ML20140E885
| ML20140E885 | |
| Person / Time | |
|---|---|
| Issue date: | 02/26/1991 |
| From: | Murphy A NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| Shared Package | |
| ML20007G200 | List: |
| References | |
| FRN-57FR47802, RULE-PR-100, RULE-PR-50, RULE-PR-52 AD93-1-006, AD93-1-6, NUDOCS 9705010195 | |
| Download: ML20140E885 (3) | |
Text
{{#Wiki_filter:....._._.2 . . , _ _ _ . ._ ~ . _ _ . . _ . _ _ . _...__ . _ _.- . _ , ~ . _ . . _ . . .
. ~ . . _ _ . _ . . . - _ . .- .
i er. . ** U4 O 4,' f UNITED STATES 1
'g.
; NUCLEAR REGULATORY COMMISSION ]
y; (?*
; Y'. L. 3' '. j _
WMmNC. TON, O C. 2W,55 ;
, a j
" . .>...J }
i 1 l FEB 2 c;199; f MEMORANDUM FOR: Distribution
- FROM
- Andrew J. Murphy, Chief -r )
i Structural & Seismic Engineering Branch j Division of Engineering, RES l
SUBJECT:
PUBLIC MEETING ON THE REVISION OF 10 CFR PART 100, APPENDIX.A L 1 A meeting with NUMARC is scheduled for 1:30PM on March 6,1991, to discuss the schedule and the technical topics for potential inclusion in the 'l l revision of Appendix A. A copy of the proposed agenda is attached. , 1 I i l l l l Y i 1 yj )
., ,. -w f ,
Andrew J. Murphy, Chief ! Structural & Seismic Engineering Branch Division of Engineering, RES
Enclosure:
As stated cc: SSEB l l i r l 9705010195 970422 PDR PR - l 50 57FR47902 PDR . d r'+ %w - -.gg- - - 14, m -. q- . - - - y.-og
- . -. _ .. _ . - - - . . = _ _ . . ._ .
l I oc l I j I AGENDA PUBLIC MEETING ON REVISION OF 10CFR PART 100, APPENDIX A l 6 March 1991 1:30 - 3:30 PM l 1:30 INTRODUCTION and NRC SCHEDULE SHAO/ MURPHY 1:45 NUMARC COMMENTS ON THE REVISION TO APPENDIX A NUMARC 2:00 TOPICS FOR POTENTIAL REVISION ; IDENTIFIED BY NRC KENNEALLY/MCMULLEN l I 2:45 OPEN DISCUSSION ALL l 3:30 ADJOURNMENT l Room 013 5650 Nicholson Lane Rockville, Maryland 20555
Contact:
A. Murphy (301-492-3860) . f
- s. ='
F FEB 2 c 199: DISTRIBUTI0ft:
". Taylor, ED0 P. Jehle, OGC G. Bagchi, NRR D. Jeng, NRR R. Rothman, NRR ,
P. Sobel, NRR A. B. Ibrahim, NMSS C. K. Chou, LLNL D. Bernreuter, LLNL R. Murray, LLNL M. Witte, LLNL i l l l l l 1 l
, # jo,, UNITED S TATES
~
8 )g g NUCLEAR REGULATORY COMMISSION PPS p WASHINGTc N, D. C. 20555 's
%,[l APR 2 31993 HEMORANDUM FOR: Lawrence C. Shao, Director Division of Engineering, RES FROM: Andrew J. Murphy, Chief Structural & Seismic Engineering Branch Division of Engineering, RES l
SUBJECT:
SUMMARY
OF PUBLIC MEETING ON THE REVISION OF APPENDIX A TO 10 CFR PART 100, " SEISMIC AND GEOLOGIC SITING CRITERIA FOR NUCLEAR POWER PLANTS" On April 17, 1991 the NRC staff met with the staff of Nuclear Management and Resources Council (NUMARC) and other industry representatives to discuss the , revision of Appendix A to 10 CFR Part 100. The meeting was held at the Hyatt l Regency, 1 Bethesda Metro Center, Bethesda, Maryland. Enclosure 1 is a list of I attendees. The principal objective of the meeting was to provide NUMARC, other industry representatives, and other interested members of the public with an opportunity 1 to express their views on the Appendix A revision. This meeting was a follow-up to a meeting held on March 6,1991 where the NRC staff discussed schedule and technical topics related to the Appendix A revision. After my brief introductory remarks I turned the meeting over to Orhan Gurbuz, NUMARC to provide an outline of topics to be discussed and to start the presentations. The following presentations were made:
- 1. NUMARC Perspective on Appendix A by 0. Gurbuz (Enclosure 2),
- 2. Proposed Revision of 10 CFR Part 100, Appendix A by J. C. Stepp (Enclosure 3),
- 3. Preliminary Recommendations on Seismic Design Aspects of Appendix A by M. Hayner (Enclosure 4),
- 4. Overview of Regulatory Guides by F. Swan (Enclosure 5), and
- 5. Position on Operating Basis Earthquake in Appendix A to 10 CFR Part 100 by J. Reed (Enclosure 6).
i The staff was informed that the views expressed by NUMARC/ industry were preliminary; they are planning to submit a written document to the staff in approximately one month. A clear message from all presenters was that the regulation (Appendix A) should be simplified; discussing only requirements, the guidance should be placed in regulatory guides. In general, NUMARC/ industry did not present any information that the staff had not already identified as a topic needing attention. The staff is carefully
- _ Q ,lm G e '.f Ik3 %hf
APR P 31931 , o reviewing their suggestions. It is obvious that they have worked very hard on these issues and their effort is recognized by the staff. They understand that the "predecisional mode" that the staff is working under limits the amount of information that we can share with them regarding our positions. However this , is not discouraging them, and they expressed an interest in having additional ! public meetings to discuss specific technical topics. ' th$w ' Andrew J. Murphy, Chief ; Structural & Seismic Engineering Branch Division of Engineering, RES
Enclosures:
As Stated (6) CC: Attached Distribution List l k 4 i T l
l t
- Distribution List, Memorandum Dated: MR 2 31991 N. Taylor, EDO E. Beckjord, RES T. Spets, RES C. Heltemes, RES T. King, RES C. Ader, RES L. Soffer, RES l
N. Chokshi, RES I R. Kenneally,sRES 2 l R. McMullen, RES~~ l E. Zurflueh, RES l J. Costello, RES G. Bagchi, NRR R. Rothman, NRR P. Sobel, NRR ! l D. Terao, NRR .
- J. McIntyre, NRR i H. Brammer, NRR I A. Ibrahim, NHSS .
! P. Jehle, OGC ! l R. Ng, NUMARC O. Gurbuz, NUMARC C. K. Chou, LLNL l D. Bernreuter, LLNL l R. Murray, LLNL ! M. Witte, LLNL P. Prassinos, LLNL , D. Chung, LLNL l l Public Document Room l l i
ENCLOSURE _1
.____._....____J.. .
.9... . . _ . _ . _ _
.._.. . [ __. _. _ _ _ . _ . _
.___.______.__..__-._____..f.___._._ _ . . . _ _ . . _. _ _ _ _ _ _ _
<u i .
kM2~- . kC .L.-._._. J. i .._... - -.. _. -__.. ._. . _. . _. -. .. . . .
~
k/c Sw.cc & QN/_ & i.s h z 5 n [ ___l_~ sy %.D,-- L n ec. k es 0 0.m_2 ns .._ NilU.bu.C31ok. ski '
.p gc}g e.S J S o Q A 9 2.__.s 1 L la
'(_3 0 S4 Q 3 k o I i Awd . Muy .lMRc/ RGS l'
._ ._ IA H . . _ _
l .L.LN A.' .J_ G ff3 -It2/ - o2. M '_. JeM._. /t . - _ __ H/46 . #A6"" (To r_)-??t-J 7//_.- .
)8Md?~~_.Arojyj c._S~o 9 - 9?9 - 6 ~/_// '
._ //Vn l.__bf61'~6Vt A '
-__. Boe __. c}Ecr2/2 s7x c ._1 Btf) -Str&.-F.282L_
. 43 .h. Cu f M .._ . . ..__ ___._ _ _. ._ ETEC._ .. . . _ . . . . _.
MtM% - 52.%,s _ I6W trA GLu SY_ _ _._ _ .._..__._...(.TrEC.__._ _ _ . _ . . 9.iEd 58 6- SM). _ . . . 4sa McL e n_ . wec./vez.____. h(o,4 492-cr7 t s._ . _. Daotc(~TE,,ao - __NR/NRR_. ___ 0192-3 s./7 .._
._..//2..)64twAd__ Nac,a/#fl._. _ . _...._(m ._(3o1) &&-0 776_ ..__
_ _ _ b M ia Ami c.k. _ M se.o ) 855 . T Go o _._ ott@ *15 .. (a s t) d h o T M _ . _ _ _. _. .E A w_A. Ida.4* L<w. ' i_..f.4.1Ro7WM4M' i i hec./h W ' { Sot)_A31.-330.6 _ _~ N lli s . S h M _._ _ _. _.Nkc/N M '3019f 2-r#_~73E.
..90YA&Jh4GCNC _.. LNAC/M&.__ _._..__hbl h 2.-A7 3 3 .--
T .-
-. - . M TN .-.- . -. - D O E- - .- . {50/J 3S3- 4 %'f ._
. .. . Evs .. b.5_._.-__. ,.......dedleb 7. _. _-.__.. ' yai. ..Vs 7 -Jo 7 Y. .- .
Mike. ScH P MA4 _ _ . ._ . .. pq a s2 - 2s ee . 1$0 F'P N L, N SON -.__. (5/0) 302" S70Y OCAWIS . . . _ . _ . .
. _ . ] W 'l~5 W L E.- -. .. - .._. - . N % N O. G.C. -.
.(.50l) 3.92 -sT FS"
~J ~= g e .~P % r _. . ._.___. 3) M: 5 B _ _ ______ .. ..(Ao.R). De 6 ST3 .
. ! klif$ Wubjecler Wak/ESA___ -..LtN)NS-WYo..._ ..
l . . E *.? '..M ! M.M_^ % A_._.. _
...._.3?.sroM{.ncar_MN_2(a*U_N Ydb .. __. _ ._ _ .. N e V B-oi n oI_
_. D O E 2o9. - 6~1Tb # 733E
._ . f o d " r . E
- Y.e .k ___
chenpc Lw<- Nus. mo sse-mi A %. . _ . . . _ _ NumAa c _ __.2o., s y - uero . . WE - 3-e mer __ _ 84b SC'. _... . .{2/6 . 20Y- 575 ..
4 J L - ii Ji --.-4 . + + . +~. _ -.A--.m,.h4 a .4 8 e i I ,
- - - - 1. . _ . . :
I j/ ....--4 .- . ___ _. .. , . l,___._,.__._.. , \ .e a _ s .. n - 1
' I
~
.I .
f h_. Boe OHotLTotD _ hH CArtoue EUne_i. Gas !
._ sol 34s Arzs._ !
, . _ .~F~/4 6 4 . -. . .-_. GermJny (j,oe/bnf ; wr-stne-ym. , O Cv v2L -
- Numimc-. : 4cz-8n-aro-0g -- .. - . . -. w u m e.c..-. _ ..(za). w-ase
<ja , -7;-eace-y sa c+ee-ysm a aima -rn2
._ . .- . _._. _ . _ . __ . _ _ . _ . l
. .M . ._ _ k.._ . e w: SW8 k_ '
h_ h_ T) h._N...ll- . _ .. (fe.e } - h9 -
. _ _ _ _ _ _ _ . .,. icd l I i e t '
t , I 4- . -- _ _
-* e. -1 h ese $ .e 9.h.pe.
*"" *.'N - **~ -e...w w. 4.ww w. ...w.
E
.- -.- .. .. ..~ _ _._ ._ . . .. _ . . _ . . , _ . . . _ . . . . . . . _ _ . _ _ _ _ . . _ _ . , . . . _ . _ .
i _.-_ . . . . _ . _ . . . . _ _ . _ _ _ _ _ _ _ ;..._ _ _ .._ _ _. _ . _ __ _. ...__.._,.__._._.._,] l
. _. . . - ..-.-A '
t
-.. - - - _ _ . _ . . _. _ _.. _ _ . . . __a l i
_4 l i l h ? .
. , i i I ' 1
[
! I l s , i l I
~. -....s h 8
, . , . t
. .. ,e-e.-_oe.- .g 6
; e i
.-. - - ._- _ . . _ . w -. ,
8 4 , f
. . . - - . , , . ,m .. _ .. .m._ . . . _ . _ _ . . . . . . . _ .. . . --- .-- .
t
~.f l l !
4 I , f .- - -- ._ _ +_ .. .-------a-t
+
i. i ; .
-. A . ._ _ _ . .
* - - - - - - J h
- . - -. >. e.-... -w.ea-...-
Ne M & " " 8 **
- e. 6.-4 bm@*Wh&W . *=*e.. -ab.4, -e -
.4 . .mW . .g e -..-.ege. ..ap. '.f*.*. ._.N.em- e
+ w .a.== e .g.w e.- .e -ee .m.m.--.m.eg We.-- - = =** **h **" ***'""' * *"' *** * * ^ * *
. Mh -ge Mw.-eh. gp h Dp+h.i. .We.*W** *^***+ *
*-'MM. ..em . sw $.g. . ...e.A~ 4-_e.w- e. M M 4 e t.+e. Name-.-e- M'. .
k
ENCLOSURE 2 - I i I i NUMARC PERSPECTIVE ON APPENDIX A E 4 4 .i
)
!' BY ! i i ORHAN GURBUZ i A. k j
- PRESENTED TO j NUCLEAR REGULATORY COMMISSION 4
APRIL 17, 1991 i l 1 d
OUTLINE o OVERVIEW 0. GURBUZ o APPENDIX A REVISIONS C. STEPP o CHANGES TO 10 CFR 50 M. HAYNER o REGULATORY GUIDES F. SWAN o OBE TREATMENT J. REED o DISCUSSION AND
SUMMARY
ALL t i 1
l 0VERVIEW l l I o PRIMARY INDUSTRY CONSIDERATIONS o RECENT NUMARC ACTIVITIES o PRELIMINARY NUMARC RECOMMENDATIONS o REGULATORY GUIDES i i l a l 2
~
l i .- i l. PRIMARY INDUSTRY CONSIDERATIONS o ASSURANCE OF " PROSPECTIVE" APPLICATION i o PLANT SHUTDOWN AND RESTART CRITERIA o CONTENT AND DETAIL OF: 1 REVISED APPENDIX A DESIGN PROVISIONS IN 10 CFR 50 REGULATORY GUIDES o NOT CLEAR IF ALL ESSENTIAL RG's WILL BE ISSUED FOR COMMENTS o TIMING AND CONTENTS OF CHANGES TO OTHER DOCUMENTS: SRPs EXISTING REGULATORY GUIDES CODES AND STANDARDS 3
i I i ! RECENT NUMARC ACTIVITIES i i o FORMED AHAC - NOV 1990 i f o AHAC REVIEWED RELATED ISSUES, IDENTIFIED OPTIONS i
- o SIWG FAVORS OPTION C, " SIMPLIFY APPENDIX A AND l ISSUE SUPPLEMENTARY REGULATORY GUIDES" i
- o NUMARC IS CURRENTLY PREPARING INPUT TO SUBMIT TO l i
NRC: l l PROPOSED APPENDIX A i i - POTENTIAL CHANGES TO 10 CFR 50 l - OUTLINES FOR TWO KEY REGULATORY GUIDES i POSITION PAPER ON OBE TREATMENT l i i i l j i
1-I PRELIMINARY NUMARC.RECOMENDATIONS i i ) o SIMPLIFY APPENDIX A: i l - INCLUDE ONLY "WHAT" NEEDS TO BE DONE TO
- i ESTABLISH SITE SEISMIC REQUIREMENTS o DEVELOP REGULATORY GUIDES TO ADDRESS "HOW" i o DESIGN RULES TO 10 CFR 50, INCLUDING PLANT
! SHUTDOWN & RESTART CRITERIA o ENDORSE EPRI REPORTS:
- NP-5930, "A CRITERION OF DETERMINING
! EXCEEDANCE OF THE OBE" l
- NP-6695, " GUIDELINES FOR NUCLEAR PLANT
! RESPONSE TO AN EARTHQUAKE" l l l 0 ELIMINATE OBE AS A DESIGN REQUIREMENT ! o APPENDIX A LANGUAGE: i NON-PRESCRIPTIVE " CONSIDERATIONS" WHICH l
- " GUIDE" THE COMMISSION IN ITS EVALUATION OF i
! A SITE l i MINIMIZE AMBIGUOUS LEGAL INTERPRETATIONS l i l AVOID CODIFYING EXPERT OPINION I l 1 AVOID CODIFYING STATE OF ART i i l
! 5 i
NEW REGULATORY GUIDES ; GEOSCIENCES:
- 1. DETERMINATION OF DESIGN BASIS GROUND MOTION
- 2. DETERMINATION OF POTENTIAL FOR TECTONIC SURFACE DEFORMATION
- 3. DETERMINATION OF SITE GE0 TECHNICAL DESIGN PARAMETERS
- 4. ASSESSMENT OF POTENTIAL FOR VOLCANIC HAZARDS
- 5. ASSESSMENT OF TSUNAMI AND WATER WAVE HAZARDS DESIGN:
l i l 1. ACCEPTABLE METHODS FOR SEISMIC DESIGN l 2. DESIGN FOR SEISMICALLY INDUCED FLOODS AND WATER l WAVES ) 3. SOIL-STRUCTURE INTERACTION ANALYSIS i 4. PLANT SHUTDOWN AND RESTART CRITERIA l 1 I 6 1
)' l ENCLOSURE 3
- j. .
i c. i 4 i I l PROPOSED REVISION i 0F l 10 CFR, PART 100, APPENDIX A ! SEISMIC AND GE0 LOGIC SITING CRITERIA ! FOR I NUCLEAR POWER PLANTS l PRESENTED TO U.S. NUCLEAR REGULATORY COMMISSION BY J.C. STEPP l APRIL 17, 1991
i
- i. -
J
~
l' REVISION OF APPEND 1X A i 1 .i o PARTICIPANTS C. CHEN - GILBERT i
- 0. GURBUZ - NUMARC l
- 1 l -
M. HAYNER - CEI T. O'HARA - YAEco ) D. OSTROM - SCE 1 - C. STEPP - EPRI I 1 F. SWAN - GE0 MATRIX i i - R. WHORTON - SCE & G 4 1
~ ___. . . _._ _ _ _ _ ._ _ _ _ _. _ ._ __ __.
i i REVISION OF APPENDIX A i 4 ! o OBJECTIVE l PROVIDE STABLE REGULATORY BASIS l i FOR FUTURE NPPs . l - AVOID LICENSING DELAYS DUE T0 i UNCLEAR REGULATORY i REQUIREMENTS i PROVIDE FLEXIBLE STRUCTURE TO
- PERMIT CONSIDERATION OF NEW l TECHNICAL UNDERSTANDINGS l
PROVIDE REGULATION IMPLEMENTATION BY ALL - TECHNICAL, LEGAL ) 1 <
- 1 4
l l 2 j
!. ) i REVISION OF APPENDIX A ! STRATEGY i I i i o SIMPLIFY REGULATION REQUIRED DETERMINATIONS ONLY i i REMOVE TECHNICAL STATEMENTS j REQUIRING INTERPRETATION i i ADDRESS ONLY SITING AND SEISMIC l DESIGN BASIS DETERMINATION l REQUIREMENTS MOVE ALL SEISMIC DESIGN ! REQUIREMENTS TO 10 CFR PART 50 IMPLEMENT REGULATION AND l REGULATORY GUIDES i DEFINE REQUIRED GUIDELINES l l IMPLEMENT STATE-GF-ART ! PRACTICE 1 j 3 i m
i l' REVISION OF APPENDIX A l STRATEGY (CONT'D) i l o SIMPLIFY REGULATION - CONT. ALLOW USE OF PROBABILISTIC l METHODS STRUCTURE REQUIREMENTS TO j MINIMIZE REQUIREMENTS FOR i UNNECESSARY INFORMATION i USE GRADED APPROACH DEPENDING
- ON INFORMATION NEEDS TO MAKE i DETERMINATION CROSS REFERENCE TO AVOID DUPLICATING INFORMATION l -
REMOVE OPERATING BASIS l EARTHQUAKE AS A DESIGN i REQUIREMENT i i i j i
! 4 i ! I
i i. 1 REVISION OF APPENDIX A i SCOPE OF REVISED APPENDIX i [ I. PURPOSE 1 ! II. SCOPE l III. DETERMINATION OF THE POTENTAIL l FOR TECTONIC SURFACE DEFORMATION i i IV. DETERMINATION OF DESIGN BASIS l GROUND MOTION ! V. DETERMINATION OF DESIGN l PARAMETERS FOR OTHER SEISMIC AND l GE0 LOGIC HAZARDS i l l l 1 1 ! 5 i l 1 A
i 4
- REVISION OF APPENDIX A l l 1
l o SCOPE l l REQUIREMENTS FOR SITING AND
- DETERMINATION OF DESIGN BASES
i SEISMIC AND GE0 LOGIC HAZARDS I l SEISMIC DESIGN REQUIREMENTS MOVED TO PART 50 ) i J j l i I i l l I I i ! 6 1 a
1 4 i- . l i-i REVISION OF APPENDIX A i o DEFINITIONS i l - SITE - LOCATION OF POWER PLANT i STRUCTURES SITE LOCALITY - R LESS THAN 1 KM l l - SITE AREA - R LESS THAN 8 KM l - SITE VICINITY - R LESS THAN ! 25 KM i l - SITE REGION - R LESS THAN 150 KM .
- I TECTONIC SURFACE DEFORMATION - A l GENERAL TERM FOR FAULTING OR ;
! FOLDING AT OR NEAR THE GROUND l i SURFACE DUE TO LARGE SCALE l TECTONIC FORCES IN THE UPPER
- PART OF THE EARTH'S CRUST i
4 1 7
- l
i i i REVISION OF APPENDIX A ~ o DEFINITIONS (CONT'D) i DESIGN BASIS GROUND MOTION i (DBGM) i ! THE FREE-FIELD VIBRATORY GROUND ! MOTION DETERMINED AT THE SITE, i CONSIDERING THE SEISMIC SOURCES ! IN THE SITE REGION AND SPECIFIC ' l GE0 TECHNICAL CHARACTERISTICS OF i THE SIT 7. SUBSURFACE MATERIALS. i l IT IS THAT GROUND MOTION FOR , i WHICH THERE IS REASONABLE l ASSURANCE OF A LOW LIKELIHOOD OF ! BEING EXCEEDED. I i ' i 4 I 1 i f
i ! REVISION OF APPENDIX A i l o REFINITIONS (CONT'D) SEISMIC SOURCE A GENERAL TERM REFERRING TO A i GEOLOGIC STRUCTURE OR REGION OF j THE EARTH'S CRUST WHERE j EARTHQUAKES ORIGINATE i i i l i 9
i - l
- l REVISION OF APPENDIX A i
I ! DETERMINATION OF SITE GE0 TECHNICAL ! DESIGN BASIS PARAMETERS 4 ! o PURPOSE DEVELOP DESIGN BASIS PARAMETERS ! FOR FOUNDATION AND FOUNDATION-l STRUCTURE SYSTEM FORMULATE PROCEDURES AND ACTIONS TO MITIGATE ADVERSE IMPACTS ON CONSTRUCTION AND OPERATION OF A NPP 1 1 10
- l. .
l.- l REVISION OF APPENDIX A l t l DETERMINATION OF SITE GE0 TECHNICAL j DESIGN BASIS PARAMETERS (CONT'D) o SCOPE OF DETERMINATIONS I i 3 i - FOUNDATION EXCAVATION AND DESIGN j BASIS PARAMETERS LIQUEFACTION AND GROUND j SETTLEMENT UNDER DBGM LOADING i SLOPE STABILITY UNDER DBGM ! LOADING SUBSURFACE CAVITIES AND JOINTS GROUND WATER CONDITIONS SLOPE STABILITY AND SETTLEMENT i UNDER STATIC LOADING i GE0 METRIC DISTRIBUTION OF ! FOUNDATION MATERIALS AND THEIR DYNAMIC MATERIAL PROPERTIES l 11
i l REVISION OF APPENDIX A 3 l DETERMINATION OF TECTONIC SURFACE ! DEFORMATION l l o PURPOSE 1 i l - DETERMINE NEED FOR AND DEVELOP
- DESIGN BASIS PARAMETERS FOR EFFECTS OF TECTONIC SURFACE DEFORMATION l
12
i j . i l REVISION OF~ APPENDIX A l i ! DETERMINATION OF TECTONIC SURFACE j DEFORMATION o SCOPE OF DETERMINATIONS ASSESS TECTONIC FRAMEWORK OF THE i SITE REGION ASSESS RELATION BETWEEN i SEISMICITY AND GE0 LOGIC l STRUCTURE WITHIN SITE VICINITY i IDENTIFY AND ASSESS GEOLOGIC i STRUCTURES WITHIN THE SITE AREA l THAT HAVE A POTENTIAL FOR ' TECTONIC DEFORMATION THAT COULD ADVERSELY AFFECT THE DESIGN l BASIS OF A NPP i
- o PERMITS USE OF PROBABILISTIC l APPROACHES 4
1 I l 13 4 m
i !" REVISION OF APPENDIX A i i DETERMINATION OF DESIGN BASIS GROUND ! MOTION (DBGM)
- o PURPOSE DETERMINE DBGM
! o SCOPE OF DETERMINATION i SEI3MIC SOURCES WITHIN THE SITE l REGION i EARTHQUAKE POTENTIAL FOR EACH l SEISMIC SOURCE l l VIBRATORY GROUND MOTION j TRANSMISSION IN THE SITE REGION i i SITE AREA GEOLOGY AND SITE i DYNAMIC GE0 TECHNICAL PARAMETERS VIBRATORY GROUND MOTION AT SITE i FROM SEISMIC SOURCES IN SITE ! REGION l ) o PERMITS USE OF PROBABILISTIC l APPROACHES . 1 j 14 J
l.. i 1 l REVISION OF APPENDIX A ! DEIERMINATION OF DESIGN REQUIREMENTS l FOR OTHER SEISMIC AND GEOLOGIC HAZARDS 0 PURPOSE l l DETERMINE NEED FOR AND ESTABLISH l DESIGN BASIS PARAMETERS FOR I OTHER SEISMIC AND GEOLOGIC HAZARDS o SCOPE OF DETERMINATION SEISMICALLY-INDUCED FLOODS AND WATER WAVES (TSUNAMI, SEICHE DAM FAILURE) VOLCANIC HAZARDS MAN-INDUCED SEISMIC OR GE0 LOGIC HAZARDS (SEISMICITY) SUBSIDENCE OR COLLAPSE 15
1 .. ENCLOSURE 4 I PRELIMINARY i RECOMENDATIONS ON SEISMIC DESIGN ASPECTS OF APPENDIX A l BY MIKE HAYNER l PRESENTED TO U.S. NUCLEAR REGULATORY COMMISSION APRIL 17, 1991
f OUTLINE o OBJECTIVES o APPENDIX A PROVISIONS THAT MAY BE MOVED TO 10 CFR 50 o TOPICS THAT NEED TO BE ADDRESSED IN 10 CFR 50 OR REGULATORY GUIDES
'l i
1
i, i 4 j .- i OBJECTIVES l 0 PROTECTION OF PUBLIC HEALTH AND SAFETY IS j PRIMARY OBJECTIVE ! o INCLUDE ONLY PROVISIONS THAT ADDRESS "WHAT" NEEDS TO BE DONE TO ACHIEVE THE PRIMARY OBJECTIVE l l 0 SEPARATE SITING AND DESIGN REQUIREMENTS ! o CLEARLY DELINEATE REQUIREMENTS FOR CATEGORIES OF l PLANTS SUCH AS: l
- CP BEFORE JAN 1, 199_
- CP OR COL AFTER JAN 1, 199_
- o ASSURE CONSISTENCY WITH AND SUPPORT OF EXISTING l SUBTIER DOCUMENTS i
j o KEEP IT SIMPLE, USABLE, TECHNICALLY BASED l ! o MINIMIZE POTENTIAL FOR DIFFICULT LEGAL i INTERPRETATIONS IN APPLICATION i l i
! 2
APPENDIX A PROVISIONS THAT MAY BE MOVED TO 10 CFR 50 o- DESIGN BASIS GROUND MOTION; REPLACES SAFE SHUTDOWN EARTHQUAKE o DESIGN FOR SINGLE LEVEL OF EARTHQUAKE; CHANGE FROM CURRENT DUAL EARTHQUAKE CRITERIA o PERMISSIBILITY TO DESIGN BEYOND YIELD STRAIN PROVIDED THAT SAFETY FUNCTIONS ARE MAINTAINED o THE REQUIREMENT THAT SEISMIC ANALYSIS MUST ACCOUNT FOR S0IL-STRUCTURE INTERACTION EFFECTS o DESIGN BASIS CONSISTENT WITH 10 CFR 100, APPENDIX A o PROVISIONS REGARDING COOLING WATER SUPPLY AND DISTANT STRUCTURES o PLANT SHUTDOWN AND RESTART CRITERIA l l 3 1
TOPICS THAT NEED TO BE ADDRESSED IN 10 CFR 50 OR REGULATORY GUIDES o DETAILS OF PLANT SHUTDOWN AND RESTART CRITERIA: ) ADOPT EPRI REPORTS USE "CAV" CONCEPT FOR FUTURE PLANTS, UTILITIES SHOULD BE GIVEN THE OPTION TO ESTABLISH SPECTRAL LIMITS FOR SHUTDOWN i o PROVISIONS REGARDING SEISMIC INSTRUMENTATION l o- DETAILS OF SOIL-STRUCTURE INTERACTION ANALYSIS , l o DETAILS OF ACCEPTABLE SEISMIC DESIGN METHODS: DYNAMIC ANALYSIS - EQUIVALENT STATIC LOAD QUALIFICATION TEST USE OF EXPERIENCE DATABASE 4
l t l l 1 IQPICS THAT NEED TO BE ADDRESSED IN 10 CFR 50 OR REGULATORY GUIDES (CONT'D) ) o OTHER DESIGN CONSIDERATIONS i DBGM IS SPECIFIED AT THE FREE FIELD SURFACE ANALYSIS OF BURIED STRUCTURES EFFECTS OF EXPECTED DURATION OF GROUND l MOTION DETAILS OF DESIGN FOR TECTONIC SURFACE I
- DEFORMATION FATIGUE CONSIDERATIONS IN DESIGN i
o EQUIPMENT QUALIFICATION o DESIGN FOR SEISMICALLY INDUCED FLOODS AND WAVE ! EFFECTS 1 i l 1 l i i 1 1 i , 4 j 5 l
1 ENCLOSURE 5 l l 1 l
, OVERVIEW OF REGULATORY GUIDES l
i i BY l F. H. SWAN l i \ I l PRESENTED T0: U.S. NUCLEAR REGULATORY COMMISSION APRIL 17, 1991 l t l 1
NEW REGULATORY GUIDES GEOSCIENCES:
- 1. DETERMINATION OF DESIGN BASIS GROUND MOTION
- 2. DETERMINATION OF POTENTIAL FOR TECTONIC SURFACE DEFORMATION
- 3. DETERMINATION OF SITE GE0 TECHNICAL DESIGN PARAMETERS
- 4. ASSESSMENT OF POTENTIAL FOR VOLCANIC HAZARDS
- 5. ASSESSMENT OF TSUNAMI AND WATER WAVE HAZARDS DESIGN:
- 1. ACCEPTABLE METHODS FOR SEISMIC DESIGN
- 2. DESIGN FOR SEISMICALLY INDUCED FLOODS AND WATER WAVES
- 3. SOIL-STRUCTURE INTERACTION ANALYSIS -
- 4. PLANT SHUTDOWN AND RESTART CRITERIA 2
i
1 i- ! i ).,- l s l DETERMINATION OF DESIGN BASIS GROUND MOTION i i
1.0 INTRODUCTION
i ! 2.0 REQUIRED INVESTIGATIONS i l 3.0 IDENTIFICATION OF SEISMIC SOURCES i ! 4.0 CHARACTERIZATION OF SEISMIC SOURCES ! 5.0 ASSESSMENT OF FREE-FIELD VIBRATORY GROUND l MOTION , i l l 6.0 DETERMINATION OF DESIGN BASIS GROUND MOTION i l l I i 1 1 l l l I 1 j i ) 3 1
. DETERMINATION OF THE POTENTIAL FOR TECTONIC SURFACE DEFORMATION :
1.0 INTRODUCTION
2.0 REQUIRED INVESTIGATIONS 3.0 IDENTIFICATION OF CAPABLE FAULTS / FOLDS (SITE AREA 8 < KM RADIUS) 4.0 CHARACTERIZATION OF CAPABLE FAULTS / FOLDS 1 5.0 EVALUATION OF GEOLOGIC STRUCTURE BENEATH THE SITE 6.0 ANALYSIS OF THE POTENTIAL FOR TECTONIC DEFORMATION AT THE SITE 4
\. - l i1 - ' KEY ELEMENTS OF PROPOSED REGULATORY GUIDES I i i o DETERMINATION OF DGBH IN A SINGLE REG GUIDE i THAT ADDRESSES THE ENTIRE PROCESS FROM SEISHIC SOURCE IDENTIFICATION THROUGH j SOURCE CHARACTERIZATION, ATTENUATION, SITE
- EFFECTS AND SELECTION OF DGBM
! o SUM 4ARIZES METHODS FOR BOTH THE EASTERN AND l WESTERN U.S. o SUM 4ARIZES CURRENT STATE-OF-THE-ART i DETERMINISTIC AND PROBABILISTIC APPROACHES i ! o GIVES GUIDANCE ON THE REQUIRED SCOPE OF l INVESTIGATIONS i ! o CLARIFIES AMBIGUOUS TERMINOLOGY AND CONCEPTS ! THAT ARE IN THE CURRENT VERSION OF APPENDIX j A l o EXPLICITLY ADDRESS THE TREATMENT OF i UNCERTAINTY j o MORE FLEXIBLE TO CHANGE; EASIER TO UPDATE j REG GUIDES TO REFLECT TECHNOLOGICAL ADVANCES i 1 l 5 l i 1- __ _
j .l ENCLOSURE 6 ) '- ; i l i i i i ! POSITION ON i ! OPERATING BASIS EARTHQUAKE i IN I APPENDIX A TO 10 CFR 100 1 . BY l l JOHN W. REED JACK R. BENJAMIN AND ASSOCIATES, INC. i PRESENTED TO U.S. NUCLEAR REGULATORY COMMISSION , APRIL 17, 1991 4 i i, ?
- INTRODUCTION 4
! o HISTORY OF OBE l o NEED FOR SECOND EARTHQUAKE AS DESIGN i BASIS i l o NEED FOR SECOND EARTHQUAKE FOR PLANT SHUTDOWN AND RESTART i o RECOMMENDATIONS FOR APPENDIX A TO 10 CFR 100 l 1
i i i 1 '. HISTORY OF OBE i ! o DEFINITIONS OF OBE IN 10 CFR 100 l APPENDIX A I o ORIGIN OF OBE i l 1 I l E e i l l l l ) l l
- 2 4
l; DEFINITIONS OF OBE IN 10 CFR 100 APPENDIX A t i o AN EARTHQUAKE WHICH, CONSIDERING i REGIONAL AND LOCAL GEOLOGY, i SEISMOLOGY AND SPECIFIC ! CHARACTERISTICS OF LOCAL SUBSURFACE ! MATERIAL, COULD REASONABLY BE EXPECTED TO AFFECT THE PLANT SITE
- DURING THE OPERATING LIFE OF THE PLANT I o OBE SHALL BE AT LEAST EQUAL TO ONE-i HALF THE MAXIMUM VIBRATORY GROUND
! ACCELERATION OF THE SAFE SHUTDOWN EARTHQUAKE l o THE OBE SHALL BE SPECIFIED BY THE APPLICANT 3
j .* j ORIGIN OF OBE ; i I 1 o IN LATE 1960'S DR. NEWMARK DELINEATED TWO EARTHQUAKE DESIGN APPROACHES: LOWER LEVEL " DESIGN" EARTHQUAKE: o EXPECTED TO BE EXPERIENCED IN REASONABLE PERIOD OF TIME o LATER CALLED "0PERATING BASIS EARTHQUAKE" LARGER " MAXIMUM CREDIBLE" EARTHQUAKE: o MAXIMUM POSSIBLE INTENSITY OF SEISMIC MOTION WHICH THE PLANT MUST SURVIVE o LATER CALLED " DESIGN BASIS EARTHQUAKE" o HOUSNER AND HUDSON ALSO PROMOTED MULTIPLE LEVEL DESIGNS 4 8 .- . - . . . . -.
ORIGIN OF OBE (CONTINUED) o ULTIMATELY A DUAL CRITERIA WERE INCORPORATED INTO 10 CFR 100 APPENDIX A UPPER LEVEL REQUIREMENT CALLED SSE LOWER LEVEL REQUIREMENT CALLED OBE o ORIGINAL INTENT WAS THAT THE SSE WOULD CONTROL DESIGN OF PLANT o OBE WAS TO PROVIDE A CHECK WHERE CONTINUED OPERATION IS DESIRED AT LOWER DESIGN LEVEL o OBE CONTROLS DESIGN FOR MANY l COMPONENTS 5
NEED FOR SECOND EARTHQUAKE AS DESIGN BASIS o SEISMIC PRA INDICATES LiEDIAN PLANT CAPACITY OF EUS PLANTS TO BE 2 TO 4 TIMES SSE o NO PROVERBIAL FAILURE " CLIFF" JUST ABOVE SSE LEVEL o MOST FAILURE CASES FOUND TO BE OUTSIDE DESIGN PROCESS (E.G., ANCHORAGE DETAILS AND INTERACTION ISSUES) i o NOT CLEAR WHAT SAFETY PURPOSE OBE SERVES IN DESIGN PROCESS ! 1
- . l I
I a i i I ' 6
\
- 1. -
i l NEED FOR SECOND EARTHQUAKE AS DESIGN BASIS ' (CONTINUED) l l ! o ANY ADDITIONAL MARGIN PROVIDED BY
- j. OBE IS UNINTENTIONAL AND VARIES:
i j - COMPONENT TO COMPONENT LOCATION TO LOCATION ! o DBGM PROVIDES SUFFICIENT MARGIN TO l SATISFY SAFETY CONCERNS i i o FATIGUE TESTING REQUIREMENTS COULD BE BASED ON DBGM INPUT i o ATTENTION TO DETAILS MORE IMPORTANT THAN DESIGNING FOR OBE l I d 7 ]
i. I j.- ! NEED FOR SECOND EARTHQUAKE FOR PLANT l SHUTDOWN AND RESTART i l o EPRI RESPONSE GUIDELINES (NP-6695) ! AND OBE EXCEEDANCE CRITERION REPORT (NP-5930) PROVIDE RATIONAL BASIS FOR
- RESPONSE TO EARTHQUAKES
PROGRESSIVE RESPONSE TO EARTHQUAKE l AhITIAL INSPECTIONS j - REVIEW SEISMIC PARAMETERS FROM i ACCELEROMETER RECORDINGS (SPECTRAL ! ACCELERATION AND CAV) ! ~ MAKE DECISIONS ON NEED FOR PLANT l SHUTDOWN AND READINESS FOR ! SHUTDOWN l o IF OBE REMOVED AS DESIGN BASIS THE EPRI OBE EXCEEDANCE CRITERION CAN STILL BE USED l i i 8
/
i j =.
- NEED FOR SECOND EARTHQUAKE FOR PLANT l SHUTDOWN AND RESTART (CONTINUED) l o SPECTRAL ACCELERATION CHECK CAN BE i ELIMINATED SINCE EXCEEDANCE REQUIRES l THAT BOTH THE SPECTRAL AND CAV i PARAMETERS EXCEED LIMXTS o EXPERIENCE-BASED SPECTRAL LIMIT CAN j BE USED AS SPECTRAL CHECK o TECHNICAL ARGUMENTS COULD BE MADE TO
! ESTABLISH AN EQUIVALENT OBE BASED ON l THE DBGM REQUIREMENTS o OPTION SHOULD ALWAYS BE PERMITTED T0 i PERFORM DESIGN FOR OBE USING CURRENT l REQUIREMENTS FOR PURPOSES OF j DETERMINING PLANT SHUTDOWN AFTER i EARTHQUAKE I 4
- 1 l
9 l
RECOMMENDATIONS FOR APPENDIX A TO 10 CFR 100 o THE OBE SHOULD BE REMOVED AS A DESIGN REQUIREMENT l o APPENDIX A SHOULD BE WRITTEN TO ! ALLOW, FOR THE PURPOSES OF PLANT SHUTDOWN, LICENSEES TO: DETERMINE AN EQUIVALENT OBE BASED ! ON DBGM REQUIREMENTS; OR DESIGN DIRECTLY FOR CURRENT OBE
- REQUIREMENTS I o NRC SHOULD PERMIT THE USE OF THE i EPRI PLANT RESPONSE GUIDELINES AND
! OBE EXCEEDANCE CRITERION IN l DETERMINING RESPONSE TO AN , EARTHQUAKE l 10
- -- y
p l != NUMARC PERSPECTIVE ON APPERDIX_A BY < ORHAN GURBUZ l l l l PRESENTED TO NUCLEAR REGULATORY COMMISSION l APRIL 17, 1991 4 E 2-Qinr^>- J l( y J U/ U l
l T 1 1 OUTLINE l 1 o OVERVIEW 0. GURBUZ i l 0 APPENDIX A REVISIONS C. STEPP o CHANGES TO 10 CFR 50 M. HAYNER 1 ! o REGULATORY GUIDES F. SWAN l 0 OBE TREATMENT J. REED o DISCUSSION AND
SUMMARY
ALL i ! l I 4 I k i I 1
l lt OVERVIEW o PRIMARY INDUSTRY CONSIDERATIONS o RECENT NUMARC ACTIVITIES o PRELIMINARY NUMARC RECOMMENDATIONS o REGULATORY GUIDES 1 i l i 2 l
t PRIMARY INDUSTRY CONSIDERATIONS o ASSURANCE OF " PROSPECTIVE" APPLICATION o PLANT SHUTDOWN AND RESTART CRITERIA . I o CONTENT AND DETAIL OF: I REVISED APPENDIX A i DESIGN PROVISIONS IN 10 CFR 50 REGULATORY GUIDES o NOT CLEAR IF ALL ESSENTIAL RG's WILL BE ISSUED FOR COMMENTS o TIMING AND CONTENTS OF CHANGES TO OTHER DOCUMENTS: SRPs EXISTING REGULATORY GUIDES CODES AND STANDARDS 3 'i
l i l' ! d
//
~ e j Nd f l 1 l i 4
/ /
# r/ s RECERLKtBULC_ACIIVITIES i l s$f'
/
e o FORMED AHAC - NOV 1990 AHAC REVIEWED RELATED ISSUES, IDENTIFIED OPTIONS j o SIWG FAVORS OPTION C, " SIMPLIFY APPENDIX A AND ! ISSUE SUPPLEMENTARY REGULATORY GUIDES" 'i o NUMARC IS CURRENTLY PREPARING INPUT TO SUBMIT TO NRC: PROPOSED APPENDIX A POTENTIAL CHANGES TO 10 CFR 50 l OUTLINES FOR TWO KEY REGULATORY GUIDES ! POSITION PAPER ON OBE TREATMENT b y[/' y / s - e .- - _ .., n ., ,, ,. 4
PRELIMINARY NUMARC RECOMIENDATIONS o SIMPLIFY APPENDIX A: INCLUDE ONLY "WHAT" NEEDS TO BE DONE TO ESTABLISH SITE SEISMIC REQUIREMENTS o DEVELOP REGilLATORY GUIDES TO ADDRESS "HOW" o DESIGN RULES TO iC CFR 50, INCLUDING PLANT SHUTDOWN & RESTART CRITERIA o ENDORSE EPRI REPORTS:
- NP-5930, "A CRITERION OF DETERMINING EXCEEDANCE OF THE OBE" l
- NP-6695, " GUIDELINES FOR NUCLEAR PLANT l RESPONSE TO AN EARTHQUAKE" .
o ELIMINATE OBE AS A DESIGN REQUIREMENT o APPENDIX A LANGUAGE: NON-PRESCRIPTIVE " CONSIDERATIONS" WHICH
" GUIDE" THE COMMISSION IN ITS EVALUATION OF A SITE MINIMIZE AMBIGUOUS LEGAL INTERPRETATIONS AVOID CODIFYING EXPERT OPINION -y M M"f#"b i
AVOID CODIFYING STATE OF ART g ,u 4 - ,it 6 5
\
i NEW REGULATORY GUIDES GEOSCIENCES: j 1. DETERMINATION OF DESIGN BASIS GROUND MOTION
- 2. DETERMINATION OF POTENTIAL FOR TECTONIC SURFACE DEFORMATION
- 3. DETERMINATION OF SITE GE0 TECHNICAL DESIGN PARAMETERS
- 4. ASSESSMENT OF POTENTIAL FOR VOLCANIC HAZARDS
- 5. ASSESSMENT OF TSUNAMI AND WATER WAVE HAZARDS DESIGN:
- 1. ACCEPTABLE METHODS FOR SEISMIC DESIGN
- 2. DESIGN FOR SEISMICALLY INDUCED FLOODS AND WATER WAVES I
- 3. SOIL-STRUCTURE INTERACTION ANALYSIS
- 4. PLANT SHUTDOWN AND RESTART CRITERIA l
6
PROPOSED REVISION OF 10 CFR, PART 100, APPENDIX A SEISMIC AND GE0 LOGIC SITING CRITERIA FOR NUCLEAR POWER PLANTS PRESENTED TO l',.S. NUCLEAR REGULATORY COMMISSION BY J.C. STEPP APRIL 17, 1991 4
. . - . .. . _ _ _ . ._ _ . _ __ n
S REVISION OF APPENDIX A o PARTICIPANTS C. CHEN - GILBERT l 1
- 0. GURBUZ - NUMARC M. HAYNER - CEI T. O'HARA - YAECo D. OSTROM - SCE C. STEPP - EPRI F. SWAN - GE0 MATRIX R. WHORTON - SCE & G 1
l l 1 j' REVISION OF APPENDIX A i i o OBJECTIVE l - PROVIDE STABLE REGULATORY BASIS i FOR FUTURE NPPs , i \ AVOID LICENSING DELAYS DUE T0 l i UNCLEAR REGULATORY REQUIREMENTS l i - PROVIDE FLEXIBLE STRUCTURE T0 i PERMIT CONSIDERATION OF NEW ! TECHNICAL UNDERSTANDINGS PROVIDE REGULATION IMPLEMENTATION BY ALL - TECHNICAL, LEGAL ; I 2
\ -- -- - +, _ _ . - . , _ - _- - .--.- ._. _ _ _ . _ . ._ _ _ , . _ . .
REVISION OF APPENDIX A STRATEGY I o SIMPLIFY REGULATION i REQUIRED DETERMINATIONS ONLY REMOVE TECHNICAL STATEMENTS REQUIRING INTERPRETATION ADDRESS ONLY SITING AND SEISMIC DESIGN BASIS DETERMINATION REQUIREMENTS NOVE ALL SEISMIC DESIGN REQUIREMENTS TO 10 CFR PART 50 IMPLEMENT REGULATION AND REGULATORY GUIDES DEFINE REQUIRED GUIDELINES IMPLEMENT STATE-0F-ART PRACTICE i l l 3
REVISION OF APPENDIX A STRATEGY (CONT'D) i o SIMPLIFY REGULATION - CONT. l - ALLOW USE OF PROBABILISTIC i METHODS i - STRUCTURE REQUIREMENTS TO ! MINIMIZE REQUIREMENTS FOR
- 1 p/ 7 UNNECESSARY INFORMATION i y'y -
USE GRADED APPROACH DEPENDING i f ON INFORMATION NEEDS TO MAKE DETERMINATION li !~ CROSS REFERENCE TO AVOID DUPLICATING INFORMATION I - REMOVE OPERATING BASIS ! EARTHQUAKE AS A DESIGN ! REQUIREMENT 4
l. REVISION OF APPENDIX A i SCOPE OF REVISED APPENDIX [ ! I. PURPOSE ! II. SCOPE i r j III. DETERMINATION OF THE POTENT 61lL i FOR TECTONIC SURFACE DEFORMATION ! IV. DETERMINATION OF DESIGN BASIS GROUND MOTION i , i ! V. DETERMINATION OF DESIGN l l! PARAMETERS FOR OTHER SEISMIC AND !, GE0 LOGIC HAZARDS ) i I d l 5 I 1
i i REVISION OF APPENDIX A l 0 SCOPE 1 REQUIREMENTS FOR SITING AND DETERMINATION OF DESIGN BASES
- SEISMIC AND GE0 LOGIC HAZARDS i -
SEISMIC DESIGN REQUIREMENTS i MOVED TO PART 50 6 l I
i REVISION OF APPENDIX A j o DEFINITIONS
~
l - SITE - LOCATION OF POWER PLANT ! STRUCTURES l ! l i - SITE LOCALITY - R LESS THAN 1 KM L o - SITE AREA - R LESS THAN 8 KM SITE VICINITY - R LESS THAN l p/ 25 KM l l q- SITE REGION - R LESS THAN 150 KM l TECTONIC SURFACE DEFORMATION - A ! GENERAL TERM FOR FAULTING OR ! FOLDING AT OR NEAR THE GROUND l SURFACE DUE TO LARGE SCALE TECTONIC FORCES IN THE UPPER PART OF THE EARTH'S CRUST 7
\
,' REVISION OF APPENDIX A i i o DEFINITIONS (CONT'D) J,
"',)
DESIGN BASIS GROUND MOTION f (DBGM) '*j y p i THE FREE-FIELD VIBRATORY GROUND ! MOTION DETERMINED AT THE SITE, CONSIDERING THE SEISMIC SOURCES
\#
(fp IN THE SITE REGION AND SPECIFIC i GE0 TECHNICAL CHARACTERISTICS OF i THE SITE SUBSURFACE MATERIALS. p j - IT IS THAT GROUND MOTION FOR WHICH THERE IS REASONABLE ASSURANCE OF A LOW LIKELIHOOD OF
- BEING EXCEEDED.
i LA bgi,LJ J s 1 4JLi t TL w~L L
&~ il ef'A" l
?
\D),-
Y"ty r 8
REVISION OF APPENDIX A i o DEFINITIONS (CONT'D) SEISMIC SOURCE A GENERAL TERM REFERRING TO A GEOLOGIC STRUCTURE OR REGION OF THE EARTH'S CRUST WHERE EARTHQUAKES ORIGINATE 9
REVISION OF APPENDIX A DETERMINATION OF SITE GE0 TECHNICAL DESIGN BASIS PARAMETERS o PURPOSE DEVELOP DESIGN BASIS PARAMETERS FOR FOUNDATION AND FOUNDATION-STRUCTURE SYSTEM FORMULATE PROCEDURES AND ACTIONS TO MITIGATE ADVERSE IMPACTS ON CONSTRUCTION AND OPERATION OF A NPP l 10 J
REVISION OF APPENDIX A . DETERMINATION OF SITE GE0 TECHNICAL
- j. DESIGN BASIS PARAMETERS (CONT'D) l 0 SCOPE OF DETERMINATIONS l -
FOUNDATION EXCAVATION AND DESIGN i BASIS PARAMETERS 3,v LIQUEFACTION AND GROUND T A"Q $
*ft,#[.
ct i SETTLEMENT UNDER DBGM LOADING y l ps SLOPE STABILITY UNDER DBGM I l LOADING SUBSURFACE CAVITIES AND JOINTS GROUND WATER CONDITIONS SLOPE STABILITY AND SETTLEMENT l UNDER STATIC LOADING GE0 METRIC DISTRIBUTION OF l FOUNDATION MATERIALS AND THEIR DYNAMIC MATERIAL PROPERTIES 11
REVISION OF APPENDIX A DETERMINATION OF TECTONIC SURFACE DEFORMATION o PURPOSE DETERMINE NEED FOR AND DEVELOP l DESIGN BASIS PARAMETERS FOR EFFECTS OF TECTONIC SURFACE DEFORMATION l i 12
- j. . .
l ! REVISION OF APPENDIX A i DETERMINATION OF TECTONIC SURFACE i DEFORMATION i l o SCOPE OF DETERMINATIONS i - ASSESS TECTONIC FRAMEWORK OF THE SITE REGION i ASSESS RELATION BETWEEN j SEISMICITY AND GE0 LOGIC l STRUCTURE WITHIN SITE VICINITY l IDENTIFY AND ASSESS GEOLOGIC
- STRUCTURES WITHIN THE SITE AREA THAT HAVE A POTENTIAL FOR !
l l TECTONIC DEFORMATION THAT COULD ! ! ADVERSELY AFFECT THE DESIGN ! BASIS OF A NPP l j o PERMITS USE OF PROBABILISTIC l APPROACHES $ ) l l 13 i
I ; l' REVISION OF APPENDIX A l DETERMINATION OF DESIGN BASIS GROUND MOTION (DBGM)
- o PURPOSE i
DETERMINE DBGM j o SCOPE OF DETERMINATION I - SEISMIC SOURCES WITHIN THE SITE 1 l REGION EARTHQUAKE POTENTIAL FOR EACH a SEISMIC SOURCE N s' - VIBRATORY GROUND MOTION ! ! #9eg TRANSMISSION IN THE SITE REGION 9y< SITE AREA GE0 LOGY AND SITE l DYNAMIC GE0 TECHNICAL PARAMETERS ) - VIBRATORY GROUND MOTION AT SITE l FROM SEISMIC SOURCES IN SITE
- REGION
- o PERMITS USE OF PROBABILISTIC i APPROACHES j 14 l
4
- j. . ..
.' REVISION OF APPENDIX A I DETERMINATION OF DESIGN REQUIREMENTS l FOR OTHER SEISMIC AND GEOLOGIC HAZARDS - ! o PURPOSE 4 DETERMINE NEED FOR AND ESTABLISH i DESIGN BASIS PARAMETERS FOR l OTHER SEISMIC AND GEOLOGIC l HAZARDS 1 ! o SCOPE OF DETERMINATION l l : i l SEISMICALLY-INDUCED FLOODS AND ! WATER WAVES (TSUNAMI, SEICHE DAM l i ' FAILURE) i - VOLCANIC HAZARDS 4 MAN-INDUCED SEISMIC OR GE0 LOGIC HAZARDS (SEISMICITY) SUBSIDENCE OR COLLAPSE 4 i ) i 15
L PRELIMINARY 1 i j BY l MIKE HAYNER l i l PRESENTED TO U.S. NUCLEAR REGULATORY COMMISSION APRIL 17, 1991
h 1 2 1 i I 1 ^ i i f DUTLINE j o OBJECTIVES l 0 APPENDIX A PROVISI6iiS THAT MAY BE MOVED TO j 10 CFR 50 ! o TOPICS THAT NEED TO BE ADDRESSED IN 10 CFR 50 OR i REGULATORY GUIDES i i i
- s.-
! Vy i cp/ sv l g J h( / l f M 1 < 1
/
i !t OBJECTIVES I o P PROTECTION OF PUBLIC HEALTH AND SAFETY IS PRIMARY OBJECTIVE l
/
! c INCLUDE GNLY PROVISIONS THAT ADDRESS "WHATN ! NEEDS TO BE DONE TO ACHIEVE THE PRIMARY l OBJECTIVE i o SEPARATE SITING AND DESIGN REQUIREMENTS i i o CLEARLY DELINEATE REQUIREMENTS FOR CATEGORIES OF j PLANTS SUCH AS: CP BEFORE JAN 1, 199_ j - CP OR COL AFTER JAN 1, 199_ i o ASSURE CONSISTENCY WITH AND SUPPORT OF EXISTING l SUBTIER DOCUMENTS o KEEP IT SIMPLE, USABLE, TECHNICALLY BASED o MINIMIZE POTENTIAL FOR DIFFICULT LEGAL INTERPRETATIONS IN APPLICATION l 2 1 i
it i APPENDIX A PROVISIONS THAT MAY BE MOVED TO ! 10 CFR 50 i o DESIGN BASIS GROUND MOTION; REPLACES SAFE j SHUTDOWN EARTHQUAKE j o DESIGN FOR SINGLE LEVEL OF EARTHQUAKE; CHANGE l FROM CURRENT DUAL EARTHQUAKE CRITERIA ! o PERMISSIBILITY TO DESIGN BEYOND YIELD STRAIN l PROVIDED THAT SAFETY FUNCTIONS ARE MAINTAINED j o THE REQUIREMENT THAT SEISMIC ANALYSIS MUST l ACCOUNT FOR SOIL-STRUCTURE INTERACTION EFFECTS i V l N j o DESIGN BASIS CONSISTENT WITH 10 CFR 100, 5 ,Jye t APPENDIX A 3 l
)
o PROVISIONS REGARDING COOLING WATER SUPPLY AND ' DISTANT STRUCTURES l 0 PLANT SHUTDOWN AND RESTART CRITERIA
/ yl)' y -@'
\S s' y_p e
/ A vi \t/ \~
g ).'r .\'t \ v Yh3 3 vh/
>ggg[ [y J
i i 3, i i i j TOPICS THAT NEED TO BE ADDRESSED IN i 10 CFR 50 OR REGULATORY GUIDES i l o DETAILS OF PLANT SHUTDOWN AND RESTART CRITERIA: i ADOPT EPRI REPORTS USE "CAV" CONCEPT FOR FUTURE PLANTS, UTILITIES SHOULD BE GIVEN THE OPTION TO ESTABLISH SPECTRAL LIMITS FOR SHUTDOWN i ! o PROVISIONS REGARDING SEISMIC INSTRUMENTATION l 0 DETAILS OF S0IL-STRUCTURE INTERACTION ANALYSIS l o DETAILS OF ACCEPTABLE SEISMIC DESIGN METHODS: l - DYNAMIC ANALYSIS l - EQUIVALENT STATIC LOAD i l QUALIFICATION TEST USE OF EXPERIENCE DATABASE [ \
@& (g &
l
/
f
./'
y ^' i (f , i Ge1 r I 4 g$g I l
s i i j TOPICS THAT NEED TO BE ADDRESSED IN i 10 CFR 50 OR REGULATORY GUIDES (CONT'D) i i 4 0 OTHER DESIGN CONSIDERATIONS l DBGM IS SPECIFIED AT THE FREE FIELD SURFACE i - ANALYSIS OF BURIED STRUCTURES i EFFECTS OF EXPECTED DURATION OF GROUND i MOTION l - DETAILS OF DESIGN FOR TECTONIC SURFACE l DEFORMATION i - FATIGUE CONSIDERATIONS IN DESIGN .l o EQUIPMENT QUALIFICATION
- o DESIGN FOR SEISMICAll.Y INDUCED FLOODS AND WAVE i EFFECTS 5
f
. e -
la A ( 1 OVERVIEW OF REGULATORY GUIDES 4 d 4 i , i i i i BY : 1 i !. F. Hs SWAN i l PRESENTED T0: l j U.S. NUCLEAR REGULATORY COMMISSION i . l APRIL 17, 1991 4 I i S I 1 1 e . _ _ _ . -
i i !I i NEW REGULATORY GUIDES i GEOSCIENCES: i l 1. DETERMINATION OF DESIGN BASIS GROUND MOTION i l 2. DETERMINATION OF POTENTIAL FOR TECTONIC SURFACE ! DEFORMATION i i 3. DETERMINATION OF SITE GE0 TECHNICAL DESIGN j PARAMETERS l ! 4. ASSESSMENT OF POTENTIAL FOR VOLCANIC HAZARDS i
- 5. ASSESSMENT OF TSUNAMI AND WATER WAVE HAZARDS
- DESIGN:
! 1. ACCEPTABLE METHODS FOR SEISMIC DESIGN i ! 2. DESIGN FOR SEISMICALLY INDUCED FLOODS AND WATER WAVES
- 3. S0IL-STRUCTURE INTERACTION ANALYSIS
- 4. PLANT SHUTDOWN AND RESTART CRITERIA 1
l 2 l l t i
l.. i l DETERMINATION OF DESIGN BASIS GROUND MOTION i, f
1.0 INTRODUCTION
i i 2.0 REQUIRED INVESTIGATIONS 4 l 3.0 IDENTIFICATION OF SEISMIC SOURCES (
- 4.0 CHARACTERIZATION OF SEISMIC SOURCES i
l 5.0 ASSESSMENT OF FREE-FIELD VIBRATORY GROUND i MOTION l ) 6.0 DETERMINATION OF DESIGN BASIS GROUND MOTION l ) L.._,L
- _4 p l g.ht. u k i
)TJ
- l a
p-3 c_. - - . - - . .-
i
* * ~
\\ DETERMINATION OF THE POTENTIAL FOR ! TECTONIC SURFACE. DEFORMATION i
1.0 INTRODUCTION
l 2.0 REQUIRED INVESTIGATIONS l 3.0 IDENTIFICATION OF CAPABLE FAULTS / FOLDS (SITE AREA 8 < KM RADIUS) l j 4.0 CHARACTERIZATION OF LE F FOLDS ! 5.0 EVALUATION OF GE0 LOGIC STRUCTURE BENEATH THE j SITE j 6.0 ANALYSIS OF THE POTENTIAL FOR TECTONIC ! DEFORMATION AT THE SITE 4 l ! l t l 4
l l ' l KEY ELEMENTS OF PROPOSED l REGULATORY GUIDES
- i
! o DETERMINATION OF DGBH IN A SINGLE REG GUIDE i ! THAT ADDRESSES THE ENTIRE PROCESS FROM l l SEISMIC SOURCE IDENTIFICATION THROUGH i SOURCE CHARACTERIZATION, ATTENUATION, SITE l EFFECTS AND SELECTION OF DGBH l f o SUM 4ARIZES METHODS FOR BOTH THE EASTERM AND l WESTERN U.S.
