ML20056A436

Procedural and Submittal Guidance for Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities.Draft Report for Comment

ML20056A436
Person / Time
Issue date:	07/31/1990
From:	NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
References
NUREG-1407, NUREG-1407-DRFT, NUREG-1407-DRFT-FC, NUDOCS 9008070367
Download: ML20056A436 (41)

Text

- - - -.,,..,,,

t NUREG-1407 Procedura1anc Submittai Guidance for the Incividual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities Draft Report for Comment U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research y

$[sha mig %;

$h.5$lf.l 9000070367 900731

$So7 E" enn

AVAILABILITY NOTICE Availability of Reference Materials Cited in URC Publicat' as Most documonts cited in NRC publications will be available from one of the following sourcost 1,1 Tho NRC Public Document Room, 2120 L Street, NW, Lower Lovel, Washington. 00-20555-1 2-

_The Superintendent of Documents, U.S. Government Printing Offico, P.O. Box 37082, Washington, DC - 20013-7082 3.

The National Technical Information Servico, Springfield, VA 22161 AlthouDh the listing that follows represents the majority of documents cited in NRC publica-j tions, it is not intended to bo exhaustivo,

' j j)

Reforonced documents availablo for inspection and copying for a 100 from the NRC Public Document Room includo NRC correspondence and Internal NRC memoranda; NRC Office of inspection and Enforcement bulletins, circulars, information noticos, inspection and investi-gation noticos; Licensoo Event Reports; vendor reports and correspondonce; Commission '

{

papers; and applicant and licensoo documents and correspondence.

The lottowing documents in the NUREO series are availablo for purchaso from the GPO Sales Program: formal NRC staff and contractor reports, NRC-sponsored conference proceed-j ings, and NRC booklets and brochures. Also evalltblo aro Regulatory Guidos, NRC regula-

' tions in the Codo 01 fodoral Regulations, and Nuclear Regulatory Commission issuances.

-j i

Documents available from the Nationt '.iinical Information Servico include NUREG series reports and technical reports prepar, oy other federal agencios and reports prepared by the Atomic Energy Commission, forerunnor agency to the Nuclear Regulatory Commission.

1 Documents'available from public ar-special technical librarios include all open literature i

(toms, such as books, journal and periodical articles, and transactions. Federal Register noticos, fedoral and stato logislation, and congressional reports can usually be obtained i

from those librarios, j

Documents such as thosos, distortations, forolgn reports and translations, and non NRC

]

conferenco procoodings' are available for purchase from the organization sponsoring the -

publication cited.

Single copios of NRC draft reports are available froo, to the extent of supply, unon written request to the Office of Information Resources Management, Distribution Section, U.S.

' Nuctoar Regulatory Commission, Washington, DC 20555.

Copios of industry codas and standards used in a substantivo manner in the NRC regulatory l

process are maintained at the NRC Library, 7920 Norfolk Avenue. Bethesda, Maryland, and are available thoto for referenco use by the public. Codes and standards are usually copy-righted and may be purchased from the originating organization or, if they aro American q

National Standards, from the Amorican National Standards institute,1430 Broadway.

Now York. NY 100t8.

4

i NUREG-1407 i

Procedural and Submittal Guidance for the Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities Draf t Repc,rt for Comment Manuscript Completed: July 1990

!) ate l'ubhshed: July 1990 Divis;on of Safety issue Resolution Ollice of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555 i

(W,.......y

~

i 1

r

p,w" ' p' t -

y+

[."

i) a-

<=,,

,ja t

i r.

1

+<

2r O:p % ' ' ;

3:

en t

v'?

e t D 7) d/ =i:4, i

f f c.

'w'.

-gi 3 wj 4.,

1i

- ABSTRACT yf.

[c llaned on a Policy Statement on Severe Accklents; the -

. document presents guidance for performing and report-licennec of each nuclear power plant is requested to per -

ing the results of the individua', plant examination of 7 form an individual plant examinathm. The plant examin.

external events (IpEEE).The guidance for reporting the :

I

,3

- istion systematically looks for vulnerabilitics to severe results of the individual plant.:xamination of internal

..H accidents and cost effective safety improvements that re-events (IPU)is presented in NUXEG-1335!

j duce or climinate the important vulnerabilities. This j

M;y,

.t ll gl j '

}[

r ;x j; fr; d'

.s.

)

,I t

't s.

1 y

)

~r

.l s..

't p,

,.,o-

y a

s If

,si 4

p

,,y q

i iT

r. '. i -j

.j y

g

. a _.

A t

,i 4

m LI$

b a

-; i y

n'

. 1 p

.i.:

~

,L Rd p

7p

/ a.

hf

'[

. -i f t

i s I

g t

s u,.s l (L"

{

m' 2;0,<

7 4

o

.c p

.2 i ;

T c,

p If

, -.: G 4

i F-ci

-p g-l n'

Q pp w.

t w.

t p[-

. lili NURCO-1407; tI

<,n

,3 s

n, Y

t';1[_Y Y

l. NkJ

~

, c,.

a;

I s

p:

4 A?

(

p

,7 t

i i.

1 J

CONTENTS Page Abstract'.................................................................................

iii

- I hecu t i ve S u m m ary......................................................................

vii k

- Acknowl edge m en t s........................................................................

ix t

e i

1 I n t r od uc t io n..................................................... '.....................

1

=1.1 ll ac k g r o u n d. -.......................................................................

1

' 1.2 I Pl !P.11 0bject ives..................................................................

1 l

1.3 Pu roose of Docu m e n t........ -................ -..........,...........................

1 f

F 2 - 11 vents 6duated for inclusion in the IPElllL................................................

2

?

n o

j?

- 2.1 - Sei sm ic ! !ven t s.....................................................................

2 2.2 I n t e r nal Fires.....................................................................

2 L

2.3 l iigh Winds and Tornadoes.......................... 4..............................

3 8+

2.4 I ht e r md Flood s....................................................................

3 2.5 - Transportation and Nearby Fm '.'lty Accidents..........................................

3 2.6 Lig h t n in g.........................................................................-

3 2.7 y Severe Temperature Transients (lixtreme iIcat. l!xtreme Cold)...........................

4-U

2. 8 Sevc ic Wea t h e r S t ormS...........,.....................,...........................

4 2.9 lhternal Fires (Forest Fires, Grass Fires)..............................................-

4 2.10 llxtraterrestrial Activity (Meteorite Strikes, Satellite Falls).............,.................

4 2.1 1 Volc4m ic Act ivity...................................................................

4 2.12 S u m m a ry.........................................................................

4

. 3 i Acceptable Methodologies for Performing the Scismic IPEEI?.................................

5 k

- 3.1 S c is m ic P R A......................................................................

3.1.1 New Seismic PR A Analysis....................................................

~ 5 3.1.1.1. O cneral Conside rations...............................................

5-

- 3.1.1.2 liaza rd Selection....................................................

5 L

. 3.1.1.3 Fragility Ihtimat ion..................................................

6 3.1.14 - Scismic PR A Methodology Enhancements...............................

6 3.1.1.5 Containment Performance............................................

7 3.1.2 U se of a n thist ing PR A.......................................................

7

[

3.2 Scisn41c Margins M ethodologies......................................................

7 3.2.1 G en eral Considerations., -....................................................

8' p

3.2..s Review lxvel I!arthquake and Associated Spc: tral Shapes.........................

8 l

3.2.3 n NRC Scismic Margins M ethodology............................................ -

8 i

3.2.4 : liPRI Scismic Margins M ethodology............................................

9' O

3.2.5 Reduced-scope Margins Method..............................................

.10 -

E 3.2.6 Containment Performance............

10 3.31 Optional M ethodologies.............................................................

-10 1

4 " Acceptable Methodology for Performing the Internal Fires IPEEE..............................

12 4.1 ; New Fire PR A Analysis............................................................

12

'4.1.1. Identify Critical Areas of Vulnerability..........................................

12

- 4.1.2 b Calculate the Frequency of Fire initiation in Each Area...........................

12 4.1.3 Analyze for the Disabling of Critical Safety Functions.............................

12 4.1.4 Identify Fire induced initiating Events / Systems Analysis...........................

12 4.21 Use of an thisting Fire PR A.........................................................

~12

4.3 E Opt ional M ethodologies............................................................

12 v

IUREG-1407 9

1. ' b '

1 G,

~

b

' i-GI ' i

(.,.i,,

soMc t

v e

,w f,

l LM F

cy.

5 Acceptahic Methodology for Performing the liigh Winds. Floods. and Transportation

- and N earby I'acility Accident IPEEE -.......................................................

12 i

, f.1 1 t h t rod uct ion....................................................,..................

13 5.2 Analytical Proced u r e............................................................... -

14:

l 0

5.2.1 Review Plant. specific 11azard Data and 1.icensing Bases...........................

14

-5.2.2 Identify Significant Changes Since OL issuance.................................. -

14 5.231 Determine if the Plant / Facilities Design Meets Current Criteria..........,.........

14-(

' $.2,4. Determine if the liarard Frcquency is Acceptably Low (Optional S'.cp)..............

114

$.2.5 Perform a flounding Analysis ; Optional Step)..............................,....

14.

1 5.2.6 Pceform a Probabilistic Risk Assessment (Optional Step).......................... -

- 14 '

-n 14 l

53 O ptional M e thodologies...........................................................,

1

. 6 Coordination with Ongoing Programs.......................................................

14

+

k J[ j' 14

61 _ I n t r od U et ton............................................................,........

~

6.2 Description of Ongoing Programs.....................................................

15 6.2.1 IPE Program Related to inter al Events.......................................

'15

~

6.2.2 - Programs Related to Externa, Events..........................................

15

?

t 6.2.2.1 S eismic Prog ra m s.....................................................

15-

. 6.2.2,2 Int ernal Fires Progn ms...................................... -.........

15 l

p' 16 l

6.3. Approach oh Coordination with Ongs ng Programs......................................

63.1 ' Coordination with Int ernal Erci ts Program (IPE)...............................

16-6 3.1.1 Preanalyses Plan ning.................................................

.16

'16 63.L2 Plant Modifications..................................................-

4 16,

j 63.13 Accident snagemnt........

4

' 63.2 Caordination Among External Events Programs.................................. -

16 i

633 Coordination with Seismic Programs............................................

17 j

63 3. l ? USl A-4 5 and 01-131....................... 4...........

17-i 63.3.2 Charleston Earthquake Israe.......................................... -

17 i

- 6 3 3.3 ' U S l A-4 6.................................. 4...............,......

17 3

c?

6.3.4 Coordination with Other issues.................... 4..........................

18 1

b

' 7 Docu m entation a nd R e porting........................................................... -

18'

' 7.1 E information Submitted to the NRC.............................................. 4...

18 1

18

7.2 Information Retained for Audit......................................................

e 18

8 r R e f e re nces..........,..............................................,,................

i'

Appendix At-Review Level Earthquake g

Appendix 11: Comparison Between a Reduced scope and Full scope Seismic Margins Evaluation

~

Appendix C;: Detailed Documentation and Reporting Guidelines f

Figure t

t-5.1 L R'ecommended IPEEll Approach for Winds, Floods, and Others '.............................

13 i

Tables 11

- 3.11 Review lxvel !!arthquake-Plant Sites East of the Rocky Mot.ntains.........................

l}

3.21 Review Level liarthquake-Western United States Plant Sites..............................

B-2

' i:

H.11 Reduced-scope Margins Method '........................................................

E C.1 ! Sta' 'rd Table of Contents for IPEllli Submittal....... c.................................

C-5 LNURiiG-1407 vi E

m.

1

EXECUTIVE SUMM AlW llackgrountl by modifying, where appropriate, hardware and proce-dures that would help prevent or mitigate severe acci-In the Commission policy statement on severe accidents dents.

in nuclear power plants issued on August 8,1985, the commission concluded, based on avaih ble information, 1(!cntification Of Exterlial Events that existing lants ptse no undue risk to the public health and sa{ cry and that there is no present basis for irnmediate action on any regulatory requiremr nts for in supporting the implementation of the Severe Accident these plants, llowever, the Commission has recogni/cd, Pol cy, a study was performed. to determine which exter-based on N1(C and mdustry experience with plant-nal initiators could be a potentially important accident specific probabilistic risk assessments (Pitas), tht initiator that may pose a threat of severe core damage or systematic examinations are beneficial m, identifying of a large radioactive release to the environment, The plant-specific vulnerabilities to severe accidents that external events considered, consistent with past probabil-could be fixed with low-cost improvements. As part of the stic risk assessments (PI(As), are those events whose implementation of the Severe Accident Policy, the cause is external to all systems used in normal operation Commission issued Generic Irlter 88-20 on November and emergency operation situations. The external events 23, D88, requesting that each licensee congioct an evaluated include seismic events, internal fires, high individual plant examination (IIlli) for internally mitiated winds and temdoes, external floods, transportation and events including mternal floodmg.

nearby facility accidents, lightning, severe temperature transients (extreme heat, extreme cold), severe winter Many Pitas indicate that the risk from external events storms, external fires (forest fires, grass fires), extrater-could be a significant contributor to the core damage in restrial activity (meteorite st rikes, satellite falls), and vol-some instances. Ilowever, the examination for externally canic activity, initiated events is proceeding on a later schedule to allow the stali to carry out additional work to (l) identify which llased on the tesults of that study, the staff has concluded external hazards need a systematic examination, (2)iden-that five external events need to be included specifically tify acceptable examination methods and develop procc-in the IPElilu seismic events, internal fires, high winds, dural and submittal guidance, and (3) coordinate the floods, and transportation and nearby facility accidents.

IPlillE with other ongoing external event pograms, in llowever, licen.ees should confirm that no other plant.

December 1987, an lixternal !! vents Steering Group nnique external events with potential severe accident (l!!iSG) was established to make recommendations re-vulnerability are being excluded from the IPl!!Il!.

garding the scope, methods, and coordination of the Examination Methods The EESG has recently completed its task. This report, Seisnile Events based upon the !!ESG recommendations, provides de-tailed guidance to the licensees on the conduct of the A seismic PI(A (l.csel I plus containment performance)

IPEEl! and on the structure and content of the IPEEE or a seismic margins methodology (SMM)is considered a submittal. It provides, specifically, the guidelines defining siable approach to identify potential vulnerabilities.

the IPEEE objectives; identify external events that G uidance is provided for licensees performing a new seis-should be included in the IPEEE; identify acceptable mic Pila or updating an existing seismic Pil A: emphasis methodologies; and identify coordination between the is placed on the identification and ranking of dominant IPEEE and the ongoing NI(C programs.

plant sequences that could lead to scismically induced core damage rather than the numerical estimate of abso-lute frequency of occurrence. Methodology upgrades in.

Objectives of the ll'EEE clude plant walkdowns, evaluation of relay chatter and soilliquefaction effects, and results in terms of high confi-The general objectives of the IPEEE are similar to that of dence tow probability of failure (IICl PF) values.

the IPli-that is, for each licensee (1) to develop an appreciation of severe accident behavior, (2) to under-G uidance is also provided for licensees using the NI(C-or stand the most likely severe accident sequences that could liPI(I sp(msored seismic margins methodology 'Ihe mar-occur at its plant,(3) to gain a qualitative understanding gins methodology screens components based on their im-of the overall probability of core damage and fission prod-portance to safety and cismic capacity. Ily design, the uct releases, and (4) if necessary, to reduce the overall methodology utilizes two review or screening levels probabilitics for core damage and fission product releases geared to peak ground accelerations of 0.3g and 0.$g.

vii NUl(EG-1407

c

= Esecutive Commary j

i

'l

[b

Review level carthquakes (RI.I!) were assigned based on 2.

Identify significant changes since operating license

.the 1.LNL and EPRI hazard estimates, sensitivity tests, issuance

' and seismological and engineering judgment.The use of 3.

Determine if the E.lant/facilitics desiEn meets cur-l the 0.3g Rll for rnost Central and Eastcrn United States rent (1975 SRP) criteria plants would meet IPEEE objectives. For some sites

' where the scismic hazard islow, a reduced scope margins if the above are not satisfied, one or more of the following j

^

ccthodology emphasizing plant walkdowns is considered steps should be taken to further evaluate the situation.

i g

e adequate For Western United States tites, other than

'r California coastal sites 0.5g RIE should be used. Meth.

Optional Effort-y odology upgrades include relay chatter, liquefaction, and 4.

Determine if hazard frequency is acceptably low plant walkdown enhancements for the NRC method; q

= guidance on alternative success paths for the EPRI Perform bounding analysis 6.

Perform a probabilistic risk assessment (PR A) ti m i s

Alternative Methods 1

. internal Fires.

