flood wave from this path does not reach the SSF. Likewise, there is no Yard inundation from the Visitor Center Swale wall and the ONS intake Canal Dike pathS because Keowee reservoir elevations do not overtop these structures.

53

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT o

I-Y~

Section 2 - Flooding Hazard Reevaluation Table 9. Comparison of 1-D and 2-D model results - time to breach and maximum water surface elevations (Elevation in ft msl and Time in hours)

(b)(7)(F)

• The ONS Intake Canal Dike crest is nominally at an elevation of 815 ft msl but has a wall built along the North End to provide protection from flood waters entering the Yard to an elevation of 824 ft msl. Figure 0-14 and Table 9 indicate that the peak water surface elevation in the ONS Intake Canal Dike reacheL7°('F) imsl with a dike crest elevation oLEmsl.

The Visitor's Center Swale wall is built to an elevation of 832 ft msl. Figure D-1 5 and Table 9 indicate that the maximum water surface elevation reached at the Visitor's Center Swale wall is 817.2 ft msl.

The nominal elevation of the Yard is 796 ft msl. The Yard would be exposed to flooding via downstream backwater from Keowee Dam (T0wtr.F~ue -13 and D-16 indicate that the Yard would be approximately inundated to elevatior b) 7 )(F),iduring the modeled dam failure.

54

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation 2.4 Storm Surge and Seiche The ONS site is located on an inland reservoir in the northwestern corner of South Carolina and is not subjected to storm surge or seiche flooding. St*orm surge and seiche flooding have been reviewed in the FERC-required evaluation of the Keowee and Jocassee Developments and are not considered credible events to produce maximum water levels at the site.

A seiche caused by an earthquake or landslide is not considered credible based on the topography and geology around the reservoirs.

Storm surge events are not expected to affect the site. The maximum wave height and wave run-up have been calculated for Lake Keowee and Lake Jocassee using USBR 1981. The

)

results of wind-driven wave run-up using USBR Wind Velocity charts are as follows in Table 10:

Table 10. Wind-driven wave run-up results Location' Wave Height ,WaVe'Run-Up* Maximum 'Fetch Keowee Main Dam - Fair Weather 6.10 ft. 5.84 ft. 2.37 miles Keowee Main Damn - PMF 3.60 ft. 3.73 ft. 2.37 miles Jocassee Main Dam - Fair Weather 6.48 ft. 7.09 ft. 3.04 miles Jocassee Main Dam - PMF 3.80 ft. 4.47 ft. 3.04 miles The results using ANS 2.8, 2-Yr Wind Velocity are as follows in Table 11:

Table 11. 2-year wind velocity results Loc~ation .. Wave Height. .WaveRun-Up Maximum Fetch Keowee Main Damn - Fair Weather 5.33 ft. 5.11 ft. 2.37 miles Keowee Main Damn - PMF 3.20 ft. 3.34 ft. 2.37 miles Jocassee Main Dam - Fair Weather 6.10 ft. 6.66 ft. 3.04 miles Jocassee Main Dam - PMF 3.80 ft. 4.43 ft. 3.04 miles The wave height and wave run-up figures are vertical measurements above full pond or PMF elevations as tabulated above. The design normal freeboard at each damn is 15 ft which is adequate to prevent overtopping of wind-driven waves at both dam sites.

The Keowee reservoir has not exceeded an elevation of 800.0 ft msl and the Jocassee reservoir has not exceeded an elevation of 1,110 ft msl since construction was completed (approximately 40 years).

During combined events of a PMF and wind-driven waves, because of the time-lag characteristics of the runoff hydrograph after a storm, it is not considered credible that the maximum reservoir elevation due to PMF would occur simultaneously with winds causing maximum wave heights and run-ups. Refer to Section 2.8 for an additional discussion of combined effects of a wind-driven wave impact combined with a PMF.

55

Duke Energy Carolinas, LLC ONS FLOODING HAZARD REEVALUATION REPORT Section 2 - Flooding Hazard Reevaluation 2.5 Tsunami Since the ONS site is not located on an open ocean coast or large body of water, tsunami-induced flooding will not produce the maximum water level at the site.

2.6 Ice-Induced Flooding Since the ONS site is not located in an area of the U.S. subjected to periods of extreme cold weather that have been reported to produce surface water ice formations, ice-induced flooding will not produce a credible maximum water level at the site and is not considered a realistic external flooding hazard to ONS.

2.6.1 Ice Effects Long-term air temperature records (1951-2011 ) available at the South Carolina State Climatology Office were reviewed to assess historical extreme air temperature variations at the ONS site. The analysis was also supported by onsite temperature data measured at the ONS site.

The climate at the ONS site is characterized by short, mild winters and long, humid summers.

Local climatology data for Pickens County, South Carolina, for a period of 30 years show a mean air temperature of 59.7O Fahrenheit.

There has not been a recorded event of significant surface ice formation on Lake Jocassee or Lake Keowee since filling in the early 1970's.