- o SUNIARIZES CURRENT STATE-OF-THE-ART l DETERMINISTIC AND P!iOBABILISTIC APPROACHES i
l o GIVES GUIDANCE ON THE REQUIRED SCOPE OF l INVESTIGATIONS o CLARIFIES AMBIGUOUS TERMINOLOGY AND CONCEPTS ! THAT ARE IN THE CURRENT VERSION OF APPENDIX l A ! o EXPLICITLY ADDRESS THE TREATMENT OF UNCERTAINTY l j o MORE FLEXIBLE TO CHANGE; EASIER TO UPDATE REG GUIDES TO REFLECT TECHNOLOGICAL ADVANCES 5
i' 1 l l POSITION ON i ! OPERATING BASIS EARTHQUAKE
- IN i APPENDIX A TO 10 CFR 100 i
i i ) i ! BY ! JOHN W. REED i JACK R. BENJAMIN AND ASSOCIATES, INC. I i ! PRESENTED TO U.S. NUCLEAR REGULATORY COMMISSION l APRIL 17, 1991 i i 1 1
i . . j I-INTRODUCTION i o HISTORY OF OBE ! o NEED FOR SECOND EARTHQUAKE AS DESIGN ! BASIS i ! o NEED FOR SECOND EARTHQUAKE FOR PLANT SHUTDOWN AND RESTART l l 0 RECOMMENDATIONS FOR APPENDIX A TO 10 i CFR 100 i I l l l 4 1
/, . HISTORY OF OBE o DEFINITIONS OF OBE IN 10 CFR 100 APPENDIX A i 1 o ORIGIN OF OBE l i h 1 1 I i l l l l l i l 2 l
1. l 2 i' DEFINITIONS OF OBE IN 10 CFR 100 APPENDIX A \ ! o AN EARTHQUAKE WHICH, CONSIDERING i REGIONAL AND LOCAL GEOLOGY, l SEISMOLOGY AND SPECIFIC i CHARACTERISTICS OF LOCAL SUBSURFACE ! MATERIAL, COULD REASONABLY BE j EXPECTED TO AFFECT THE PLANT SITE i DURING THE OPERATING LIFE OF THE l PLANT i o OBE SHALL BE AT LEAST EQUAL TO ONE-HALF THE MAXIMUM VIBRATORY GROUND . ACCELERATION OF THE SAFE SHUTDOWN ! EARTHQUAKE I o THE OBE SHALL BE SPECIFIED BY THE i APPLICANT i ! o van na L p k LJM J shy l l l 3
1 i j' ORIGIN OF OBE
- o IN LATE 1960'S DR. NEWMARK DELINEATED TWO EARTHQUAKE DESIGN
! APPROACHES: i l - LOWER LEVEL " DESIGN" EARTHQUAKE: i ! o EXPECTED TO BE EXPERIENCED IN j REASONABLE PERIOD OF TIME i o LATER CALLED "0PERATING BASIS l l EARTHQUAKE" { \ i - LARGER " MAXIMUM CREDIBLE" } EARTHQUAKE: i o MAXIMUM POSSIBLE INTENSITY OF SEISMIC MOTION WHICH THE PLANT MUST SURVIVE ' o LATER CALLED " DESIGN BASIS EARTHQUAKE" o HOUSNER AND HUDSON ALSO PROMOTED MULTIPLE LEVEL DESIGNS 4
i !' ORIGIN OF OBE (CONTINUED) ! o ULTIMATELY A DUAL CRITERIA WERE i !' INCORPORATED INTO 10 CFR 100 APPENDIX A l - UPPER LEVEL REQUIREMENT CALLED SSE i LOWER LEVEL REQUIREMENT CALLED OBE I ! o ORIGINAL INTENT WAS THAT THE SSE l WOULD CONTROL DESIGN OF PLANT o OBE WAS TO PROVIDE A CHECK WHERE CONTINUED OPERATION IS DESIRED AT LOWER DESIGN LEVEL o OBE CONTROLS DESIGN FOR MANY COMPONENTS 5
}.- NEED FOR SECOND EARTHQUAKE AS DESIGN BASIS i ! o SEISMIC PRA INDICATES MEDIAN PLANT l CAPACITY OF EUS PLANTS TO BE 2 TO 4
- TIMES SSE
! o NO PROVERBIAL FAILURE " CLIFF" JUST i ABOVE SSE LEVEL ! o MOST FAILURE CASES FOUND TO BE l OUTSIDE DESIGN PROCESS (E.G., j ANCHORAGE DETAILS AND INTERACTION l ISSUES) l 0 NOT CLEAR WHAT SAFETY PURPOSE OBE ' l SERVES IN DESIGN PROCESS i Sgg b6P'
\ )<sT 0 0* /
d l '; S pr p h8') g b '" I
\v P Plr o-p $gr 6 x ov
l. i* l NEED FOR SECOND EARTHQUAKE AS DESIGN BASIS l (CONTINUED) , o ANY ADDITIONAL MARGIN PROVIDED BY ! OBE IS UNINTENTIONAL AND VARIES: i COMPONENT TO COMPONENT LOCATION TO LOCATION ! o DBGM PROVIDES SUFFICIENT MARGIN TO l SATISFY SAFETY CONCERNS l 0 FATIGUE TESTING REQUIREMENTS COULD , ! c' BE BASED ON DBGM INPUT $ /
/
l o ATTENTION TO DETAILS MORE IMPORTANT
/ THAN DESIGNING FOR OBE i
/ s. ,
seh.) ..,Wh , t' 9"s g/ s3, a-s 7
- j. -
i i. i NEED FOR SECOND EARTHQUAKE FOR PLANT j SHUTDOWN AND RESTART i i l o EPRI RESPONSE GUIDELINES (NP-6695) i AND OBE EXCEEDANCE CRITERION REPORT (NP-5930) PROVIDE RATIONAL BASIS FOR j RESPONSE TO EARTHQUAKES: !l i PROGRESSIVE RESPONSE TO EARTHQUAKE INITIAL INSPECTIONS I - REVIEW SEISMIC PARAMETERS FROM i ACCELEROMETER RECORDINGS (SPECTRAL i ACCELERATION AND CAV) i MAKE DECISIONS ON NEED FOR PLANT i SHUTDOWN AND READINESS FOR l SHUTDOWN i o IF OBE REMOVED AS DESIGN BASIS THE
- EPRI OBE EXCEEDANCE CRITERION CAN l STILL BE USED i
- 8 l
i 1 _ _. _ . _ . _
l. l' NEED FOR SECOND EARTHQUAKE FOR PLANT l l SHUTDOWN AND RESTART (CONTINUED) I l l t o SPECTRAL ACCELERATION CHECK CAN BE ! ELIMINATED SINCE EXCEEDANCE REQUIRES l THAT BOTH THE SPECTRAL AND CAV ' l PARAMETERS EXCEED LIMITS ! 12 l 0 EXPERIENCE-BASED SPECTRAL LIMIT CAN 7)d;,7 l BE USED AS SPECTRAL CHECK i i o TECHNICAL ARGUMENTS COULD BE MADE TO l ESTABLISH AN EQUIVALENT OBE BASED ON i 7 THE DBGM REQUIREMENTS i /
- /
[o OPTION SHOULD ALWAYS BE PERMITTED TO PERFORM DESIGN FOR OBE USING CURRENT i / REQUIREMENTS FOR PURPOSES OF i / DETERMINING PLANT SHUTDOWN AFTER j i EARTHQUAKE i gavw I 9
8 *, i ; RECOMMENDATIONS FOR APPENDIX A TO 10 CFR 100 i i o THE OBE SHOULD BE REMOVED AS A l DESIGN REQUIREMENT l o APPENDIX A SHOULD BE WRITTEN TO l ALLOW, FOR THE PURPOSES OF PLANT ! SHUTDOWN, LICENSEES TO: i - DETERMINE AN EQUIVALENT OBE BASED l ON DBGM REQUIREMENTS; OR i l - DESIGN DIRECTLY FOR CURRENT OBE ! REQUIREMENTS i i o NRC SHOULD PERMIT THE USE OF THE i EPRI PLANT RESPONSE GUIDELINES AND i OBE EXCEEDANCE CRITERION IN DETERMINING RESPONSE TO AN EARTHQUAKE l l I 10
f -CKc2 n , / - l8 W u . 3 , Ap93-1 OFFICIALTRANSCRIPT OF PROCEEDINGS nR v1 ( x. f
) .
'~
n- C
- y. ,# 1
> >u G' ' , t3 f- /
wg h'
' .c
.O -
c q, ,f j p. j Q'" . AgCDCp U.S. Nuclear Regulatory Commission Advisory Committee On Reactor Safeguards Yldf: Meeting of the Subcommittee on Extreme External Phenomena Docket No.
...a .w . .
.-',n . .- .
. ; ;- + :.. c, ~wu..
1 -
- m. ... .t_. . . . . _ _-
- 4. . .-,,, , g . .y.c.
, .- m ,-
. . . . . ;&+, a .~-w .
!;; . ~ -
;; ; .t.oCAnom , . ., .q, .j.3.e th e s ' a.d , Maryland, .
p , e s ..w+~ . .u r,y.L .:n e y3 . .. d.w'.c.% g;p:'gcpp$n$iNflif.f;:::.8l. .;%z
$ Y % V ! f ; *-h b yff" i g A p'i g . . :. w - ;d ..., y%,s aQ' - ' ; .,
@l,DdWrg10M1991+;Sl%
$gh,,m. g
- 0. ~4 uesde .~D e cemb e r'T 3,
q
, y. ..
-e . %;A+. . 7,C ( seE,s.,..
.... . l. ,- 2 7 5 s = 9 g,..a s. , .m -.
, s. ;x sa ,.,
ist .& Y!;V;5?f e.HM'&s9W. 9 s. ,v _e,smqp. - -n . v ~.R
- a&
$e e ** '
;y k
#, $m, ; . .. . ! Q.
4 o . r t ({'he*,< 4
, ~
m, e h- u:;4c, h .p., .*: ' W.WW
. ,,.3,g 1r m<
q D r' f -
, Jjj l m
' ~
~..-.
., - . .- ~
.cP x -
ps i * ,.,. ml . ,[ , y y,.
- n ; = - ;
~ ~ ' -
al ..l j
.- l'
.. :. ' , p.' '. *
{
- J PP
, m. ,..r..
.,,.5.*.
- c, +- ,..m M j . ,. . J. ., I, j,j 7 t-L. L
% -r i*s j
u . .. ,....a
-.N 'u--.en. e -~e...A4,~t.n.s. a "*-v** ~*
- gj 4.. %
. . . a.'C,
- esp ', -.s tY.?'~4W. . . - w . 4'; : .*.*~ - r~-. -- ~ l
.*- . ~* *., '.y. . - ~ [ /*
.).,
.'?
$ '.. . . /7, ' ~~
- .-- y -~ z hm v. 43 3
- x, . ; .- :4 , .t.M.X,y.: ,,
." ; y ,g : e.q;.;. ,C; j g % ,& .; _y w . Q ;; - : ,
+
,. + nggig, l* Mhbh 'l.QfM. ~
&&,w'i WW+ j$l$f@M.
l p .g:?$ d h;tG ANN RII1Y & ASSOCIATES, I.TD. 0 H!i<.TB v
>~~ W~ ~ E '
M - %.aM. .. 4$E gg9 G ?Mf.g D 1 ^ v. g g@i;6u M ahin m gse w y s. D.C 10006 A..., w. . i.: - .i..~.----- Ob n n(.. ' (M M3M ;!N- b I 6 h , . O' ' E.' i
- b.W
@ ACRS "W ^ , ' , , '1979 PDR
^ ++: E[ bN4QJW.m%..< a. i~.
. a.1 *^ M' ~. .,2
~'. -
.W<
70fN ?,p,rE!r""5$ f h. . & m&~%).h,?k;& % W W - b . . . . _ _ . _ _ _ _
7 L. OmAL 1
- Ap93-1 i
L OFFICIALTRANSCRIPT OF PROCEEDINGS m 4 j h.
.s s s
t 4 .e .. 4 t (, .
- o s .
1 V q s u,- ,
- p. ,
k\ ..: p 4-e i d", ,
, l i'p /-
7., ,e4: i g ':@ / h 4/ , f.'s 1 O.g'C.b'..'.y.+'*. . G -
. C.-
-cA ' -
4 - - i . i AgCDCK U.S. Nuclear Regulatorv. Commission
- j. Advisory Committee On Reactor Safeguards
'T % f j ad: Meeting of the Subcommittee On i Extreme Extern &1 Phenomena i
i e i1 Docket W.
.~:., .s' ..
4 s 4 4 *i'
, , g. g 4c. , .. p 5 ? 3;, 'e '.
, . ,, hW'q " ? x . ; ( . ' s s J,4 [ .
. l, J- , u H lS'P: '~ ' * ' " W
. ,,hd~; c ; " j /' *;';7 M.W. 7,z ft-T y -
,1 !r4~ . ix. 4r Q~*
Lna+f.,J.O :.n
- V '.( -d%. . n: .w. -w , , ..
..u....
p .. f,,,,..:Q 4.:x4t'w.d. v,&n*~ ;;' t:3;_,,,3, ~
., ,.:,.~,. ., ,
n 4.~: e ** q. s w,,1-3. A yw;.; w y;,n ;-,, ., ,n ,. . .
, , -._ ,. ..[; , _
x, . ,dQw: ef' ... ,. , ,
; 43 7 ~.-...%..
w.s,.M .w , * / - 8 ,.w.,.:..>.. .; , w ,,
.,a:., M ;.a.aUSetha"s3aiiM l '%m
;:h.;1 *..Q ,$.[!.OCMIOff,. , M,,,r9; 7' ~ ~.. y, e s.
Xa.f, s %.Rf % m A! e W Y l; O T, p.W g n M,;&_
'ydA.,..var'yland;ipp:%,h
_W . . . ff'. g Al' woume Y $ 3 fhM'&w?:.M, Y, e==4 glfd$$,;$.)e.kkf&p'h. g n 'e==bers:t1m or,ess;.;.sj.273 *cV'O . i. ks..W.m&q,g j 1 . , . , i
- is .
.~ 9 ' m"
't
. ~ . . a ,
. ..- hr.t N. I w %. %5wdC ~ c a.m -' . .
- 1 e$w ? f. - } A %*QY%g$Cd'lY
.s r $4
% gC h n +<a., W 4 ,.
- n V.L4. .mW '
<-~9.VpM :
~
' e li\
t.-x%M.g. 4*t . g.',s, p.3 / '
~se, v
, . ;, s-rD* ' e
~ . . ,. .
n
~__4 2
~-~. w.i :>m .
+
v e a l .m?$ k. &&+b&w:'H0*Ed'e$ 0 !- l
!~ D ?sh .-
'- ~g$,N O '? " '
8 {J. .g b 4 [.s.;.MMADNVD, s.m.*/1,r j P; , L~ 'i'
.c.sW,..W:;sGr.l
~
m w = s a'0.'. w ..y. u l,,, s.g"
'k, ,, ..
%.c & ,
. ,f , a .n. ac L, & ,w.m. c ,, n+ :
7 a,- , . 1, , , , . . _ . . . - . . . . - . . y
,, s.na. w.l j
. w. n. .. +- ..c . . u. a. . -. :. .,
e e p e < . . .,, ( -* l-~- 48 Mt - c ,- , t : a % - 5 34- -
-*:v.r
;V *mw*:. a. *;~:g<
' . ;,%. 7 r -
" s e
((
.n...
- 1. _ ~ j .l sl j .- ,
p:u .ra:x .u , . -w.,owsw . .,m+_ J l,
, r ; . .i
.n mpw s J . J ,f ),l
)
- e. w p z~ +.: m, m.un . wy.1 r=v w-~.<.zew.m.am.v =i aa. a..,ra
.; ; ye + ; e o . , ,. .
- ***;~,***'-~
~ <. y. y - n y. , ~*~.y s .y T'. 7 "* W7' L, .1* . M7!7, T' '!,.
r y-*:-(7W.*?J~"'N;&.", w.m,nt,' A V., l ,.. .t, '"
% .cJ '.1 - .
.~ A'7"~ =
~~~;!.?;""~1C+wf ~~. ..m?,' ? :s 7 3,. u. ~~_;.,- '
&.[ ,, h, y%, y ;*~. W 'G * ' u Q'xf[% ", N..h., y4Q/[ v y.,-b]d.g;/M,99Q.?p[' (* s.R6k% 6N .;,-r...Q. c w: ' y;; ; c:.
.a +
. ,.y r..c,,.y p
, .< ~ - < , .
~
ks.g * }:D $? .fhh.Y*h'*h$ h~h !;,, l$.-:: a .s
)#ha Y <
.u f *$, sN4Mbf!&%} MG0WWW hM:WW ':
- g
%k$;Dyh@$Mh' MP ANN REEY & ASSOGATE .
J~ . ?.
@,. p pc5.&s h yh&$.E
.W-~Ww a -{
m? iti.,
;Q W~urann.gse3ae< w u -
O.(& 5.,e b . . h e.ls.l . ._? , ' . '
~
0 L' c w
~.
z
~ :
O M ' ** '
* ' ~
i2.iI 5?*~S$ 5 ~~ f''". "h' '*\'";.~.'f' ~~; y'*? **- , , PDR ACRS 3 4 h,f C~'
~
J"~ T-1879 v.. p m m =2*6p M # %
' ~PDR d : g p..Mrm '
g'f S h & &Au;a.Abac&alWip p. 1 w~ AL
PUBLIC NOTICE BY THE UNITED STATE NUCLEAR REGULATORY COMMISSION'S ADVISORY COMMITTEE ON REACTOR SAFEGUARDS DATE: Tuesday, December 10, 1991 The contents of this transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards, (date) Tuesday, December 10, 19gl as Reported herein, are a record of the discussions recorded at the meeting held on the above date. This transcript has not been reviewed, corrected or edited, and it may contain inaccuracies.
- 1- UNITED STATES OF AMERICA l
- 1 2 NU7 LEAR REGULATORY COMMISSION 1
3 ******* I 4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 1 l 5 ******* !e 6 MEETING OF THE SUBCOMMITTEE ON l 7 EXTREME EXTERNAL PHENOMENA !* 1 8 ******* 9 10 11 Nuclear Regulatory Commission 12 Conference Room P-110 13 7910 Norfolk Avenue 14 Bethesda, Maryland 15 1 16 Tuesday, December 10, 1991 17 18 19 The above-entitled proceedings comm,enced at 8:30 e 20 O' clock a.m., pursuant to notice, C. Siess, ACRS 21 Subcommittee Chairman, presiding. 22 23 l 24 l l ! 25 1 i i i l
I
'2 !
l 1 PARTICIPANTS: 2 3 C. SIESS, ACRS Subcommittee Chairman 4 C. MICHELSON, ACRS Member I 5 C. WYLIE, ACRS Member e 6 D. HOUSTON, ACRS Cognizant Staff Member 7 B. BOSNAK, NRC/RES , 8 A. MURPHY, NRC/RES 9 N. CHOKSHI, NRC/RES 10 D. BERNREUTER, LLNL 1 1 11' R. McMULLEN, NRC/RES i 12 P. SOBEL, NRC/NRR 13 R. KENNEALLY, NRC/RES 14 M. WITTE, LLNL 15-5 16 i 17 l 18 19 l
.. i 20 21 22 '23 l 24 \
I
c 1 PROCEEDINGS 2 [8:30 a.m.] 3 MR. SIESS: The meeting will come to order. 4 This is a meeting of the ACRS Subcommittae on 5 Extreme External Phenomena. 6 I am Chester Siess, Subcommittee Chairman. 7 - We have two other members in attendance today: 8 Mr. Michelson on my left and Mr. Wylie over here. 9 The purpose of the meeting is to discuss the 10 staff's proposed revisions to 10 CFR Part 100, Appendix A, 11 " Seismic and Geologic Siting Criteria for Nuclear Power ) 12 Plants." 13 Mr. Dean Houston on my right is the Cognizant ACRS 14 Staff Member for this meeting. 15 The rules for participation in the meeting have 16 been announced as part of the notice previously published in i 17 the " Federal Register" on Monday, November 25. 18 A transcript of the meeting is being kept and will 19 be made available as stated in the " Federal Register" e 20 Notice. It is requested that each speaker first identify 21 himself or herself and speak with sufficient clarity and 22 volume so that he or she can be readily hard. 23 That means use microphones. And since we've got a 24 small group here today, I'd suggest that at least some 25 members of the staff that are going to be most actively ) i ____-_-__ - ---- - -------- - --------- - ----
1 involved might come up and sit up at the table here. 2 There's more microphones available that way. So you can 3 decide which ones. I don't know whether you want everybody 4 up or not. I don't care. But it'll be handy for everybody 5 that's going to talk to have a mike. 6 We have received no written comments or requests 7 to make oral statements from members of the public. We will 8 be happy to hear from members of the public at appropriate 9 times if they'll simply let me or Mr. Houston know. 10 I have an announcement for members of the 11 subcommittee and our staff, and any of the NRC staff that 12 may try to exit through the door to the classified area. 13 That door is now alarmed. It has not been alarmed for the 14 last six months, but somebody decided it should be, and if 15 you go through that door, you have to use your key card or 16 it makes terrible noises and you get chided by one of.the 17 guards. It's already happened to me. There's a sign on 18 there that reminds you. 19 We have an agenda that covers a lot of territory, 20 with the usual optimistic times. And I assume that the 21 members of the subcommittee have received an axtensive . 22 collection of paper. 23 I have received something that you may not have. 24 You have a copy of the new Appendix B and the new Appendix 25 S, both of which are lined in, lined out, and particularly 3
'5 1 for Appendix B there was an awful lot more taken out than 2 was put in.
3 I've got clean copies of both of those. Do you 4 have any extras of those? 5 MR. CHOKSHI: No, but we can make sor.e. 6 MR. SIESS: Yes. Proposed as is withaut the If you can get those for us. These are 7 lined-out material. 8 marked up. 9 MR. CHOKSHI: May I have those, and I'll Xerox 10 them. MR. SIESS: These are marked up. I'd rather not. 11 but I 12 It might be helpful to have all my comments on it, 13 doubt it. MR. CHOKSHI: We will get copies. 14 MR. SIESS: We have a number of documents, of 15 We have the Appendix B, Appendix S, and I guess 16 course. Guides and a Draft Standard Review Plan, all 17 four draft Reg. 18 of which replaces the existing Appendix A to Part 100. 19 I think that, other than try to review all of this 20 and repeat what I think Bob Bosnak is going to say, we'll 21 just let Bob start in with the introduction and over'tiew. 22 But before that, we've made no plans yet, selected 23 no date to take this to the full committee, partly because I 24 don't think they're ready for it until they have also looked 25 at the other changes that they're looking at in the siting. l
\
l
i 1 1 I think the full committee probably should get the whole 2 picture at one time, not necessarily, but it may be i 3 convenient, and because I'm not sure how far we're going to 4 get with this stuff today. 5 If it looks like we need another subcommittee 6 meeting, we've tentatively picked what, February 5, which is 7 the Wednesday before the next full committee meeting, not 8 the next one, but the February full committee meeting. j 9 But we'll see.how far we get today. And I'll tell l 10 you right now, it's going to be my recommendation, subject l 11 to what the other people think, that the Reg. Guides l l 12 probably ought to go out for public comment, and have 13 another look at them after that. That's particularly true l l 14 of what I call the small ones, 10-16, 17, and 18. On the 15 10-15, I don't know, we really don't have that yet. 16 Okay. With that introduction, Charlie, do you or 17 Carl have anything to say aright now? l 18 MR. MICHELSON: No. 19 MR. WYLIE: No. 1 20 MR. SIESS: Okay. Mr. Bosnak is substituting for ! 21 Mr. Shao, I guess, on the agenda. [ Slide.) 22 23 MR. BOSNAK: What I am going to ba presenting to l 24 you is an overview to Appendix A to Part 100, just exactly 25 what we're trying to do.
1 The presentation, my presentation will be followed 2 by a series of presentations covering the extent of the 3 revision, and the presenters are members of the staff and and we have people 4 also Lawrence Livermore National Lab, 5 here from the office of Research and NRR. 6 (Slide.) MR. BOSNAK: I think the principal message here 7 8 with respect to the background is that the bases for what is In 9 now Appendix A to Part 100 goes back to the late '60s. 10 other words, we're talking now almost 30 years of a period 11 of time that has passed since we've made any changes to the 12 regulation. The regolation, as you see, became effective on 13 14 December 13, 1973.' 15 And there have been many advances in the areas of 16 seismology and geology, and also in the concurrent engineering area, that requires us to take a look at what we j 17 18 have and the difficulties in applying the current Appendix A 19 to plant construction. 20 (Slide.] 21 MR. BOSNAK: Now, a little bit of, if you want to 22 call it philosophy. 23 We expect that the new, it will be called Appendix
;14 B, will be part of 10 CFR 100, but Appendix A will remain on the books. This is a forward fit development.
~
25
l 1 l l So Appendix A will remain on the books. 2 The new Appendix S and the new Appendix B will t 3 both be forward fit. 4 MR. SIESS: Bob, there is nothing optional about 5 that, is there? 6 MR. BOSNAK: Optional? 7 MR. SIESS: Could a plant decide they wanted to 8 1 pick an OBE and an SSE based on the new values? 9 MR. BOSNAK: Do you mean an existing plant or a 10 new plant? 11 MR. SIESS: Existing plant. 12 MR. MURPHY: Chet, I'll touch on that, but we have 13 considered ths case-by-case option, if you want, of trying j 14 to resolve issuas-that may arise with currently operating 15 plants using the new regulation. It would be something that 16 we would consider, the licensing staff would consider, on a 17 case-by-case basis. It is not being set up with a 18 particular procedure to be followed, or anything. 19 MR. SIESS: But if somebody for some reason wanted 20 to develop an SSE or OBE for equipment qualification 21 purposes or license renewal purposes, or something, using 22 the new criteria, could they? 23 MR. MURPHY: Again, it would be something we would j 24 consider on a case-by-case basis. It will not be { 25 specifically allowed or excluded by the regulation. l t f l ! )
'9 MR. SIESS: But if you wanted to, you can't?
1 MR. MURPHY: Right. We're not proposing any kind 2 3 of a backfit with this. MR. SIESS: Okay. 4 MR. BOSNAK: It would have to be on a case-by-case 5 6 basis. 7 MR. SIESS: Okay. MR. BOSNAK: But as this proceeds through the 8 9 process, it is a forward fit only, Fo We will not be doing a 10 backfit analysis. MR. MICHELSON: Roughly when do you think Appendix 11 12 B might appear in the regulation? 13 MR. BOSNAK: Well, if we talk about getting it out 14 for public comment as a proposed rule this Spring, that 15 period that you have for public comment probably will take 16 60 to 90 days. MR. MICHELS0H: But by the end of '92. 17 18 MR. DOSNAK: By the end of '92 we should be ready 19 to go forward with the final rule. MR. MICHELSON: Then how do you think this applies 20 21 to ABWR? Doesn't apply to -- oh. Appendix B 22 MR. SIESS: doesn't apply to ABWR. Appendix S might. Appendix B will 23 24 apply to the first applicatLon for a site which is expected 25 -- i
?
1 i0 MR. MICHELSON: Yes, and it won't apply for the 2 certification itself, but for the first plant you build 3 under it. 4 MR. SIESS: No. 5 I think the thinking is now, in the five-year strategic plan, that there will be a site 6 approval application sometime late in '92, is that right? d 7 MR. BOSNAK: Supposedly. 8 That's what I've heard. MR. SIESS: And Appendix B is strictly site. 9 MR. MICHELSON: Yes. Yes. 10 MR. SIESS: , Now, nothing that we're looking at is 11 going to apply to a standard plant, because their "g" 12 value is already fixed. 13 MR. MICHELSON: I meant a site for a plant. It 14 will apply to the'first site that's selected, I think. 15 MR. SIESS: The first site, if somebody selects it 16 for a plant, it will apply to that, or if they ask for a 17 preliminary site approval without a plant, it'll apply to 18 that. 19 MR. MICHELSON: If you apply it before you make 20 this c part of the regulation, if you apply it, say, next 21 Summer. It's just under the present regulations, then; this 22 wouldn't apply at all? . Is that the way you view it? or are 23 you going to kind of apply it anyway? 24 MR. BOSHAK: Well, let's say this was out as a 25 proposed rule this Spring.
n . . . .- _ _ ._.-. - .- - . . . . . . .. .. 11 1 MR. MICHELSON: Yes. 2 MR. BOSNAK: And if somebody on a, call it on a i 3 case basis, wanted to use some of the material, I'm sure we 4 would -- i 5 MR. MICHELSON: You'd just sort of ad hoc that l 6 one. But you wouldn't require it. They could apply under 7 the present regulation, if they wished? 8 MR. BOSNAK: That's right. l I 9 MR. SIESS: By the time they got it, the new ones 10 would be in effect. l 11 MR. BOSNAK: The last bullet that I wanted to 12 point out takes care of what is now in VI in Appendix A to 13 Part 100, and it will become part of 10 CFR 50, and new 14 Appendix S, with respect to all of the engineering design , I 15 associated with seismic activity. 16 [ Slide.] 17 MR. BOSNAK: And the last thing that I have here 18 is just to give you a quick order of presentations ! 19 Appendix B, the overall approach, will by Andy 20 Murphy. 21 Probabilistic criteria, which is something that we 22 have not had before and we feel is obviously something that 23 has been needed. Nilesh Chokshi. 24 The application of the new approach will be Mr. 25 Bernreuter from Lawrence Livermore. i
12 1- Geologic investigations, McMullen.
'2 Ground motion, Phyllis Sobel.
3 Engineering criteria, Roger Kenneally. 4 The OBE/SSE investigations, Witte from Livermore. 5 And finally again on the Reg Guides for t 6 engineering, Roger Kenneally. 1
)
7 MR. SIESS: Just leave that up there for a minute, , l 8 Bob. 9 MR. BOSNAK: Sure. 10 MR. SIESS: I'm sure Andy is going to get into 11 this, but the existing Appendix B -- I mean the existing I l 12 Appendix A is being split up and part of it is going into 13 Part 50. That's the Appendix, the Appendix S. 14 Now there isn't much change. That's mainly a 15 physical, an editorial move, shall I say. 16 MR. BOSNAK: Yes. 17 MR. SIESS: All right. Now the three Reg Guides I 18 1016, 17, and 18, where is that now? 19 MR. MURPHY: The one Reg Guide in instrumentation L 20 was there. The othe'r materials, in effect, are new 21 material. 22 MR. SIESS: Okay. But none of that was in 23 Appendix B -- Appendix A. 24 MR. MURPHY: Right. t j- 25 MR. SIESS: So the contents of Appendix A are 5 l l , __ _ - _ . - -. _, ,
I3 which is 1 basically being replaced by Appendix S and Part 50, and the new Reg Guide or 2 the part that deals with the plant, 3 the new Appendix B and Reg Guide 1015. 4 4 MR. MURPHY: Right. 5 MR. SIESS: Okay. And, of course, the new 6 Appendix B leans heavily on 1015, all of the stuff that -- 7 MR. MURPHY: Yes, sir. 8 MR. SIESS: -- we wished we could have changed ten 9 years ago is now in the Reg Guide. 10 MR. MURPHY: That's correct. 11 MR. SIESS: Putting it rather crudely. 12 As I recall, one of the main objectives of the 13 original Appendix A and, I think, one of its more successful 14 applications was to define the scope of the exploration 15 necessary to establish the earthquake design basis or SSE 16 earthquake. 17 As I recall, back in those days there was a real 18 problem with somebody trying to site -- pick a site to know 19 how far out to go, what to look for, how to get there, when 20 to stop, and there was a lot of give-and-take between -- 21 well, I won't say give-and-take. They'd present it; you'd 22 send it back. 23 To what extent has the criteria for the scope of 24 the investigation changed as we go from existing Appendix A 25 to new Reg Guide 1015?
/ i4 ' 1 MR. MURPHY: To a very large extent, the scope has , 2 remained basically the same. The definitions, the ' t 3 inferences, the conflicting interpretations -- , 4 MR. SIESS: And how many miles out you go and -- 5 MR.-MURPHY: Yes, sir. 6 MR. SIESS:
- Okay. '
~
MR. MURPHY: Basically what's happened is that has 8 moved from '_se regulation to the Reg Guide. Some of the 1 9 definitions that were put into the regulation, the original 10 Appendix 7, have been moved and changed to some extent in ! 11 the new Reg Guide 1015. 12 MR. SIESS: There's a lot of change in terminology i . 1 13 there. i i 14 NF. MURPHY: Yes, sir. 15 KR. SIESS: Now I understand you've had some 16 examples worked out of -- does that include that part as 17 well as the selection of the ground motion? 18 MR. MURPHY: The examples that have been worked
)
l 19 out have been associated with the probabilistic analysis l 20 part of it mainly. 21 MR. SIESS: Okay. I 22 MR. MUMPHY: We have not done, if you want, a 23 trial case on the new deterministic investigations. It's 24 something we are contemplating. 25 MR. SIESS: Okay. So the big change, then, other i t
15 1 than the shifting around of where things are, is in the 2 selection, quantification of the design basis earthquake, or 3 let's say the earthquake motions? 4 MR. MURPHY: Yes, sir. That's where the apparent 5 changes are. 1
- i 6 MR. SIESS: And that's the difference between the l 7 existing Appendix A and the new Appendix B and DG-1015.
8 MR. MURPHY: Right. The major changes are in the 9 placement of them and straightening out the language and the 10 descriptions that are used to describe the materials that 11 are requested. 12 MR. SIESS: Now you are going to have something 13 today from Liverr. ore en how you select the design basis -- 14 well, let's say -- what's your new nomenclature? Safe 15 shutdown earthquake ground motion? 16 MR. MURPHY: Yes, sir. 17 MR. SIESS: How you select that with this 18 combination of probabilistic and deterministic criteria? 19 MR. MURPHY: Yes, we're going to have something on 20 that today. 21 MR. SIESS: Applied to specific sites that have 22 been through the old process? 23 MR. MURPHY: Yes. 24 MR. SIESS: Okay. Because that's the one part 25 that is the biggest change. It's also the one part that I t _
l l l l l i6 1 can't understand. ] 2 MR. MURPHY: Okay. We'll try to make that clear. 3 MR. SIESS: I'd like to say, if you could make me ) 4 understand it, it will be clear for anybody, but I doubt
]
l 5 that. 6 Okay. Are you through, Bob? l l 7 MR. BOSNAK: Yes, I'm through. Andy is going to , i 8 -- 9 MR. SIESS: Okay, thank you. 10 Are there any significant changes in the material l . 11 that got moved to Appendix S? 12 MR. CHOKSHI: The portion regarding OBE is 13 significant. l 14 MR. BOSNAK: I think as the presentation goes on, 15 we will pick up on that. . 16 MR. SIESS: But the biggest change is the OBE i 17 definition, then? 18 MR. CHOKSHI: Yes. 19 MR. SIESS: Well, not the definition, but the way 20 it is used in design. 21 MR. CHOKSHI: That is correct. 22 (Slide.] 23 MR. MURPHY: That simply tells you who I am, why l 24 we are here today,and my phone number. 25 (Slide.] , i (
I7 1 MR. MURPHY: Basically, what I am going to do for 2 the first part of this discussion this morning is just to 3 lay down some of the background material as to why in effect 4 we are here. Why are we revising Appendix A. 5 And I think what the first bullet does is tell us, 6 fine, we all think Appendix A has warts and wrinkles on it, 7 but what we must recognize is that for some large majority l 8 of the power plants that are out there now, they were 9 regulated and licensed under this, and basically it worked. 10 They're out there. We don't think there are any 11 really serious mistakes. We don't think there are any 12 moderate mistakes that have been made in licensing these l' 13 facilities. So, the bottom line is that Appendix A has i i 14 worked in the past. l l 15 It has been a trifle inconvenient and cumbersome 16 at times, but it has worked. - 17 MR. MICHELSON: Have you had enough earthquake 18 experience with the present-day plants to know whether or i l j 19 not it really works? ] 20 MR. MURPHY: It has not been challenged by a major . l 21 earthquake. 22 MR. MICHELSON: Right. 23 MR. MURPHY: And really, that's one of the things 24 that are in its favor. 4 25 MR. MICHELSON: Well, in a way. I l l
- . - . - _ _ - . ~ . . . . . _ . - . - - - . - . . . - . - - . - . - . - . - . - . . . . - .-- . ... _ .-
i8 1 MR. SIESS: But your statement must be based on 2 such things as the seismic margin studies. 3 MR. MURPHY: Yes, sir. There's lots of things 4' that have gone if you want to make that the background so I 5 that this is a rationale and a reasonable statement to make. j 6 MR. SIESS: What we can't say is that Appendix A l 7 has lead to a-uniform level of seismic design. l 8 MR. MURPHY: No, I don't think it has. 9 MR. SIESS: That is some -- 10 MR. MURPHY: There has been interpretation and l 11 changes in interpretation from the time it was written until l l 12 today. 13 MR. SIESS: But if I did a PRA on every plant out i 14 there, I probably wouldn't get the same seismic risk on all 15 of them. 16 MR. MURPHY: No. I don't think so. I think it's 17 quite correct to say that the -- if you want, the margin has 18 changed with time. i l 19 MR. SIESS: But if I don't get the same seismic -- 20 I'm not sure I want the same seismic risk on all plants, but 21 I guess if I had to define a uniformity, that would be the i 22 number, the quantity that I would pick. 23 But if I don't get the same seismic risk, say, on 24 two plants, take A and B, there could be more than one l 25 reason. One is I wouldn't have the same seismic hazard i
_ _ _ _ _ . . . - _ . _ _ . _ _ . _ _ . _ _ _ _ _ _ _ _ . . _ . _ _ _ _ _ ~ _ . _ _ _ . 4 1 1 J l 19 ! 4 1 probability. I could pick the hazard probability wrong.
- 2 The other would simply be that the structural 3 design --
I 4 MR. MURPHY: Has changed. l 5 MR. SIESS: Was made to changing section 3 8 ' 6 criteria, damping criteria, something else. ! 7 MR. MURPHY: Yes, sir. 8 MR. SIESS: What would you say our objective is in ! 9 seismic siting? l 10 MR. MURPHY: Oh, lordy! ! 11 MR. SIESS: Well, let me help you. Before we - 12 started talking certification in standard designs, one 13 objective might be a uniform hazard, right? Some uniformity I 14 of the seismic risk. 15- MR. MURPHY: I would go along with the uniformity 16 of the seismic risk rather than uniformity of the hazard. ; 17 MR. SIESS: Yes. Okay. The seismic risk. But as l 18 far as what we are talking about today, everything except 19 Appendix S is dealing with hazard, not with -- i 20 MR. MURPHY: That's correct. , 21 MR. SIESS: Plant design. But now that all future 22- plants'-- and this isn't a strictly correct statement -- but 23 at least some people hope that all the future plants will be 24 standard plants and will be designed to 0.3g, any talk about , 25 uniformity of seismic risk is meaningless, isn't it? [ i
. - - . - _ ~ - . . . . - - . - ~ . - . . - , _ . - - - _ - - . _ _ _ _ _ . _ . . _ , _ _ _ _ .
l bo ) 1 MR. MURPHY: In my mind. 2 MR. CHOKSHI: The seismic risk is governed by ; 3 hazard. And if you put the same -- 4 MR. SIESS: The combination of the hazards and the 5 0.3g.
.6 MR. CHOKSHI: It will be different from site to
, 7 site. 4 ! . l 8 MR. SIESS: So what now would we say the objective 9 is? 10 MR. CHOKSHI: In my presentation I am going to l 11 show the -- how the current plans look in terms of 12 probability effects in design. i 13 MR. SIESS: Forget about the current plants. We 1 14 are starting fromoscratch now. I 15 MR. CHOKSHI: I think that we are trying to make 16 sure that the new plants -- that siting design basis at 17 least is as compatible with the current generation of 18 plants. 19 MR. SIESS: Okay, but suppose we didn't have a
~~
20 current generation of plants? Suppose we were starting out l 21 fresh with a series of standard plants all designed to an 22 SSE of 0.3g equivalent, and we are developing a method for 23 selecting sites at which those plants can be built? j 24 MR. MURPHY: Then we would definitely be taking a 25 different tack in the probabilistic side of the new
$1 1 regulation than we are to date.
2 What we are doing to date depends upon having that ; current generation of plants out there and if you want to ; 3 4 say, an experience base -- 5 MR. SIESS: That's methodology. ; That's methodology. 4 6 MR. MURPHY: 7 MR. SIESS: I am, again, saying what's the , 8 objective? And the only objective you could have, if all my 9 plants are standard plants, all of them are designed for. j 10 0.3g, the only objective I can justify from a regulatory ; 11 point of view and public safety point of view, is that the 12 site selected will be one in which the. probability of 13 exceeding the 0.3g is low by some -- 14 MR. MURPHY: Some standard. i 15 MR. SIESS: Some specified amount. Right? : 16 MR. MURPHY: Right. 3 17 MR. SIESS: I mean, it might be 10 to the minus 18 five probability of exceeding 0.3g at one site, and 10 to 19 the minus eight at some other site, as long as it's less 20 than some number which we haven't stated, obviously. 21 MR. MURPHY: Right. ' 22 MR. SIESS: For the future, then, this is the 23 objective. Simply to be sure that a site is acceptable for 24 a standard plan. Right? 25 MR. MURPHY: That's correct. t
22 1 MR.-SIESS: Now, you are really setting that level l 2 of acceptability for future plants in terms of what exists 3 for present plants. 4 MR. MURPHY: That's correct. That's part of it. 5 MR. SIESS: Now, although I don't understand it l 6 all, I am thinking you are saying it should be less -- 7 MR. MURPHY: We will explain some of that. 8 MR. SIESS: Less than the median of the present 9 plants. 10 MR. CHOKSHI: That's right. 11 MR. SIESS: So, you see. You are still not going 12 to abandon the present plants. 13 MR. MURPHY: No, sir. 14 MR. SIESS: If you were starting over, you would 15 have to arrive at a number -- 16 MR. CHOKSHI: Based on the current generation. 17 MR. SIESS: We are looking for some probability of 18 exceedance not to be greater than any plant in the future? 19 MR. MURPHY: Right. Without coming anywhere close
~~
20 to specifying what that greater than number is. 21 MR. SIESS: Well, you will. You've got to specify 22 it. 23 MR. CHOKSHI: When -- I'll show it. l 24 MR. SIESS: That's what I am going to be looking l l 25 for. l i i i
$3 1 MR. CHOKSHI: Yes.
2 MR. MURPHY: Okay. 3 MR. SIESS: Just keep that in mind.
'4 MR. CHOKSHI: Both myself and Don will talk about 5 the use of the number.
6 MR. SIESS: Because I am going to take the 7 approach -- I am looking for future plants, 0.3g, standard 8 designs, and what these criteria mean. 9 MR. CHOKSHI: Right. 10 MR. SIESS: In terms of those plants. 11 MR. MURPHY: Fine. 12 MR. SIESS: And the only way I think I can see it, 13 now, somebody else may offer a different approach -- is what 14 is the probability of exceedance,'if you wish. ; 15 MR. CHOKSHI: For 0.3g. 16 MR. SIESS: For the 0.3g, however it is i 17 interpreted. l' 18 MR. CHOKSHI: Yes, I have a viewgraph on that. 19 MR. MURPHY: Okay. One other point is that while 20 this regulation and the writers -- the Staff -- had the 21 probability of nonstandard plants in the future being very 22 low, the regulation in itself is not discriminate, i.e., we 23 can have standard or nonstandard plants -- ,
?
24 MR. SIESS: Oh, yes. 25 MR. MURPHY: Licensed under this regulation. I i
d4 1 want to make sure that was understood. 2 MR. SIESS: Oh, yes. 3 MR. MURPHY: Okay. Then the rest of the bullets-4 on here indicate some of the issues we have had over time 5 with Appendix A. The prescriptiveness in the regulation 6 itself is something that we want to pull out. . 7 These conflicting interpretations which have led 8 to significant discussions and difficulties among the 9 various bodies that have reviewed the documents -- the 10 current regulation, as Bob has indicated, doesn't take into 11 account the advances that have been made in seismology and 12 geology over the time, nor the advances that have been made 13 in the probabilistic analyses in handling the uncertainties 14 associated with these things. 15 One of the big problems with the engineering 16 aspects in the current Appendix A is that the multiple 17 definition of the operating basis earthquake. 18 Those are some of the things that we anticipate 19 taking care of with the new regulations. 20 (Slide.] .- 21 MR. MURPHY: And here if you want, just some 22 background on -- I'll call it the " legalisms". That's why 23 we are here. 24 There have been the two SECY papers that have 25 discussed the difficulties and problems associated with the
$5 1 current regulation. They go back to approximately four 2 years after the regulation was promulgated, and I think 3 about half the folks on the staff here were hired in some 4 sense to assist in the revision of Appendix A.
5 I know that that was one of the things that I was 6 brought on board 12 years ago to start working on. 7 MR. SIESS: It took us four years to find out we 8 had a problem, and another 26 to come up with a solution. 9 (Laughter.] 10 MR. MURPHY: That's the right order of magnitude. 11 It was not just these groups. The SECY papers and so forth, 12 but also the piping review committee pointed out some 13 difficulties with it as well. 14 The last two bullets are simply the in-house 15 requirements to get it on the books so that we could get 16 started working on it. As noted there, the EDO staff has 17 indicated that this rulemaking very definitely has a high 18 priority and that we are supposed to be significantly in 19 gear and working on this thing. 20 (Slide.] 21 MR. MURPHY: Those are basically the objectives. 22 If you look back at the issues, that's what we are taking 23 care of. We want to remove the uncertainties in this l 24 interpretation in the current regulation. We want to cut 25 the 26 years down to two and a half years if there are cases
_ _ _ . _ . ,. _ . .__ _ . _ . _ _ _ . _ _ . _ . _ _ _ _ . _ . . . _ _ ~ ~ _ _ _ . . . . _ _ __.
$6 1 where we need to update what's out there.
2 The difficulties of having the prescriptive 3 material in the regulation and not in reg guides has 4 obviously been a thorn in everybody's side, and just an all-t ! 5 around problem. l . L 6 One other thing the industry folks have been 7 poking and prodding us about as they look forward to the 8 next. generation of reactors, is they are looking for some 9 sort of stability within the licensing process. l l 10 We are hoping to provide the stability.in the l l i l 11 regulation and the flexibility within the reg guides. Our i 12 instructions are to complete this action for use for early 13 site review. There is some discrepancy back and forth as to-1 14 when the first application is expected. At this time, we l 15. are anticipating, as I will show you in the other slide -- l 16 we're anticipating finishing this rulemaking probably.in l l 17 early '93. l 18 [ Slide.] ! 19 MR. MURPHY: We have touched on -- l
.- I t
20 MR. SIESS: Do you have anything firm on when ' 21 there might be an early sie review? I keep seeing a 22 reference to a DOE application. Are they simply sponsoring 23 somebody to make an application? 24 MR.-MURPHY: I will offer you my interpretation of , 25 what I think'is going on. Possibly the folks from NUMARC or ' s }'
$7 1 DOE may have some additional thoughts.
2 But it was my understanding that there has been 3 some sort of an RFP and a consortium, I believe, of three 4 utilities and NUMARC, I think, have put together a response 5 to that proposal. They are -- in that proposal is like a 6 year's worth of study being proposed on what is going on 7 with the siting regulations. And then at the end of that 8 year, they may put together an applicant package. 9 Sort of the feeling that I had was that '92 was 10 the study year, and the -- potentially the applicant was in , 11 '93. But this is -- I'll call it mainly hearsay at this 12 moment. 13 MR. SIESS: I've got a feeling that the seismic 14 aspects of an early site review might not be the dominant 15 feature. I have seen figures that have taken two or three 16 years to get a site' approved. 17 Suppose -- I an away from the early site review, I 18 guess, for a moment. But in looking at our siting criteria, 19 which would be the rag guide 1015 -- DG-1015 -- suppose l 20 somebody came in and asked for a site review at Millstone? , 21 What would we do? Apply the new criteria? Or -- 22 MR. MURPHY: I think that you end up with a mixed 23 bag. Let's say it's probably going to take a year or two 24 anyway for the site review at this new Millstone. I guess 25 that a lot of what will happen -- in my mind -- would be l l
,, -- -- . . ~ --
1 B
$8 ;
1 that the current regulation would be applied, but after this 2 thing is published as a proposed -- 3 MR. SIESS: No -- let's Juppose this has all been 4 approved. 5 MR. MURPHY: This has all been approved? 6 MR. SIESS: This has all been approved. It's in 7 effect. Effective date exists. And somebody says I want to j i 8 put up Millstone 4. 9 What do we use for an earthquake? Seismic design 10 basis. 11 MR. MURPHY: I am not familiar enough with 12 Millstone -- how Millstone turned out on the probabilistic 13 analysis either in the Livermore or the EPRI study, but -- 14 MR. SIESS: The answer is you take another look at 15 it. j 16 MR. MURPHY: You take a look at it. First you p 17 take a look at the geology and geophysice that was done i 18 initially for the first three plants on that site. You look 19 at that material, review that material. 20 And see what you come up with. Effectively follow 21 -- I'll call it the neoclassical Appendix A approach. 22 Then you go ahead and you run the calculations 23 either with the EPRI or the Livermore study and see what 24 those numbers tell you.
~
25 I will ask you if I can defer further discussion
i
$9 1 of that until I get to my last viewgraph this morning --
2 MR. SIESS: Keep in mind.that Millstone will have 3 had a seismic PRA cone. On three plants. 4 MR. MURPu7: Uh-huh. l 5 MR. SIESS: By then. ' 6 MR. MURPHY: Right. 7 MR. SIESS: And I am just trying to figure -- how 8 does this apply to sites that already exist, that already 9 have plants, that already have been looked at? I 10 Do we start over? I think -- J l 11 MR. MURPHY: No. We don't have to start over from ! 12 ground zero, but we do have to review the material that we 13 do have in hand. 14 And let me beg your indulgence. Let me get the l 15 last slide, and then we will talk it through. 16 I think we can show you where the differences i 17 would be from a "whole new application" and an application 18 for a new plant at an existing site. 19 MR. SIESS: Keep in mind, now, that the l 20 regulations do not include the reg guides. 21 MR. MURPHY: That's right. 22 MR. SIESS: The regulation that would apply to an 23 application for construction at Millstone would be what's in l 24 Appendix B. 25 MR. MURPHY: That's right.
I 1 so 1 MR. SIESS: Right? 1 2 MR. MURPHY: Yes, sir. 3 MR. SIES: Okay. Go ahead. I 4 [ Slide.] 5 MR. MURPHY: I think we have touched on a good bit 6 of this already. . Particularly the first bullet. The 7 revision of Appendix A involves a new Appendix B that will 8 contain the geology and seismology. 9 l A new Appendix S to Part 50, which as we all said, l 10 will contain the design of the engineering parts of the old 11 Appendix A. l ! 12 In order to carry this out, and what we are doing 13 is simply list the various pieces of it -- but what we are 14 doing for the earth sciences for Appendix B is that we end 15 up with a new reg guide and a revision to the standard l 16 review plan section. l 17 This new reg gaide is the one 1015, which 18 Dr. Siess has been referring to, and the SRP section is the l 19 one that Phyllis will be talking about on the ground motion. l 20 The engineering -- we end up with one revised reg 21 guide, which is the one on instrumentation. And two new reg l 22 guides. These are the ones that describe pre-carthquake l i 23 planning, and then the restart guidance following an 24 r,a rthquake . 25 [ Slide.] 4 l
$1 1 MR. MURPHY: This viewgraph indicates the 2 applicability of the new regulation. Basically it is a 3 forward fit. There is no intention to backfit this 4 regulation onto operating facilities. We will be licensing 5 probably, most probably, standard plants in the future, and 6 that has inf?.uenced some of the thoughts that went into the 7 regulation, the new regulation.
4 8 The other thing to take note of, as we've said 9 earlier, is that we anticipate that if there are licensing 10 issues associated with the siting of the facility, setting 11 the design levels and so forth, that if there are new issues 12 for existing plants, we will consider on a case-by-case 13 basis using the new technology in the two new regulations. 14 (Slide.) ) 15 MR. MURPHY: Scheduling. It says here "were i I 16 integral.with the source term part of Part 100." This was a ; l 17 decision that was made by the commissioners, and this l 18 decision was in order to have these regulations in place for 19 any early site review. 20 And obviously we are meeting with you folks today, 21 and as noted there, and as needed in the future. 22 We anticipate issuing the whole package, which is 23 the two new regulations and -- I'll call them the five 24 guidance documents, the four revised or new Reg Guides and 25 the Standard Review Plan -- that whole package, the whole i
b2 1 inch and a half, two inches in front of you, to be issued 2 for public comment this spring. We're looking for a comment 3 period through the spring and the summer and then back for a 4 final issue in early 1993. 5 We have established a technical panel to provide 6 us with guidance and comment. We have met with those folks 7 three times. . 8 MR. SIESS: At what stage in your documentation i i 9 have those three meetings been held? Has that panel seen 10 the material we've got? 11 MR. MURPHY: They have a copy of the material that i l 12 you've got there. They have seen earlier versions of it. 13 The'first meeting that we had with them I'd say was 14 basically like a brainstorming kind of meeting. The next 15 two meetings were based upon some draft material that we had 16 proposed, and we have gotten their comment back on that. 17 MR. SIESS: The provided comments in writing to l 18 you? j 19 MR. MURPHY: Some of it has been in writing. Some l .- 20 of it has been oral at our meetings and in phone l 21 conversations subsequent to those meetings. 22 MR. SIESS: What additional meetings do you plan? 23 MR. MURPHY: I would say that depending upon the l 24 outcome of this meeting and our meeting with CRGR, that 25 there will be additional interchange at that time. I would 4 3
$3 1 presume that after we've gotten the public comment back and 2 the Staff has compartmentalized what the comments were, that 3 we would get back to them again at that stage, and then at 4 least once more when the Staff came up with close to a final 5 draft of the public-commented document.
6 So I would guess another three or four meetings 7 with these folks. . 8 MR. SIESS: Can you keep us informed of those 9 meetings? 10 MR. MURPHY: Sure. 11 MR. SIESS: Could we send a representative? 12 MR. MURPHY: Yes, sir. 13 MR, SIESS: At least to listen. It's not likely 14 to be me en that schedule, but -- l l 15 MR. CHOKSHI: Just one clarification. They 16 haven't seen the trial applications. They haven't seen the 17 calculations we have done for site, which we'll be showing 18 to you. So we will be sending them those. 19 MR. MURPHY: In fact, the material that Don will 20 be presenting. 21 We've had a couple of public meetings, like I 22 said, principally with.the folks from NUMARC and EPRI. As I 23 understand it, they have set up an ad-hoc group at NUMARC, 1 l 24 but it hasn't been too interactive over the last three to l 25 four months, and we've done the usual good things of
3' 4 1 1 coordinating with the appropriate groups within the NRC. 2 (Slide.] 3 MR. MURPHY: These are the experts that we have 4 been meeting with. We've gone back and picked out some of 5 the old names, the folks that have worked with Appendix A at 6 various stages, have been around when the warts and the . 7 wrinkles were discovered, and we've been, like 1 said, we've 8 been making use of these folks. . 9 MR. SIESS: I think Jim Devine might have been 10 involved in the original. 11 MR. MURPHY: Yes, sir. 12 MR. SIESS: I'm not sure. I was. Dave Okrent and 13 I were at all the meetings on the original Appendix 7., and 14 I'm pretty much recalling Devine was, too. I'm not sure 15 about anybody else there. Among the original perpetrators. 16 MR. MURPHY: Well, there are some of the original 17
-- no perpetrators, the ones that had to live with it as it 18 was first being applied.
19 (Slide.) 20 MR. MURPHY: This is my last viewgraph. I think 21 this one contains the meat of the matter as to what's going 22 to happen in the future. . I presume that you can see most of 23 it. 24 Basically we're talking about a two-path system. 25 The deterministic path'is very, very close to what we have
$5 1 today. There have been some changes inLthe terminology, and 2 Dick McMullen will tell you about that, that we are using in 3 the determinir, tic path. There have been some changes 4 associated with things like tectonic province that were 5 necessary. There have been some changes associated with 6 things like maximum credible earthquake, redefining and 7 renaming these things.
