Boska. John

From:Sent:To:Subject:Attachments:

Follow Up Flag:Flag Status:

Hubbard, Dean M <Dean.Hubbard@duke-energy.corn>Tuesday, April 23, 2013 11:11 AMBoska, JohnSecurity Sensitive VersionNRC Presentation March 25. 2013 FINAL Rev C (Security Sensitive).pdf

Follow upCompleted

John, Attached is the security sensitive version for the other side of Adams. I have added the security sensitivestatement at the bottom of the slides.

Dean

1

' DukeOEnergy

Fukushima -

Flooding Hazard Reevaluation

Upstream Dam Failure Analysis

NCRTechnical

Presentation

NRC Headquarters

One White Flint North

Rockville, MD

March 25, 2013

For Information Only"Wthl.Bmphi i~ie,, ,• ••.23

9 DukeUEnergy

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

For Information Only.. tho ... r.....m.. P ,••ublcD~ ,• u,,u•, 10 CFR

DDuke Aed1 j(ergy Agenda

,:, Current Dam Failure Analysis -January 28, 2011

ý Breach Analysis Summary

ý 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

For Information Only

9 Duke 2011 Breach Analysis Summary

.:. Breach parameters developed using regression methodology andtechnical papers:

ý Froehlich 2008

ý Walder & O'Connor

ý MacDonald & Langridge-Monopolis

. Breach analysis focused on maximizing flooding levels to provide avery conservative and bounding analysis:

Breach dimensions maximized to assume loss of most of the damembankment.

Froehlich breach time of 5 hours was reduced to 2.8

Maximum peak outflow was selected from all methods

, Breach times of Keowee dams/dikes adjusted to maximize waterdirected at the siteTailwater effect below Jocassee dam was not considered

For Information Only;;tA" fm P.le-1ý15re ýuner •

lukeSnergy

2011 Breach Analysis Summary

Jocassee Dam (postulated dam failure)

Initial breach derived primarily from Froehlichequations.

regression

*:, Breach dimensions were adjusted based on physicalconstraints of natural valley

o:o Jocassee breach parameters:ý Top Width - 1156 (64% of overall crest)ý Bottom Width - 431 feet

ý Bottom Elevation - 800 msl

ý Breach Formation Time - 2.8 hrs,ý Peak outflow 5,400,000 cfs

For Information Only

DukeEnergy

2011 SE Jocassee Dam BreachProgression and Stage-Discharge Hydrographs

For Information Only-.Wl*ithh•ld from Pubklic DI'•._ ...,,,,. ,..,..,,.,.o,,;. uiuJ~r 10 CFR 2'.39

R Duke 2011 Breach Analysis SummaryOEnergyKeowee Dam/Dikes (postulated cascading dam failures)

,:o Overtopping failure trigger of two feet over the crest

,:o Cascading dam/dike failure on Keoweeý Keowee main dam- 2.8 hrs

ý West Saddle Dam-0.5 hrs

ý Intake Canal Dike- 0.9 hrs

ý Little River Dam - 1.9 hrs

*:. Conservative assumptions were made to maximize the waterdirected toward the power block

For Information Only,,- .- ,.. OWMT!eP::z

9 DukeUEergy

Model DevelopmentHEC.RAS ID Model

•/BJoaum"" .w" mD.. ..

NOTE:bo

For Informa-,Withl .. Pab'i. DB=,tmuwure undeM.

*9e Model DevelopmentSRH 2D Model

(57 thousand elements)1nergyREW1ICOMIEATIONALfE

For Information OnlyWithhold from DPulic rle.o..re 4undcr 10 -FR 9-O

Puke.nergy

2011 Breach Analysis Summary2D Model

WATER SURFACE ELEVATIONS AT KEOWEE DAM

Jocassee-Keowee Dam Breach StudyPool Elevations at Keowee Dam

860

840

E

C0

4p

o20- Case 10OW

Case 0OOW 1-Case 100W2

--- Case 10OW-3

Case 100W_4-- 4------

780

. . . ..

0 1 2 3

760

740

4 5 6 7 8 9 10 11 12

Model Time (hrs)

For Information Only:th hold from Db.lic Dieclosur^ under-10 n R .3- 0

OPukewnergy

Updated DamFailure Evaluation

For Information OnlyWithk•Id _rnrr public- nlkeIn_•,,rp ,-_-_r 1.0 CF.R 2.Q

* Duke Updated Dam Failure Evaluation#Energy 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 recognizedincluding materials and method/quality of construction

,:, Overtopping and Seismic are confirmed from the 2011 SE asnot being credible failure modes

For Information OnlyTwe un ,

P Duke Updated Dam Failure EvaluationVEnergy Fukushima 2.1

Overtopping of the Jocassee dam was confirmed not to be a credible failure mode

*:o The Jocassee dam and dikes include 15 feet of freeboard

o:o The Jocassee watershed is small relative to storage capacity - 148 square miles