'the staff recognizes that other methods capable ofidenti-

'Ihe internal fires IPl!EH can be accomplished by per-fying plant specific vulnerabilities to severe accidents

- forming a level I fire PR A. 'those issues identified in the may be acceptable. A licensee may request a review of any l

= Fire Risk Scoping Study should be addressed using plant

• other systematic examination method to determine its specific data nnd a specially Iallored walkdown procedure.

acceptability for IPlilill purposes.

j "Ihc guidance does not address a simplified fire methodol-Coordination-U ony currently being developed by the Nuclear. Manage.

ment and Resources Council (NUM ARC). 'lhis method-Guidance is provided to coordinate the IPElill process op

, be eviewed by the staff when it is completed with ongoing programs. 'Ihc first coordination level is a

among the major clements of the severe accident policy implementation, that is, cc:~lination among the IPl!EE ;

Higli Winds, Floods, and Transportullon and the internal events IPl!, containment performance im-L M

Nearby Facility Accidents provements, and accident management.'Ihe second coor-dination level is among the major elements of the IPlitiB,

.the recommended overall approach consists of a pro' that is, scismic events, fires, and high winds, floods, and

! gressive screenmg. the screenmg criterion for reportmg others. 'the third level of coordination is within cair potential severe accident sequences is consistent with major cicment of the IPElill.

mternal event IPUs. the steps in the progressive screen-

ing approach represent a scrics of' analysis in increasing Programs subsumed into the IPl!EE include the scismic

. level of detail, effort, and resolution, llowever, the licen-aspect of USl A-45 (Decay lleat Removal), G1-131 (In-see may choose to bypass one or more steps so long as the Core Flux Mapping System), and the Charleston liarth-1-

vulnerabilitics are cather identified or demonstrated to bc

. quake issue. Programs that ned to be coordinated with insignificant. The screening approach consists of the fol-the IPE!!B include USI A-46 (Scismic liquipment Quali-p lowmg steps:

f cation which also covers the scismic spatial interaction -

of USI A-17 and the concern of USl A-40 for the scismic AH Plantst capability oflarge safety related above ground tanks)and i 1.

Review plant. specific hazard _ data and licensing GI-57 (liffects of Fire Protection System Actuation on :

bases Eafety Related Equipment).

NURl!O-1407 vnt i

a

=

l ACKNOWLEDGEMENTS

'this document represents the staff position on the Indi-contributors to the process; they are named below. In vidual Plant fixamination for severe accident vul-addition, significant input was received from consultants nerabilities due to external events (IPl!!!!!). Representa-and contractors to the NRC, who are also named below.

tives of both the Office of Nuclear Regulatory Research lidward llill of NRC provided technical editing.

and the Ol'fice of Nuclear Reactor Regu'ation were at,tive NRC

'Ihomas !!. hiurley

' thomas L King l'emetrios 1. llasdekas Iawrence C. Shao Nilesh C, Chokshi Pei Y. Chen Warren ) *'.auers David C. Jeng P.T.Kuo Glenn 11. Kelly Roper hi. Kenneally William D. Ileckner

' thomas hl. Cheng Jack Strosnider Kazimieras hi. Campe Rex 0. Wescott Conrad hicCracken Adel A. lil.llassioni John 11. Flack David P. Notley Jocelyn A. hiitchell

'!honms ht. Novak Jarnes !!. Itichardson it. Wayne Ilouston Guy A. Arlotto lirian Sheron Goutam llacchi Andrew J. hiurphy Leon itciter T. Y. Chang John T. Chen Consultanis R. J. Iludnitz l'uture Resources Associates, Inc.

O. I!. Cummings 1 awrence 1.ivermore National 1.aboratory R. P. Kennedy Structural hicchanics Consulting. Inc.

hi. K. Ravindra

!!Q17. lingineering. Inc.

Contractors and Subcontractors P. Amico S:icnce Application International D. L 11ernreuter lawrence 1.ivermore National laboratory ht. P. Ilohn Sandia National laboratory R. J. Iludniti Future Resources Associates,Inc.

D.11. Chung Iawrence Livermore National laboratory

11. C. Davis lawrence 1.ivermore National Ir.boratory G.S.liardy lawrercc 1.ivermore National laboratory J. R. hicDonald Texas Tech University R. C. hiurray lawrence 1.ivermore National 1.aboratory A. hl. Nafday 11Ql! lingineering, Inc.

P. G. Prassinos lawrence 1.ivermore National laboratory

. ht. K. Ravindra 11011 lingineering. Inc.

J. Savy lawrence 1.ivermore National laboratoiy I

ix NUR11G-1407

i 1 INTRODUCTION 1.1 llackground 1,2 IPEEE Objectives The objectives of the IPll!!!!, which are similar to the On August 8,1985, the Nuclear Itegulatory Commission objectives of the internal event IPli, are for each licensee:

issued a policy statement on severe accidents (NitC, 1985). 'the Commission concluded, based on available 1,

to develop an appreciation of severe accident L

information, that existing plants pose no undue risk to the

behavior, public health and safety and that there is no present basis for immediate action on any regulatory requirements for 2.

to understand the most likely severe accident se-these plants. Ilowever, the Commission recogni/cs, quences that could occur at the licensee's plant, based on NitC and industry experience with plant-specific probabilistic risk assessments (Pitas), that sys.

3.

to ghin a quMitative understanding of the overall tematic examinations are beneficial in identifying plant-probability of core damage and fission product re-specific vulnerabilities to severe accidents that could be leases,and fixed with low-cost improvernents. As part of the implementation of the polky statement, the Commission 4.

if necessary, to reduce the overall probabilities of issued Generie letter 88-20 (NitC,1988 and 1989),

core damage and radioactive material releases by requesting that each licensee conduct an individual plant modifying, where appropriate, hardware and proce.

examination (IPl!) for internally initiated events.

dures that would help prevent or mitigate severe accidents.

Itisk assessments indicate that the risk from external Ilowever,it was recognized at the outset that the external events could be a u,pmficant contributor to the core dam-event initiators could not necessarily be treated in exactly the same manner as internal event initiators in the are in some instances. Ilowever, licensees were re-quested to proceed with the examinations only for mter-implementation of the Severe Accident Policy. 'llus is because the sources and treatment of uncertainties in nally mitiated events (incigding internal floodmg) in Oc-estimates of core damage frequencies for external and nenc 1.ctter 88-20. I!xammation of severe accident vul-internal events can be quite different. In addition, some nerabilitics due to externally imtiated events (IPlilil!)is proceeding separately and on a later schedule to allow the methods have been developed for evaluating external staff to carry out auditional work (SliCY-88-147) to hazards and identifying vulnerabilities that do not pro-(1) identify w hich external hazards need a systematic exa-duce estimates of damage frequency. For examp!c, seis-mic margins methods produce estimates of seismic hazard mination, (2) identify exammation methods and develop procedural guidance, and (3) coordinate the IPlilili with levels of high confidence tow probabihty of failure (llCLPF) for a plant rather than estimates of core dam-other ongoing NitC programs that deal with various ti age probability.

pects of external event evaluations to ensure that there is no duplication of industry efforts *

'therefore, the staff determined that an explicit estimate of core damage frequency was not needed to me;t the intent of the Severe Accident Policy and would not be a To accomplish these objectives the staff established the requirement of :he IPl!IIII.Thus, objectis e 3 above would

!!xternal Fvents Steering Group (!!CSG) in December be addressed only indirectly for some methods to be used 1987 to make recommendations regardmg the scope, in the IPlilih.

1 methods, and coordination of the IPlil!!! (lleckjord, 1987, 1988). Specifically, the liliSG is responsible for developmg broad guidance for dealing with (1) cxternal W Pul'WSO Of I)M'ulMid events on a generic basis both oigannationa'ly and tech-

'Ihe purpose of this document is to provide guidelines for nologicahy and (2) the implementation of the 4evere acci-conducting the individual plant examination of external dent policy with respect to external events. The lillSG events (IPlilil!) and to provide guidelines on the struc-established thtee technical subcommittees dealing with ture and content of the IPl!!Ils submittal. Th; external carthquakes (seismic events), internal fires, and high events recommenoed for inclusion in the IPlilili are l

winds, floods, and *other" external events. The subcom-identified in Sectmn 2. Acceptable methodologies for mittees were chartered to define the scope of the external performing an 1?lilill along with upgrades to reflect J

events examination, identify acceptable examination state of-the-art i.nprovements are identified in Sections 3 methodologies, and coordinate ongoing issues and activi.

through 5. Sect on 3 addresses the seismic portion: Sec-ties (for example, Umesolved Safety issues and Generie tion 4 the intetnal fires portion; and Section 5 the high Issues).

winds, floods, and other portion of the IPlil!!!.

e I

NUlt!!G-1407

Coordmation between the IPElil! and the internal events electrical equipment, diesel peripherals, structural IPl!, other external events, and ongoing programs within failures, and equipment anchorages.

cach external event are provided in Section 6. A summary of documentation and reporting guidelines is provided in 2.

New data such as the occurrences of larger than Section 7, anticipated carthquakes and the development of new hypotheses indicate that the plant specific scis-mic hazard may be quite different from that envi-2 EVENTS EVAL,UATED FOR sioned at the time of licensing and makes it difficult INCLUSION IN Tile IPEEE to rule out scismic events on the basis of initiating

'lhe external events considered, consistent with past probabilistic risk assessments (PRAs) are those events 3.

Itased primarily on their vintage, the current popu-whose cause is external to all systerns used in normal lation of plants exhibit various levels of seismic de-operation and emergency operation situations. Internal sign requirements and margin. Some of the very fire and internal flood are external to the " system" and early plants have been backfitted under the System-therefore have been considered as external events in past atic !! valuation Program to ensure certain rnargins PRAs. Ilowever, internal floods are being considered in for safe shutdown using criteria different from cut-the internal events IPE process (NRC,1988).

rent licensing criteria.

In supporting the implementation of the Severe Accident 4.

'!here have been modifications to plants since their Policy, a study of the risk of core damage to nuclear power original designs, for instance, the reduction of snub-plants in the United States due to externally initiated bers at some plants. 'Ihese changes, in part, have events was performed. *lhe objective was to determine relied on existing conservatism or risk based argu-which esternal initiators have the potential of initiating an ments (e.g., l OCA + SSE combinations). The sys, accident that mry lead to severe reactor cue damage or tematic examination of plants by the IPE and IPE!!E large mdioactive release to the environment. Scismically will give an integrated picture of plants as they exist.

ini;iated events are investigated in NURl!G/CR-5042, it will also allow an mtegrated evaluation of the Suppl.1; nternal fires, high winds / tornadoes, external effects of individual changes made to plants over flomis, and transportation accidents are investigated in time.

NUREG/CR-5042; "other external events" are investi-gated in NUREG/CR-5042. Suppl. 2.The "other exter.

5.

'lhere are unresolved safety issues and generic is-nal events" covered are nea 9 industrial / military facility sues (e.g., USl A-45, USI A-46) that are in various acciden's, on. site hazardous material storage accidents, stages of implementation.The IPE/IPElili provides severe temperature transients, severe wea'ncr storms, a convenient as well as meaningful framework for lightning strikes, extemal fires, extraterrestrial activity, addressing many of these issues.

volcanic activity, carth movement, and abrasive wind-

. stonus.

6.

PR As and seismic margins evaluations have resulted in cost-effective plant specific improvements.

Some external events may not pose a significant threat of

.iherefore, the seismic external hazard should be m. -

a severe accident to all plants, some events may have been cluded in the IPEEh, considered in the plant's design to a sufficient degree, and some events may have been or will be reviewed under ongoing programs at some plants.The staff's evaluation 2.2 1niertial Fires and recommendations are contained in the following sectione liased upon the examination of past fire PR As, the contri-bution of internal fires to the probability of core damage may be significant and is very plant specific (NUREG/

2.1 Seismic Events CR-5042).110 wever, the numerical results always con-tain large uncertainties. 'the fire risk scoping study The following are based upon an examination of current (NURl!G/CR-50SS) further confirms the following:

seismic design critena, prev ous and ongomg seismic is-sues and programs, and scismic PRAs:

1.

The overall fire-induced core %.ge frequency for the four plants studied (Seahrook, Oconce, Limer-L Mean seismic core damage frequencies calculated ick and Indian Point) increased from the original from past PRAs (NURliG-1150, NUREG/

PR A studies even though, for certain fire scenarios, CR-5042, Suppl.1) have been found to be in the there was a net decrease. For all plants reviewed, range of IE-4 to IE-6 per year. Identified vul-fire continues to represent a dominant risk nerabilities are plant specific and include yard tanks, contributor.

NURiiG-1407 2

2.

Most initiating event ficquencies were increa,ed 2.4 External Fl(xxis based on a much more complete data base avai'ab4 on fire occurrences in nuclear power plants. Under l'or plants designed against current criteria as described currently applied risk assessment methodologies, in llegulatory Guide 1.59 and applicable Standard lle-this increase in initiating event frequen,y alone re-view Plan Sections, particularly Section 2.4, fkxxis pose sults in a direct increase in overall fire-induced core no significant threat of a severe accident because the damage frequency with all other factors remaining exceedance frequency of the design basis fhxx1, excluding constant.

fl(xxis due to failure of upstream dams, isjudged to be less than III-5 (Chery,1985), and the conditional core dam.

3.

Use of an expanded data base on historical fire sup-age frequency for a design basis flood is judged to be less pression times for nuclear power plants resulted in a than ll!-1. *lhus core damage frequencies are estimated suppression probability distribution with a lower to be less than 111-6 per year for the plant designed probability of suppression within a given time than against NI(C's current criteria.1lowever, the latest prob-that assumed in the original risk assessments. Under able maximum precipitation (PMP) criteria published by current methodologies, this again results in an in-the National Weather Service (NWS) call for higher rain-crease in fire. initiated core damage frequency with fall intensities over shorter time intervals and smaller all other factors rem dning constant, areas than have previourly been considered; this could result in higher site fimxling levels and greater roof pond.

U xlated information on the ignition and damage ing loads than have been used in previous design bases 4.

I thresholds of mble insulations in some cases re-(GI 103). Licensees are requested to assess the effects of sulted in lower thermal damage limits. In some applying these new criteria to their plants in terms of cases, no change in damage limits was required. A onute flooding and roof ponding. Also, some older plants decrease in the assumed thermal damage limits may have higher potential risk and need systematic exami.

would, in general, be expectet,I to lead to increased nations for plant specific vulnerabilities.

f estimates of fire-initiated core damage frequency.

5.

Plant modifications made as a result of Appendix R 2.5 Transportalion and Nearby i

requirements redui.d core damage frequen':y at in-Facility Acciderits I

dian Point and 1.imerick for the acquantified areas by factors of ten and three, respectively. For

.Ihese ewnts consist of accidents related to inmsporta-l Scabrook, the identified Appendix R plant moJifica-t,on and accidents at mdustrial and military facilities.

tions did not affect the requantified core damage Plants designed against N RC's current eriteria (NURl!G/

scenarios for internal fires. The Oconce PRA had CR-5042) should have no signiheant vulnerability to se-alreadyincorpotuted Appendix R modifications and vere accidents from these events because the imtlators s

no mothfications subsequent m its performance considered in the design should have a recurrence fre-werc identified. Ilence no Appendix R impact cculd quency less than 11!-6 or have been shown through a j

boundmg analys,s not to affect the plant, flowever, be identified for either Seabrook or Oconce.

i changes mt i aave occurred since the ongmal design and j

6.

A number of issues that were not addressed in the there, may oc neeptions that need some systematic past fire PRAs (effectiveness of fire brigade, effee, exammation, Also, spmc older plants may not meet the tiveness of fire barrier, seismic / fire interactions, NRC's current critena and need systematic examinations i

control system interactions, and effects of fire for plant specific vulnerabilities.

suppressants on safety equipment) could increase vulnembility to fire.

3,(, ygnjng Therefore, based on the above studies, the internal fire 1.iphtning has been gerienced in many nuclear power harard should be included in the IPlilili, plants in the United States (NUR EG/CR-5042, Suppl. 2; I

Ai!OD,1986; ACRS,1989). The impact of lightning on j

2.3 1ligli Wiiids arid Torriadoes plant opemtion and the vulnerability of plants to a severe j

accident due to lightning have been exammed. ihe major For plants designed against NRC's current criteria, these conclusion is that the primary impact of lightning on nu-events pose no significant threat of a severe accident clear power plants is the loss of olfsite power. The loss of because the current design criteria for wind are domi-offsite power is included as part of the internal events nated by tornadoes having a frequency of exceedance of IPli, and exammation for vulnerabilities due to this espect f

about 111-7. Ilowever, older plants and some modern of lightning is therefore already included in the Ipli proc-i plants having facilities not designed against these criteria ess.The staff has concluded that, in general, other effects need a systematic examination to identify plant specific of lightning on nuclear power plants are insignificant.

vulnerabilities (N URl!G /CR-5042).