8 But basically we follow the same path that has 9 been there in the past. We identify the seismic sources, 10 and this is a now term that Dick will define for you. For 11 those seismic sources, we, in effect, determine what we are 12 calling right now just as a term " deterministic earthquake"; 13 in effect, the largest expected earthquake within that 14 source zone. 15 Then based upon that and using the same 16 groundrel .9s that we have in the past, we cbtain the 17 deterministic controlling earthquake. These are the 18 earthquakes that provide us the ground motion for the 19 particular facility, the particular site of the facility. 20 And there right now we are characterizing those in terms of 21 an M and a D, a magnitude and a distance. 22 Here we're talking about a site that may have one l L 23 or two of these. There are some sites that are partially l 24 controlled by a moderate nearby earthquake and also 25 influenced or controlled by a' larger magnitude earthquake at i
1 56
- j. 1 a greater difference.
i, 2 ! MR. SIESS: What about a site that doesn't have ' ! 3 any nearby earthquakes, central stable? They used to be i j 4 called "ce c.';ral stable. " i 5 MR. MURPHY: f Than that would likely be controlled 6 by a larger, more distant earthquake. l 7 MR. SIESS: I'm not sure I get that. Take a site ? - 8 l -- take Dresden, for example. What would control Dresden? l 9 MR. MURPHY: Phyllis, do you know Dresden? 4 J l 10 MR. SIESS: Take Clinton. That's a fairly recent j 11 one. Somebody ought to-be around. I know that New Madrid - I f 12 - I mean, the Wabash Valley fault zone controls some low-
- )
j 13 frequency stuff there, but that's not really important. ! i l 14 What gave you the SSE for Clinton? ) I
- 15 MR. MURPHY: Can you answer that one, Phyllia?
j 16 MS. SOBEL: My name is Phyllis Sobel. I'm with l i 17 NRR. 1 18 MR. SIESS: Get close to the mike. 19 MS. SOBEL: I think for most sites licensed in the 20 1980s in the central part of the country that were not near 21 the New Madrid seismic zone, we used an intensity 7 or a 22 magnitude 5.3. 23 MR. SIESS: Yes, but that wasn't what Ar.dy said 24 would be a nearby or a faraway earthquake. Now you've just 25 told me what you did, but neither one of those is related to
$7 1 any existing earthquake. 1 2 MS. SOBEL: Okay. I think by " faraway", he may j have meant that we would consider a seismotectonic province j
3 4 that was large enough that it would encompass an event of 5 that size. 6 MR. SIESS: Okay. So North Anna, Ohio or r 7 something like that was the basis for your 5.3 and 7. 8 MS. SOBEL: Right. 9 MR. MURPHY: Right. i l 10 MR. SIESS: Okay. That's what we're doing now. 11 MR. MURPHY: That's right. I 12 MR. SIESS: Okay. 13 MR. MURPHY: What we're proposing -- 14 MR. SIESS: That's not quite the way you described 15 it. 16 MR. MURPHY: Okay. l 17 MR. SIESS: It's not into that so-called central 18 stable province. We simply said there have been some 19 earthquakes in areas like this, and that's what we're going 20 to use. , 21 MS. SOBEL: I think the deterministic path that 22 Andy has described will, in spite of the changes in 23 definitions, will result in similar earthquakes. I 24 MR. SIESS: It's not any more deterministic nov 25 than it was then. l
l l
$8 1 MR. MURPHY: No. But it's a name that's '
I 2 recognized. That's all we're depending upon there. t l 3 MR. SIESS: In other words, if I look at those two l 4 l l paths, if I went that righthand path with what we know now, 5 I would have a real problem at a site like Surry onca l 6 Charleston was unleashed, wouldn't I? ! 7 MR. MURPHY: That's right. If Charleston were 8 unleashed, you'd have a problem at Surry. 9 MR. SIESS: But if I go down the lefthand side, I i 10 can come up with an acceptable value for Surry. ; I ! 11 MR. MURPHY: That's correct. l 12 MR. SIESS: Okay. ' 13 MR. MURPHY: Keeping -- if you want to say keeping I 14 charleston in its place. 15 The lefthand path, the probabilistic analysis 16 path, starts off with simply carrying out either an EPRI or l 17 the Livermore seismic hazard calculations for that site, 18 taking into consideration the information that is available. 19 We're not talking about modifying those documents at this 20 time, but simply using them as they exist today. 21 once we have that information, we simply make a 22 comparison with that information for the current facilities, 23 and right now we're saying that the new plants, the new l 24 sites, should be in the lower half of the population for the 25 current sites. ! I l l
i
$9 1 Nilesh's next presentation will get into the ,
2 details of-how this is carried out, but basically what we're 3 saying is that we make a comparison, and we expect the 4 hazard for the new sites to be in the lower half of the 5 population for the existing sites. 6 And then after we made those calculations, we i 7 daaggregate the information that was collected there to ' 8 obtain the comparable set of magnitudes and distances to 9 describe the earthquakes that, in effect, are controlling i 10 the dtsign through the probabilistic side. 11 When we have those two sets of magnitudes and , 12 distances, one from the deterministic path and then one from 13 the probabilistic path, we make a comparison of those. And 14 what we are anticipating is that, as we have seen in some 15 very preliminary trial cases, is that the numbers turn out 16 to be quite comparable and that the deterministic approach 17 seems to envelope what we are seeing with the probabilistic. 18 So based upon making this comparison, we take and [ 19 do a classical calculation of the SSE ground motion that i 20 will be used for design at that site. 21 MR. SIESS: That I'm not going to understand until 22 I see some examples. 23 Let me ask you another question that is not , 24 related at all to this, but it does relate to something I i 25 noted as I read through this stuff, and it's going to come i ( l l i I
80 1 up later. 2 We have a term that we use.very frequently in a 3 very important way called " seismic category 1 structures, 4 systems, and components." Where are those defined now? 5 MR. KENNEALLY: Roger Kenneally. Regulatory Guide 6 1.29 defines some of the functions and also within the 7 Appendix A, the definition of the -- 8 MR. SIESS: They're defined in Appendix A? 9 MR. KENNEALLY: Yes. 10 MR. SIESS: That definition is being moved to 11 Appendix S, am I right? 12 MR. KENNEALLY: It will be moved to S, and part of 13 it is also contained in -- 14 MR. SIESS: But it is not referred to there as 15 defining a seismic Category 1 system, structure, or 16 component. That same definition -- you know, it's a.three-17 part definition. 18 MR. KENNEALLY: Right, correct. 19 MR. SIESS: That same definition appears at least ! 20 once in every Reg Guide you've got here, in some cases ; 21 twice, and I keep wondering why can't somewhere you define a 22 seismic Category 1 SSE and simply use that terminology in 23 the remaining documents? 24 I'll point this out later, but I just want to be 4 25 sure that Appendix A was the only place we defined it, and I
kl 1 don't think even that' called it seismic Category 1. 2 Where does that terminology appear in the 3 regulations? 4 MR. CHOKSHI: I think the Reg Guide 1.29 is the -- 5 MR. SIESS: Is it in 1.29? And it uses the term 6 " Category 1"? 7 MR. CliOKSHI: Category 1. 8 MR. SIESS: Because that doesn't appear in 9 anything I've seen of the revised material. 10 MR. CHOKSHI: No, we are just using the same 11 definition. 12 MR. SIESS: And yet that definition gets repeated 13 about five times. I've lost track of them. But one Reg 14 Guide has got it twice, or one Appendix has it twice. 15 MR. KENNEALLY: That is correct. 16 MR. SIESS: Well, let's watch for that as we go 17 through, because that's a very important category. 18 MR. MURPHY: Right. 19 MR. SIESS: It gets used for a lot of other things 20 besides seismic, as I recall. 21 You're on for the next, Item 2, Andy or -- 22 MR. MURPHY: Pardon? 23 MR. SIESS: I said you are on for the next, Item 24 2. Or is this Appendix B overall? 25 MR. MURPHY: This is the Appendix B.
l ; l + i i !~ I d2 , 1 MR. SIESS: HDo you combine those two items? ' L 2 MR. MURPHY: In effect. i l 3 MR. SIESS: Okay, fine. t 4 MR. MURPHY: Okay. So next Nilesh will be talking 5 about the probabilistic criteria. 6 Just to keep or add to the confusion, this is i 7 Appendix A to the Reg Guide. 8 MR. SIESS: I'm just deciding whether it's too 9 early for a break. I think it is. . 10 Does anybody need a break? 11 (No response.] ' 12 ER. SIESS: Okay. Let's proceed. l i 13 [ Slide.] ' i 14 MR. CHOKSHI: Good morning. I am going to talk , 15 about the middle box on the flowchart that is shown on the I 1 16 lefthand side. I will be discussing about the probabilistic ! 17 criterion which goes with this proposed division. ( 18 I think Dr. Sisss already raised that question on i 19 who you're going to define the probability, some limit in 20 terms of exceedance. l 21 MR. SIESS: Wait just a minute. We're sort of 22 getting into the middle here somewhere. You're talking l 23 about Appendix A to Reg Guide DG-10157 i 24 MR. CHOKSHI: Yes. 4 , 25 MR. SIESS: Where are you going to talk about the j t l
l i k3 , 1 rest of it? i 2 MR. CHOKSHI: The part of the DG which covers the i 3 geologic investigations, Dick McMullen will cover that later 4 and the definitions, the changes in the -- 5 MR. SIESS: Well, let's take just a minute as I i s 6 glance through here. 7 (Pause.) 8 MR. SIESS: How much of this is valid? $ 9 MR. MURPHY: How much of it is valid? 10 MR. SIESS: Yes. It says: "This draft is : 11 currently under further revision," the copy I have. , 12 MR. MURPHY: A considerable amount of it is valid. 13 The principal changes will be associated with materials that 14 Don Bernreuter will be describing after Nilesh is done. i l 15 It's a matter of setting the tone. i 16 MR. SIESS: Okay. Now this draft does not have an 17 Appendix A. , 18 MR. CHOKSHI: No. , 19 MR. SIESS: What I have that has an Appendix A is , 20 the Summary of Draft Regulatory Guide. 21 MR. CHOKSHI: Yes, I think -- 22 MR. MURPHY: That's correct. L 23 MR. SIESS: And I've got ten pages of summary for 24 the 36 pages of guide. 25 MR. MURPHY: ~That's correct. L
L . l 44 1~ MR. SIESS: And those ten pages summarize those 36 2 pages. And than I've got an appendix -- I'm having trouble 3 with-the microphone and the paper -- I have an appendix l
)
'4 which tells me how to do the probabilistic part, as you 5- said, the lefthand side of the chart.
i a ! 6 MR. MURPHY: Right.
- 7. MR. SIESS: Assuming that the deterministic part .
8 is the same that we've been doing it, except for all the new ! 9 nomenclature. 10 MR. MURPHY: That would be a good way to 11 characterize it, right.- 12 MR. SIESS: Yes. Now at what point are we going j 13 to'have somebody explain the new nomenclature and the other
- 14 things that are changed in Appendix A to Appendix B7 l
15 MR. NURPHY: Dick McMullen will be explaining 16 that. The order that we have at this stage, and we took it l 17 by controversy, if you want. We figured that you were going 18 to be most interested in seeing what the probabilistic 19 analysis was going to be doing. That is what Nilesh chokshi 20 will be presenting. And then following that will be -- I'll 21 call them the case studies or examples that Don Bernreuter l l 22 will discuss, and then Dick McMullen will describe what 23 changes have happened to the classical investigations in i 24 Appendix A, the changes in the procedures and in the l 25 nomenclature. l l i
i
$5 f
1 MR. SIESS: I think that will work. 2- MR. NURPHY: I mean, we can change it if you want. : 3 MR. SIESS: No. I think you may have selected 4 about the best order, sort of working from the newest back 5 to the oldest, except right now I don't'have too much of a ;
- I 6 feel as to how old the old is, how much of the old is new.
7 MR. MURPHY: I understand. ,
, 4 8 MR. SIESS: In other words, whether seismogenic 9 provinces are similar to tectonic provinces or quite 10 different from a tectonic province and so forth. But I 11 think we can hold that order that you have here. <
L 12 Let's try it this way, if you don't mind. 13 MR. CHOKSHI: I think as you pointed out, Dr. 14 .Siess, that's probably one of the big challenges on how to 15 come up with a probabilistic criterion, particularly given i 16 the status of the current estimates of hazards in the. j 17 Eastern United States. , 18 Most of our discussion, I think a lot of the 19 discussion in the expert panel meetings was focused on how 20 to combine the so-called semi-probabilistic/ deterministic 21 approach. And what I want to discuss is the way we came up 22 with this criteria and some of the rationale behind it. , 23 One of the reasons why the Appendix A was 24 highlighted in summary was because it has also a lot of 25 discussion about the rationale and basis for that, and we
46 1 might go with a similar format for the public comment, so 2 that everybody understands how we arrived at that criteria. 3 [ Slide.] 4 MR. CHOKSHI: Now the proposed Appendix B 5 explicitly calls for more deterministic and probabilistic 6 evaluations to be performed, to be conducted, and it also 7 talks about the probability criterion, that it be similar to 8 that shown for the lower half of the population of the 9 currently operating plants. 10 (Slide.) 11 MR. CHOKSHI: Given this proposed regulation, the 12 objective of tnAs appendix is to outline a procedure, so one 13 can determine the probability of exceeding the design basis 14 and also establish a criterion to which it can be compared. 15 Such a tool would also allow for initial 16 screening, and I think one of the most important uses.is , i 17 Item 3 in the scheme of the things which we are now working I 18 on, is to establish probabilistic controlling earthquakes, 19 and Don will go into a lot of details on how exactly that is 20 done. 21 The appendix also discusses about the 22 probabilistic hazard to be used in a PRA. I'm not going to 23 talk much about that today, because I think the other part 24 is the one which has a lot of interest. 25 The appendix further goes on reiterating some of l
k7 1 the objectives of why to do -- why consider both 2 probabilistic and deterministic. I guess the point we tried 3 to emphasize is that more than anything else, more than 4 numerical criteria, the probabilistic will lay out 5 information which was somewhat implicit in a lot of old ' 6 Appendix A decisions. The recurrence period, the zones, and 7 all of that will be transparent, and it will be better for 8 people to judge. 9 MR. SIESS: Your last bullet, the hazard estimates 10 to be used in a PRA, for the certified plants, is a PRA done 11 before it's even built? 12 MR. CHOKSHI: Yes. A design-specific PRA. 13 MR. SIESS: Is there a requirement that another 14 PRA be done after'the site is selected or after the plant is 15 built? 16 MR. CHOKSHI: Well, right now that is in flux, but 17 as a minimum they would have to develop a site-specific 18 hazard to compare with what was used in the design time. 19 They use some sort of bounding curves, and as far as I can 20 recall, there is an interface requirement proposed on a 21 number of reviews that a site-specific hazard be calculated 22 and compared to assure that it is enveloped.
)
23 MR. SIESS: If I made a PRA using, say, the 24 seismic hazard curve for Site A and I put it at Site B, how 25 easy is it to come up with the PRA for Site B? l
i i I8 1 MR. CHOKSHI: If the plant is the same, systems ! 2 and components, then I think it is relatively easy to come ' 3 up with the PRA for Site B, unless there is -- the two which ! 4 could impact would be the site-specific features like the ' 5 soil conditions. 6 MR. SIESS: Yes. Leaving those out. Those would 7 be important. That has to be done after the site selection. ,. 8 MR..CHOKSHI: And if the hazard has a different, 9 drastically different slope characteristic, that that might 10 order -- change the order of some of the dominant ; 11 components, but it would come out of the analysis. 12 (Slide.) 13 MR. CHOKSHI: Okay. Let me go to the procedures. l 14 I think in terms of a probabilistic design basis, I think we ; 15 decided that the current generation of plants in terms of 16 the probability of exceeding design basis is safe enough. 17 That's where we are making our calibration consistent with a 18 lot of engineering practices when you calibrate yourself 19 against an existing design or core requirements. 20 And given the current status that we have the two 21 methods, which are using quite a bit different estimates for 22 a site, the focus is on using this hazard in a consistent 23 but relative fashion. 24 So I think that's one of the very important -- the i
- 25 criterion goes with th's method. If you use the Livermore i
. -. . . .. . . . . - - . = . . . - . ~ . - - . - . . . . .- . . . -- -.. .-
l N9 1 method, there's one criterion; if.you use the EPRI method, 2 there's another criterion. , 3- MR. SIESS: Say that again. 4 MR. CHOKSHI: The probability of exceeding design ,
.5 basis,'the criterion, it's pretty much established for one 6 for each method. There is one set-of numbers for Livermore, 7 one set of numbers for EPRI.
8 MR. SIESS: One set of what? 9 MR. CHOKSHI: For the Livermore hazard curves, if 10 you use the Livermore hazard curve, there is one set of 11 numbers you will compara against. And I'll show you in a 12 minute. 13 MR. SIESS: Do you consider the EPRI or Livermore
- 14 curves uniform hazard spectra?
I 15 MR. CHOKSHI: Yes. In the development of this 16 criterion, we considered them, right. 17 MR. SIESS: Maybe I'm not sure I know what -- 18 because I read something where you were. talking about 19 uniform hazard spectra as something different. What is a , 20 uniform hazard spectrum? 21 MR. CHOKSHI: Okay. Let me switch just for a 22 minute to the next viewgraph. . l 23 [ Slide.] 24 MR. SIE3S: That's hard to read on the screen. 1 25 Okay, we've got it here. Fine. It's hard to read there. j 1 l
I 50 1 MR. CHOKSHI: Okay. These are an example of 2 uniform hazard spectra for a site for five different periods 3 of years or five different probabilities of accidents -- 4 10,000 years, 5000, 2000 and 1000 and 500. l 5 Superimposed on that is the design spectrum. 6 MR. SIESS: Uniform hazard spectra, to you -- each 7 one of those lines is a uniform hazard spectrum? . 8 MR. CHOKSHI: Right. Associated with a different 9 period of years. One is'for the 10,000 -- the topmost is 10 for the 10,000-year period. 11 MR. SIESS: Okay. And if I have a ic 000-year 12 spectrum for each plant, that's a set of uniform hazard 13 spectra, or is this a uniform? 14 MR. CHOKSHI: This is a uniform hazard spectra for 15 a site. So for each site, you will have a series of uniform 16 hazard spectra depending on the -- l 17 MR. SIESS: " Uniform" means constant? 18 MR. CHOKSHI: The probability of exceeding is a l 19 constant on that curve. 20 MR. SIESS: " Uniform" relates to that particular 21 site and that particular curve. l 22 MR. CHOKSHI: Right. 23 MR. SIESS: Not to uniformity among sites. 24 MR. CHOKSHI: That's right. j 25 MR. MURPHY: That's correct, Doctor. 1 ? a
__ .- - . - - ~ - - _ . . _. - - . . - .
$1 1 MR. SIESS: Why do you call it " uniform"?
2 MR. CHOKSHI: Because at each frequency, the 3 probability of exceeding the spectra is constant. 4 MR. SIESS: Is constant. 5 MR. CHOKSHI: Right. 6 MR. SIESS: So why not call it " constant"? Thank 1 7 you, because I was completely misled somewhere in my 8 education. I thought uniform hazard spectra for the United 9 States, for example, referred to a whole group of sites, but 10 you are using it for a particular location. 11 MR. MURPHY: Don, do you want to add anything? 12 MR. SIESS: Uniform hazard pl.ots. 13 MR. BERNREUTER: Don Bernreuter with Lawrence 14 Livermore National Labs. 15 I think the terminology, " uniform hazard spectra", 16 just ought to -- 17 MR. SIESS: Louder. I can't hear you. 18 MR. BERNREUTER: The terminology, " uniform hazard 19 spectra", I think developed years and years ago. Actually 20 what it is, is an equal probability of exceedance spectra. 21 MR. SIESS: For a site. 22 MR. BERNREUTER: For each site. That's correct. 23 MR. SIESS: Well, now what about uniform hazard 24 maps? Aren't there such things? 25 MR. MURPHY: Yes, there are.
~ . . - - . . . . -.._.
$2 1 MR. SIESS: That's what I thought.
2 MR. MURPHY: Yes. 3 MR. SIESS: But that now would be for one 4 probability covering all sites in an area. 5 MR. MURPHY: Tha 's right. 6 MR. SIESS: Okay. That's what was confusing me. 7 [ Slide.) 8 MR. CHOKSHI: Okay. I think as Andy showed on the 9 lefthand side, if you use the Livermore or EPRI method, you 10 will generate a set of uniform hazard spectra like I showed 11 you before for a site. 12 So the Step One is to generate this uniform hazard 13 spectra, and then Step Two consists of three parts. The 14 first is to estimate the annual probability of exceeding the 15 design spectrum, and we have defined two discrete 16 frequencies. 1 17 This issue was discussed quite at length with our I 18 expert panel, and the selection of two frequencies important 19 to the structures and components are selected, 5 and 10 20 hertz. 21 So in order to carry out this calculation, 22 pictorially you would superimpose the design spectrum on 23 these curves, and then you would go -- 24 MR. SIESS: I'm sorry. I'm finding this extremely 25 difficult to follow expressed verbally. There must be' l l l
$3 1 examples you can give of numbers for these two procedures 2 that can make sense, because the words involve too many 3 words that you define one way and I might define another 4 way, and I just can't follow it.
5 MR. CHOKSHI: Okay. Let me go to -- 6 [ Slide.) $ 7 MR. CHOKSHI: This is for a site, for one of the 8 current sites. 9 MR. SIESS: I need to have an example in terms of 10 a site and all of the information that is used for that 11 site, all the way down through the process of coming up with 12 expected maximum earthquakes, everything right on down. 13 MR. CHOKSHI: Yes. Don is going to go through 14 that. l 15 MR. SIESS: And philosophically there may be 16 differences that can be explained, but these are not 17 philosophic statements. These are procedural outlines here. 18 MR. CHOKSHI: Right.
- 19 MR. SIESS
- And it's just impossible to follow in 20 words.
21 MR. CHOKSHI: Yes. I think -- ! 22 MR. SIESS: Once you've seen the numbers, it may 23 be that the words make sense, but they don't know. 24 MR. CHOKSHI: I think the actual calculation, I 25 think Don is going to go through on this probability of
.- . ~ . -- .
54 1 exceeding. 2 MR. SIESS: But now these are simply two steps, 3 not two procedures. 4 MR. CHOKSHI: Right. 5 MR. SIESS: And I could apply these steps with 6 either the EPRI or the Livermore -- 7 MR. CHOKSHI: Right.
. i 8 MR. SIESS: -- and I will get two different 9 answers.
10 MR. CHOKSHI: Right. l 11 (Slide.) I 12 MR. CHOKSHI: Now this is the -- I think this is i 13 where the criterion comes in. 14 MR. SIESS: All right. Now that's an interesting 15 curve. l 16 MR. CHOKSHI: Okay. This is a plot for the 17 current plants. On the -- 18 MR. SIESS: How many sites? 19 MR. CHOKSHI: About 69 sites and 110 plants, 20 something like that. 21 MR. SIESS: Well, it's not the plants that counts. l 1 22 It's the sites. l l 23 MR. CHOKSHI: This is the probability of exceeding l l 24 -- for current plants exceeding their design basis. It uses 25 the same calculational procedure which was on the previous l . 4
$5' 1 graph, and Don will go through it.
2 MR. SIESS: All right. This is the probability of 3 exceeding the design basis expressed in terms of what? Frec 4 field, zero period, acceleration? 5 MR. CHOKSHI: Free period, 5 and 10 hertz, an 4 6 average of 5 and 10 hertz spectral ordinates. Instead of 7 going to PGA, we are using the spectral measures, because 8 that more reflects -- 9 MR. SIESS: What spectral value? Acceleration? 10 velocity? 11 MR. CHOKSHI: Acceleration. 12 MR. SIESS: Acceleration. 13 MR. CHOKSHI: Right. l 14 MR. SIESS: This is the average of the free field 15 response spectrum acceleration at 5 and 10 hertz? 16 MR. CHOKSHI: Right. ; 17 MR. SIESS: Not free field response spectrum, ; 18 because it's -- 19 MR. CHOKSHI: This is one spectrum. Each site has 20 -- 21 MR. SIESS: Okay, okay. 22 MR. CHOKSHI: For example, for a site under the 23 Reg Guide 1.60 spectra, this would be the 1.60, not whatever 24 PGA, but the response at 5 and 10 hertz. 25 MR. SIESS: But it's not the PGA.
\
$6 1 MR. CHOKSHI: It's not the PGA, no. And that's 2 one of the issues which we are looking into, that what is 3 the effect of looking at different frequencies, PGA.
4 MR. MICHELSON: Could you help me for just a 5 minute? 6 MR. CHOKSHI: Yes. 7 MR. MICHELSON: Since you're much more acquainted ' 8 with this subject than I am. 9 If I determine a certain probability of exceeding 10 a given value, my interest then is, so what? What effect l 11 will it have on the plant? And it depends upon how much you ;
- 12 exceed --
l 13 MR. CHOKSHI: That's right. 14 MR. MICHELSON: -- that value. All this tells me I l 15 is that I have a certain probability of exceeding, but it 1 i l 16 doesn't tell me the consequence. 17 MR. CHOKSHI: No. 18 MR. MICHELSON: Because I don't know by how much 19 you've exceeded, and I don't know the equipment. 20 MR. CHOKSHI: That's right. We are strictly here 21 dealing with siting. l 22 MR. MICHELSON: Yes. And with the probability of 23 exceeding, but I'm trying to relate that somehow to risk. j 24 MR. CHOKSHI: You can't. 25 MR. MICHELSON: I just can't do it? l l 1 l
( 57 i 1 MR. SIESS: You can. All we get here is the I 2 earthquake to design for, the ground motion to design for. 3 Now how much margin you've got and what the I ( 4 consequences of exceeding becomes fixed by either the l 5 conservatists in picking the earthquake, which really gets : i - ! 6 wiped out.if you're talking about exceeding it, the l l 7 conservatisms in design, allowable stresses versus limits. 8 And the only way you get any feel for that is by 9 looking at the seismic margin studies that have been made. 10 MR. MICHELSON: But I'm trying to determine how I 11 know I've got sufficient confidence in the number that we - 12 come up with here. 13 MR. SIESS: Well, if plants are designed in the 14 future the way they have been in the past, our seismic
- 15 margin studies tell us that it takes two to three times the l
l 16 design-basis earthquake to have a significant probability of 17 core melt. i 18 MR. MICHELSON: Well, if you*re dealing only with 19 effects on structures -- 20 MR. SIESS: Structures, systems, and components. 21 MR. MICHELSON: If you're dealing with relay 22 chatters and things like that, it isn't like two or three. 23 MR. SIESS: Well, but again, the seismic margins , 24 and relay chatter -- 25 MR. MICHELSON: Yes, but it's a different issue, d 4 _. - - -= _ .,.
l l 58 1 and it may become the limiting factor. 2 MR, CHOKSHI: Yes, but I think, Dr. Michelson, 3 when you are doing a new design, you are going to require 4 them to do a PRA anyway. So this is strictly -- this-5 approach would make it -- 6 MR. MICHELSON: But in the PRA they do, they're 7 going to start out first with the probability of exceeding . 8 the design basis. 9 MR. SIESS: No, no. 10 MR. CHOKSHI: Full hazard curve. 11 MR. SIESS: For the PRA, they use the full hazard I 12 curve. This just locates it with respect to existing 13 plants. 14 MR. MICHELSON: But we don't have the full 15 response of equipment to the full hazard curve necessarily. 16 It depends on -- 17 MR. SIESS: That's what you do in the PRA, the 18 fragility curves. 19 MR. MICHELSON: Yes, but we don't -- well, we 20 hopefully have the data, but -- 21 MR. SIESS: Well, if you don't have them, you 22 can't do a PRA. l '23 MR. MICHELSON: Yes, that's right. The problem 24 is, by how much are you going to exceed these design bases. 25 2:en you can tell whether you've got data for that
I 59 1- acceleration. , 2 MR. CHOKSHI: As I think Dr. Siess said, one of 3 3 the reasons is that the plant -- we have a different margin J 4 and different design. 1 5 .There's another reason. You know, when we are 6 looking at the probability of exceeding design basis, it's ) 7 one acceleration level, at one point on the hazard curve, 8 and the slope and other things could have a tremendous 9 impact on the overall core damage or the PRA results. 1 10 So I think we have gone to great pains to explain ; 11 that in the appendix, because I want to put it out for f 12 public comment. So that point is very clear, that we are : 13 not making any statement about plant core risk or -- 14 MR. MURPHY: That's right. 15 (Slide.) 16 MR. CHOKSHI: Okay, going back -- 17 MR. SIESS: I guess you're going to leave it to 18 somebody else to apply this now to an actual plant-specific i 19 PRA. 20 MR. CHOKSHI: Right. 21 MR. SIESS: That's not your job here today. ( , 22 MR. MURPHY: That's right. 23 MR. CHOKSHI: What this covers is again the 24 probability of exceeding design basis for the current 25 plants, and this is basically the plants which fall below 1
i I l
$0 1 these limits.
2 MR. SIESS: Now I don't know whether those -- ! 3 there's an item on there that says " ordered using median" -- 4 "med", which could be median, I guess. 5 MR. CHOKSHI: It is the median. 6 MR. SIESS: Log hazard? 7 MR. CHOKSHI: These are the medians. Okay. These 8 were -- the design bases were compared with median uniform 9 hazard spectra. 10 MR. SIESS: Median what? 11 MR. CHOKSHI: Uniform hazard spectra. 12 [ Slide.] I 13 MR. CHOKSHI: Okay. These uniform hazard spectra l 14 -- it's a probabilistic spectra, and each line here has a 15 distribution around it. It has a median, mean, 85th 16 percentile and differiant radius measures. 17 The plot I ahowed you was comparing median spectra l 18 with the design basis. In the appendix that I will -- 19 MR. SIESS: Median? 20 MR. CHOKSHI: Median uniform hazard spectra. 21 MR. SIESS: Hazard spectra at -- which hazard a 22 spectra? 23 MR. CHOKSHI: The uniform hazard spectra. 24 MR. SIESS: The uniform hazard spectra. l 25 MR. CHOKSHI: The one which -- this plot was
i
$1 i
i derived by. comparing median uniform hazard spectra with the 2 design basis. That's why it says median. 3 MR. SIESS: Well, the uniform hazard spectrum is 4 at one probability, according to your definition. 5 MR. CHOKSHI: Right. 6 MR. SIESS: How do I get a median of a 10 to the i
-7 -4 uniform hazard spectrum?
8 MR. CHOKSHI: It's still a 10 to the -4, but there 9 is a median, mean, 85th percentile. There's a distribution 10 around 10 to the -4 response. 11 MR. SIESS: You'd better draw me a picture of 12 that. 13 [ Pause.) 14 MR. CHOKSHI: If I have a 10 to the -4 -- 15 MR. SIESS: Label the axes. 16 (Pause.) 17 MR. CHOKSHI: Yes. 18 MR. SIESS: I used to have a professor that always , 19 plotted without labeling the axes. We called it his 20 universal curve. I 21 MR. CHOKSHI: If I were to draw a uniform hazard 4 22 spectra for only the 10,000-year period -- 23 MR. SIESS: All right. That tells me that's the 24 spectrum for that probability. 25 MR. CHOKSHI: Right. But that would only be one l J
N2 I 1 curve because of uncertainties in different parameters. I 2 would have a series of curves, okay, with the different -- 3 they would all be before the 10,000-year period but with 4 different probabilities. 5 MR. SIESS: Okay. 6 MR. CHOKSHI: So you have a distribution around 7 the uniform hazard spectra. 8 MR. SIESS: And so you are using the median? 9 MR. CHOKSHI: Right. And in the appendix, we talk 10 about different cases -- median, mean, 85th percentile. And = 11 when Don is going through his example, he will show what are 12 the implications of those things. 13 MR. SIESS: Okay. So now if we take those medians 14 -- 15 MR. CHOKSHI: Right. 16 MR. SIESS: What are the rest of those notes on 17 there: frequency 7, based on 5 g-experts? 18 MR. CHOKSHI: Okay. These are the Livermore with 19 the Ohio ground motion experts. Frequency 7 means it's a 5 20 and 10 hertz average. 21 MR. SIESS: So the computer just doesn't know when 22 to stop plotting. It also doesn't know how to spell. 23 MR. CHOKSHI: [ Laughing.) Yes, I see that, 24 "probilistic", yes. 25 So this is the current probability of exceeding y _ _ _ _ _ _
- _-- - - - _ _ - _ - -. .~. . - . . - . - - - . . . .
53 i 1 median spectra for the -- 2 MR. SIESS: Okay. i 3 MR. CHOKSHI: The proposal'is to that for the new ! 4 design, that it should be less than where 50 percent of the 5 plants currently are, and that roughly for this curve is 1 x 6 10 to the -4. 7 MR. SIESS: Now you obviously have a lot of 8 choices. 9 MR. CHOKSHI: Right. 10 MR. SIESS: You can rule out zero. You probably 11 can rule out 100. How (lid you pick 50? l 12 MR. CHOKSHI: Escause it's just a judgment, and 13 the expert panel and everybody said that's an even number, 14 because it's half <of the plants are currently below that. 15 And as I'll show in the next viewgraph, what does 16 it mean when you look at the standard design. l 17 MR. SIESS: You know, 50 is an unusual number. 85 18 is the number we used, 84. 19 MR. CHOKSHI: Well, I think there are a couple of 20 things that are characteristic of this curve. 21 MR. SIESS: Well, this says -- what we're saying 1 22 is that these new sites are going to be worse than half the 23 sites we've got now. 24 MR. CHOKSHI: That's true. , 25 MR. SIESS: Right?
l
$4 l 1 MR. MURPHY: ,'Etally we're saying it the other 2 way around.
3 MR. SIESS: You ca, but I differ. 4 MR. CHOKSHI: There n e a number of ways to look 5 at this. For example, if we ha.d 1000 plants, if you 6 strictly follow this limit, this distribution could approach 7 something like this in limit. It will be all centered . 8 around this criteria. 9 So I think what we are trying to do is guard 10 against that the -- you know, the curve doesn't go farther 11 out, too much out to the right. It will be -- on the upper 12 portion, it will be always to the left. 13 MR. SIESS: What you are saying is that the 14 probability of exceedance -- and this is the probability of 15 exceedance of the SSE -- l 16 MR. CFOKSHI: SSE, right. , 17 MR. SlESS: At each plant. 18 MR. CHOKSHI: At each plant, yes. 19 MR. SIESS: And if we had been real smart 25 years 20 ago and picked 10 to the -4 as a basis for siting, then all 21 of our sites would have a g-value such that the probability l
- 22 of exceedance was 10 to the -4; that's what your blue curve 23 is.
24 MR. CHOKSHI: Right, i 25 MR. SIESS: Now we didn't do that. In fact, if we 4
l l
$5 1 had tried to do it, I don't have the slightest idea of what 3 1
2 number we would have come up with. Twenty-five years ago, <
~
3 some of us thought -- well, 24 years ago -- some of us J 4 thought the probability of exceeding the SSE was zero, and 5 some of us didn't even think in probabilistic terms. l 6 But if we had had to pick a probability -- well, I 7 can think of at least one person on the ACRS at that time ! 8 that would have picked 10 to the -6. That wasn't me. 9 So now rather than trying to pick a number, we're ; 10 trying to backcheck and see if we can arrive at a number by-11 looking at what we ended up with. ; 12 MR. CHOKSHI: That's right. That's exactly right. 13 MR. SIESS: And I don't see why 50 percent is any ; 14 better than any other number. 15 MR. CHOKSHI: There's really no solid reason for 16 why 50 percer.t, except for that, you know, that's -- l 17 MR. SIESS: It's just the median. If we had a PRA ; 4 18 for every one of these plants, a seismic PRA, you know, some i 19 of the plants have got only -- have got a very high ! ~ 20 probability of exceeding the earthquake; 10 to the -3 is f 21 fairly high considering other things. l 22 I'd expect their seismic risk to be a little 23 higher, maybe unacceptably high. I don't know. - 24 Some of them have, you know, a very, very small 25 probability of exceeding it. There I would expect the t
?
---4.e --, s ,e-4he.4. m e aJ ,id- ,-eed-M- d +A & & -w---m. ,4L..a A ++J -m44a-- lau+44Md.*-2sia_AJL.i..+#a. 4 g .m JaA4 .Ja 4. sw_d. 3,,#4.,,,J,,,yd. 4& w e
-4*. awa, am.mm- am 3 l
$6 1 seismic risk to be negligible. That won't necessarily be, l 2 but I might expect it to just looking at this. l l
3 MR. CHOKSHI: But there is another reason to take 4 something -- 5 MR. SIESS: Why couldn't I look at seismic risk as 6 a basis for thinking what's a. good -- 7 MR. CHOKSHI: That would -- for one thing, that 8 would couple design and the site. We would have to have the i 9 full information in order to -- 10 MR. SIESS: Well, I've got two -- I've got at 11 icsst one standard design out there. 12 MR. CHOKSHI: Right. I'll show the next curve, 13 which will illustrate, you know, what happenn with .3 g. 14 MR. MICHELSON: You don't know enough about that 15 standard design to do a seismic PRA on that design. That's l 16 the problem. 17 MR. MURPHY: That's part of what -- what we were i 18 instructed to do was to be able to decouple the two. , i 19 MR. SIESS: All right. But I could still -- I j l 20 could take a standard design and make whatever assumptions I 21 needed to do a seismic PRA, and then I could set myself some ' s 22 sort of goal for seismic risk. If I don't want it 23 contributing more than 10 to the -6 or 6x10 to the -6 or l 24 some other number to the core melt probability, or I don't j 25 want it contributing more than 10 to the -7 to the larger i l i
- --e &-- 3,-y- m- - -, , , - - - - - -- - -
$7 1 leak probability, and then I could find out.what hazard 2 curve, as a distribution, would give me that bottom line 3 safety, and I could arrive at a number.
4 It might not be 10 to the -4. It might be 10 to 5 the -2. 6 MR. CHOKSHI: But, Dr. Siess, there are two 7 points. This number is not an absolute number. If I -- for ' a the EPRI, this may be 10 to the -6. So we are trying to P 9 avoid to come down to -- because right now we don't have a i 10 stability in the number. 11 MR. SIESS: No, it's not an absolute number, and I 12 guess the only way we could handle that would be to get 13 five or six or eight or ten or maybe 50 other organizations 14 to do what EPRI and Livermore have done and then get a , 15 distribution on it. I mean, we've got two points. 16 MR. CHOKSHI: Right, we've got two points. 17 MR. SIESS: If there are reasons for those numbers 18 being different other than randon -- if they're non-random i 19 differences, we ought to be able to settle them. - l e i 20 That's out of the realms of probability. If t 21 somebody is conservative because they're on the regulatory l . 22 side, and somebody is less conservative because they were on 23 the industry side, those are non-random, non-probabilistic l 24 things that we ought to be able to factor into our safety 25 thinking. !
, - , . - . ~ ,
d
$8 1 I've got no problem with 50 percent. I'm just 2 trying to --
3 MR. CHOKSHI: But also there is -- 4 MR. SIESS: Because it does tell me that these 5 plants are worse'than 50 percent of the plants out there, 6 whether I'm using EPRI or Livermore. 7 If I judge the safety of a plant against 8 earthquakes, by the EPRI scheme I get one answer, but still 9 whatever I come up with, if that governs, is going to be 50 10 percent, worse than 50 percent -- 11 MR. CHOKSHI: This will be a theoretical limiting 12 case in reality, because this curve contains data from the 13 very low seismic side, old vintage side, which used the 14 hazard spectra and things in reality with the design, but 4 15 the curve is going to be like this, you know. 16 MR. SIESS: This is going to be skewed by the
,17 minimum .1 g.
18 MR. CHOKSHI: . Right, exactly. So that's the 19 reason, you see, if you fit too low on this curve, you're 20 going to really severely, you know, be basing your decision 21 on a very low-level hazard site, and in reality the curve 22 would be, you know -- you'll be better than, I think -- auch 23 better off than the current generation. 24 MR. SIESS: I'm just trying to visualize your 25 explaining this to the Hearing Board.
$9 1 MR. CHOKSHI: I think or.e of the earlier questions 2 you raised now that we have standard designs and .3 g --
3 MR. SIESS: But for all practical purposes, for a 4 Livermore type analysis, it comes out 10 to the -4 5 probability of exceedance as your probability-based design 6 SSE. 7 MR. CHOKSHI: Right. 8 MR. SIESS: For EPRI, it will be probably what, 10 9 to the -5 or 10 to the -67 10 MR. CHOKSHI: Yes. On the median. 11 MR. SIESS: So whatever technique, if you use one 12 set of curves, you use one number. 13 MR. CHOKSHI: Right. 14 MR. SIESS: And if you use the other, you use 15 another number. 16 MR. MURPHY: Right. 17 MR. SIESS: The ought to give you the same SSE, 18 because they've got different probabilities of exceedance. l 19 MR. MURPHY: Yes. And here it's not just a matter 20 of having the median, we're also looking at the mean and the 21 84th percentile as well for both sets. So in effect, there 22 are three numbers for each set for two sets. l 23 [ Pause.] l l 24 MR. SIESS: I think the process may be better than l l 25 the number you'get. l l
P h I 90 1 MR. CHOKSHI: Yes. The number is strictly a 2 relative measure. But as you will see -- and I think this l 3 is just to give you an idea, if I -- this is the same 4 information, but as giving that each current plant is 5 designed for .3 g 0098 median spectra. ; l ' 6 (Slide.] ; j 7 MR. SIESS: Okay. Now just a minute. This is the , ! 8 probability of exceeding .3 g at current sites? l 9 MR. CHOKSHI: Right. 0098 spectra, median l 10 spectra. And this is not a direct comparison of .3 g and 11 Reg Guide 1.60, but this will give you an idea. The limit ' l 12 over there was 1x10 to the -4, so you will see that most of 13 the plants using even median spectra would meet those 14 limits. 4 15 MR. SIESS: There is a considerable difference in { 16 the shape of these two curves. 17 MR. CHOKSHI: Yes, because the other one is a 18 mixed bag. 19 MR. SIESS: The other curve looks like it would be ) 20 like, you know, almost the normal type distribution, if you 21 plotted the distribution. 22 MR. CHOKSHI: That is on a different scale, so it i 23 would be hard to -- let me see. 24 [ Pause.] 25 MR. CHOKSHI: This was the earlier one. 4 i 4 I
i I i il 1 MR. SIESS: It's just the difference in shape. 2 MR. CHOKSHI: Well, the scales are different, too. I 3 MR. SIESS: Yes, but they're both log scales. 4 MR. CHOKSHI: Right. But this was just to 5 illustrate what it means, you know, a replication of this 6 with respect to standard design. For the standard design .
-7 using Reg Guide 1.60 spectra, the curve will be even to the 8 left of this.
9 [ Slide.) 10 MR. CHOKSHI: Okay. I think as I mentioned 11 several times -- 12 MR. SIESS: Oh, is this second one also the l 13 average of the 5 and 10 hertz? 14 MR. CHOKSHI: Yes, the same, exactly the same, 15 calculated the same way, 5 and 10 hertz. 16 MR. SIESS: Okay. l 17 (Slide.] 18 MR. CHOKSHI: Okay. My final viewgraph, Don is l 19 going to talk about the trial application of this criterion 20 and the concepts, and as I think several times during this, 21 things came up, and you will see that these are sensitivity 22 studies, and I think it's important to understand the 23 various statistical measures. If you use mean, median, how c 24 does the controlling earthquakes change, and what frequency, 25 proper frequency, do you have to use to get at the problem
l 72 1 of the controlling earthquake. And you will see a few 2 examples of this. 3 MR. SIESS: Now if I had had this procedure in 4 existence and had used it, I could have ended up with all my 5 planto at 10 to the -4 probability of exceedance? If we had 6 used this approach, we would have had a uniform, a real 7 uniform hazard, I guess. 8 MR. CHOKSHI: Right. At least what would happen, 9 that you would try to make sure that you are below that. 10 But since we are using the -- 11 MR. SIESS: Now if I had started out and said I 12 want to design for earthquakes using 10 to the -4 13 probability of exceedance, and I'd standardized on my 14 spectra, those are conservatisms, you sea. I developed 15 spectra and made them conservative this way, squaring them 16 off, and then taking an 84 percent and all of that. Those 17 would have been the conservatisms that were built into the 18 probability of core melt or large release. 19 MR. CHOKSHI: Right. 20 MR. SIESS: But the probability of the earthquake, 21 I suppose -- suppose I had done that using essentially the 22 Livermore hazard curves for all the existing sites and 23 arrived at earthquakes for design, they would have a uniform 24 probability of exceedance, right? 25 MR. CHOKSHI: Design basis, right.
i3 1 MR. SIESS: Now if I went back and did all those l 2 sites for the EPRI hazard curves, obviously they would have 3 a different probability of exceedance. Would they still be ! 4 uniform? 5 MR. CHOKSHI: If you follow the same procedures, 6 yes. The only thing, the number you will be keying will be 7 different. One will be 10 to the -4 -- 8 MR. SIESS: I've already picked my sites. I've 9 already picked my earthquakes, and I've got a 10 to the -4 . 10 probability of that earthquake being exceeded, that ground 11 motion being exceeded at every site, and I arrived at that 12 using Livermore. 13 MR. CHOKSHI: Okay. 14 MR. SIESS: Now I go back and examine every site 15 using EPRI, and I get a different ground motion at every 16 site than I did with Livermore. 17 MR. CHOKSHI: Right. 18 MR. SIESS: And I'll get a different probability 19 of exceeding. Will the probability be the same for all 20 sites or just the magnitude of the earthquake? 21 MR. MURPHY: As I understand your question, the 22 probability would be different at each site, as they are in 23 the calculations. 24 MR. SIESS: It would be different at each site, 25 but would it be the same at all sites? It wouldn't be'10 to l l I
i4 1 the -4 using EPRI. 2 MR. MURPHY: Right. It would not be. If you used 3 just EPRI or -- let's say, if you just used EPRI, the number 4 would not be the same probability at each site. 5 MR. SIESS: You see, I've already picked my ground 6 motions for all the sites using Livermore at a 10 to the -4 7 probability of exceedance. That's my design ground motion. 8 Now I.want to use the EPRI to see what my 9 probability of exceedance is. 10 MR. CHOKSHI: It would have to be different. 11 MR. SIESS: And it will be different. 12 MR..CHOKSHI: Right. 13 MR. SIESS: It would be different at every site, 14 too, I think. 15 - -- -- - - - MR. CHOKSHI: Yes, because the curves are 16 separate, so, you know, you will have to -- if you want the 17 same probability of exceedance -- 18 MR. SIESS: I'm just trying to get a feeling for i 19 how robust this calculation is. 20 MR. CHOKSHI: That's why we are doing all of the 21 sensitivity studies. 22 MR. SIESS: It's a very elaborate process to come 23 up with a number. 24 - 3Gt. CHOKSHI: In a sense, you will get the feeling for that, because if you are doing -- as Don is going to 25
~
f '. 6:) 4 LA G"O
i I I l i5 1 describe, he has done Livermore with median and mean. You 2 can look at that as EPRI and Livermore or as some constant j 3 measurement. So you will get an idea when the curves are 4 separate what kind of effect it has on coming up with the 5 maximum magnitude and distance. 6 MR. SIESS: Well, again, this is a fairly complex 1 7 procedure, and if I asked you, can I-put a new plant at ! 1 8 Millstone, I've still got this procedure to go through. And 9 I'm trying to get some feel as to whether I'm going to get 10 any better answer than I'd get by taking whatever we did for 11 Millstone-3 and saying: Go ahead; that's under .3 g; put 12 the plant there. 13 MR. CHOKSHI: I don't think that will change too 14 auch. But what you would have is a full description of the 15 effect of various sources, various hypotheses, the rate of I 16 recurrence explicitly accounted for. 17 MR. SIESS: Oh, I know the geologists and 18 seismologists wouldn't know a lot. Obviously they know a 19 lot about the Millstone site. If there's anything wrong l 20 with the Millstone site, we've got three plants that ought j i 21 to be shut down. ! 22 MR. CHOKSHI: Right. I 23 MR. SIESS: And obviously we don't think they are. { 24 In fact, Bosnak said we didn't. ! 25 [ Laughter.) , l
-. . _ . .. - -_L
i i Y6 1 MR. SIESS: That was his opening statement. But l 2 I'm just trying to get a feel whether we're getting answers 3 as good as the complexity that this thing indicates. l 4 Now I'm not proposing that we go back to what we 5 did a long time ago, and in the central stable province, we l 6 said the worst thing anybody has seen is an Intensity 6, so 7 let's take an Intensity 7 and design for that, convert it to , 8 a magnitude, and, of course, we built a lot of plants that l 9 way. 10 I don't think they're too bad, frankly. I mean, 11 I'm trying to get a feel -- what is this going to give me 12 for Florida? What's it going to give me for South Texas? 13 MR. CHOKSHI: I think you will see four regions -- l l 14 MR. SIESS: You know, where nobody has looked at 15 anything within 100 miles of it. I mean, I'm obviously not 16 going to put a plant at New Madrid. 17 [ Pause.] 18 MR. SIESS: Okay. Let's take a break. Let's take 19 fifteen minutes. Let's be back at 10:30. 20 (Brief recess.) 21 MR. SIESS: Let's reconvene the meeting. I think 22 you can start. 23 (Slide.] 24 MR. BERNREUTER: Thank you. I'm Don Bernreuter 25 with Lawrence Livermore National Laboratory, and I'm going i l t
i7 I 1 to be talking about some elements of the application of the 2 new approach, although due to time constraints and trying to 3 put all of this information together, I haven't completely 4 followed the theology that we laid out there, but it sort of 1 5 gives a feel for what we're doing, and I'll briefly comment 6 as we go along on some places where I have departed from it. 7 [ Slide.) 8 MR. BERNREUTER: First, just a few background 9 statements, although we've already heard these. But, you 10 know, we've introduced a relatively complex procedure of how 11 to go through this, and the primary reasons that we did 12 this, the first is that both the Livermore and the EPRI . 13 probabilistic studies show that there is considerable , 14 uncertainty in defining the seismic sources, talking about 15 not only the geometry but the recurrence models and also the j 16 maximum earthquakes or expected maximum earthquakes that one i 17 might see in each of those sources, and there is a very wide 18 variation among various practitioners and experts that it's 19 hard to reconcile in a simple deterministic analysis. 20 In addition there has been a major ongoing study 21 primarily funded by EPRI looking at trying to address this , 22 issue of expected maximum earthquake. What is the largest ' 23' earthquake you might see in a source zone? 24 And to do this, they've been looking at 25 tectonically analogous source zones throughout the world, i
is
)
1 and those studies indicate to a certain extent that the 2 postulated maximum earthquake in many seismogenic sources l l 3 that you might say are somewhat tectonical]y analogous to 4 parts of the United States -- and that's on a very loose
)
5 definition of what might be tectonically analogous, because j
. 1 6 it's hard to pin it down -- but those might have been larger i 7 or may be larger than was used in past licensing I 8 applications.
! 9 And so effectively the new proposed rule sort of l ( 10 allows for the explicit inclusion of this uncertainty and 11 these alternative models in a probabilistic sort of way, and 12 these factors then can be explicitly included in the models, 13 whereas in the simple deterministic path, it was very 14 difficult to get these in in other than a very simplistic l 15 sort of way. It sort of led you to a certain amount of 16 difficulties. 17 (Slide.) 18 MR. BERNREUTER: It might be worthwhile once again 19 looking at Andy's slide here. I'm probably going to -- most f 20 of my talk is going to be centered about talking about this l 21 right here. I'm briefly going to touch on some of the other 22 elements of it. This is sort of the deterministic path, and 23 this is sort of the probabilistic path. i 24 (Slide.] 25 MR. BERNREUTER: Now the deterministic path we l r -w-- - - - ___ -___ -____ - .-
i9
- 1. view as will remain relatively similar to the current or the 2 classical approach that's been used. However, we view that 3 the implementacion of it will be simplified, because you now 4 no longer have to agonize so much about uncertainty.
5 The real problem in the past with.trying to apply 3 6 Appendix A, you had introduced such terms as maximum 7 credible event and other terms which'you couldn't really ; 8 define, and you were always sort of torn about how to handle 9 uncertainty. 10 Well, now the uncertainty elements are all put in 11 the probabilistic path. That's how you handle or take care 12 of your uncertainty that you have about what's going on and 13 how you handle alternative source zones and your uncertainty 14 about maximum magnitudes and what have you. All that is put 15 in the probabilistic path, and so the deterministic path is 16 simplified. 17 MR. SIESS: I'm sorry. I don't quite follow that ! 18 line of reasoning. What an engineer does to take care of 19 uncertainty is to provide conservatism. If there's no 1 20 uncertainty, there is no need for conservatism, so the 21 amount of conservatism you put into a design, the amount of 22 margin you put into a design, is there because of 23 uncertainty. 24 And a good engineer adjusts the amount of margin 25 or conservatism to his level of uncertainty.
'80 1 So I don't see how we change anything by moving 2 uncertainty from one place to another.
3 MR. BERNREUTER: Well, I think -- 4 MR. SIESS: And I don't know how to quantify it. 5 I mean, the only place uncertainty gets quantified is 6 through a PRA, but a PRA is made after the design is made. 7 MR. BERNREUTER: That's true, but it's a first 8 step in the quantification of the uncertainty. If you ; 1 9 follow the old approach, you're really total at a loss of i 10 where you are in the entire hazard space. 11 MR. SIESS: What's the object -- I mean, let's ) 12 back off a minute What's the object of this whole process? l 13 The object is to design and build nuclear power 14 plants for which there is a reasonable assurance that there 15 is no undue hazard to the health and safety of the public. 16 Now " reasonable assurance" may be what we're 17 talking about. That's something that's related to 18 uncertainty, I assume. 19 But are we keeping in mind what the object is, or 20 are we looking for numbers to put into a PRA? 21 MR. BERNREUTER: Well, I think we're trying to 22 keep in mind that we're effectively trying to get this ) 23 reasonable assurance. We're trying to get it in a manner , 1 24 that's a little more defensible than in the past, that we 25 have a certain basis for saying that we have achieved this I
'l o 4 81 1 goal. . 2 MR. SIESS: How defensible is it to go before a 3 Hearing Board and say that this site is worse than half the 4 sites out there that you've approved in the past or that 5 we've approved in the past? 6 Is that a defensible position, simply because 7 you've done probabilistic analyses? I 8 MR. BERNREUTER: I think I'll leave that one to l 9 the Staff to address. 20 MR. SIESS: " Defensible", you know, I've got to be 11 able to defend it to myself as a designer, but in this ) 12 business, you usually have to defend it before a Hearing 13 Board and sometimes before a court of law. 14 MR. MURPHY: That's correct, and that's part of 15 the process. I don't want to -- I think this is a bad 16 terminology. But what we're depending to a large extent on 17 in this process is the deterministic path, if you have been 18 satisfied with the results that have come out of that, and 19 those have been. defendable.
~
20 What I believe Don -- you'd have to say Don is 21 referring to here is that after the process has been 22 developed, has occurred, and you've got an answer in hand, 23 you then are in a position of having to defend that. And 34 because it's the most troublesome, you worry about the case 25 of a Charleston earthquake, an earthquake that is large and
$2 1 may have the potential of. wandering into your site or baing 2 placed.at your site.
3 And what we're saying here is that the uncertainty 4 that is associated with that, we are principally going to 5 attack, reduce, by going the probabilistic path. There is 6 always uncertainty on the deterministic side, as we've been 7 _using it, and what we're saying is, let's not try to make 8 our principal defense or attack on that uncertainty through 9 the deterministic side. 10 We have a much better vehicle in the probabilistic 11 analysis to put that rogue earthquake into its proper 12 perspective by carrying out the probabilistic analysis. And 13 that's what we are attempting to do. 14 The design specification that we need to give to 15 the engineers, in effect, comes mainly down the 16 deterministic path, as it has in the past. It is not.really 17 possible to pull that fully through the probabilistic path 18 yet, because the numbers that we are working with there are 19 still relative. We do not have the absolute numbers that
~~
20 would make it possible to, you know, set a safety goal, if 21 you want, for seismic and then go to meet that through a 22 probabilistic analysis. 23 MR. SIESS: I guess you have two choices in 24 defending. You can defend the result, or you can defend the 25 process.
$3 1 Now what you're saying is that you think you have 2 a better justification with one process than you do with the 3 other, that the probabilistic process is more -- gives a 4 more defensible result than the deterministic process.
5 MR. MURPHY: I don't think I want to -- 6 MR. SIESS: And by " defensible", I mean somebody 7 argues: Well, why can't it be twice that? 8 MR. MURPHY: That's correct, because in the 9 probabilistic process, you have taken and tried to put twice 10 that earthquake into the process, to put that into its 11 proper perspective. How often is that going to occur? What 12 are the community's feelings about that earthquake 13 occurring? And you take that into consideration. 14 And if it turns out that that event should be 15 considered at the 10 to the -7 or -8 probability, it is 16 factored into it. And it basically currently would show 17 that it does not have a significant or a meaningful 18 contribution to the hazard of that site. 19 MR. SIESS: Do you really think there's a 20 community feeling in terms of probabilities? 21 MR. MURPHY: I'm not sure there's a community 22 feeling about probabilities or a probability number itself, 23 but I think there is a community feeling about many of the 24 parameters that go together to make these things. 25 MR. SIESS: We had a meeting cut in San Luis
$4 1 Obispo, which sits right on top of the Los osos fault, which 2 has a potentiality of, I think, of 6.8, and those people 3 were worried about the Diablo Canyon power plant 20 miles 4 away on another fault, but they weren't at all worried about 5 their fault.
6 Now you think there's a community feeling-in terms 7 of probabilities. I don't think so. , 8 I mean, somebody says: Why can't the earthquake 9 be twice what you picked? Now this is a Hearing Board. 10 You've got to answer: Yes, it can be twice that, but the 11 probability is low. 12 MR. MURPHY: That's right. 13 MR. SIESS: Now if you've got probabilistic-minded 14 people on the Hearing Board, that's one thing, but they're 15 not out in the community. And I will never will forget that 16 one of the German RSK members said: Don't ever try to get 17 up in a court and talk risk. He says: Everybody 18 understands conseguances; nobody understands probability. 19 And if the conseguances are big enough, you can never 20 convince them that the probabilities are low enough. 21 So I don't think that community feeling gives me 22 auch comfort. Maybe as an engineer I should get some 23 comfort. ; 24 MR. CHOKSHI: I think what we are trying to do is, 25 once you establish for any probability level what is l l
f B5 1 contributing sources, it strictly becomes at that level l i 2 which source contributes most -- i 3 MR. SIESS: And that's your biggest uncertainty.
- 4 The biggest uncertainty in the whole process is the hazard 5 curve, and the biggest uncertainty in the hazard curve is j 6 which sources out there are capable of what.
i 7 Maybe the sources are not the biggest i~
- 8 uncertainties, but what they're capable of. That's where 9 you'c e2 pert
- disagree. They will all agree that there's a l
10 source there. 4 11 FN. MURPHY: I don't think the level of . 12 disagreement is as catastrophic to the process, showstopping s . 13 to the process, as you are inferring. I think there is 14 disagreement. There is no question about that. 15 But I think that a procesr that has been used, 16 developed by the EPRI and the Livermore, begins to put some 17 of that disagreement into some sort of a perspective, so 18 that you can better weigh this. 19 Does that mean that there isn't a fault out there 20 that's going to turn around and bite you in the pipeline? 21 No. But we're living, you know -- we've got to be 22 living with the risks that are there, and we try co get our 23 best handle on what those are. And by balancing what we're 24 calling deterministic versus probabilistic, we're trying to 25 put those risks, hazar'ds, into some sort of a reasonable
$6 1 perspective.
2 MR. SIESS: Except I think some people are getting i 3 more comfort out of it than they should. l l 4 The ACRS many years ago asked questions which led 5 to the seismic margins study. The question was simply: If l l 6 the earthquake is bigger than we design for, what happens? l 7 Is the plant going to fail if it's'.21 instead of .20, 1 ~ i l 8 because we weren't -- nobody had any confidence that we had ; l 9 picked the maximum earthquake, the maximum probable or 10 maximum possible? You know, we didn't have all of this. 11 The question was: What happens if we are wrong? l 12 And we get most of our comfort on seismic siting parameters l 13 from the fact that seismic margin studies have shown that we 14 do have margins. We're not on the edge of a cliff. We can 15 have a lot bigger earthquake. We can have an earthquake as 16 big as they've had in California in Illinois, and we 17 probably still don't fail certain things. That's an 18 exaggeration, I believe, but there are areas I could cite. 19 The PRAs have shown, yes, we have vulnerabilities 20 to earthquakes, particularly offsite power sources, but
- 21 everybody knew that from the begirning, and if you want to 22 fix~it, you can.
