ML091380424

Slides for Closing Meeting May 11, 2009 with Duke on the Oconee Flooding Issue

ML091380424
Person / Time
Site:	Oconee
Issue date:	05/11/2009
From:	Duke Energy Carolinas
To:	Office of Nuclear Reactor Regulation
Thompson, Jon 415-1119
References
FOIA/PA-2013-0004A
Download: ML091380424 (39)

Text

Oconee Nuclear Station External Flood NRR Meeting Rockville, MD May 11,2009 Withhold from public disclosure 1 under 10 CFR 2.390

Duke Attendees

~ Mike Glover, ONS General Manager, Nuclear Plant Projects

~ Rich Freudenberger, ONS Safety Assurance Manager

~ Bob Meixell, ONS Regulatory Compliance Manager

~ Tim Brown, ONS Project Manager

~ Ray McCoy, ONS Principal Engineer

~ Brad Keaton, Duke Hydro Licensing Engineer

~ David Cummings, Duke Assistant General Counsel

~ Ed Luttrell, HDR/DTA Senior Vice President

~ Chris Ey, HDR/DTA Civil Engineering Manager Withhold from public disclosure 2 under 10 CFR 2.390

Agenda

~ Morning Session

-Opening Remarks

-Inundation Analysis Results oSummary of Previous Inundation Analyses oSummary of Current Inundation Analysis Withhold from public disclosure 3 under 10 CFR 2.390

Agenda

~ Afternoon Session

• Oconee Site External Flood Licensing Basis o Jocassee Design o External Flood Licensing Basis o Precedence o Status of Commitments

• Objectives of Risk Study

• Response to NRC Questions

• Summary

• Closing Remarks Withhold from public disclosure 4 under 10 CFR 2.390

Opening Remarks

~ Duke has completed a new detailed inundation study, given a failure of Jocassee. The results of this analysis are significantly different than the results reported from previous 1992 inundation study.

~ Today, we will describe a summary of the inputs to the new study and the results. We will then compare those inputs and results to those of the previous inundation study.

~ Duke has located additional design information regarding the Jocassee design, including inspection records from construction.

~ We have completed additional licensing research that demonstrates that the AEC understood the design of Jocassee during the original licensing of ONS.

~ We maintain that Jocassee is a well designed, well constructed, well maintained project, and meets all FERC requirements.

~ We will describe the seismic analysis of Jocassee, as well as PMP/PMF analysis that have been completed, submitted to and accepted by FERC.

Withhold from public disclosure 5 under 10 CFR 2.390

Opening Remarks

~ We will discuss Regulatory Guide 1.59 and ANSI N170-1976 and their possible application to this issue.

~ We will then state our position regarding the licensing basis of the ONS for external flooding.

~ We will discuss precedence established at another NRC licensed facility regarding external flooding following the postulated failure of an upstream dam.

~ We will provide a status of commitments.

~ We will describe what additional actions we plan to take.

~ Finally, we will provide the objective of the risk analysis and how the results of that analysis will be used in the ONS risk assessment model for external events.

Withhold from public disclosure 6 under 10 CFR 2.390

Morning Session

~ Inundation Analysis Results

• Summary of Previous Inundation Analyses

• Summary of Current Inundation Analysis Withhold from public disclosure T under 10 CFR 2.390

Summary of Previous Inundation Studies

~ 1983 Study

• DAMBRK Software

• Focused on Oconee for early PRA work.

• No Keowee Reservoir Dam failures assumed

• Level pool rise assumed over Keowee Reservoir basin

• Breach parameters were not well documented

• Results in 2.45 ft of water overtopping Keowee Dam and Oconee Intake Canal Dike

• Overtopping of Intake Canal Dike produces yard water height of 4.71 ft, considering a hydraulic jump.

~ Action: Initial construction of 5 ft. SSF flood walls, as a PRA enhancement.

Withhold from public disclosure 8 under 10 CFR 2.390

Summary of Previous Inundation Studies

~ 1992 FERC EAP Study

• DAMBRK Software

• Focused on Emergency Action Plan (EAP) flood routing and inundation maps.

• Single breach size ('1992'): Bottom elev. 800 ft. msl., bottom width 250 ft., 1:1 side slopes, average width 575 ft.

• A failure time of 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.

• Jocassee failure calculations submitted and accepted by FERC for EAP.

• Domino-effect assumes Keowee Dam failure also.

• Keowee connecting channel characteristics modeled as off-channel levee with EAP focus on downstream impact.

