ML13123A204
| ML13123A204 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 03/25/2013 |
| From: | Baxter D, Ehasz J, Ey C, Hubbard D, David Jones, Mccoy R Duke Energy Carolinas |
| To: | Office of Nuclear Reactor Regulation |
| Boska J | |
| References | |
| FOIA/PA-2013-0264, FOIA/PA-2016-0071 | |
| Download: ML13123A204 (33) | |
Text
Fukushima -
Flooding Hazard Reevaluation Upstream Dam Failure Analysis NCR Technical Presentation NRC Headquarters One White Flint North Oconee Nuclear Station Rockville, MD March 25, 2013 For Information Only
Dave Baxter, VP, Regulatory Project Completion Dean Hubbard, Oconee External Flood Licensing Manager Ray McCoy, Principal Engineer, ONS Civil Design Chris Ey, Civil Engineering Manager, HDR Dana Jones, Oconee Fukushima Engineering Supervisor Joe Ehasz, VP, URS Program Manager - Water Resources 2
For Information Only
Agenda Current Dam Failure Analysis - January 28, 2011 Breach Analysis Summary Model Development Updated Dam Failure Evaluation - submitted March 12, 2013 Models Considered Selection of Xu & Zhang Update Breach Parameters Sensitivity Analysis Independent Review Comparative Analysis - Large Modern Dam Failures Modifications Scope 3
For Information Only
2011 Breach Analysis Summary Breach parameters developed using regression methodology and technical papers:
Froehlich 2008 Walder & OConnor MacDonald & Langridge-Monopolis Breach analysis focused on maximizing flooding levels to provide a very conservative and bounding analysis:
Breach dimensions maximized to assume loss of most of the dam embankment.
Froehlich breach time of 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> was reduced to 2.8 Maximum peak outflow was selected from all methods Breach times of Keowee dams/dikes adjusted to maximize water directed at the site Tailwater effect below Jocassee dam was not considered 4
For Information Only
2011 Breach Analysis Summary Jocassee Dam (postulated dam failure)
Initial breach derived primarily from Froehlich regression equations.
Breach dimensions were adjusted based on physical constraints of natural valley Jocassee breach parameters:
Top Width - 1156 (64% of overall crest)
Bottom Width - 431 feet Bottom Elevation - 800 msl Breach Formation Time - 2.8 hrs, Peak outflow 5,400,000 cfs 5
For Information Only
2011 SE Jocassee Dam Breach Progression and Stage-Discharge Hydrographs 6
For Information Only
2011 Breach Analysis Summary Keowee Dam/Dikes (postulated cascading dam failures)
Overtopping failure trigger of two feet over the crest Cascading dam/dike failure on Keowee Keowee main dam- 2.8 hrs West Saddle Dam - 0.5 hrs Intake Canal Dike- 0.9 hrs Little River Dam - 1.9 hrs Conservative assumptions were made to maximize the water directed toward the power block 7
For Information Only
Model Development HEC-RAS 1D Model 8
For Information Only
Model Development SRH 2D Model (57 thousand elements) 9 For Information Only
2011 Breach Analysis Summary 2D Model 10 For Information Only
Updated Dam Failure Evaluation 11
Updated Dam Failure Evaluation Fukushima 2.1 Attributes of updated and refined dam failure analysis Updated methodology and present day regulatory guidance Performed to meet NUREG CR/7046, 2011 & ANS 2.8, 1992 Realistic but still conservative assumptions Physical characteristics of the dams/dikes recognized including materials and method/quality of construction Overtopping and Seismic are confirmed from the 2011 SE as not being credible failure modes 12 For Information Only
Updated Dam Failure Evaluation Fukushima 2.1 Overtopping of the Jocassee dam was confirmed not to be a credible failure mode The Jocassee dam and dikes include 15 feet of freeboard The Jocassee watershed is small relative to storage capacity - 148 square miles The top of the spillways are located at 1110 (full normal level)
Four diverse methods of assuring spillway gate operation Rigorous spillway gate maintenance and surveillance testing as required and monitored by FERC Lake management procedures require consideration of lower level to anticipate additional storage needs for significant storms Weekly rain forecast are prepared by Duke Energy to project rainfall for the basin Precipitation monitoring has assured that no overtopping of the spillway gates has occurred in 40 + years of operation PMF using current HRR-51,52 results in 3 feet of freeboard margin 2011 SE also concluded that overtopping was not credible 13 For Information Only
Updated Dam Failure Evaluation Fukushima 2.1 Seismic Failure of the Dam was confirmed not to be a credible failure mode Seismic evaluation based on current FERC criteria using the 1989 EPRI Hazard Curves The Jocassee dam is designed to a 0.12 g horizontal ground acceleration (Oconee site is designed to a 0.1g horizontal ground acceleration).