2.6.2 Ice Jam Events There are no recorded ice jam events in the upper reach of the Savannah River based on a search of the USACE's Ice Jam Database. Water temperatures in this area of the southeast U.S. consistently remain above freezing.

2.7 Channel Diversions Due to the location of ONS on the banks of Lake Keowee and the upstream topography of the reservoir, channel diversion is not a credible flooding event. The controlling external flooding event is upstream dam failure.

2.8 Combined Effects Section 9 of ANS 2.8 outlines general criteria to be reviewed for addressing combined flood-causing events. Lake Jocassee and Lake Keowee are man-made impoundments located in northwestern South Carolina protected from coastal events as well as extreme cold weather events. Both reservoirs are considered "enclosed bodies of water" as defined in ANS 2.8 Section 7.3.3. Based on this definition and guidelines noted in ANS 2.8 Section 9.2.3.2 for the streamside location of ONS, Alternative II was determined to be the most limiting case and was used for reviewing the possible combinations producing maximum flood levels. This includes:

56

Duke Energy Carolinas, LLC f ONS FLOODING HAZARD REEVALUATiON REPORT Section 2 - Flooding Hazard Reevaluation

1. PMF.

2. 25-year surge or seiche with wind wave activity.

3. 100-year or maximum controlled level of water body, whichever is less.

Damn failure as defined above in Section 2.3.1 was also reviewed.

Combined flooding effects (PMP, PMF, dam failure and/or wind-driven waves) are concluded in Table 12. LIP was reviewed for the ONS site using a state-of-the-practice, coupled 1-D and 2-D model, including reviewing water impounding on roofs of buildings, with and without site drainage features. Due to the detailed 2-D analysis, variable water surfaces were developed for the entire Yard. The maximum elevation noted in Table 12 is for a single reporting point.

Flooding in upstream reservoirs was reviewed using modern computer programs for hydrology and hydraulics and the current HMR 51 rainfall. PMF flood elevations do not produce overtopping at Keowee or Jocassee. Dam failures were simulated for Jocassee and Keowee as detailed in Section 2.3. The failure of Jocassee Dam *with the cascading failure of Keowee Dam produced a flood inundation elevation ofb)(7)(") imsl. The darn*Lr breach inundation exceeds the Yard nominal elevation of 796 ~ft msl yapproximatel*j~j Storm surge and seiche-generated waves were not considered a credible event for evaluation of an ONS flood hazard. A combined event of the PMF plus wind-driven waves was reviewed for both Keowee and Jocassee. As shown in Table 12, Keowee maintains approximately 2.8 ft of freeboard when combining the peak PMF elevation with wind-driven waves. A PMF for Jocassee coupled with wind-driven waves (Section 2.4) would produce some lapping of waves over the 20-ft-wide crest of the dam; however, due to the large riprap that comprises the upstream and downstream shell of the Jocassee rock-filled dam, the breaking of waves over the crest of the dam is not expected to have any significant impact on slope stability.

Table 12. Combined effects flood elevations Reevaluation Combined Reevaluation Combined Effects Effects Flood Elevation Flood Causing Mechanism Flood Elevation Keowee ~ Jocasseel Local Intense Precipitation 800.39 ft msl (Note 1) N/A Run-up______p__us___PMF__ ___ 12.2______t_____l__1126.4 ft msl Tsunami _ __ __ _ __ _ _ __ _ _ N/A_ __ _ __ _N/A Ice-Induced Flooding _____ /A__________N/A Channel Diversion N/A N/A Notes: 1 location of recorded maximum Yard elevation is based on location in Figure B-l, West 13.

57

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 3 -Comparison of Current Licensing Basis and Reevaluated Flood-Causing Mechanisms Section 3 - Comparison of Current Licensing Basis and Reevaluated Flood-Causing Mechanisms Table 13 below summarizes the comparison of current licensing basis and reevaluated flood-causing mechanisms, which includes wind effect for flooding in reservoirs and dam failures.

The table outlines a comparison for current licensing basis and reevaluated flood-causing mechanisms for the ONS site including the dam failure mechanism that considers the postulated fair-weather piping breach failure of Jocassee Dam to be the enveloping dam failure event.

Table 13. Current licensing basis and reevaluation flood elevations

'Current Licensing, Reevaluation Reevaluation Basis Flood Flood *Flood Delta from Flood-Causing MechaniSm Elevation, Elevation Lice nsing:'Basils 800.39 ft msl Local Intense Precipitation 798.17 ft msl (Note 1) +2.22 ft.

Flooding in Reservoirs 808.0 ft msl 889ftml+.f.

Dam Failures (Note 3) N/A (Note 4) __Note____ 2)_______

Storm Surge and Seiche/Wind-Wave Run-up N/A N/A____N___A Tsunami N/A N/A___NiA _

Ice-Induced Flooding N/A N________/A _

Channel Diversion N/A N/A N/A Combined effects N/A 812.2 ft msl N/A Notes:

Location of recorded maximum Yard elevation is based on location in Figure B-i, West 13 2Location of recorded maximum inundation elevation is shown in Figure D-1 1, Keowee 1W 3 Enveloping dam failure scenario is fair-weather Jocassee Damn and cascading Keowee Damn failure.