• 'Sunny Day' dam failure resulting water elevation of 823.3 at the Keowee Dam (8.3 ft above dam crest).

• 'Sunny Day' failure resulted in a yard (tailrace) flood elev. of 805.5 ft. (12.5 ft.

above yard elevation).

~ Action: PRA updated with results for IPEEE submittal.

Withhold from public disclosure 9 under 10 CFR 2.390

Summary of Inundation Studies

~ September 2008 Study

• DAMBRK Software

• Focused on Oconee - resurrected 1992 FERC EAP model.

• Model results focused on the '1992' [bottom breach elev. 800, average breach width 575 ft.]

• The Jocassee breach parameters accepted by FERC for 1992 EAP work.

• Domino-effect assumes Keowee Dam failure.

• Jocassee and Keowee water levels varied for resultant sensitivity.

• Using normal lake levels, 'Sunny Day' Jocassee Dam Failure resulted in:

o Flooded tailrace (ONS yard) to elev. 807.2, or 11.2 ft. of water in the yard

• Using 9/2008 lake levels, 'Sunny Day' Jocassee Dam failure also resulted in:

o Flooded tailrace (ONS yard) to elev. 803, or 7 ft. of water in the yard

~ Action: Designed and constructed an extension of the SSF Flood Wall to 7.5 ft to provide additional margin.

Withhold from public disclosure 10 under 10 CFR 2.390

Summary of Inundation Studies

~ December 2008 Study

• HEC-RAS Software

• Oconee focused

• 1992 breach [bottom breach elev. 800, average breach width 575 ft.],

accepted by FERC for EAP.

• Domino-effect assumes Keowee Dam failure.

• Jocassee and Keowee water levels varied for resultant sensitivity.

• 'Keowee Connecting Channel' modeled with a focus on flow conveyance between Little River arm and Keowee arm.

• Significantly enhanced topography modeled upstream/downstream of ONS.

• Model is 44 miles long, 149 input cross sections, and 468 interpolated cross-sections in model.

Withhold from public disclosure 11 under 10 CFR 2.390

Summary of Inundation Studies

~ December 2008 Study I continued

• New results compared to DAMBRK results for insights/direction.

• Using normal lake levels, 'Sunny Day' Jocassee Dam Failure resulted In:

o Flooded tailrace (ONS yard) to elev. 793.3, or 2.7 ft below the yard level.

o Flood overtops the Keowee Dam by 16.5 ft (water surface elev. 831.5).

o Water surface elev. at the Oconee Intake Canal Dike is not directly calculated.

~ Action: Additional study warranted due to:

• New and different results with regard to tailrace elevation.

• Water elevation at Little River Dam and Oconee Intake Canal elevation now of interest.

• Assess sensitivity of the cross sections, flow path, and Manning's In' values.

Withhold from public disclosure 12 under 10 CFR 2.390

Summary of Inundation Studies

~ March 2009 Study

.- HEC-RAS Software

• Based on December 2008 model with increased detail and extended model to Lake Hartwell Dam to determine downstream backwater impacts and sensitivity of cross section geometry.

• Areas of focus determined by review of December 2008 model:

o Verify storage volume in Keowee and Little River arms.

o Assess sensitivity of Manning's 'n' values.

o Confirm / modify active and inactive flow areas.

o Model ONS Intake Canal to determine flood elevation.

o Use momentum and energy equations, as needed, for more accurate results o Assume a breach in ONS Intake Canal Dike.

o Assure all flows from ONS Intake Canal Dike and Keowee Dam are combined and routed to the tailrace for complete backwater effect.

Withhold from public disclosure 13 under 10 CFR 2.390

Summary of Inundation Studies

~ March 2009 Study I continued

• Dam Failures/Breaches o Modeled two different breaches - but results focused on '1992'.

o 1992 EAP model breach is 250 ft wide, bottom elevation 800, time to failure 4 hrs., average width 575 ft.

o Jocassee Breach discharge rate compared to previous computed values for validation and previously suggested non-linear flow was not exercised.

o Failures of Keowee Dam, ONS Intake Canal Dike, and Little River Dam initiated due to overtopping.

Withhold from public disclosure 14 under 10 CFR 2.390

Summary of Inundation Studies

~ March 2009 Study / continued

• Flow Modeling o Jocassee and Keowee water levels varied for resultant sensitivity.

o Keowee Connecting Channel modeling examined to simulate realistic flow behavior using HEC-RAS built in options.

o All significant junctions re-examined for momentum vs. energy equation modeling considerations.