2007 Updated Fragility Analysis High Confidence of a Low Probability of Failure (HCLPF) of the dam by sliding 0.305 g Evaluation was performed by Applied Research & Engineering Sciences (ARES) Corp., formerly EQE, a respected consulting firm in the area of seismic fragility The ARES report concluded the median centered fragility value for failure of the dam is 1.64 g.
Maximum Probabilistic Peak Ground Acceleration for a 2% probability of being exceeded within a 50 year period is 0.197 g (using the United States Geologic Service hazard maps applicable to Jocassee).
Jocassee dam is included in the seismic model of the Oconee Probable Risk Assessment.
The combination of the updated seismic fragility with the seismic hazard curve results in a negligible risk contribution from seismic events.
In a letter dated 11/20/07 and in the 1/28/11 SE report, the NRC concluded that there is a negligible risk 14 For Information Only
Models Considered Regression Analysis Froehlich 2008 Walder & OConnor MacDonald & Langridge-Monopolis 1984 Xu & Zhang 2009 15 For Information Only
Selection of Xu & Zhang 2009 Basis Most current regression method developed and validated with the largest data base of dam failures:
182 earth and rockfill dam failures compiled 75 failures w/ sufficient info to develop breach regression models Empirical formulas that account for physical characteristics of dam/reservoir: dam type, failure mode, height, dam erodibility, reservoir shape/storage) 33 of the 75 failures were on large dams ( > 15 meters )
Applies to multi-zoned dams Method yields realistic but conservative breach parameters Recognized by industry experts 16 For Information Only
Breach Parameters Fukushima Update Jocassee Dam - Xu & Zhang Starting reservoir elevation 1110 (normal full pond)
Rockfill dam with low erodibility classification Piping failure initiating at 1020 feet msl (Sunny Day Failure)
Breach parameters:
Top Width - 701 (39% of overall crest)
Bottom Width - 431 Bottom Elevation - 870 Breach Formation Time:
Xu & Zhang - 29.2 hrs.(13.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> piping +16.0 open weir)
Froehlich - 16.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> (open weir)
Peak outflow: 1,760,000 cfs 17 For Information Only
Jocassee Dam Low Erodibility Classification Diagram removed due to security sensitive information 18 For Information Only
Fukushima Model JOCASSEE DAM BREACH PARAMETERS Reservoir Breach Crest Bottom Breach Bottom Average Time to Top of Starting Right Side Left Side Breach Initiation Structure Elevation Failure Mode Elevation (ft Breach Width Breach Failure Breach Elevation (ft Slope (Zr) Slope (Zl) Progression Elevation (ft (ft msl) msl) (ft) Width (ft) (Hr) Width (ft) msl) msl)
Jocassee 1125 1,110 Piping 870 431 566 0.53 0.53 29.2 701 Sine Wave 1,020 Dam Breach Formation Time Xu & Zhang definition: 29.2 (13.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> piping + 16.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> open weir)
Froehlich definition: 16.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> open weir 19 For Information Only
Fukushima Model Jocassee Dam Breach Progression and Stage-Discharge Hydrographs Breach Formation Time ; Xu & Zhang definition: - 29.2 (13.2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> piping + 16.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> open weir) Froehlich definition: -16.0 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> open weir 20 For Information Only
Breach Parameters Fukushima Update Keowee Dam Starting reservoir elevation 800 (normal full pond)
Homogeneous earth fill dam Overtopping failure trigger of two feet over the crest at 817 msl by rapid rise of Keowee reservoir over the crest Multiple simultaneous breach initiation formation points across the Keowee dam and West Saddle dam Cascading dam/dike failure on Keowee Keowee main dam- 0.75 hrs West Saddle Dam - 0.5 hrs (shorter than main dam, ratio of height) 21 For Information Only
Fukushima Model Keowee Dam Breach Progression HEC-RAS 1
0.9 0.8 0.7 Relative Breach Progression 0.6 0.5 0.4 0.3 0.2 0.1 0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Relative Time Progression 22 For Information Only
Fukushima 1D Modeling Keowee Dam - Headwater and Tailwater Stage Hydrographs Final BEP LE 1-D Model Performance 830 820 810 800 790 780 770 760 Elevation - feet msl 750 740 730 720 710 700 690 680 670 660 650 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 Model Time - hours BEP LE HW BEP LE TW 23 For Information Only
Fukushima 2.1 2D Modeling Keowee Dam Breach Progression 24 For Information Only
Fukushima 2D Modeling Velocity and Flow Pattern at 17 hrs.