SThere were no dam failures postulated inthe original licensing/design case basis of the plant. On Jan 28, 2011 a SE was issued descr'ibing a new flooding hazard for the site. This is a Beyond design basis event. See Section 1.3 for further discussion.

3.1 Local Intense Precipitation The current licensing basis case ONS flood elevation from the LIP is 798.17 ft msl (Section 1.2).

The Beyond design basis case reevaluated flood elevation is 800.39 ft mnsl, is 2.22 ft more than the current licensing basis flood elevation, and 3.89 ft above the elevation of safety-related SSCs of 796.5 ft msl. The elevation reported here is specific to the point near the Unit 1 Reactor Building (West 13). Because of the detailed analysis using a 2-D model, water elevations around the Yard vary significantly. Reference Table B-I for additional flood inundation elevations at selected points around the Yard.

58

Duke Energy Carolinas, LLC I ONS FLOODINGHAZARD REEVALUATION REPORT section.3 -Comparison of Current Ucensing Basis and Reevaluated Flood-Causi'ng Mechanisms-3.2 Probable Maximum Flooding There is no current licensing or design basis ONS flood elevation from probable maximum flooding in upstream reservoirs with Coincident wave run-Up.'

The reevaluated flood ,elevation is 808.9 ft msl or 812.2 ft ms! with coincidental wave run-up.

The PMF elevation increased by 0.9 ft from the current flood elevation and does-not impact safety-related SSCs.

3.3 Dam Failures There are no current licensing or design basis ONS dam failure flood elevations.

The reevaluated flood hazard elevation due to a combined Jocassee'and Keowee.amsflzaIilure assuming full reservoir elevations at the initiation of the Jocassee Dam breach i{j 7 ~x) msl.

This elevation is due to the initial failure by piping of the Jocassee Main Dam and thie Cascading failure of the Keowee Main Dam and West Saddle Dam due to overtopping. With the Yard at a wnlelevation of 796 ft msl, this inundation Would produceaa depth of water approximately LZ) round the SSF . . .. .....

3.4 Storm Surge and Seiche Storm surge and seiche are not expected to affect the site for the reasons listed above in Section 2.4.

3.5 Tsunami Tsunami-nduced flooding is not expected to affect the site for the reasons listed above in Section 2.5.

3.6 Ice-Induced Flooding Ice-induced flooding is not expected to affect the site for the reasons listed above in Section 2.6.

3.7 Channel Diversion Channel diversions are not expected to affect the site for the reasons listed above in Section 2.7.

3.8 Combined Effects Ther~e are no current licensing or design basis combined site flood hazard ;effects. Combined flooding effects are discussed in Section 2.8.

59

Duke Energy Carolinas, LLC I ONS FLOODINGHAZARD REEVALUATION REPORT

- Section 4 - Interim Evaluation and Actions Taken or Planned Section 4- Interim Evaluation and .Actions Taken  !

or Planned Based on the results listed above, only two of the flood-causing mechanisms produce flood waters that could potentially impact the site; the LiP rainfall event and the combined Jocassee and Keowee Dams failure.I I.

Actions have been taken for both the LIP event and the postulated fair-weather failure of the Jocassee Dam. ONS was in the process of analyzing both of these flood hazards prior to the release of the March 12, 2012, NRC 10 50.54(f) request for information.

The reevaluated flood elevation for an HMR 51/52 generated LIP event is 800.39 ft msl. This is' 2.22 ft more than the current licensing basis flood elevation, and 3.89 ft above the elevation of safety-related SSCso0f 796.5 ft msl.. The elevation repoi'ted here is specific to the point near the Unit 1 Reactor Building (West 13).

The reevaluated flood hazard elevation due to a combined Jocassee and Keowee Dams failure, assuming full reservoir elevations at the initiation of the Jocassee Dam breach. **(tF'Is in the Keowee tailrace and (b)7)( Imsl at the Standby Shutdown Facility (SSF.).

The 2.1 Haztard Reevaluation has shown that the Beyond design basis flooding hazards for LIP exceeds the current licensing basis flooding levels. LIP maximum flooding levels are below the SSF flood walls and current'floodirng response procedures adequately address the flooding caused by the new Beyond design basis event (BDBE). As such,-added interim actions are not  !

required to respond to the maximum LIP flooding levels.

The flooding levels caused by an upstream dam failure are within the BOBE as defined'by "Oconee's Response to Confirmatory Action Letter (CAL) 2-10-003" issued to the NRC on

'August 2, 2010 (ADAMS ML102170006). Compensatory actions that were completed in response to the June 22, 2010, Confirmatory Action Letter (CAL 2-10-003) will remain in place Until the CAL is super'seded by regulatory action related to the Fukushima response. The existing interim actions identified in CAL 2-.10;-003 are Sufficient in mitigating the effects of flooding as predicted in the FHRR.