However, for certain sites where lightning strikes are t

I 3

NURiiG-1407 I

s

likely to aficct more than just loss-of-offsite power, fur-2.8 Severe Weather Storms ther examination on lightning effects may be warranted.

Severe weather storms (icestorm, hailstorm, snowstorm, llased up(m an examination of historical data on light.

duststorm, sandstorm) accompanied by strong winds have ning, as well as knowledge of plant systems, the staff caused several complete and partial losses of offsite concludes:

power (NUltliG/CR-5042, Suppl. 2). The potential ef.

fect of loss of offsite power and station blackout will be addressed in the internal event IPli; thus severe weather 1.

IJghtn.mg has typ.ically caused partial or complete loss of offsite power, which is the main impact of storms need not be repeated in the IPl!!!!!.

lightning and which is already being examined as part of the internal events IPIs.

2.9 External Fires (Forest Fires, Grass Fires) 2.

I.ightning is much less likely to affect the onsite power system.

'these are fires occurring outside the plant site boundary.

Potential eff' cts on the plant could be loss of offsite 1.i hinin8 asaffected safety-related ecllsi 3 ment and pown an ce adon t

Nant undahn and 3*

8 h

I has caused reactor trips, but these events have not possible control room evacuation. Usually, external fires been significant in terms of impact on the plant.

are unable to spread onsite becausc of site clearing during the construction stage (NUltliG/ Cit-5042, Suppl. 2),

llowever, there has been one instance during which a 4.

Safety systems (e.g., <!iesel generators, electrically nearby forest fire caused a partial loss of offsite power, powered pumps) are not normally in operation.

• lhe effect ofloss of offsite power will be addressed in the

'thus, while control systems may be damaged, the internal events IPli and need not be repeated in the safety systems are less susceptible to damage and IPl!!!!!. 'the other effects have been evaluated during may be manually activated.

operating license (OL) review against suf ficiently conser-vative criteria; thus they do not need to be reassessed in 5.

Redundang of safety systems and the capability for the IPlilifi.

recovery of systems (replacing fuses or resetting breakers) further reduce the likelihood that the low frequency of lightmng damage events will result in a 2.10 Extraterrestrial Activit)'

severe r,ecident, (Meteorite Strikes, Satellite Falls) 11xtraterrestrial activny is considered to be natural satel-

'lhe staff has judged that the probability of a severe acci, lites such as meteors or artificial satellites that enter the dent caused by lightning (other than one due to loss of carth's atmosphere from space. Because the probability offsite power) is relatively low and further consideration of a meteorite strike is very small (less than 111-9) on lightning dieets should be performed only for plant (NURiiG/CR-5042, Suppl. 2), it can be dismissed on the sites where lightning strikes are likely to cause more than basis of its low initiating event frequency.

Just loss of-offsite power.

2.11 Volcanic Activity

'2.7 Severe Temperature Transients HS' ""'3" ' P *" P ""' SU'S "" '" I ' ""ay from l

(Extreine IIcat, Extreine Cold) active volcans to expect any effect at the plant, so most Severe temperature transients may affect nuclear power licensees need not consider the vokanic effects. Ilow-plants in the United States (NURiiG-1032).1-lowever, ever, a few sites may need to consider the effects of the effects are usually limited to reducing the capacity of volcanic activity (NURI!O/CR-5042, Suppl. 2) because the ultimate heat sink and loss of offsite power (NURiiG/

of the locations of active volcanoes.

CR-5042, Suppl. 2). ' Die capacity reduction of the ulti-mate heat sink would be a slow process that allows plant 2.12 Stiniinary operators sufficient time to take proper actions such as reducing power output level or achieving safe shutdown, in summary, based on the above evaluation, five events if necessary, and maintaining the plant in a safe shutdown need to be included by all licensees m the IPf!!!!!: scismic condition.*lhe ot'ier potential impact on the plant, loss of events, internal fires, high winds, floods, and transporta-offsite power, will be considered within the realm of the tion and nearby fac;lity accidents. All licensees should station blackout rule (NRC,1988b) and the internal event confira, however, that no plant unique external events IPl!."Iherefore, the temperature transients need not be known to the licensee today with potential severe acci-addtpssed in the IPl!!ill, dent vulnerability are being excluded from the IPlilifi.

NURl!G-1407 4

i 3 ACCEL'TAlllE METilODOLO-PRA calculations that account for all uncertainties are C3' '3 "*"Ptable. Howem, the staff beheves that, for 7

GIES I?OR l'ERFORMING Tillt the scismic 11,lil!!!, it is not necessary to carry out com-SEISMIC li'EEE, plete uncertainty quantifications defining a distribution of comdamage frequendes in order to identify vul-For the purposes of an IPlilil!, two methodologies are nera hues.

an Mnt esumau,on usmg a single hazard considered acceptable to identify potential seismic vul-C".c (rather than a family of hazard curves) and a single nerabilities at nuclear power plants.*Ihe first is a seismic fragtlity curve (rather tigm a family of fragtlity curves) for probabilistic risk assessment (NURl!G/CR-2300, NURiiG/CR-2815, Vol. 2), the second is one of the scis-C".2 component is sufficient to get insights into potential seismic vulnerabilit,es, hican pomt estimates usmg haz-i mie margins methodologies (Shihi) described in

"'d curves described in NURiiG/CR-5250 and til RI NUlti!G/CR-4334 and I!PRI NP-6041 or the reduced N1,-6395D should be obtained.,Ilus wdl encourage the Shih1 described later in this section.

most pertment results/msy; hts from the scismic portmn E

U in meeting the objectives of the IPl!!ill, the examination should focus on qualitative insights from the systematic

'the above point estimation approach is valid only be-plant examination rather than only on absolute core dam-cause of the iPlilil! objective: to identify dominant se-age frequency estimates. Guidance for performing the quences and components and where possible rank them, scismic IPlilill using a PRA or margins methodology is (this point estimate should not be confused with a " Phase provided below.

1" type PRA analysis where point estimate calculations are used only to define t copes for more detailed Phase 11

'*d Ph"S" 'll "'"di'5)' I" agilities used in this point esti-

• 1*1 Seismic PRA mate.where possible, sho tid be plant specific and rigor.

'lhis discussion deals with the use of PRA techniques in ous to be able to identify d iminant components and rank the seismic IPlilili.'lhe PR A should be at least a Irvel 1 them. Correlations and oil er aspects of analysis should plus containment performance analysis. The basic cle.

be treated so that, when a tean seismic hazard curve is ments are (l) harard analysis, (2) plant system and r.true, used with the mean plant fragility curve, the resulting ture response analysis,(3) evaluation of component fra.

core damage estimate approximately represents the mean estimate that would be derived from the full uncer-gilities and failure modes,(4) plant system and sequence analysis, and (5) containment and containment system tainty analysis.

analysis including source terms, to identify unique seismic The recommendation of performing point estimation sequences or vulnerabihties different from the mternal type calculations is made primarily to highlight insights event analysis. Specific guidance and enhancements are needed for the severe accident behavior perspective.This provided for licensees performing a new I R A or updating sould not be construed as deemphasizing or igno ing an existing seismie 1 R A.

tmeertainties. Analysts are encouraged to ??.ke ca eful study of the origins of the possible 'wrtaintics. i".clud-3.1.1 New Seismic l'RA Analysis ing those that arc hardest toquan7.anyof tneinsights obtained from a PRA analysis a udtained by trying to 3.1.1.1 General Considerations gain a better understanding of the uncertainties. Consid-1.icensees choosing to do a seismic PR A built on an inter.

cration of uncertainties may affect how the results of a nal events PRA should be aware of important considera.

PRA are implemented in plant changes, tions that, if incorporated in the planning of the internal 3.1.12 llazard Selection events PRA, will minimize their resource expenditure and speed the staff reviews. For example, (1) a well.

For the Ur.ited States east of the Rocky hiountains, two organized walldown team and a properly planned highly sophisticated scismic hazard studies were con-walkdown will enable many issues to be addressed at the ducted by Ir,wrence 1.ivermore National laboratory same time (2) the independent peer review group should (1.1.NL)(NURiiG/CR-5250) and the lilectric Power Re-consider the need to review both internal and external search Institute (l!PRI)(llPR1 NP-6395-D). For many event analyses:(3) fault tree analysts for internal events sites, these studies yield significant differences at the should be aware of spatial interactions (including internal low-probability and high level ground motions. The in-flooding effects) and common.cause effects and the cull-itial PR As carried out using these estimates (Surry and ing or pruning of trees should be done with these consid-Peach llottom in NURl!G-1150) indicate that, despite crations in mind; and (4) internal event models should be large differences in absolute numerical estimates, the developed knowing that, in the seismic analysis, the fra-identification, nmking, and relative contributions of the gilities of a component are sensitive to elevation. Also, a dominant seismic sequences are virtually the same for component and its peripheral equipment may have differ-both 1.1.NI,and !!PRI hazard estimates.'this equivalence ent fragilities.

is apparently due to the fact that the slopes of the seismic 5

NURl!G-1407

hazard curves are not significantly different over those 3.17.3 I ragility Estimation ground motion levels that,in conjunction with the fragili-

'lhe follow ng guidance on fragility estimation is included ties, control the relative distribution of seismically in-to clarify the use of fragility in the context of the

• point

)

duced core damage frequencies. Although these results esGmation approach discussed above. Details and meth-are very encouraging, there is no guarantee that this wdl ods for fragility and high coafidence-low probability of be true for all sites in the Central and liastern United failure (llCl. Pit) calculations are discussed in a num-States.

ber of references, for example, NURiiG/ Cit-2300, NUltliG/CI(-4334, I!PRI NP 6041, and NURl!G/

'lhe staff position is that, while a full scismic hatard un-c)(.5270. lt is recogniicd that large uncertainties exist in certainty analysis is not neccuary in performing a scismie fragilities estimation (NUltlIG/ Cit-5270). A perspective PI( A for the lPl!!ill, mean (arithmetic) Farard estimates on how this uncertainty affects the results of analysis from both the Ll.NL and EPlti should be used to obtain (numerical and other insights, for example, dominant se-two Mferent point (mean) estimates.The use of both of quences and components, should be maintained.

these estimates will sene to identify dif ferences,if any,in the delineation of dominant scisinic sequences (minor Consistent with the point estimation approach, one can variations in contributions and rankings are anticipated).

use a single mean component fragility curve for each Such differences would have to be addressed by the licen-compenent and hence for sequence level and plant level see in its identification and listing of vulnerabilities.The assessments. This mean curve is defined by the median 2

use of both the 1.LNL and !!Plti mean hatard curves has capacity, tt, and composite uncertainty,Ik, where Sc -

another advantage in that the extent of uncertainty will BR + lhd, when Br and Bu are estimated separately Br become obvious and the emphasis on the bottom line and Bu represent random uncertainty and modeling tm-numbers is reduced.

certainty, respectively. IL is also acceptable to use a family of fragility curves instead of a si. gle curve.

i For plants in the Western United States, for which there gle mean frag lity curve is available, ilCLPf,-

Men a sm.

i are no counterparts to the 1 LNL and liPlti studies, a licensee should conduct its own study to define the mean capacuy for a comp (ment (sequence or plant) can be ap-harard estimate for use in the IPlil II. The licensee P"? mated by -2.3 Bc below he median (i.e.,1% com.

posite probability of failure is essentially equivalent to Should also provide reasonable assurance that any signifi-959c confidence of less than 5% probability of failure).

cant uncertainty in those elements of hazard (for exam, plc, slope) that control the identification, ranking, and While developing sequence-level and plant level fragdp ties, licensee should retain the ability to report ilCLPI s relative contribution of seismic contributors to core dam-wnh ami without nonseismic failures and human actions.

age is addressed in sensitivity studies. As in the Central and liastern United States, the identification and listing 3.1.L4 Seismic PI( A Methodology I.nhancements of vulnerabilities should take this uncertairity into ac-

count, liased on a review of past seismic PI(As, certain areas have been treated inconsistently or not at all.The follow-Most seismic PilAs use peak ground acceleration as the ing areas should be included:

hazard parameter. If this is done, spectral shapes that are L

Plant Walkdowns. Walkdowns are I1erformed to find consistent mth current estimates of ground motion as. designed, as-built, and as operated seismic weak-should be used. In the Central and Eastern United States, nesses in plants liach l,censee should perform a i

current spectral estimates can be found the LLNL and walkdown consistent with the guidelines described I!Pitt studies. Since similar spectral shape are obtained in the EPitt Scismic Margins Methodology (liPill from Lt.NL and 1 Plti hazard studies, separate analyses NP-6041)(team composition, documentahon, etc).

using both spectral shapes are not needed. Medium spec-tral shapes of 10,000 year return period provideel in 1

Relay Chatter. An acceptable procedure for address-NUltliG/ Cit-5250 along with van, ability estimates me ng relay chatter issues is described in liardy et al.,

recommended for use m the analyses. Other site specific 1989, llelays, in this context, include components spectral shape estimates may be proposed (that is. derived such as electric tclays, c(mtactors, and switches that from a suite of appropriate recorded carthquakes) 1 or are prone to chatter.De examination of the relay the Western United States, site-specibe spectral shape chatter effects (for example, the llatch margins should be established and used, evaluation) has resulted in large resource expendi-tures in terms of staff hours. Therefore, careful if an upper bound cutoff to ground motion at less than planning and use of generic insights, if they are ap-1,5 g peak ground acceleration is assumed, sensitivity plicable to the plant, are necessary, studies should be conducted to determine whether the use of this cutoff affects the delineation and ranking of 3

llCLPF Calculatiota Licensees should report seismic sequences, plant level, sequence level, and cornponent level i

NUltl10-1407 6

l

IICLPFs. In several PR As (for examplc, Millstone 3 in past PR As. It clays, in this cont ext, inc!ude com pm and Diablo Canyon), IICLPF estimates are reported nents such as electric relays, contactors, and along with other PRA results 'Ihese PRAs ;an be switches that are prone to chatter Licensees should used for guidance to derive llCLPFs from fragilities.

analyze the effect of relay chatter or determine that ilCI.PFs are to be reported both with and without the type of relays used in the safety systems are not the effects of nonscismic failures and human subject to relay chatter. Re~ "ts of this effort that actions.

lead to a PR A revision or p.

t fixes should be re-ported. Additional guidance is provided in liardy et 4.

Liquefaction. 'the potential for soil liquefaction and al.,1989.

associated cifcets on the 1,lant need to be examined for some sites because of specific site conditions.

4.

Nonscismic Failures and Human Actions. In several

'the impact on plant operation should be assessed seismic PR As, nonseismic failures ( c.g., failures of from the point of view of both potential for and the auxiliary feedwater system and failure of feed consequences of lique faction. Procedures for assess.

and bleed mode of core cooling, battery depletion, ing soil liquefaction are described in !!PRI NP-604 L power operated relief valve failures)and human ac-tions (e.g., delays or failures in performing specified actions, or operator misdiagnosis of a situation re-3.1,L5 Containment Performance sulting in an improper action that is not be related to the actual, current plant situation) have been impor-

'lhe primary purpose of the containment performance tant contributors to seismically induced core damage evaluation is to identify scquences tmd vulne abilitics that frequencies or risk indices. Unless nonseismic fail-involve containment, containment functicas, an I con-urcs are considered, improper decisions may be tainment systems (c.g., fans, sprays, igniters and suppres-made regarding plant modifications or procedural sion pools) seismic failure modes or timin g that are sig-changes.

nificantly different from those found in th: IPli internal eventr, evaluation. Additional guidance is presented in The licensee has the option to expand its PRA or Section 3.2.6.

demonstrete that the exclusion of nonseismic fail-ures will not significantly alter the PRA results or 3.1.2 Use of an histing Pila insights. 'lhe scope of nonseismic failures and hu-man interactions that might affect seisnue sequences

'the use of an existing seismic PRA to address the seismic should be defined by the licensee based on the inter-IPI!!!!! is acceptable provided the PR A reflects the cur.

nal events analyses.

rent as-built and as-operated condition of the plant and 5.

Liquefaction. The potential for soil liquefaction and some of the deficiencies of past PR As discussed below are adequately addressed.

associated effects on the plant need to be examined for some sites because of specific site conditions.

1, Hazard Sclection. For PR As at sites east of the Rocky The impact on plant operation should be assessed Mountains that did not use the LLNL and liPRI from the point of view of both potential for and hazard estimates, sensitivity studies should be con, consequ,ences of liquefaction bed in !!PRI NP-60 Procedures forasses.v ing soil hquefaction are descri ducted todetermineif the use of these results would affect the delineation or ranking of seismic se-quences. For PRAs in the Western United States' 6.