23 So the uncertainties have always been there, and 24 they're not going away, and we're not quantifying a whole 25 lot better. p w . , , e-- -- w -- g--- ,,-r-- - - - . < - ----,r- - ,, -- - ~ - -
-.--=_.---- _ --_-.- -_.._..----...-. . --..__-
1 1 I 87 i 1 Now I'm not objecting to what you're trying to do, f 2 but I think you're taking more credit for it than it j 3 deserves. 4 4 MR. MURPHY: That may be so. But I think we're 5 also looking at with the same eyes that you're using, that j 6 what we're saying is, we appear to have done a good job, or 7 a relatively good job, in siting these facilities'and ! 8 providing the specifications to be used for design. We're l' ] 9 looking at the ways. We made those decisions. And to a i 4 10 large extent, we're saying we're going to be using those 11 same criteria to make decisions in the future. 12 MR. SIESS: I might put the question a different 13 way, because you've answered it already. 14 Why don't you simply go to a purely probabilistic 15 basis? Why don't we simply say 10 to the -4 probability of 16 exceedance or 10 to the -5, and just do that and forget 17 about this so-called deterministic thing? 18 This is only for future plants. All we really 19 have to establish is that it's less than .3 g, unless 20 somebody is going to come in and want to design the onsite 21 site-specific facilities for .1 and the plant for .3, and I 22 don't know whether anybody is going to let them do that. 23 But why not go strictly probabilistic? That's , 24 what we've been using now for the last 15 or 20 years every 25 time we look at a plant.
58 1 MR. MURPHY: Every time we wanted to check what we 2 did at the plant, we looked at it probabilistically. 3 MR. SIESS: That's right. 4 MR. MURPHY: We didn't make our basic decisions -- 5 MR. SIESS: So why not go all probabilistic? 6 MR. MURPHY: My personal opinion is that at this 7 stage the science or the art of probabilistic analysis is 8 not mature enough to allow us to do that? 9 MR. SIESS: It's mature enough for us to decide 10 whether a plant can continue to operate. 11 MR. MURPHY: We're using it there to check 4 12 decision that we made by a different means, and we're simply 13 checking ourselves. We're not making -- nominally, we're 14 not making the initial decision based upon the probabilistic l 15 analysis. 16 MR. SIESS: Is that why you wanted Diablo Canyon i 17 to come up with a deterministic set of values as well as a ) 18 PRA, or only because the regulations deal in the 19 deterministic numbers? 20 MR. MURPHY: The regulations deal in deterministic 21 numbers. 22 MR. CHOKSHI: I think there is also an issue about 23 ground motion to be used in design. A purely probabilistic 24 approach will give you something like uniform hazard 25 spectra. Each has little to do with ground motion.
- . . - . - . . . - - - .... ._- . - ~. - . . _. _-_ -.-
59 1 MR. SIESS: Well, it has something to do with 2 ground motion. 3 MR. CHOKSHI: Something, really, but it doesn't 4 look like -- 5 MR. SIESS: We still always start with that. 6 MR. CHOKSHI: For a different -- for engineering 7 design, we would probably tweak that motion anyway, because 8 I don't think we would be fully comfortable with that. 9 MR. SIESS: Well, let's face it, for engineering 10 design on standard plants, that's,all they're doing. 11 MR. CHOKSHI: Right. 12 MR. SIESS: They've got a spectrum and that's it. ; 13 This is the ground motion we're designing the plant for. , 14 Now if that's all we have to site in the future, 15 our only problem is to determine whether that spectrum is 16 exceeded by anything that happens at the site. And if.the 17 site is well below it, maybe we're home free. i 18 I don't know how much the Staff has thought about 19 whether the site-specific things, which are fairly important 20 parts of the plant -- I've got a plant design for .3 g. I 21 come up with a .15 g. Are you going to let them design the 22 circulating water system and all of the service water 23 systems and those things for .15 when the rest of the plant l 24 is designed for .37 I don't know. 25 I mean, there are some category 1 systems that are
so 1 site-specific. 2 MR.'BOSNAK: If you had a good hazard analysis of 3 the component systems and structures that you have, 4 including the non-seismic, and you had some idea of the 5 uncertainties, a better-quantification of the uncertainties 6 that you had with respect to seismic hazard, you might be 7- 'able to -- I'm net saying this is incorporated into the 8 cystem that we have right now -- but you might be able to' 9 then to focus on the margin that you have in those pieces of 10 the plant hardware that are the greatest contributors to 11 risk ar.d build a little bit more margin into those 12 components. 13 MR. SIESS: Obviously if I did a PRA -- I'm the 14 l
'cwner now; I'm the guy that's going to put the money into i 1
15 it, and I've got a plant that will take .3 g, but I've got i 16 all my site facilities that only need to be designed for .15 l 17 -- a PRA could tell me whether I'd gained anything by upping 18 them to .3 or .2 or .25. 19 I don't know if anybody is going to do that or 20 not. But the Staff would certainly let them design those 21 things for .15, if that's what you come up with Appendix A - 22 - Appendix B, Appendix S, whatever. 23 MR. MURPHY: Now to back up just a little bit, an 24 additional part of the answer to the question, why not use 25 probabilistic solely and forget the deterministic part of
'91 I
1 the review or the deterministic part of the investigations, 2 is that the probabilistic assumes that the material and the ! ! 3 information that are produced by the deterministic analysis 4 are available and accurate, that the probabilistic makes the 5 assumption that what we would have learned from the l 6 deterministic analysis is there and available. It's part of l, 7 the basic assumption in the probabilistic process. It 8 depends on that information being there, being confirmed, to 9 nove forward, just, if you want to say, a basic assumption lo or starting point on that analysis. 11 MR. SIESS: What information, for example?
)
12 MR. NURPHY: The information, you can get down to ; l 13 the specifics as to whether or not there's a fault onsite, 14 and what do you know about that fault that's onsite. The 15 probabilistic analysis -- 16 MR. SIESS: Well, I'm sorry, but Appendix B.says 17 you can't put it next to a fault, doesn't it? 18 I'm not even into the Reg Guide now. I'm sticking i e 19 with Apt indix B. Doesn't it say I can't site it if there's 20 a fault, an active fault? 21 MR. MURPHY: Not right underneath. But I mean, 22 let's say we're talking about -- 23 MR. SIESS: It used to. 24 MR. MURPHY: Let's say we're talking about a 25 situation where you're offsite, five miles, ten miles > l i I
-. - .. _. ~.-. -. - .. - - - .- _ _-._ -.. - . - . - -.. - -.. - . ~ -- ..-._ - . -
1
$2 1 offsite, and you've got a large fault there.
2 In order for the probabilistic analysis to be 3 meaningful, that fault has to have been discovered, its 4 characteristics determined, recurrence time, length, 5 magnitude, that sort of stuff. 6 MR. SIESS: Now you're talking about investigation 7 and exploration. 8 MR. MURPHY: Yes, sir. 9 MR. SIESS: Well, I'm not talking about that. I'm 10 talking about determining the safe shutdown earthquake 11 ground notion. I don't remember the chart, but didn't it 12 start with investigation in both cases? l 13 w _R- FURPEi* : That's right. 14 MR. SIESS: You have to-know what's there. 15 MR. MURPHY: Right. l 16 (Slide.) l 17 MR.,MICHELSON: Let me-ask in the case of an ABWR,
- 18 which I'm a little more acquainted with, the site that I l
19 might pick for an ABWR will be looked at without even ( 20 considering what goes on the site, and those are tha rules ) l 21 we're talking about today. 22 I guess it's entirely possible that although the 23 ABWR might be arbitrarily designed, the part that the vendor 24 designs, which is most of the nuclear island and that sort f 25 of thing, will all be designed, say, for .3 g ground motions i i l
'93 1 or accelerations.
2 The site-specific items, which is the ultimate
^
3 heat sink, which they don't design, which is designed by the 4 user of the site, might only have to be designed for .15, if 5 it happens to be a low seismic site. 6 Let me caution you that from the PRA viewpoint and , 7 so forth, at the time of certification of this design, 8 nothing is known about that ultimate heat sink. At the time 9 the site is approved, nothing is known about the heat sink. 10 And I thought once you have an approved site and approved, 11 certified design, that yo'u'can't make any changes, that 12 indeed it is a .15 g that's going to be used for the 13 ultimate sink. 14 MR. SIESS: No, I don't think it's required.. 15 , MR. MICHELSON: It is. 16 MR. SIESS: It's not required. ; 17 MR. CHOKSHI: Somebody might propose to do it that 18 way. 19 MR. MICHELSON: Well, if somebody says my site is ! 20 a very low seismic site, I only need -- my ultimate heat ; 21 sink, I don't need to design that for .3. 22 MR. MURPHY: That's right. l 23 MR. MICHELSON: And included in that is the piping l 24 from the ultimate heat sink down to the basement of the 25 control building. That will also be --
_ _ _ . _ _ _ _ . _ _ _ _ _ . _ . ~ . _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _. _ -. _ _... _ ._ l
$4 1 MR. SIESS: The Staff says that would be !
- 2. permitted.
i 3 MR. MICHELSON: And that will be permitted for l 4 that piping in the control building to be designed for .15, l l 5 even though all the rest of it on the other side of the heat y 6 exchangers is designed for .3. : 7 MR. SIESS: And you can't change that, if I : 8 . understand the certification process. l 9 MR. MICHELSON: You can't change the requirement. 10 MR. SIESS: They can't change the requirement. 11 MR. MURPHY: Right. 12 MR. MICHELSON: That if they wish, they can go in I 13 with .15. 14 MR. SIESS: That would probably be a pretty good 15 plant, because if you look at the seismic contribution to 16 core aalt, a bit chunk of it is loss of offsite power, and 17 the part that would be due to those systems is small. 18 MR. MICHELSON: Well, it depends on what the ; 19 utility later designs and whether he uses lots of metal 20 bellows that take care of his expansion problems in the i 21 basement. I think there are a lot of things that you don't 22 know anything about.
- 23 MR. CHOKSHI
- But currently, you know, for i
i 24 example, for the ABWR, I believe that a plant walkdown when l 25 actually designed, and all site-specific features have to be , t
I
$5 j 1 looked at and make sure that there is no adverse impact on -
2 - 3 MR. MICHELSON: But looked at at .15 g. 4 MR. CHOKSHI: As long as they show there is no l I 5 adverse impact both in terms of PRA results, as well as the I
~
6 design certification. 7 MR. MICHELSON: And, of course, in PRA space, if I i 8 design for .15, by these rules you'll have some certain 9 probability of exceedance -- ' 10 MR. CHOKSHI: Right. 11 MR. MICHELSON: -- of that. But the consequence 12 will still be acceptable, I gather. 13- MR. CHOKSHI: Right. It depends on what the other 14 things arn. 15 MR. SIESS: You've got some margins, you know. ; 16 MR. MICHELSON: Yes. But none of this you'll know 17 anything about until somebody comes up and says this site 18 and this design I'm going to wed together, and I further am 19 going to claim that you've made all your final design 20 decisions, licensing decisions, and unless I deviate from i 21 the licensing requirement in those decisions, you can do it. 22 MR. SIESS: Keep in mind the alternate. They 23 could go in and put a custom plant there, the whole plant 24 design for .15. 25 MR. MURPHY: Right. 4
l l '9 6 1 MR. MICHELSON: They could, yes. 2 MR. SIESS: So what we're getting out of l 3 standardization is a .3 plant and a .15 of the other. 1 f 4 MR. MICHELSON: For the water system. j 5 MR. SIESS: And it shouldn't be worse. I say that I l \ . l 6 cautiously. j 7 MR. CHOKSHI: But the procedure described here.is i 8 general, you know. That is the presumption. But I think if 9 somebody comes in with a standard plant, it is probably 10 relatively easy to go through these steps and show that, you l 11 know, except for investigation, the rest of the process l 12 might be much more smoother. 13 MR. SIESS: But you shouldn't be penalized for 14 having a standard plant. If I want to put a custom plant on 15 that site, and you'll let me do it at .15 g, then there's no 16 reason for anybody to tell me to do it differently if.it's a 17 standard plant designed for .3 g. 18 MR. MICHELSON: Yes, there is a reason. The l 19 reason is that it was certified at .3 g, and it's to be 20 built that way. 21 MR. SIESS: Only the plant was. 22 MR. MICHELSON: Well, yes, but that was 95 percent 23 of it. 24 MR. SIESS: But that plant has got to be better , 25 than -- I can put a .15 g plant there. Now I shouldn't be l
~ . _ _ . . . . - . . _ _ _ _ . . . _ _ _ _ _ . . _ _ _ - . _ . _ _ _ _ _ _ _ _ _ _ .. _ _ _ .. ..._ _ _ _ . _ i I
, 1
'97 1 penalized if I want to put a .3 g plant for the rest of the 2 site.
l 3 MR. MICHELSON: But I thought the. inference was, ; l 4 well, now I can come back and maybe my relays that I want to i 5 'use were found to chatter a little at .3, but since it's l . i 6 only a .15 site, why can't I -- 7 MR. SIESS: Oh, no. That's something else.
- l l
8 MR. MICHELSON: You know,.then there are all those L 9 stories. 10 MR. CHOKSHI: Then the problem becomes quite 11 complex. 12 MR. SIESS: That's not a siting problem. That's a
- 13. plant design problem.
14 MR. MICHELSON: But the argument you just finished l l 15 presenting, thought, Chat, I can argue this site is a .15 g 16 site. I don't need those .3 g relays, even though that was 17 in the certified design. I don't need them. 18 MR. SIESS: That's how they enforce it. I don't 19 know. ,
~
20 MR. MICHELSON: Because it was, you know, a 21 blanket certification to cover all sites. But it gets real 22 sticky. 23 MR. SIESS: They would have an argument, but then 24 it wouldn't be a standard plant. 25 MR. CHOKSHI: Yes. What it does to the process of 1 1 J
.- . . . . - _ _ - - - . . . . - - - ~ . - - - . _ . - . . - _ - - - - _ .
( i' e 98 j 1 .stendardized design, I think -- I don't know. I 2 MR. SIESS: If they want to come in with an ABWR I. 3 certified plant, but change the relays and license it as a i 1 4 custom plant, they certainly could.
- 5 MR. MICHELSON
- But see, what you're going to run i
! 6 into is, you're going to have a cold climate, and you're ? j 7 going to need steam heating for the reactor building and so . 8 forth. They're going to come in with a .15 g steam heating j' 9 plant. 10 MR. SIESS: All they've donc is invalidate the 11 PRA, I guess. i j 12 MR. CHOKSHI; I think that's the key to it. i ) 13 MR. SIESS: I guess I really haven't heard a real e; 4 14 good arg** ment not to go all probabilistic for future plants. 3
- 15 MR. MICHELSON
- If you know how to do it. I mean, j 16 you can do it in a pure sense for the site, I guess. I'm i
] 17 not that expert -- 18 MR. SIESS: Well, this procedure they know how to 19 do. We're going to hear the details of it in a minute. L - l 20 MR. MICHELSON: They're going to explain to me how I f 21 I'm going to take and apply it now to a design of a 5 - k 22 particular system in an ABWR7 i ! 23 MR. SIESS: That's something else. 1 l 24 MR. MICHELSON: Well, that's what this about. l 25 MR.-SIESS: No. i-
, c _n..
$9 1 MR. MICHELSON: In the final context, nobody cares 2 a dann about this until you put a plant on a site..
3 MR. SIESS: That's right. l 4 MR. MICHELSON: Worrying about all of this for a l 5 site alone is fun, but it doesn't become important until you , 6 put a nuclear plant on the site. , 7 MR. SIESS: No, don't say it's fun, because the 8 Commission has set up a procedure for pre-approved sites. ; 9 MR. MURPHY: Right. You will do this whole 10 procerr. You get a stamp of approval, and you may never put 11 a plant on that site. It's just sitting there -- 12 MR. SIESS: There's nothing safe about that site. 13 All we know is that for that site, this is the seismic 14 design basis that:the Commission will accept for that site, t la assuming that new information doesn't come out. 16 You know, if you've had an earthquake ten mi.las 17 away that you didn't know about between the time you 18 approved the site arJ the time the application was made to r 19 put a plant on its they're not going to get it. 20 MR. MURPHY: No, sir. 21 MR. SIESS: I don't know what the rules say, but I 22 know they're not. This is simply how to determine the 23 seismic basis at a site. 24 MR. MURPHY: That's correct. , 4 25 MR. SIESS: What you do with it is very important.
1b0 1 But whether you get it by this method or.that method, as I 2 said, I haven't heard an argument that would convince me 3 that we need two methods. 4 MR. MICHELSON: Now when you go to do the -- when 5 you go to place the plant on the site, do you think you can 6 do that probabilistically also? l 7 Once you go and put the plant on the site and you , 8 want to design certain features like site-specific features, 9 like the ultimate heat sink, I don't know that we're well 10 enough skilled yet to do that probabilistically. 11 MR. SIESS: It's already been done. You arrived i 12 at .15 g as an acceptable safe shutdown earthquake design 13 motion basis for this site, and that's true, no matter what
- 14 plant you put there.
i 15 MR. MICHELSON: That's right. But now when I go ! 16 tt put a bellows on a pipe, and the part that I'm designing, i 17 I have to know the fragility of it, and I'm not sure that I 18 know that I can apply probhbilistic techniques at that ] 19 point. 20 MR. SIESS: There's nothing probabilistic about 4 21 it. 22 MR. CHOKSHI: Since once it becomes .15 g, that is ) 23 deterministic ground notion. 1 24 MR. SIESS: Yes. Then you just forget all this 25 probability and you go back to doing what you've been doing. 1
l'b1 1 But then when you get all done, then you like to go back and 2 say now what's the -- 3 MR. MICHELSON: There's no place in the severe 4 accident consideration here. 5 MR. SIESS: There is no place in the ASME Code for 6 anything but putting in a number. A load is a load. 7 Now if you want to do a PRA after you get through, 8 that's something else. If you want to do a PRA and you've 9 done it by the lefthand side, you've got a basis. If you've 10 done it by the righthand side, you can't do a PRA, but you 11 can do a seismic margin study, right? 12 MR. CHOKSHI: Dr. Siess, to the question about 13 probabilistic and deterministic, this came up a number of , 14 times in our discussion with the expert panel. The current 15 status of the probabiljotic methods, you have two methods to 16 use for qui':e different reasons, EPRI, Livermore, so there ' 17 is still an issue to be settled. The matter is not -- 18 MR. SIESS: Oh, not. If you want pure 19 probabilistic, you'd have to decide one or the other. 20 MR. CHOKSHI: Right. So then we are not -- l 21 MR. SIESS: And you'd get around that. 1 . 22 MR. CHOKSHI: There is another aspect that ..c 23 probabilistic is considering uncertainties, ordinate models, 24 a larger model of information which might smear some things. 25 And here you have an experience based on the deterministic l l
162 1 approach which looked at the thing, the past experience, and 2 it's tried to verify the judgment, and what probabilistic 3 gives without the deterministic project, I think everyone on 4 the panel sort of wasn't comfortable, and our staff is not 5 comfortable that we are at the stage where we can handle all 6 the problems. 7 I think that is the biggest -- 8 MR. SIESS: Well, you see, what you're doing using 9 the median, below the median probability, gets you out of 10 the problem of what to do with EPRI and Livermore, because 11 you take the median of each one. a 12 MR. CHOKSHI: Each one, right. I l 13 MR. SIESS: And they might give you slightly ; 14 different answers, but they're not two orders of magnitude 15 difference. So that's one reason, I guess, for going 16 strictly probabilistic, unless you want to bite the bullet 17 and say -- 18 MR. CHOKSHI: Just use one. 19 MR. SIESS: -- we believe it's -- 20 MR. MURPHY: We're not going to bite that bullet. l 21 MR. SIESS: That bothers me, too, that we're off 22 by two orders of magnitude on some of these things, and you 23 don't feel strongly enough about what you've got to say the 24 other people are crazy. ! 25 MR. MURPHY: You can't say that in public. l
l 103 l 1 MR. SIESS: Dr. Lewis wanted you to split the t 2 difference. And your argument about not going all 3 deterministic is not that you get a. batter answer the way , l 4 you're doing it now, but you get a more defensible process. J 5 MR. CHOKSHI: And I think the one thing the ,
. l 6 probabilistic explicitly considers is the recurrence, which l J
7 notion is never clear in the deterministic. That brings it- l 8 out very clearly, how the rate of recurrence affects the l ! l 9 hazard, and that part is no different -- 1 10 MR. SIESS: Well, it allows you to make the i . 11 probabilistic argument that, yes, the earthquake could be ] 12 ten times as large, but the probability is lower, and some l 13 of us find that convincing and -- 14 MR. CHOKSHI: That's true. 15 MR. MURPHY: Right. l l 16 MR. SIESS: You know, a lot of us thought that the 1 17 SSE was probably down 10 to the -6 probability in the 18 earlier plants, and when we got into the probabilistic stuff 19 and found it was only 10 to the -4, 10 to the -5, and 20 believed it, that still, you know -- geologists are sort of 21 -- probabilities are sort of new to geologists, too. 22 But when we found it was 10 to the -4, 10 to the - l 23 5, we got worried. That's when we wanted to now look not i 24 just at that but at the margins, what is the probability it l l 25 will do something to the structure. l , i l
I I i 164
'l -Okay, let's get back --
2 MR. BERNREUTER: Back to the presentation? r 3 MR. SIESS: Back to that and to some numbers. 4 It's the philosophy that these guys are going to have defend l 5 sometime, not the numbers. 1 . 6 EMR . BERNREUTER: In any event, the deterministic 7 path, as I sugges":d, I'm not going into today, because we
-8 envision that pretty much, at least the net result would be 9 what we have today, basically. That's the' idea of it.
10 In my view, it's somewhat simplified, in that you 11 . don't have to worry so much about the uncertainties, because 12 they're moved over to the probabilistic path. 13 My presentation is trying to focus more in the 14 probabilistic path. l 15 Now, if you note, the probabilistic path -- which 16 is-my next viewgraph, but it might be a little easier.to 17 understand here -- really, you might say, has two elements
- 18 to it.
i l 19 One element is the step right here that Nilesh l 20 described previously, where you simply look at and compara 21 your SSE that you come out with, however you come out with 22 it, to the bulk of the plants and show that, indeed -- and 23 the suggestion was made that you took the average of the 5 24 and the 10 hertz and that it's below the 50th percentile 25 there. It was down in the bottom half. I
l 165 1 That's, you might say, a pure probabilistic path, i 2 if we take the results and show whatever you have come with,
]
3; design, however you arrived at it, indeed it meets this 4 criteria. 5 So, that's relatively simple criteria, and I j 6 really don't plan to go into that step then. 7 MR. SIESS: That's where we've been. 8 !(R . BERNREUTER: That's where we've been. 9 Now, we've introduced, at the suggestion of that 10 panel that was put together -- there was sort of a mixture 11 of probabalists and determinists in there, and so, they 12 suggested strongly thst a very interesting thing to do and 13 what you ought to look at is also to try to use your 14 probabilistic analysis to tell you which earthquakes you 15 should be looking at; that is, what are the controlling i i 16 earthquakes, what earthquakes are controlling the l 17 probabilistic design. 18 We'll go back and look at those, and we can do two 19 things from that. l
~
20 One, of course, you can sort of look at that and 21 see whether or not that makes sense to you physically, and
~
l 22 number two, then you can use that information to get a 23 better assessment of the ground motion that you would see at l l 24 the site. l 25 MR. SIESS: That means going back up to the first f 1
.- - . - - - ..-. ~ . . - . -. - . - . - . . . . . _ _ . - . . - - . - _ . . . - . - . . - -
! l'0 6 l 1 box 7 l 2 MR. BERNREUTER: That's this box right here.
'3 MR. SIESS: Once I've decided that earthquake A at ]
i 4 ragion X is really dominant, does that mean I might want to j i 5 go back and do more geologic seismological investigation? 6 MR. BERNREUTER: You might also want to do more i l 7 geologic work, or.you might just simply, if you don't think . ! 8 you know enough about it, then you want to use that 9 information to go into a better way of deterministically 10 determining your ground motion at the site, using your best 11 ground motion models or other ways of modeling that 12 particular earthquake that would give you the information 13 that you need at the site, which sort of gets all mashed 14 together in a probabilistic analysis. 15 MR. SIESS: If it's appropriate, it would help me 16 to relate some of this to what was done in the Diablo Canyon 17 long-term seismic program, which I am quite familiar with, 18 and I think that some of the stuff that's in that box is ; l 19 similar to that. I believe so. I think that they - 20 MR. BERNREUTER: 21 came up with scenario earthquakes and such. . 22 MR. CHOKSHI: I think, in Diablo, they used 23 probabilistic calculations to show that the Hosgri dominated ) 24 the risk of the seismic hazard at the site. 25 MR. SIESS: They did both things, of course. l 1_-__________-____ _ . - -
l 1 l'07 1 MR. BERNREUTER: And came up with basically the -- ! 2 you have uncertainty in how large that earthquake can be and l l 3 where it can be'and the type of mechanism and such. 1 i ! l 4 I think it allows you then to sort through all of i l 5 those and see which of those scenarios really is most Io j 5 important, or it may be that their equally important. Then l 7 you have to worry about both of them, and then some 8 judgement has to be made by the staff. ) ? l 9 MR. SIESS: There are some suggestions in the -- I l ! 1 10 don't know whether they're in the draft reg guide 1015 or in
- 11 the summary of it -- which reference the Diablo Canyon l
12 techniques, which I could understand. 13 The thing is it wasn't clear to me whether it's iI 14 really intended that people go out and spend that kind of 15 money on a new site or not. 16 MR. BERNREUTER: I think Diablo Canyon nnedmd to 17 do it because there were no -- I mean basically no regional 18 studies. Unless you move into the very active tectonic 19 region like Diablo Canyon, clearly you cannot depend upon 20 regional studies, because you have to do site-specific. 21 MR. SIESS: Okay. So, if you were eastern U.S., 22 then you don't think you need to do that much. 23 MR. BERNREUTER: In the eastern U.S., presumably 24 once you've gone down the deterministic path and have done, i l 25 shall we say, these necessary geological and seismological I I l
4 I i i ! 168 ' i , 1 investigations around the site, I think the supposition is ! 2 that that is not going to show you any active tectonic 3 faulting, and so, you fall back, then, on-the standard l 4 regional studies that you've done, and now you can make use 5 of these regional studies, like the EPRI and the Livermore
- 6 study.
t , ! 7' MR. SIESS: Would you have these two paths for a I 8 west coast site? f 9 MR. NURPHY: You would have them, but they would i , ! 10 not be as simple as this one, in that you could not make a l l 11 comparison with existing studies. 4 l 12 NR. SIESS: That's right. j 13 NR. NURPHY: But you would have a probabilistic j 14 hazard analysis.
- I t
i 15 MR. SIESS: Yes, but you wouldn't have any use for i 16 that middle box on the left then. 1 i 17 MR.'NURPHY: No. 18 MR. SIESS: Okay. l 19 MR. NURPHY: There is nothing to compara it to. 20 It just wouldn't be there. 21 NR. BERNREUTER: Well, you compare this and this, . 22 but you wouldn't have this step here. 23 NR. SIESS: That's right. 24 MR. NURPHY: Right, very definitely not. 25 MR. SIESS: Yes.
4 ! i j 109 l 1 MR. CHOKSHI: You probably would have to lock at ), J j 2 full hazard spectrum to see.what's controlling. 3 MR.-SIESS: And really, 90 percent of your l 4 thinking in this whole thing is eastern U.S. sites. 1
- 5 MR. BERNREUTER: My whole thrust is strictly i
} 6 eastern United States. , i l 7 MR. SIESS: Yours would be, sure. j l _ 8 MR. BERNREUTER: Okay. I 9 MR. SIESS: Okay. l 10 MR. BERNREUTER: And I think we're driven more 4 l j 11 this dual path in the eastern United' States than we are in l' 12 the western United States. l 13 MR. SIESS: Well, that doesn't bother me, because 14 I don't think there are going to be anymore sites in the f
- 15 west anyway.
16 MR. BERNREUTER: Yes. They, generally speaking -- 17 you feel much more certain about your deterministic path i 18 there. ! 19 MR. CHOKSHI: Most of the standard designs are L - 20 designed for eastern United States. 21 MR. SIESS: And yet, I feel that the uncertainties 22 are actually less in California. They probably had 23 something close to the biggest earthquake they're going to 24 have. In the east, you know, we keep worrying about the one 25 that's three times as big. l
i i 4 liO 1 [ Slide.) 2 MR. BERNREUTER:' As I say, the probabilistic path 3 effectively really sort of bifurcates it. One is the check 4 that Nilesh discussed earlier, and then there is the path 5 that we put in talking about the -- trying to define the 6 probabilistic controlling earthquake. 7 (Slide.] . 8 MR. BERNREUTER: We thought -- it was the 9 recommendation of the group that we put together, the panel, 10 that it was worthwhile looking at what I'm calling the PCE, 11 the probablistic controlling earthquake path, was introduced 12 for several reasons. 13 One thing, it identifies significant earthquake 14 sources for comparison with the deterministic earthquake. 15 It makes use of all of the probablistic data and 16 uncertainties to gain very useful insights. That is,,of all 17 of the uncertainties, how do they affect the factfinding of 18 the analysis, and where is that leading you. 19 Lastly, in a sense, the deterministic people at 20 least always argue, "Well, the probablistic approach is 21 really nice, but it -- because you are treating 22 uncertainties in there, but when doing those uncertainties, 23' you do have to make a lot of, once again, a different set of 24 assumptions, which could be very wrong, so that your 25 relative hazard curve may be very well, but where you'are in
111 1 space may be very wrong. 2 So if you came in at -- say you were going to come l 3 in at the ten to the minus fourth level per se, and you did 4 that in an absolute sense, you could be very wrong. 5 So the idea of this path is that you sort of get 6 that out of the process in a sense. J 7 [ Slide.] 8 MR. BERNREUTER: Okay. Now, let's look at the 9 probablistic controlling earthquake, and I'm going through 10 just exactly -- it's a four-step process is the idea of what 11 it is. It's really -- the idea was that you de-aggregated 12 the results to determine which earthquakes are dominating 13 the -- or controlling the hazard. 14 That's all well and good, but what do you mean by 15 that? So the purpose of this is to try to step you through , 16 conceptually just exactly what it is. So it's a four.or t 17 five-step definition. 18 The first step is that if you're going to do that, 19 you do have to have some sort of relative probability level 20 to enter your curves. That's simply what Nilesh was 21 discussing previously, that simply as a matter of 22 convenience, you simply now lay out this cumulative ( 23 distribution of the probability of exceeding the SSE for l 24 whatever estimate you might happen to want to use. t 25 There are two elements of it. One is the hazard - I rv
._ < . _ - - . _ . _ ~_.._ _ .-___ _ - . . . _ . . _ . . _ _ _ _ _ . ~. _ . . . _ . . ___.
li2 1 - the ground motion parameter, which could be PGA, the 2 average of the five and ten hertz spectral velocity, the 3 average of five, ten, fifteen hertz, whatever you want. 4 That would be the ground motion parameter. 5 Then you have several possible estimated that you
- 6 could use. That is, the mean hazard, your probability of i
7 exceeding the mean hazard, your probability of exceeding the , 8 median hazard, or the probability of exceeding some 9 percentile, like the 85th percentile or the 84.1 hazard. 10 Ideally, I should have done this, which Nilesh 11 described, the average of the five and the ten hertz. 12 Unfortunately, because we only had a relatively short time 13 to throw all this together and do a lot of calculations, I l 14 needed to look atesingle individual estimators at this l l 15 stage. So I've done it for the peak acceleration and five I
)
16 hertz. 17 At the next step, I'm going to do for the average 18 of the five and the ten hertz. But I thought it would be 19 interesting to look at what happens whether you use the five 20 hertz or the PGA to unfold your probablistic determined 21 earthquake, because they are telling you different things. 22 (slide.] 23 MR. BERNREUTER: At any rate, you go into this 24 distribution, and the next viewgraph just specifically -- 25 using the Livermore results. If you go in and try to find
li3 i 1 the probability of exceeding the ASSE for all the sites, 2 it's 1.19 times E to the minus four, or if you go to the 3 five hertz, the median is 5.43 E to the minus five, or at 4 the mean, is 1.32 E to the minus three. 5 MR. SIESS: You have lost me with your notation. 6 Would you like to restate the significance of those figures 7 using words? 8 MR. BERNREUTER: Okay. Well, the idea is that -- 9 remember, I just simply went in here somewhat arbitrarily to 10 40, you know, so that -- so that whatever site we came out 11 with would be better than half the sites or -- 12 MR. SIESS: This is one site? 13 MR. BERNREUTER: This would be one site. Well, 14 actually, this is universal for all sites. One example -- 15 MR. SIESS: I'm sorry, I do not understand what 16 you are telling me. 17 MR. BERNREUTER: Okay. 18 MR. SIESS: Now, if you don't understand what I 19 don't understand, I'll try to explain it to you, but right
~
20 now, I just don't understand what you're telling me, whether 21 we're talking about one site, a thousand sites. 22 MR. BERNREUTER: Okay. 23 (Slide.] 24 MR. BERNREUTER: What we did, just to be explicit, l l 25 in this particular case, if you remember what we did, we ) 1
I li4 1 performed a hazard analysis -- 2 MR. SIESS: That, I understand. 3 MR. BERNREUTER: Okay. 4 MR. SIESS: That figure, we've been through. l 5 MR. BERNREUTER: All right. And that's what this 6 step was. Then we said -- 7 MR. SIESS: And this is the one based on -- . 8 MR. BERNREUTER: This for the average of the five 9 and the ten hertz, I mean this particular curve. 10 MR. SIESS: Yes, but the previous one, the other 11 figure had different parameters. 12 MR. BERNREUTER: My figure? 13 MR. SIESS: No. l 14 MR. BERNREUTER: Oh. 15 MR. SIESS: Nilesh's. l 16 MR. BERNREUTER: Oh, he had another figure which 17 was -- one was for the median and the other one you saw was 18 for the mean. 19 MR. SIESS: Okay. And this is for the design
.- \
20 basis? This is the .3, isn't it? I l 21 MR. BERNREUTER: Oh, excuse me. He didn't show ? \ l 22 the mean. What was it? .3 23 (Slide.] i 24 MR. BERNREUTER: Oh, that's Level 3. You're ; l
- 25 right. Excuse me. Excuse me. Wrong figure.
l
1I5 1 MR. SIESS: All right. Now what number did you 2 take off of that? 3 MR. BERNREUTER: Okay. In this particular -- 4 MR. SIESS: We're still talking about the 5 equations that are on the following sheet. 6 MR. BERNREUTER: Okay. Now, here, this particular , 7 graph was for the average of the probability of excantang 8 the average of the five and ten hertz spectral -- 9 MR. SIESS: I know what that graph is. 10 MR. BERNREUTER: Okay. 11 MR. SIESS: I'm asking you what the number is 12 represented by A, by ASSE, by -- 13 MR. BERNREUTER: Okay. Here, I was trying to 14 explain was that ideally, I should have followed Nilesh's 15 logic all the way through it and used always the average of 1 i 16 the five and ten hertz, okay, to do my comparison and 17 unfolding. But, as I was trying to explain, unfortunately, j 18 that required a fair amount of reprogramming for-me to do 19 that in this short period -- to prepare for this meeting, 20 and so it was simpler to trent individual estimators. 21 MR. SIESS: I'd be glad to take this up at another 22 meeting if it would help you get the answers that we could 23 use. 24 MR. CHOKSHI: Dr. Siess, it is for the probability 25 of exceeding PGA for each site.
li6 1 MR. SIESS: Okay. 2 MR. CHOKSHI: It will be a similar curve. The 3 number will be different, but it will be the same, similar 4 type of curve. If I plot probability of exceeding PGA, I am 5 generating -- 6 MR. SIESS: Look, what I'm looking for for the 7 subcommittee and for me, I hope, is two things. First, an 8 example or more than one example of how the process p 9 described in the regulatory guide is applied to a specific
)
10 site -- 5 I 11 MR. CHOKSHI: Right.
- 12 MR. SIESS
- -- or two or three or four specific 1
13 sites. That's one thing I'm looking for, an explanation of 14 how the method that we're talking about publishing as a i 15 requirement of the NRC is applied, not how some other method
/
4 16 is applied. , 1 17 The other thing we're looking for -- that's I l 18 examples. Let's see numbers, diagrams, seismogenic zones, 19 whatever. 20 The other thing we're looking for is how that 21 method or the combination of the two deterministic and 22 probablistic methods applied to existing sites give -- what 23 design ground motions they give compared to the ones that we 24 used for the design of that particular plant at that 25 particular site. Two separate things which can't be
II7 1 combined. 2 Now, if you are not at a stage where you can give 3 that kind of information, I'd be happy to defer that 1 4 discussion to another meeting. 5 MR. BERNREUTER: Well, I think that particularly 6 looking at just using the five hertz spectra alone going 7 through with it gives you really what's going to happen when 8 you use the average of the five and ten. It'll really show 9 you basically -- 10 MR. SIESS: So far, I'm not even sure where we 11 are, whether we're in the second box on the lefthand side or 12 the third box on the lefthand side. 13 MR. BERNREUTER: Actually, we're trying to be 14- right here in this third box, okay.- l 15 MR. SIESS: Determine the probablistic controlling 16 earthquake. 17 MR. BERNREUTER: We wanted to determine the l d 18 probablistic controlling earthquakes. 19 MR. SIESS: Now,'for that, I've got to have a
~
20 site. l 21 MR. BERNREUTER: First you have a site. 22 MR. SIESS: And I haven't seen a site. You've 23 started -- 24 MR. BERNREUTER: Oh, okay. I see your problem. 25 Okay.
. _ . _ . . ~ . . _ _ . _ . ~ . . _ _ . . . _ . _ . _ . _ . _ _ _ _ _ . _ . _ _ _ . _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ . . .
li8 l 1 MR.-SIESS: Yes. I'm just lost. l I 2 MR. BERNREUTER:. Okay. I think I-understand what 3 the problem here is. Okay. l 4' First, what you do is you pick a site. 5 MR. SIESS:_ All right. Let's pick one. 6 MR. BERNREUTER: . Pick a site.. l 7 MR. SIESS: Have you got one? 8 MR. BERNREUTER: And I picked four sites, l 9 actually. 10 MR. SIESS: Okay. l 11 MR. BERNREUTER: Okay? l 12 MR. SIESS: Do you have a map that shown where 13 they are? 14 MR. BERNREUTER: Well, I didn't exactly put then 15 on a map, but I've got -- okay. Actually, I picked four l 16 sites to look at. One was a site in New England, one 17 -- l 18 MR. SIESS: Is that at an existing location? l 19 MR. BERNREUTER: These are all at existing sites. 20 MR. SIESS: Can you name it or is it classified? l 21 MR. BERNREUTER: I asked them whether they wanted
~
22 me to actually name these sites or not, and it was sort of 23 decided not to. l 24 MR. SIESS: Well, I want the names or we can stop j l j 25 now. I 1 l l
. . ~ -
m _ _ _ . _ . . . . .- __ _ _ . _ - . _ _ _ . _ _ _ _ _ _ _ . . . _ _ _ _ _ li9 1 1 MR. MURPHY: Go ahead. l 2 MR. BERNREUTER: Okay. This was the Seabrook 3 site. 4 MR. SIESS: Seabrook. 5 MR.,BERNREUTER: Seabrook. This is Vogtle. This
~
6 cite was Bellefonte, and this site was Perry. 7 MR. SIESS: Thank you. 8 MR. BERNREUTER: Now, you pick a site. Actually, 9 in the actual applications, it would be a new site, but I 10 picked the four -- 11 MR. SIESS: Not necessarily. 12 MR. BERNREUTER: Not necessarily. All right. So 13 you pick a site. Now you perform your hazard analysis, 14 okay? You have your hazard curves and all your hazard 15 information. 16 Now, what you want to do is ask the question with 17 the idea of the probablistic controlling earthquake, Which 18 earthquakes are most dominating the hazard? P 19 MR. SIESS: Now, is that part of the requirements?
~
20 21 MR. MURPHY: Part of the guidance, yes. That 22 would be part of the process. 23 MR. SIESS: This would be described in Reg Guide 24 1015? 25 MR. MURPHY: Yes. , i l__-- _ _ - _ _ _ _ _ _ _ - - - _ - - - - _ _ - _ _ _ _ _ _ - _ _ _ _ _ - - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
r I 150 1 MR. SIESS: Give me a reference. I just want to 2 follow this now. 3 MR._ MURPHY: It is not in there fully described 1 4 today. You'd have to follow it through the summary of that i 5 rag guide. As you will see, to some extent, it limps 6 through that summary. It's not full-blown.
. l I " 7 !
MR. SIESS: I mean, we're not doing a typical ) 8 hazard analysis like was .done using the Livermore studies in j 9 the seismic margin. 10 MR. MURPHY: That is the first step in the 11 process. 12 MR. BERNREUTER: That's the first step. 13 MR. MURPHY: That's Box 1 on the left. 14 MR. BERNREUTER: We performed a Livermore and -- 15 MR. SIESS: I'm talking about now at a site. 16 MR. BERNREUTER: At a site. You go to a site -- 17 MR. SIESS: I've already got that 50 percentile 18 thing. 19 MR. MURPHY: Right. 20 MR. BERNREUTER: And you -- ' 21 MR. SIESS: I already know what I'm aiming for, is 22 ten to the minus four, right? 23 MR. CHOKSHI: The 50 percentile criterion defines 24 the probability of accidents level. 4 25 MR. SIESS: Okay. So now I've got my ten to the l
~ - - . , . .
121 1 minus four. I'm down to the third box. 2 MR. CHOKSHI: Now you go to the site specific 3 hazard -- 4 MR. SIESS: All right. 5 MR. CHOKSHI: -- enter with that probability and 6 find your ground motion level. 7 MR. SIESS: I want, to find the ground motion level 8 at ten to the minus four prcbability of -- 9 MR. CHOKSHI: Right. 10 HR. BERNREUTER: That's right. That's this right 11 here. 12 MR. CHOKSHI: And once you find that level, then < 13 you de-aggregate to identify what magnitude and distance 14 contributes to that level of hazard. That is the key. 15 MR. SIESS: Fine. 16 MR. CHOKSHI: Once you define the ground motion 17 level -- say, for example, it's a PGA, okay, ten to the 18 minus four, I look at the peak ground acceleration, it's 19 .25g, and what I'm going to do is we're going to look at l' 20 what magnitude and distance contributes most to the .25g 21 earthquake level. Is the magnitude five at 20 kilometers or 22 a magnitude of six at 80 kilometers. 23 MR. SIESS: Okay. 24 MR. CHOKSHI: That is the process. 25 MR. SIESS: Now, after I do that, what? l t_____
1$2 1 MR. CHOKSHI: Then you're going to compare with 2 your deterministic derived magnitude. For example, the 3 deterministic told you about 5.3 -- 4 MR. SIESS: My deterministic, I put something and 5 something, some M and D. 6 MR. CHOKSHI: Some M and D, right, which comes 7 down the path and come up with the M and D. 8 MR. SIESS: But this was a specific M out here at 9 some location and some distance? 10 MR. CHOKSHI: Some distance. 11 MR. SIESS: For my probablistic one, suppose it 12 comes out that it's this M and this D over here is dominant? 13 MR. CHOKSHI: It's going to come out most likely 14 more than one. As Don goes through the example, you will 15 see. There is a set of probablistic M and Ds and there is a 16 set of deterministic M and Ds, and at that point, in many 17 cases you are able to make determinations given the 18 proximity and the magnitude with the probablistic M and D 19 and then you can just go to the ground motion calculations. 20 MR. SIESS: Well, I want to pick which M and D to 21 use so that I get the right attenuation. 22 MR. CHOKSHI: It is determined deterministically. 23 But I think when Don goes through the examples basically 24 usi,ng the probablistic to come up with the ground motion 25 parameter for the site, I mean the level, and then unfold l
1$3 ! I what is contributing to that level. 2 MR. SIESS: I now have an example. 3 (Slide.] 4 MR. BERNREUTER: So, as I say, you come in at that 5 probability of exceedance that you've chosen, and you read 6 off the ground motion level corresponding to that. 7 MR. SIESS: That ground motion is still defined by - 8 a spectrum? 9 MR. BERNREUTER: It could be whatever ground 10 motion parameter you wanted to' pick. You could pick the 11 average of the 5 and 10 hertz or you could pick PGA. 12 MR. SIESS: Do I have a choice, or does the reg 13 guide tell me what to do? , 14 MR. BERNREUTER: Right now, the reg guide says, 15 provisionally, use the average of the 5 and 10 hertz. ; i I 16 MR. SIESS: Let's stick to what the reg guide ; 17 says. 18 MR. BERNREUTER: Okay. 19 Now, part of what we were trying to do in our 20 analysis was to assure ourselves, since it's somewhat slight l 21 arbitrary, not completely, but the choice of the 5 and 10 I~ 22 hertz, average of these two together, was to look at several 23 different parameters, just to see -- you know, other choices 24 of ground motion parameters, to see where that might lead 25 you. 1
l 124 1 Unfortunately, I think some of those are on a few l 2 of the viewgraphs. , 1 3 MR. CHOKSHI: What we have done is a sensitivity l i 4 study. Don looked at two ground motion parameters, one 1 5 spectral ordinate and PGA. So, we know what are the
~
6 ' implications of using, you know, different ground motion 7 1 parameters. So, he will show the results of both, j 8 MR. BERNREUTER: The idea of the PCE, or the l 9 probabilletic controlling earthquake, is, indeed, the j 10 probabilistic controlling earthquake are those earthquakes, 11 in terms basically of magnitude and distance, which most 12 significantly contribute to the hazard at this particular I l 13 ground motion level. That's what we wanted to find. l
)
14 (Slide..) ; I l 15 .MR. SIESS: Is PCE a new term? j l 16 MR. CHOKSHI: Yes. i i 17 MR. SIESS: I see EME. That's different, right? 18 MR. BERNREUTER: Right. 19 KR. SIESS: Okay. 20 MR. BERNREUTER: Okay. 21 Now, there was a single hazard model and a single L 22 ground motion model. The de-aggregation, of course, would 23 be very simple and easy to do and would lead to unique 24 results. 25 For example, in the Livermore study, we had'11 4
155 1 seismicity experts; in EPRI, there were six teams. So, we 2 had five ground motion oxperts with multiple models. EPRI 3 only had basically just three ground motion models, but 4- nonetheless, there was a multiple use of models and experts. , 5 MR. SIESS: Now, how is that different from the 6 so-called deterministic method? 7 MR. BERNREUTER: In deterministic, you have a 8 single model, and basically, in the end, you choose a single 9 ground motion model. 10 MR. CHOKSHI: You basically start with a single 1 11 source zone model or map. 12- MR. SIESS: I don't think we have ever had 11 13 seismicity experts, but we've certainly had more than one. 14 MR. BERNREUTER: But somehow, the alternatives had 15 to be jettisoned, because they were relatively low, too low, 16 or whatever. You had to make these qualitative judgments to 17 simply zero in on a single noclel, in the end, a single final 18 model that you're going to use. 19 MR. CHOKSHI: Those dacisions are made at each 20 discrete point, zones, ground motion model. You consider 21 them, but then select a model. 22 MR. SIESS: As I recall, we always had two models. 23 One was the applicant's, and the other was the staff's. 24 MR. BERNREUTER: That often was the case. 25 (Slide.) l a
,, _. - -. - = - _ - _ . ~ . - - - . - . _ - . . . - . - -. .- - ...-- -
1$6 1 MR. BERNREUTER: Because of this complexity, then 2 the next step of de-aggregating the curve is a bit complex, 3 and there is probably more than one way to do it. 4 Now, as I say, because of this complexity, then we 5 have to go through a slightly complex procedure to de-6 aggregate the curve, and effectively, what we want to do is 7 plot the entire hazard surface.
~
l 8 We want to first do it for each cf the seismicity l l 9 experts, and then we want to combine them, so that we get 10 the relative weights of the seismicity experts in there, and 11 then look at that combined hazard surface. 12 So, you have to do that in a numerical process to l 13 do it in .s discrete nummer of points. l l 11 4 So, as a initial step, I broke it up into, 15 effectively, 16 discrete points that plot out this hazard 16 surface. 1 17 MR. SIESS: Are you describing the process that 18 has to be gone through to follow the procedures required by l 19 new Appendix B, or is this a research project? 20 MR. CHOKSHI: This is, conceptually, how one would 21 come up with a -- l 22 MR. SIESS: You mean a licensee is going to have l 23 to go out and have 11 experts? 24 MR. CHOKSHI: No. Tnat's why we are saying use 25 Livermore and EPRI. That all is done. The data bank is l l l
i l2'7 1 there. 2 MR. SIESS: Well, then if that's already been 3 done, why are we discussing it now? I'm lost again. l I 4 MR. CHOKSHI: Don was just describing that he has 4 5 to come up with -- $ 6 MR. SIESS: Are you describing how you got the 4 l, 7 Livermere? l 8 MR. BERNREUTER: Well, the Livermore and the EPRI j 9 results, at least as you grind through them, simply gave you r 5'
- 10 the aggregated, over all experts and everything.
. 11 MR. SIESS: What do I need to calculate the 4 12 seismic design basis for my plant? Do I have to de-4 13 aggregate something? 14 MR. BERNREUTER: That's correct. To come up with 15 the PCE, you have to de-aggregate that set of calculations. , 16 MR. SIESS: I can't just do -- is that what'.s been 17 done every time somebody has looked at a site with the 18 Livermore or EPRI curves? 19 MR. MURPHY: No, it is not. 20 MR. BERNREUTER: No. This is something new. This is something different. I 21 MR. SIESS: o 22 MR. BERNREUTER: This is something different. 23 In the Livermore study, we partially did this for 24 the staff's benefit, for them to understand, because they l 25 were interested in where the' hazard was coming from, and not 4
128 1 only that, we did something like this, although not quite 2 thi: extrouive, for the staff, for them to use in coming up 3 with their resolution of the Charleston issue. 4 So, this is just a continue of that idea. That 5 was it. The staff has used this de-aggregated idea in their 6 work. 7 That is, they use the de-aggregation of some 1 8 hazard curves to resolve to their satisfaction the 9 Charleston issue. 10 MR. CHOKSHI: More than anything else, Dr. Siess, 11 there is the display of information which is already in the 12 Liveracre and EPRI calculations. In order to Cisplay this 13 information or pull this out, Don is describing the steps. 14 MR. SIESS: If I'm doing this, where would I look 15 for that information? 16 MR. CHOKSHI: You ses part of that in two or three 17 cases of your report. 18 MR. SIESS: Are those refere*.ced in the Reg Guide? 19 MR. BERNREUTER: Not yet in this Reg Guide. It's 20 going to be in there. 21 MR. SIESS: Will they be referenced in the Reg I 22 Guide? 23 MR. BERNREUTER: Yes. 24 MR. MURPHY: Yes. 25 MR. BERNREUTER: And/or examples will be given, I I
13 9 1 think. Some of this doesn't exactly per se exist in quite 2 this form anywhere that I know of. J 3 MR. SIESS: I don't know whether this procedure is 4 as complicated as'it sounds. 5 MR. BERNREUTER: It's really not that complicated. l 9 6 MR. SIESS: Or whether you jtot can't explain it i 7 to me. , 8 MR. BERNREUTER: I think I'm not doing a good job i 9 of explaining it. [ 10 MR. SIESS: Unfortunately, I figure if somebody 11 knows what they're doing, they can explain it, and I'm not i 12 getting it. And I'm beginning to wonder whether we can ever 13 choose a site by this process; whether it wouldn't be better 14 if the staff would simply say that if you will have a site, 15 we will determine the safe shutdown earthquake groundmotion 16 and contract it out to Livermore. Why not use a third-17 party? Are we going to go through this process -- that the 18 licanuee is going to come in, after going through what I 19 consider a very complicated and, at this point, in my mind, 20 obscure process and say this is my proposed groundmotion and 21 the staff is going te say oh no, we don't like that one, go a 22 back and try again? And the staff will presumably have some 23 basis for saying it, but they won't tell the licensee , 24 because they never have. 25 Wouldn't it be better if we could agree on a
-130 1 third-party that would go out and decide these things?
2 There's only going to be a couple of dozen sites in the next 3 few years while we work this thing out. But this -- there's 4 just something wrong about what we're doing here. It's not j 5 transparent to me what we're trying to do. Now, maybe I 6 just need to have something written out.
- l 7 I went through Appendix A to the Reg Guide and I 8 couldn't follow that. I'm not following this because I 9 can't get numbers. I see diagrams up there that presumably 10 -- I get 11 seisnic experts to fill in, I'm not sure.
11 That's what I'm hearing. All I know is what I'm hearing. 12 MR. BERNREUTER: The point was of the 11 experts 13 was to indicate it's not a completely simple process to 14 deaggregate this because of the fact that you do have 11 15 experts and a number of groundmotion models. You have to go 16 back and look at your hazard curves and do some 17 calculations. 18 MR. CHOKSHI: Maybe, Dr. Siess, maybe if we go to 19 the specific example, and then come back to the general 20 procedure, it might be easier. 21 MR. SIESS: That's what I'm asking for. 22 MR. CHOKSHI: Then why don't we start. 23 MR. SI?SS: The sensitivity studies -- I'd rather 24 see those after I see one case. l MR. CHOKSHI: Then come back -- how those M&D were i 25 l I
. ~ . - . - .-. . _ . . - - . - . . _ - - . - . _ . . - . - - - - . - . _ . . - - - . - . - - -
151 i f 1 derived, and after we look at some results. 1 2 MR. BERNREUTER: That might be the easiest thing f 3 to do. I think the whole point of this was just simply to j 4 point out that this was done -- in a sense -- I looked at i 5 the hazard surface in 16 discreet boxes, and this was sort l* 6 of defining the boxes there. 1 7 MR. CHOKSHI: Start with your 14 maybe. 8 MR. SIESS: I'd much rather approach this thing l l 9 the way I think somebody would do it -- perhaps that could i j 10 select it -- that is, they'd have a map and they'd have a j i 11 site and they'd have a zonation at Livermore or EPRI, and 12 they could come up with some numbers. And I just can't i 13 follow annotated tables. I want you to try to show me 14 something. 15 If you take what's on the screen right now, I i 16 don't know what it means. I don't know what those numbers 17 are. 18 MR. MURPHY: Dr. Siess, let me offer a suggestion. 1 19 We seem to have been considerably bogged down here in making 20 this clear to you as to what's going on. I'll basically 21 assume that that's our fault in trying to be too complicated
.. \
22 about this. 23 MR. SIESS: It could be that I'm just stupid. 24 MR. MURPHY: No, no, no. I'm not stupid on 25 everything, but as far as this goes, I'm starting off'on a i
1$2 1 pretty elementary level. 2 MR. MURPHY: We, I think, need maybe to sit down 3 ani vork through this process maybe over the' lunch hour and 4 then co.'e back to you later this afternoon and then explain 5 to you what's going on. Because, obviously, at this stage, 6 we have gone ahead and made this thing far more complicated 7 than it actually is, at least in my mind. 8 MR. SIESS: I think you just assumed I knew a lot 9 more than I did. But, let me ask you something quick. Take 10 Seabronk site. What will this come out with for a "G" 11 value? You've got some curves at the back that I think mean 12 something. 13 MR. BERNREUTER: I have got the Seabrook site, if 14 you want to go back to it. 15 MR. SIESS: Well, there are some curves with S, D 16 and M on them. i 17 MR. BERNREUTER: All right. Let me gat that 1 18 figure. 19 (Slide.) 20 MR. BERNREUTER: This happens to be the Seabrook. l 21 Okay. Take a look at it. What this is -- this is the 22 design -- the -- l 23 MR. SIESS: First, let's start off -- what are the l i 24 axis and -- 25 MR. BERNREUTER: This is response spectra. I I
1$3 1 MR. SIESS: These are velocity spectrum? 2 MR. BERNREUTER: Velocity spectrum. This is 3 period. 4 MR. SIESS: Okay. 5 MR. BERNREUTER: Okay? 6 MR. SIESS: Why velocity and not acceleration? i 7 MR. BERNREUTER: For some reason-or another, we 8 chose when we did -- NRC chose -- oh -- the primary reason 9 for the velocity spectra was because NRC generally had to -- 10 often had their spectra plotted out in the velocity spectra, 11 so they could use triparte lot paper to come up with the -- 12 triparte lot paper. So, they wanted the spectra that we 13 generated, in terms of velocity spectra rather than 14 acceleration spectra. But, it generates just as long terms 15 of acceleration. 16 MR. SIESS: For some people, they can look at one 1 17 and see the other. I can't. < 18 MR. BERNREUTER: Okay. 19 MR. CHOKSHI: Engineers like it. 20 MR. SIESS: Engineers think in terms of 21 acceleration. 22 MR. BERNREUTER: I chose the velocity, but I could 23 have just as well have -- unfortunately, I just -- 24 MR. SIESS: In general, if the velocity is higher, 25 the acceleration is going to be higher.
l l'34 1 MR. BERNREUTER: This is the design spectra. The 2 curve labeled "D" is the design spectra for the Seabrook 3 site. 4 MR. SIESS: That was used for the Seabrook site? 5 MR. CHOKSHI: That was used for design. I think 6 it was -- I think -- 7 MR. CHOKSHI: That's the .25g. 8 MR. BERNREUTER: .25g site. 9 MR. SIESS: The shape of the spectra for Seabrook 10 would have been determined by the Reg Guide? 11 MR. CHOKSHI: 160. 12 MR. BERNREUTER: Right. Presumably. 13 MR. SIESS: And the anchor point would be where? 14 MR. CH9KSHI: .25g. 15 MR. SIESS: And where would that be in terms of 16 velocity? 17 MR. BERNREUTER: Well, it would be -- well, 18 actually it would have to come out a little bit -- come down 19 about right here.
)
20 MR. CHOKSHI: At about .33 it would be. J 21 MR. BERNREUTER: Yes. It's almost there. So,
\
22 you'd have to extrapolate this out to .33. I i 23 MR. SIESS: So, the anchor point would be i l 24 somewhere off to the left? , 25 MR. BERNREUTER: Yes, come down here, about right 4
1$5 1 here. 2 MR. CHOKSHI: Pretty close to B. 3 MR. SIESS: Okay. 4 MR. BERNREUTER: Now, in using the -- for this l 5 -particular site, using the -- deaggragating, we determined, 6 in this particular case, that the -- one of the 7 probabilistic earthquake worked out to be a magnitude 5.9 at 8 effectively 19 kilometers from the site over that area. 9 . Sort of an. average. That sort of worked out to be the 10 average of the box. Okay? 11 MR. SIESS: That would have been the earthquake 12 that would have been used under deterministic? 13 MR. BERNREUTER: No, no, no. That's what the 14 probabilistic controlling -- that's the piece -- yes, that - 15 - if we're using this method -- 16 MR. SIESS: It's the biggest earthquake? This one 17 contributes to most of the groundmotion at the site, or does l 18 it contribute tc Til of the groundmotion at the site? 19 MR. BERNREUTER: Most. ~ 20 MR. CHOKSHI: There are some others. 21 MR. SIESS: Okay. So, you picked one earthquake 22 that contributes most of the groundmotion? And then what 23 have you done with it? 24 MR. BERNREUTER: There are a lot of ways, and I 25 think Phyllis will later discuss actually how they will
l 136 1 carry that completely through the analysis. 2 I chose, just because I had it available to me, to 3 use the ground motion models that the experts supplied. 4 They supplied us with five. 5 Since there were five ground motion experts, each 6 one chose what he felt was the best ground motion model, . 7 said this is what I think is the best ground motion model to l 8 use in the eastern United States. ! 9 MR. SIESS: You chose to do this. Could an 10 applicant chose to do it this way? J 11 MR. CHOKSHI: The procedure is, once you come up ! 12 with this M&D, you basically use our SRP 2.5.2 procedures to 13 come up with the response spectra. , 14 ! once, you know, you have this magnitude 5.9, then 15 it's detailed upward same as a deterministic earthquake, as 16 opposed to, from a deterministic, you came up with 5.3 at 25 17 kilometer. j 18 The ground motions are available using 30 percent. 19 There is no more probabilistic after that. l
- 1 20 MR. SIESS: Okay.