'60¸

Duke Energy Carolinas, LLC I*ONS FLOODING H-AZARD REEVALUATION REPORT Section 5 -Additional Actions Section 5- Additional Actions Early in 2012, a question arose about the LIP analysis for the Yard, which Was entered into the ONS CAP. This led to a thorough investigation of the current status of various structures on site and their ability to resist flood waters from an LIP event. A walkdown was performed for all Class 1 structures and other structures deemed critical to safe shutdown. The walkdown surveyed the building envelope to identify weaknesses in flood protection. It was determined that ONS is in nonconformance with the Licensing Basis as described in UFSAR Section 3.4.1.1 because water may exceed a 6-inch sill in a few areas surrounding the Class 1 structures. All other areas surrounding the structures were determined to be acceptable. Interim actions were taken in the form of adding Flood Barrier Sandbags and Gryff01yn coverings. These barrier locations and heights are based on a preliminary analysis that was performed on local flood heights in the Yard. The LIP data discussed in Section 1.2.1 of this report finalizes and validates the concern from the preliminary analysis. The sandbag flood barriers continue to serve as an effective measure against the finalized current licensing basis LIP ev*ent.

ONS submitted its Flood Hazard Reevaluation. Report to the NRC on March 12, 2013, pUrsuant to the NRC's 10 CFR 50.54 (f) letter. By letter dated March 20, 2014, the NRC submitted a request for additional information regarding the FHRR consisting of 15 questions. Specifically, the last two questions were industry-wide requests. The revision and resubmission of the FHRR causes the need for ONS to revise the original response to PAI's 14 and 15. These responses are presented below:

5.1 RAI-14

Hazard input to the integrated assessment: Flood event duration parameters

Background:

The March 12, 2012, 50.54(f) letter, Enclosure 2, requests the licensee to perform an integrated assessment of the plant's response to the reevaluated hazard ifthe reevaluated flood hazard is not bounded by the current design basis. Flood scenario parameters from the flood hazard reevaluation serve as the input to the integrated assessment. To support efficient and effective evaluations under the integrated assessment, the NRC staff will review flood scenario parameters as part of the flood hazard reevaluation and document results of the review as part of the staff assessment of the flood hazard reevaluation. The licensee has provided reevaluated 'flood hazards at the site including LIP flooding, probable maximum flooding on contributing watershed, flooding in streams and rivers, and flooding from breach of dams. The LIP flooding is reported to exceed the current licensing basis and subsequently the

• licensee has committed to perform integrated assessment.

Request: The licensee is requested to provide the applicable flood event duration parameters (see definitiOn and Figure 6 of the Guidance for Performing an Integrated Assessment, JLDISG-201 2-05) associated with niechanisms that trigger an integrated assessment using the results of the flood hazard reevaluation. This includes (as applicable) the warning time the site will have to prepare for the event (e.g., the time between notification of an impending flood event and arrival 61

Duke Energy Carolinas, LLCI ONS FLOODING HAZARD REEVALUATION REPORT Section 5 - Additional Actions of floodwaters on site) and the period of time the site is inundated for the mechanisms that are not bounded by the current design basis. The licensee is also requested to provide the basis or source of information for the flood event duration, which may include a description of relevant forecasting methods (e.g., products from local, regional, or national weather forecasting centers) and/or timing information derived from the hazard analysis.

5.1.1 Response to RAl-14 RAI-14 is requesting flood event duration parameters including duration from the FHRR to define key inputs to the Integrated Assessment. ONS has two external flooding events that are included in the FHRR that will be addressed in the IA. Below are the requested parameters for each of the external flooding hazards:

Upstream Dam ,FailureFlood DurationParameters

1. The warning time the 'site will have topeare for the event - Tewrigtime for a Sunny Day failure is assumed to be[l )U*Fj Ibased on a minimumo*b()F time after detection, an (b()(7)(F) Ibetween a significant pipe collapse in the Jocassee Dam .(internal breach formation process) and downstream dam breach flood waters in the Keowee Tailrace reaching the elevation of nominal site grade. 796 ft msl. This assumption is based on early warning due to an aggressive monitoring program for the dam and historical piping failures. The dam's early warning and monitoring program includes seepage collection points that monitor flow volume and turbidity, regularly scheduled visual walk-down inspections, forebay and talirace electronic level alarms, and 24-hour video surveillance of the dam and adjoining abutments that enhance early detection of unusual conditions including possible piping through the dam.

2. The period of time the site Is inundated - The site grade at a nominal 796 ft msl is inundated by floodwaters from a fair-weather failure of the Jocassee dam for approximatel (b)(7)(F) The period of time where the flood impacts the site belOw nominal site grade as defined by the overtopping of the Keowee dam until floodwaters recede below the turbine building drain invert is aprxmtlF()F 3.. The basis -or source of information for the flood event duration - The information used to c*alculate the duration includes the hydrograph for the breach and the HEC-RAS model that routes the flood waters to the site and beyond.