HCLPF Calculation.1.icensees should extract and the sensitivity studies should be carried out to deter-report plant level, dominant sequence level, and mme the effect of uncertainty m hazard on the de-dominant-component level llCLPFs both with and imcation and ranking of scismic sequences.

w thout the effects of nonscismic failure and human 2.

Walkdowns. Since a walkdown is considered to be one of the most important ini.redients of the scismic 7.

Containment Perfonnance. Licensees should ensure IPlifill, a supplementary walkdown in conformance that the performance of containment and contain-with the procedures described in the liPRI margin ment systems are addressed. Sectie-3.2.6 contains methodology (EPRI NP-604 !) (team composition,

guidance, documentation, etc.) should be perforrned. Il may be necessary to amplify the earlier analysis based on the 3.2 Scistille Margins Methodologies walkdown outcome, 'Ihese results should be re-ported.

This discussion deals with the use of the seismic margins methodology in the seismic IPEli!!. Specifically, guidance 3.

Rchiy Chatter. Relay chatter effects either have not and enhancer ents are provided for a licensee using been considered or were assumed fully recoverable either the NRC or liPRI margins methooology.

7 NURl!G-1407

1 l

3.2.1 General Considerations sites cast of the llocky Mountains are not subject to the same level of carthquake hazard. For some sites where "the seistnic margin methodology is considered accept-the seismic hazard is low, a reduced-scope ma.

ap-able for addressing seismic concerns in the severe acci-proach centered on waltdowns is acceptable. I a 'wo dent policy implementation.Two rnethodologies are cur-liastern U.S. sites where the staff studies indicated.aat rendy available: one developed under NitC sponsorship the seismic hazard is relatively high and a 0.3g R1Ji is and the other developed under lipRl sponsorship 'lhe judeed not adequate. llecause the component capacity stafI has determined that both methods (with the noted data sets associated with the margins methods are catego-enhancements) will adequately address IPlilill objec-ri/ed at two screening levels,0.3g and 0.5g, a 0.5p itLl!

tives.*lhe two methods use different system analysis phi-should be used. l'or western sites other than California losophies. The NitC method is based 01 an event / fault coastal situ, a 0.5g Ill.li should be used for the rnargin tree approach todelineate accident sequences.1%r exam-approach.The Ill.lis defined for U.S. sites, as w ell as sites plc for PWits, two safety functions are considered to be that can perform a reduced scope SMM,are presented in most hnportant to plant seismic safety: reactor sub-Tables 3.1 and 3.2 'the seismic margins evaluations criticality and early emergency core cooling. lf these func-should utili/c the NUltliG/ Cit-0098 median rock or soil tions are ensured for a given earthquake, there is high spectrum anchored at 0.3g or 0.5g depending on the g-confidence that core damage would not occur at that level and primary condition at the site. Further discussion level.The FPiti methodology is based on a systems "suc-on the review level earthquake is presented in Appen-cess path" approach.This approach defines and evaluates dix A.

the capacity of those components required to bnng the plant to a stable condition (either hot or cold shutdown)

The ground motion should be considered at the surlace in and maintain that condition for at least 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. Several the free field. If secondary conditions such as shallow soil possible success paths may exist, conditions are being considered, appropriate procedures should be used to determine the free field motion in the

!!ach licensee should examine its plant critically to ensure vicinity of those affected structures and components, that the generic insights used in matgins methodology and the capacity evaluation of structures and compo-development to identify critical functions, systems, and nents should take into account soil structure interaction success path logie are applicable to its plant. This is par-effects.

titularly vital for older plants where systems and func-tions may differ greatly Irom the plants considered in the llecause recent ground motion estimates such as those development of the margins methodologies (NUltliG, included in the 1.1.NL and !!Pld hazard studies indicate Cit-4334, NUlti!G/ Cit-5076, and liPHI NP4041).

relatively higher ground motion at frequencies greater than 10 ilz than that shown in the NUltIIG/ Cit-0098 Iloth NltC and I!Pitt methmis should include an inde-spectrum, the margins evaluation of only nonductile com-pendent peer review to ensure proper implementation of ponents (if required), for instance, relays, that are sensi-the margins methodology. A peer review group meeting tive to high frequencies should be performed as discussed prior to a plant walkdown will provide insights into the in Section 3.2.3(1). No plant specific response analysis is appropriateness of the proposed plant walkdowns. Peer anticipated to address high frequency ground motion review group endorsement of the final results adds to the concerns. Ilowever,if a licensee decides to evaluate plant credibdity of the iICl.PF numbers. lleview group mem-response for high frequency ground motion, the response bers should have combined experience in the areas of spectral shapes derived from the appropriate sit e-specific systems engineering, seismic capacity engineering, scis-median uniform hazard response spectra (10,000 year mic Pit As, and seismic margins methodologies.

return period) shown in NUllEG/ Cit-5250 anchored at 0.3 or 0.5g should be used.

3.2.2 Iteview Level Earthquake and Assocl-nted Spectral Shapes 3.2.3 NitC Seismic Margins Methodology The scismic margins methodology was designed to dem-Several enhancements to the NltC methodology are onstrate sufficient margin over SSi! to assure plant safety needed before it can be used for IPlilitiimplementation.

and to find any " weak links" which might limit the plant

'these enhancements, along with the means to accom-shutdown capability to safely withstand a seismic event plish them, are described below; bigger than SSli.The scismic margins me thod utilizes two review.or screening levels geared to peak ground accel-1.

Rclay Chatter. llelays, in this context, include compo-erations of 03g and 0.5g. It is the staff's judgement that nents such as electric relays, contactors, and the use of a 0.3g review level carthquake (ItLli) for most switches tha' are prone to chatter.The NltC method of the nuclear power plant sites in the Central and liast-as originally developed did not address the relay crn United States (east of the Rocky Mountains) would chatter issue because of the lack of information on scrye to meet the objectives of the IPl!!!!!. Ilowever, all the subject. Ilardy et al.,1989, has summarized NUltliG-1407 8

a.

t

]

a 1

f results of several efforts in this area and has pro.

4.~

riant Walkdown. The walkdown shsutd be per.

[

- vided guidance to address this issue in an IPEEE forrned and documented in accordance with the rec-context. Relay chatter anal is co;.d be very re-ommendations contained in EPRI NP-604L source intensive, and caref I planning and use of genericinsights if they are applicable to the plant are 5.

1. CLPF Calculations. Two approaches, fragility necessary, analysis (FA) and conservative deterministic failure 3

margin (CDFM), for computing component and -

l

' Attempts to address high frequency ground motion plant ilCLPF6 are acceptable. By developing come amcerns by analysisis vc likely to entail extensive plete fragilities for components that remain in the ~

efforts including the deve opment of new and rnuch plant.levellloolean equations if the CDFM method L

- more complex building models that transmit and l is chosen to calculate plant ilCl.PF, it is possible to I

amplify high frequency input and generate accurate Inake plant ilCLPF statements with and without the mclusion of nonscistmc failures and human actions, i

l and meaningful floor spectra at high frequencies.

b f Estimates of high.ftcquency amplification in cabi-As noted in EPRI NP-6041, use of the Ocner;c -

I

. nets containing relays wdl also have to be developed.

I!quipment Ruggedness Spectrum (GERS) to esti-p Rather than usm, ganalysis.the followingaller. native mate llCLPFs should take into account the latest '

appsoach is acceptable:

results for ongoing work on the reconciliation of

~

GERS and liCLpF.

a.

Prepare a list of relays that are known to have j

high.frequeng sensitivity.

3.2.4 EPRI Seismic Margins Methodology bc Screen relays that are known to have very high Several enhancements to the EPRI methodology arc IICLPFs (that is, they can be climinated from needed before it cim be used for IPEEE implementation, s

further consideration without performing spe.

These enhancements along with the means to accomplish -

cific response calculations),

them are described below:

j 1.

tion of Mtunut Sumss Parla, eEM L c.-

) Assume that the remaining relays will chatter J at the review level carthquake.

methoddogy as c"~endy amsthuted requim t

evaluation of a pre. sed path and an alternative path.The NRC panel that reviewed the EPRI meth.

d.

Screen the remaining relays by either showing odology recommended:

that the 'clectrical circuitry is insensitive to high. frequency chatter or that they can be re-

" that a reasonably complete set of potential covered from changes of state and associated :

success paths be set down initially, rather than a

. false alarms.

very small number, since limiting the number of success paths too quickly can prevent the identifi.

m~':

e.

Finally, replace the remaining relays with ones cation of some plant. level llCLPF insights, and.

that are.not sensitive to high frequency (an can mask plant differences regarding defense in.

altertmtive approach is to show that the re.

depth.The Panel believes that preliminary analy.

maining relays are rugged by conducting tests).-

sis to narrow the number of paths to the required L

~ two or three should begin with the fuller set, and it.

Although stated in the context of high frequency recommends that this narrowing be documented i

ground motion, the above approach can be used to in detail."

]j i

iaddress the relay chatter issue.'

For iPEEE purposes,it is desirable that, to the maxi.

mum extent possible, the alternative path involve-

2. - Ltquefactmn.- An evaluation of the potential for operational sequences, systems, piping runs, and 1

hquefaction should be conducted; however, no guid' components 'different from those used in the pre.

-NP-6041 etm be used for gu}idance to carry out aance was provided m the NRC ' SMM. l!PR ferred path.The procedure used in the trial applica.

tion of the EPRI methodology (EPRI NP-6 M9) pro;

+

c liquefaction analysis.

vides an acceptable approach for use in selecting N - Nonseismic Failures and Human Actions:These ac-tivities should be included; guidance on including 2.

Nonscismic Tallures 'and Human Action Success s

a~

nonseismic failures and human actions is provided in paths are chosen based on a screening criterion ap.

4 NUREG/CR-4826 (Maine Yankee evaluation) and plied to nonseismic failures and needed human ac.

O lludnitz,1989a (extension of the Maine Yankee, tions. It is important that the failure' modes And 3

4

~

human actions are clearly ident;fied smd have low Catawba, and Ilatch analyses).

tfy 9

NUREG-1407 e

i l

enough failure p%1ities to not affect the seismic buckets) and seismic failure modes that are significantly margins evalua ~

3 screening criteria used in ddferent from those found in the IPE internal events the Maine Yan

j. ins evaluation (NUltEG/

evaluation.

Cit-4826) addresmg both single-train and multi-train systems is an acceptable approach. *lhe redun-

!!ach licensee should develop a plan to address contain.

dancies along a given success path should be specifi-ment performance during a seismic event. Some general cally analy/ed and documented when they exist. (in a guidance is provided based on past PRA experience and complementary sense, where a single component is some generic capacity estimates of typical components truly "alone" in perforrning a vital function along a involved in containment systems. I rom a survey of past success path, this should be highlighted too.) 'this PRAs (Amico,1989), it appears that high consequence information will sv.c to indicate the extent to which sequences involve gross structural failure of the contain-a single failure would or would not invalidate the ment itself or failure of major equipment or structures plant's ability to respond safely to a given earth-within f.he containment at very high accelerations quake !cvel.

(HCI.PF values gicater than 0.$p) and isolation failure due to seismically induced relay chatter.

3.

Mrof GERL Although not an enhancement per se, the use of the Generie liquipment Ituggedness Generally, containment penetrations art etsmically rug-Spectrum (G Fits) to estimate 1ICI.Pl s should take ped; a rigorous fragility analysis is needeu only at review into account the latest esults for ongoing work on levels greater than 0.3g, but a walkdown to evaluate for the reconciliation of Gl!RS and 1ICl.PF.

unusual conditions is recommended. An evaluation of the backup air system of the equipment hatch and personnel 4.

Relay Chaner for High-Ftrquency Gmumi Motion, lock th it employ inflatable seals should be performed at Guidance for addressing the high. frequency ground all review levch. Also, some penetrations need moimp, motion issue is discussed in Section 3.2.3(1).

and the possit ility and consequences of a cooling loss caused 'cy an c arthquake should be considered.

3.2.5 Iteduced scope Margins Method Valvea myolved in the containment isolation system are For sites where the scismic hazard is low, a reduced-scope expet ted to be seismically agged(NUltliG/ Cit-4734). A seismic margins alethod emphasizing the walkdown is walkdown to ensure that they are similar to test data and adequate.*lhorough walkdowns have demonstrated to be have known high capacities and that there are no spatial the most important tool for identifyig,scismic weak links interaction issues will suffice. Seismic failures of actua.

whose cor rection is highly cost beneficial. Applicable sites tion and control systems are more likely to cause isolatica are identified in Table 3.1, system failures and should be included in the examina-tion. For valves relying on a backup air system, the The initial steps of the full-scope margins methodology air system should also be included in the seismic up to and including the initial plant walkdown are per-examination, formed regardless of method selected (NRC or IIPRI).

liasically, pertinent activities up to and including the in-Components of the containment heat removal / pressure itial plant walkdown need to be performed. These activi.

suppression functional system that are not included else-

. here and are not known to have high capacities should ties include gathering system information, classifying w

frop. 'ine systems and identifying front line components, be examined. An example of such a component might be a cle ;ying support systems and identifying support sys-fan cooler unit supported on isolator shims. The tem components, and identifying plant unique features.

walkdown should include examination of such compo.

nents and their anchorages. Similarly, support systems Further guidance on the differencesbetween the reduced and other system interaction effects (e.g., reiay chatter) scope and full scope margins methods, that is, elements should be examined as applicable.

preserved and elements climinated are provided in Ap-pendix 11.

For Mark I and ice condenser containments, it is not feasible to screen out components (e.g., torus, ice bucket The evaluation should be documented in a walkdown support) on a generic capacity basis. The potential for team report and subjected to a peer review, accident sequences initiated by a containment functional failure should be examined.

3.2.6 Containntent Perfortnance The primary purpose of the evaluation for a seismic event is to identify sequences and vulnerabilities that involve A licensee may request a review of any other systematic containment, containment functions, and containment examination method to determine its acceptability for systems (e.g., fans, sprays, igniters, suppression pools, ice IPl!EE purposes.

NURl!G-1407 10

iiJ i

Table 3.1 Review level Earthqual.e-Plant Sites East of the Rocky Mountains c

Reduced Scope Program

. liig Rock Point

'Duane Arnold South Texas Turkey Point i

' Comanche Peak Grand Gulf St. Lucie Waterford

' Crystal River itiver llend 0.3g Arkansas -

Dresden Maine Yankee Robinson i

licaver Valley Farley McGuire Salem licilefonte Fermi Millstone Sequoyah liraidwood -

Pit / patrick Monticello Shoreham Ilrowns Ferry Fort Calhoun Nine Mile Point

' Summer lirunswick Oinna

' North Anna Surry Ilyron Iladdam Neck

'Oconec Susquehanna j

Callaway llatris Oyster Creek

'Three Mile Island Calvert Cliffs llatch Palisades Vermont Yankee

' Catawba llope Creek Peach llottom Vogtle Clinton Indian Point Perry Watts liar Cook Kewaunee Point llcach Wolf Creek i

Cooper IaSalle Prairic Island Yankee Rowe Davis Ilesse IJmerick Quad Citics Zion l

0.5 g#

Pilgrim Seabrook

.t N(Til!S:

' Sgweial attention to shallow soil conditiom b needed at these heations (See Section 3.2.2L

1. ILned on the staff studies, review level entthquakes Freater than 03; nre needed for these two sites. Ilecaws the comguncut capacity data sets awriated with the margins rnethals are categosized at two scieening hvch,03g and o.5g,it is necessary that the 111.11 for these sites be set at 0.5g.

I Table 3.2 Rniew Level Earthquake-Western United States Plant Sites

~0.5g

' Trojan

' Rancho Seco

' Washington Nuclear

'Palo Verde Seismic Margins Methods do not Apply to Pollowing Sites:

Diablo Canyon San Onofre

'N(rili

• Indicates a Western tJnited States sine whose defauh bin is 0.5g unless the licenser can demonstrate that the site haraid is similar to those sites cast of the llocky htountains that are found in the 03g bin.

Changes in the review kvel carthquake from 0,$g to 03g should te approved g rior to doing significant analysit i

i1 N UIE G-1407

4 ACCEPTAllLE METilODOLOGY 4.1.2 Calculate the Frequency of I? ire initia.

FOR PERFORMING Tile INTER, tion in Each Area NAL FIRES IPEEE

'this calculation is sensitive to location within a larger area, particularly if fuel loading conditions, cross zone For purposes of an IPElili, a lxvel 1 probabilistic r:,sk spreadmg potential, or other idiosyncrasics are consid-assessment (Pit A) is considered acceptable to identify cred. Also, the data base on fires in various areas should gotential internal fire vulnerabilities at nuclear power be coupled with location. specific information obtained plants. Some fire issues identified in the Fire Itisk Scop-from the plant walkdown, nd other experience to account ng Study, (1) seismic / fire interactions, (2) effects of fire for uncertainties.

suppressants on safety equipment, and (3) control system interactions, should be addressed in the IPhlill. The walkdown procedures of the 11lil!!! should address the W M)ze,for the Disabling of Critical above issues and shonid be specifically tailored to assess Safety I'unelions the potential vult abilities related to these issues. The 1)etermine the likelihood of equipment being disabled by licensee should use a plant. specific data base on fire bn-a fire. The areas to be addreased include:

gade training in the IPElill to assess the chectivenesuif manual fire fighting to determine the response time for 1.