*:o The top of the spillways are located at 1110 (full normal level)

ý Four diverse methods of assuring spillway gate operation

ý Rigorous spillway gate maintenance and surveillance testing as required andmonitored by FERC

*:o Lake management procedures require consideration of lower level to anticipateadditional storage needs for significant storms

ý Weekly rain forecast are prepared by Duke Energy to project rainfall for the basin

ý Precipitation monitoring has assured that no overtopping of the spillway gates hasoccurred in 40 + years of operation

*:o PMF using current HRR-51,52 results in 3 feet of freeboard margin

*:o 2011 SE also concluded that overtopping was not credible

For Information Only1 UII•itho I 4111 I V GFRi. .

S DUke Updated Dam Failure EvaluationVEnergy Fukushima 2.1

Seismic Failure of the Dam was confirmed not to be a credible failure modeo:, 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 a0.1g horizontal ground acceleration).

*:o 2007 Updated Fragility AnalysisHigh 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, arespected 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 yearperiod is 0.191 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.

ý The combination of the updated seismic fragility with the seismic hazard curve results in a negligible riskcontribution from seismic events.

ý In a letter dated 11/20/07 and in the 1/28/11 SE report, the NRC concluded that there is a negligible risk

For Information OnlyWiVllLihJ fiui._ Pdbch. DIcosu.viw1ef414FR+

9e Models ConsideredRegression Analysisw.fnergy

ý Froehlich 2008

ý Walder & O'Connor

ý MacDonald & Langridge-Monopolis 1984

ý Xu& Zhang 2009

For Information Only!h f um P ,•lluure uner 10 CFR 2.30

R DUke Selection of Xu & Zhang 2009lEnergy Bsis

*:, Most current regression method developed and validated withthe largest data base of dam failures:

ý 182 earth and rockfill dam failures compiledý 75 failures w/ sufficient info to develop breach regression models

.:. Empirical formulas that account for physical characteristics ofdam/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

For Information OnlyW•thhold eo-Publi isc7osure un-der iU iIK IM.3

IDuDkePOF~nergyBreach Parameters

Fukushima Update

Jocassee Dam - Xu & Zhang

ý Starting reservoir elevation 1110 (normal full pond)

ý Rockfill dam with low erodibility classification

ý Piping failure initiating at 1020 feet msl (Sunny Day Failure)

ý Breach parameters:/Top Width - 701' (39% of overall crest)v/ Bottom Width - 431'/ Bottom Elevation - 870'/Breach Formation Time:

A Xu & Zhang - 29.2 hrs.(1 3.2 hours piping +16.0 open weir)

, Froehlich - 16.0 hours (open weir)-/Peak outflow: 1,760,000 cfs

For Information OnlylI IiIi~ f"I uIIIIIU P c r uJ i~scosure Underld C 2-

Puke,nergy

Jocassee DamLow Erodibility Classification

(b)(7)(F)

mor inormauon unly

DukeEnergy Fukushima Model

V,

JOCASSEE DAM BREACH PARAMETERS

Reservoir BreachCrest Bottom Breach Bottom Average Time to Top of

Structure Elevation Starting Failure Mode Elevation(D BreachWidth Breach Right Side Let Side Failure Breach Breach Initiation

(a msl) Elevation ( msl) (fa) Width (0) Slope (i) Slope() (r) Width (at) Progression Elevation(ftms!) Imsl)

1125 1,110 P0i4g 870 431 566 0.53 0.53 29.2 701 SieWave 1,020Dam

Breach Formation TimeXu & Zhang definition:Froehlich definition:

29.2 (13.2 hours piping + 16.0 hours open weir)16.0 hours open weir

For Information OnlyWithhold from Public D!losure unaer it CFR 2,39-

luke Fukushima Model Jocassee Dam Breachrnergy Progression and Stage-Discharge Hydrographs

Breach Formation Time; Xu & Zhang definiton: -29.2 (13.2 hours piping + 16.0 hours open weir) Froehllch definition: -16.0 hours open weir

For Information OnlyWithidfrom. Publi Difkyu e Ud 10 CFR 2.390

Duke Breach Parameters(lEnergy Fukushima Update

,:, Keowee Dam

ý Starting reservoir elevation 800 (normal full pond)

ý Homogeneous earth fill dam

Overtopping failure trigger of two feet over the crest at 817 msl byrapid rise of Keowee reservoir over the crest

Multiple simultaneous breach initiation formation points across theKeowee dam and West Saddle dam

o:* Cascading dam/dike failure on Keowee

ý Keowee main dam-0.75 hrs

ý West Saddle Dam. 0.5 hrs (shorter than main dam, ratio of height)

For Information Only•ihoHfom Piihlin flki. sIr, l n dcr 10'C," 2. "

PDukeOihergy Fukushima Model Keowee Dam

Breach Progression HEC-RAS

0.9

0.8 .