4. Relevant forecasting methods - FERC and NRC requirements for upstream dam 1monitoring include video camera monitoring of the dam/abutments and at seepage leak off points, forebay and tailrace alarms, 24-hour staffing at Jocassee and at Hydro Central, inspections immediately after receiving 2 inches or more rainfall, inspections following any felt seismic event and weekly dam safety inspections. Monitoring and site notifications are included in Duke Energy Hydro and ONS procedureS.

Local Intense PrecipitationFlood DurationParameters

1. The warning time the site will have to prepare for the event - The major storms that have sufficient atmospheric moisture to deliver the 18.95 inches of rain can be reliably forecasted 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> in advance due to the size of the storms. The major, storm types that could contain sufficient moisture include Tropical Cyclones, Synoptic Storms, 62

Ouke Energy Carolinas, LLCj ONS FLOODING HAZARD REEVALUATION REPORT Section 5 -Additional Actions; and. Mesoscale Convective Complexes. Isolated thunderstorms, do not have the ability to approach LIP level rainfall events due to the location of ONS where no orographic lift is present.

2. T[he period of time the site is inundated - The site Will start developing water accumulation in and around the power block early in the assumed 72-hour duration rainfall event as modeled using a two-thirds loaded distribution. Water levels are projected to decline to nominal levels within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> after the rainfall has subsided.

3. ,The basis or "source of information for the flood event duration -- The 72-hour event was based on the event that Produced the most Conservative prec;ipitation for the site drainage basin and the most conservative temporal distribution based on the results of the runoff model.

4. Relevant forecasting methods - The forecasting for the LIP magnitude storms is based on using the National Weather Service (NWS) Quantitative Precipitation Forecasts (QPF's) for medium range forecast and Probabilistic Quantitative Precipitation Forecasts (PQPF's) for shOrt- range forecast. A monitoring threshold Would be set based on the appropriate level of extreme rainfall for the basin where the ONS is located. Six inches in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> is considered an extreme rainfall event With a return rate :of less than I in 1000 years. This monitoring threshold would be applied to the medium forecast prior to the event. If this threshold is met based on medium range QPF forecaStS (3 to 7 days prior to the event), the nuclear site would be notified by the fleet meteorologist (staffed 7 days a week) which would then initiate site monitoring once per shift as directed by site, procedure. A mitigation action trigger would be established u~sing .the 95th Percentile PQPF value of 6 inches in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 'n the short-range forecast (Day 1).

The action trigger is conservatively based on one-half the maximum historical 24-hour precipitation event. When this trigger is met, the site would initiate flooding .response actions 'including securing flood gates and doors. Using one-half of the 24-hour histOric precipitation event provides a conservative trigger that will be activated well in advance of a storm capable of delivering LiP rainfall of a more extreme value of 18.95 inches for a 1-hour/I-mi2* precipitation event.

The maximum historical precipitation event from 1973 to 2012 in North Carolina and Upstate South Carolina was 12.32 inches in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> occurred on August 27, 1995 (Tropical Storm Jerry) based on the highest recorded levels at Greenville, South Carolina: The highest 48-hour and 72-hour total for the same event was 14.47'inches. No consequential flooding resulted at the ONS'site during this event.

5.2 RAI 15: input to integrated assessment: Flood height and associated effects

Background:

The March 12, 2012, 50.54(f) letter, Enclosure 2, requests the licensee to perform an integrated assessment of the plant's response to the reevaluated hazard ifthe reevaluated flood hazard is not bounded by the current design basis.* Flood: scenario parameters from the flood hazard reevaluation serve as the.input to the integrated assessment. To support effcient and effective evaluations under the integrated assessment, the NRC staff will review flood 63

Duke Energy CarolinaS, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 5 -Adlditional Actions scenario parameters as part of the flood hazard reevaluation and document results of the review as part of the staff assessment of the flood hazard reevaluation. The licensee has provided reevaluated flood hazards at the site including LIP flooding, probable maximum flooding on contributing Watershed, flooding in streams and rivers, and flooding from breach of dams. The LIP flooding is reported to eXceed the current licensing basis and subsequently the licensee has committed to perform integrated assessment.

Request: The March 12, 2012, 50.54(f) letter, Enclosure 2, requests the licensee to perform an integrated assessment of the Plant,'s response to the reevaluated hazard ifthe reevaluated flood hazard is not bounded by the current design basis. The licensee is requested to provide the flood height and associated effects (as defined in Section 9 of JLD-ISG-2012.-05) that are not described in the flood hazar'd reevaluation report for mechanisms that trigger an Integrated Assessment. This includes the following quantified information for each flooding mechanism (as applicable):

• Hydrodynamic loading, including debris;

• .Effects caused by sediment deposition and erosion (e.g., flow velocities, scour):

• .Concurrent site conditions, including adverse weather; and

• Groundwater ingress.