Fire growth and spread, including the treatment of the manual fire fighters. '!ht licensee should also show hot gases and smoke.

the effectiveness of fire barriers in the IPliHi!. The cur-rent fire Pila method has its limitations (NUllI!O/

2.

Detection / suppression ef fectiveness and reliability, CR-5088,1489) and that significant

• engineering judg-ment" must be brought to bear once the PRA has betn 3.

Component fragility to fire and combustion pnxt-accomplished to allow for sensible application of the re-ucts.

sults. 'ihe staff believes that the type of acngineering judgment" needed to interpret the results of a Pila is 4.

Probability estimates (distributions) for fault tree fully within the competence of most fire safety experts, quantification.

including experts within the regulatory staff. Further, despite current limitations in the methodology, a fire 4.1.4 Identify l# ire induced initiating lhents/

" vulnerability scarch"in the spirit of the Severe Accident Systetus Analysis Policy Statement and the IPl! exercise is feasible, and such a vulnerability search need not wait for the comple-Perform in a fashion similar to an internal. initiator Pit A.

tion of further methodology development. Finally, in meeting the objectives of the IPlilill, it is desirable t 4.2 Use of an Existing Fire PRA focus on relatwe msights rather than on absolute core damage irequency.

'the use of an existing fire Pil A to address the internal fire IPlillE is acceptable provided the Pil A reflects the cut-rent as built and as operated status of the plant and the 4.1 New l,,re PRA Analys.is denciencies of past PRAs, identified in the fire risk scop.

u There are several different approaches for the analysis of ing study (NURl!G/CR-5088), are adequately addressed.

fires (NURiiG/CR-2300, 1983, NUREGiCR-2815, 1985, NURl!G/CR-4840,1990, and NUltliG/CR-5259, 4.3 Optlonal Methociologies 1990). A logic approach is described in Null!!G/

A licensee may request a review of any other systematic Cit-5259. Although not all fire 1 ras delineate their examination method to determine its acceptability for analysis steps m exactly the same way, the following steps, IPlil!!! purposes.

in one form or another, must be part of any analysis.

5 ACCEPTAllLE METilODOLOGY 4,1,1 Identify Critical Areas of Vulnerability FOR PERFORMING Tile lilGli

'lhe criterion is whether a fire could compromise impor.

WINDS. FLOODS, AND TRANS-tant safety equipment. limphasis should be placed on PORTATION AND NEARllY areas yhere multiple equipment could be compromised, FACILITY ACCIDENT IPEEE m particular, several trains of redundant equipment to perform the same safety function. Attention should be For the purposes of an IPliEli, the staff is recommending given to the potential for cross-zone spread of fire and the a progressive screening approach to idemify potential likelihom! that transient fuels might supplement fuels high winds, floods, and transportation and nearby facility already present in a zone, accident vulnerabilities at nuclear power plants. "Ihe i

NURl!O-1407 12

owners of the Trojan and Washington Nucicar Plant 2.

events will have been screened from further considera.

who are requested to evaluate the effects of volcanic tion by the staf f. For those external events not in either of activities in at.sessing severe accident vulnerabiistics, these categories, further consideration using the pro-should determine tf the secommended screening ap-gressive screenmg approach shown in Figurc 5.1 is proach is applicable to their unique situation, rec.ommended.

5.1 Introcluction 5.2 Analytical l'rocedure It is assumed that the 11'11 for internal events will be m.

progress or completed when the high winds, floods, and

'lhe steps shown in Figure 5.1 represent a series of analy-transportation and nearby facility accident portion of the ses in increasing level of detail, effort, and resolution.

IPill!.li is being performed. Some external events will be Ilowever, the licensee may choose to bypass one or more addressed in the internal events IPli analyses (e.g., the of the optional steps so long as the 1975 Standard iteview primary offeet of lightning is loss of offsite power, w hich is Plan ertteria are met or the potential vulnerabilitics are included in the internal events analyses); other external either identified or demonstrated to be insignificant.

RECOMMENDED IPEEE APPROACH FOR WINDS, FLOODS, AND OTHERS (1) Review Plant Specific Hazard Data and Licensing Bases (FSAR)

V (2) Identify Significant Changes,if any, since OL lssuanco I

Y NO (3) Does Plant / Facilities Design Moot 1975 SRP Criteria?

YES m

(Quick Screening & Walkdog{

OR > (4) is the Hazard Frequency Acceptathy Low?

YES f

NO OR m (5) Bounding Analysis YES m

(Response / Consequence) 7 NO I Y

U" > (6) PRA I

Y (7) Documentation 2'

(incl. Identified Reportable items and Proposed Improvements Note: Steps 4,5 and 6 are optional, liigure 5.1 llecommended ll'lilill Approach for Winds.1:loods, and Others 13 NUltllO.407

l l

l In peneral, the containment structure, equipment hatch.

Otherwise, one or more cf the steps in Sections 5.2.4, personnel air lock, and other penetrations are designed 5.2.5, and 5.2.6 should be taken to further evaluate the and constructed to have high capacities in resisting the situation.

cf fccts of high winds, floods, and overpressure induced by transportation or nearby facility accidents.'therefore, no 5.2.41)etermine if the llazard Frequency is additional cutainment performance assessment (beyond Acceptably lany (Optional Step) that discussed for the seismic portion of the IPlil l! m Sections 3.1.1.5,3.1.2, and 3.2.6) is needed unless a If the original design basis does not meet current regula-licensee predicts or identifies plant unique uccident se-tory requirements, the licensees may choose to demon-quences different from those determined by the internal st rate that the original design basis is sufficiently low (i.e.,

events IPl!.

less than 11!-5 per year)and the conditional core damage frequency is judged to be less than 11!-1.

5.2.1 1(eview I'lant specific llazard I)ata and if the original design basis hazard combined with the 1,leensing flaseS conditional core damage frequency is not sufficiently low, i.e., lower than the screening criterion of Ili-6 per year.

Alllicensees should review the plant design hanud infor-additional analysis inay be needed.

mation and the licensing bases, including the resolut.on of each event.

5.2.5 l'erform a llounding Analysis (Optional Step) 5.2.2 Identify Significant Changes Since 01, sanalyu... iendedtoprovidea conservat.ive calcula-Issuance

.I'lu.

sis m tion showing that either the hanud would not result in All licensees should teview the site for any significant core damage or the core damage frequency is below the changes since the issuance of the operating license with r eporting criten,on.The level of detail is that level needed respect to (1) military / industrial facilities within 5 miks, to demonstrate the point; judgment is needed for deter-(2) onsite storage or other activities involving hazardous mmmg the proper level of detail and the needed effort.

materials, (3) transportation, or (4) developments that could alicet the original design conditions.

5.2,6 l'erform o l'robabilistle Itisk Assess.

ment (Optional Step) 5.2.3 I)etermine if the I'lant/ Facilities 1)esign A probabilistic risk assessment (PR A) consists of the fol.

Meets Current Criteria lowing key elements: hazard analysis, Iragility evaluation.

plant systems and accident analysis (event / fault trees),

All licensees should compare the information obtained and radioactive material release analysis. The detailed Irom Sections 5.2.1 and 5.2.2 for conformance to current procedure is described in NUltliO/ Cit-2300, NUlli!G/

criteria and perform a confirmatory walkdown of the CR-2815, and NUltliG/ Cit-5259. A core damage fre-plant 'the walkdown would concentrate on outdoor fa-quency less than Ili-6 per year would screen the event cilities that could be affected by high winds, onsite storage from further consideration.The level of detailis that level of ha7ardous materials, and offsite developments,if the needed to demonstrate that the core damage frequency is comparison indicates that the plant conforms to the cut-low or to find vulnerabilities.

rent criteria (1975 NitC SitP criteria) and the walkdown reveals no potential vulnerabilities not included in the 5.3 Optional Methodologies original design basis analysts, it is judged that the contn-bution from that hazard to core damage frequency is less A licensee may request a review of any other systematic than 11!-6 per year and the IPlilili screening criterion is examination methcd to determine its acceptability for

met, IPlilill purposes.

)

NUltliG-1407 14 l

1 6 COORDINATION WITil ON, staff identified alternative approaches to certain GOING l'ROGRAMS dnign pnindms and modWe dons ni Om NIC criteria in the Standard lleview Plan to reflect the current state of the att and industry practice. 'the 6.1 IHirOduCl,lon concern for the seismic capaeny of safety related above ground tanks (at the SSli)is included in USl If unnecessary duplication of cffort is to be avoided, coor-A-46.

dination with ongoing programs is necessary. The first coordination level consists of the three major elements of the severe accident policy implementation, that is, coordi-3.

USI A-45, " Shutdown Decay Ileat itemovat lle-nation of the IPillil! with the internal events IPli and quirements," has the objective of determining accident management. The second coordination level whether the decay heat removal function at operal-consistsof the threemajorelementsof theIPlil!!!,thatis, ing plants is adequate and if cost beneficialimprove-seismic events, internal fires, and high winds, floods, and mentcouldbeidentified.USI A-45wassubsumedin others. 'lhe third coordination level consists of each the lPl!(Gl.88-20); therefore, the seismic adequacy rnajor element of the li'lil!!!, for example, scismic events of the decay heat removal system should be included and the ongoing programs related to : hat element.

in the seismic events review of the IPlil!!!.

t m A4. Seismie Qualification of liquipment in 6.2 I)CScriptioll of Origolilg l'rogratiis Operating Plants,,, has developed an alternative 6.2.1 li'E l'rogram 1(elated to Internal Events

"* @"d ""d "* "'"" I' I" ('" "" " '"' I I"SI"E requ rements) to venf) the seismic adequacy ol-equ5nwnt in operating plants with construction in Generic 1 etter 88-20 (NRC,1988), each existing plant pum upp anon o ted Mom abut E AH was requested to pe form a systematic examination to

" " b""

*d to th eMng sak identify any plant +pecific vulnerabilities to sevete acci-t@wn cadquaWM'Du'wopwM A4>

dents and report the results to the staff.The process was as n expan to com Ow Msmic spadal sys-defined as an individual plant examination (IPli).1.icen.

^

""d """"#" "f USI sees were requested to proceed with the examinations for A

for the seismic capability of large safety-internally initiated events only (including internal flood-I""

ing'). lixamination of externally mitiated events would proceed separately and on a later schedule. Ilowever, while perf orming the IPli for internally initiated events, 5.

GI-131, " Potential Seismic Interaction involving licensees w ere advised to document and retain plant +pe-the hlovable In Core 171ux Mapping System Used in cific data relevant to external events so that they can be Westinghouse Plants." was identified because por-readily retrieved in a convenient form when needed for tions of the in core flux mapping system that have later external event analyses, not been seismically analyzed are located directly above the seal table.1:ailure of this equipment dur-ing a seismic event could cause multiple failures at 6.2.2 l'rograms llelated to External Events the seal table and could produce an equivalent small break l_OCA.

6.2 2.1 Seismic Programs The following is a brief descriptial if the programs re-6.

The " Charleston liarthquake issue" came about as a lated to seismic events:

result of a U.S. Geological Survey letter in 1982 that pointed out the possibility that large damagint 1.

USI A-17,

• System lmeractions in Nuclear Power c:Mhquakes have some likelihood of occurring at Plants," addresses NitC's concerns regarding the locations not formerly considered in past licensino interaction of various systems with regard to decisions. The staff initiated the Seismic llazard whether actions or consequences could adversely Characterization Project (l_1.NI.), which providcJ affect the redundancy and independence of safety probabilistic seismP Sazard estimates for all nucle; r systems. 'lhe evaluation of system interactions re-power plant sites cast of the Itocky h1ountains. A lated to internal events and internal floods is in-similar project was carried out by !!Pitt for the elec-cluded in the IPl! (01. 88-20). The evaluation of trie utility industry.The staffs purpose in evaluating spatial system interaction under seismic conditions the probabilistic studies has been to identify plac (the SSli)is included in USI A in the Central and liastern United States where past licensing decisions may have r alted in d.dr being 2.

USl A-40, "Scismic Design Criteriac investigates outliers with respect to sciss hazard, that is, the selected areas of the seismic design process. The likelihood of ucceding thek _esign bases.

15 NURl!G-1407

6.2.2.2 Internal l' ires Programs identify possible interactions. liighlights of the inajor findings in the seismic area are:

'the following is a brief description of programs related to interrud fires:

1.

In general, the modifications proposed as a result of the internal events analysis would not adversely af-1, NUlt!!G/ Cit-5088, " Fire 1(isk Scoping Study" Icct the seismic risk provided they do not become identifies some fire issues that had not previously weak links.

been addressed in the fire Pitas: fire growth code, seismic / fire interaction, fire barrier elfcctiveness.

2.

In general, the modifications made could potentially manual fire fighting effectiveness, effects of fire contribute to an increase in risk at the plant in the suppressants on safety equipment, and control following ways:

system interactions. A plant. specific analysis (in.

cluding a specifically tailored walkdown) should be a.

Many of the modifications proposed may in-performed to assess the actual risk impact of these volve adding valves or suction lines to existing issues at a plant.

systems. *lhus the possibility of a violation of the pressure loundary and potential diversion l

2.

GI-57,*liffectsof Fire Protection System Actuation exists if the modification were to fait during an on Safety Related liquipment," assesses the impact carthquake. Also, modifications may involve of inadvertent actuation of fire protection systems routing different trains of electrical power or on safety syst ems.This is one of the issues identified power from adjacent units. 'the possibility ex-in the Fire itisk Scoping Study. 'the industry, ists that the circuitry could be designed in such through I! Pill, has a program collecting data on the a way that failure of non4afety.related electri-effects of suppressants on the safety equipment.

cal components could actually defeat the cir-cuitry that was desired to provide redundancy,

""d 6.3 Approach on Coordination willl Ongoing l'rograillS b.

The possibility that inadequate anchorage could defeat the planned redundancy.

If duplication of effort by the staff and licensees is to be avoided, it is important that the above programs be coor-3.

The potential adverse effects of the modifications

dinated, include:

6.3.1 Countination with Internal Events a.

Poor accessibility for maintenance, l'rogram (ll'E) b.

Stiffening of systems leading to higher stress

'the coordination between the internal events IPli and due to thermal cycles during normal plant op-the IPlilill can be categorized into three phases:

eration.

preanalyses planning, plant modifications, and accident The cited study (llohn,1989) provides specific examples management.

of fixes and their impact on other initiating events.

6.3.1.1 Preanalyses Planning 6.3.1.3 Accident Management Considerations that would enhance efliciency between internal event and seismically related external event Guidance on the integration of findings from the IPillili and accident management is being developed analyses were discussed in Section 3.1.1.1.These consid, crations include (1) definition of elements and their (SliCY-89-308 Oct,1989).

boundaries,(2) walkdown procedures and spatialinterac-tions, and (3) composition of the peer review group. It is 6.3.2 Coonlination Among External Events likely that the IPl! will precede the IPlilill. Careful plan-l'rograins ning taking into account the above considerations will The issue of m.iegration between seismic and other avoid a duplication of effort by the licensee.

external events pnmanly involves mteractions between seismic events and fires tmd seismic events and floods, it is 6.3.1.2 Plant Modifications necessary to address seistmeally mduced fires and flooJs Since the IPli and IPlilill are likely to be performed as part of the IPl!!!!i.The effects of seismically induced separately,it is impcotive to examine the impact of modi-fires and the impact of inadvertent actuation of fire pro-fications identified during the IPli on external events and tection systems on safety systems should be addressed.

vice versa.*lhe staff examined several Pit As that included The effects of seismically induced external flooding and i

both internal events and external events (llohn,1989) to internal flooding on plant safety should be included.1he NUltl!G-1407 16

l coordination between the scismic and the fire or flood implement the seismic IPl!Illi. Thus resolution of these analysts should be based on the following:

issues can be casily accomplished during the scismic IPlilli! evaluation.

1.

The seismic analysts should generally scarch for and identify the initiating events (certain specific

'the potential interaction betw een the seal table and non-scistnically initiated failures of equipment or struc.

Category I seismic sygems associated with the movable

~

tures) that can cause fires or floods, and in core flux mapping system can be identified during the seisraic walkdown of the IPlil!!!. If needed, the compo-2.

The seismic and fue or fhul analysts should also nent capacities or consequences of component failure can discuss other concur rent seismically induced f ailures be evaluated using the same procedures that are used in or possible cifects on human actions and then, pro, the seismic IPlilili.

ceed to complete the rest of the IPlilill analysis.