21 MR. CHOKSHI: Now, what Don has done, in order to 22 show you some calculations, he has gone and he has used one 23 ground motion model attenuation at that distance, and these 3 24 are the research response spectra resulting from those 25 calculations.
- =- . .
l$ 7 1 M is the median spectra, S is the median -- I mean 2 mean spectra, and S is the mean plus one standard deviation, 3 which we use -- I mean the deterministic. r 4 What it shows is that, for the most part of the 5 frequencies significant to engineering structures and 6 components, deterministic is enveloping quite a ways to 7 probabilistic. , 8 MR. SIESS: Does that I mean that I have been too 9 conservative, then? 10 MR. CHOKSHI: That is one way to interpret it, but 11 if you wanted to stay in the 50 percent probabilistically, 12 you will be able to relax some. Whether, deterministically, 13 we will do that is a different question. 14 MR. SIESS: Now suppose I have a site and I find 15 that there are two zones, seismic zones; each contributes l 16 about half. 17 MR. CHOKSHI: Right. 18 KR. SIESS: And one of them is 20 kilometers from 19 the site and the other is 50 kilometers from the site. 20 Obviously, I'll get different attenuations and different 21 magnitudes. How do I handle that? 22 MR. CHOKSHI: Same way. 23 MR. SIESS: Take both of them and go back to the 24 deterministic method? 25 MR. CHOKSHI: Exactly. That's right.
-- - .. - . .... . .-. . - - _ - - . - - - . - - _ . . . ~ _ - . - _ _ _
i 1 1$8 i i 1 MR. SIESS: And use.the ground motion model? l 2 MR. CHOKSHI: Same as deterministic, and what will 'i 3 happen, in most cases, that certain part of the spectra 4 frequency in the spectra will be governed by one and certain 5 part of the frequency will be governed by the other, just 6 like in our deterministic. 7 MR. SIESS: Now, in this case, one earthquake 8 contributes all of it or most of it. 9 MR. CHOKSHI: Yes, most o'f it. That's why this 10 illustration was the only one. 11 MR. SIESS: But if that earthquake only 12 contributed 80 percent, then I go back to the deterministic 13 and up it to the 100 or something? 14 MR. CHOKSHI: Or just to see. 15 If you already selected deterministic and, for 16 example, it was much closer in that what probabilistic come 17 out, then you know that the deterministic ground motion 18 prediction will be higher than the probabilistic, and you 19 don't have to go through that step. 20 In many cases, that might be obvious once you 21 determine M&D. In some cases, you will have to do the 22 ground motion calculation and compare the two. 23 MR. SIESS: So, if I have two earthquakes that 24 contribute, each contribute 50 percent, I obviously don't 25 assume they both occur. I take whichever one gives me the 5
__ _ .._. m . _ . _ . _ _ _ . - _ _ - . _ _ _ . _ _ _ _ _ . _ . . _ . _ _ _ _ _ _ _ _ . . . _ _ . _ f 159 1 worst case and use that. ! 2 MR. MURPHY: In effect, if it's that clear-cut, l l 3 yes. Apparently, t's not that clear-cut. Apparently, one 4 earthquake will dominate at low frequency and another one l 5 will dominate at the high frequency. 6 MR. CHOKSHI: In some cases. 7 MR. McMULLEN: An interesting result, when you !~ q l 8 look at these results, if you base it all on looking at the , 9 . median hazard curve, generally speaking it's all controlled 10 by close-in, nearby earthquakes. 11 The hazard surface sort of has a sort of sharp 12' spike like that. Basically, one earthquake or maybe two j 13 earthquakes will basically dominate in a sense. 14 If you go to the mean hazard estimats and use that 15 to look at your information, then the situation becomes much 16 more complex. Then you will have several earthquakes.all l 17 contributing about the same amount. The hazard curve is a 18 bit flat. The hazard is coming from three or four different 19 sources. 20 MR. SIESS: What if I'm out in the middle of 21 Indiana? What's the nearby, close earthquake? Is that 15
* , l 22 miles out, take the 7 and put it 15 miles out?
23 MR. BERNREUTER: Well, the deterministic analysis 24 would be basically using an intensity 7, converting it to a 25 magnitude 5.5 or thereabouts, at about 15 kilometers from i i i 1 l i 1 i
.- - ~ ~.. - ..-. -. --. - - . .--- - _ . -...~..-.-. -.- -. . . . - . -- . .
l a l' 40 1 the site. 2 MR. SIESS: Yes, 15 kilometers. Now, what do I
'3 get out of this, though?
4 MR. BERNREUTER: Well, out of this thing, let's 5 see. We don't have Indiana, but we do have Ohio. 6 [ Slide.) 7 MR. BERNREUTER: Here is Ohio. Let's take a quick 8 look at it. This is just one way of getting the -- here, we 9 found that one of the controlling earthquakes was -- using 10 the median hazard estimator, the controlling earthquake was 11 actually found to be magnitude 5.6 at about 20 kilometers. 12 You have to take some of these distances a little 13 bit with a grain of salt in the sense that sort of a crude 14 analysis went into it. 15 MR. SIESS: What was used for Perry? 16 MR. BERNREUTER: Excuse me? 17 MR. SIESS: What was used for Perry? 18 MR. BERNREUTER: Intensity 7. 19 MR. SIESS: Was it a random 7? 20 MR. BERNREUTER: Random 7. I think that's 21 correct. 22 MR. SIESS: So, it was a typical, central, stable 23' -- put a 7 and don't put it right under the plant. 24 MR. BERNREUTER: The only place in the central
; 25 stable region, I think, that's really different than that i
- i i
141
- 1 were some plants up there in north central Illinois were 2 given intensity, I think, 8.
3 MR. SIESS: Well, something got Quad Cities up to 4 .22 g. 1 5 MR. BERNREUTER: Well, that was an earlier site, 6 actually, as I recall. I think that was also -- in fact, it 7 was a soil site, and I think they gave it some amplification 4 8 for the soil. 9 What you see is that what happened is, of course, 10 that the -- here is the design, which is -- I'm pretty sure 11 it's the reg guide spectra for Perry. Nes. 12 The eastern U.S. ground motions, these have had 13 some correction put into them for the fact that you didn't 14 see a lot more high frequency in the east than you do in the 15 west, and so, you see that you, generally speaking, wind up i 16 exceeding the design. l 17 MR. SIESS: Now, this ground motion model that you j 18 used for this -- j 19 MR. BERNREUTER: Actually, what I used was -- 20 MR. SIESS: It consistently gives a different 21 shape than the reg guide ground motion. \ 22 MR. BERNREUTER: Right. 23 MR. SIESS: Is that true of all the ground motion 24 models? . 25 MR. BERNREUTER: No ground motion model will give
152 1 you a reg guide shape. 2 MR. SIESS: Okay. 3 MR. BERNREUTER: Unless you happen to choose your 4 ground motion model artificially and say that, since you 5 didn't know too much about spectral information in the east, i l 6 you will simply devise a ground motion model where you take l i 7 the reg guide shap,e and shift it by PGA, and so, then you - 8 would get that. l 9 But other than that, none of the ground motion 10 models givefyou a reg guide shape, because it's an average 11 of a bunch of earthquakes that were all scaled to 1 g and 12 then you averaged them all together. So, it's not a real l 13 shape and had a lot of big earthquakes in it. 14 MR. SIESS: Let's follow Andy's suggestion, and we 15 will come back to this after lunch, or a little later. 16 In the meantime, let's take an hour off for. lunch, 17 and we can see where we pick up after that. 18 We will come back at five after one. 19 [Whereupon, at 12:06 p.m., the meeting recessed 20 for lunch, to reconvene this same day, Tuesday, December 10, 21 1991, at 1:05 p.m.] 22 23 24 25
4 I 143 1- AFTERNOON SESSION 2 (1:23 p.m.] 3 MR. SIESS: We're back in business. 4 I have proposed to Andy that we might get out of 5 the way a few items that I think are relatively clean, 6 technically, but I still have some questions about, and to
, 7 be perfectly frank, they need some editing, and I thought we 8 could take Appendix S, new Appendix B, and then three draft 9 reg ' guides, 1016, 17, and 18, and probably get through them 10 fairly quickly.
11 At least, I could express my concerns, you could 12 respond as necessary, or some of them, you may agree with 13 and say that needs to be fixed or something. 14 Is that all right? [ 15 MR. MURPHY: That sounds fine. 16 MR. SIESS: Let's start with Appendix S, whi'ch is l 17 all the engineering criteria that were in old Appendix A and 18 that really belong in Part 50 and particularly to be 19 consistent with the separation of source term and siting, to 20 get that stuff over into Part 50, where it belongs.
. 21 I didn't have a great deal of problem with this.
22 It does need editing. I don't know whether a technical 23 editor has been at it or not. I j 24 MR. KENNEALLY: They have not looked at it yet, 25 Dr. Siess.
144 1 MR. SIESS: My experience has been that they do a 2 pretty good job. 3- On page three, I guess, there's a definition of 4 zero period acceleration that I'd never seen.before. I 5 thought I knew what zero period acceleration was, and here , 6 it's defined as an asymptote. Can somebody explain, by a l 7 sketch, how you get the zero period? - 8 MR. CHOKSHI: Same as PGA. There is no I l 9 amplification. It remains constant. 10 MR. SIESS: This is where you have a complete I 11 spectrum, not an arbitrary spectrum. l 12 MR. CHOKSHI: Complete spectrum. It's the same as 13 PGA. l 14 MR. SIES'S: Okay. , 15 MR. KENNEALLY: The definition ch.. from one of 16 the national standards. I think it was ANSI ANS 2.2 and 17 again 2.10. I thought that it would be a little better 18 using one I thought was recognized. 19 MR. SIESS: That's a good reason. Having written , ; 1 20 national standards, they're not always that smart either, 21 you know. - I 22 In the next part, on page three, there's some l 23 stuff that needs to be fixed. There is a line that says 24 " derived from a free-field ground motion response spectra," 25 and there still seems to be confusion between what's
l l \ . 145 1 singular and what's plural. , 2 I'm afraid " data" has become plural in even modern , 3 dictionaries, but I don't think " spectra" has become I 4 singular. 5 At'the bottom of that, it says, "A nuclear power 6 plant shall be designed so that, at the safe shutdown
" I assume it means certain . 7 earthquake ground motion . . .
8 systems are okay. I think it means for the surface. . 9 MR. KENNEALLY: What page are you on, Dr. Siess? 10 MR. SIESS: At the very bottom It says "at the 11 ground motion." That's a location. I think it means if the 12 safe shutdown -- 13 MR. KENNEALLY: Could you please direct us to a , 14 line number? 15 MR. SIESS: Bottom line. You can get Wordperfect 16 to number your lines consecutively through the document, so 17 you don't need page numbers and line numbers. 18 It just says "at the safe" -- you know, for the -- 19 during the safe shutdown, following it, or whatever. It's 20 just that "at" isn't the proper word. 21 MR. KENNEALLY: All right. 22 MR. SIESS: And then, at the top of that page, we 23 now have a definition of seismic category 1 systems 24 components, which isn't called that. , 25 MR. CHOKSHI: This is carried over from the l
I i e' 146 l l I 1 current definition. 2 MR. SIESS: I know. I know. But we find it l 3 elsewhere. In fact, if you turp to page five, beginning l 4 with line 23, you will find the same thing defined. Do you
.5 see it? It's identical.
6 MR. KENNEALLY: I think I see one of the l 7 differences, Professor Siess. You're working on the reduced . i 8 text version. We have the comparative text that was sent to l 9 the staff. 10 MR. SIESS: Well, I've got the reduced text 11 version. i l 12 MR. KENNEALLY: Right. I 13 MR. SIESS: Well, I'll give you the designation. 14 It's under 4A1, and the second paragraph says, " Nuclear 15 plants shall be so designed that, at the safe shutdown, 16 earthquake ground motion, structures, systems, and l 17 components will remain functional. These systems, 18 structures, and components are those," one, two, three. 19 MR. KENNEALLY: Yes. . 20 MR. SIESS: Okay? l -21 Now, you go down to 2, operating basis earthquake, - 22 and 11 under that -- I'm sorry, go down to B, 4B. Got it? 23 MR. KENNEALLY: Yes. 24 MR. SIESS: Surface deformation, "These ! 25 structures, systems, and components are those necessary to . i l
- . - . . .. . - . - - - . . - . . . . . . . . - . - - ~ . -
4 I: 4 147 1 ! 1 . .
; 2 Now, if, somewhere, we had a definition of seismic 3 category 1 or whatever you called it, then it would simply
, 4 say those seismic category 1 systems, but here we've got the
- 5 same definition spelled out two places, with no name 6 assigned to it.
7 Now, that's only two. They happen to be in the 8 same one. There's at least three others somewhere. 9 On the top of Page -- let's see. I'll have to l 10 give you the section. In (ii), if an applicant chooses an 11 operating basis earthquake later than one-third the safe 12 shutdown earthquake ground motion, an explicit suitable 13 analysis -I don't know what " explicit" means there. It 14 must mean something because it doesn't say a " suitable 15 analysis" -- 16 MR. CHOKSHI: I think it's contrasting with 'the 17 item above, where there is n,o requirement for an explicit 18 analysis. 19 MR. SIESS: And again, things like this -- you're 20 comparing an operating basis earthquake with a safe shutdown s
- 21 earthquake ground motion.
22 MR. KENNEALLY: It should be ground motion. 23 MR. SIESS: See, the first thing that hit me is 24 that all of a sudden, the safe shutdown earthquake ground 25 motion has a new term, which is all right. I see the reason
148 1 for it. .But then the OBE is not defined as a ground motion, 2 and maybe there's a reason for that. I sort of got used to t 3 it, but I never figured out why, and here, when it compares i
- 4 one to the other.
j 5 My question is, is an OBE earthquake equivalent to , 6 an SSE ground motion? It may be that you expect them to '
.7 take an OBE ground motion rather than an OBE. .
L 8 MR. SIESS: Yes. Well, what you're'saying is that o i ! 9 in the second case, the OBE ground motion will be one-third l 10 of SSE, a simple scaled SSE ground motion to one-third.
]
11 MR. SIESS: It doesn't say that. i 12 MR. CHOKSHI: Yes. It doesn't say that. 13 MR. SIESS: So it's little things like that. 14 MR. CHOKSHI: Right.
- 15 MR. SIESS
- And then under required plant 16 shutdown, demonstrate to the Commission that no functional 17 damage has occurred. I would have said that a function has 18 not been impaired. To me, that's not quite the same as no 19 functional damage. No damage -- you'd have to define ,
20 " damage." 21 MR. CHOKSHI: You are right. The -- - 22 MR. SIESS: The functional damage I think was 23 intended to give the idea-that it w'as damage that impaired 24 the function. Why not say it? This is additive. ' l l 25 Under surface deformation, the design basis for i
i 149 1 surface deformation shall be taken into account. Now, is 2 surface deformation defined independent of an earthquake? 3 This would be the SSE and the OBE? 4 (Staff conferring off the record.] 5 MR. CHOKSHI: I think the surface deformation is 6 not necessarily tied to the seismic. ;
, 7 MR. SIESS: Let me read something the way I think l 8 you mean to say it. The potential for surface deformation 9 -shall be taken into account in the design of a nuclear 10 plant. Right now, it says the design basis for surface i i
11 deformation shall be -- 12 MR. CHOKSHI: Yes. 13 MR. SIESS: You haven't defined a design basis. 14 MR. CHOKSHI: No, we haven't. 15 MR. SIESS: Okay. This is editorial, but again, I
- 16 want to be sure just what you mean. Then again you repeat 17 the definition of seismic Category 1. '
t 18 Now, all your footnotes keep referring to 19 regulatory guides. Is it expected that there'll be a- ; 20 specific reference to the guide?
. 21 MR. KENNEALLY: Yes. We will put in the 22 appropriate regulatory guide number.
23 MR. SIESS: Okay. That's all I had there, mostly 24 stylistic and rhetoric. But this to me seems to me a fairly 25 straightforward translation and transposition from what's in
lb0 1 there now. 2 In Appendix B, and again, I've got the clean 3 version, which refers to things by sections, you have 4 definitions under III, Definitions, the expected maximum 5 earthquake is the largest earthquake that can reasonably be , 6 expected to occur in a given seismic source. Because of the l 7 uncertainty, the expected maximum earthquake is described by - 3 8 a distribution. 9 Now, for a distribution, I've got to have more 10 than one, and somehow I didn't get from that given source to 11 a distribution. I think it means something about a 12 distribution of the EMES from all sources, or am I wrong? 13 MR. MURPHY: No. It's -- 14 MR. SIESS: See, max.4 mum is one. 15 MR. MURPHY: That's correct. 16 MR. SIESS: So what is the universe for which I'm 17 getting the distribution? - 18 MR. MURPHY: What you're saying is there is a 19 maximum which, if you want, is the peak on a bell-shaped , 20 curve with a distribution as a log normal around that 21 maximum, so that you're not just picking the peak, but - 22 you're also describing what the distribution -- what you 23 expect that maximum to be. You want the probability. 24 That's the current maximum. ! 25 MR. SIESS: Well, then the words need to be "
\
)
151 1 changed tremendously. 2 MR. MURPHY: Okay. 3 MR. SIESS: Because you have defined it, the 4 expected maximum earthquake is the largest. That's the 5 peak. And then you say it is described by. But I've 6 already described it. I've described it as the peak. 7 MR. MURPHY: I see where you're coming from. 8 MR. SIESS: See? You've got two statements and 9 they don't agree. You really want to say something to the 10 effect that the expected maximum earthquake for a given 11 seismic source is described by distribution, something. I 12 don't know -- think it through. But right now, there are two 13 statements that just -- 14 MR. MURPHY: All right. 15 MR. SIESS: Now, the next item, Item C, safe 1 16 shutdown earthquake ground motion, is the earthquake f'o r 17 which category? The seismic category 1 structures. And we 18 now have the definition repeated. 19 You see, at least in the existing regulations, we l 20 define them once in Appendix A, and everywhere else, we 1 l . 21 refer to seismic Category 1. Here, we've got them defined 22 every time we refer to it and nowhere do we name them. 23 I thought maybe somebody just didn't want to call 24 that seismic category 1. 25 MR. KENNEALLY: Well, there was a little thought 1
152 1 at one time of cleaning it up the way you had,~and I 2 personally changed because I was scared of introducing 3 another term that would have farther confused the issue. 4 MR. SIESS: Well, the term is already there 5 somewhere. 6 MR. KENNEALLY: Right. Yes. 7 MR. SIESS: And it's used -- I suspect it's used . 8 in more places than what we're changing here. I'll bet you 9 I can find references to seismic Category 1 elsewhere in the 10 regulations. 11 It was always a little awkward. You had to go to 12 Appendix A to find the definition, but it was in -- it was e 13 in the Appendix S part of Appendix A. 14 MR. KENNEALLY: That's correct. 15 MR. SIESS: Now we've got it in the Appendix B. 16 It's not a bad situation, but again -- then there's a 17 definition for the operating basis earthquake, and I 18 couldn't find a reference to the OBE anywhere in Appendix B. 19 I don't really think it's necessary to define it. . 20 MR. CHOKSHI: We can take it out. 21 MR. KENNEALLY: At least for the start, we thought - 22 to maintain a little consistency, but I can see the 23 confusion it's going to bring in. , 24 MR. SIESS: Now, in one of the others, I found 25 things defined that did not appear in the text and things in i e
153 1 the text that weren't defined. Since you've got this on a 2 computer, you can find those. 3 In Item G, surface displacement is distortion of 4 soils and rocks. I think you mean "or" rocks. 5 MR. MURPHY: Yes. 6 MR. SIESS: "And/or." 7 MR. MURPHY: "And/or." 8 MR. SIESS: Next page, I, seismogenic source. I l 9 come down to the second sentence, "The seismogenic source is 10 not expected to cause surface displacement." Why not? I 11 MR. MURPHY: Why not? ) 12 MR. SIESS: Uh-huh. I i 13 MR. MURPHY: That's the differentiation between I l 14 the two that we're using now, the capable tectonic source 15 and the seismogenic source. The seismogenic source does not 16 have any capable features in it, or it's not expected, as it 17 says here, it's not expected to have capable -- 18 MR. SIESS: Well, not expected to -- 19 MR. MURPHY: -- to have capable features in it. 20 MR. SIESS: If you're wrong and it does cause . 21 surface displacement, does that make it a capable tectonic 22 source? 23 MR. MURPHY: Yes, sir. Yes. 24 MR. SIESS: So seismogenic source is a portion of 25 the earth that has uniform earthquake potential different 1
1 154 1 from the surrounding area, distinguishable, will not cause 2 surface displacements, and I changed your wording. Because 3 if it does -- now, a capable tectonic source can develop 4 both earthquakes and surface deformation. 5 MR. MURPHY: That's correct. 6 MR. SIESS: So if I have everything that meets 7 seismogenic source, but it does cause surface displacement, , l l 8 that makes it a tectonic -- I 9 MR. MURPHY: That's correct. 10 MR. SIESS: So you don't want to say it's not 11 expected to. Will not. 12 MR. MURPHY: Will not. .
< 1 13 MR. SIESS: And if you are wrong and it does, you
] 4 8 14 just named it wrong. 15 MR. MURPHY: Yes. And obviously the consequence f
! 16 of the analysis that we've done on the basis of naming it i 17 wrong.
i ' 18 MR. SIESS: Yes, but then you're going to have to 19 go back and change your category. That's just -- I thought , 20 that was editorial. 21 A capable tectonic source characterized by surface . l 22 deformation the last 500,000 years, at least once in the 23 Jer?. 50. That's changed from 35. Why? l . j 24 MR. McMULLEN: Richnrd McMuller. The 35,000 was j 25 based on the upper limits of one of the techniques most 4 a t
t 1b5 1 often used in determining the age of strata and-faults, and 2 the resolution of'that technique has been greatly improved 3 through the years. , 4 In fact, the resolution in most techniques had 5 been improved. 30 it's still somewhat arbitrary, but it's 6 --
. 7 MR. SIESS: Okay. But you were after something ;
8 fairly recent -- 9 MR. McMULLEN: Yes. 10 MR. SIESS: Geologically recent. 11 MR. McMULLEN:- Right. 12 MR. SIESS: Recent to me is-yesterday. And 35,000 13 was what you could measure; now you can measure 50. 14 The 500,000 you get on what basis? 15 MR. McMULLEN: That's another technique. 16 MR. SIESS: Different dating, but it -- 17 MR. McMULLEN: Potassium Argon, which dates older 18 aaterials. d
, 19 MR. SIESS: Okay. You didn't define " recurring 20 nature" before. You're not defining it here, and I guess 1 - 21 it's probably just as well.
22 MR. MURPHY: Right. 23 MR. SIESS: Two would be recurring. 24 MR. McMULLEN: Right. 4 25 MR. SIESS: Fifty would be recurring. Does it i I
lb6
'l make any~ difference?
2 MR. MURPHY: Yes, it would make some difference. 3 MR. SIESS: If it's two or 50. 4 MR. MURPHY: If it's two or 50. If it's not 5 distinctly different, it's something that you could work , 6 with in the probablistic analysis, the recurrence rate. 7 MR. SIESS: The only thing -- let's go back and do , 8 it. What wouldn't be capable? Hadn't moved in the last 9 50,000 years. That's not capable. 10 MR. NURPHY: Right.
'll MR. McMULLEN: Unless there's been more than one 12 movement. I 13 MR. SIESS: And if it hadn't moved -- I'm talking 1 14 about -- I'm doing the reverse. If I've got nc motion in 15 the last 50,000 years and I can't find more than one in the 16 last 500,000, it's not capable?
17 MR. McMULLEN: Yes, that's right. 18 MR. SIESS: Okay. I just want to be sure it works I 19 both ways. .. 20 MR. MURPHY: It does. 21 MR. SIESS: Editorial again. Item 3, a structural - 22 association with a capable tectonic source according to 23 Characteristics 1, having Characteristics 1 of this 24 paragraph. You have characteristics meeting the 25 requirements of Characteristic 1, but not at -- not
/
157 1 according to -- if that was a definition, you say according i 2 to the definition, but not according to a characteristic. 3 Next item -- I'm all the way down to the last
-4 item, o, in definitions. I guess we're still in 5 definitions. I would suggest that somebody check that all
.6 of these terms are actually uc-d in here.
, 7 MR. KENNEALLY: Yes, sir.
8 MR. SIESS: Zero period acceleration is a 9 numerical value that corresponds to the acceleration level 10 of the input design earthquake response spectra at
-11 frequencies where -- what bothered me a little bit was 12 defining -- using " period" in one place and " frequency" in i l'3 the other.
14 Is there anything wrong with making them both , 15 period? 16 MR. NURPHY: All right. L
'17 MR. SIESS: I had a comment down under required ,
l 18 investigations of the both probabilistic and deterministic. 1 l 19 That's a much broader issue that we're still debating. I'm 20 sure we're going to and up that way. >
- 21 In the section headed " Required Investigation,"
22 'we've got vibratory groundmotion, tectonic surface 23 defamation, non-tectonic defamation, seismically-induced 24 floods and waterways and volcanic activities. When you read L l 25 that, as an editor reada it, they're going to find out that
t l 158 1 E volcanic activities starts off in a way that's nonparallel 2 to the others. It's just -- the style changed all of a 3 sudden. That's just editorial. 4 Roman V-V " seismic and geologic design bases. C, 5 determination of earthquake groundmotion for the seismic 6 design basis." The first sentence says: "The safe shutdown 7 earthquake groundmotion." Should that be in the title? , , 8 MR. CHOKSHI: Yes. 9 MR. SIESS: Okay. I would suggest that you do 10 that. We might as well start to think of safe shutdown 11 earthquake groundmotion as one phrase. Safe shutdown j 12 earthquake groundmotion is defined by response spectra not i I 13 determined by. Determined from something else. 4 14 And I've got this envelope of composite and that 15 would come out late, it just wasn't clear to me. 16 When you talk about the lower half of the 17 population, do you think that won't mean simply saying the i 1 18 median, below the median? I 19 MR. CHOKSHI: We can make it that. ,. 20 MR. SIESS: One problem a lot of people have, some 21 of us, when we talk about lower than or greater than, in . l l 22 terms of probabilities, and we tend to forget the minus on I l 23 the exponent. And I think that -- 24 MR. CHOKSHI: We can make it clearer. 25 MR. SIESS: -- if you simply said it has a lower
19 1 probabilit*j of exceedance than the median, follow.the i , 2 others, and then you sort of got the picture stated. Just f 3 words. 4 Now, come over into -- still in five, C, D, E, F, 5 F, has subparts one, two, three, four. The determination of ; 6 other design conditions. And four is distant structures,
. 7 which I assume means things like water intakes. Those ;
8- structures which are not located in the immediate vicinity ., 9 of the site but which are safety-related. What's safety-f 10 related? Safety-related seismic category one -- three ' 11 ) requirements that we've listed separately? l 12 MR. CHOKSHI: Yes, it would have to be. 13 MR. SIESS: Now, the other definition of safety-14 related, I think, in in Appendix B. Isn't safety-related 15 and important to safety -- they give you a word by defining 16 things. I'd define that one. 17 MR. CHOKSHI: Yes.
.18 MR. MURPHY: Yes. I think that's one in the l
19 advanced reactors. It's one that's being debated. 20 MR. SIESS: Well, you know, we've been debating
. 21 for years safety-related versus important to safety, versus 22 seismic category one and, of course, the distinctions are l 23 blurring like mad, you know. We're beginning to find out 1
24 that there are a lot of things that can lead to failures ( 25 that we didn't have in there before. I e
160 1 Those were the kind of things I had. And I don't 2 have much problem with that, if it gets edited to make it a 3 little bit clearer what you -- Charlie, do you have any ; 4 questions? 5 MR. WYLIE: No. , 6 MR. SIESS: It could read a little bit better. 7 MR. MURPHY: Agreed. . 1 8 MR. SIESS: If you think about this business of j 9 maybe defining something once and not having to repeat it l 10 six times. It's twice in one of them and once in the other. i 11 And I'm not through. l j 12 1016,-whish is Rev 2 to Reg Guide 1.12, the old 13 seismic instrumentation. I didn't go back and look at 1.12, 14 but I have a sneaking suspicion that you have deviated and 15 that you're using a decimal numbering system, which I 16 commend you for, that are the combinations of upper case and 17 lower case Roman letters and little numbers in parenthesis, 18 and big numbers in brackets, and never knowing where the 19 hell you are. ,. , 20 And, in fact, I'm pretty sure you did it, because 21 when I looked at page four, item five, up at the top, it - 22 refers to Regulatory Position 1biv, which is 1.2.4. 23 MR. KENNEALLY: Right. We noticed that as well. 24 NR. SIESS: Didn't catch them all. There are some 25 other changes that you've done from typical Reg Guide
[ 161 1 formats. For years we worked a lot with the Reg Guides. I 2 had the Reg Guide Subcommittee. And I tried to convince the 3 staff that there should be a one-to-one relationship between 4 the discussion and the position. And the discussion was 5 really a commentary on the position. I'm used to writing an 6 ACI code in the commentary. And they gradually went to
, 7 that.
8 Now, you've deviated from that, and I'm not 9 complaining about it. Because your discussion here isn't a 10 discussion at all, it's a scope statement. It's a very good 11 statement. It tells you what you're going to have in here. 12 And in your regulatory position, not always, but frequently, 13 your first paragraph is a discussion -- an explanation of 14 why you have the position. It reads very well, so I'm not 15 complaining about that, I'm just pointing it out. 16 I was a little concerned with the item in the 17 discussion -- have we got the same copy now? 18 MR. KENNEALLY: I believe so. October the 9th? 19 MR. SIESS: Line 31 on page two. You know, you 20 will find it easier if you number your lines all the way
. 21 through the document. Wordperfect will do that. I say it 22 because ACR3 numbers letters that way when we're drafting 23 them, and it saves time. You can say line 134 instead of --
24 MR. KENNEALLY: Okay. I I 25 MR. SIESS: And you've got this point that you
l 1 ib2 1 elaborate on later. The instrumentation system should be 2 operable at all times. In other words, you didn't want to 1 3 give many -- there are too many cases where an earthquake 1 4 and instruments weren't working. And my first thought there ! 5 is, that's fine. It ought to be. But, let's don't shutdown i 6 plants simply because the instruments aren't working. i 1
- 7 There's a damn small risk of an earthquake at any point in .
l l 8 time. Whether it should be an LCO or two weeks or something l 9 like that, I don't know. 10 But, it goes on to say that "the guidelines that 11 will be followed by the NRC staff if the seismic 12 instrumentation is inoperable are identified." Which is l l 13 true, but they're identified in 1017, not in 1016. 14 MR. KENNEALLY: What I was trying to do was also 15 get the attention to the -- actually the footnote on page l l 16 five. 17 MR. SIESS: On page what? 18 MR. KENNEALLY: Page five. 19 MR. SIESS: Yes. ,. 20 MR. KENNEALLY: That needs clarification. 21 MR. SIESS: When you get to there you say it. At - l 22 this point, I started looking for it. 23 MR. KENNEALLY: Yes. I understand. ) 24 MR. SIESS: And for some reason, I think I missed 25 the footnote. But, anyway, page four, this is part of the 4 3 a
lb3 1 locations. Internal containment structures, which is 2 defined in the definitions as a structure internal to the 3 primary or secondary containment is supported by the 4 containment foundation.
)
But an internal containment ' 5 structure to me means it's part of the containment I 1 6 structure. You're really talking about structures inside 7 containment. 8 MR. KENNEALLY: Yes, sir. 9 MR. SIESS: And it hasn't got anything to do with 10 the containment? 11 MR. KENNEALLY: That's correct. 12 MR. CHOKSHI: Except on the.same common line. 13 MR. SIESS: Yes, but that's almost impossible not 14 to have if it's inside. But, on first reading, it was a 15 little confusing, and I don't think it needs to be 16 confusing. You could find a little better word. l l 17 MR. KENNEALLY: Yes, sir. 18 MR. SIESS: And five there is where you have the 19 wrong reference. g 20 MR. KENNEALLY: Yes. l
. 21 MR. SIESS: If seismic isolators are used, l 22 instrumentation should be placed on the rigid and isolation 23 portions at approximately the same elevations. Would that i 24 mean you wanted something above and below the isolator?
l 25 Below the isolator over here, and 50 feet over there?
l l L i 14 1 MR. KENNEALLY: The only case we have--- 2 MR. SIESS: Or was it all right as long as it was 3 the same elevation? 4 1(R . KENNEALLY: The only case we have right now I 5 for proposed seismic isolation might be the prism, where i 6 within that particular building there's a portion of the I 7 building that is isolated, not the entire building. . 8 MR. SIESS: But you wouldn't object to having it 9 180 degrees apart -- one on the supported structure and ona 10 on the supports? 11 MR. KENNEALLY: That's correct. 12 MR. SIESS: Wouldn't that be at the same azimuth? 13 MR. KENNEALLY: No. 14 MR. SIESS: Was it rigid enough at that level? 15 MR. KENNEALLY: Yes. 16 MR. SIESS: I had a note that maybe you'd say 17 placed on both the rigid and isolated portions, but it's 18 unnecessary. If it's on this and this it has to be both, 19 that's correct. 20 MR. KENNEALLY: Yes. 21 MR. SIESS: Now, 1.3 is a little bit misleading, - 22 because it starts off saying the location should be j 23 determined to obtain the most pertinent information. But i ! 24 the real thrust of 1.3 is ALARA. ,
- I s 25 MR. KENNEALLY: Yes.
i 4
_ _ . _ _ _ . _ . _ _ . _ - . _ . . _ . . _ . _ . _ - . . _ . _ _ _ . . _ . . . . . . - _ . ~ _ . _ _ . . . _ _ i , ! 165 l 1 MR. CHOKSHI: It's two different thrusts. i- 2 MR. SIESS: It left me a little bit unclear what j 3 governs. I've got the most pertinent information, the 4 seismic data, and I've got ALARA. How do I balance? Now, i , 5 by putting'the sentence you do first would tend to give that l 6 sort of the first priority. But there's no bridge in there
, 7 however.
8 MR. CHOKSHI: I guesp we need to say consistent 9 with the ALARA principles. 10 MR. SIESS: I mean, ALARA is nice. And actually, 11 in terms of overall risk, it might be more important than 12 the other, because the number man rem people are going to 13 get from maintaining seismic instrumentation may be a lot 14 bigger than they'll ever get from an earthquake. But you're 15 right. 16 These things should not be in horrible place's to 17 get to or places where people are going to be reluctant to 18 go in there and do it right. And there's no reason for them 19 to be. Have they ever put them in places like that? No one 20 has ever put one down underneath the reactor where all of l l
, 21 the heat force go through, at the bottom of the pit?
22 MR. CHOKSHI: Many of them, I think, are hard to l l 23 maintain. l 24 MR. SIESS: Item four you see is a little bit j 25 different. It's not related to ALARA as such. Maintenance I
. _ _ _ . . . . - - . _ - . _ . _ = . _ _ - . _ _ _ _ . ..__.._._.___._ _ _ _ _ __ _ _ _ _ _ _ l 1 166 1 installation. Again, it says " consistent with this 2 regulatory position," which I'd say consistent with one -- l 3 get it where you need the information, two, reduce the dose i 4 rates, three, facilitate maintenance.
]
i 5 Now, there's no indication of weighting. , l 6 MR. KENNEALLY: Where are we, Professor-Siess? ! 7 What item in particular? , 8 MR. SIESS: Item four. 4 1 9 MR. KENNEALLY: Okay. Okay. 10 MR. SIESS: At the top of page five. 11 MR. KENNEALLY: Yes, sir. 12 MR. SIESS: See, if you look at the organization 13 of 1.3, what you see is the'first sentence that says: "Put 14 them where you need to get the information, and that's what i 15 they're there for." The second sentence says, "try to keep 16 the exposures down." And it has three subheadings under 17 keeping the exposures down, right? 18 MR. KENNEALLY: Yes. 19 MR. SIESS: ALARA design, a low dose rate area, , 20 minimum maintenance and service inspection. That's time, 21 right? - 22 MR. KENNEALLY: Yes. ~ 23 MR. SIESS: That's one, two, three. 24 Now, four is also locate and facilitate
- 25 maintenance.
- It's not clear whether those are ALARA l
167 1 governing just to simplify maintenance. Because you've 2 already said put them where there's a low dose rate and l 3- minimum maintenance and in-service. Right? i 4 MR. KENNEALLY: Yes. 5 MR. SIESS: So, just what is behind for isn't 6 clear. And is it a part of 1.3 really? l
, 7 MR. MURPHY: We have to take a look at that.
8 MR. KENNEALLY: We'll have to look at that and , 9 clarify it. 10 MR. SIESS: I think you should look at that. 11 You'll see you've given several things, and they're not all 12 related. The relation among them and the priorities ancng 13 them isn't -- I'm not sure that I disagree with any, but l's 14 not sure that four shouldn't be a 1.4. l l 15 Again, a little nit maybe. In item two there, 16 multi-unit sites, the last sentence. In the case of 17 separate control rooms, annunciation is specified in seven l l 18 should be applicable to both. Annunciation shouldn't be ] 19 applicable. The requirements might be, or simply the 20 annunciation should be provided in both control rooms. l
- 21 Okay.
22 Now, incidentally, under item three, seismic -- 23 I'm on page five, seismic instrumentation operability. 3.1 24 is a good discussion. It's not a requirement in it. It's 25 an explanation for the requirements, which is what we used
168 1 to put in the discussion. I've got no objection. I think 2 it reads pretty good. The names might be wrong. 3 In 3.2, " Instrumentation should be maintained in 4 operation during periods of plant shutdown." 5 " Maintenance and repair procedures should make . 6 provisions for keeping the maximum number of instruments in 7 service during plant operation and shutdown." . 8 When it says, " Instrumentation should be 9 maintained . . ., " all of it, some of it? It says " maximum" 10 on the next page. 11 The footnote which talks about inoperable 12 instrumentation -- where is the reference? I see the 13 footnote, but I don't see -- 14 MR. KENNEALLY: I can see it. I'm looking right 15 now. ! 16 ' MR. SIESS: It probably belongs there. Okay'. ) 17 What's in here is good, but there are words the j l 18 somebody -- it is a reg guide, but again, you commit to the i 19 maximum. What's the maximum? ,. l 20 Page six, 5.3, under installation, " Protection 21 should be provided against accidental impacts." on the next - 22 page, 6.2, " Spurious triggering should be avoided." 23 Now, those may not be strictly the same, but the 24 main reason for avoiding accidental impacts is spurious 25 triggering. l
l 169 1 MR. KENNEALLY: That was one event, and also, it i 2 was damaging to the instrumentation itself. ! 3- MR. SIESS: It's a big enough impact, I guess. I 4 MR. KENNEALLY: There have been cases where the l l
)
5 people on the standards committee have told me that the 6 instrumentation has physically been hit during their I 7 walkdowns. They have noticed it during those walkdowns. I 8 MR. SIESS: Page seven, item 7, remote indication, 9 says, "On actuation of free field or any foundation level 10 time history, a remote indication of the. control room should 11 be activated." 12 Now, if you remember, earlier, it said that l 13 annunciation is specified in reg position seven. Is this 14 what was in mind? j 15 MR. KENNEALLY: Yes. 16 MR. SIESS: This is annunciation. 17 MR. KENNEALLY: Yes. 18 MR. SIESS: A remote indication is an 19 annunciation. 20 MR. KENNEALLY: Yes.
. 21 MR. SIESS: Is there any reason not to use the 22 same world?
23 MR. KENNEALLY: No. 24 MR. SIESS: Because everything is remote 25 indication. i l
l l 170 1 8.2, "The frequency of maintenance is . . . 2 Either the frequency of maintenance "should be" or the 3 " required" frequency of maintenance is, and I don't guess it 4 can be required, because this is a reg guide. 5 I'm over to page nine, Appendix A definitions. I 6 checked most of them. I'm not sure I found " safe shutdown 7 ground motion" in there, but anyway -- I'm not sure I found . 8 " acceleration sensor" in the text, but I didn't look, and 9 not having it on computer, I wasn't going to bother. 10 Look at the one that says " containment 11 foundation." I guess I wasn't sure why that needed to be 12 defined. 13 "For the foundation which supports more than just 14 the containment structure or reactor building, the aren 15 which is within the close proximity of the containment shall 1 16 also be considered as part of the containment foundation." 17 In the first place, "also" in addition to what?
- 18' If I've got a mat with the containment and the aux building 19 on it and a few other things -- and we did have a - ,.
20 Waterford is built that way -- I'd consider that the 21 containment foundation. -
'22 MR. KENNEALLY: Yes.
23 NR. SIESS: " . . . within close proximity of the
" -- I've got a 180-foot diameter 24 containment shall . . .
25 containment. Does that mean I can't put my instrument out
I l 171 1 in the middle of it? I've got to put it over near the 1 2 shell? Think that one out. 3 I am not sure that you need a definition of 4 " containment foundation." It's that chunk of concrete the 5 containment is sitting on. 6 Throughout this thing, you have-defined the j i 7 operating basis earthquake by its function. "The operating , i 8 basis earthquake produces the vibratory ground motion." l 9 "The operating basis earthquake is that earthquake 10 which produces the vibratory ground moti.on," and I think 11 you've been consistent with that. An editor will-pick that 12 up. 13 You see, you've got to get the OBE to the ground 14 motion. 15 MR. KENNEALLY: Right. 16 MR. SIESS: But you don't use the OBE ground 17 motion, as such, in your checks. 18 MR. KENNEALLY: Right. 19 MR. SIESS: You do use the SSE ground motion. . 20 Everything is built around it. I see the reason for it, and , , 21 I think it's helpful, once I'm getting used to it. At 22 first, it just bothered me. 23 Okay. That's the kind of nits I had in there, 24 except I find no requirement in here on the range of the 25 instruments that should be provided. Are we interested only
1 i 1 l 3 1h2 1 in the OBE? 2 I'm not even sure the SSE is even mentioned. It's 3 defined in here, but I don't think I saw a reference to the 4 SSE, and we certainly want instrumentation well beyond the 5 OBE, right? 6 MR. KENNEALLY: Yes. 7 MR. SIESS: Shouldn't this be covered? . 8 MR. KENNEALLY: Yes. I 9 MR. SIESS: Is it covered in 1.12 now? 10 MR. KENNEALLY: I believe it's covered in 1.12 11 mainly because the amount of instrumentation is detsrained l 12 as a function of the SSE. There is a cutoff. l 13 MR. SIESS: I think that stuff should be. 14 MR. KENNEALLY: Here, we have not made that fewer 15 instruments below a certain acceleration level. The plants I 16 will more or less all have the same instrumentation. 17 MR. SIESS: See, the whole emphasis here, taken in 18 conjunction with the other ones, is what instrumentation do 19 we need-to have to know whether to shut down, but there are ,. 20 other reasons for having that instrumentation. 21 MR. KENNEALLY: That's correct. - 22 MR. SIESS: If we have a nice, big earthqucke and 23 we get all this good information, it's going to be useful. 24 When we're putting anything in the plant, let's 25 take advantage of the fact that we're going to have an
. . _. . . _ ~ . . - . - - _ - _ - - - - . . _ . . .
1 3 1 instrumented structure there when the next good earthquake 2 comes along. l 3 We'll learn more from a well-instrumented plant 4 than anything else. So, you might look at the range. 5 There is some compromise between sensitivity and 6 complete range, but that should not be too big a problem. l
, 7 Okay. 1017, okay, and then we have 1018. I guess I read l 8 them in reverse order, because I've got a note that 1017 is 9 better than 1018. It does need editing.
10 You know what would help a little bit, I think, is 11 in the introduction. You always run it one great big long I 12 paragraph with all these 10 CFR parts in it. To get through 13 the part that says tliese are the parts of the regulations 14 that this relates to, and then make a new paragraph that 15 says what this Reg Guide is covering. 16 For example, on this particular one, I guess'it's l 17 down to the last five lines. These are two separate 18 thoughts, particularly when it's related to an industry 19 document. 20 MR. CHOKSHI: Yes.
- 21 MR. SIESS: Now, this one is really e Reg Guide
[ 22 which endoraes an industry standard, with some exceptions. i 23 MR. KENNEALLY: It's basically -- you're on 1018 I l l t 24 or 1017? 25 MR. SIESS: I'm on 1017. I picked them up in l a
l l t 174 l 1 order. I'll take the 1018 if you want. I 2 MR. KENNEALLY: No. It's either one. 1017 is 3 basically -- it's not an industry standard. It's more of an 4 EPR report. 5 MR. SIESS: Well, let's call it that. 6 MR. KENNEALLY: Fine, it is a standard with 7 exceptions, then. . ! 8 MR. SIESS: There is NCANS 210 is mentioned in . i 9 here, but I don't think it's endorsed; is it? l 10 MR. KENNEALLY: No, it is not. The 1979 version 11 is the one that's currently available. They were hopeful of 12 having the revised document available before we went out to 13 public comment, at which time, it would have made my , 14 endorsement a little simpler. 15 MR. SIESS: Yes, but it says on page 3 that the 16 Reg position is a combination of several items, and th's 17 first is the ANSI. 18 MR. KENNHALLY: It quotes it as far as being the 19 source of what I considered, but I'm not endorsing it. ,. 20 MR. SIESS: Okay. 21 MR. KENNEALLY: That particular one is a little - ! 22 history. The revision of that got hung up with the OBE [ 23 exceedance, and they're still struggling to get that through 24 their last balloting, and it looks like Spring or whatever,
- 25 it will probably be on the street.
4 4 I
I i i 175 l 1 Maybe when this came back, there's a final guide,
~
2 after the public comment resolution and there may be an 3 endorsement of it. We'll have to see what it looks like. 4 MR. SIESS: Okay, I guess what bothered me in that 5 paragraph was, it says, the regulatory position is a i 6 combination of several items. I think I was expecting there 7 to be comments-on those particular items. That isn't what l 8 you meant; you meant that your position is based on some of 9 the stuff in ANSI, and then in the two NPs? 10 MR. KENNEALLY: Yes. 11 MR. SIESS: But actually, you used the material in 12 the NPs, but not the other? 13 MR. KENNEALLY: That's correct. , 14 MR. SIESS: It's all right, I think. Now, NP-6695 15 is referenced here, but is it covered at all? 16 MR. KENNEALLY: I believe, for an example, o'n page 17 6, -- 18 MR. SIESS: Because it's also in 18. 19 MR. KENNEALLY: Yes, that's correct. On page 6, 20 starting on line 28, we're describing those paragraphs
. 21 within 6695 as being acceptable.
22 MR. SIESS: Okay, you're right; they're endorsed 23 as being acceptable. 24 MR. KENNEALLY: And there is another section where 25 we do likewise.
- - - - ~ . - .- . - - . - . - . . ..
Ih6 1 MR. SIESS: So you've really got three EPRI 2 documents and two Reg Guides and one of th,tm is part of one 3 of them, and one of them covers parts of the other two and, 4 believe it or not, both of them cover one item that's the 5 same? 6 MR. KENNEALLY: That's right. 7 MR. SIESS: That's why I'm not sure. Table 5.1 is . 8 endorsed by both? 9 MR. KENNEALLY: Yes. One reason for that is, on 10 the selection of equipment that would be, routinely checked 11 during normal plant walkdown, you have the status of what's 12 going on. That's endorsed in this guide, that procedure of 13 knowing what your plant is like. 14 That's also the table one would look to do a ! 15 better evaluation after the fact you've had your earthquake, 16 in trying to bring it back on line. 17 MR. SIESS: Okay, now, for some reason that is not 18 at all clear from what I've read, you revised some ; 19 definitions, and in B, you say you revised them, and you , 20 almost give a reason. At the top of page 4, the felt 21 earthquake is revised. You don't say why. - 22 The next paragraph where you delete that sentence I 23 that they should consider the need for power in deciding 24 whether to shut down, you've deleted that and you've said - 25 why. So that's all right.
i i 177 1 Do you think you really need, in a reg guide, to 2 repeat those OBE definitions? 3 MR. KENNEALLY: We have done it because the 4 philosophy has changed so much over the OBE in the past. We 5 have a particular instance where they may have elected a 6 value or they may have used the -- o 7 MR. SIESS: The philosophy is still there. The 8 first paragraph says what the OBE is. That hasn't changed. 9 MR. KENNEALLY: No. 10 MR. SIESS: The second one is simply give them two 11 ways to set it, which is a difference in procedure. It's 12 not explained, the justification for saying take a third and 13 you're home-free, don't even have to check it, right? 14 Nowhere is that explained, and this is not 15 necessarily the place to do it. What's that in? Appendix 16 S? 17 MR. KENNEALLY: Yes. 18 MR. SIESS: I like what you have done, but I'm 19 just wondering, you've changed it, but it's just a flat-out 20 change. Anybody is going to remember now I can take a third 21 or two-tenths. 22 MR. KENNEALLY: That's a good point. 23 MR. SIESS: The definition is still good. 24 MR. KENNEALLY: Yes. I 25 MR. SIESS: Turn to page five, " Definitions are l l l
i 9. 178 ' 1 contained in Appendix A." I don't know why. You had them 2- in the text on the previous one. You had a whole heading 3 called " Definitions." 4 MR. KENNEALLY: Right. 5 MR. SIESS: The thing is that, if I look ahead . 6 just a little bit, to page seven, which is under the 7 " Regulatory Position," and I see a definition of " felt . 8 earthquake" and "OBE" and "SSE" and they're also in tha list 9 of definitions in the appendix. 10 MR. KENNEALLY: I agree with the repetition. One l 11 reason why it's in this particular place is we're endorsing 12 that document. Now maybe it could be just done with -- l-l 13 replace those definitions with the ones we have used. 14 MR. SIESS: I think that your regulatory position 15 that says that we want to change the definitions, I'll argue i 16 with that later, but I think that's appropriate to be in the 17 position, but if they are in the body of this thing, they 18 don't have to be in the Appendix. 19 MR. KENNEALLY: Oh, okay. Fine. . 20 MR. SIESS: I look them up in the appendix because l 21 I don't know what they mean. I'll come back to that. - l ! 22 Then I finally got around to finding what to do if 23 the instruments aren't working. i 24 Page six, 3.2, needs some editing. "A file of all 25 seismic instrumentation should be kept." You don't really
/
179 1 mean to-keep the instruments in a file. 2 MR. KENNEALLY: No. I 3 MR. SIESS: Item 2, " Building and equipment plan 4 views and vertical. sections'. . ., " you don't really mean j 5 that, either. i 6 You mean plan views and vertical sections' showing- l J
, 7 the location of each instrument and the orientation should )
i 8 be in there, but the first sentence, you know, covers a ! 9 couple of vaults full of documents. r 10 Then there's "A complete service history of each 11 seismic instrument should be kept at the plant." Since the 12 heading of the section is "should be kept at the plant," 13 that's sort of redundant. i 14 Now, 3.3 is an endorsement. It says 5.3.1 and 15 5.3.21 are acceptable. Right? f 16 MR. KENNEALLY: Right. ! 17 MR. SIESS: On the next page, you have, again, an. , t 18 endorsement subject to an addition of some definitions, 19 right? 20 MR. KENNEALLY: Yes.
- 21 MR. SIESS: And you did the same thing in the 22 other one.
23 MR. KENNEALLY: Yes. l 24 MR. SIESS: And what I found awkward here -- and f l 25 again, this is editorial -- I think you'd be better off l i f k i
i i 180 i I 1 saying that sections 4.3.1, 4.3.2, etcetera, etcetera, are { 2 acceptable and then take up the definitions section, saying 3 that's acceptable with the following changes. 4 You listed everything that's acceptable, subject ' i 5 to the following definition changes, and the reason you've - ; 6 subdivided it, the first batch are for the evaluation requirements, and the second batch are for the -- they're b 7 l . 8 the same thing. l 9 The first paragraph says that this material -- the ! l l 10- selection of equipment and structures is acceptable. You 11 could say the guidelines for immediate earthquake actions _in 12 this and this would be acceptable. You can say the 13 definitions are acceptable subject to the following change. 14 MR. KENNEALLY: Okay. ; 15 MR. SIESS: Because all you change is the 16 definitions, j 17 MR. KENNEALLY: Correct. l 18 MR. SIESS:- Which gets me down to something with a l 19 little meat to it. It didn't look to me like the changes in , 1 l 20 the definitions were that substantive. Am I missing 21 something? - 22 MR. KENNEALLY: On the first one, on page seven, ; 23 4.1, felt earthquake, the key operator there was the "or" l l 24 between number 1 and number 2.
- 25 MR. SIESS
- Yes.
I {
I i i ! 1 - 181 , 1 Now, where do I find their definition? It's in i 2 here. :
- . 3 MR. KENNEALLY
- Yes. It would be on XIII.
1 4 MR. SIESS: They say, (a) " ground motions felt" l 4 5 and so forth and so forth, and (b) "for plants with operable 6 seismic instrumentation, the seismic switches are
. 7 activated," and you don't want to give them that -- for plants with operable seismic instrumentation. ;
8 9 MR. KENNEALLY: It's basically the "and" is tae -- l
-10 substituting an "or."
11 MR. SIESS: It is an "and" but only for that case. 12 It's a limited "and." ; 13 MR. KENNEALLY: Yes. 14 MR. SIESS: And you did'a few other minor changes, 15 but that's really not -- what's the case where yours would - 16 apply and theirs wouldn't? 17 MR. KENNEALLY: Mine would be a case where 18 operators may have thought there was an earthquake. The a 19 instrumentation wasn't activated. 20 MR. SIESS: Let's take the case the 21 instrumentation isn't there, it's not operating. 22 MR. KENNEALLY: Yes. ) 23 MR. SIESS: And the operators didn't think they 24 felt it. 25 MR. KENNEALLY: If the instrumentation'didn't I
- j
r l I- ;
^
l. i 182 1- operate.and the operators felt that they didn't --
- l. 2 MR. SIESS: Didn't feel one.
l i 3 MR. KENNEALLY: -- have one, there is no action. ! 6 ! 4 MR. SIESS: Under? ( l 5 MR. KENNEALLY: Under either one, I believe. 6 MR. SIESS: Under either one. l 7 MR. KENNEALLY: Theirs or mine. , 8 MR. SIESS: Now, the instrumentation is op -- is l 9 not operative. The first one was it was operative. , 10 MR. KENNEALLY: Okay. , i 11 MR. SIESS: Right? I'm trying to draw a tree. 12 MR. KENNEALLY: Yes. I ! 13 MR. SIESS: What was the first one I said? The l . l 14 instrumentation is operative? 15 MR. KENNEALLY: Uh-huh. i 16 MR. SIESS: And.the operators feel it. l l 17 MR. KENNEALLY: Uh-huh. 18 MR. SIESs: But the instrumentation doesn't l 19 trigger. ,. l MR. KENNEALLY: Right. 20 21 MR. SIESS: Under your definition, would that . 22- require a shutdown? 23 MR. KENNEALLY: Not a shutdown, to initiate the ! 24 proceedings. 25 MR. SIESS: Okay. Would that require an action? (
i l l 183 , 1 MR. KENNEALLY: Yes. ! 2 MR. SIESS: Under their definition, that-would not ! 3 require an action, either? 4 MR. KENNEALLY: Right. , 5 MR. SIESS: Sol if the instrumentation -- now, the 6 instrumentation is active. An operator feels it, but the j 7 instrumentation doesn't trigger. 8 MR. KENNEALLY: Uh-huh. 9 MR. SIESS: What does that require? 10 MR. KENNEALLY: According to ours, they'll , 11 initiate the procedure. According to theirs, I don't 12 believe they'll do anything. I 13 MR. SIESS: The operator feels it is an "or," so 14 that would initiate. ' i 15 MR. KENNEALLY: Uh-huh. : 16 MR. SIESS: In that case, it is applicable 17 instrumentation. If it is not activated, it would not. 18 'Okay. So that's the difference. i 19 MR. KENNEALLY: Yes, sir.
- l 20 MR. CHOKSHI: They clearly say as. defined by them, l . 21 "If it doesn't happen, we don't have an action."
l 22 MR. SIESS: Yes. Okay. There is a difference? i l 23 MR. CHOKSHI: Yes. 24 MR. SIESS: Now, on the CBE definition, if I can . 25 find that one --
l o 184 1 1 MR. KENNEALLY: That's repeated because within the l 2 context of the revision of Appendix S, we had changed the I 3 definition to OBE. ! 4 MR. SIESS: Yes,.but how did you change it from 5 theirs? If they had an earthquake that could reasonable be 6 expected to occur at the plant site, and you leave it at L 7 that. . i 8 MR. KENNEALLY: i That's the difference. 9 MR. SIESS: But it wasn't limited it to that. It 10 is the earthquake that produces the vibratory ground motion' 11 for which those features at a nuclear plant necessary for 12 continued operation of the design to remain functioning. So l 13 if they take out the first sentence -- in other words, no l 14 matter whether it could be reasonably expected or not -- 15 MR. CHOKSHI: They are operating under the old 16 definition, the Appendix A definition. 17' MR. SIESS: Theirs is a definition; yours isn't. 18 Your definition properly stated it would be the operating 19 basis earthquake is that which produces a ground motion for .. , 20 which those features -- l 21 MR. CHOKSHI: Right. - 22 MR. SIESS: But it doesn't tell you how you 23 selected it. 24 MR. CHOKSHI: No. That's right. 25 MR. SIESS: That's what you do with it. m , _. _. _ _ . . . .-. _ .. _,. _ ,._ . -,_ -4 . . _ . .
185 1 Now, a real definition of the OBE is the one you 2 had back here before. It's either one-third of the safe 3 shutdown earthquake or two-tenths g or.something else. 4 You know, it's just that they have a definition; 5 you don't. 6 MR. KENNEALLY: And both of them are extracted , 7- from what the appendix currently -- either the revised one 8 in ours or.the existing one for theirs. We will clear it up 9 all the way through. 10 MR. SIESS: Yes. Your revised one defines an OBE 11 as something you select and givee you some guidelines. 12 MR. KENNEALLY: Uh-huh. 13 MR. SIESS: You can actually select it to be 14 anything you can work with, can't you? 15 MR. CHOKSHI: Right.
'16 MR. SIESS: If you take it at least a third of the 17 SSE,.you don't have to do anything.
18 MR. CHOKSHI: Right. 19 MR. SIESS: I assume somebody did some analyses 20 that showed that with present Section 3 and ACI and ASCE
- 21 margins at a third of the SSE, it won't govern. Is that 22 right?
23 MR. CHOKSHI: Yes, there was a study of generally, 24 what is the effect of taking away this. 25 MR. SIESS: Yes. And then it says -- let's see --
+
1 l 186 l 1 a third of the SSE, alid if you don't take a third of the 1 2 SSE, what are your options? Greater than a third, then you ! J 3 need an analysis -- 4 MR. CHOKSHI: Right. 5 MR. SIESS: -- to show that no functional -- . 6 MR. CHOKSHI: Right. ' lou basically do the current 7 procedure. . 8 MR. SIESS: Yes. I can't find in here where it 9 says two-tenths. 10 MR. CHOKSHI: Oh, that goes to,the shutdown. 11 MR. SIESS: Okay. That's right. Okay. Do you 12 think anybody will ever teko one greater than a third? 13 MR. CHOKSHI: Maybe on the West Coast, maybe. j 14 MR. SIESS: Pardon? 15 MR. CHOKSHI: On the West Coast. I 16 MR. SIESS: Diablo stated two-tenths. 17 MR. CHOKSHI: They stated two-tenths, right. 18 MR. SIESS: But that's mainly because they had 19 analyzed the whole thing for two-tenths. If they started , 20 from scratch, they wouldn't have had any problem taking .25 21 or one-third of the SSE. - 22 MR. CHOKSHI: Oh, yes. Right. 23 MR. SIESS: That was their design and they had 24 been through all those calculations, and wanted to stick 25 with it.