5.2.1 Response to RAI 15 SThe mechanisms that exceeded the current licensing basis or design basis and triggered the integrated assessment at ONS are the LIP event and the sunny-day failure of Jocassee Dam and the cascading failure Of Keowee Dam.

Local Intense Precipitation Hydrodynamic Effects Section 2.1 and Appendix C of the FHRR provided detailed results (flood depths and durations of inundation) at various locations throughout the site. Flow velocities from the 2-D model were reviewed at relevant door openings to safety-related structures to determine whether' hydrodynamic loading iS of concern at any of the critical locations. The results indicate that maximum velocities are generally below i ft/sec, With occasional exceedance at locations where flow is constrained between twO buildings. Furthermore, the velocities reported by the model do not represent velocities at the maximum flood stage and the velocity'vectors are generally not orthogonal to the doors. Since hydrodynamic loads are a function of flow velocity and flood depth, these toads are expected to be minimal and well within the margin of safety provided for the respective flood protection features. ASCE/SEI 7-10 standard provides a recommended

'approach for estimation of dynamic effects of moving water with flow velocities below 10 ft/sec.

Based on this approach, dynamic effects of moving water can be converted into equivalent hydrostatic head by increasing the design flood elevation by an equivalent surcharge depth, dh, equal to dh =aV2/2*g, Where V = average velocity of water in fl/sec, g = acceleratio~n due to gravity, 32.2 ft/sec2 , a = coefficient of drag The ayerage maximum velocity at ONS critical str~uctures is 1.13 ft/sec. Per F EMA 259, the most ConServatiVe coefficient of drag for building 64

i.,onagns oecurity *enu5uve ,nrorinpUon; wImnno~u irom purnu pUr ,u t.rr* £'.O3uIUd)i)F Duke Energy Carolinas,, LLC I ONS FLOODING .HAZARD REEVALUATION REPORT

~~Section 5 -Additional Actions width/flood depth ratio is 2. Using these extremely conservative values, the equivalent surcharge depth is equal only to 0.04 ft.

Debris Effects The areas within the protected area that could potentially provide a source for debris are either paved or covered with gravel or paved surfaces with little vegetation or loose materials available. The protected area is also surrounded by vehicle barrier system and security fences which would significantly minimize the potential for an~y debristo impact safety-related Structures. In addition, relatively low velocities would minimize the movement of debris throughout the power block. Therefore, debris effects at ONS were considered negligible.

Effects caused by sediment deposition and erosiOn As described previously, the average maximum velocity throughout the power block is 1.13 ftlsec, with a single highest velocitY' of 6 ft/sec. Since most areas within the power block are paved, no erosion is expected. because maximum values of flow velocity that can be sustained Without Significant erosion are' an order of magnitude higher than the average maximum velocity. The LIP event is a localized flOoding event, which is not expected to car'ry significant amount of sediment typical for riverine flooding. Therefore, sediment deposition at ONS was considered negligible.

Concurrent Site conditions The meteorological events that could potentially result in significant rainfall of the LIP magnitude are squall lines, thunderstorms with .capping inversion, and mesoscale convective systems.

These meteorological e~vents are typically accompanied by hail, strong winds, and even tornadoes.

GroundwaterIngqress The LiP is a localized, short duration event, which is not expected to increase groundwater levels on site. Furthermore;, ONS is protected against groundwater ingress.

-y Sunny-day Failure of JocasseeDam I-Ivdrodynamic and Debris Effects The sunny-day failure of Jocassee Dam and the cascading failure of Keowee Dam results in

.flooding hazard elevation of~b(;)F) n the Ke*Owee tailrace. The flooding hazard elevation is above the nominal elevation of the power block (796 if) and, therefore, safety-related structures throughout the power block will be affected by the associated effects Of the dam failure. The results of the 2-D modeling indicate that the. flow velocities in the Keowee tailrace immediately downstream Of the breach are approximately 40 ft/sec. TheSe high velocities are limited to areas within the immediate breach opening and dissipate to less than 30 ft/sec within 1,000 'ft downstream of the main dambreach. These localized high velocities; will .result in significant hydrodynamic forces and debris Carried by the breach wave .and will result in impact forces on structures located within the immediate breach inundation zone defined by the natural river segment downstream of Keowee Dam leading into the upper reach of Lake, Hartwell, Because the exact flow paths for debris cannot be predicted, any structures located directly downstr'eam of the Keowee Dam breach, will be :considered lost due to hYdrodynamic and debris impact 65

Duke Energy Carolinas. LLC J ONS FLOODING HIAZARD REEVALUATION REPORT Section 5 -Additional Actions loads. Conversely, flow velocities adjacent to the Intake Canal dike and the east slope of the power block are significantly lower (inundation due to backwater impacts) and do not exceed 4 ft/sec.

ConcurrentSite Conditions The results of the FHRR indicate that all access routes from the southeast direction will be inundated during the event and existing infrastructure will likely be severely damaged due to high velocities and impacts from debris and mud flow.