6.3.3.2 Charleston liarthquake issue The coordination should include a meeting, prior to scis-As a result of work carried out to resolve the Charleston mic walkdown, in w hich the fire and flood analyrts should I!arthquake issue, probabilistic seismic hazard estirries discuss the key issues, how the analysis will be done, and exist for all nuelcar power plants cast of the Itocky Moun-what to look for. The fite or flood analyst may need to tains.These should be used directly by any licensee in that participate in parts of the seismic walkdown or revisit the region opting to satisfy th( seismic IPiil!!! with a seismic areas identified during the seismic walkdown to grasp the Pi(A. The hazard estimates also played a key role in issues hom the seismic. capacity point of view.

determining the review levcl earthquake used in the scis-mic margins methoJology option. Therefore, the IPl!!!!!

6.3.3 Coonlination with Seismic Progntms will constitute resolution of the Charleston liarthquake issue.

A number of programs telated lo seismic events tequiring licensee action have been identified. Mr ny of these 6.3.3.3 USl A-46 programs have arisen as a result of the changing percep-Implementationof theUS! A-46programinvolvesplants tion of hatards and revisions in the design and qualifica-with construction permit applications uocketed before tion criteria. There are two caicpories of seismic pro-about IW2.The USI A-46 plants thus form a subset of all grams as they relate to the seismic IPlil!!! The first the nuclear power plants in the U.S. that are requested to categoryinvolves programs, USl A-45 " Shutdown Decay perform the seismic IPlil!!!.

lleat itemoval itequirements,"01-131, " Potential Scis-mic Interaction involv.ng the Movable in-Core 17)ux The most efficient way to address the ongoing seismic Mappint. System Used in Westinghouse Plants," and the programs for US! A-46 plants is to conduct the A-46

• Charleston liarthquake issue" that have been subsumed review and walkdown to gather relevart information for into the IPli/IPillili and should therefore be specifically the seismic IPlilill. In order to facilitate this approach, addressed as part of the seismic IPlilitt The second cate-the activities of USI A-46 and the seismicIPlilill need to pory involves prograrns. c.p., USl A-46,"Scismic Qualifi-be coordinated, and the plant walkdown needs to be well eation of liquipment in Operating Plants," that can be planned.Sevemiinherent differencesbetween the A-46 coordinated with the seismic IPlilili.The coordination of program and the scismie IPlilill should be noted at the these programs with the seismic IPlilill is most beneficial outset befor e attempting to coordinate the two programs.

in reducing the resources spent by the licensee and staff.

1irst, the objectives are quite different.%c USl A-46 6.3.3.1 USl A-45 and GI-131 program has licensing implications on plant operation; this propram will assess and ensure the seismic rugged.

The methodolagy used in the seismic IPlilil! can also be ness of safety-related equipment in a plant to withstand used to address USl A-45 and G1-131.The systems and the SSti.The seismic IPlilill, on the other hand, gener-components for addressing USl A-45 will have been de-ally tries to identify plant vulnerabilities when subjected termined by the internal events 11'11, and the purpose of to carthquake levels higher than the SS!! design basis, the seismic IPlilili is to identify any significant and unique seisn.ic vulnerabilities in the decay heat removal Second. the scope of the reviews is different. USI A-46 is function. In addition, the seismic 1Pl!!!!! w ill evaluate the concerned with only one success path (with some require-poiential seismic interaction of the movable in core llux ment on equipment redundancy) of equipment needed to mapping system used in Westinghouse plants.

bring the plant to safe shutdown in the event of an carth-quake and maintain it there for at least 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />. The Capacities of decay heat removal components can be scenario considers an earthquake of the SSillevel with a established using either the fragility analysis (FA) or m poss.ible loss of offsite power because of this cat thquake.

servative deterministic failure margin (Cl IzM) a;-

The probabilities of a seismically induced i>OCA (small proaches depending upon the methodology chosen to or large) and a high.cnergy line break (IIlit ll) occurring 17 NUllI!G-1407

are pidred to be low cnough that thcu uirNJeration at Q4 (hordination with (llkr hes this ear thquake lesci n not wartanicJ Pqium tuhmn atal structures wdl bc exammed dor my a w.dkJown onk if in aAbtion to the spnif h i sis a'klt is discusscJ ab m e.

they have the potential to cose u rsnot micras tnin with it, durity its ll'l i I a l u c r o.e t < 1 i disa n t is a notable s u!nclaI'llll\\ Ibat Is I 'ph a!I) aW N Lited willi an) tithc!

the equllMiicnl icview ed drni cause daniaf e tti [hn equjji ment. ihe leview of abost-proui J tanks tas u t of l 51 l N or ( il anJ propmes rocawu cs to dnptsc of the spc-A-4b) n an cueption t dn sdely nsuc or (2 ) conslaJes lluit no s ulnerabihty esnts at its plant that n topically assocuitcJ with any l%I

"'('L h 'I

"'"*""*'Ih'l

'" I 'l r noh ed Ior a lhe scoma IPl i I n wru ined with the sulnctabilitic s plant upi n u siew iuid au cplana of ik ic,uus of the i

of the w} nile plant not just the eqmpment Alsis esala

'I

'"""* "T 'h' M J be d nx u m J

=

atuins are rencralk made at locls abose the desin hen At ttus k s el of carthquake. st ntmcalh mJut ed I ( n &

.hr al%ty of the th e idohips to iJchtils s ul i

a ar c (tiliskieled. and rnitiratiny usterra aini equipri: crit t i in r abn,itics assot iated w ith the l N1 or ( il bemy ad addrew thn untiator are 'niewed l'her clor c, es en il ttu Ju ssed 1.1,NI sentlth Inar f uis incthihliaif) n utall/cd li unple t

trient the Ncisnile IPl i I (Nun c it n quil( AguiLH lii ih-g l

q q. g g

I N A th n Juatam t addinonal eqmjonent would necJ t yu.io or memlh pior s momimt FWun to be iniewed m(t that it tjuurd hu unplemeritmp 1 Si acg yq.,,g,am;ggupnpp A 4h used I hird, lhe In t }s til les icw arid w,tlkdi'wil arc ditIctrill l }n+ [n }; rig al lniys lg q [cu dsing th( iv iic

(

'l bc Sennut (Juahin ation l lihts ( uoup (M)11 e i ath!

l PRI h.n c Jo clopn! a detailed ( n ner n impicna nta 7 p()ClNENTATi()N AND tion Pro < cJui c H ilPi lot the l 51 A 4h i n ic w and gg) g; walkdow n ttuit wa' iniewed by the N R( stall, and a Salet) I s aluation Repor t,NI R i we iwucJ lla t ill'

! hc IPl i I shoulJ bc documentt J m a tras cabic manner should be f ollowed in pctlottning the l'SI A 4h r eview to pim idc the baus f or tht bnJmps I hn t an be deah v.ith anJ walkdow n t he rmdchncs assos iated with the sentuk most ellhicntly bs a tw o oct appi o.u h lla lust tici PR A or senmh maryms methoJolors are not a spectf u annis of the t esults of the curnma, ion, w hh h will be as those o the (ilP l o illuu r ate ihn pomt, m the reported to the NRl f or inicu l he sci ond tici is the walLJow n rs view of espaniaon ath hot bol's ( ilP calls lot documentation of the cuniunition itsell, whh h should be the bolt tif tilcw chn ks f etioncJ hs the turnset hu the duratiim of the hcense the use td a wilih h test hit h

whereas the mat yms mdkdow n ensut ts onk that the an unicss supersedcJ

~

t bor bolts are aJcquate to hold down the nlmpment a-designeu with no specu h testmp r equuements to t onlu m I hc mlot unition subnutted to the N R( shoulJ bc orran.

,anhor capacity. Ihc completion of the scisnuc IPl 1 I i/cd and picscoted m accotJaine with AppenJa (

lhe does not automatically mean that the l'51 A 4h inicw n subnuttal ina) enable mariy issues to be drah with m the satisf actotih wmpleted IPl i I iniew Pertment nsues at t d ussed in Section h.

m I or some issues f or cumple. l!51 A 4h. a detailed docu o

mentation r equucment custs and that should be f ollowed

.I.here may be m citaps or Jiller eni.cs m the eqmpment m the broaJ Iramewoik ol Il,i 1 I subimttals Specibe scope lot llSI A 4h anJ the scisunc II'l 11 I oi equip-mloimation televant to patucular issues. e g l' Sis and ment that is willun the scope of (,$1 A -46 or the scismic t ils, should be klentificJ IPl.11 only, it is clear that either (ilP or IPI l.1 puhle-Imes, respectnely, should apply I or the overlappmp equipment, the ef ficient approach is to use the UIP for 7.1 Inf,orination Subm,tted to the NRC i

both walkdowns, however, the IPl l I should use the A deuulcJ int of udor mauon to be submitted to the NRC review level earthquake.

n poviJed m Appenda C in summary, it is recommendcJ that licensees coordmate M lHformation htained br Mdit the mlormation collection f or the USl A-46 and seismic IPEl 1 teview and walkdown m order to mmmu/c or Retamed documentation should melude appheable event avoid dupheation of eff ort by the licensees and siaft. Care trees and fault trees, cuttent versions of the system note-should be exercised in the coordmation to ensure that the books (ti appheable L w alLJown ieports, anJ the results of L

requirements and objective'.4 both programs are ful-the examination in peneral, all documents essential for a hlied Coordmation of f'. iwo programs has been shown ptactitioner m the fielJ to understand what was done m to be f(asible m the trial evaluation of the Ilatch plant the IPlil I should be retamed in addioon, the manner in usmg the 1:PRI seismic margins methodologs.

which the vahdity of these documents has been ensured NllRl R 1407 18

I should be documented. II credit is allowed in the IPlil!!!

NI(C,1985," Policy Statement on Severe f(cactor Acci-for any actions taken by the operators, the licensee should dents." Federalllegister, Vol. 50, p.32138. August 8, have established plant procedures to be used by the plant 1985.

staff responsible for managing a severe accident should one occur. Procedures i.hould provide assurance that the NI(C,1988,

• Individual Plant lixamination for Severe operators can and will take the proper action.

Accident Vulnerabilities-10 C11150,54(f)," Generic letter No. 88-20, November 23,1988.

8 REFERENCES NI(C,1989, " Initiation of the Individual Plant lixamina-tion for Severe Accident Vulnerabilities-10 CI'It ACitS,1989, hiemorandum frorn C. Daily to lt, Savio and 50.54(f)," Generie Ietter No. "" "* Supplement bl. Stella,

Subject:

Assesstnent of issue Concerning No. l., August 29,1989, Operating l(eactors: IJphtning induced iteactor

!! vents, dated August 2,1989.

Sl!CY 88-147 Integration i

f Severe AliOI),1986, A1:OD lingineering livaluation lleport, "1.ightning livents at Nuclear Power Plants," April I!Piti NP-6041,"A hiethodo

' (Nu-

'.188.

1986.

clear Power Plant Seismic Amico,1989,

• Containment Considerations for the Use I! Pit! NP-6359, "Setsn.ic h1argin., onent of the of the Seismic h1argins hiethods for 1(isk Screening,"

Catawba Nuclear Station," Vols 1 & 2, April 1989.

Science ApplicationsInternationalCorporation, I.ct.

ter Iteport, J une 30,1989, liPill NP-6395-D, "Probabilistic Seismic lla7ard livalu.

ation at Nuclear Plant Sites in the Central and F. astern Appendix it. Title 10 Code of Trderal Regulations. Energy United States: 1(esolution of the Charleston issue,"

(/0 CFR l' arts 0 to 19W, itevised as of Januaiy 1,1989, April 1989.

10 Cl?lt 50, Appendix it,' fire Protection Program for Nuclear l ower Facilities Operating Prior to january 1, NUlti!G-1032, *1! valuation of Station tilackout Acci-1974," Office of the Federal 1(epister, National Ar.

dents at Nuclear Power Plants," June 1988, chives and 1(ecords Administration, Washington, NUlt!!G-ll50," Severe Accident 1(ists: An Assessment D.C.

for I ive U.S. Nuclear Power Plants," Vols.1 & 2, June 1989-llohn,1989, " Status 1(eport on issues 1(elated to Inter-nal/l!xternal livsnt Interaction and Decay lleat lle-moral itequirements for IPlis," Draft, hiay 1989.

NUltliG/ Cit-0098, " Development of Criteria for Sc.is-mie 1(eview of Selected Nuclear Power Plants," hiay 1978' lleckjord,1987, hiemorandum from liric S. lleckjotd to l..C. Shao,

Subject:

lixternal livents Steering Group' NUltliG/ Cit-2300,"PI(A Procedures Guide," January Dated December 21.19b7,

g9g3, lleckjord,1988, hiemorandum from lirie S. lleckjord to NUltliG/ Cit-4334,"An Approach to the Quantification I awrence C. Shao,

Subject:

F.xternal 1: vents Steering of Seismic hiargins in Nuclear Power I'lants," August Group, Dated hlay 31,1988.

1985.

Iludmtz,1989, Ixtter from 1(obert J, iludnitz to Conrad NUlti!G/ Cit-4482, "ltecommendations to the Nuclear li. hieCracken,

Subject:

hiodification to my hiemo-llegulatory Commission on Trial Guidelines for Seis-randum of 12 July 1988, dated 3 January 1989.

mie h1argin Iteviews of Nuclear Power Plants," h1 arch Chery,1985, hiemorandum from D. Chery to D. hioeller and D. Okrent,

Subject:

llydrologic lingineering i tes-NUlt!!G /CI(-4734, " Seismic Testing of Typical Contain-entation to Combined hiceting of the ACI(S Subcom-ment Piping Penetration Systems," December 1986, mittees on Site !! valuation and lixtreme livents Phe-nomena, dated October 9,1985.

NUlti!G/ Cit-4826, "Scismic hlargin 1(eview of the hiaine Yankee Atomic Power Station," Vols.1-3, Ilardy et al.,1989. " Guidance on 1(elay Chatter liffects,"

h1 arch 1987.

11Q11, Draft, June 1989 NUltt!G/ Cit-5042, "livaluation of lixternal llazards to 1(avindnu 1989," Integration of Various Seismic issues."

Nuclear Power Plants in the United States," Decem-I!Qli, Diaft, June 1989 ber 1987.

i 19 "Lill!G-1407

NUlti'.0/ Cit-5042, Suppl.1 *livaluation of lixternal NURiiG/ Cit-5088. "I%c Itisk Scoping Study," January llazards to Nuclear Power Plants in the United 1989 State-Scismic liarard," April 1988.

N Ultt!G/ Cit-5250,

• Seismic 1lanstd Characterization of NURl!O/ Cit $N2, Suppl. 2. "!! valuation of lixternal

<>9 Nuclear Power Plant Sites 11ast of the llocky llazards to Nuclear Power Plants in the United Mountains," Vols.1-8, January 1989.

States-Other !!xternal livents," February 1989.

NUltEG/ Cit-5259, " Individual Plant 11xamination for

- NUld!O/ Cit-5076,"An Approach to the Quantification Iixternal livents: Guidance and Procedures," Draft.

of Seismic Marrins in Nuclear Power Plants: The Im-portance of IlWit Plant Systems and Functions to NUld!O/ Cit-5270," Assessment of Seismic Margin Cal.

Scismic Margins," May 1988.

culation Methods," March 1989.

NUllP.G-1407 20

Al'I'ICNDIX A REVil?W l.l VEL ICARTilQUAKl?

The scismic margins methodology was designed to dem-A.1 Intrmluction onstrate sufficient margin over hSli to assure plant safety and to find any " weak links" which might litmt the plant The specification of a ceview level carthquale (RLii) for shutdown capability to safely withstand a seismic event use in carrying out an individual plant examination for bigger than SSli.The methmlology involves the screening external events (IPElil!) has been a complex problem of components based on their importance to safety and involving the search for con %ncy. It would be prefer-seismic capacity.The scismie margins methml utilizes two able if the RI.lis were mmpletely consistent with the review or screening levels sm,rtd to peak ground accel.

individual plant e. mination (IPli) for internal events, crations of 0.3g and 0.59. In areas of low to malerate the IPlil!!! using a seismic PI(A, and the inherent seismic hazard, most plants that hive been evaluated 5ttengths of the seismic margins methodologies, but it is using PRAs or margins studies have been shown to have very dif ficult to satisiy all of these elements in any rigor-1ICLPh at or below 0.3g. Past experience indicates that, ous quantitative sense. Thus, f or example, attempting to at the 0.3g sercening level, a small number of " weak links" equate the review level carthquake to the reporting crite-are likely to be identihed, efficiently defining the domi.

ria in the IPl!(mean sequence frequency leading to core nant contributor s to scismically induced cor e damage. It is damage of Ili-6 per year)is fraught whh difficulties be-the staff's judgment that the use of a 0.3g review level cause of the large uncertainties in numerical estimates of earthquake for most of the nuclear power plant sites in seismically induced core damage, the inappropriateness the Central and liastern United States (east of the Rocky of companson between numerical estimates of seismical.