0.7

• 0.6 i. ...

0.5

"0,.4C3

0.2

0.1

00 0.1 0.2 0.3 0.4 0.5 0.6 0,7 0.8 0.9

Relative Time Progression

For Informaton Only"ucDisc osureun.

Puke"nergy Fukushima ID Modeling

Keowee Dam -Headwater and Tailwater Stage HydrographsFinal BEP LE 1-D Model Performance

830

820

810

800

790

780

770

- 760E

750

S740

730

720

710

700

690

680

670

660

6500 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36

Model Time - hours

-BEP LEHW -BEP LE TW

For Information Only0Mý IMCSCý. ,

Fukushima 21 2D ModelingKeowee Dam Breach Progression

1

0.9

0C8C

-i 0.790.6

I0.5

2 0.40.3_m0,5-

02 -

0.1

016.2 16.3 16.4 16.5 16.6

Time (hr)16,7 16.8 16.9 17

For Information Only-VllUl. U rom ull DisdusumUit: • iJ 1 FR H90'

Dukenergy

Fukushima 2DModeling Velocityand Flow Pattern

at l7 hrs.

t: 17.0Val. Mal

0 2 4 6 8 10121416182022242628303234363840WRTý

For Information Only"Winhhoia from rubIk. Didusoure unde, 10 C;R 2,39 -

PDukeUEnera

Fukushima 2DModeling Velocityand Flow Pattern

at 2O hrs.

0

viý

2 4 6 8 10121416182022242628303234363840

For Information OnlySltý c Discoueudrt

PDukelEnergy

Fukushima JD-2DModeling Results

- :4

Breaching

Keowee Dam Intake Dike

HEC-RAS 2-D HEC-RAS 2-D

Elevation Decimal lime 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 DecimalTimejElevation Decimal Time

818.4 16.53 820.1 16.58 10 17.17 807.2 17.67

Maximum Water Surfaces

Swale Tailwater

HEC-RAS 2-D HEC-RAS 2-D

Elevation DecimalTime Elevation Decimal Time Elevation Decimal Time Elevation Decimal Time

817.5 16.55 815.51 1653 17.52 790.4 18.41

For Information OnlyWiToi trom uu Dbcli'r e 10 F"

~u.KenUrgwiergy Sensitivity Analysis

Model Peak Oufflow (cfs)McDonald & Langridge-Monopolis 1984 1,566,381Costa, 1985 1,634,480Xu & Zhan 2009 1,760,000Evans, 1986 1,803,331SOS, 1981 2,647,711Bureau of Reclamation, 1982 3,046,462McDonald & Lang ridge-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

For Information OnlymAml$1 l l i m i

Ivihonwa TromTTubuc MEMwOur under 10 CrK 4,19

N Independent ReviewBreach Parameters,nergy

I Independent Peer ReviewJoe Ehasz, P.E.David Bowles, Ph. D P.E. P.H.

, FERC Board of Consultant ReviewGonzalo Castro, Ph.D., PRE.James Michael Duncan, Ph.D., PRE.James F Ruff, Ph.D., RE.Gabriel Fernandez, Ph.D., P.E.

For Information Only- °un e e

b Duke Comparative AnalysisrEnergy Large Modern Dam Failures

.', Taum SaukOvertopping failure initiated by human error (previous overtopping events had occurred)

> Random rockfill embankment supporting the inner concrete liner loosely placed by end dumping the materialwithout compaction except for the top 16' of 84' height

> The embankment was constructed on a very steep downstream slope of 1.3H to IV with a 10 high concreteparapet wall along the crest of the dam

> Embankment was highly erodible and contained over 45% sand sized material (also evident in unusuallevel of surface erosion from rain events)

*:. Tetonearthen 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 ofsand, 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 passedthrough the construction diversion tunnel

> After overtopping of the core started, the dam took 26 hours to complete the breach and empty the upstreamreservoir For Information Only

* .ub ....... 190

' Duke Modification ScopeEnergy 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):

ý Transformer relocation

ý Diversion walls and drainage canals

ý Aux building and Turbine building protection

For Information Onlyit.'l , €Idfrom .P Wdhlisclosuwro• ,, G. d FR1C .

lukeInergy

Modification Options

)*,. Jocassee Dam

•'• 7p ...

fl,-,./' . . ,

-N

*.1

4-

7/

/i

For Information Onlyil

v"znnOgla -1:F F - Im reu 1 w m6C4 JI i i C" a*apar 10 C rz -? _

DukeEnergy

Questions and Feedback

For Information OnlyWitkk-I,,l~t- - u l der~teFR ...... J