• Manning's In' o Multiple Manning's In' values used to address changing flow channel conditions.

o Manning's In' sensitivity completed using peak water surface cases.

o Confirmed assumed In' values used by consistent outcomes when In' was varied over a range per 'Open Channel Hydraulics' by Chow.

Withhold from public disclosure 15 under 10 CFR 2.390

Summary of Inundation Studies

~ March 2009 Study / continued

• Geometry o Significantly increase number of input cross sections to capture changes in topography and resulting impact.

o Model is 44 miles long, 1411 input cross sections, and 5762 cross-sections interpolated in model.

o Lake Keowee Connecting channel bridge abutments (major contributor to possible flow restriction) included in the model, bridge piers and deck were not.

o Computed reservoir storage/volumes were verified using input cross sections:

- Lake Keowee computed storage volume within 2.6% of the full pond storage from file data.

- Lake Hartwell was within 4.1 % of the full pond storage.

Withhold from public disclosure 16 under 10 CFR 2.390

Summary of Inundation Studies

~ March 2009 Study I continued

• Using Normal Pond levels, Jocassee '1992' breach results in:

o Water overtopping the Keowee Dam by 15.7 ft (water surface elev. 830.7).

o Water overtopping the Oconee Intake Canal Dike by 4.5 ft (water surface at elev. 819.5).

o Tailrace floods to elev. 792.5, or 3.5 ft below the yard level - therefore no ponding in yard (from tailrace).

Withhold from public disclosure 17 under 10 CFR 2.390

Summary of Inundation Studies

~ March 2009 Study comparison to 1992 Study Result 1992 3/2009 Tailrace elevation 12.5 ft. above 3.5 ft. below yard (elev. yard (elev.

805.5 ft. msl.) 792.5 ft. msl.)

Intake Canal Dike overtop water elev. nla 4.5 ft. above dike (elev.

819.5 ft. msl.)

Keowee Dam overtop water elev. 8.3 ft. above 15.7 ft. above dam (elev. dam (elev.

823.3 ft. msl.) 830.7 ft. msl.)

ONS Yard Elev. 796 ft. msl.

Withhold from public disclosure 18 under 10 CFR 2.390

Summary of Inundation Studies

~ March 2009 Study comparison to 1992 Study

• Factors contributing to differences:

o Software capabilities o More detailed topography o Number of cross sections and detail per cross section o Connecting channel flow conveyance o Capability of modeling intersections and storage areas o Manning's In' capability and use

• High confidence this model realistically predicts potential flooding at the Oconee site due to detailed internal reviews by expert consultants, including an independent review, in addition to a owner's review by Duke.

Withhold from public disclosure 19 under 10 CFR 2.390

Afternoon Session

• Oconee Site External Flood Licensing Basis o Jocassee Design o External Flood Licensing Basis o Precedence o Status of Commitments

• Objectives of Risk Study

• Response to NRC Questions

• Summary

• Closing Remarks Withhold from public disclosure 20 under 10 CFR 2.390

Jocassee Design PMP/PMF:

~Oconee SER Appendix D:

-AEC Consultant's Analysis conformed with AEC requirements, as summarized in the February 1972 draft, 'Standard Format and Contents of Safety Analysis Reports for Nuclear Power Plants'

-Runoff depth 26.6 inches in 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.

-PMP hydrograph based on an infiltration rate of .05 inches per hour.

-Peak inflow was 245,000 cfs

-Peak reservoir stage was 1122.5 ft. msl. (2.5 feet of freeboard)

-Peak reservoir stage assumed flow through both spillway gates, with no units operating. Peak outflow through both gates was 72,000 cfs.

Withhold from public disclosure 21 under 10 CFR 2.390

Jocassee Design PMP/PMF:

~Latest (1991):

• Average Probable Maximum Storm (PMS) depth of 36.41 inches in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />

• PMP hydrograph based on an infiltration rate of .093 inches per hour, equivalent to a runoff eN of 60.

• Peak inflow was 522,734 cfs

• Peak reservoir stage was 1122.0 ft. msl. (3.0 feet of freeboard)

• Peak reservoir stage assumed flow through both spillway gates, and four operating units. Peak outflow through the gates and units was 85,400 cfs.

• However calculated outflow was based on a spillway gate width of 30 feet. Gates are actually 38 feet wide. The difference in flow over the spillway given a gate width of 38 feet is equivalent to two operating units.

Withhold from public disclosure 22 under 10 CFR 2.390

Jocassee Design Seismic:

~ Seismic Design of Jocassee was understood by AEC during the original licensing of ONS.