1 7 1 MR. CHOKSHI: I doubt seriously'anybody would go 2 'beyond.that. 3 MR. SIESS: You said the fact that Table.5.1 is
-4 covered in both of these was deliberate.
5 MR. KENNEALLY: Yes. o 6 MR. SIESS: Okay. 5-1. 7 Now, I.have another safe shutdown earthquake. How 8 did that get changed? 9 MR. KENNEALLY: Ground motion. The title had 10 changed. 11 MR. SIESS: It is expressed in terms of ground 12 motion and it defines seismic category 1 structures again. . 13 MR. KENNEALLY: Uh-huh. 14 MR. SIESS: That's, I think, five we counted. . I 15 don't know the way-out, but it's just getting -- it bothers 16 se to have something defined so many places because 17 somewhere somebody is going to skip. One of them is not 18 going to agree. I mean, it's something we can call it. 19 MR. KENNEALLY: Well, I think the solution on part 20 of that is if we did simplify that definition, then we would
. 21 . add a second one or reference something else, and we would 22 go through all of these documents repeating that as well.
23 So it would be repetitive. 24 MR. .SIESS: You just' define a seismic Category 1 25 structure.
. _ .~ _ .. . . _ . _ _ _ _ _ . . _ ._. _ _... . . . _ _ _ _ ___ _ _ _____ _
i l~
- i 188 l 1 MR. CHOKSHI: Structures, systems and components.
I 2 .MR..SIESS: Incidentally, the first definition, 3 cumulative absolute velocity, the time interval of the 4 absolute acceleration time history over a prescribed 5 duration of the earthquake record, which I guess is light, , 6 if you want to call it that, but you've got a much better 7 definition of it in the text. You tell them how to compute . , l 8 it. So why does it need to be defined in the definitions? ! 9 By the time I get finished reading the text, I 10 know exactly whatfit is and I'm not even, going to look for 11 it in the definitions. So that's what I mean. Maybe an j 12' editor will catch those things. 13 I skipped a couple of things, I think. Let's see. 14 Appendix B -- oh, just a minute. Page 10. Maybe I don't i 15 understand the cumulative absolute velocity check. CAV , 16 check'is exceeded if any one of the three components -- two 17 horizontal.and one vertical -- do they really have to be 18 horizontal and vertical?
- 19 If I'*ve got three components, don't they describe ,
20 the effect no matter which direction it's in? I mean, it's 21 three-dimensional. Once I have three orthogonal components, - 22 I don't have to describe everything, do I? I mean, this is 23 conventional; I'm not objecting.
.z 24 MR. CHOKSHI: Yes.
25 MR. SIESS: Any is larger than .16g seconds. Why l l- {
. , _ ~ . . . , _ = _ _ _ . . ._ _ _ . . _ _ . _ . . _ .
189 1 isn't this dependent on the level of the OBE or SSE? It's 2 an absolute'.16gs That's true for all sites that we've had? 3 MR. CHOKSHI: This is the integration of the time 4 history, the absolute velocity. 5 MR. SIESS: Yes. 6 MR. CHOKSHI: So that number is -- pressure is . 7 constant. But for a different time history, the number will 8 be different. 9 MR. SIESS: Okay. But this is supposed to be a 10 measure of the damage potential -- 11 MR. CHOKSHI: Right. 12 MR. SIESS: -- no matter what the plant is 13 designed for? You mean .16g seconds is going to be a same 14 damage potential for a plant defined for .15g as for one 15 designed for .16g? 16 MR. CHOKSHI:- This measure is derived from l'ooking 17 at the engineer building and conventional cord, nothing to 18 do with a nuclear power plant. The damage to those ,- 19 buildings was used as a surrogate for this criteria. So 20 this relates to the MMI seven, eight level of damage. 21 The actually records -- you know, they went to the 22 historical records of the damage that occurred to the 23 engineer structures and set the criteria. For nuclear power 24 plants, this is a threshold. You would not expect any 25 damage at these levels.
. . - . . . - . - - . . . . ~ - - . . . - - . . - - . . - ._. -. -
l l 190 1 MR. SIESS: I sure wouldn't if this is the 2 threshold for conventional -- 3 MR. CHOKSHI: Yes. 4 MR. SIESS: Okay. If you are below this, you 5 forget about it? 6 MR. KENNEALLY: Yt s. 1 i 7 MR. SIESS: So it is just a good safe lower bound? . 8 MR. CHOKSHI: Yes. 9 MR. KENNEALLY: Yes. 10 MR. SIESS: Okay. 11 The next page is the OBE exceedance or damage. 12 Where is damage defined? 13 MR. KENNEALLY: I believe it is in the EPRI j 14 report. i 15 MR. SIESS: I do not think it's in the list of 16 definitions. I 17 MR. KENNEALLY: It is out of the context of the 18 EPRI. document. It describes that they have walkdowns, and 19 if they see changes in state. . 20 MR. SIESS: Is that in one of the things that is 21 endorsed? - 22 MR. KENNEALLY: It's endorsed. 23 MR. SIESS: Okay. 24 MR. KENNEALLY: Yes. A change in state over what 25 they --
l l 191 i 1 MR. SIESS: I will have to admit I did not get l i 2 through all of the EPRI documents. In your list of 3 references, NP6695 I think belongs in there. ,
-4 MR. KENNEALLY: Yes. l l
5 MR. SIESS: In Appendix B, you need some editing. ] = ; 6 I guess that's all. i . 7 MR. KENNEALLY: Okay. 8 MR. SIESS: The first sentence, the plants at. I l 9 which only instrumentally determined foundation level data l l 10 are available, I think that means for pl. ants at which l 11 instrumentally determined data are available only at the 12 foundation level. 13 MR. KENNEALLY: Uh-huh. 14 MR. SIESS: And there's a footnote that talks ] i 15 about higher quality damage or lack of damage. reports, and I 16 think the quality refers to the reports and not to the~ f 17 damage, the higher quality reports of damage or lack of 18 damage. 19 MR. KENNEALLY: Oh, yes. I see. 20 MR. SIESS: I'd like to say I think I know what it l l 21 means. That's on the footnote on Page 16. 22 MR. KENNEALLY: Right. , 23 MR. SIESS: Again, I mention them, and you see l 24 that the editor catches them. You used to have a good l l 25 editor over there. l l 6 l l l
. - -. .- . _. . . . - - _ _ . - _ . . - - - . _ . . -. -. . _ ~ . - . . I e i 192 l 1 Now, that's essentially the kind of questions I've 2 got. These, I think, are straightforward. S comes up . 3 pretty clean. I have no problem with the OBE/SSE stuff. I 4 think it would be nice to see how you arrived at the one- ! l 5 third. 6 MR. KENNEALLY: Yes. 7 MR. SIESS: But it looks reasonable. If you want . l 8 to talk about that, you know, I'd be glad to hear about it. 9 Appendix B. The big hang-up is essentially what 10 is in DG-1015 right now. , 11 MR. KENNEALLY: Right. 12 MR. SIESS: But for these things, my 13 recommendation to the subcommittee and, if they agree, to 14 the full committee -- they're all going in for public 15 comment. Get them out, but I don't think you can get them 16 out without -- 17 MR. KENNEALLY: Right. 18 MR. SIESS: -- 15, and I would like to recommend 19 those go out because I don't see any major safety issues. , 20 They tend to be procedural issues. The OBE/SSE thing could 21 be a potential se*ety issue depending on how well it's been - 22 thought out. Bu ,ast off hand, I feel that if we lose some 23 of the cases where OBE governed, it is not going to hurt un 24 that much because I've always been bothered by that, as you 25 know. e
- y. . . _ _ _ . _ . _ , . - - . _ . _ _ . . . _ _ _ _ . _ . _ _ _ ~ _ _ _ _ _ . _ . .
4 l~ , l 193 1 The. basic philosophical question of a dual ! 2 approach to depicting the ground motion -- I.think you've ! i 3 got good arguments for doing it both ways. I think to say, i
<4 why don't you do it all probabilistic, you gave me a good 5 answer. If you do, you're stuck with EPRI and Livermore, l 6 two approaches which will give two different answers and you
- . 7' just don't know which one is best. i t
8 MR. MURPHY: Right. l 9 MR. SIESS: You've developed a scheme that will j 10 let you make use of some of the insights.of the l 11 probabilistic without having to use the damn numbers,
- 12 because by setting them both to the 50 percentile, they l 4
13 should come out reasonably the same. . l c l l 14 To go back completely to the deterministic l 15 approach, I don't hear as good an argument because the l 16 arguments that you lose something from the probabilist'ic l 17 approach is true, but I'm not sure that what you lose is all l l ~ 18 that important. I mean, I look at the evidence of what
. 19 we've accomplished for 180 sites or so, without the 20 p:vbabilistic approach and stumbling along and changing as j
!- 21 we went, ans learning as we went, we still came out with 22 some fairly reasonable ground motions for design. 23- There's not any plants out there, I think, that < 24 are miles off. I can go back and look at Big Rock Point i
- 25 that didn't have any seismic design. When you looked at i
f
-ny w .+-- - -- - - -
194 1 that thing with'an SEP and only one pipe was straight 1in-the 2 whole plant, it still didn't come out too bad.. But it's 3 only 67 megawatts, but there's nothing around it but trees. j 4 Again, you just look at that'and -- we made bigger l l 5 mistakes with Diablo Canyon, but there's nothing in what , 6 we've got that would have found the Hosgri before the Hosgri 7 got found. And there was a lot of work done on Diablo to . 8 find out if there was anything bad out there. They moved as 9 far away from San Andreas as they could, you know, that type !
- 10. of thing. l 11 So, I look at the arguments for not doing that, 12 and I don't find them terribly convincing, so far, that I'm i
[ 13 giving you too much chance to argue back. I tend to get the 14 argument, gee, we've learned a lot from the probabilistic 15 stuff, so let's use it. l
?
16- But I do think it's worth thinking about. Do we ! 17 really gain something in safety, in confidence, maybe, both l i 18 in the reasonable assurance and in the. undue risk part of l 19 the thing, from having to -- and if the answer is we do, and , ! t 20 it isn't that much work, but some of the things that we . 21 don't think are a lot of work, and up being awfully messy - 22 before we get through with hearings.
.23 So, as far as I'm concerned, and I think as far as 24 ACRS might be concerned, that's really the only place we've 25 got an issue. I think we need a little more convincing
I, 1 / i [ 195 l L 1 arguments on the OBE, and we'd like to hear from the 2 industryfon the OBE. We'd like to hear from the industry on 3 this other stuff, on the dual system, and see what people { 4 think..
- 5. I don't think it's going to add a lot to anybody's 6 cost. I'm more concerned as to how many years it adds to f
+
. 7 the licensing process. Frankly, I think you've done a 8- pretty straightforward job of getting this thing cleaned up ;
9 and separated out along with this separation of siting from i 10 plant design. It's done fairly neatly,,and I've got no 11 problems with cleaning up the geology stuff. 12 I knew what a tectonic province was. If I knew 13 what a capable fault was. Now I've got to learn some new 14 definitions, but so what? ! 15 MR. MURPHY: Right. 16 MR. SIESS: Now, what does that leave us to get l 17 into in a little more detail? Do you want to get back into l 18 the stuff of the application of the new approach and the ; i
, 19 OBESSE? Is that all we have left? Unless you've got I 20 problems, I say we've covered H, Appendix S, right?
21 MR. MURPHY: Right. . t l 22 MR. SIESS: And that's got the OBESSE in it, but 23 then the details of that, who's got that? The Engineering j 24 Regulatory Guides, that's the ones we've looked at. Ground l 2 i 25 motion, Phyllis has talked about ground motion and Dick ' i
l lh6 l 1 hasn't talked about geologic. What do we have on those to l 2' cover, on the geologic. investigation? That was the scope, I 3 the extent, how far out to go? 4 MR. MURPHY: That was some of it. Dick was also ! 5 going to mention some of the definitions that we have a L 6 changed, and which ones we've changed and how.we've changed j t 7 them, some that we have deleted, but part of it was simply - 8 an. enumeration of what we're still requiring by way of the 9 geological and geophysical investigations. 10 MR. SIESS: I think we' ought ,to get something on l 11 that. If you want to say nothing's changed except L L 12 definitions, but I think we ought to have some basis for l l 13 understanding the different terms, maybe, and why they're 14 there. I haven't been following that and reading all the 15 public meetings and groups where people came in and griped i 16 about them. 17 What does Phyllis have? i l l 18 MR. MURPHY: Phyllis had basically the markup for 19 revisions on the Standard Review Plan 2.5.2. . g 20 MR. SIESS: Oh, I didn't even look at that. I l 21 figure if the regulations are there, you guys can write your - 22 own Standard Review Plan. 23 MR. CHOKSI: It describes the deterministic ground i 24 motion procedures. 25 MR. SIESS: I mean, if we understand it from a i
.~. _. - - .-
197 ! i 1 what's in the Reg Guides, do we really need to look at the : 2 Standard Review' Plan? I 3 MR. CHOKSI: I don't think it has changed in the 4 way we calculate ground motions. : 5 MR. MURPHY: Basically, the changes that are in. , i 6 the Standard Review Plan section are those that principally f , .7 are associated with putting in the new definitions of OBE 8 and so forth. There's not been -- : 1 9 MR. SIESS: I mean, there's nothing original in 10 the SRP changes; they're just changes to conform. 11 MR. MURPHY: That's it. 12 MS. SOBEL: And the markup you have was done in 13 September, which was before some of these new terms had ' i 14 developed, so the markup you have is really a very 15 preliminary one. ! 16 MR. SIESS: Let's. skip that for the time being. ; 17 Phyllis is always getting cut out. 18 MR. MURPHY: She didn't object. 19 MS. SOBEL: I might go draw your attention to one 20 thing, and that's in the Standard Review Plan on page 10, I - 21 think it is. 22 MR. SIESS: Let me find that. 23 MS. SOBEL: The last time the Standard Review Plan 24 was revised -- that revision became effective in 1989 -- we 25 did have some probabilistic statements in that.
198 1 MR. SIESS: We've got our handout. 2 (Slide.] 3 MS. SOBEL: In the handout, I have quoted from the l 4 Standard Review Plant. 5 In 1989, the revision of the Standard Review Plan , j 6 was designed to reflect current staff practice during the 7 1980s, and as part of current staff practice, we were using , ; l 8 probabilistic seismic hazard results, and this paragraph was i 9 added to the Standard Review Plan to reflect the way we were i 10 using them at that time. 11 MR. SIESS: What year was this? 12 MS. SOBEL: The year it became effective'was 1989. 13 It was part of the resolution of unresolved safety issue A-14 40. 15' MR. SIESS: And we haven't had a plan since then. 16 MS. SOBEL: No. p 17 MR. SIESS: So, it's never been used. , 18 MR. CHOKSHI: No. This was used in the '80s. 19 MS. SOBEL: It reflects what was used in the ,. f 20 1980s. 21 MR. SIESS: Oh, okay. Yes. People were bringing . 22 in that sort of stuff. 23 MS. SOBEL: We were using the probability of 24 exceeding the SSE during operating license reviews and l 25 comparing it to the probability of exceeding the SSE at 4 I
- 199 1 other sites at that time.
2 MR. SIESS: So, it doesn't represent a change from 3 practice. 4 MS. SOBEL: Some of the concepts you have been 5 discussing today, such as the probabilistic controlling 6 earthquake, that is a change from the way we were discussing , 7 things then. , 8 MR. SIESS: Basically, what you've got in the 9 proposed Standard Review Plan is the implementation of the 10 changes that we're making in the reg guides and the 11 regulations -- 12 MS. SOBEL: Right. That's our main intention, 13 NR. SIESS: -- like references and stuff. 14 MS. SOBEL: Yes. 15 MR. SIESS: That's the trouble with the Standard 16 Review Plan. It tends to narrow your options, but the good 17 thing is that people ignore it. They can't hold you to a l 18 Standard Review Plan anymore than you can hold somebody to a , 19 reg guide unless he is committed to it. j j 20 Who is getting ready to go up there? j - 21 MR. MURPHY: I guess the question would be what 22 would you like to hear next? 23 MR. SIESS: Whatever we've got the best chance of 24 finishing up, because we can have another meeting. J 25 MR. MURPHY: Right now, it seems our problems are
m p.
~
200 1- principally related with 1015. Why don't we hold that and 2 let Monica talk about the OBE, SSE investigations that she
=3 is taking care of, and then maybe we can finish up totally 4 the engineering aspects here and then just have --
5 MR. SIESS: Then we can have some geology and some 6 explanations of the new terms. 7 MR. MURPHY: Right. If we've got time after 3 8 Monica is done, we can go on to the geology. If not, we'll 9 pick that up next time. 10 MR. MICHELSON: Do you want to,take a break? 11 MR. SIESS: Do you want to break now? Let's take 12 a break. 13 [ Recess.) 14 MR. SIESS: Let's reconvene. 15 [ Slide.) 16 MS. WITTE: I am Monika Witte, and I'm from 17 Lawrence Livermore National Laboratory. And I'm going to 18 talk'about the OBE SSE investigations that are ongoing. 19 MR. SIESS: Excuse me, Monika. -Her handouts are , 20 about six page down the list -- combined list we have here, 21 if everybody finds them. I've got it. Fine. Go ahead. . 22 [ Slide.) 23 MS. WITTE: First, I'm going to talk about some of 24 the options for Appendix S, a little bit about the new 25 regulation in Germany, the risk study that's ongoing, a
-. - - . . . ....- - .-.~ . - .. . . - - . _ - . . - - . . . . . - . _ . - . . - . - . - . _ . - -
! l i I i .- j 201 I l j 1 survey of utilities and designers that we undertook, a 5 2 review of codes and standards and work that's in progress. I 1 \ , I i 3 [ Slide.] j
'4 MS, WITTE: A number of options have been )
5 considered for changes to the engineering aspect of Part 1, , l 6 Appendix A. Previous suggestions regarding the OBE have i
- c. 7 included: Decoupling the OBE and SSE; setting the OBE based 8 upon a reasonable return period; another one has been to 9 leave the OBE and SSE coupled, as the current regulation j 10 requires; and to modify the load factors used in the' load 11 combinations, equations and modify the damping requirements I
12 so that the design process would be streamlined and the OBE 13 would no longer govern. And the third one, which is the one 14 that's pretty much been selected is to set the OBE shutdown 1 15 for inspection level at some factor of the SSE and to leave 16 the option to the utility to design for the OBE or.to assume 17 the risk of an inspection shutdown at some lower-level, such 18 as one-third.
, 19 (Slide.)
20 MS. WITTE: So, just to give you some background, 21 I talked to Helmut Schultz, from the German Regulatory Body 22 and the information we received is that Germany has recently 23 revised their seismic regulations. The seismic regulations 24 are provided in KTA Safety Standard 2201, which was first 25 issued in 1975 and not revised until recently. I
.. . . . . - - - . - - , ~_ . - . ~.- . - -- -.. .- -.-- - . -
1
.. i 202 1 The standard originally was based upon 10 CFR 100, l 2 Appendix A. And the new 2201 requires only one design basis 3 earthquake, the so-called Bemessungserdbeben, which is 4 analogous to our SSE. And they have an inspection-level 5 earthquake for shutdown which has been set at .4 SSE. They ,
6 set the level so that the corresponding vibration would not 1 7 induce stresses exceeding the allowable stress limits , 8 originally required for the OBE. They have done a series of 9 studies showing that .4 is an appropriate level to produce 10 this result. 11 [ Slide.) 12 MS. WITTE: Ravi Ravindra and Wen Tong of EQE in 13 Costa Mesa, are performing a risk study, under contract to 14 LLNL. And the first step of this study included the 15 following: An identification of components, which might be 16 affected by change in the OBE design criteria, based on a 17 survey of the industry; and secondly, an identification of 18 generic categories of components which are potential seismic 19 risk contributors. . ( 20 MR. MICHELSON: Is that contributors at the OBE 21 level? - I 1 22 MS. WITTE: No. Just seismic risk contributors in 23 general.
- 24 MR. MICHELSON
- Okay. -
{ d 4 25 [ Slide.] i 1
203 1 MS. WITTE: So, I have a list of components that 2 fall into each of those categories. 3 The first list include the components which might i 4 be affected by a change in the OBE design criteria, based on ; 5 a survey of AEs and designers. . And this list consists'of t 6 the following components: Concrete frames, including walls,
, 7 floors and roofs in the aux building and in the fuel-8' handling building; piping and pipe support systems -- [
9 MR. MICHELSON: Excuse me. Why doesn't the SSE 10 bound any building construction and so forth, except the 11 non-seismic buildings? 12 MR. SIESS: You might start back with that. Why 13 does the OBE govern? , 14 MS. WITTE: The OBE governs for a variety of 15 reasons. For the concrete, it's because of the 1.9 load . 16 factor which was introduced :Ln 1975, I think. 17 MR. MICHELSON: Okay. ,. 18 MS. WITTE: I need to say that this survey was ! 19 taken -- 20 MR. MICHELSON: Okay. I see.
. 21 MR. SIESS: We had different load factors, 22 depending on the probability, but they weren't the correct ,
23 factors in terms of the probability. . 24 MR. MICHELSON: Yes, okay. Now, in the case of s 25 piping and support systems -- ,
204 l' MR. SIESS: Same thing. Different damping? 2 MS. WITTE: Right. Different damping has been l '3 used. l
.)
4 MR. MICHELSON: Yes. .But not different stress j 5 levels. 6 MS. WITTE: I need to back up and say that these 4 71 ' comments are -- 5 l 1 8 MR. SIESS: Different stress levels. l l 9 MR. MICHELSON: In piping? l 10 MR. SIESS: Yes. Piping for OBE would be designed 11 for one stress level or -- 12- MR. MICHELSON: Okay, the different level. You're 13 going to level C on the SSE and Level A or B on the -- 14 MR. SIESS: Yes. Plus difference in damping 15 factors. 16 MR. MICHELSON: Okay. 17 MS. WITTE: Right. I 18 MR. SIESS: It's the same thing as the load 19 factors for concrete. , 20 MR. MICHELSON: Yes. I see. 21 MR. SIESS: Concrete keeps the same allowable, but - 22 raises the load multiplier. 23 MR. MICHELSON: Well,the first thing you know, the 24 OBE controls the design and not the SSE. I see. i I 25 MS. WITTE: Right. f { . . . - -
i . t i { 205 ; ! 1 MR. MICHELSON: Okay. I've got it. i , 2 MS. WITTE: But these are judgments, okay? i i 3 MR. CHOKSHI: Yes. There is one thing to be -- } ! 4 when we say control designs, it's just in terms of. ! l 5 allowable stresses. It does not mean the physical size is ;
; e 6 going to be governed in all cases by OBE.
j 7
, MR. MICHELSON: Yes, because the.OBE infers that i
8 things could continue to operate properly. l 9 MR. SIESS: We always had the -- it was always ! i 10 possible for a designer to use an inelastic analysis for the i ! b j 11 SSE and show no loss of function, but nobody did -- j 1 l 12- MR. CHOKSHI: Right. 1
- 13 MR. SIESS: -- except in checking turbine building j 14 at Diablo. In design they never did it because inelastic 15 analysis of piping, for example, is an extremely expensive
, 16 process. You can't do it more than once if you do it then. 17 The structure is the same way. So, they ended up both being { 18 designed elastically with different damping factors, Y 19 different load combinations and load factors. j 20 MR. MICHELSON: Just getting educated.
. 21 MS. WITTE: Okay. Another series of components 22 here. Spherical steel containments and heads on cylindrical 23 steel containments might be governed by the OBE; 4
24 containment penetration reinforcement; reactor vessel and
; 25 steam generator internal components; large bore component
I a l 206 I 1 support snubbers -- large bore component supports, in 2 general, and embedment plates. 3 [ Slide.] i 4 MS. WITTE: The next list 3ncludes components from I 5 -- which have been shown to be potential seismic risk 6 contributors based on existing seismic PRAs, and this list 7 consists of the following: NSSS supports, control rod drive
- 8 mechanisms, valves, heat exchangers, ground-mounted storage 9 tanks, small indoor tanks, batteries and battery racks, 10 motor control centers, switchgears, transformers, diesel 11 generator system components, pumps, and fans and cooler 12 units.
13 MR. MICHELSON: Now, the arguments I heard just a 14 moment ago, I think I appreciate when you're pushing stress 15 levels, but when it comes to things like motor control 16 centers, how does the OBE -- 17 MR. SIESS: It didn't. These are possible 18 contributors to seismic risk, independent of how they are 19 designed. . 20 MR. MICHELSON: Oh, that's all that's meant. 21 MR. SIESS: Motor control centers are contributors - 22 to seismic risk not because they were designed for the wrong 23 loading but because they weren't supported properly. 24 MR. CHOKSHI: Exactly. 25 (Slide.]
207 1 MS. WITTE: So, the second step of the study was 2 to select a subset of components from lists A and B for 3 careful review, and if, for these components, the designs 4 are governed by the load combination equations, which 5 include the OBE loads, then the study will evaluate the 6 effect on the component capacity for the hypothetical
, 7 condition of no OBE design analysis.
8 MR. SIESS: Your definition of " governed," at the 9 bottom of the page, I think it means what I would mean, and 10 that is it would be governed by the OBE if the component or 11 system or structure designed for the SSE had to be changed 12 because of a calculation by the OBE. 13 MS. WITTE: Well, I'm actually not going to that 14 level. I don't know whether or not the design had to be 15 changed or not. 16 MR. SIESS: If it was right up to the limit, 17 though, it would be, by your definition. 18 MS. WITTE: Right. 19 MR. SIESS: Okay.
- 20 MS. WITTE: Right.
. 21 MR. SIESS: You could have had margin built in, 22 because they selected the next largest size or something.
23 MR. CHOKSHI: In the early days, I think the 24 differences are not that big. I think they probably won't 25 have any impact on the capacities.
/
208 l 1 MR. S,IESS: But if it didn't require a change, it l 2 would require a difference in margin. 3 MS. WITTE: Right. 4 MR. SIESS: You calculated the margins in two 5 ways. 6 MR. CHOKSHI: That's the third step. 7 MR. SIESS: Yes. ~ 8 MS. WITTE: Right. 9 The third step'will be to determine the effect on 10 the seismic risk if the capacities are r, educed, and the 11 fourth step will be to suggest changes to the component 12 design process if there is an increase in the seismic risk. 13 [ Slide.] 14 MS. WITTE: So, from the first step, we have 15 obtained data from P.K. Agrawal at Sargent & Lundy and his 16 staff for 32 components on 13 component types, and the 17 following components have been reviewed: 18 Two containment penetrations, and in these cases, 19 upset load combinations tended to govern these two designs, ,. 20 but it was not clear that this was due to earthquake loads. 21 Thermal expansion loads are a large contributor.
- 22 The SSE loads are not explicitly included in the 23 faulted analysis. Instead, the containment penetration is j 24 subjected to the maximum pipe break load, and that's how the 1
( l 25 faulted loads are generated.
/
i 1 2 9 1 MR. SIESS: Now, let me get something clear. When l l 2 you say an " upset load," that's a load combination or a load 3 multiplier? i l 4 MS. WITTE: A load combination. 5 MR. SIESS: Say ASME code. 6 MS. WITTE: Right. - 7 MR. CHOKSHI: That's for the OBE. 8 MS. WITTE: Right. 9 MR. SIESS: And " faulted" is another different 10 load combination. 11 MS. WITTE: That's right. 12 MR. SIESS: Which is higher? Faulted is higher 13 than upset? 14 MS. WITTE: That's right. And the faulted 15 includes the SSE load, typically, and the upset load 16 combination typically includes the OBE. 17 MR. SIESS: Yes. Okay. 18 MS. WITTE: So, we looked at containment , 19 penetrations, and we looked at piping and pipe supports. 20 For pipitig systems, we looked et four systems. 21 All four systems had been reanalyzed with the PVRC damping 22 during the snubber reduction program, and OBE governed all 23 four cases for the pipe stresses, and this is different than 24 what was predicted when we took the survey of the industry. 25 MR. SIESS: Why?
210 1 MS. WITTE: Because of the particular plants that 2- the systems came from. It just happened, at those plants, 3 that the OBE was proportionally higher at the elevations at 4 which we looked at the piping systems. 5 For the 44 supports involved in those four 6 systems, nine of them were governed by the OBE. 1 7 MR. SIESS: Supports include snubbers? ~ ) 8 l MS. WITTE: I don't remember. l 9 MR. SIESS: That's all right. 10 [ Slide.] 11 MS. WITTE: We looked at four shear walls. The 12 SSE governed all four, but the OBE would have governed one 13 if the load factor had been 1.9, as it presently is'. The i 14 survey predicted that the OBE would have governed all four 15 with a load factor of 1.9. 16 MR. SIESS: What was the load factor? 17 MS. WITTE: I think it was 1.4, actually. These 18 are old plants. But I compared them all as if the load 19 factor had been 1.9, and in only one case would OBE have , 20 governed. l 21 We looked at three support structures, and the SSE - ! 22 governed two out of the three. 23 We looked at two horizontal pumps. The OBE { 24 governed one pump design and governed some aspects of the l 25 other pump design. i
1 211 1 MR. SIESS: What kind of differences were there? 2 OBE governed, say, by 10 percent or by 50 percent? l 3 MS. WITTE: Sometimes it was by a very small 4 margin, and sometimes it was by more than that. You have to 5 look at the individual cases. 6 MR. SIESS: Was there any pattern to it, to what , 7 types of things the OBE -- where the differences were large? 8 MS. WITTE: Well, a general pattern, I think, is 9 that the OBE tends to govern if the earthquake contribution 10 to the total load is small, and the SSE tends to govern if 11 the earthquake contribution is large. , 12 MR. SIESS: Okay. The other contributions being 13 dead load, accident loads, pipe whip. 14 MS. WITTE: Right. 15 MR. MICHELSON: Did you do the pump analysis, or 16 did you review somebody else's? 17 MS. WITTE: We only reviet:sd -- no new work was 18 done. We only reviewed components which had been designed 19 for existing plants. 20 MR. MICHELSON: Now, in looking at a pump - 21 analysis, what did they analyze? Did they analyze the 22 bearing load during the OBE and the pump shaft deflection 23 and interference between the bladings and things of that 24 sort? 25 I've seen seismic analyses that did nothing more
-. . ~ . _ - - _ . . . - . - . - - - _ . - . . - _ - . _ . - - . - -
t 2k2 ~! 1 than look at anchor bolts, you know. i > L 2 MS. WITTE: Yes. 3 MR. MICHELSON: So, I don't know what this is, how. > l ! 4 good -- you know, what' kind of analysis we're looking at l 5 when it said "OBE governed." 6 MS. WITTE: That's a good point. We looked at the j 7 smallest -- the portion of the analysis that had the i 8 smallest margin of safety against the allowable. 9 MR. MICHELSON: Where did they find the pinge ! 10 point to be? What was giving them trouble? 11 MS. WITTE: For the pumps, I think it was almost I i 12 always the anchor bolts. 13 MR. MICHELSON: Maybe that's all they looked at, 14 too. Did they look at pump shaft deflection? Those big, 15 long horizontal pumps get some interesting interferences 16 from a little vibration like that. l 17 MS. WITTE:- Right. We looked at shaft -- not i , 18 deflections but loads, stresses in one case. l 19 MR. MICHELSON: Well, the deflection may be all it , l 20 takes. It may not exceed the allowable stresses, but it may 21 chew up the blading. - t 22 MS. WITTE: Yes. 23 MR. MICHELSON: It depends on the type of i l 24 compressor, the amount of clearance there is, and a number l 25 of factors. ' i l { i
213 1 MS. WITTE: Yes. 2 MR. MICHELSON: Same way on your chilled water 3 compressor on the next page. I haven't seen a decent 4 chilled water analysis yet. I just wondered if all they did 5 was look at anchor bolts. 9 6 MR. SIESS: But the question is would the , 7 differences be other than the difference in the loading? 8 MR. CHOKSHI: Yes. I think, in looking at the 9 existing calculation, there might be things where the way 10 OBE is envelopad, there may be some effect on the way the 11 things were done. You're not going to get a really true 12 picture of whether the OBE is large or is small. 13 MR. SIESS: Are there any differences between the 14 OBE and the SSE other than the loading, spectral 15 differences? 16 MR. CHOKSHI: Damping, because of the dampin'g. 17 MR. SIESS: Well, damping, of course, ends up in 18 affecting the load. 19 MR. CHOKSHI: Right. 20 MR. SIESS: But are there shifts in the spectra?
. 21 MS. WITTE: Yes.
22 MR. CHOKSHI: In a soil structure where the site 23 is sensitive to soil / structure interaction, you might see it 24 because of the foundation stiffness at different levels. 25 MR. MICHELSON: Are these richer in the high end
, a 214 1 or low end at OBE?
2 MS. WITTE: I don't recall. I have the spectra 3 but I just don't remember. 4 MR. MICHELSON: You would think that the SSE is 5 always more severe from the viewpoisit of vibration. But if . 6 you shift the spectrum enough, it might be that you're 7 reaching the right frequencies to give you a -- it depends
- I 8 on the natural frequency of that shaft.
9 MR. CHOKSHI: In general, the structures remain I 10 elastic. So, I would not expect too many changes in the i 11 frequencies, in many cases, because of the scale of the OBE. 12 MR. SIESS: If that's true, then of course it is
]
13 fairly straightforward. 14 MS. WITTE: We did get a shift in frequencies 15 though. I just don't remember which direction it was. 16 MR. CHOKSHI: Only for a particular piece. In 17 general, it is an elastic system. 18 MS. WITTE: Yes. 19 MR. MICHELSON: It is like trying to predict relay , 20 chatter with a calculation. It's kind of hard to do. l 21 MR. CHOKSHI: Yes, it is.
- 22 MR. SIESS: Get rid of them is the only answer I 23 can see to relay chatter.
24 MR. MICHELSON: Yes. 25 MR. SIESS: Otherwise, we'll spend the rest of our l l l I
l l l 215 1- lives trying to figure that out. 2 MR. MICHELSON: I was quite interested i l 3 particularly in the refrigeration unit which is on the next l l 4 page. Because they do involve some fairly tight clearances-5 in.the compressor. I just wondered if anybody has really I 6 ever done a seismic analysis on the compressor and not other.
, 7 than the anchoring of it and so forth? !
o 8 MR. SIESS: They are not a qualification. Does , ! I l 9 the refrigeration unit compressors have to be -- i l 10 MR. MICHFJMGh : Yes, but they don't run it when ' 11 they're shaking it.- That's the problem. 12 1 MR. SIESS: They don't? 13 MR. MICHELSON: Most penple don't run these l 14 components when they're shaking it. It's a huge component, j 15 and it's hard to set up enough to set a complete closed ! l 16 cycle refrigeration system in operating order. So, most of ( 17 them shake it and see if it comes apart. j 18 MR. SIESS: They did them at Surrell. l l 19 MR. MICHELSON: Well, you can do it. It's a very 20 large set up. It's about like trying to do a pump underflow
]
,- 21 while you're shaking it.
22 MR. SIESS:- A pump the size of this table. 23 MR. MICHELSON: The refrigeration unit is bigger j j 24 than this table, Chet. . We're talking about very large i ! 25 compressors, several hundred tons of refrigeration. This is i l f l I l
r 216 1 not the kind you stick in your home air conditioner by any f 2 means. These are enormous units. They're not quite as big i 3 as this table,,about two-thirds of.the width of this table 4 and.about as long. They have big heat exchangers on them, 5 the whole bit.
, 3 6- MR. SIESS: That stuff can be -- they tested stuff ;
! 7 like that at Surrell. . 2
- 8 MR. MICHELSON
- Well, it's about like shaking a
, l 1 9 diesel engine. You can do it, if you've got yourself a l 10 shake table somewhere big enough. ) l 11 MR. CHOKSHI: The Japanese have a main diesels on j 12 the shake table. I 13 MR. MICHELSON: They have a shake table big enough j I 14 for their whole diesel? 15 MR. CHOKSHI: Yes. 16 [ Slide.] 17 MS. WITTE: The component that had the smallest 18 margin of safety that we-looked at was the refrigeration 19 unit. And the portion of it that was governed by the -- or , 20" that had the smallest margin of safety were the anchor 21 bolts. So, those are the. sections of it that we looked at
- l 22 most carefully.
23 MR. MICHELSON: Did somebody check though to see I
- . 24 if they did the bearing loadings under seismic and that sort h 1 25 of thing? Stress on the bearings? Shaft deflection is one l
l
l l I 217 l l 1 problem, and bearing stresses are another problem. I 2 MS. WITTE: Yes. I have copies of the 3 calculations. I don't remember whether they did the bearing 4 calculations. j i ! ! 5 MR. MICHELSON: The ones that I have seen I 6 generally don't.
. 7 MS. WITTE: So, it's worth looking.
8 MR. MICHELSON: I'd just look at the anchor bolts, 9 because that's easy. 10 MS. WITTE: Okay. , i 11 MR. MICHELSON: Yes. 12 (Slide.] 13 MS. WITTE: In addition to the risk study, we 14 performed a survey of utilities and designers, and this is 15 actually where that risk -- that list A came from. And I 16 contacted staff people from a wide variety of sources -- 17 from utilities, consulting firms, AEs, code organizations, l 18 in order to get-feedback on the feasibility of eliminating 19 the OBE from design. The input was qualitative and the 20 input was occasionally contradictory.
. 21 And, as I pointed out earlier, it didn't always 22 agree with the data that we collected from Sargant & Lundy. i 23 MR. SIESS: Are there real differences or just 24 differences in people?
25 MS. WITTE: Well, some people said we should l
i l i t k I I 218 l 1 eliminate the OBE from design, and some people said we 2 should not. So, there were some definite black and white. 3 MR. SIESS: Did they give reasons? 4 MS. WITTE: Yes. l l 5 MR. SIESS: Some of them just didn't want to , 6 change? i . 7 MS. WITTE: Some people don't want to change, some ,
- 8 people believe that the OBE contributes to safety, and 9 others didn't. There were some diametrically opposed views.
4 i 10 4 : 1 11 [ Slide.) 12 MS. WITTE: We tried to obtain input on structural j i 13 and mechanical components: Concrete structures, steel
)
14 structures, piping, embedment plates, components supports, 15 in-line components, containment penetrations. And we also l 1' 16 obtained much more limited data or input on electrical 17 components.
- - 18 MR. MICHELSON
- Let me ask, just to get educated
- 19 on this business a little more. Large, dynamic equipment, ,
20 like a pump, for instance, in case of an earthquake, it's 21 fairly likely that it may receive a start signal during the . 22 earthquake. Earthquakes don't -- they last 30 seconds or, 1. 23 as I understand it, or up to a minute. Do we understand 24 that we have qualified those pumps such as we know they can .
; 25 start up, accelerate and operate under SSE conditions?
i l I
l l l l 219 ] 1 MS. WITTE: Actually, one of the main comments 2 that we received on components that are -- not -- this isn't 3 going.to be a direct answer to your questions, but that they j 4 -- that we show proof of operability during earthquakes. 5 That if we don't have a design for a lower-level earthquake, i 6 then we need to show operability. ! l 7 MR. MICHELSON: Yes. See, operability means they 8 won't damage themselves in the process of accelerating, 9 because they were told to start during the earthquake, nor i 10 will they be damaged under continuous op,eration during the l 11 earthquake. You have to cover both. And start-up is quite i 12 a stressful thing for some-of the components to begin with. 13 And those start-up loads ought to be added to the seismic 14 loads during the start-up portion of the cycle. But I don't ! l 15 know that people ever thought or refined it enough to do it. I i 16 But the real world is. That's when they're going to s' tart. ) i 17 MR. CHOKSHI: I don't know whether they considered ; 18 that or not. 19 MS. WITTE: You may have brought it up earlier, , l l 20 that it just has to do with testing capability. . 21 MR. MICHELSON: It can be done, it's just very 22 expensive. Big equipment to test it in, and setting up 23 fluid and circulating fluid systems is not cheap. 24 MR. SIESS: Well, there are an awful lot of 25 components that are qualified by analysis. Where in NRC are
220 1 those analyses reviewed? NRR7 i 2 .MR. CHOKSHI: Yes. 3 MR. SIESS: Mechanical Engineering Branch? 4 MR. CHOKSHI: Yes, sir. It ought to be in Section 5 3.9. !
, 1 6 MR. MICHELSON: They would be the ones we need to l 7 ask. i 8 MR. CHOKSHI: Sections 3.9.
)
9 MR. SIESS: I thought the stuff they test is I 10 usually easy to test stuff, not -- 11- MR. MICHELSON: Yes. They test the little stuff 12 because that's not expensive. Big -- l l' 13 MR. SIESS: I don't think anybody ever tested long 14 pump shafts. 15 MR. MICHELSON: I don't know that anybody -- yes, 16
~
that's one of them. Another one -- I'm not sure -- big RHR 17 2,000 horsepower RHR-type size pumps have been tested under l 18 dynamic conditions, you know, with water circulating through 19 them and starting them and so forth. Yes, it's a very , 20 expensive test. ; 21 MR. CHOKSI: We have to look at the whole motor - 22 systems. 23 MR. MICHELSON: Is that because they probably 24 appreciate that it's beyond analysis? 25 MR. MURPHY: The Japanese do have the capacity and
221 1 they are talking about doing some of this.
.2 MR. MICHELSON: Do they put loops on their shake 3 tables then and run them in circulation modes?
4 MR. CHOKSI: Within the main feedwater. ' 5 MR. MICHELSON: That's a big loop. 6 MR. SIESS: Yes, and they've got big facilities, , 7 too. 8 MR..MICHELSON: Big money, too. 9 [ Slide.] 10 MS. WITTE: Okay, I tried to summarize very 11 briefly, what we learned from talking -- taking the survey 12 and talking to the utilities and designers. And one point 13 is that the impact of OBE design depends, to a large extent, 14 on the philosophy of the designer. I've given you two 15 examples, but there are more. 16 Some designers use different damping to generate > 17 response spectrum for SSE and OBE, and some use only one -- 18 excuse me, use only OBE damping for both OBE and SSE to
, 19 allow for scaling, since the determination has been made >
20 that OBE may govern anyway.
- 21 So, in that case, it's difficult to assess. If we t 22 look at calculations where designers used that philosophy, 23 it will be difficult to assess what the impact will be.
24 MR. MICHELSON: What do you mean, allow for 25 scaling? What scaling are you referring to?
222 1 MS. WITTE: A scaling of the analysis results, 2 seismic analysis results. 3 MR. CHOKSI: Instead of doing an SSA analysis, you 4 double the OBE. 5 MR. MICHELSON: You scale the loads. 6 MR. CHOKSI: Make it twice. 7 MR. MICHELSON: Does the shifting analysis - 8 recognize the spectrum may be different? Did they include 9 that much? 10 MR. CHOKSI: If the system is linear elastic, 11 there won't be that much shifting of the peaks. 12 MR. MICHEL3ON: My intuition says, I thought, that 13 the SSE always governs if you haven't changed any of the 14 frequency content. 15 MR. SIESS: Not if you use different allowable 16 stresses, it doesn't. 17 MR. MICHELSON: But keeping in mind my little 18 earlier rule that says this equipment has to operate under 19 the SSE, it started during the earthquake, is running, and , 20 by golly, it's supposed to curvive the earthquake in a 21 nominal condition. = 22 MR. SIESS: The thing is, the requirement -- the 23 thing is, if I designed for half the SSE and half the 24 allowable stress -- I'd say, half the SSE and one third the 25 allowable. stress, then the OBE is going to govern. It's not
l l l I '
, 223 1 going to fail at the OBE; it will just reach its specified l 2 allowable stress.
3 MR. MICHELSON: Chet, for static numbers like 4 buildings and so forth, that's fine. But for a dynamic 5 component like a pump and rotating equipment and so forth, 6 that doesn't work. You can't -- but the people aren't
, 7 allowed three times the stress, unless you only have to 8 assure pressure boundary integrity.
9 MR. SIESS: Somewhere, you have to make a 10 distinction between the design which is what we're talking 11 about now, and the qualification. Now, equipment is 12 qualified, not for an OBE. Seismic qualification equipment 13 is always based on an SSE. 14 It's based on the larger earthquake, and there, 15 it's no loss of function. It's not some calculated stress. 16 MR. CHOKSI: I think these kind of philosophy is 17 as much a function of level of earthquake. If you have 0.1 18 G, doubling -- 19 MR. SIESS: Yes, you don't design and qualify to 20 the same. l
. 21 MR. MICHELSON: If I remember, the ASME Code, 22 correctly, I can go too a Level C, for instance, on 23 earthquake, if it's k pipe. I can't go to Level C if it's 24 the shaft of a pump.
25 MR. SIESS: That's right.
224 1 MR. MICHELSON: I've got to stay in Level A, 2 because it's got to work, unless you declare that you don't 3 need the pump's pumpability, only its physical ~ integrity, 4 and then you can go to Level C. 5 MR. SIESS: There ASME for that pump that I know L . 6 of; is there? 7 MR. MICHELSON: Well, there are for certain , l 8 components, yes. l 9 MR. SIESS: Components. 10 MR.-MICHELSON: Shafts and cas,ing. ' 11 MR. SIESS: But again, the qualification, 12 equipment qualification, is based on the SSE. 13 MR. MICHELSON: Yes. 14 MR. SIESS: You either analyze it or you test it, 15 and the criterion is not always some allowable stress; it's 16 a function. 17 MS. WITTE: Right. 18 MR. SIESS: You see, the thing is this: it used j 19 to always bother people that here I've got the OBE and it , 20 governs. It has a lower damping, but before this thing can ! 21 fail, it's got to get up in the inelastic range and so forth - 22 which will be a higher damping. You can't go from the OBE - 23 - it doesn't fail at the OBE, because we're not designing h 24 with the failure limit state; we're designing with some i ! 25 allowable lower level limit state for some arbitrary stress. j l i . . _ .,
- - - . . . . - -- .. - - -- .. _ ~ _- ..
1 225 1 It might be the yield. l l 2 In spite of the fact that ASME will let you go to 3 1.5 yield, that doesn't mean it necessarily gets to 1.5 1 4 yield. l 5 MR. MICHELSON:
, When you go into yield, it's got 6 to operate.
l 7 MS. WITTE: Right. l , 8 MR. MICHELSON: It's the dynamic component. j You 9 can't go into the yield. 10 MR. SIESS: Yes, on the dynami.c component, that's i 11 not your limit there. There are other things that can be. 12 In piping, it clearly can. l 13 MR. MICHELSON: In piping, it can. 14 MR. SIESS: Of course, we've got miles of piping 15 out there designed for a code based on a mode of failure i 16 that no piping's ever existed in an earthquake, which is 17 even worse. It's much worse to design for wrong mode of 18 failure than it is for the wrong earthquake. 19 MS. WITTE: That's right. Another example is that 20 inline components are frequently designed for SSE loads and
. 21 then compared to OBE allowables. Of course, if there's no 22 design requirement for OBE, then most likely, that won't i
23 change. j 24 MR. SIESS: That's backwards from what people . 25 thought it was going to be?
226 1 MS. WITTE: Right. 2 MR. SIESS: You were going to design elastically 3- for one, and you were going to check for function for the 4 other. It didn't come out that way. 5 MS. WITTE: Okay, and then the third summary, I. F 6 suppose, is that the OBE might govern the design of those 7 components that I listed earlier in List A. So, I've just , i 8 repeated the same list here aga'in. 9 (Slide.) , 10- MS. WITTE: We also looked at , changes, required 11 changes to codes and standards. John Stevenson reviewed 12 codes and standards for us, and he looked at those that I've ! 13 listed here, ASME Section III, ACI 349-85 and 349.1R, ANSI 14 N690 for 1991, ASCE 1-82 and 4-86 and IEEE 323 and 344, 15 which is still in progress. l , 16 MR. SIESS: We have done a lot of seismic margin i 17 studies. If we change this OBE/SSE relationship, it 18 probably would reduce some -- anyplace the OBE governed, $ 19 there is at least a chance that the seismic margin would be e i
'20 reduced, but we don't know how much, do we?
21f MR. CHOKSHI: We don't know, right, and I'think we - 22 are looking at, so far, what we have found. It's not an 23 easy process. I think we probably would have to do some 24 design' calculations ourselves, because the way the people 25 designed, there's a lot of fudge factors. 1
. . . _ . , , . _ _ _ _ _ . , . _ . _ _ . . _ _ _.___ _ _ . _ _ _ - . _ _ _ _ _ _ __._.1
r r . 227 ' 1 MR. SIESS: If you go look at all the seismic 2 margins. If you only look at the lower end of the seismic l 3 margins, it might not be so difficult. 1 4 MR. CHOKSHI: That's what we're going to do, is j 1 5 select a few components and look at them. 6 MR. SIESS: Because if you're looking at the I
, 7 codes, the changes that would have to be made would be 8 mainly to change the load factors'or whatever, so that, with 9 OBE at one-third the SSE, OBE wouldn't govern.
10 MS. WITTE: Right. 1 11 MR. SIESS: Now, you get some feel for seeing how 12 big those changes have to be. 13 MR. CHOKSHI: Part of this is to identify whether 14 we need those changes and go back and do the -- 15 MS. WITTE: Right. 16 MR. SIESS: Okay. 17 (Slide.] 18 MS. WITTE: This review hasn't looked at that 19 aspect of it, whether or not we need to change load factors. 20 The review that John Stevenson did was just to look at them
- 21 and see what obvious changes need to be made if the ,
22 definitions are changed, and there are no surprises. 23 As I say, some definitions need to be changed. 24 References to dual earthquake loads and other minor wording 25 changes might have to happen. l i
. . --- .- . .-~ ----. -..~- .- - - -. - - - . -
l
't 228 !
1 Fatigue and seismic anchor movements are now { 2 reviewed in the -- for the OBE loads, and so, they'll have 3 to be included in the design somehow, and OBE is no longer - 4 included, and there will be some no-longer-relevant load 5 combinations in ACI and a few others that will need to be , 6 modified. 7 MR. SIESS: The proposal is that, at one-third the . 8 SSE, I don't need to check for the OBE. 9 MR. CHOKSHI: For the most part.. Now, there are l
- I 10 areas, currently only OBE, the design basis only exists in I 11 terms of dBE, like anchor motion.
12 MR. SIESS: Let's stick to the other for a minute. 13 Mow, wouldn't I have to check the codes to see if 14 the allowables or load factors or whatever would still be 15 valid if I -- at one-third the SSE? 16 MR. CHOKSHI: I think we are hoping that, by'doing 17 some more looking into this, that we will be able to 18 conclude that, if.you use one-third, it's generally true and 19 will not require everybody to go through it. . 20 The Germans have gone through that exercise, and 21 they have determined .4 as a factor at which everything will - i 22 remain within the allowable limits, except some few local 23 places, but we still haven't come to that point. l , 24 MR. SIESS: You're telling the designer, at one- { 25 third you don't need to check. We'll assure you that it I t
1 2 9 1 won't occur. I 2 MR. CHOKSHI: That's right. i 3 MR. SIESS: And if OBE would govern it, that we're 4 willing to let the SSE take over. 1 j 5 MR. CHOKSHI: Right. i 6 MS. WITTE: Right. It's a two-step process. 1 , 7 First we need to look and see where OBE governs and, if 8 there is any increase in risk, what that would be, and how { 9 do we solve that problem, and the second step is to look at . 10 what happens if you look at one-third of the OBE?
- 11 Then, does the stress that would occur.w,ith that j 12 one-third still meet the existing upset load combination !
I 13 allowable? 1 14 ~MR. MICHELSON: Now, if you're looking at this i 4 1 15 from the risk. viewpoint, let me ask, how do you presently i 16 look at the somewhat arbitrary rule that we don't assu'me we ; 1 4 17 get a pipe break concurrent with an earthquake, for l , l 18 instance? ! l 1 , 19 For deterministic design purposes, that's kind of 20 the rule we use. We do all of this, and then we turn around
- 21 and say but the ECCS equipment must survive the earthquake, 22 even though no pipes break that they would be needed for.
23 But do you have any probabilistic basis of some 24 sort to believe that rule, too, and the same for the OBE? 25 You know, we're trying to relate this probabilistically. I ; i
i 1 230 i 1 don't know how you do that. 2 MR. CHOKSHI: For example, even if you lose, say, 3 margin in the piping somewhat, it's not going to have an 4 impact, because there are many components much weaker than 5 the piping. So, in that case, probably, we will not have to 6 do anything with the piping. 7 But the study was going to the components which . 8 will tend to dominate, and seismic risk is sort of used here 9 in terms of -- 10 MR. MICHELSON: Well, let me just ask the simple 11 question: What's the probability of getting a pipe break 12 during one of these SSE's? 13 MR. CHOKSHI: It's very low. 14 MR. MICHELSON: Now, what basis do you have and 15 how low are you -- you're saying "very low," but has 16 somebody done some studies to arrive at that conclusion, 17 keer,ing in mind that not all piping is even seismically 18 qualified in the plant to begin with, and you won't get any 19 " pipe break? 20 MR. SIESS: They reviewed many, many -- 21 MR. MICHELSON: Yes, I know what they did, but - 22 those weren't SSE levels. 23 MR. SIESS: What do you mean they weren't? 24 MR. MICHELSON: It depends on the plant. 25 MR. CHOKSHI: EPRI and NRC tests, which were
d , 231 1 completed -- 1 2 MR. SIESS:- They looked at big earthquakes. 3 MR. MICHELSON: If you don't have any pipe breaks 4 during an earthquake, then why do we insist on aismic i 5 survivability of this equipment? It's there just for the 6 purpose.of addressing pipe breaks.
. 7 MR. CHOKSHI: Part of it is pipe rupture l
8 locations, yes. 1 9 MR. MICHELSON: No, no, there aren't going to be 10 any pipe ruptures, and it's there to address large breaks, 11 most of it is. Some of it is not. Some of it is shutdown l l 12 heat removal. I can understand why that's needed, but how 13 about your high-pressure injection systems and so forth? 14 They're clearly there to flood a reactor vessel in l 15 case of a pipe break, and if we aren't going to get any pipe I 16 breaks during an earthquake, why do we seismically qualify a 17 high-pressure injection pump? 18 MR. SIESS: In spite of everything we have
, . 19 learned, people are not satisfied.
20 MR. MICHELSON: Yes, I know why we do it, but when 21 people start telling me they're going to do some 22 probabilistic business in this, then I ask, well, what's the 23 probability and, you know, why are we even qualifying this 24 equipment for earthquakes, since the probability is so low 25 of having that situation arise?
. __ _=
, i I l r 4 4 232 l 1 MR. CHOKSHI: But that system may be called upon l. 2 after the earthquake for high-pressure injection. You might 3 still have a small hole. 4 MR. MICHELSON: Yes,-but we don't design for any l -5 pipe breaks as a consequence of an earthquake. There is no . 6 concurrent LOCAs and earthquakes. LOCA can be anything from
- 7 a small break on up to-the biggest break, and none of them .
8 are in consideration. 9 So, I don't need -- I don't think I need a high-l 10 pressure injection, as long as I can remove the decay heat. 11 MR. CHOKSHI: That's true.
'12 MR. MICHELSON: And I've got a number of other 13 mechanisms.
L 14 MR. CHOKSHI: Large LOCA and seismic has generally 15 not been dominant, because of other reasons, support 16 failures and other things. 17 MR. MICHELSON: Well, when you do a PRA, then, do j 18 you go through and analyze all the piping to see which one 19 might break? We're not talking about the random occurrence , f '20 of a break during an earthquake. That ought tt be l 21 incredible. - ' I 22 We're talking about an earthquake-induced failure
'23 of the piping.
i 24 MR. CHOKSHI: In most of the PRAs, that's
- 23 considered a much higher capacity type of event.
L
1 233
-1 MR. MICHELSON: It's not included in the analysis.
2 MR. CHOKSHI: Most of the time, in the PRA, you 3 have a small break. 4 MR. MICHELSON: But not concurrent with an 5 earthquake. 6 MR. SIESS: I think the feeling is now that the
. 7 probability of a pipe break due to inertial forces -- that's 8 the pipe vibration -- is as close to zero as it's ever going 9 to get.
10 Pipe breaks due to differential support motions, 11 equipment not anchored, pipes not anchored, are not 12 negligible,-but they are still very low. 13 If we anchor equipment, if we do the thing that 14 everybody is putting the emphasis on now, if you watch heavy 15 weights on small pipes, the probability of pipe break during 16 an earthquake is low, and that probability is not likely to 17 Hbe much of a function of whether it's SSE or OBE. It's 18 going to be a function of whether you thought of it or you
, 19 didn't think of it.
20 If you design those things for any earthquake, you 21 are in darn good shape, because there's pretty good margin 22 there, but if you don't anchor the darn thing down or if you 23 don't support a valve on a -- a large valve on a three-inch 24 pipe, you can get it, and I think that's about where we 25 stand.
t 234 ; i 1 But inertial forces, which we still worry about -- 2 hangers and snubbers a stuff -- very, very low probability 3 of causing a pipe break. 4 Again, it's a question of what you think about, 5 not how big-it is. That's true of a lot of things. , 6 MR. MICHELSON: When I see a PRA that says this is i 7 found to be credible, I wonder if they have analyzed the . 8 piping from the viewpoint of support points and all the i 9 other details. They have to have a lot of understanding and 10 detail to do that. ! 11 MR. SIESS: PRAs depend an awful lot on somebody's 12 imagination, and if they can't think of the good things, 13 they're not going to get done. If you don't analyze them, 14- they don't happen. 15 MR. CHOKSHI: If the right person walks down the 16 plant and understands it, it will get discovered. 17 MR. SIESS: Yes. Onward. 18 [ Slide.) 19 MS. WITTE: Okay. To summarize, or anyway, to , 20 talk about the future, I guess, we expect to complete a 21 detailed review of steel containment design next year. EQE - 22 will do that for us as well. 23 We hope to obtain more data from several sources 24 on component loads in order to be able to evaluate more 25 fully the adequacy of the proposed OBE level. It turns out
i 235 1 to be very difficult to get the data that we're looking for , 1 2 on the existing components from existing plants, because of l 3 the way the data has been handled. 4 And we will complete the risk study, with that 5 data that's been collected. 6 MR. SIESS: What's the steel containment issue? 7 Buckling? 8 MS. WITTE: Buckling. Right. 9 HR. SIESS: We got them designed for that hydrogen 10 explosion, and we still have to worry ab,out buckling. Ji MR. CHOKSHI: I think we have to look at the whole 12 picture. 13 MR. SIESS: Most of what you have said has been 14 the question of OBE versus SSE and so forth.
- 15 How much of it has been quantitative enough to 16 justify the .33 that's in the present proposal, other than 17 it's less than the Germans' .4?
18 MS. WITTE: So far, we can't quantify it. We hope o 19 to be able to do that in the coming year. 20 MR. SIESS: I don't know whether you'll ever be . 21 able to quantify it. 22 MR. CHOKSHI: I don't think it will be complete. 23 MR. SIESS: I don't know what would constitute a 24 complete justification. If you could show that at .33 under 25 existing design parameters and allowables, an OBE at .33
1 s i i l
- \
' 236 = l 1 l with SSE would never govern, that would be a complete 2
. justification, and there's no way you could ever get to that 3 point in a finite' length of time.
4 If we could look at a lot of cases and say that in 5
-no cases would the OBE be more than a 10 percent or 15 .