Groundwater Inoqress Groundwater levels on site will not increase due to the sunny-day breach. Furthermore, ONS is protected against groundwater ingress.

66

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT

,, ~Section 6 -References Section 6 - References AMEC Environment and Infrastructure, Inc. 201 4a.Jocassee Dam Breach Parameter Analysis Report. Prepared for =Duke Energy Carolinas, LLC. November 24, 2014.

• (2014b). Data for Site Amplifications-- Rev.i, Fukushima 2,1 Seismic Reevaluation Project, Oconee Nuclear Station, Sen~eca, SC, AMEC Project No. 6234-1t3-0075, Letter from J. Li and C. Sams to R. Keiser dated August 27, 2014.

AmeriCan Nuclear Society (ANS). 1992. ANS-2.8-1992, "Deternmining Design Basis Flooding at Power Reactor Sites." 2.8 Hydrologic Dam Failures," Sections 5.1.3, 5.5.1, and 5.5.4.

1992.

BMT WBM. 2013. TUF!-OW FV Science Manual.

BMT WBM. 2014. TU!FLOW FV User's Manual.

Bohman, L.R. (USGS), Determination of Flood Hydrographs for Streams in~South Carolina:

.Volume 1, Simulation of Flood Hydrographs for Rural Watersheds in South Carolina, Water Resources Investigations Report 89-4087, United States Geological Survey, SWater Resources Division, June 1989.

CEUS-SSC (2*012). Central and Eastern United States Seismic Source Characterization for Nuclear Facilities, U.S. Nuclear Regulatory Commission Report, NUREG-21 15; EPRI Report 1021097, 6 Volumes; DOE Report#. DOE/NE-0140.

Chow, Ven Te, Maidment, David R., Mays, Larry W., Applied Hydrology, McGraw Hill, I1988a.

• 1988b. Open-Channel Hydraulics. 3rd Edition. McGraw-Hill, Incorporated, New York, New York.

Duke Energy, ONS Un~its 1, 2 & 3, Final Safety Analysis Report, Revision. 18, September 12, 20*12.

• .ONS Units 1. 2 & 3, Updated Final Safety Analysis Report, Revision 1, June 25, 2012.

Ehasz, P.E., Joseph L. and Bowles, P.E., Dr. David S., 2013. uJocassee and Keowee Dams, Breach Parameter Review." February 2013.

EPRi. (2013a). Seismic Evaluation Guidance Screening, Prioritization and Implementation Details (SPID) for the Resolution of Fukushima Near-Term Task Force Recommendation 2.1: Seismic, Elec. Power Res. inst. Rept 1025287, Feb.

.(201 3b). EPRI (2004, 2006) Ground-Motion Model (GMM) Review Project, Elec. Power' Res. lnst, Palo Alto, CA, Rept. 3002000717, June, 2 volumes.

Federal Energy Regulatory Commission. 2007. Engineering Guidelines for the Evaluation of Hydropower Projects, Chapter VI, Emergency Action Plans, Office of Hydropower Licensing, Washington, D.C. October 2007.

67

Duke Energy Carolinas, U.C I ONS FLOODING HAZARD REEVALUATION REIPORT Sec~on 6 - References

-- 1993. Engineering Guidelines for the Evaluation of Hydropower Projects, Chapter I!,

Selecting and Accommodating Inflow Design Floods for Dams, Washington, D.C.

October 1993.

Froehlich, D.C.* 2008. "Embankment Dam Breach Parameters and Their Uncertainties.' ASCE Journal of Hydraulic Engineering. December 2008. pp. 1708-1721.

-- 995a. "*Embankment Dam Breach Parameters Revisited." Proceedings Of the 1995 ASCE Conference on Water Resources Engineering, San Antonio, Texas. August. pp.

887-891.

-- 995b. "Peak Outflow from Breached Embankment Dam." Journal of Water Resources Planning and Management, vol.* 121, no. 1, pp. 90-97.

HDR. 2014. Oconee Nuclear Station, External Site Flooding Evaluation, Fukushimna *Study, Jocassee-Keowee Dam Failure Assessment, 1-D HEC-RAS Model. Report prepared for Duke Energy Carolinas, LLC. December 2014.

LCI (2014). Project Report- Oconee Composite profile 1 (Jocassee Dam) seismic hazard using the CEUS seismic source model, Oconee Nuclear Station Hazard, OEP001-PR Revi.

MacDonald, T.C. and J.L. Langridge-Monopo!is. 1984. "Breaching Characteristics of Dam Failures." Journal of Hydraulic Engineering, no. 110., p. 567-587.

N~elz, S. and pender, G. 2013. Benchmarking the Latest Generation of 2D Hydraulic Modelling Packages, Report - SCi120002. UK Environment Agency. Horison House, Deanery Road, Bristol, BSI 9AH. August 2013.

U.S. Department of Commerce, National Oceanic and Atmospheric Administration, U.S.

Departmentof the ArmyNOAA Hydi'ometerologoical Reprot No. 52, Application of Probable Maximum Precipitatio Estimates - United Steates Each of the i Otth Meridain.