Mountains) woukt serve to meet the objectives of the ly and internally imiuced core damage (the source and IPlilll!.

treatment of uncertainty can be quite different) and the inherent difficulties in relating the output of a seismic

• "E#

All sites cast of the 1(ocky Mountains, however, are not N" "

' "" "" "N subject to the same level of earthquake hazard. The re-

" "'""' !" " "N #"

9" cent studies by ILNI. (NUl(l!G/CR-5250) and I! Pill n ea e n'a m e ume manner as internal (l!PRI NP-6395-D) show significant differences deIiend-initiators in implementing the Severe Accident Policy, it ing on locatmn and specihc s.te conditions, llecause the should be noted that the Rt.lidefines a reportinglevel. A i

two studies do not necessarily agree wlith each other, it IICLPF value lower than RLl! does not necessarily rep-was deerned necessary to use both studies m determining resent a plant vulnerability, I lowever, the licensee should which review level caythquake should be assigned to each assess the significance of iICl.PF values lowcr than 1(Lli site. Ilazard compansons were made using the median, and take any necessary actions and make other improve-85th percentile, and mean from the site specific results ments that are deemed appropriate by the licensee, proviJed by the LLNL and I!PRI studies. liased on the sensitivity tests and engineering and seismological judg-

@ Heral h llua W,) PI w e itre ment, the staff has defined the review level earthquake for each site (0.3g,0.5p, or reduced scope) in Table 3.1.

g g ggg The staff has recommended three review level carth-The sites in the Western United States (west of the Rocky Mountain Front) are t cated differently Those sites in quakes to be used when applying the seismic margins coastal California where the seismic ' hazard is much rnethodolm to nuclear power plants east of the Rocky Mmmtains for the lPlil!!!. The review levels or " bins" are higher and tue resulting design bases are greater than 0.5g cannot make use of the margins methodology. The 0.5g,0.3g, and a reduced scope level. The basie informa-tion used was the lawrence 1.ivermore National labora-other plant sites in the West should use a 0.5g review level carthquake unless it can be demonstrated that the seismic tory (ILNI.) hazard study (NURiiG/ Cit-5250) and the harard level at a particular plant site is consistent with the

!!!cetne Power Research Institute (l!PRI) hazard study scismic hazard at the 0.3g bin plant sites cast of the Rocky (l!PRI NP-6345-D). These studies represent state-of-the art estimates of seismic hazard. llecause the two stud-Mountains.The results of the binning for the plants in the Western United States are presented in Table 3.2.

tes do not necessarily agree w;th each other, it was deemed necessaiy to use them both in determinmg which bin a particular site belonged in.

.l'he rationale for the selection of the review level carth-quakes (Rt.lis)and the grouping of the plants cast of the in the ILNI. study (i4URiiG/CR-5250), it w.

.oted Rocky Mountains is discussed below.

that, for some sites, the mean estimates of seismic hazard 1

NUlt!!G-1407

Appendix A were dominated by the input of one ground motion expert lated to the types c.f motion that could cause damage at

- (No. 5). This dominance was caused by the low attenu.

nuclear power plants. Unit weights (2 nth each) were ation, high uncertainty, and relatively high motion on rock assigned to the likelihoods of exceeding spectral response found in this expert's input. This input has received a ordinates at 2.5, 5, and 10 llz, One-half unit i < ight great deal of attention, and smne have argued that it is (Inth) was assigned to the likelihood of exceeding the inconsistent with the data. The staff requested ILNL (as

PGA, a sensitivity study) b calculate the hazard at nuclear power plant sites cast of the Rocky Mountains icaving out A.2.4 Ranking Criteria the input of this expert.

Emphasis was placed on the relative ranking of sites with Data from the Saguenay Event in Quebec, Canada (No.

respect to other sites using the same scismic hazard study, vember 1988), the largest carthquake in castern North statistic, and ground motion measures. Extensive use was America in 50 years, appeus to be qtute different from made of a clustering methodology developed by LLNL previous data sets and has not helped to resolve the con.

for the NRC (Bernreuter et al.,1989a,1989b). Fora given troversy, At this time, in order to avoid relying exclusively hazard study, statistic, ground motion measure and refer-on the Ll.NL results that include the input of mpert ence level, this methodology divides the ensemble of sites No. 5, the staff is treating the LLNL hazard es. mates into groups so that the sites in any one group are "close' based on the other four ground motion experts as a sepa.

to each other with respect to scismic hazard. For examp!c, rate study when binning nuclear power plant sites for the sites may be divideo into groups based on mean esti.

IPEEE.

mates of exceeding 0.5g PGA from the EPRI study or median estimates of exceeding the 2.5 llz spectral re-A.2.2 Compan. son Procedure sponse (associated with the NUREG/CR-0098 response spectrum anchored at 0.3g) from the LLNL five expert flazard comparisons were made using the mean, median, study. Although there were a fixed mlmber of groups, no 8 8 wcre reqwed in a group, and and 85th prcentile from the site-specific results provided m

s me gmups mntm.ned M y a s m, by @ fi.NL ed EPRI studies. Each of these pieces of information represents a different way of characterizing the distribution of seismic hazard estimates at each site as A.2.5 Spectral Sliape determined by a particular study.

'Ihe spectral shape associated with the 0.3g screening level was assumed to be the median NUREG/CR-0098 Mcan: The mean is a commonly used statistic that spectrum anchored at 0.3g.There has been some discus-can be assigned actuarial significance. However, be-sion that the screening level should actually be associated cause of the skewed nature of the distrioution, it is

.cith a somewhat higher ground motion (the Seismic also a highly unstable (with respect to methodology Qualification Utility G roup (SQUG) bounding spectra m) and input assumptions) view of hazard.The mean is but in this relative comparison, the use of this alternative highly sensitive to the characterization of the cx-spectrum would make little or no difference, tremes of the distribution.

UN hdm ifcdiar.: Tbc median is more stable thN the mean

.nd shows the greatest agreemem,between the A.3.1 Initial llinning Evaluation LI.NL and EPRI studies. However, it is only the 50th percentile of the hazard and is insensitive to the As the first step, sites that consistently fell into the group extent of uncertaliQ.

that had the highest likelihood of exceeding the 0.3g NUREG/CR-0098 5% damped median spectrum were 85th Percentile: An alternative candidate to the mean conditionally assigned to the 0.5g bin. Sites that fell into is the 85th percenti!c. It reflects uncertainty in that it the group that had the lowest likelihood of exceeding the indicates the breadth of the distribution but it is less 0.3g NUREG/CR-0098 5% damped spectrum were as, tensitive to extreme outliers.

signed to the reduced-scope bin.

A.2.3 Weigitting Criteria The ground motion measure compared was the weighted combination of 2.5 Hz,5 Hz,10 Hz and PG A.The individ-In the past, great emphasis has been placed on the likeh,-

ual coasistency criteria used were:

hood of exceediag peak ground acedcration (PGA). In 1.

Agreement among the LLNL five. expert, LLNL this evaluation, site ha7ard comparisons were made using four ernert, and EPRI studies, and respanse spectra and PG A. The likelihoods of exceeding

~

spectral response accelerations in the 2.5 to 10 Hz range 2.

Agreement between the median ::nd either maan or were examincel base these frequencie.c src more re-85th percWile statistics.

. NUREG -1497 2

Appendix A This resulted in a comparison of nFe separate hazard made good seismological sense and f here was no need to grot. pings (three pieces of information for each of the include additional sites in these bins. In conjunction with three studies).

this examination, limited sensitivity tests were also car-ried out to determine the impact of slight relaxations in For exarnp.e. if a particular site fellin the top group (0.5g the consistency criteria.

bin) for all of the criteria except the EPRI median, it remained in the 0.3g bin. The conclusions must be sup-A.4 Hinning of Sito-Results ported by all the hazard studies. On the other hand, if a particular site fell in the bottom group for all cf the A.4.1 Reduced scope Margins Methodology 3

IllH

. criteria except for the LIMI/four. and LIEL five expert mean estimates, it was included in the reduced-scope bin.

The consistency criteria outlined in Section A.3.1 were J

Only one measure of uncertainty, mean or 85th percen.

slightly modified to idc ntify sites for the reduced-scope tile, needs to be satisfied.

bin.The two bcatom median groups were included rather than the bottom group alone. When this was doac, five The candidates for the 0.5g and reduced scope bins were sites (South Texas, Comanche Peak, Waterford, River then subjected to additional evaluation by the staif.

Hend, ar'd Crystal River) were identified as belonging to the reduced scope bin.

A.3.2 Subsequent llinning Evaluations Als added to this bin were several sites for which no The candidates for the 0.5g bin were first exantined to liPRI calcol tione -e avaDable but were in the bottom provide some assurance that although the hazard was gr ups in b th the LI.NL four-and five-expert studies.

relatively high, it was high enough to warrant inclusion in They are Duane Arnold, Big Rock Point, Grand Gull, St.

th;s hm.

1.ucie, and Turkey Point. The ten candidate sites in the reduced-scope bin lie in ar eas of low seismic hazard along As a test, it was considered appropriate that a site be-

" ^"

"PP" longed in the 0.5g bin if a hypothetical nuclear power

{ ides' plant at that site was assurned to have a llCIEF of 0.3g

=

and the mean annual core damage frequency associated A.4.2 0*Sg Ilin with that hypothetical plant was 111-5 or higher. The work g

cited in Ravindra 1989b showed that the mean annual As a result of the evaluations cited above, two sites (Pil-core damage frequency was roughly an order of magni.

grim and heabrook) were identified as belonging la the tude lower than the mean annuallikelihood of exceeding 0.5g bm.

the plant ilCl.PF and ve7 roughly equal to the median annual likelihood of exceeding the plant ilCLPF.

A.4.3 0.3g Ilin Z

All sites not identified as belonging in the 0.5g or re-llased on these estimates, the staff assumed that inclusion duced scope bins were assigned to the Q 3g cin.

in the 0.5g bin would be supported if:

s A.4.4 Other Considerations 1.

The mean or 85th percentile annual likelihood of exceeding the 0.3g spectrum frorn all three studies The grouping was made assuming that each h) cation was was 10-4 or greater, and associated with one site condition (rock or varying depths of soil). Some twelve plant sites east of the Rocky Moun-2.

The median annual likelihood of exceeding the 0.3g tains whose main Category I structures are Leated on spectrum from all three sGes was IE-5 or greater, rock also have some Caugory I structures or components located en shallow or intermediate depths of soil. Since

- This evaluation should " viewed as a " sam.ty check,'; it shallow soil, less than about 80 feet thick, can significantly should not be viewed as a plant-specific statement on core amplify gnund motion, these sites should j erform soil damage frequencies. The reasons are:

amplificauon studies to determine the effect, t

1.

The uncertainty and generic nature associated wah In particular, for four of the sites included in the 0.3g bin k

the correlation m Raymdra,1989b, (on the basis of their primary site conditions), the hazard 2.

The use of spectral estimates rather than peak for structures or components on the secondary site condi-F gn,und acceleration, tions is equal to or higher than the hazard associated with

' 3.

' The inclusion of the 85th percentile estimates, and those plants in the 0.5g bin. Licensees should, if site-L specific arnlysis indicate, use the 0.5g screening tables for 1

4.

Ali the previously mentioned problems associated elements affected by soil amphfication. Similarly, for onc with bottom line numbers.

site in the reduced scope bin, site-specific analysis should K

Finally, the staff examined the candidates for the 0.5g and be carried out to determine the effcets on those elements

' reduced-scope bins to assme itself that the classifi"ation affected by soit amplification.

3 NUREG-1407 i

ii iei-i,

w i

1 Al'I'ENDIX 11 COMI'ARISON llETWisEN A REDUCICD SCOl'E AND liUI 1 SCOl'E Si?lSMIC MARGINS EVAL,UATION i

'Ihere are differences between the reducedsope and should involvc interactions among seismic capability

[

full scope rnargins evaluation both in the extent of the evaluation engineers, systems engineers, and the systerns amdysis and in the amount of quantification of licensee's plant operations personnel. *lhe j

- HCLPF values for equipment identified in the walkdown, walkdown team should visually inspect pertinent The comparison is presented in Table il 1. 'the emphasis structures, equipment, and anchorages consistent on walkdown and not on quantification also applies to the with the full scope NitC or I!PRI methodolopy. If performance of containment and containment systems, potentially vulnerable cornponents are found during USI A-45 (Decay licat Removal Requirements), and the walkdown, a capacity check may be necessary 01-131 (Iri Core Flux Mapping S; stem).

using the applicabic NURl!G/CR-0098 ground re-sponse spectra. 'lhese results should be docu-mented. Data sheets similar to those found in Ap-II.I l$l011101119 l'reserveci pendix i of iteRi Ne-co41 shouid be used io docu.

• " ' ^ *

• The followin8 elements of the scismic mar 0 ns methodol-ings for structural details that can not be seen m the i

ogy must be preserved, that is, they must be identical in field should be performed.

the reduced scope and futhscope evaluation:

4.

While the post walkdown assessment effort for a 1.

For either the NRC or liPRI methodology, the sys-reduced scope evaluation should be identical m tems enginects.nus perform significtmt pre-walkdown work that uhould be preserved in a re-quality to that in the full scope margins methodol.

duced scope evaluation in the NRC methodology, w, hs Gmst and im1 M cHort am Mfmnt b this involves defining initiating events, defining cause sequence;y (NE) or sucem paMml NI) '

8*

""I be gmputed. Instead, its event trees and the safety functions involved, and identifying systems and components necessary to emphasis should be on identifymg possible weak-carry out these functions. In the !!PRI methocology, link itcms that may need strengthenmg.

this involves defining success paths (pnmary and alternative) and the systems and componen's in-11.2 RecltictiOIIS volved in these paths. For both methodologies, the thrust of this work is to narrow the scope and focus

'Ihe following, although needed in the full scope margins l

the effort on the key element of the review, the methodology, are not needed in a reduced scope margins walMow" ev@;aion:

3.

For either the NRC or the liPRI methodology, the seismic capability evaluation engineers must per-H.2.1 NRC Seismic Margins Methodology form significant pre-walkdown work that should be preserved in the reduced scope evaluation, in ersh 1.

The systems engineers need not prepare or quantify methodology, this involves developing an under.

fault trees and lloolean expressions representing standing of the seismicinput to the plant, tho scismic accident sequences. Also, since fault trees will not design basiv und realistic ground and floor response bc < developed, these engineers need not combine spcetra. It also involves pre-walkdown screening of nonscismic failtue basic events with seismically initi-i the key systems and components identifiad by the ated failures in t/.y rigorous fashion, although the systems engineers so as to make the walkcovm its elf existence of those non-seismic failures,if identified, most efficient.'Ihe thtust of the screening is to iden.

should be noted md their importance assessed in tify items thoMt to have very bigh IICLPF \\ alues, tha course of the margin evaluation.

items suspected of having low ilCLPF va'tues, and therefore lists of items to be cumined at various 2.

The seismic capability evaluation engineers need levels of detail during the walkdown.

not develop I ICLPF eapacity values for all of the key equip.nent items that would be represented on the 3.

The r educed-scope evaluation should be identical in seqoence level !!oolcans (which will not be devel-quality and effort to that required in the full scope oped). It follows that they can not develop a plant-margins methodology. One crucial feature is that it level iICLPF capatify value.

I NURiiG-1407

Appendix il

~ ll.2.2 EPRI Seismic Margins Methodology (primary and alternative) being studied, it follows that they can not develop any sucess-path-level The seismic capability evaluation engineers need llCLPF capacity values that would be taken as not develop llCLPF capacity values for all of the representing the plantelevel llClfF capacity.

key equipment items found on the success paths Table 11.1 Reduced scope hlargins hiethod liased on NRC Seismic Margins Methodology (NUREG/CR-4482, Chapter 4)

Step No.

Description in Reduced Program?

I Selection of ::arthquake Review Level Not applicable, NRC designates sites that qualify 2

Initial Systems Review Yes, in entirety 3

Initial Component ilCLPF Categorization Yes, in entirety 4

First Plant Walkdown Yer, in entirety 5

Systems Modeling Finalize Event Trees:

Yes Fault Tree Development:

No 6

Sceond Plant Walkdown Only as needed 7

. Systems Mode' Development No 8

Margin Evaluation of Components, Plant No liased on EPRI belsmic Margins Methodology (EPRI NP-6041, Chapter 2)

Step No.

Description In Reduced Program?