6 percent factor, that would be a pretty good justification, 7 because that's as closo as we are anyway. . 8 Or, if there are a few cases where it's large, I 9 think I'd take a look at what's' wrong. 10 MR. CHOKSHI: I think the marg.in comes from the 11 people's practice. 12 MR. SIESS: But you know, for the kinds of things 13 that are covered by the basic codes -- concrete, steel, ASME 14 -- you ought to be able to simply take the load combinations 15 plus allowables for a range of ratios of those loads, which 16 we ought to be able to find out what that range is from 17 experience, and just tell what happens at .33, once you know 18 what the range is.
.19 The trouble is, if you get six different load .
1 3 1 20 combinations and they all have a range, you have, you know, 1 21 a fairly large number of combinations and permutations. But ' 22 again, with a computer, that only takes a few minutes. ! 23 MR. KENNEALLY: The other issue is to get the l l 24 actual other loads that go into it, the thermals or l l l 25 whatever, so you do realistic calculations. ' t I
l l 1
/
237 l 1 MR. SIESS: Yes. You have to know their relative l 2 values -- 3 MR. KENNEALLY: Correct. 4 MR. SIESS: -- and what combinations. 5 MR. KENNEALLY: That's probably one of the hard 6 parts. 7 MR. SIESS: And some of them, you can assume a 8 zero, because they're not going to change. 9 MR. KENNEALLY: That's right. 10 MR. SIESS: It's just like in a simple dead load 11 case, all you have to do is a range of dead loads / live 12 loads, and you can compare anything you want. There a range 13 from zero to one takes care of it. Here it's a little more 14 complicated. 15 MS. WITTE: We hope to get data on maybe two 16 examples of each component that we have on those lists for 17 which we can get data. So we never will have a large body 18 of information. We'll always be limited. But we hope to be 19 able to gain from it some insight into how those components 20 function. 1 . 21 MR. SIESS: Some of these outfits, like Bechtel, { 22 must have computer programs for designing these things. 23 MR. CHOKSHI: Yes. And I think you've been i 24 talking with Bechtel. 25 MS. WITTE: Yes, I've been talking with Bechtel. !
,.- .. . . - - - . . - . . - . ~ - - - - - - . . . - - . . . - . . - ---.-. - . 1 l l , , 238 ) 1 MR. SIESS: And have them stick in some variables. 2 If they wanted to, they could do it. Some of them, I l 3 suspect, have already done it, and found out it didn't make 1 l 4 enough difference to fight about. ! 5 What the Germans use is their basis for their .4. . 6 MR. CHOKSHI: They have some calculation. And I 7 have a letter here, and it is proprietary. So, so far, s 8 we're just looking at the study itself. 9 MR. SIESS: Who did it? 10 MR. CHOKSHI: Do you remember who did the work? 11 The gentleman's name. I will show it to you. I can't 12 pronounce it. From GRS. I 13 MR. SIESS: Oh, from GRS. All right. I 14 MR. CHOKSHI: What they conclude is that .4, 15 except for some non-linearity, they don't expect anything l 16 unusual in the structures. On the whole, I think that'their 17 conclusion is that .4 won't be the number for the mechanical l 18 and instrumentation. 19 MR. SIESS: I really can't see how you can get in . 20 any trouble, because the margins are so large, and the L 21 components I remember where the margins are small are not - 22 likely to be ones where that's going to be a big deal, if 23 you feed them into a PRA. 24 MR. CHOKSHI: We are also talking about the t
- 25 standard design, which is designed to a much higher .g, and l
l _ ~
i l l 239 1 OBE for that could be 1.1. 2 MR. SIESS: See, places like those tanks -- let's 3 see. The tanks got an OBE SSE? 4 MR. CHOKSHI: I believe so, yes. 5 MR. SIESS: They just had an error in the earlier 6 design. , 7 MR. CHOKSHI: Yes. I i 8 MR. SIESS: If the OBE governed that, that will 9 not happen in the future, you see? And those are one of our 10 critical spots. 11 I think what you're doing is just fine. I think 12 this is an approach. And I think your biggest problem is l l 13 going to be getting the codes straightened out. 14 MR. CHOKSHI: Yes. That's true. In fact, EPRI l 15 has proposed to help us on that, and they also prepared a l 16 proprietary document, indicating where more changes ar's
- 17. needed in the Regulatory Guides. I propose some alternative 18 positions.
19 MR. SIESS: I know what you're going through, i .o l 20 because we've been trying to do the same thing on building.
. 21 We've been trying to go through and change the concrete code 22 to accommodate new ANSI 858 loads. We're trying to come up 23 with a concistent set of loads for steel and concrete. And t 24 we want to get the same answer with the new loads that we -
l 25 got with the old ones, i
/
240 1 Okay. Next. Thank you very much. That was 2 helpful. 3 In your discussions, if we use one-third the SSE, 4 what number of people do you think would opt for that or 5 would want something lower? , 6 MS. WITTE: I haven't heard anybody say that they 7 would want it lower than one-third. - 8 MR. SIESS: So, most of them would opt for a 9 third? Of the two options that were given, you can do a 10 third with no calculation, or you could take something 11 different. 12 MS. WITTE: Take something higher and do an 13 analysis. We did get the comment that there should be no 14 inspection level shutdown at all. We heard that from some 15 sources. That we should be able to see any earthquake up to 16 SSE before there's an inspection level shutdown. 17 MR. SIESS: Did you hear that from a utility, that 18 there should be no shutdown level? 19 MS. WITTE: I don't remember hearing it from a . 20 utility, but from a designer. 21 MR. SIESS: Oh, well. I can't imagine a utility - 22 that wants to take a chance with its plant after -- an OBE i 23 is not a small earthquake. After all, Diablo Canyon had one 1 24 that measured .0015 and they declared it an unusual i 25 occurrence -- unusual event in California. l 1 1
241 1 MS. WITTE: A bigger concern that we heard 2 actually was that we need to define very precisely what's 3 involved in the inspection so that plants -- utilities know 4 what they're up against when they accept the risk of the 5 shutdown and inspection with no analysis. l 6 MR. SIESS: Well, now, thEt's in there. , 7 MS. WITTE: Right. They need to know if they're i 8 going to be shutdown for a year or a month or what. 9 MR. SIESS: And how do they expect to find that 10 out? 11 MR. MICHELSON: How many plants have seen an OBE 12 so far -- experienced an OBE? 13 MR. SIESS: Perry exceeded their OBE one 14 frequency. 15 MR. MICHELSON: Well, you're not talking about a 16 very frequsnt occurrence to begin with. 17 MR. CHOKSHI: Not so far. 18 MR. MICHELSON: Well, there's no reason to believe 19 seismic activity is picking up, is there? Their past 20 history now ought to be a pretty good indicator of our 21 future experience. 22 MR. SIESS: We've got a lot of reactor years out 23 there, and what has exceeded an OBE. Of course, the OBE was 24 set a little high, but nobody is even coming close. 25 MR. MICHELSON: Not something to worry about too
2 2 1 much, because it's very unlikely. If it does happen though, 2 then you'd have to sort of ad hoc the inspection and 3 whatever needs to be done, which you would-have to do 4 anyway. 5 MR. SIESS: It wasn't even operating that. I , 6 don't think it shut down. 7 MR. CHOKSHI: It will be a new composed guide. We - 8 probably won't shut it down. 9 MR. SIESS: As I recall, about half the plants --- 10 I haven't made a survey recently, but about half the plants 11 would trip the turbine on vibration. And I guarantee you an 12 OBE would trip the turbine. Now, all of them didn't. Some 13 of them just indicate. But, we used to talk about seismic ! 14 trip, and it turns out that there's a pretty good seismic I 15 trip on a lot of'the plants, it's just depending on -- I l 16 guess that philosophy of whether you trip the turbine on 17 vitiation of annunciate is the same kind of thinking. It's l 18 just how the company thinks about safety and their 19 investment, I guess. . 20 Okay. l j 21 (Slide.) - 22 MR. McMULLEN: Maybe we would have less problems 23 if we built the plant and then did the investigation. 24 MR. SIESS: Well, on some of them that's darn near
- 25 what we did.
I
a 243 1 MR. McMULLEN: Right. 2 [ Slide.] 3 MR. McMULLEN: The primary goals of Appendix B is 4 to identify, define and characterize seismic sources that 5 are significant to estimating a seismic and surface 6 displacement hazards for a nuclear power plant; define , 7 reasonable and site subsurface conditions as to their 8 vibratory ground motion propagation characteristics and 9 stability under seismic loading. And all of this to provide 10 basic input to deterministic and probabilistic hazard 11 analyses. 12 Appendix B is a revision of Appendix A in which, 13 as Dr. Murphy said, where the -- containing the criteria, i 14 whereas the description of methods and more prescriptive 15 type things are in the regulatory guide. 16 This is to guide applicants and NRC reviewegs. j 17 [ Slide.] 18 MR. McNULLEN: This is a slide just to show that 19 the definitions and the terms -- the terms and the phrases 20 that were changed in Appendix A. Term capable fault, j 21 tectonic province, tectonic structure, two earthquakes. And 22 some of the new terms are seismic source, which encompasses 23 both seismogenic source and capable tectonic source. 24 Seismogenic source -- 25 MR. SIESS: What replaces capable fault? l 1
244 1 MR. McMULLEN: Capable tectonic source. The 2 ressen for that is it's not too much a different definition 3 than that of capable fault, it's just been expanded to 4 include other kind of surface and near surface deformation 5 that's caused by faulting at depth. Like what happened at , 6 Coalinga in 1983 and William Harris a couple of years ago, 7 and the Loma Prieta earthquake where surface folding uplift - 8 was caused by a fault at depth. And that fault did not 9 actually break surface itself. 10 MR. SIESS: Can we have an earthquake that isn't 11 caused by a fault? Are the seduction zone earthquakes 12 fault-caused? 13 MR. McMULLEN: Yes. 14 MR. SIESS: It's considered a-fault? 15 MR. McMULLEN: That's a capable tectonic 16 structure. 17 MR. SIBSS: Yes. But is a seduction zone 18 earthquake caused by a fault? 19 MR. McMULLEN: Yes. . 20 MR. SIESS: Do you consider that a fault? 21 MR. McMULLEN: Yes. Either the interface between - 22 the two plates are faults associated with that interface. 23 MR. GIESS: Okay. Really what you got rid of was 24 the word capable and, in the process, got rid of the word 25 tault.
245 1 MR. MURPHY: We kept capable, but got rid of the 2 faults. 3 MR. SIESS: Oh, that's right. You do have capable 4 tectonic source. 5 You said Loma Prieta wasn't a fault? 6 MR. McMULLEN: Oh, yes, it was a fault. But the 7- fault that caused the earthquake didn't rupture the ground 8 surface. 9 MR. SIESS: Yes, but none of the Eastern U.S. 10 earthquakes have ruptured the ground sur, face either, but 11 we've always referred to those as faults. 12 MR. MURPHY: But not capable faults. 13 MR. SIESS: New Madrid is not a capable fault? Is 14 that what you're saying now? 15 MR. MURPHY: New Madrid, I would say we have not 16 specifically identified a fault at New Madrid. 17 MR. SIESS: Okay. 18 MR. MURPHY: We have identified a structure or a 19 source in New Madrid. 20 MR. SIESS: So, you have a capable tectonic {
- 21 source?
22 MR. MURPHY: That's correct. 23 MR. SIESS: I'm going to still call it New Madrid. 24 I live closer to it than you do. 25 MR. MURPHY: That's okay. But I went to school in I l
.. . . . .. -- . . - . . - .- - .- .. .-.. - . . . . - - . . . . - . - . ~ _ . . .
246 1 St. Louis. 2 MR. MICHELSON: That's no excuse. 3 MR. SIESS: All the seismologists call it New 4 Madrid that I know of. 5 MR. MICHELSON: People who live there call it 6 that. 7 MR. SIESS: In the Eastern U.S. we have things 8 like Charleston and New Madrid, that we know are capable 9 tectonic sources. 10 MR. McMULLEN: Well, the actua.1 source hasn't been 11 identified. Those would be seismogenic sources. 12 MR. MURPHY: They are seismogenic sources because 13 there has not been the surface displacement and surface 14 rupture associated with them. 15 MR. SIESS: Okay. Tectonic source has to break 16 the surface? 17 MR. MURPHY: Capable tectonic source has to break 18 the surface. 19 MR. McMULLEN: It has to deform the surface. , 20 MR. MURPHY: Deform. 21 MR. SIESS: Deform it. - 22 MR. McMULLEN: Yes, by the actual -- 23 MR. SIESS: New Madrid deformed the surface. It 24 formed a lake and all sorts of things. . 25 MR. McMULLEN: I'm sorry, which --
i r l , i 1 l
- I 1
247 ' 1 MR. SIESS: New Madrid deformed the surface. I 2 don't get a lake without deforming the surface. 3 MR. McMULLEN: Yes. It also caused an uplift. 4 But that hasn't been identified -- that hasn't been related l 5 to the specific fault that caused the earthquake. 6 MR. SIESS: No, just it occurred at the same time.
. 7 That's a pretty strong indication. The ground shook and the 8 ground went down, subsided at the same time, so we think --
9 I think there's a pretty good chance it was caused by the 10 earthquake. 11 MR. MURPHY: But we're -- maybe we'rt arguing 12 semantics right at the moment. 13 MR. SIESS: No. What -- 14 MR. MURPHY: What we're talking about -- 15 MR. SIESS: No, we're arguing definitions. 16 MR. MURPHY: Okay. 17 MR. SIESS: I want to know what a seismogenic 18 source is.
, 19 MR. MURPHY: My interpretation of why we're not 20 calling New Madrid -- what we saw there was that that may 21 have been secondary ground motion, not primary ground 22 motion, secondary -- primary ground motion being directly 23 related to the movement on the fault; that if the fault
( 24 moved and the rocks that were on the surface, that lay on th 25 surface, draped over the moved fault, that's primary, but if i l
. . . . . ..#.- -m __
_4...m , ,. . . . - .- l l 1 248 1 we're talking about slumping -- 1 2 MR. SIESS: From liquefaction or landslides, it's i 3 not. ; 4 MR. MURPHY: That's right. 5 MR. SIESS: Now I understand. , l 6 MR. MURPHY: Okay. I 1 7 MR. SIESS: Is that clear if I read the reg guide? .
)
I 8 MR. MURPHY: If it's not clear, we'll make it 9 clear. 10 MR. SIESS: Okay. Now, seismic source is
)
11 anything. That's a generic source for seismogenic and l 12 capable tectonic source. j 13 MR. McMULLEN: That's right. I 14 MR. MURPHY: That's correct. 1 15 MR. SIESS: Now, capable -- is there a tectonic ! 16 source that isn't capable, or does capable and tectonic 17 always go together? 18 MR. MURPHY: As we're using them in the 19 regulation, they always go together. , J 20 MR. SIESS: Okay. 21 MR. McMULLEN: There are many tectonic sources --
- 22 tectonic structures throughout the eastern US, but they are 1 23 not -- they may not be seismogenic sources or capable 24 tectonic sources.
25 MR. SIESS: Well, yes. I mean, there are a lot of
i S I 249 1 tectonic structures that are ten million years old. ) 2 MR. McMULLEN: Right. 3 MR. SIESS: A mountain forming and -- 4 MR. McMULLEN: Yes. 5 MR. SIESS: I'm still -- every once in a while, I 6 spend a few minutes thinking about some way of finding a new
.. 7 name for those two things you call deterministic and 8 probablistic. I have no problem with probablistic, but 9 deterministic ain't deterministic, and I keep trying to 10 think of a better way of doing it, and I, haven't come up ,
l 11 with it. Maybe Dr. Lewis will think of something better. 12 MR. MURPHY: Okay. 13 MR. SIESS: Because every time you say 1 14 " deterministic," I'm afraid he's going to jump down your 15 throats. You agree they are misnomers. l 16 MR. MURPHY: Yes, sir. 17 MR. SIESS: They are the old method and the new 18 method, and I just find some -- it just bothers me. 19 MR. McMULLEN: The rest of these newer definitions 20 we talked about in quite a bit of detail earlier. Do you
. 21 want to discuss them anymore? ,
22 MR. SIESS: No. I want to go down another couple 23 -- you really haven't deleted operating basis earthquake; l 24 you've just redefined it.
- 25 NR. MURPHY: Redefined it, yes.
l l l \
_ .__._ ._ _ . . _ _ _ _ _ _ _ _ . _ _ . . . . - _ . . . _ _ . _ . . - . _ . _ _ . _._.___m . ) i i s 250 . 1 MR. SIESS: And you don't really have an operating , 2 basis earthquake ground motion as such. You have invented a i i 3 new term called the safe shutdown earthquake ground motion, l 4 but this is not important. 5 Now, the maximum credible earthquake has been - 2 : { 6 replaced by the estimated maximum earthquake? Is that i i ! j 7 right? . i 8 MR. McMULLEN: Deterministic earthquake. That's l 9 an earthquake you come up with -- 10 MR. SIESS: What has happened to the maximum 11 credible earthquake? Has it got a new name or is it no 12 longer a concept? 13 MR. McMULLEN: W 11, it's no longer -- I guess 14 it's still a concept, but we're not using it. I liked it i 15' myself, but apparently it causes too much confusion. It 16 implies that the maximum earthquake possible, and -- 17 MR. SIESS: It doesn't say possible; it just says
)
18 credible. But people don't believe that. 19 MR. McMULLEN: Yes. The people interpret that -- , 20 MR. MICHELSON: Think of one that is incredible. 21 MR. SIESS: The thing is, different people have - 22 different levels of credibility when it comes to 23 earthquakes. 24 MR. McMULLEN: It was, I think, the recommendation 1 25 of the panel, too, that we not abide by that.
. .__. . . - - ._ ~ - ..
j 1
, 2b1 ;
i 1 MR. SIESS: Does the estimated maximum earthquake l 2 correspond more'or less to what some of us used to call the i 3 maximum credible earthquake?' Is it a more clearly defined t ! 4 term? l l 5 MR. MURPHY: It's the closest approximation we 6 have left.
- i
. 7 MR. SIESS: And that's -- !
l 8 HMR.. MURPHY: It should be expected maximum. We've i 9 gou s' typo there. s 10 MR. SIESS: It's expected maximum. l 11 MR. MURPHY: Expected maximum, not estimated l l 12 maximum, L ! 13 MR. SIESS: And that's used in the probablistic 14 approach or just in the deterministic? I 15 MR. MURPHY: That's used in the probablistic 16 approach. 17 MR. SIESS: But only there, or is it used also in 18 the deterministic?
. 19 MR. MURPHY: It's principally used in the 20 probablistic approach, but it does influence, if you want, l - 21 in some cense, the selecting of what we're calling up here 22 the deterministic earthquakes that are associated with --
23 MR. SIESS: Okay. l 24 MR. MURPHY: -- the various seismic sources. { 25 MR. SIESS: I remember now.
l l 1 l 252 l N 1 MR. MURPHY: Okay. 2 MR. SIESS: And the SSE ground motion. Okay. Now, l i 3 did we cover everything? I didn't look ahead, but go ahead. I 4 I was looking at the next slide where you were defining the 5 ones I was -- 6 MR. McMULLEN: Yes. I was going on by that when 7 we -- , 8 MR. SIESS: I think I've got those two 9 straightened out. Are these the definitions right out of , 1 10 the document? ' 1 11 MR. McMULLEN: No. These are abbreviated 12 definitions. l 13 [ Slide.) l 14 MR. McMULLEN: I just abbreviated them to show the 15 difference between seismogenic source and capable tectonic 16 source. 17 MR. SIESS: That's still got that "not expected 18 to" language in there. 19 MR. MURPHY: Yes, sir, it does. . 20 MR. SIESS: It weakened the whole thing, in my 21 mind. - 22 MR. MURPHY: We can fix that fairly easily. 23 MR. SIESS: Okay. Yes. 24 [ Slide.]
- 25 MR. MaF.ULLEN
- This is a list of most of the i
l
l 253 1 information needed about seismic sources: Location, ! 2 geometry and distance from site. i 3 MR. SIESS: Wait a minute. Wait a minute. You 4 had a page in here that's excerpted from -- 5 MR. McMULLEN: Yes. 1 i 6 MR. SIESS: Can I ask you something about that?
. 7 MR. McMULLEN: Yes. I've got it here.
! 8 (Slide.) 9 MR. SIESS: In the first place, that's the one j 10 that defines the capable tectonic source, and we've talked , 11 about it a little bit earlier. 12 The last paragraph on there -- have we had 13 anything like that in the rule previously? 14 MR. McMNLLEN: Yes. That's pretty much the same 1 15 wording as was in the earlier one. We still recognize the 16 difficulty of determining whether an Eastern -- 17 MR. SIESS: It says if you dig a hole and you find 18 a fault but it's pre-Quaternary, you can forget about it. 19 MP. McMULLEN: After investigating it to determine 20 that that's its characteristic.
. 21 MR. SIESS: Well, now that's the part I don't 22 understand, because if I find a fault, no matter what its 23 age -- let's see. .How old is Quaternary?
24 MR. McMULLEN: Two million years.
- 25 MR. SIESS
- Two million years. And if I know it's 3
4
,-.. . - ,, -- - - - - ~-
254 1 two million years old, do I still have to go up and look for 2 the last 500,000 or the last 50,000? 3 MR. McMULLEN: No. The reason Quaternary is in 4 there, we figured what's going on today -- tectonic regimes 5 don't change overnight. If you consider it that far back, 6 you have a better chance of finding, you know, what's going 7 on today. Not that everything that old is active today. . 8 MR. SIESS: I find a fault -- I'm thinking of a 9 couple of places like we dug a hole and found a fault -- one 10 of them was in Puerto Rico -- and it was a fault, clearly. ; 11 MR. McMULLEN: Right. 12 MR. SIESS: If I could date that as pre-13 Quaternary, can I stop there? 14 MR. McMULLEN: As not affecting pre-Quaternary 15 strata? 16 MR. SIESS: Well, all I know is it was strata 17 there or -- it says here, " ... geologic structural features 18 which are geologically old." Now, I'm not quite sure what ' 19 that menne. I just assume that -- . 20 MR. MURPHY: Let's say that's pre-Quaternary for 21 the moment, more than two million years old. - 22 MR. SIESS: Yes. So what do I mean by " geological 1 23 structural features that are old"? l 24 MR. <%?ULLEN: Folds. 25 MR. SIESS: I have got a fault. You know, it does l
l l 255 1 this in the hole. 2 MR. McMULLEN: Right. l 3 MR. SIESS: How do I -- that's a geologic 4 structural feature. 5 MR. McMULLEN: Yes. 6 MR. SIESS: Is there any way I could determine l 7 that that occurred two million years ago and not 500,000 l 8 year ago? ) i 9 MR. MURPHY: A simple way would be that if you had ! 10 the pre-Quaternary rocks broken by the fault and then these ) 11 pre-Quaternary rocks were overlain by much more recent 1 12 rocks, maybe 50,000 years old, and those were unbroken, that i l 13 would give you a date that this thing hasn't moved, you i 14 know, let's say in the 50,000 years, or if it was overlain 15 by something older, if you could date what was the age of i 16 the unbroken rocks or sediments on top, you then have a 17 handle on the age of the last movement on that fault. 18 MR. SIESS: But now wouldn't that be covered by 19 the item just up in (j ) (1) ? 20 MR. McMULLEN: What this last paragraph is, is if
. 21 you find a fault and you can't absolutely date it, like many 1
22 faults in the East, you can, by the characteristics of that l 23 fault, relate it to the tectonic regime that created a fault ! 24 that is similar to that that you do have an age on. I mean, ! 25 you don't have to absolutely date that fault. You relate it l i
. . ~ . - . . .. - . - , . . - . - - - . - _ . . . . . . . . . . . - . - . . - -
)
I 1 256 j 1 to a similar fault that was obviously created under the same I 2 stress regime. 3 MR. SIESS: I see. That's the structural 4 association of a structure with geologic structural features 5 that are old. In other words, it's tying this one to . 6 another one that you can date? l 7 MR. MURPHY: That's right. . ! 8 MR. SIESS: Okay. That's the key part of this. l l 9 MR. McMULLEN: Right. t 10 MR. SIESS: It's not the fact that it's old, it's 11 the fact that you can relate it to another structure. 12 MR. MURPHY: That's right, and that you can date 13 the other structure.
- 14 MR. SIESS: I guess I read past that a little too !
i 15 fast. I thought the emphasis was on the age, and it's not. i 16 The emphasis is on tying it to another -- 17 MR. MURPHY: Yes, sir. It's on the association. - 18 MR. SIESS: Okay. 19 [ Slide.] , I 20 MR. McMULLEN: The information needed about 21 seismic sources, the reason we're doing the investigation: - 22 location, geometry and distance from site; history of 23- Quaternary displacements and information in that category -- 24 age of last offset and ages of previous displacements, 25 rupture length, rupture length per event, rupture area -- f ~ - w , - - -
2b7 1 now, these are things that apply to the Western US rather 2 than the Eastern US -- recurrence intervals and slip rate; 3 associated seismicity and paleoseismicity; the relationship - 4 of the structure to other potential seismic sources; EMES , 5 and recurrence models; deterministic earthquakes; other 6 factors that contribute to the characterization of a seismic
. 7 source; characteristics of soils and rocks between the 8 source and the site, and, of course, the site-specific 9 characteristic.
10 MR. SIESS: This says the seismic source. I've , 11 got a site located in the middle of Indiana again. 12 MR. McMULLEN: Indiana? l 13 HR. SIESS: Yes. It's right near Indianapolis. I ; 1 14 don't know where they're going to get the water, but we'll . 4 15 put it there anyway. I 16 MR. McMULLEN: How is the Anna, Ohio area? 17 , MR. SIESS: That would be a seismic source. 18 Wabash Valley Fault Zone would be a seismic source.
. 19 You have to get all of this information. Now, at 20 what point does the distance of the site get large enough? - 2.1 New Madrid is close enough to have some effect.
22 MR. McMULLEN: New Madrid would certainly be a 23 source there, too. 24 MR. SIESS: Now, these are known seismic sources, 25 right? i
2h8 1 MR. McMULLEN: Right. 2 MR. SIESS: Do we have to go out and look for any 3 more? 4 MR. McMULLEN: That's the purpose of the regional 5 investigation, out to a radius of 200 miles. 6 MR. SIESS: Now, does the regional investigation 7 come before this? ., 8 MR. McMULLEN: Well, this is the things you're 9 looking for in your investigations, including the regional 10 investigation. . 11 MR. SIESS: Okay. Now, are you going to tell us 12 about the regional? 13 MR. McMULLEN: Yes. 14 MR. SIESS: Okay. Seems like we're in the wrong 15 order. 16 [ Slide.) 17 MR. McMULLEN: These are just major subjects that 18 are focused on in the investigation: seismicity, geology, 19 geophysics, and geotechnical investigations. _ I 20 (Slide.) 21 MR. McMULLEN: These are investigations that are - 22 -necessary to validate all sites, no matter what part of the 23 country. It doesn't say anything about the level of detail. 24 These are just subjects that have to be addressed. 25 Certainly the investigation for Susquehanna would be much 4 l
t 259 1 different from investigations for Diablo Canyon. 2 MR. SIESS: And how would I know that, so that I 3 do the right' thing the first time? 4 MR. McMULLEN: You learn that in-looking at 5 available data. You know that there's -- 6 MR. SIESS: I want to know what I have to do so
. 7 that when I bring it to you, it will be acceptable'and you 8- won't tell me to go back out and get something else.- Where 9 do I find out how far I have to go out, what I have to look 10 for, when to stop?
11 MR. MURPHY: That will all be in the Reg. Guide 12 1015. 13 (Slide.) 1 14 MR. McMULLEN: The last slide -- 15 MR. SIESS: And what we're looking at here is 16 strictly what's in Appendix A? l 17 MR. MURPHY: What you're looking at here is in the 18 Reg. Guide.
, 19 MR. SIESS: Okay.
20 MR. McMULLEN: It's not in the Reg. Guide you
. 21 have. It's in the later-on draft of that.
22 MR. SIESS: The Reg. Guide I've got, I thought, 23 either that or the summary of it had a lot of information 24 about how far out you had to go.
- 25 MR. NURPHY
- That is in the summary, in the 4
b 4 i _ ._ _
260 1 updated version that you will get, if you want, in February. 2 MR. SIESS: Okay. 3 MR. McMULLEN: I beg your pardon. There is a 4 discussion about radiuses of 200 miles and 25 miles and 5 5 miles in the one you have. i l 6 MR. SIESS: It's in the summary or in the draft l 7 guide? , 8 MR. McMULLEN: Yes. It's in the Section 2, ' 9 summary in the introductory part of Section 2. 10 MR. SIESS: Reconnaissance and other sources. 11 MR. McMULLEN: I believe it was also in the 12 positions in the back. 13 MR. S1ESS: You know, this Dg 1015 is a textbook 14 or.a tutorial. It really ought to be a NUREG with a short 15 Reg. Guide attached to it somewhere. 16 MR. McMULLEN: When we first looked into this a 17 year ago, we decided it would take at least nine Regulatory 18 Guides to address all the subjects that had to be addressed. 19 In the two years we had to do it, we didn't think -- , 20 MR. SIESS: Well, I can see the staff issuing a 21 NUREG and then one Regulatory Culde which refers to it, . 22 because your position section is very short. 23 I don't guess there's any harm, but it is unusual 24 for a Reg Guide to be this much of a tutorial. 25 MR. MURPHY: That's right.
l i i 261 i 1 MR. SIESS: And the discussion which is the 2 tutorial part, is not keyed in any direct way -- well, the f 3 introduction is pretty long, too, if I'm not mistaken, i 4 MR. MURPHY: Yes, sir. , 5 MR. SIESS: They are not keyed too directly to the : 6 positions.
, 7 MR. MURPHY: Yes.
8 MR. SIESS: This thing is still in a state of 9 flux. You might consider, were there any implications of 10 that? I don't know that it's bad, excep.t that, again, I i 11 think you can revise a NUREG a lot faster than you can 12 devise a Reg Guide. 13 MR. MURPHY: Yes, but like you say, the 14 implications then are how much value or guidance is actually , i 15 in the NUREG in any kind of a regulatory sense, and how much ; 16 is in a Reg Guide? ! 17 MR. SIESS: But an awful lot of this is just 18 discussion, explanation, definitions of terms and all of 19 that, none of which is regulatory. A definition of a tern 20 is not a regulation until you refer to the term in the :
. 21 regulation. i 22 MR. MURPHY: Right. I 23 MR. SIESS: Anyway, it worth thinking about.
1 24 MR. MURPHY: And we will. I 25 MR. SIESS: Again, the weakness in our process is l i
262 1 the guy starts out and he doesn't know when to stop. He 1 2 thinks he knows when to stop, but when -- and in the early ) 3 days, that was what happened. 4 Every time somebody did the geology or seismology 5 for a site, ha had to go back and do some more and do some 6 more. One of the first things that got accomplished by 7 Appendix A was defining the scope of that work, and it held , 8 up pretty good, I think. 9 MR. MURPHY: Yes, sir. 10 MR. SIESS: I want to be sure that the same kind 11 of thing is still in there. Okay. Some of this sounds more 12 like Diablo canyon than we used to do. 13 MR. McNULLEN: Well it is an effort to include 14 what we're doing now. 15 MR. SIESS: But you see, what was done at Diablo 16 was for a plant whose design earthquake had gone from .4 to 17 .75 in a few years, and there were an awful lot of other 18 problems. That, to me, was a whole different than going out 19 and looking for a site to put a plant designed for .3G in 20 the Eastern U.S. where we have never had a size .3 G for any 21 custom plant. . 22 I'm approaching it from a different direction and 23 there will be no time, I think, in the Eastern U.S. where a 24 Diablo Canyon type of investigation would be justified, ] 25 unless you wanted to get very close to Charleston or New _, , -.. , , , , . . . - - - ~, _
l 263 1 Madrid or places like that. l 2 It will probably always be something like that on 3 the West Coast. l 4 MR. MURPHY: Right. i 5 MR. SIESS: Now, what was the one out in the 6 desert that got as far from San Andreas in the other j , 7 direction? 8 MR. CHOKSI: Sun Desert. 9 MR. SIESS: Of course, it had the advantage of ! 10 being internal. The mistake of Diablo was putting in on the . 11 coast with the whole Pacific Ocean out there to hide another 12 fault. But unless it's a site like that, you know, I think 13 you're going to take a pretty good investigation. 14 I just want to -- we don't want the Diablo Canyon 15 type of thing for a site in the Eastern U.S., unless it's in 16 a very bad spot. l 17 MR. McMULLEN: Their reason our investigation in 18 the Eastern U.S. may be just going back and looking at the 19 available data -- 20 MR. SIESS: It may be. There has been a lot of
, 21 good data out there.
22 MR. McMULLEN: Fill it in with specific 23 investigations, if they're necessary. 24 MR. SIESS: I'd have to go through this list to ! l [ 25 see if this requires more than would be available for a l
l 264 1 fairly good recent, wcll-studied site. For example, what's 2 TVA's western-most site? Isn't it Yellow Creek? 3 MR. McMULLEN: Bellefonte. 4 MR. MICHELSON: Further east. 5 MR. SIESS: Further east than what? 6 MR. MICHELSON: Than Yellow Creek. 1 7 MR. SIESS: Yellow Creek was the most western . 8 site? 9 MR. MICHELSON: Yes. 10 MR. SIESS: They did a terrific amount of geology 11 and seismology for Yellow Creek, trying to put an eastern 12 bound on New Madrid. 13 MR. MICHELSON: Yes. 14 MR. SIESS: I'd be interested, for example, in 15 whether I could take that Yellow Creek two or three volumes l 16 of geology and seismology and satisfy all of this. Its 17 almost in the extreme case, I think, but it's old enough, 18 you know, that it wouldn't meet some of our more recent 19 knowledge, maybe not even about New Madrid. I don't know. , 20 But that would be a fairly simple test of what you 21 just said, you know. . 22 MR. MURPHY: What more would be necessary? , 23 MR. SIESS: I mentioned earlier, are there some 24 sites out there that have been abandoned after the siting 25 was done? I don't know whether Cherokee and Perkins -- how
265 1 much were done on those? Those were getting over closer to 2 New Madrid -- I mean, to Charleston, weren't they? 3 MR. McMULLEN: They are in North Carolina. 4 MR. SIESS: They are in North Carolina. ! 5 MR. McMULLEN: They dug the excavations for one of 6 them, at least.
, 7 MR. SIESS: Yes. You know, one of them was partly 8 built. They've been making movies at it. That was the one 9 under water?
( 10 MR. WYLIE: Cherokee 1. j 11 MR. SIESS: What was the movie that was all under
; 12 water? They filled the containment half full of water and -
13 used it. ' 14 MR. MICHELSON: They used it for a movie set? 15 MR. SIESS: Yes. In fact, they sold it to 16 somebody and leased it out for a movie set. I 17 MR. MICHELSON: It was good for something. ' i 18 MR. SIESS: Okay, let's go on, 19 (Slide.] 20 MR. McMULLEN: This is a continuation of your
. 21 other slide with investigations.
22 I just wanted to point out that in the eastern and
.23 central U.S. the defining seismogenic sources is most 24 important because capable tectonic sources are very rare.
25 There have been two documented cases in eastr. n North
2b6 1 America, the Meers Fault and the surface displacement that 2 accompanied the Ungava earthquake two years ago up in 3 Quebec. 4 You always have to -- beg pardon? 5 MR. SIESS: If I'd picked a site 20-30 miles from 6 the nearest fault 20 years ago, would I have found the Meers 7 Fault as a part of my siting review? , , 8 MR. McMULLEN: If investigations were done to the 9 level of detail that Wolf creek was done and comanche Peak, 10 yes. t 11 MR. SIESS: Okay, and those are fairly recent i 12 plants. 13 MR. McMULLEN: Yes. 14 MR. SIESS: They would have picked it up. 15 MR. McMULLEN: Yes. 16 The heavy dependence on seismicity and geophysics 17 because the structures aren't at the surface. 18 MR. SIESS: What replaced tectonic province? 19 MR. McMULLEN: Seismogenic source. , 20 MR. SIESS: Seismogenic source. 21 MR. MURPHY: The seismic source. There wasn't a . 22 direct replacement for tectonic province.
- 23 MR. SIESS: The thing is province always had a ,
24 large area and source to me starts talking about a small, 25 but the seismic source could be a large? -
1 l 267 i 1 MR. MURPHY: Yes, sir. 1
) 2 MR. McMULLEN: If a fault is encountered, similar i
i 3 methods must be used just like in the' western U.S., although i l 4 we've learned through the years that in the eastern U.S. a l 5 fault is innocent until proven guilty and just the opposite j - 6 in the western U.S. 7 I think it was a lot of overkill in the past. For , i
; 8 example, Virgin Summer, many faults were found in the i 9 excavation there and each display of every fault is i
10 investigated in great' detail and as time,went on we started l 11 in other sites lumping similar faults together and just i 12 regarding it as a fault zone and just doing it once rsther 13 than the overkill that was done there. 4 14 MR. SIESS: Now the Mears Fault was detected fron l 15 surface features, was it not? 16 MR. McMULLEN: Yes. 17 MR. SIESS: I mean I guess it's not proven but was 18 suggested by surface features and then studied in more 19 detail to show that it really was active or had been active? 20 MR. McMULLEN: It's been shown to have been
, 21 active.
22 MR. SIESS: But that was done for more than just 23 surface features? 24 MR. McMULLEN: Well, it has been traced . 25 geophysically to -- I don't remember how deep --
2h8 1 MR. SIESS: Okay. 2 MR. MURPHY: It was trenched and other things to 3 look at to verify that it had movement on it within the 4 definition of a capable fault. 5 MR. SIESS: You know, I've had the theory for a 6 long time that nuclear plants attract faults. 7 MR. MURPHY: No, sir. I think it's seismometers , 8 attract faults. j 9 MR. SIESS: Oh. Well, without seismometers you 10 don't have earthquakes. 11 MR. MURPHY: That's true. 12 MR. SIESS: Don't have a lot of little ones. 13 MR. MURPHY: If Anna, Ohio is a case, putting 14 seismometers out stops earthquakes as well. 15 [ Laughter.) 16 MR. SIESS: That's frustrating to seismologists, 17 isn't it? 18 MR. NURPHY: Yes, sirl 19 [ Slide.] , 20 MR. McMULLEN: In the western U.S. the focus was 21 on both seismogenic sources and capable tectonic sources. . 22 There's more emphasis on the surface investigative 23 methods than in the eastern U.S., however there are V 24 unidentified sources in the western U.S. that don't come to 25 the surface, and we have talked about them already, the
269 N 1 blind faults and the plate interface. 2 MR. SIESS: Was Loma Prieta an unidentified 3 source? Or was it on a known -- 4 MR. McMULLEN: It's been pretty much characterized 5 by the after-shock patterns but I don't think there's been 6 any geophysical -- 7 MR. SIESS: Did they know it was there? 8 MR. MURPHY: Was it mapped beforehand is what he 9 is asking. 10 MR. McMULLEN: Oh -- 11 MR. SIESS: Is it on the maps? 12 MR. McMULLEN: The buried part, no. 13 The part that revealed itself in the after-shock 14 pattern -- it's a part of the San Andreas Fault and there's i 15 a bend there -- I l 16 MR. SIESS: I know, yes. i 17 MR. McMULLEN: The surface part of that was. 18 MR. SIESS: Even an area like that of the San i 19 Andreas wasn't as well known as some people thought it was? 20 MR. McMULLEN: No. , 21 MR. SIESS: Because of that hook in there. 22 MR. MURPHY: Yes, the Whittier Narrows was a 23 better example of one that wasn't know before the earthquake , l 24 occurred on it. 25 After the earthquake occurred on it they were able
l l 270 1 to get out and then map it and if you want to say, well, 2 maybe we should have seen it beforehand. 3 MR. SIESS: So it would just be safe for me to l 4 assume that anywhere in California I could have an i 5 earthquake, eh? l l i ! 6 I was thinking about moving out there but -- I 7 okay. 8 MR. McMULLEN: That is the last one I have. 9 MR. SIESS: You have another one here. 10 MR. McMULLEN: The area one? 11 MR. SIESS: Yes. I didn't see it. That's all 12 right. I've read it. 13 [ Slide.) 14 MR. SIESS: How :nch does that change from 200 15 miles? I don't remember anything like that in there before. i l 16 Was that just implicit? I 17 MR. McMULLEN: That's the distance in Appendix A. 18 It's the same -- 19 MR. SIESS: Does this thing go out 200 miles from , 20 the plant? 21 MR. McMULLEN: Yes. . ! 22 MR. SIESS: I didn't remember that. So, the 23 biggest changes then are simply -- in this part, are simply 24 in some terminology.
- - 25 MR. McMULLEN
- In this part?
271 1 MR. SIESS: In what you've presented? These 2 revisions to Appendix A are primarily in terminology?
~
3 MR. McMULLEN: Yes. 4 MR. MURPHY: Yes, sir. l 5 MR. SIESS: What I actually have to do is pretty l 6 auch the same? 7 MR. MURPHY: That was the intent. , 8 MR. SIESS: Yes. So, we have made some people l 9 happy who didn't like tectonic provinces and capable faults. 10 Okay. Thank you. l 11 .In this a good quitting point for today? ) 12 MR. MICHELSON: For the. day? What do you mean for i 13 the day? This will start again tomorrow, won't it? 14 MR. SIESS: No. I said for today. 15 On February 5th, which is the next meeting we've 16 scheduled, and I don't think we've got much chance to'do 17 anything before then, we would expect to continue with the 18 use of the probabilistic criteria and its application to 19 specific cases. 20 I think we'd be interested in reviewing a more
, 21 complete, more final draft of DG 1015, which, if I start 1
22 with the regulatory position, isn't an awful lot. And as I l 23 work back, I'm not sure whether I'm in a regulatory guide or 24 a tutorial. You see, that's what part of my point was. 25 We always get into a problem of what we were i 1 i
l l , i l 272 i 1 reviewing. And, if you will recall, in the old system which 2 we still operate in, I think, ACRS only endorsed the 3 regulatory position in the Reg Guide. For the rest of it, 4 we satisfied -- limited ourselves to correcting grammar, 5 rhetoric and things of that sort. But the ACRS endorsement , , 6 is only the position. And the position isn't that much. It 7 wouldn't be any problem. But I think we ought to have a . 8 look at the whole thing. So, those two things would be the 9 subject of the next meeting. Then I would try to go to the 10 full committee after that. 11 Now, the Full Committee in January is having a l 1 12 couple of day meeting on general siting source. term stuff, l 13 in which the Appendix A, B, S is part of the issue. It's l 14 separable but still part. ! 15 I would~ recommend that you -- I don't know what 16 your plans are, but if you can go out -- you want to go 17 public with the Reg Guide 1016, 17 and 18, I would have no 18 problem with that, and I don't think anybody on th e i 19 Committee would. , 20 MR. MICHELSON: No. 21 MR. SIESS: If you'd like a letter from us saying
- 22 those are okay to go out for public comment. As far as I'm 23 concerned, Appendix S and Appendix B are okay to go out.'
24 The only thing we haven't really looked at a lot is this 25 thjng here.
273 1 Now, what is in Appendix S is the OBE -- but , 2 again, I think this is close enough -- I'm satisfied enough 3 with that to see it go out for public comment. I think 4 you're going to get some comment, and it may be useful. So, l 5 would you like a letter saying that we have looked at things 6 and are satisfied for them to go up to the next stage, for 7 the CRGR public comment? l 8 MR. MURPHY: We would appreciate that. 9 MR. SIESS: And we'd like a chance to see them ; 10 after the public comment. 11 MR. MURPHY: That's the intent for sure. 12 MR. SIESS: In the meantime, we'll continue our 13 review of the DG 1015. 14 MR. MURPHY: Okay. It sounds good at this end. 15 MR. SIESS: We'll do what we can. This is the 16 darndest mess of a bookkeeping, to make this simple change.
- 17 We would always appreciate comments from the i
18 industry, either individual utilities, NUMARC, EPRI, what 19- have you. I understood that they were not prepared to 20 present any comments at this meeting, because they had not
, , 21 seen scue parts of this. And I didn't try to decide to find 22 out which parts they have and haven't seen. But they will 23 have had a chance to see everything except possibly DG 1015 24 by the February meeting, and probably would have seen all of 25 that. I think they have been involved in some of that.
, . _ _ ~ -- --r --
274 1- So, we would invite industry people to come in at 2 that time with any comments they have. If they have made 3 comments as part of public comments -- public comment would 4 not be finished by then. of course, when we look at stuff 5 after public comments, then obviously people can come in. , 6 MR. MURPHY: Procedurally, I'm not sure what's 7 going to happen next. But, as I understand it, this will . 8 not go out for public comment until after we've gone through 9 CRGR and the folks in Warren Minner's organization are 10 controlling this -- are holding them as a unit at this time. 11 So, they may not be out -- probably will not be out for 12 public comments -- or the public comment period -- 13 MR. SIESS: Until after you've finished 1015? 14 MR. MURPHY: Yes, sir. 15 MR. SIESS: I don't know why some -- they're just 16 giving people a bigger job. An awful lot of this is 17 reviewable without seeing 1015. 18 MR. MURPHY: Yes, sir. 19 MR. SIESS: Just like we did. But that's up to , 20 you. We'd like to come back, and, you know, we'll come back 21 and talk about this whenever you're ready. - 22 MR. MURPHY: Okay. 23 MR. SIESS: Now, if CRGR has any major changes, we 24- always like to know what they've come up with. . 25 MR. MURPHY: Yes, sir. I
1 275 1 MR. SIESS: Good or bad. 2 MR. MURPHY: We understand that. 3 MR. SIESS: Okay? Anything else? ; 4 MR. MURPHY: Then we'll see you on the fifth to 5 talk about the Reg Guide? 6 MR. SIESS: I hope so. 7 [Whereupon, at 5:11 o' clock p.m., the above-8 entitled meeting was adjourned.) 9 10 11 1 l 12 13 14 15 16 17 18 19
- 20 I
, 21 i 22 23 , 24
.25
REPORTER'S CERTIFICATE This is to certify that the attached proceed-ings before the United States Nuclear Regulatory Commission in the matter of i l NAME OF PROCEEDING: Extreme External Phenomena DOCKET NUMBER: PLACE OF PROCEEDING: Bethesda, Maryland were held as herein appears, and that this is - the original transcript thereof for the file of the United States Nuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court report- ) ing company, and that the transcript is a true I and accurate record of the foregoing proceedings. f)tt ib J l Official Reporter Ann Riley & Associates. Ltd. 9 l 4
I OVERVIEW APPENDIX A TO 10 CFR PART 100 REVISION l l PRESENTED TO THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS EXTREME EXTERNAL PHENOMENA SUBCOMMITTEE
. DECEMBER 10,1991 I
ROBERT J. BOSNAK, DEPUTY DIRECTOR l DIVISION OF ENGINEERING 1 U.S. NUCLEAR REGULATORY COMMISSION l OFFICE OF NUCLEAR REGULATORY RESEARCH l (301) 492-3850 !
i : l BACKGROUND i THE BASES FOR APPENDIX A TO PART 100 WERE I ESTABLISHED IN THE LATE 1960's i i 4 ! THE REGULATION BECAME EFFECTIVE ON DECEMBER 13,1973 ADVANCES IN THE SCIENCES OF SEISMOLOGY AND GEOLOGY, ALONG WITH THE OCCURRENCE OF SOME ISSUES IN LICENSING NOT FORESEEN IN THE DEVELOPMENT OF APPENDIX A HAVE CREATED SOME DIFFICULTIES IN THE APPLICATION OF THIS REGULATION 4
. . _ - _ ~ .. . . _-.. _. . . . - - .._ - . _ _.. - . - - .-._ - . . . - _ - . . _ . . . _ - - - - - _ .
PHILOSOPHY , l THE REVISED CRITERIA PRESENTED IN THE PROPOSED ! REGULATION W3LL NOT BE APPLIED TO EXISTING PLANTS l l i
- )
l l THE PROPOSED REVISED CRITERIA ON SEISMIC AND GEOLOGIC SITING WOULD BE DESIGNATED AS A NEW t APPENDIX B TO PART 100 l j l CRITERIA NOT ASSOCIATED WITH THE SELECTION OF THE. SITE OR ESTABLISHMENT OF THE SAFE SHUTDOWN EARTHQUAKE GROUND. MOTION HAVE BEEN MACED INTO l l PROPOSED APPENDIX S TO PART 50 l' l l l _ _ _ _ _ _ _ _ _ _ _ _ __ _. ___
---- -. .- . ~ . . . . - . _ - . ~ - . - _ . . . -... . _ -
! ORDER OF PRESENTATIONS APPENDIX B - OVERALL APPROACH MURPHY USE OF PROBABILISTIC CRITERION CHOKSHI i . APPLICATION OF NEW APPROACH BERNREUTER l } GEOLOGIC INVESTIGATIONS McMULLEN GROUND MOTION SOBEL 1 ^ ENGINEERING CRITERIA KENNEALLY i OBE/SSE INVESTIGATIONS WITTE l ENGINEERING REG GUIDES KENNEALLY 4
}
-__ -. . . - . - .=. .
l l OVERVIEW OF APPENDIX A REVISION PRESENTED TO THE EXTREME EXTERNAL PHENOMENA SUBCOMMITTEE ADVISORY COMMITTEE ON REACTOR SAFETY
=y
[ l
\..../
December 10,1991 Andrew J. Murphy, Chief Structural and Seismic Engineering Branch Division of Engineering Office of Nuclear Regulatory Research 301-492-3860 O e
'l REVISION OF APPENDIX A TO 10 CFR PART 100 LANTS 9 ISSUE e WHILE THE CURRENT APPENDIX A HAS GENERALLY BEEN SUCCESSFUL IN THAT THE MAJORITY OF CURRENT NPPs WERE LICENSED USING IT, THERE HAVE BEEN DIFFICULTIES.
e APPENDIX A TOO DETAILED, INFLEXIBLE, AND LACK OF CLARITY IN SOME SECTIONS LEADING TO CONFLICTING INTERPRETATIONS. e CONFLICTING INTERPRETATIONS OF APPENDIX A OFTEN LED TO TIME CONSUMING DISCUSSIONS AND ADJUDICATION BY THE LICENSING PANELS t e CURRENT APPENDIX A ISSUED IN 1973 DOES NOT REFLECT ADVANCES IN THE SCIENCES OF SEISMOLOGY AND GEOLOGY (FOR EXAMPLE, PALEOSEISMIC DISCOVERIES AND PROBABluSTIC HAZARD ANALYSIS), AND DOES NOT REFLECT EVOLUTION OF THE LICENSING PROCESS THROUGH ADJUDICATORY PROCESS. e THE MULTIPLE DEFINITION OF THE OBE (FUNCTIONALITY, LIKEllHOOD OF OCCURRENCE, AND MINIMUM FRACTION OF SSE) HAVE, IN SOME CASES, RESULTED IN THE OBE CONTROLLING DESIGN. 0 9 O O
REVISION OF APPENDIX A (CONT'D1 i e IDENTIFICATION OF NEED FOR REVISION OF APPENDIX A e " GEOLOGICAL AND SEISMIC SITING POLICY AND PRACTICE FOR NUCLEAR POWER PLANTS", SECY-77-288A, AUGUST,1977. e "lDENTIFICATION OF ISSUES PERTAINING TO SEISMIC AND GEOLOGICAL SITING REGULATION, POLICY AND PRACTICE FOR NUCLEAR POWER PLANTS", SECY-79-300, APRIL,1979. ' e "
SUMMARY
REPORT OF THE SYMPOSIUM ON SEISMIC AND GEOLOGIC l SITING CRITERIA FOR NUCLEAR POWER PLANTS", NUREGICP-OO87, OCTOBER,1986.
# USER REQUESTS e MEMORAND.UM FROM NRR (GILLESPIE) TO RES (BECKJORD) DATED AUGUST 2,1990.
e MEMORANDUM FROM EDO TO RES DATED SEPTEMBER 6,1990 APPROVING RULEMAKING ACTIVITIES.
- HIGH PRIORITY STATUS.
- REVICdD REGULATION GENERAL IN NATURE, DETAILS IN REG. GUIDES.,
- SITING / DESIGN DECOUPLING.
i
l 1 t l REVISION OF APPENDlX A (CONT'D) l 1 9 OBJECTIVES OF THE REVISION e REMOVE SOURCES OF CURRENT MISINTERPRETATION e INCREASE THE EASE FOR UPDATING THE TECHNICAL GUIDANCE e PROVIDE STABILITY IN LICENSE REVIEWS i o COMPLETE REVISION FOR EARLY SITE REVIEWS f
i
~
I [ REVISION OF APPENDIX A (CONT'D) f O SCOPE ; e REVISE APPENDIX A (NEW APPENDIX B), NEW APPENDIX S TO PART 50 TO DECOUPLE DESIGN FROM SITING. l 1 CREATE NEW GUIDE, IDENTIFICATION AND CHARACTERIZATION OF SEISMIC I SOURCES WITH PROBABILISTIC ACCEPTANCE CRITERIA. l e REVISE SRP SECTION 2.5.2, VIBRATORY GROUND MOTION. ; e REVISE GUIDE ON NUCLEAR POWER PLANT INSTRUMENTATION FOR l EARTHQUAKES. ! l
- DEVELOP TWO NEW GUIDES: PRE-EARTHQUAKE PLANNING AND RESTART !
FOLLOWING AN EARTHQUAKE. i e REVISIT EXISTING REG. GUIDES; IDENTIFY THOSE REQUIRING TECHNICAL l AND EDITORIAL CHANGES. ? i SUMMARIZING: ! o EARTH SCIENCES - 1 r.EG. GUIDE & 1 S.R.P. SECTION i e ENGINEERING - 1 REVISED REG. GUIDE & 2 NEW REGJ GUIDES ,, t
REVISION OF APPENDIX A (CONT'D) O APPLICABILITY t e NEW NUCLEAR POWER PLANTS e USE FOR RESOLUTION OF ISSUES AT CURRENTLY OPERATING PLANTS ON A CASE-BY-CASE BASIS I t 5 7
. s . .
i i REVISION OF APPENDIX A (CONT'D) O SCHEDULING l e SCHEDULE INTEGRAL WITH THE REVISION OF PART 100 TO SUPPORT EARLY LICENSE REVIEW. e MEET WITH THE ACRS SUBCOMMITTEE ON DECEMBER 10,1991 AND AS NEEDED. e REVISIONS WILL BE ISSUED FOR PUBLIC COMMENTS IN SPRING 1992. i li e FINAL RULE TO BE PUBLISHED BY EARLY 1993. l l
# COORDINATION ,
I e A REVIEW PANEL CONSISTING OF EMINENT EARTH SCIENTISTS AND ' ENGINEERS REVIEWING WORK IN PROGRESS, THREE MEETINGS SO FAR. e TWO PUBLIC MEETINGS PRINCIPALLY WITH NUMARC AND EPRI. NUMARC HAS ESTABLISHED AN AD-HOC GROUP. ! e REVISIONS ARE COORDINATED WITH OGC, NRR, EDO, AND PART 100 f ACTIVITIES.
- t 1
d EXPERT TECHNICAL PANEL Robert Budnitz Future Resources Assoc. Kevin Coppersmith Geomatrix Allin Cornell Stanford University James Devine U.S. Geological Curvey Walter Hays U.S. Geological Survey Robert Kennedy SMA Paul Pomeroy Rondout Assoc. i
SITE PROBABILISTIC DETERMINISTIC ANALYSIS ANALYSIS Geologloal, Selsmological and Investigatione I
'" "* "" Identify Selsm!c LLNL Setemic Hazard Sources Assessment Compare to Operating Determine Deterministic Plants to Set Earthquake for ProbablIIty of Each Source Exceedence Level Determine Probabilistic Determine Deterr-inistic Controllistg Earthquakes Controlling Eartisquakes to Obtain Their Es & Ds Ms and Da i i l
i Compare I Compute SSE Ground Motion
APPENDIX A TO DG-1015 PROBABILISTIC CONSIDERATIONS IN ESTIMATES OF VIBRATORY GROUND MOTION p i j PRESENTED TO THE
- ADVISORY COMMITTEE ON REACTOR SAFEGUARDS l EXTREME EXTERNAL PHENOMENA SUBCOMMITTEE l ,
DECEMBER 10,1991 1 3 i, NILESH C. CHOKSHI U.S. NUCLEAR REGULATORY COMMISSION DIVISION OF ENGINEERING OFFICE OF NUCLEAR REGULATORY RESEARCH (301) 492-3816 4
PROPOSED APPENDlX B REQUIREMENTS , 9 "EACH APPLICANT-----SHALL INVESTIGATE ALL SEISMIC AND GEOLOGIC FACTORS THAT MAY AFFECT THE DESIGN AND OPERATION OF THE PROPOSED NUCLEAR --
- . BOTH DETERMINISTIC AND PROBABILISTIC EVALUATIONS SHALL BE CONDUCTED." (ll)
O "THE PROBABILITY OF EXCEEDING THE SAFE SHUTDOWN EARTHQUAKE GROUND MOTION IS CONSIDERED ACCEPTABLY LOW IF IT COMPARES . FAVORABLY (THAT IS, SIMILAR TO THAT SHOWN FOR THE LOWER HALF OF THE POPULATION) TO THAT AT OPERATING NUCLEAR POWER PLANTS." (OPERATING . PLANTS AS OF THE EFFECTIVE DATE OF THIS REGULATION). (V.(c)) l l l 1 l ~ l l l l l 1
i b l
- l l
e l 1 i j OBJECTIVES OF APPENDlX - l
- i l
] 9 THIS APPENDIX TO DG-1015 OUTLINES A PROCEDURE ,
- TO' CALCULATE THE PROBABILITY OF EXCEEDING THE ,
I j DESIGN BASIS, AND ESTABLISHES ACCEPTANCE ! CRITERION. O INITIAL SCREENING OF A SITE. l l 9 ESTABLISH CONTROLLING EARTHQUAKES.
~
! O ESTABLISH POSITION ON THE PROBABILISTIC HAZARD ! TO BE USED IN A PRA. i l OBJECTIVES OF THE PROBABILISTIC HAZARD ANALYSIS S TAKE INTO ACCOUNT UNCERTAINTIES IN VARIOUS < FACTORS INVOLVED IN ESTIMATING GROUND MOTION. 9 IDENTIFY SIGNIFICANT CONTRIBUTORS IN TERMS OF l MAGNITUDES AND DISTANCES TO THE HAZARD. 9 DEMONSTRATE THAT THE PROBABILITY OF EXCEEDING - I THE DESIGN BASIS COMPARES FAVORABLY WITH ; THOSE OF OPERATING PLANTS. , i O HAZARD ESTIMATES TO BE USED IN A PRA. 2
PROCEDURE 9 EASTERN U.S. USE OF HAZARD AS A RELATIVE MEASURE. CONSISTENT CALCULATIONS FOR ALL SITES. ~ PROCEDURE HAS BEEN REVIEWED BY THE EXPERT REVIEW PANEL. STEP 1. CALCULATE UNIFORM HAZARD SPECTRA FOR THE : SITE. STEP 2. (A) ESTIMATE THE ANNUAL PROBABILITIES OF , ! EXCEEDING THE DESIGN SPECTRUM AT TWO l DISCRETE FREQUENCIES. j (B) CALCULATE THE COMPOSITE ANNUAL PROBABILITY OF EXCEEDANCE. , j (C) COMPARE COMPOSITE PROBABILITY OF i EXCEEDING THE DESIGN BASIS WITH THOSE OF CURRENTLY OPERATING EUS PLANTS (COMPARE WITH LIMITS GIVEN IN APPENDIX). l l l S WESTERN U.S. A SITE-SPECIFIC HAZARD ANALYSIS USING SUITABLE METHODOLOGIES SHOULD BE CARRIED OUT. NO SPECIFIC CRITERIA SUCH AS THOSE FOR EASTERN U.S. SITES. 3
1 I l PROCEDURE TO COMPUTE PROBABILITY OF EXCEEDING l DESIGN BASIS ' l COMP. PROB. = 1/2
- a1 + 1/2
- a2 ,
l SM . 1000. 2000. . 5000.. 10000. YtArs RtiURN PCR100 o w w a5 W( -
#\
h" _ "O~ ~ ~ E$ ua
- - u3 .
g _ .u 3 4 d e
., , . . - . . . , . . - ,. e . . -
~
7g 73 etasoo (ste)*e e 4 6 4
l PROBABILITY OF EXCEED.ING DESIGN BASIS OF CURRENT PLANTS ;
. i based on 5 g-experts freq= 7 eq. level = 1 probilistic metric
- 1. 0 ,
ordered using med, log, haz. :
.9 -
C - . O - -
.e -
.o : :
*: .7 - -
m . . o .6 -
.I
- 3 -
LIMIT 1E-4 _ C3 - , - E - -
= .4 - -
o : :
.3 .- -
.2 -
.1 -
. . . . i , . . , i . . , , i
[ '
- 0. u n we a u n we , u n we n 8 I I Da3 H Ee3 l
prob. of exceedance e 5 -
PROBABILITY OF EXCEEDING NUREG/CR-0098 SPECTRA , ANCHORED TO 0.3G FOR CURRENT SITES l I 1 based on 5 g-experts ' freq= 7 eq. level = 3 probilistic metric
- 1. 0 . . . . . . .