25 For Information Only
Fukushima 2D Modeling Velocity and Flow Pattern at 20 hrs.
26 For Information Only
Fukushima 1D-2D Modeling Results Breaching Keowee Dam Intake Dike HEC-RAS 2-D HEC-RAS 2-D Elevation Decimal Time Elevation Decimal Time Elevation Decimal Time Elevation Decimal Time 817 16.28 817 16.24 n/a n/a n/a n/a Maximum Water Surfaces Keowee Dam Intake Dike HEC-RAS 2-D HEC-RAS 2-D Elevation Decimal Time Elevation Decimal Time Elevation Decimal Time Elevation Decimal Time 818.4 16.53 820.1 16.58 810 17.17 807.2 17.67 Maximum Water Surfaces Swale Tailwater HEC-RAS 2-D HEC-RAS 2-D Elevation Decimal Time Elevation Decimal Time Elevation Decimal Time Elevation Decimal Time 817.5 16.55 815.5 16.53 787.4 17.52 790.4 18.41 Maximum Water Surfaces SSF SSF HEC-RAS 2-D HEC-RAS 2-D Elevation Decimal Time Elevation Decimal Time Depth Decimal Time Depth Decimal Time n/a n/a n/a n/a 0 n/a 0 n/a 27 For Information Only
Sensitivity Analysis Model Peak Outflow (cfs)
McDonald & Langridge-Monopolis 1984 1,566,381 Costa, 1985 1,634,480 Xu & Zhang, 2009 1,760,000 Evans, 1986 1,803,331 SCS, 1981 2,647,711 Bureau of Reclamation, 1982 3,046,462 McDonald & Langridge-Monopolis 1984 5,093,603 (upper envelope)
Froehlich (with additional conservatism), 2008 5,440,000 Data in this table based on Wahl 2004, January 28, 2011 SE and updated Xu & Zhang data 100+ HEC-RAS studies performed with varied breach parameters and control variables Erodiblity was the most significant factor influencing the breach parameters for Xu & Zhang 2009 Bias of conservatism with realism 28 For Information Only
Independent Review Breach Parameters
- Independent Peer Review Joe Ehasz, P.E.
David Bowles, Ph. D P.E. P.H.
- FERC Board of Consultant Review Gonzalo Castro, Ph.D., P.E.
James Michael Duncan, Ph.D., P.E.
James F Ruff, Ph.D., P.E.
Gabriel Fernandez, Ph.D., P.E.
29 For Information Only
Comparative Analysis Large Modern Dam Failures Taum Sauk Overtopping failure initiated by human error (previous overtopping events had occurred)
Random rockfill embankment supporting the inner concrete liner loosely placed by end dumping the material without compaction except for the top 16 of 84 height The embankment was constructed on a very steep downstream slope of 1.3H to 1V with a 10 high concrete parapet wall along the crest of the dam Embankment was highly erodible and contained over 45% sand sized material (also evident in unusual level of surface erosion from rain events)
.Teton earthen dam with majority of dam constructed of highly erodible windblown silt (infant mortality event)
No transition zones (sand and/or fine filters) were included between the silt core and the sand & gravel Thin layer of small rock fill on both up and downstream faces with a majority of protection relied upon mix of sand, gravel and cobble Piping failure at 130 below the crest due to inadequate protection of impervious core trench material Breach top width 781 (~25% of overall crest)
Hell Hole True rockfill dam,with upstream sloping impervious core with massive rock fill sections up and down stream to support and protect the core.
Failure caused by overtopping during construction due to an intense rain event that could not be passed through the construction diversion tunnel After overtopping of the core started, the dam took 26 hours3.009259e-4 days <br />0.00722 hours <br />4.298942e-5 weeks <br />9.893e-6 months <br /> to complete the breach and empty the upstream reservoir 30 For Information Only
Modification Scope Updated Modifications for protection from dam failure (under review):
- 1. Relocation of external backup power transmission line
- 2. Intake Dike embankment protection
- 3. East embankment protection
- 4. Discharge Diversion wall Modifications for Local Intense Precipitation (under review):
Transformer relocation Diversion walls and drainage canals Aux building and Turbine building protection 31 For Information Only
Modification Options Jocassee Dam 1
32 For Information Only
Questions and Feedback 33