~ Slope Stability Analysis (As reported to FERC):

• Methodology: Spencer's Method of Limit Equilibrium.

• Pseudo-static EO analysis assumed a horizontal ground acceleration of

.12 g.

• Core shear strength parameters used in the analysis were based on saturated tri-axial shear tests made during construction.

• Calculated factors of safety at full pond: 1.24 (upstream slope), 1.13 (downstream slope) versus required factor of safety of 1.0 Withhold from public disclosure 23 under 10 CFR 2.390

Jocassee Design Core Material:

~ Design Test Criteria:

• % Compaction: Greater than or equal to 93°~ of Standard Proctor

• Dry Density: Greater than or equal to 95 Ibs/ft3

• Moisture Content: + 3°~ of optimum on Standard Proctor Curve

~ Testing:

• Soil tests occurred periodically during construction and were evaluated monthly.

• All test results were presented to and approved by the Consulting Board.

Withhold from public disclosure 24 under 10 CFR 2.390

Jocassee Design

~ Rock:

• Design - Compacted Unit Weight: 140 Ibs/ft3

~ Random Rock:

• Design - Compacted Unit Weight: 124 Ibs/ft3

~ Testing:

• Compaction tests occurred periodically during construction and were evaluated monthly.

• Due to variability in field density tests, periodic shear tests were performed.

• All test results were presented to and approved by the Consulting Board.

Withhold from public disclosure 25 under 10 CFR 2.390

External Flood Licensing Basis

~ The ONS site flood licensing basis is predicated on the ability of both the Keowee and Jocassee to withstand natural phenomena.

~ The ONS site flood licensing basis is based on the amount of precipitation that could fall directly on ONS.

~ Random failure of either dam is beyond licensing basis, however the failure of each has been included in the risk assessment for external events.

Withhold from public disclosure 26 under 10 CFR 2.390

External Flood Licensing Basis

~ The Oconee Nuclear Station (ONS), the Keowee Hydroelectric Station, and the Jocassee Pumped Storage Plant form part of the greater Keowee-Toxaway Project. All of these facilities were built at approximately the same time.

~ Duke's licensing research indicates that the Atomic Energy Commission (AEC) understood the design of Jocassee during the original licensing of ONS.

~ These licensing documents indicate that the AEC requested information regarding the seismic design of Jocassee and the facility's ability to hold and pass a PMP/PMF.

~ The AEC requested and obtained an independent consultant's hydrological analysis of the Jocassee reservoir.

~ This analysis confirmed the ability of Jocassee to hold and pass a PMF.

~ The consultant's report can be found in Appendix D of the ONS SER.

Withhold from public disclosure 27 under 10 CFR 2.390

External Flood Licensing Basis

~ NUREG 965 (10/31/83), "NRC Inventory of Dams" list those dams that could be considered in the NRC licensing process if dam failure could lead to flooding of a nuclear power plant.

~ Jocassee is listed among a group of dams that could be considered in the NRC licensing process.

~ However, the NUREG also notes that Jocassee is among a group of dams regulated by other Federal agencies (FERC or TVA).

~ The group of dams regulated by other federal agencies were not included in the group of dams for which the NRC was considered to be

'responsible. '

~ Jocassee meets all FERC requirements.

Withhold from public disclosure 28 under 10 CFR 2.390

External Flood Licensing Basis

~ Regulatory Guide 1.59 (8/1977), "Design Basis Flood for Nuclear Power Plants:

• "Consideration of the most severe natural phenomena."

• "Nuclear Power plants should be designed for the most severe flood conditions that can be reasonably predicted to occur as a result of severe hydro-meteorological conditions, seismic activity or both."

• "Flood conditions that could be caused by dam failures from earthquakes should also be considered."

• "Consideration of seismically induced floods should include the same range of seismic events as is postulated for the design of the nuclear plant."

~ Applicability: Although ONS is not committed to this REG Guide, Jocassee is designed to withstand the most severe natural phenomena events applicable to the facility, including a seismic event, and a PMP/PMF event.

Withhold from public disclosure 29 under 10 CFR 2.390

External Flood Licensing Basis

~ American National Standard ANSI N170-1976, "Standards for Determining Design Basis Flooding at Power Reactor Sites":

• "Guidelines are to be developed to establish design basis flooding at Power Reactor Sites as a result of river, stream, seismically-induced dam failure, surge, seiche, and/or wave action flooding".

• Scope: "Methodology is described to evaluate the flood having virtually no risk of exceedance that may be caused by: precipitation and snowmelt, and any resulting dam failures; seismically induced dam failures; surge or seiche and attendant wind-generated wave activity; or a reasonable combination of these events."