.9 - ordered using med. log. haz :
C - . O - .
.e .- .
o : y - : u . _ v, . . c .6 .-
= : - .5 .- .
c . . E - .
= .4 -
o :
.3 .-
"~ .
! 2
. 7
.1 -
i
- 0. ~
o . .
. . . . i i
. ~ o .. ,
i : w i ! w - i prob. of exceedonce l l 1 4 2 6
l REFINEMENTS OF CRITERIA G TRIAL APPLICATIONS BEING PERFORMED FOR SEVERAL - SITES. S SENSITIVITY STUDIES BEING CARRIED OUT TO ; EVALUATE SOME OF THE FOLLOWING PARAMETERS: e SENSITIVITY TO VARIOUS STATISTICAL MEASURES. o SENSITIVITY TO PGA AND RESPONSE SPECTRAL 4 ORDINATES. 7
- , j NMP A NuclearSystemsSafetyProgram E r
1 I APPLICATION OF NEW APPROACH i i i i i Presented by l i Don L. Bernreuter i Lawrence Livermore National Laboratory FTS 532-0305 (510) 422-0305 j i ACRS Meeting on Revision of Appendix A Meeting j l Nuclear Regulatory Commission Bethesda, MD i December 10, 1991 .. EGP '8DLB-1 i
i l t
Background
i Both the LLNL and EPRI studies show that there is considerable uncertainty in defining seismic sources (geometry, recurrence relations and EMES). i On-going world-wide studies of tectonically analogous regions indicate that the postulated maximum earthquake in many , seismogenic sources may be larger than used in past licensing decisions. New proposed rule allows for the explicit inclusion of uncertainty and alternative models. Such factors can only be implicitly incorporated into the current procedure. . t i EG91-078DLB-2
The deterministic path will } Remain similar to current NRC practices. j Implementation simplified because uncertainty is primarily incorporated in the probabilistic path.
- Seeks to determine the most likely " worst" case.
. Used in conjunction with the probabilistic path.
I EG' 3DLB-3
The probabilistic path actually has two paths l
. The completely probabilistic path - Typically the LLNL and EPRI results. This was described by the previous speaker.
. Outcome of the probabilistic analysis hazard curves and uniform hazard spectra at different p.eriods.
. The path leading to the selection of Probabilistic Controlling Earthquakes (PCEs).
l .. 1 EG91-07BDLB4
The path defining the PCEs is introduced 1 for several reasons ; Identifies significant sources of earthquakes for comparison with the
- deterministic earthquake. -
Makes use of the probabilistic data and uncertainties to gain , useful insights. i
. Because it is " relative" It is not dependent on the " absolute correctness" of many of the parameters used in the model.
I t 4 9 ego" '78DLB-5 1
Concept of the Probabilistic Controlling Earthquake (PCE)
- 1. As previously described using the cumulative distribution of exceeding the SSE (e.g., Sv at 5 hz) for a given hazard estimator (e.g., mean) of all NPP, select appropriate probability of exceedance (PE) for the PSSE at the 0.5 level.
Jk .
- 1. ------------
l i l l
= .
I 1 O O PE Probability of Exceedance for Estimator EG91-078DLB4 i ___-_______-________________-__________-_-_________________________-_______________--_--_-___I
. . . - - - - - ~ . . . - ..w -.-n .a-- .- sx -----....a._
- -. . - . - . . . - . - . . - - - . ~ _ . . . . - - - - - - - - _ _ ,
.=
'i,-
Y t 1 b l N n-e vS ) v d
- W r f .
4 V v rn n C m m 4 w v E v
! w o
o i i p # 2 V e w e s a = f g :: In 7 W cw y
~ vt -
II 1A ~ 9 l 6 r Al x h 2 9
% 4 -
MI m@ ~> gl t.n ' Y s D mi o o 1 El A d ! E
Concept of the Probabilistic Controlling Earthquake (PCE) (continued)
- 2. Enter the hazard curve (at the PE value) and obtain the GM value.
i i 8 m 8 i 8 N 3 y
~
a io k I Ground Motion
- 3. The PCE are those earthquakes (in terms of magnitude and distance) l which most significantly contribute to the hazard at the above determined GM level.
l ' l l EG91-078DLB-7
1 i l If there were only a single seismicity model (with , uncertainties) the problem of determining the PCE l
- would be relatively simple and straight forward.
However, this is not the case as the LLNL study has i t 11 Seismicity experts, each with multiple alternatives 4 5 Ground Motion experts with multiple models ;
- Several possible hazard estimators (median 85%, mean) j i
EC '78DLB-8 l
The procedure for determining the PCE for a site In order to account for the relative likelihood provided by each expert and to appropriately weight the input of any given expert, we have assumed that we want to use the results obtained by aggregating over all experts and scenarios.
- 1. To do this we developed the contribution to the hazard for each S-expert and all G-experts in 16 bins so that the hazard surface could be examined as a function of magnitude (M) and distance (D) at the ground motion level previously determined.
i 5 to 5.5 5.5 to 6 6 to 6.5 l > 6.5 ; l 0 - 25 t i 25 - 50 ! l 50 - 100 i
> 100 l
-. j i
EG91-078DLB-9 i
i The procedure for determining the PCE for a site f (continued) t
- 2. Combined over all S and G-experts for each bin to compute the relative hazard frorr- each bin. (This gives 16 hazard curves for each i estimator, e.g. median).
- 3. Enter each hazard curve at the correct GM value previously i determined to obtain relative contribution for that bin to the PCE ground motion.
h _ \
\
J \ _ Bin 1 Bin 3 E Bin 4
/';
Ground Motion EG' 78DLB-10
i 1 i [ The procedure for determining the PCE for a site (continued) ' f i
- 4. The bins which contribute the most define the PCEs for the site (note that we get sets for the median, mean and Ba ; hazard estimators, also for each GM parameter (PGA, Sv) used.
__ t
- 5. We can also compute a M, D value by using the hazard level in each bin as the appropriate weight. We call this the scenario PCE. ,
i I t i j EG91-078DLB-11
We applied this approach to 4 existing NPP sites as
- a trial using PGA hazard and 5 hz hazard curves to select the appropriate GM level to use i
One site in New England (NE) i One site that might be affected by the Charleston earthquake (SE) One site in the Southern Central EUS (SCUS) (away from both Charleston and New Madrid)
. One site in CUS (away from New Madrid)
EC 78DLB-12
For each of the 4 test sites we
- 1. Determined the ground motion level to use for PGA and 5 hz spectral velocity based on the median,85%, and mean hazard estimates (6 cases).
- 2. Computed the hazard curves ~ for the 16 bins for the 6 cases in (1).
- 3. Using the GM levels determined in (1) and the hazard curves in (2), we developed a matrix of relative contributions which we used to select 6 sets of PCEs.
- 4. Computed M, D values for the 6 cases.
EG91-078DLB-13
, Matrix of relative contribution to the hazard (based on PGA) for NE site McLa n . ; 6 to 6.5 > 6.5 ! 5 to 5.5 5.5 to 6~ o9 i. .s s o. M 5.c O - 25 o oE 0- 83 ' 55 '8 25 - 50 i 50 - 100 oo 0#
- Oo3 o-
=
o.o o- ' O'
> 100 Mean 5 to 5.5 5.5 to 6 6 to 6.5 > 6.5
8 3' ' g: s.6 O .25 098+ -
~
- o. s s o- " 5: 46 25 - 50 o .8 I i.
- o. 44 .vs 7 o se a o.oss 50 - 100
- 0. a is .57I .c 5 o. m i > 100 ,
EG91-078DLB-14
- y
Matrix of relative contribution to the hazard _ (based on 5 hz Sv) for NE site Mea ia n. 5 to 5.5 5.5 to ~6 6 to 6.5 -
> 6.5 -
,a r. .s3 o.o p1= 5.9 0 - 25 25 - 50 o oo3 O' 'N oo 5"N-
- O- + o-
50 - 100 o.o
> 100 o- -
00 o. o
~
tAea n 5 to 5.5 5.5 to 6 6 to 6.5 > 6.5
~~
O - 25 o 76 69 'T E = s .9
.o o.4G '3; 25 - 50 o 4' B = st
- o. ss o.86 o,1, 50 - 100 0% r o S' o.76 05l O.25
> 100 EG91-078DLB-18
. Summary of results for the NE '
Bins with most significant Scenario M. D contribution to the hazard
~5ssed on Based on Based on Estimator PGA Elu Based pn PGA 5 hz M 6 E D M D M D '
Median '7 5.G 18 5.9 t9 5.7s 57s '7 S.z s tt c.zs '7 Mean 5.15 38 5,7s 38 E6 46 59 51 5 75 17 5,25 17 s.zs ,8 5.s 't 5 75 iso 1s iT
- s,7s- 78 i
h h
Summary of results for the SE E Bins with most significant Scenario M D contribution to the hazard Based on Based on Based on Estimator PGA 8 hr Based on PGA 5 hz ~ M B M D M D M D Median ' 0 E' ' E G g, g y9 5. S 3: 5.25
- g. z s n Mean g.,5 ,g o c .z s ugo c,4 lot g ,4 gg c.75 <7 (o.75 28
(
L I i Summary of results for the SCUS
,_. _ Bins with most significant Scenerlo M, D contribution to the hazard Based on Based on Based on Based on l Estimator PG A Elu PGA 5 hz E B E D M D M D
- 5. 25 17 5 ~75 '1 Median 5.(o 9 5,8 i3 5.15 I*1 i
61 43 g .g. 47 6.75 15o 6,75 87 Mean G.75 36 6 75 150 38
~
5 15 .
- 5 , 13 I"7 >
9 0
Summary of results for the CUS Bins with most significant i Scenario M, D contribution to the hazard - Based on Based on Based on Based pn ' Estimator PGA 5lu PGA 5 hz N D E D M !D M D T. M '1 5.~75 '7 Median 54 :r. 3 5. G so 87
- 5. z.5 i
y a5 87 S.75 85 o Mean 5,3 78 g,q s5 5.15 38 5. 15 'T . 5 75 8 ^7 S. 75 38 ! G S '5* 5 25 38 l' 5 . .es s .75 78 i
*. j I
a Just to see the implication of what types of GM levels these PCEs might lead to relative to current SSEs t Developed GM estimates using the BE GM models provided by LLNL G-experts. 1 Computed the average GM at the median level of GM and of the 1- i sigma level of GM. Compared these estimates to the design value EG91-078DLB-26 I i P
L ho,j. g c,oc i I ' 3*') m a- 65 Gt1 a.dek Co wW EO CSE (D) - NE Shz med r=19 indy. models model l used: 52 37 12 12 27 magnitudes plotled 5.90 E+3 . . . , 5- "
. 4 3- .
2- . E+2 - 4 5- . 4- . 3 . 2 - o E+1 - N 5. e - u 4 . x 3- .
- 2 -
u O
= E+0 -
5- . 1 4 . ! 3- . l 2- . I i {_g a a e a l e e e e I e e , , n u n -o - ~ n no , u n ,o I l + (
- M M En3 l
e period (sec) l \ i e
g - _3
< n.a .m c>n -.a S= a** ( l'E'S ei Gn he&A-NE Shz median r=19 model J used: 52 37 12 12 27 D ::- SSE De.si9 et magnliudes plolled 5.90 E+3 . .,
4-3- 2 - E+2 - b 5- - 4
- 3. .
e e
- g. a a .
u- u o E+1 - M iu s 4 x 3 - _~ 2 . o . I
- E+0 -
6- - 4 .
- 3. .
2- - E-1, 4 A 4a 1 4 A 4a 1 4 A 4a
. l M M M d period (sec) l l
NE Shz mean r=51 model f used: 52 37 12 12 27 magnitudes plolled 5.90 E+3 . . . . , . . . . 5 - . 4 . 3 . 2 - - E+2 - - M ' 5- - 4 . 3- -
- 2. .
o E+1 -
- e. c.
O M N 5- " " - E -
= m o 4 - , 3. .
C 2- - o O
. E+0 - -
6- - 4 - 3 - - 2 - - E-1 u n' .' u' n' n' o' ~ n' , u' e' n a'
- l l ->
M M M
- e period (sec) [
1 1 l l l 1 1 l l SE Shz median r=31 l I model f used: 52 37 12 12 27 magnitudes plotted 5.90 l E+3 , , , ,
, , , , , l
- 5 - i 4- .
3- . 2- . E+2 - 5- . 4 . ! 3 . 2- . e e a J E+1 -
,, i, -
- u. m ,
bl
\ $.
n - u 4- , . m 3- .
- 2- .
o O
- E+0 -
s- . 1 4- - ; 3- . i i 2- . gg n u n ne _. I
~
n na , t u n na I 1 4= M M M j a period (sec) l 1 i 1 1 i [
4 \ i l 1 N SE Shz mean r=62 l model J used: 52 37 12 12 27 magnitudes ploited 6.40 E+3 , . . , 5- . 4 - 3 . 2- . !* E+2 - 5- - 4- - 3- - 2- 3 S - ! o o , u E+1 -
=
i M l \ 0-i . m " u 4- - x 3- - u 2- . i o i
- E+0 -
i 5- - 4 -
- 3. .
2- . g_3 u n
-o 1
u
. . .. I . . . .
n , n ne a u n no , O 0-b s.a s.Ia s.a l I 1 1 ; l i
- I period (sec) l 1
f I i
SCUS Shz med r=19 model f used: 52 37 12 12 27 magnitudes plolled 5.80 E+3 ' ' ' ' s . ... , , 5-4- 3- , t- . E+2 - 5-3- n . 2- C
" e a .
E+1 -
.o .u,.
tts M -.
) o 5-4 -
x 3- -
~
o 2. . O
; E+0 -
5-4- 3- . 2- . E-1 " n
? 4 A 4a I
~
m , o o .. Saa + M e Period (sec)
.s
1 1 l SCUS Shz mean r=47 model f used: 52 37 12 12 27 magnitudes plofied 6.20 E+3 . , , , 5- _ 4 . 3- . 2- . , E+2 - [ l l 1 i: 3-l l 2- e e . a a . I o E+1 - o u l
. - ]
OPB i N 5-i E - I o 4 - 5 3- .
- 2- .
u O , l I- E+0 - 5- . 4- . 3- . t- . i l E-1 n
' ' ' ' ' ' ' ' - - I
~ n' ** ~ n' <= , u' n' .= !
I e J M M M j a period (sec) , l l i k i
r CUS Shz mean r=55 model f used: 52 37 12 12 27 magnitudes plolled 5.90 E+3 , , , , 1 5- " 4- ' ' 3- -
\
- g. ;
. i e i E+2 -
1 l l 5 - 4 - n 4 ' 3 - 2- . l o E+1 - e e _
. e s . M 1 :: " e u :
x 3 - o-2- . O
= E+0 -
6 - 4- . l 3- . t- . I E-1 n
, a' l n a' n'
~ ~ n' n e' , u' n' l I +
W M M 4 a period (sec)
CUS Shz med r=20 model f used: 52 37 12 12 27 mognitudes plolled 5.60 E+3 . . . . , . . . . . . , . 5- -
, 4- -
- 3. .
2- - a , E+2 - 4 4 ." . 3- .
- g. .
e e
~ "
o E+1 - e Li a ng N m a 5- - u 4 - x 3- - _~ 2 . o . O I, E+0 - 5 .
- 4. .
- 3. .
g.
. . . . l t E-1 , u n , .a
_ u n
,o
, u n
,o I a +
- M M M e
period (sec)
REVISION OF APPENDIX A 10 CFR PART 100 GE0 LOGICAL, SEISMOLOGICAL AND GEOPHYSICAL INVESTIGATIONS 1 PRESENTED TO THE ACRS SUBCOMMITTEE ON EXTREME EXTERNAL PHENOMENA DECEMBER 10, 1991 R. B. MCMULLEN, GEOLOGIST STRUCTURAL AND SEISMIC ENGINEERING BRANCH DIVISION OF ENGINEERING ! l 0FFICE OF NUCLEAR REGULATORY RESEARCH (301) 492-3808 12-3.RMC * ' , . [
l I u 1 DRAFT REGULATORY GUIDE w 2 DG-1015 - I u
. 3 IDENTIFICATION AND CHARACTERIZATION l E !
O 4 0F l 5 SEISMIC SOURCES AND DESIGN GROUND MOTION 6 TABLE OF CONTENTS , 7 7 Section Title Pace No. 8 A. Introduction............................................. 1 s 9 8. Discussion............................................... 12 10 1. Purpose of Seismic Sources.......................'.. 12 11 2. Investigations to Identify Seismic Sources.........14 l 12 a. General...................................... 14 13 b. Reconnaissances and Other Sources of Preliminary l 14 Information.................................. 16 l
> 1
- 15 c. Detailed Investigations...................... 16 5
16 d. Distinction Between Tectonic and Nontectonic 17 De fo rmat i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 1 18 3. Identifying and Determining the Geometry of Seismic i l 19 Sources............................................ 25 3 20 4. Estimation of the Expected Maximum Earthquake and the -
'21 Deterministic Earthquake........................... 29 l 22 a. Stabl e Conti nent Regi on. . . . . . . . . . . . . . . . . . . . . . 31 I
! 23 b. Active Plate Margin Region................... 32
- 24 5. Consideration of Recurrence Models for Western United 5
i 25 States Sites....................................... 35
- 26 C. Regul a t o ry Po s i t i o n . . . . .'. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 9 27 D. -I mpl eme n t a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 4 r%)
I l I THE PRIMARY GOALS OF APPENDIX B TO 10 CFR PART 100 ARE: TO IDENTIFY, DEFINE AND CHARACTERIZE SEISMIC SOURCES THAT ARE j SIGNIFICANT TO ESTIMATING THE EARTHQUAKE AND SURFACE l DISPLACEMENT HAZARDS TO A NUCLEAR POWER PLANT SITE i TO DEFINE REGIONAL AND SITE SUBSURFACE CONDITIONS (PROPERTIES OF j SOILS AND ROCKS) AS TO THEIR GROUND MOTION PROPAGATION l CHARACTERISTICS AND STABILITY UNDER SEISMIC LOADING l l TO PROVIDE BASIC INPUT TO DETERMINISTIC AND PROBABILISTIC HA7ARD I ANALYSES l l APPENDIX B IS A REVISION OF APPENDIX A THAT CONTAINS GENERAL l i CRITERIA. THE REG. GUIDE PROVIDES RECOMMENDATIONS, DESCRIPTIONS j OF METHODS, AND REFERENCES TO GUIDE APPLICANTS AND NRC REVIEWERS i IN GETTING THE NEEDED INFORMATION. i I I
) ! a-3.anc . - , . 3
I
. e , ,
i i TERMS OR PHRASES ELIMINATED FROM THE REGULATION OR REVISED l CAPABLE FAULT TECTONIC PROVINCE i TECTONIC STRUCTURE
- OPERATING BASIS EARTHQUAKE MAXIMUM CREDIBLE EARTHQUAKE TERMS OR PHRASES ADDED TO THE REGULATION OR SUPPORTING GUIDANCE DOCUMENTS SEISMIC SOURCE SEISMOGENIC SOURCE CAPABLE TECTONIC SOURCE DETERMINISTIC ANALYSES DETERMINISTIC EARTHQUAKE PROBABILISTIC ANALYSES PROBABILISTIC CONTROLLING EARTHQUAKE ESTIMATED MAXIMUM EARTHQUAKE SSE GROUND MOTION vus12-3.m c 4~
I SEISMIC SOURCE A " SEISMIC SOURCE" IS A GENERAL TERM REFERRING TO BOTH SEISM 0 GENIC SOURCES AND CAPABLE TECTONIC SOURCES. SEISMOGENIC SOURCE A "SEISMOGENIC SOURCE" IS A PORTION OF THE EARTH WHICH IS CONSIDERED TO HAVE UNIFORM SEISMICITY (SAME EXPECTED MAXIMUM EARTHQUAKE AND FREQUENCY OF RECURRENCE) DISTINCT FROM THE SEISMICITY OF THE ADJACENT AREAS. A SEISMOGENIC SOURCE IS NOT EXPECTED TO CAUSE SURFACE DISPLACEMENT. CAPABLE TECTONIC SOURCE A " CAPABLE TECTONIC SOURCE" IS A TECTONIC STRUCTURE WHICH CAN GENERATE BOTH EARTHQUAKES AND DEFORMATION SUCH AS FAULTING OR FOLDING AT OR NEAR THE SURFACE IN THE PRESENT SEISM 0 TECTONIC REGIME. 12-3.RMC
- J (i). A "seismogenic source" is a portion of the earth's crust which is assumed -
to.have uniform earthquake potential (same expected maximum earthquake and frequency of recurrence) distinct from the earthquake potential of the surrounding area. A seismogenic source is not expected to cause surface displacements. Seismogenic sources cover a wide range of possibilities from a well-defined tectonic structure to simply a large region of diffuse seismicity (seismotectonic province) thought to be characterized by the same earthquake recurrence model. A seismogenic source is also characterized by its envolvement in the current tectonic regime as . reflected in the Quaternary (approximately the last 2 million years). (j) A ' capable tectonic source" is a tectonic structure which can generate I both earthquakes and tectonic surface deformation such as faulting or '
)
folding at or near the surface in the present seismotectonic regime. It l is characterized by at least one of the following characteristics: l (1) Presence of surface or near surface deformation of recurring nature of landforms or geologic deposits within the last 500,000 years or at least once in the last 50,000 years. (2) A reasonable association with one or more large earthquakes or sustained earthquake attivity which are usually accompanied by significant surface deformation. (3) A structural association with a capable tectonic source according to characteristics (1) of this paragraph such that movement on one could be reasonably expected to be accompanied by movement on the other. In some cases, the geologic evidence of past activity at or near the ground surface along a particular capable tectonic source may be obscured at a particular site. This might occur, for example, at a site having a t deep overburden. For these cases, evidence may exist elsewhere along the structure from which an evaluation of its characteristics in the vicinity of the site can be reasonably based. Such evidence shall be used in determining whether the structure is a capable tectonic source within - this definition. ' Notwithstanding the foregoing paragraphs III(g) (1), (2) and (3), structural association of a structure with geologic structural features , which are geologically old (at least pre-Quaternary) such as many of those found in the Eastern region of the United States shall, in the absence of conflicting evidence, demonstrate that the structure is not a capable tectonic source within this definition.
INFORMATION NEEDED ABOUT SEISMIC SOURCES t l - LOCATION, GEOMETRY, AND DISTANCE FROM SITE HISTORY OF QUARTERNARY (LAST 2 MILLION YEARS) ACTIVITY - AGE OF i LAST OFFSET AND PREVIOUS DISPLACEMENTS, RUPTURE LENGTH AND l RUPTURE LENGTH PER EVENT, RUPTURE AREA, RECURRENCE INTERVALS, !. AND SLIP RATE l - ASSOCIATED SEISMICITY AND PALEOSEISMICITY I - RELATIONSHIP OF THE STRUCTURE TO OTHER POTENTIAL SEISMIC SOURCES l - EXPECTED MAXIMUM EARTHQUAKES (EME) AND RECURRENCE MODELS l - DETERMINISTIC EARTHQUAKES (DE) i OTHER FACTORS THAT CAN CONTRIBUTE TO THE CHARACTERIZATION OF THE l SEISMIC SOURCE SUCH AS: ORIENTATIONS OF THE SOURCES RELATIVE T0 i THE ORIENTATIONS OF TECTONIC STRESSES CHARACTERISTICS OF SOILS AND ROCKS BETWEEN THE SOURCES AND THE / SITE 4 i 4 j 12-3.RMC
- l i
l 1 SUBJECTS ON WHICH INVESTIGATIONS SHOULD FOCUS SEISMICITY - HISTORICAL AND INSTRUMENTAL i i GEOLOGY - STRUCTURAL, STRATIGRAPHY, LITH 0 LOGY, GE0 MORPHOLOGY, ! AND PALE 0 SEISMICITY i j - GEOPHYSICS i - GE0 TECHNICAL 1 l VUS12-3.RMC
INVESTIGATIONS GENERAL (ALL SITES) - THE LEVEL OF DETAIL OF INVESTIGATIONS IS GOVERNED BY THE CURRENT AND LATE QUATERNARY TECTO i COMPLEXITY OF THE SITE AND REGION SEISMIC HONITORING AND GROUND MOTION ANALYSIS ANALYSIS OF SEISMICITY - HISTOP,ICAL AND INSTRUMENTAL REMOTE SENSING INVESTIGATIONS AND ANALYSIS GE0 LOGICAL RECONNAISSANCE OF REGION AND SITE MAPPING OF TOP 0 GRAPHY, GEOLOGY, GE0 MORPHOLOGY AND HYDROLOGY - REGIONAL SCALE IDENTIFICATION AND EVALUATION OF VERTICAL CRUSTAL MOVE 2 GROUND OR WATER GEOPHYSICAL INVESTIGATIONS DETAILED GE0 LOGIC AND GE0 MORPHIC MAPPING 0F SPECIFIC A (INCLUDING SITE) ANALYSIS OF STREAM PROFILES
~
*e
.2-3.RMC . =
8
- INVESTIGATIONS (cour.)
AaALYSIS OF 0FFSET OR ANOMAi.00S LAND FORMS ANALYSIS OF QUATERNARY DEPOSITS WITHIN OR NEAR TECTONIC ZONES IDENTIFICATION AND ANALYSIS OF SEISMICALLY INDUCED DEFORMATION FEATURES AND DISTINCTION BETWEEN THEM AND NONTECTONICALLY INDUCED DEFORMATION DETERMINATION OF AGES OF FAULT DISPLACEMENTS CORE BORINGS AND IN HOLE GEOPHYSICS TRENCHING, LOGGING AND AGE-DATING MATERIALS SYNTHESIS OF DATA INTERPLATE REGION - EASTERN AND CENTRAL U.S. DEFl?iING SEISMOGENIC SOURCES MOST IMPORTANT (CAPABLE TECTONIC SOURCES USUALLY ABSENT) HEAVY RELIANCE ON SEISMICITY AND GEOPHYSICS
- IF A FAIILT OF UNKNOWN AGE IS ENCOUNTERED, SIMILAR METHODS USED
~
IN WESTERN U.S. ARE USED TO ASSESS IT VUS12-3 RMC
l l l i
- INVESTIGATIONS (cour.)
I ACTIVE PLATE MARGIN REGION - WESTERN U.S. T THE FOCUS IS ON BOTH SEISMOGENIC AND CAPABLE TECTONIC SOURCES l i MORE EMPHASIS ON SURFACE OR NEAR SURFACE METHODS } - THERE ARE ALSO UNIDENTIFIED SOURCES IN WESTERN U.S. !. - BURIED FAULTS (BLIND THRUST FAULTS) AND SUBDUCTION ZONE l INTERFACE IN PACIFIC NORTHWEST
- FOR BURIED SOURCES - GE0DETICS, GEOPHYSICS, SEISMICITY AND l
! ANALYTICAL METHODS (SUCH AS QUANTITATIVE STRUCTURAL GE0 LOGIC 4 CROSS-SECTION ANALYSIS) ARE IMPORTANT I l 1 l 12-3.RMC
- e e ,
l AREAS OF INVESTIGATION - DEPENDS ON CURRENT AND LATE QUATERNARY TECTONIC REGIME AND GEOLOGIC COMPLEXITY ' i l REGIONAL GEOLOGICAL, GEOPHYSICAL AND SEISMIC STUDIES - RADIUS OF 200 i MILES (DATA PRESENTED AT A SCALE = 1:500,000 OR SMALLER). I l INVESTIGATION IN GREATER DETAIL WITHIN A RADIUS OF 25 MILES (MAY BE l LARGER IN AREAS NEAR CAPABLE TECTONIC SOURCES, HIGH SEISMICITY OR COMPLEX GEOLOGY) (SCALE - 1:50,000 OR SMALLERi. i l DETAILED GE0 LOGICAL AND GEOPHYSICAL STUDIES - RADIUS OF 5 MILES (MAY i BE LARGER IN AREAS NEAR CAPABLE TECTONIC SOURCES, HIGH SEISMICITY OR , COMPLEX GEOLOGY) (SCALE - 1:5,000 OR SMALLER). l ! MOST DETAILED INVESTIGATIONS AT THE SITE - APPROXIMATELY 1 SQ. MILE l (SCALE - 1:500 OR SMALLER) . i VUS12-3.RMC
i PROPOSED REVISION TO e STANDARD REVIEW PLAN 2.5.2 VIBRATORY GROUND MOTION i , i i i 1 I PHYLLIS SOBEL, GEOPHYSICIST ! STRUCTURAL AND GEOSCIENCES BRANCH, NRR i 301-492-7038 1 i 1 d 4 ACRS SUBCOMMITTEE ON EXTREME EXTERNAL PHENOMENA i DECEMBER 10,1991 i ~ r
SRP 2.5.2
~
AREAS OF REVIEW j Seismicity l 1 . Geological and tectonic characteristico of the site and
- region l
Correlation of earthquake activity with seismic sources Maximum earthquake potential l Seismic wave transmission characteristics of the site (geotechnical engineering analysis in SRP 2.5.4) Safe Shutdown Earthquake ground motion
- i l
i i l STANDARD REVIEW PLAN 2.5.2 ! PROBABILISTIC STUDIES i ) j . ! PAGE 2.5.2- 10: 1 l e I
"Probabilistic estimates of seismic hazard should be .
- calculated ... and the underlying assumptions and i associated uncertainties should be documented to assist
! in the staff's overall deterministic approach. The l probabilistic studies should highlight which seismic sources are significant to the site. ... The probability
- of exceeding the SSE response spectra should also be estimated and comparison of results made with other l probabilistic studies."
I 3 . 4 i i 1
PROPOSED CHANGES TO SRP 2.5.2 : VIBRATORY GROUND MOTION ! Reference proposed Appendix B and Seismic Sources RG OBE related changes Add new concepts to the text - Seismic Sources i Seismotectonic Sources l Capable Tectonic Sources Expected Maximum Earthquake Probabilistic Controlling Earthquake Deterministic Earthquake 1 Revise the use of probabilistic and deterministic results l i ! Update references . I i 1 <
~
l
SRP 2.5.2 < GROUND MOTION ESTIMATION SSE response spectra are acceptable if they equal or exceed the estimated 84th percentile ground motion 1 spectra from the controlling earthquakes Staff independently evaluates seismic sources and ground motion estimates ., Site-specific response spectra preferred i Statistical analysis of recorded strong motion data l i. Similar source, path and recording site properties I I Free-field conditions i l if insufficient strong motion data available, site-specific spectra may be scaled for appropriate magnitude, distance, and foundation conditions.
)
If only peak ground acceleration data is available, , a peak acceleration can be used as the high frequency asymptote to standardized response spectra such as i RG 1.60. I Theoretical-empirical models may be used, especially for l sites near capable tectonic structures or deeper sources *j
/
1
9 I r ENGINEERING ASPECTS ASSOCIATED WITH THE REVISION OF . 10 CFR PART 100, APPENDIX A < APPENDIX S TO 10 CFR PART 50 KENNEALLY OBE/SSE INVESTIGATIONS WITTE I ENGINEERING REGULATORY GUIDES KENNEALLY
\
APPENDIX S TO 10 CFR PART 50 EARTHQUAKE ENGINEERING CRITERIA - FOR NUCLEAR POWER PLANTS PRESENTED TO THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS EXTREME EXTERNAL PHENOMENA SUBCOMMITTEE DECEMBER 10,1991 l 1 1 l I l ROGER M. KENNEALLY - l l U.S. NUCLEAR REGULATORY COMMISSION DIVISION OF ENGINEERING OFFICE OF NUCLEAR REGULATORY RESEARCH (301) 492-3893 4
ENGINEERING CONTENT OF APPENDlX A TO PART 100 DEFINES THE SAFE SHUTDOWN EARTHQUAKE AND THE OPERATING BASIS EARTHQUAKE - lil(c) and lil(d) . l DEFINES SAFETY RELATED STRUCTURES, SYSTEMS AND
. COMPONENTS - Ill(c), Vi(a)(1)(i to iii), and Vi(a)(2) -
DEFINES THE MINIMUM VALUE OF THE SSE - Vi(a)(1)v ESTABLISHES THE OBE/SSE RATIO - V(a)(2) REQUIRES PLANT SHUTDOWN IF THE OBE IS EXCEEDED V(a)(2) VIBRATORY GROUND MOTION DEFINED BY RESPONSE SPECTRA AT ELEVATIONS OF THE FOUNDATIONS - Vi(a)(1) and Vi(a)(2) IDENTIFIES ACCEPTABLE ANALYTICAL METHODS - Vl(a)(1) and Vi(a)(2) i
- a DESIGN FOR SURFACE FAULTING - Vl(b) i DESIGN FOR SEISMICALLY INDUCED FLOODS AND WATER WAVES - Vl(c) l l
SOIL CONSIDERATIONS - V(d)(1 to 4) l
5 4 l 1 1 l PROPOSED CHANGES TO PART 50 i i l
- 150.34 CONTENTS OF APPLICATIONS; TECHNICAL INFORMATION l
l ADD PARAGRAPH (A)(12) . i STATES THAT EARTHQUAKE ENGINEERING CRITERIA ARE IN f APPENDIX S TO PART 50 - l 550.54 CONDITIONS OF LICENSES i j ADD PARAGRAPH (EE) i j PLANT SHUTDOWN REQUIRED IF THE OPERATING BASIS l EARTHOUAKE IS EXCEEDED l j PROPOSED APPENDIX S TO PART 50 l ! GENERATED BY MOVING SECTION VI, APPLICATION TO l ENGINEERING DESIGN, OF APPENDIX A TO 10 CFR PART 100 TO PART 50
i i PROPOSED APPENDIX S TO PART 50 APPLICAB!!lTY REV! SED CF.lTERIA WILL NOT BE APPLIED TO EXISTING PLANTS I e CRITERIA APPLIES TO PART 52 OR PART 50 APPLICANTS ON OR AFTER THE EFFECTIVE DATE OF THE REGULATION PART 52 APPLICANTS DESIGN CERTIFICATION l COMBINED LICENSE i PART 50 APPLICANTS CONSTRUCTION PERMIT OPERATING QCENSE e
PROPOSED APPENDIX S TO PART 50 SAFE SHUTDOWN EARTHQUAKE GROUND MOTION DESIGNATION CHANGED TO SAFE SHUTDOWN EARTHQUAKE GROUND MOTION (ACRONYM REMAINS SSE) AMBIGUITY IN THE CURRENT REGULATION HAS BEEN l REMOVED. THE FOLLOWING PHRASES HAVE BEEN REMOVED FROM THE REGULATION: j "THAT EARTHOUAKE" - i i " MAXIMUM EARTHQUAKE POTENTIAL" 1 i ! THE CONTROL POINT MOTION HAS BEEN CHANGED FROM I "THE ELEVATIONS OF THE FOUNDATIONS" TO "THE FREE ! l GROUND SURFACE OR HYPOTHETICAL ROCK OUTCROP, AS . i APPROPRIATE." ! THE HORIZONTAL SSE AT THE FOUNDATION LEVEL SHALL BE " ! AN APPROPRIATE RESPONSE SPECTRUM OF AT LEAST 0.1G I i j
i i ! PROPOSED APPENDIX S TO PART 50 ; OPERATING BASIS EARTHQUAKE i. I ! ELIMINATE DIVERSE DEFINITIONS OF THE OBE THAT ARE IN THE CURRENT REGULATION l-j FUNCTIONALITY OF STRUCTURES, EQUIPMENT AND 1 COMPOhlENTS ]e l LIKEllHOOD OF OCCURRENCE (AFFECT PLANT SITE l DURING OPERATING LIFE) MINIMUM PERCENTAGE OF THE SSE. . i i ! THE OBE WILL BE ASSOCIATED WITH THE FUNCTIONALITY OF STRUCTURES, EQUIPMENT AND COMPONENTS. \ - l RESPONSE ANALYSIS REQUIREMENTS ASSOCIATED WITH l THE OBE MAY BE ELIMINATED IF THE APPLICANT ELECTS TO j SET THE OBE VALUE TO THAT SPECIFIED IN THE REGULATION i l DESIGN PROVISIONS ASSOCIATED WITH, FOR ! INSTANCE, FATIGUE, ARE DISCUSSED IN REGULATORY l GUIDES HIGHER OBE VALUES MAY BE SELECTED, HOWEVER, AN l j EXPLICIT SUITABLE ANALYSES AND DESIGN SHALL BE j PERFORMED. l STAFF STUDIES RELATED TO THE MINIMUM VALUE OF THE i OBE THAT WILL APPEAR IN THE REGULATION WILL BE i DISCUSSED IN A SEPARATE PRESENTATION l
l t i OBE/SSE INVESTIGATIONS ! PRESENTED TO THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS l' EXTREME EXTERNAL PHENOMENA SUBCOMMITTEE DECEMBER 10,1991 MONIKA C. WITTE
, LAWRENCE LIVERMORE NATIONAL LABORATORY t i
FTS 543-0203
] (510) 423-0203 -
+ . ,
j i
!! !!li i s
r s e dr n a
. g s d
n y n i s n s e o a D e t a r g i t S m S o a x r e d n d r p t i G a n n d n n s a n e e e s i r s p p i n i t e o e r A o i l i d o d p i t t U C e r o a y n l u d f f o n f f g u o a o s n e r t S y e w l p y i o k v i e kr r t p w s r v o a e i u e ^ O N R S R W m m * * * * *
- n e
u s e S m
Options have been considered for changes to the engineering aspects of Part 100 Appendix A. Previous suggestions regarding the OBE have included: i
- Decouple the OBE and SSE; set the OBE based upon a reasonable return period.
- Leave the OBE and SSE coupled as the current regulation '
requires; modify the load factors used in the load combinations equations and modify the damping requirements so that the design process would be ,
- streamlined, and the OBE would be no longer govern.
- Set the OBE-shutdown-for-inspection-level at some factor of the SSE. Leave the option to the utility to design for the OBE, or assume the risk of an inspection-shutdown at some lower level.
~. .91 329-3 , 4 e t
E
,S s eS s l
yi t kr au 4. n er t en t ot f u
. uo q ais t c a A hot tera e i Sd e x t r sbb l
a s i i m As d n e o u sv a s Tiv e i
) g w i
e Ker p sg nnl o s nt p i so wdail c i o A aa l one r dn bn doh i e 0 na t u pt e dd i n 0 1 g", hs eg. h t v o a R i sn ee s ri r rnE r5 p7 F db odB fo c eeO 4 d C e ( e9 eb e s r1 a n 0 1 nd ore kh ea ce t xh i n at et v si o ys u oso r nd l e r i oeu p u ng onu h qs ef s t ass se ss re sdr t t y l ui d e as e s ts eri e l t gt s i r e u a e a q n es l rr b umq e vwee e cf i s e rB el e uy cr c i m1
, i l
nvdl e s0 dr 1 0d nal r i2 e2 d a 2e 2l l i t oeinl s ti s sd n a ciog wc ehni r a nry a e pT do h: a al t t s s n mdn n os n.lut s r ne yo s e iEoi etace h he nShm i nit h GSr T Tt ASsi aa l mlu rg * * *
- 4-9 e e 2 3
1 9 Gr L P A - L
Risk Study The first step of this study included the following:
- A. Identification of components which might be affected by t
a change in the OBE design criteria. B. Identification of generic categories of components which are potential seismic risk contributors. i t s
& }
Risk Study (con't) List A - Components which might be affected by a change in the OBE design criteria, based on survey of AEs and designers. This list consists of the following components. i Concrete frames including walls, floors, and roofs in auxiliary building, fuel handling building etc. Piping and piping support systems Spherical steel containments and heads on cylindrical steel containments Containment penetration reinforcement
- Reactor vessel and steam generator internal components Large bore component support snubbers I
Embedment plates APL91-3NS
Risk Study (con't) List B - Generic categories of components which are potential i seismic risk contributors, based on seismic PRAs. This list consists of the following componentu: i NSSS Supports Control Rod Drive Mechanism i Valves - Heat Exchangers l Ground Mounted Storage Tanks Small Indoor Tanks Batteries and battery racks Motor Control Centers i Switchgears l Transformers l Diesel Generator System (i.e., day tank, air receiver tank, DG Control Panel) l Pumps (horizontal and vertical) l HVAC Fans and Cooler Units l l
- l l
l 91-329-7 , 4 , ,
i Risk Study (con't> The second step of this ntudy was to select a subset of components from lists A and B for careful review.
- If, for these components, the designs are governed
- by the load combination equations which include the OBE loads, then the study evaluates the effect on the component capacity.for the hypothetical condition of no OBE design analysis.
The third step of the study will be to determine the effect on seismic risk, if the capacities are reduced. The fourth step of the study will be to suggest changes to the component design process if there is an increase in the seismic risk.
- Governed by the OBE in this case means that the total upset load, including the OBE load, divided by the upset allowable is greater than the total faulted load, including the '-
SSE load, divided by the faulted allowable. , APL91-329-8
. I!:, l!!r! l!ltIi! i;!!jiljjl; i.
e e gh e h t E r a t r B 2 l n e r r f o O 3 o ad i ee y r o wo er e w b s, b t t be f e eadu sus d d e s u sl mna e sc e n y n ed dc hsai n t er r r i a t s yht ou e a nsl io y v t b d ri l t s f og m o e ehni r gl , n w vt oc i u ia nl e m ot i or r ga sp l e e nx f ud e u e i v . d ot h ae lde n w s b e r et rp t l A gr 9 s s n wda x o ne - a e nl ee n ,
.iv d t h ) e i d po e e mg s v e eclae l
e v u tr s oe b n r so ma t t r s w l o s f p e o e nn res t r edaE B v o i e r f et y v a i t a r o p eCOv n st h r e i tiTa shd p rR i e w ni.l o p in t at u rpmeV . s he s t r e t r n t n n sir sE .S e a o P nbe o mv d a Si e e s p e p whr t s oo f p n i o e oSs t g. p am o t n a t. cwl yP i sios u t s ta o p e t d ohkon r .l era d ed m wre ps s i r d c 3 m o me a at a nt sen es 44 n n o s,u u a zor F n g (1 c i pt n qibd g sly tyit el e g t a e gh r e n r ac sht . i ss n n o sistrnl tt i p u nue y et o w pe a o ai u ad p r o eei o i dne w C T Ud e cf P FrrpF d dn o , t u l eop l l o i f S t am e k Dc eo h T . i s . R . ,iI! lllll , lj j'
V e gCe / w* = I e +0 i c W $$o
- f. . M Je l
6 g M ma m m u Om m i o e l i
-. es c
c= V e E e 3 e "eE - E 0 3 l>
~
~- ~
m
'5 e 3e a
g) .- t
' -- u l m 2 e
e o a e se wC O $ :
= g e a R. ge 3 g e og e$ ,
m 3 $ - em no , e # 6 e c e b .c s= mt , 6 = s > c ac u g m O Od m- O 15 so Em ; g 3 a te s l
- w 5 .>o"c E m -> a .E -
y; e O -. e V m O so mm e 2 e n De o O g *e2
+.*
g e C Ru & E E 5" a w a e li; 3
> sm oc 'e *"u s s a
.tf a E
'Em E = w o
O O Sw w mc e Ogg
, m gg j E gw M
** N hW o M W .0.,
EC $ g
'> c Og5 Rc e
V w m e u s m OS e u fOO O M
>OO !
m & O y w
& 4 h l
M * * . g
.V.). ' ; ;
k i l f
i 5 Risk Study: First step results summary. , i
)
- Refrigeration Unit j i
One chiller was reviewed. It's design was governed by the OBE.
.91 329-11 ;
- A
- Survey of Utilities and Designers. _
i
- I contacted staff people from utilities, engineering consulting firms, AE's, code organizations, and utility consortiums in order to obtain feedback on the feasibility of eliminating OBE design.
i
- Input was qualitative
- Input was occasionally contradictory i
l I i I i t r
"~
l i
- We obtained input on structural & mechanical components:
l Concrete struClures steel structures . i Piping
- embedment plates t component supports in-line components such as valves containment penetrations
- We also obtained more limited input on electrical components: :
i relays motor controllers instrumentation f valve operator-aimis , < 3 .
l i . t l ENGINEERING REGULATORY GUIDES l ASSOCIATED WITH i APPENDIX S TO 10 CFR PART 50 j - t i l PRESENTED TO THE i- ADVISORY COMMITTEE ON REACTOR SAFEGUARDS i EXTREME EXTERNAL PHENOMENA SUBCOMMITTEE i DECEMBER 10,1991 i i i 4 i
- ROGER M. KENNEALLY ll U.S. NUCLEAR REGULATORY COMMISSION
( DIVISION OF ENGINEERING OFFICE OF NUCLEAR REGULATORY RESEARCH (301) 492-3893 l 5 i
NEW REGULATORY GUIDES DRAFT REGULATORY GUIDE DG-1016 SECOND PROPOSED REVISION 2 TO REGULATORY GUIDE 1.12, NUCLEAR POWER PLANT INSTRUMENTATION FOR EARTHQUAKES - DRAFT REGULATORY GUIDE DG-1017 PRE-EARTHQUAKE PLANNING AND IMMEDIATE NUCLEAR. POWER PLANT OPERATOR POST-EARTHQUAKE ACTIONS DRAFT REGULATORY GUIDE DG-1018 RESTART OF A NUCLEAR POWER PLANT SHUT DOWN DUE TO A SEISMIC EVENT
~
1 l
- OVERVIEW OF EARTHQUAKE RESPONSE i
i i PRE-EARTHQUAKE 1 i ! OPERABLE SEISMIC INSTRUMENTATION t i i i SELECT / INSPECT STRUCTURES AND EQUIPMENT i - 1 i i 4 4 s POST-EARTHQUAKE 1 1 l WALKDOWN i i EVALUATION OF GROUND MOTION RECORDS l DETERMINE OBE EXCEEDANCE i ! v i I 4 i 1 4 l _ .. _ ~ . _ _ . _ _ _
i i l [ l FELT EARTHOUAKE i r W PLANT U TRIPPED IWWEDIATE IMWEDIATE OPERATOR OPERATOR
- ACDONS ACTIONS 1 t i r 1 r 1 r OPERATOR EVALUATION OPERATOR EVALUATION WALKDOWN OF GROUND WALKDOWN OF GROUND INSPECTIONS WOTION RECORDS INSPECDONS WCfTION RECORDS I I I I i f I r l
OBE OBE N EXCEEDED E N EXCEEDED E OR DAMAGE OR DAMAGE FOUND FOUND l 1 r 1 r 1 r i r CONTINUE PRE-SHUTDOWN POST-TRP POST-SHUTDOWN l OPERATION INSPECTIONS REVIEW INSPECTIONS AND TESTS SEE FIC 3-2 1 i r i NN REMT SHUTDOWN ' f l 1 r
- %ST-SHUTDOWN lNSPECTIONS AND TtSTS SEE FIG 3 _2 Figure 3-1. Flow Diagram of Short-Term Actions
PRE-EARTHQUAKE PLANNING k \ 4 l
, OPERABLE SEISMIC INSTRUMENTATION l l
PROCESS DATA WITHIN FOUR HOURS ; ALTERNATIVE STAFF POSITION IS DEFINED IF INSTRUMENTATION IS INOPERABLE r PRE-SELECTED SAMPLE OF STRUCTURES AND EQUIPMENT TO BE INSPECTED . PERFORM VISUAL BASE LINE INSPECTIONS. UNDERSTAND PRE-EXISTING CONDITIONS. I i I
IMMEDIATE POST-EARTHQUAKE ACTIONS I ACTION INITIATED AFTER A " FELT EARTHQUAKE" 1 1 EXCEPTION TO THE DEFINITION IN THE REPORT: j l v i CONSENSUS OF CONTROL ROOM OPERATORS gg - , l FREE FIELD OR FOUNDATION LEVEL SEISMIC INSTRUMENTS ARE ACTIVATED ( - l SPURIOUS ACTIVATION THAT CAN BE LINKED TO A NONSEISMIC EVENT DOES NOT DENOTE ACTIVATION RESPOND TO PLANT ALARMS - STABILIZE PLANT WALKDOWN INSPECTIONS \
)
i j NUCLEAR POWER PLANT INSTRUMENTATION FOR EARTHOUAKES ! SOLID-STATE, STATE-OF-THE-ART DIGITAL TIME-HISTORY : l l ACCELEROGRAPHS i INSTRUMENTATION SENSORS PLACED IN THE FREE-FIELD, ,, BUILDING FOUNDATIONS, AND AT ELEVATION IN THE j BUILDING ! c ! SENSORS- ARE NOT PLACED ON EQUIPMENT OR EQUIPMENT l SUPPORTS i i ! EMPHASIZES THAT INSTRUMENTATION SHOULD BE OPERABLE OTHER FACETS ASSOCIATED WITH THE INSTRUMENTATION, FOR INSTANCE, INSTALLATION, ACTUATION, REMOTE INDICATION, MAINTENANCE, AND MULTI-UNIT SITES ARE SIMILAR TO THE CURRENT REGULATORY GUIDE (1.12 REV.1) AND ANSI /ANS-2.2-1988 I v (
l PRE-EARTHQUAKE PLANNING AND IMMEDIATE NUCLEAR POWER PLANT OPERATOR POST-EARTHQUAKE ACTIONS l ENDORSES SPECIFIC SECTIONS OF EPRI NP-6695,
" GUIDELINES FOR NUCLEAR PLANT RESPONSE TO AN EARTHQUAKE" ,
REGULATORY POSITION IS ALSO BASED ON: l ANSI /ANS-2.10-1979, " GUIDELINES FOR RETRIEVAL, . REVIEW, PROCESSING AND EVALUATION OF RECORDS OBTAINED FROM SEISMIC INSTRUMENTATION," EPRI NP-5930, "A CRITERION FOR DETERMINING EXCEEDANCE OF THE OPERATING BASIS EARTHQUAKE." EPRI REPORT TR-100082, " STANDARDIZATION OF CUMULATIVE ABSOLUTE VELOCITY FOR USE WITH THE EPRI OBE EXCEEDANCE CRITERIA."
)
i
- Summary of input from Utilities and Designers.
- Impact of OBE design depends to a large extent on philosophy of designers.
Examples: come designers use different damping to generate response spectrum for SSE and OBE. Sorne use only OBE damping for both OBE and SSE F allow for scaling fince determination has been made that OBE may govern anyway. In-line components are frequently designed for SSE loads, then compared to OBE allowables. .
- Streamlining the seismic design process is a worthy goal.
- OBE might govern the design of those components listed in Risk Study List A.
Concrete frames including walls, floors, Reactor vessel and steam and roofs in auxiliary building, fuel generator internal components handling building etc. Large bore component Piping and piping support systems support anubbers Spherical steel containments and heads Embedment plates on cylindrical steel containments Containment penetration reinforcement , AFit1-32914
Required Changes to Codes and Standards: Documents reviewed include:
- ASME BPVC Section 111 Appendix N Subsections NB, NC, ND, NE, NF, NG, CB & CC Codes Cases N47, N411, N451, N462, N468
- ACI 349-85 and 349.1R-80
- ANSI /AISC N690-1991
- ASCE 1-82 N-725 and ASCE 4-86 IEEE 323 and IEEE 344 (the review of these two is not yet complete)
L91329-15 , g >
- Required changes to Codes and Standards:
(con't) There are no surprises in the review. Suggested changes include:
- definitions
- references to dual earthquake loads and other minor wording changes
- fatigue and seismic anchor movements must be included in the design somehow
- deletion of no longer relevant.. load combinations in ACI 349-85, and modification of others as necessary APL91329-16
Work planned or in progress. We expect to complete a detailed review of steel containment design in 1992. We hope to obtain more data from several sources on component loads in order to be able to evaluate more fully the adequacy of the proposed OBE level. We will complete the risk study. i
~. .91 -329- t 7
l 1 EVALUATION OF GROUND MOTION RECORDS j POSITION BASED ON ANSI /ANS-2.10-1979 , THIS STANDARD IS BEING REVISED AND SHOULD BE PUBLISHED IN MID-1992 r THE REVISED STANDARD CONTAINS A SECTION ON OBE EXCEEDANCE
DETERMINE OBE EXCEEDANCEi ,. POSITION BASED ON EPRI NP-5930 AND EPRI REPORT TR-100082 (UPDATE OF EPRI NP-5930) BASED ON EXCEEDING A RESPONSE SPECTRUM AND A CUMULATIVE ABSOLUTE VELOCITY CHECK RESPONSE SPECTRUM CHECK THE RESPONSE SPECTRUM CHECK IS EXCEEDED IF ANY OF THE THREE COMPONENTS OF THE 5 PERCENT ' DAMPED FREE-FIELD GROUND MOTION RESPONSE SPECTRA IS LARGER THAN THE CORRESPONDING DESIGN RESPONSE SPECTRA OR 0.2G, WHICHEVER IS GREATER FOR - FREQUENCIES BETWEEN 2 TO 10 HZ, OR A SPECTRAL VELOCITY OF 6 INCHES PER SECOND , FOR FREQUENCIES LESS THAN 2 HZ STAFF EXCEPTIONS TO THE RESPONSE SPECTRUM CHECK i ADDED SPECTRAL VELOCITY CHECK AT FREQUENCIES LESS THAN 2 HZ
- ELIMINATED THE REQUIREMENT THAT A SECOND SPECTRAL ACCELERATION MUST BE AT LEAST 2/3 '
OF THE LIMIT VALUE CUMULATIVE ABSOLUTE VELOCITY (CAV) CHECK 3 THE CAV IS THE TIME INTEGRAL OF THE ABSOLUTE ACCELERATION TIME HISTORY OVER A PRESCRIBED DURATION OF THE EARTHQUAKE RECORD THE CAV CHECK IS EXCEEDED IF ANY OF THE THREE COMPONENTS IS LARGE THAN 0.16 G-SEC0ND ,
PLANT SHUTDOWN CRITERIA OBE EXCEEDANCE RESPONSE SPECTRUM CHECK AND CAV CHECK WERE EXCEEDED THAN THE OBE WAS EXCEEDED
.QB DAMAGE ^ WALKDOWN INSPECTIONS DISCOVER DAMAGE PRE-SHUTDOWN INSPECTIONS ORDERLY PLANT SHUTDOWN ASSURE THAT SHUT DOWN EQUIPMENT IS OPERATIONAL t
RESTART OF A NUCLEAR POWER PLANT SHUT DOWN DUE TO A SEISMIC EVENT INITIAL FOCUSED INSPECTIONS AND TESTS
- EXPANDED INSPECTIONS AND TESTS BASED ON ,
l INSPECTIONS OF SAFETY-RELATED EQUIPMENT AND STRUCTURES i . l EXPANDED INSPECTIONS AND TESTS BASED ON ! INSPECTIONS OF NONSAFETY-RELATED DAMAGE . INDICATORS 0 LONG-TERM EVALUATIONS l l I l [
l FOCUSED INSPECTIONS 1 r DETERMINATION OF EPRI DAMAGE INTENSmf e i r DAMAGE TO FE5 A. SAFETY-RELATED D PANDED 4 6 lNSPECTIONS EQUIPMENT 1 r DAMAGE W . W _ DAMAGE
-Em -
Em - Ri s DAMAGE PE5 ?ES E SURVEILLANCE - E TO EARTHQUAKE TESTS ' 4 g. DAMAGE 1 ' ' REPAIR REPAIR INSPECT RV DAMAGE 044 AGE MTERNALS 1 AND FUEL TES RESTART , , , , W DAMAC U LEAK REPMR 3CPRI 1 DAMAGE nSTS
.LONG-TERM EVALUATEWS y SE FIG 3-3 3 r i r RESTART SURVM1ANCE CONTMNMENT ,
I TESTS LEAK RATE i TESTS NOTD THESE ACTIONS ARE LONG-TERM PERFORMED ONLY F 08E IS EVALUATIONS ' ' D CEEDED OR DAMAGE FDUNO SEE FIG 3-3 RESTART 4;(SE FIG. 3-1) LDNG-TERM gygg (SEE nG s-s)
' tong-nRu NTIONS SE N' 3~3 SURVDL1ANCE nSTS i r RESTART !
Figure 3-2. Flow Diagram of Post-Shutdown l Inspections and Tests
EXISTING REGULATORY GUIDES , l l THE FOLLOWING GUIDES WILL BE REVISED TO INCORPORATE EDITORIAL CHANGES. FOR EXAMPLE, THE TYPE OF CHANGE l CONTEMPLATED WOULD BE TO REFERENCE NEW PARAGRAPHS IN APPENDIX S TO PART 50 1.57 Design Limits and Loading Combinations for Metal Primary Containment System Components . 1.59 Design Basis Floods for Nuclear Power Plants ' l 1.60 Design Response Spectra for Seismic Design of Nuclear Power Plants . 1.83 Inservice inspection of Pressurized Water Reactor Steam Generator Tubes 1.92 Combining Modal Responses and Spatial Components in Seismic Response Analysis 1.102 Flood Protection for Nuclear Power Plants 1.121 Bases for Plugging Degraded PWR Steam Generator Tubes i 1.122 Development of Floor Response Spectra for Seismic Design of Floor-Supported Equipment or Components l PUBLISHED WITH THE FINAL RULE i l
EXISTING REGULATORY GUIDES (CONTINUED) l THE FOLLOWING GUIDES WILL BE REVISED TO MAINTAIN EXISTING DESIGN OR ANALYSIS PHILOSOPHY. FOR INSTANCE, THE TYPE OF CHANGES CONTEMPLATED WOULD BE TO CHANGE THE OBE TO A FRACTION OF THE SSE. 1.27 Ultimate Heat Sink for Nuclear Power Plants 1.100 Seismic Qualification of Electric and Mechanical a Equipment for Nuclear Power Plants l 1.124 Service Limits and Loading Combinations for Class 1 Liner-Type Component Supports - 1.130 Service Limits and Loading Combinations for Class 1 : Plate-and-Shell-Type Component Supports 1.132 Site investigations for Foundations of Nuclear Power Plants ' 1.138 Laboratory investigations of Soils for Engineering Analysis and Design of Nuclear Power Plants 1.142 Safety-Related Concrete Structures for Nuclear Power 7 Plants (Other than Reactor Vessels and Containments) 1.143 Design Guidance for Radioactive Waste Management
- Systems, Structures, and Components Installed in Light-Water-Cooled Nuclear Power Plants PUBLISHED WITH THE FINAL RULE i
_ _ _ _ _ _}}