1982.

U.S. Army Corps of Engineers. 2011. HEC-Ge0RAS GIS Tools for support of HEC-RAS using ArcGIS, Version 4.3.93. Hydrologic Engineering Center. U.S. Army Corps of Engineers, Institute for Water Resources Hydrologic Engineering Center, Davis, CA. February 2011.

-. 2010. "HEC-RAS, River Analysis System User's Manual, Version 4.1,;" Hydrologic Engineering Center, January 2010.

..2010. HEC-1, Hydrologic Modeling System, Version 3.5.0, Hydrologic Engineering Center, August 20.10.

-. Ice Jam Database, Total Number of Events Ranked by State, https:llrscqisias.crrel.usace.armv.millapex/f?p=2 73:39:1 4230682549501.

68.

Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 6- -References 1982- NOAA HydrometeorologicalReport No. 52, Application of Probable Maximum PrecipitationEstimates - United States East of the 105th Meridian. U.S. Department of commerce, National Oceanic and Atmospheric Administration, August 1982.

.1978. NOAA HydrometeorologicalReport No. 51, Application of ProbableMaximum PrecipitationEstimates - Uni/ted States East of the 105th Meridian. U.S. Department of Commerce, National Oceanic and Atmospheric Administration, June 1978.

U.S., NUclear RegulatOry Commission. 2014. E-Mail, Request for Additional Information- Oconee Flooding Hazard Reevaluation Report (TAG NOS. MF1012, MFi013, AND MFi014),

dated September 15, 2014, (ADAMS Accession No. ML142588222),

,* Letter. 2014. Oconee Nuclear Station, Units 1, 2, and 3, Request for AdditiOnal Information Regarding Fukushima Lessons Learned Flood HaZard ReevalUation Report (TAG Nos. MF1012, MFi01 3, and MF1014), dated March 20, 2014, (ADAMS Accession No. MLMLi 4064A59 1).

• ,2013. "Japan Lessons-Learned Project Directorate .. 213

.-.... Gudhc

.1) o Assessment of FlOoding Hazards Due to Dam Failure,. Interim Staff Guidance, Revision

0. 2013.

-. 2011i. Design-Basis Flood Estimation for Site Characterization at Nuclear Power Plants in the United States of America, NUREG/CR-7046, Office Of Nuclear Regulatory Research, Richland, WA. NOvember 2011.

• .2011. Safety.Evaluation by the office, of Nuclear Reactor Regulation Related to Duke Energy Carolinas, LLC, Confirmatory Action Letter - Commitments to Address External Flooding Concerns, Closure of Inundation Site Results, Oconee Nuclear Station Units 1, 2, and 3 (ONS), dated-January 28, 2011.

.2010. Letter From Luis A Reyes to Dave Baxter. "Confirmatory Action Letter - Oconee Nuclear Station, Units 1, 2, and 3 Commitments to Address External!Flooding Concerns (TAC Nos. ME3065, ME3066 and ME3067)," dated June 22, 2010.

• 2008. Letter from Joseph G. Giitter to Dave Baxter, "Information Request Pursuant to 10 CFR 50.54(f) Related to External Flooding, Including Failure of the Jocassee Dam, at Oconee Nuclear Station, Units 1, 2, and- 3, (TAC NOS. MD8224, MD8225, MD8226)", dated August 15, 2008.

• .2007. Office of NuclearoRegulator-Research,' Regulatory Gu~ide 1.208, A Performance-Based Approach To Define the Site-Specific Earthquake Ground Motion," (March 2007.

Walder, J. and J. O'Connor. 1997. "Methods for Predicting Peak Discharge of Floods Caused by Failure of Natural and Constructed Earthen Dams." Water Resources Research, vol. 33, no. 10, p. 2337-23*48, October 1997.

Wilson 'Engineering, 2013. "Independent Technical Review of HEC-RAS and SRH-2D Modeling," Wilson Engineering Project WE 13002. March 2013.

69

.......... * :t,. c lw LH t..m :m , .;LU*

a .lJrl..~ U~ .~

• Duke Energy Carolinas, LLC I ONS FLOODING HAZARD REEVALUATION REPORT Section 6 - References Youngs, R.R. (2014). 'Review of EPRI 1021097 Earthquake Catalog for RIS Earthquakes in the Southeastern US and Earthquakes in South Carolina Near the Time of the 1886 Charleston Earthquake Sequence." Report transmitted by EPRI letter on March 6, 2014*

Reference Calculations:

• JOC-233493-03 Rev 0, Jocassee Dam-Seismic Dam Performance

• ONS C-00392,192503-003-001 Rev 0 Hydraulic Analysis of External Flood Due to Fair Weather Breach of Jocassee Dam .using TUFLOW FV

• JOC-233493-02-01 Rev, 0, Keowee Reservoir Volume CapaCity - Combined Saddle.

• Dike 1 and Saddle Dike 2 Appurtenant Structure Failure 70