~ Applicability: Although ONS is not governed by the standard, Jocassee is designed such that it will not fail during either precipitation events or seismic events.

Withhold from public disclosure 30 under 10 CFR 2.390

External Flood Licensing Basis

~Conclusions:

• The external flood licensing basis of ONS is based on the consideration of natural phenomenological events and the analysis of these events for the Jocassee dam.

• As demonstrated the Jocassee dam can withstand a seismic event and PMP/PMF event.

• Consideration of random failures of Jocassee is outside of the licensing basis for ONS Withhold from public disclosure 31 under 10 CFR 2.390

Precedent

~ Yankee Rowe Nuclear Plant - Integrated Plant System Assessment Systematic Evaluation Program

• Potential to inundate the Yankee Rowe site following a failure of the upstream Harriman dam (in Vermont).

• Yankee Rowe Nuclear Plant demonstrated that the Harriman dam met all FERC safety criteria

• Dynamic stability analyses demonstrated that Harriman was stable under seismic loads imposed by a composite spectrum with a peak ground acceleration of .1 g.

• Hydrological analyses demonstrated that Harriman would not be overtopped and therefore that the dam does not fail during the maximum rainfall event.

• Ultimately the NRC decided any corrective actions would be determined by the FERC Part 12 program for dam safety.

Withhold from public disclosure 32 under 10 CFR 2.390

Status of Commitments Commitment Completion Date Perform flooding studies using HEC-RAS for 12/2008 (Complete) comparison with previous DAMBRK models.

Create interim guidance to address mitigation of 2/2009 (Complete) postulated flood events that could render the SSF inoperable.

Implement short term modifications to extend the 2/2009 (Complete) height of the existing SSF flood walls to 803 ft msl.

Complete Risk Assessment Study by RAC and 2/2010 New Flood Study using HEC-RAS, by HDR/DTA.

Withhold from public disclosure 33 under 10 CFR 2.390

Risk Study

~Objectives of Risk Study:

• Determination of risk profiles for each failure mode.

• Determination of inundation level at the Oconee site for each failure mode.

• Assignment of risk for each failure mode /

inundation level.

• Results of study to be incorporated in the ONS risk assessment model for external events.

Withhold from public disclosure 34 under 10 CFR 2.390

Future Actions

~ Determination of inundation level in the ONS yard caused by overtopping of the Intake Canal Dike

• Conservative evaluation using a simplified methodology

• 2D modeling

~ Summary revision of UFSAR:

• Description of Jocassee seismic analysis

• Description of Jocassee PMP/PMF analysis Withhold from public disclosure 35 under 10 CFR 2.390

Response to NRC Questions

~See attached handout.

Withhold from public disclosure 36 under 10 CFR 2.390

Summary

~ Jocassee is a well designed, well constructed, well maintained project, and meets all FERC requirements

~ Based on its design, Jocassee is not postulated to fail during either a PMP/PMF or a seismic event.

~ Based on our licensing research, we believe the AEC understood the design of Jocassee during the original licensing of ONS.

~ Our review of regulatory guides and precedent indicate that only phenomenological events should be considered in the failure analysis of dams upstream of nuclear facilities. Postulation of

'Sunny Day' failures is not included in these guides and not reflected in the precedent.

Withhold from public disclosure 37 under 10 CFR 2.390

Summary

~ Duke has increased the height of the current SSF walls to provide additional safety margin against external flooding.

~ Interim Guidance

~ Duke has commissioned an independent risk study to determine appropriate dam failure probabilities and resulting inundation levels, for inclusion in the ONS External Events PRA.

~ Duke plans additional actions to quantify the amount of inundation in the ONS yard, following a postulated failure of the Intake Canal Dike and to update the ONS UFSAR to more adequately describe the external flood licensing basis.

Withhold from public disclosure 38 under 10 CFR 2.390

Closing Remarks

~ Duke has completed a detailed flood analysis focused on ONS, given a failure of Jocassee. The analysis demonstrates that the site will not be inundated from the

~eowee tailrace, based on previously evaluated breach sizes.

~ Duke maintains that a deterministic resolution of a random failure of the Jocassee dam is not warranted.

~ Given Duke's view of the safety significance and the complexity of the analysis to resolve this issue, we believe our actions have been timely.

~ Duke believes that resolution of this matter should be considered

. . for transition from a site specific issue to a genenc Issue.

Withhold from public disclosure 39 under 10 CFR 2.390