1 Selection of Seismic Margins Earthquake Not applicable, NRC designates sites that qualify 2

Selection of Assessment Team Yes, in entirety 3

Pre-Walkdown Preparation Work Yes, in entirety 4

Systems and Element Selection Walkdown Yes. in entirety 5

Seismic Capacity Walkdown Yes, in entirety 6

Subsequent Walkdowns Only as needed 7

Seismic Margin Assessment Work No l

l NUREG-1407 2

APPENDIX C DETAILED DOCUMENTATION AND REPORTING GUIDELINES This appendix provides the guidelines for detailed docu.

2.

A list of PI( A or SMh1 studies on similar plants that mentation and reporting format and content for the the 1PEEE team has reviewed along with a list of IPElin submittals.'!he major parts of this appendix are important insights derived from these reviews.

the guidelines for seismic analysis (Section C.2), internal rire analysis (hection C.3), other analyses (Section C.4),

3.

A concise description of plant documentation used specific safety features and plant improvements (Section in the IPEEE (e.g. the FSAlt, system descriptions, C.1.4), and the licensee review team (Section C.I.5). 'lhe procedures, and licensee event reports) and a con-licensees are requested to submit their IPERE reports cise discussion of the process used to confirm that using the standard table of contents given in Table C.1 or the IPEfil! represents the as-built, as ope-provide a cross reference. This will facilitate review by the plant. 'the intent of such a confirmation is ns NI(C and promote consistency among various submittals.

propose new design reverification effoi ts on the pm.

'the contents of the elements of this table are discussedin of the licensces but to account for the impact of sections below, previous plant modifications or modificatioA, con-ducted within the IPEEE framework.

'the level of detail needed in the documentation should 4.

A descriptie

.he coordinat. ion activities of the be sufficient to enable NI(C to understand and determine 11:: validitv of all input data and calculation models used, IP.EEE teams among the external events (e.g., for to assess tiie sensitivity of the results to all key aspects of scismically induced fires).

the analysis, and to audit any calculation. It is not neces-Col.4 Subm.ttal ol' Spec fic Safety Features i

i sary to submit all the documentation needed for such an NitC review, but its existence should be cited and it and Potentitti Plant Improvements' should be available.in easily usable form. The guideline

'Ihc licensee should provide a discussion of the criteria for adequate retamed documentation is that independent used to define vulnerabilities for each external event expert analysts should be able to reproduce eny portion of evaluated.'the licensee should list any poteatialimprove-the results of the calculations in a straight forward, unam-ments (including equipment changes as well as changes in biguous manner. l'o the extent possible, the retamed maintenance, operating and emergency procedures, sur-documentation should be organized along the h,nes iden' ve llance, staffing. and training programs) that have been tified m the areas of review. Any mforme.,on that is selected for implementation based on the IPEEE (pro-comparable to that provided under the IPE for internal vide a schedule for implementation)or that have already events can be incorporated by reference, been implemented, include a discussion of anticipated benefits as well as drawbacks to any improvements.

C.1.5 IPEEE Team and Internal Review C.1.1 Conformance with Generic Letter and Supporting Material The basis for requiring the involvement of the licensee's staff in the IPEEE review is the belief that the maximum Certification that an IPEEE has been completed and

"

• I"*

"P I an EEE would be documented as requested by Generic Letter 88-20, Sup-realized if the licensee s staff were.mvolved m allaspects plement 3. The certification should also identify the o

ga nahon an m

ment wodd faMtate measures taken to ensure the technical adequacy of the integration of the knowledge gained from the examinat,e IPEEE and the validation of results, including any t'ncer-n mto operating procedures and training programs, tainty, sensitivity, and importance analyses.

.l'hus the submittal should describe licensee staff partice C.1.2 General Methodology patan and the extent to which the licensee was involved m all aspects of the program.

Provide an overview description of the methodology em-ployed in the IPliEE for each external event.

The submittal should also contain a description of the internal review or peer review performed, the results of C.1.3 Information Assembly the review team's evaluation, and a list of the review team members.The maximum benefit to the licensee will occur Iteporting guidelines include:

f the combination of persons involved in the original 1.

Plant layaut and containment building information analysis and the internal review, taken as a group, pro-not contained in the Final Safety Analysis Report vides both a cadre of licensee personnel to facilitate the (FSAR).

continued use of the results and the expertise in tne 1

NUl(EG-1407

Appendix C methods to ensure that the techniques have been cor-6.

The estimated core damage frequency (for both the rectly applied. Furthermore, an internal 1eview provides LLNI. and liPRI hazard curves, if available for the quality control and quality assurance to the IPlilill site) and plant damage state frequencies, the timing process.

of the core damage, including a qualitative Acus-sion of icertainties and how they might affect the final results, and contributions of different ground C.2 Seismic Events motions to core damage frequencies.

Section C.2.1 describes submittat information guidelines 7.

Any seismically induced containment failures and for licensees who choose the s:ismic I RA for seismic 11 lifil!, while section C.2.2 describes mformation guide-other containment performance insights. Particu-lines for licensees who choose the seismic margin method larly, vulnerabilities found in the following three systems / functions: penetrations, isolation, and con-for the seismic 11 f!!!!L l'he submittal should be pre-sented in conformance with lable C.l.

tainment heat removal / pressure suppression. flarly containment failures that might result in high-consequence sequences or may initiate accident se-C.2.1 Seismic PRA Methodology quences. Also, computed fragilities and ilCLPFs of

• E "#"

'lhe following information on the scismic IPl!!!!! is the mini n that should be documented and submitted to 8.

A table of fragilities, both generie and plant specific, used for screening as well as in the quantification.

The estimated IICLPF for the plant, dominant se-1.

A description of the methodology and key assump-quences, and components with and without nonscis-tions used in performing the seismic IPl!!!!!.

mic failures and human actions.

2.

'the hazard curves (or table of hazard values) used 9.

Documentation with rcgard to other seismic issues and the associated spectral shape used in the analy-(Section 6) addressed by the submittal, the basis and sis. Also,if an upperbound cutoff to ground motion assumptions used to address these issues, and a dis-of less than 1.5g peak ground acceleratmn is as-cussion of the findings and conclusions, fivaluation sumed,Ihe results of sensitivity studies to determine results and potential improvements associated with -

whether the cutoff affected the delineation and the decay heat removal function and movable in, ranking of seisrnic sequences, core flux mapping system (for Westinghouse plants) should be specifically highlighted.

A ummary of the walkdown findings and a concis

10. When an existing PR A is used to address the scismic desenption of the walkdown team and the procc-IPf!!!!!, thc licensee should describe sensitivity stud-ies related to the use of the initial hazard curves, supplemental plant walkdown results and subse-4.

All functional / systemic seismic event trees as well as quent evaluations, and relay chatter evaluations, data (including origin and method of analysis). Ad-The licensee should examine the above list to fill in dress to what extent the recommended enhance-those items missed in the existing seismic PRA (See ments have been incorporated in the IPl?Illi. A de-Section 3.1.2).

scription of how nonseismic failures, human actions, dependencies, relay chatter, soit liquefaction, and C.2.2 Seismic Margins Methodology seismically induced floods, fires are accounted for.

Also, a list of important nonscismic failures with a The following information on the seismic IPI!!!!! is the rationale for the assumed failure rate given a seismic minimum that should be documented nnd submitted to event.

the NRC for a full scope SMM review:

5.

A description of dominant functionc/ systemic se-1.

A description of the methodology and a list ofimpor-quences leading to core damage along with their tant assumptions, including their basis, used in per-frequencies and percentage contribution to overall forming the seismic 11III!!!. Addrev the extut to seismie core damage frequencies (for both LLNL which the following were taken into account: non-and !!PRI hazard curves if available for the site).

seismic failures, human actions, dependencies, relay Sequence selection criteria are provided in chatter, soil liquefaction, and seismically induced GL 88-20 and NURPG-1335. The description of floods / fires. Also, a list ofimportant nonscismic fail-the sequences should include a discussion of specific ures with a rationale for the assumed failure rate assumptions and human recovery action.

given a seismic event NURl!G-1407 2

4

2 4

W' q Appendix C

'y 2.1 : A summary of the walkdown findings and a concise results and potential improvements associated with description of the walkdown team and procedures the decay heat removal function and movabic in-used.

core flux mapping system (for Westinghouse plants) should be specifically highlighted.

3. ' All functional /8ystemic seismic event trees data (in-cluding origin and method of analysis) when NRC C.3 Internal Fires SMM is used.-

The information on the internal fires IPEEE identified e

- 4.

- A description of the most important sequences and below is the minimum that should be documented and

^_

important _ minimal cutsets (for both scismic and-submitted to the NRC.

- nonseismic failures) leading to core damage (NRC method) or a description of the success paths and 1.

A description of the methodology used in perform-procedures used for their selection and of each com-ing the fire IPEEE and a discussion of the status of

- ponent' in the controlling success path. (EPRI Appendix R modifications.

method).

2.

A summary of the walkdown f' dings and a concise

.l m

5. - Any scismically induced containment failures or vul-description of the walkdown team and the procc-i nerabilities and other cor.tainment performance in-dures used.This should include a descri

C of the sights. The procedure for defining the scope of the efforts to ensure that cable routing used i o ly.

- containment performance review, Results of isola-sis represents as-built information anc + mt-tion system, penetration,' and heat removal system ment of any existing dependence betw.

w ie reviews.

shutdown and control room circuitry.

y 6.

A table of fragilities, both gen' etic and plant specific, 3.

A discussion of the criteria used to identify critical used for screening as well as in the quantification.

fire areas and a list of critical areas, including (a)

The estimated ilCLPF for the plant, dominant se-single areas in which equipment failures represent n-quences, and components with and without nonscis-serious crosion of safety margin and (b) same as (a) mic failurcs t ad human actions.

but for dorble or multiple areas sharing common y

l'arriers, penetration seals, llVAC ducting, etc.

1 (Section 6) addressed by the submittal, the basis and

- 4.

tF 7.

Ikcumentation with regard to other scismic issues.

_ A discussion of the criteria used for fire size and 1

assumptions used to address these issues, and a dis-duration and the treatment of cross-zone fire sprcad '

- cussion of the findings and conclusions. Evaluation and associated major assumptions.

results and potential improvements associated with the decay heat removal function and movable in-5.

A discussion of the fire initiation data base, includ-ing the plant. specific data base used, Provide docu.

core flux mapping system (for Westinghouse plants)~

mentation in each case where the plant-specific data 1

should be specifically highlighted.

used is less conservative than the data base used in l

'the following is the minimum information that should be the approved fire vulnerability methodologies. De-

documented and submitted to'the NRC for a reduced.

scribe the data handling method, including major D

scope SMM review:

assumptions, the role of expert judgment, an(.he identification and evaluation of sources of data un-

l... A description of the procedures used to identify -

certainties.

systems (md components for the walkdown in per-forming the seismic IPEEE.

6.

A discussion of the treatment of fire growth and-spread, the spread of hot gases and smoke, and the I

. 2.

A summary of the walkdown findings and a concise analysis of detection and suppression and their asso.

description of the walkdown team and procedures ciated assumptions, including the treatment of sup-j used.

pression induced damage to equipment.

1 i

3, A discussion and the results of any speciGr em po-7.

A discussion of fire damage modeling, including the j

nent capacity evaluations performed, + m. nods definition of fire-induced failures related to fire bar-d used, and assumptions.

riers and control systems and fire induced damage to cabinets. A discussion of how human intervention is

. 4.

lbcumentation with regard to othe scismi'.. sues treated and how fire-induced and non fire induced (Section 6) addressed by the submittal, tu bt.sb and failures are combined. Identify recovery actions and assumptions used to address these issues, and a dis-types of fire mitigating actions taken credit for in cussion'of the findings and conclusions. Evaluation these sequences.

.3 NUREG-14d7

,e, i

Appendix C 8,

I! vent trees and associated fault trees for fire-4.

Results of plant / facility design review to determine initiated sequences. Discuss the treatment of man-their robustness in relation to NRC's current ual suppression, including fire fighting procedures, criteria.

fire brigade training and adequacy of existing fire brigade equipment, and treatment of access routes 5.

Results of the assessment of the hazard frequency versus existing barriers.

and the assochted conditional core damage fre-quency if step 4 of Figure 5.1 is used.

9.

'the estimated core damage frequency, the timing of the associated core damage, a list of analytical as.

6.

Results of the bounding analysis if step 5 of Figure rumptions including their bases, and the sources of 5.1 is used, uncertaintics,if applicable.

7.

All functional event trees, including origin and C,4 liigh Winds, Floods, and Others method of analysis (PRA only).

The following information on the high winds, floods, and 8.

A description of each functional sequence selected, others portion of the IPliEllis the minimum that should including discussion of specific assumptions and hu-be documented and submitted to the NRC:

man recovery action (PRA only).

1.

A description of the methodologies used in the 9.

'IT.e estimated core damage frequency, the timing of examination.

the associated core damage, a list of analytical as-sumptions including their bases, and the sources of 2.

Information on plant specific hazard data and li-uncertainties, if applicable (PRA only).

censing bases.

10. A certification that no other plant-unique external 3.

Identified significant changes (See Section 5.2.2), if e/ent is known that poses any significtmt threa I any, since OL issuance with respect to high winds, severe accident within the context of the screening floods, and other external events.

approach for "High Winds, Floods, and Others."

NURIiG-1407 4

pq+

p yr Appendix C

?

Table C.I Standard Table of Contents for IPEI:I; Submittal i

1.

Executive Summary 1.1 llackground and Objectives 1.2 Plant Familiari/.ation 1.3 Overall Methodology 1.4 Summary of Major Findings 2.

lixamination Description 2.1 Introduction 2.2 Conformance with Generic Letter and Supporting Material

{

2.3 Ocneral Methodology 1

2.4 Information Assembly l

I 3.

Seismic Analysis 3.0 - Methodology Selection (PRA or SMM) 3.la Seismic PRA 3.1.1 Ilazard Analysis 3.1.2 Review of Plant Information and Walkdowr 3.1.3 Amilysis of Plant System and Structure Resyse l

3.1.4 Evaluation of Component Fragilities and Failure Modes l

3.1.5. Analysis of Plant Systems and Sequences

[

3.1.6 Analysis of Containment Performance 3 lb Seismic Margins Method (SMM)(NRC, IIPRI, or Reduced SMM) 3.1.1 Review of Plant Information, Screening, and Walkdown 3.1.2 System Analysis 3

3.1.3 Analysis of Structure Response 3.1.4 I! valuation of Seismic Capacities of Components and Plant 3.1.5 Analysis of Containment Performance 3.2 USI A-45, GI-131, and Other Seismic Safety Issues F

4.

Internal Fires Analysis j

4.1 Fire llazard Analyas 4.2 Fire Growth and Propagation i

4.3 = Fire Detection and Suppression 44 Analysis of Plant Response'-

= 5.

liigh Winds, Floods, and Others 5.1 High Winds f

5?

Floods-Transportation and Nearby Facility Accidents Others

{

i 6.

Isensec Participation and Internal Review'l cam

.. IPHfiE Program Organization

. 6.2 Composition of Independent Review Team 6.3. Areas of Review and Major Comments 6.4 - Resolution of Comments 7.

Plant improvements and Unique Safety Features 3; Summary and Conclusions (including proposed resolution of USIs and GIs) 5 NUREG-1407

CHC sopiv 336 U S. NUCL( AR R(0VLQ10RY COMMIS$10N L HiPuhI NUMBt h Mc$itW, e o" ion nee fIf v7 N

w.m BIBLIOGRAPHIC DATA SHEET q,,,n,,on.,w v.,,,,,,,,e NUREG.1407 Lillit AN0 5UB fl1LL Procedural and Sutxhittal Guidance for Individual Plant Examination of External Events (IPEEE) for Severe Accident 3

oAtt areoai rvatissio Vulnerabilities j

'a^*

m '~

Jul y 1990 Draf t Report f or Comment

..,,N on caANt NuMet n b AVIHQH(Li

6. T YPE OF REPOH1 1, Pt HIQD CoviH E D ono ww ca,es.n

'J"l,*a"LQL",",""*"*"~"*"'^""^""""'"'*"'****~""'"'**"*""**'"*'**""""**"'"*""""'"""*""""*

Division of Safety issue Resolution-

.l Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555

9. SPONSOH ING ORG ANilAllON + N AM E AND ADDR t$5 ff,4RC, erw ~4.me en atew", si rent,scro,, e,orsik NRC D,rtsea,0,'ne o, A*,soa. U S A whs, NeMsto,y Comep**pa.

. Omdf mtdshald astsfrus4J 1

Same as above j

10 $UPPLIMENT ANY NOTES

11. AB S T R Ac t rite.oro, o, =w)

Based on a Policy Statement on Severe Accidents, the licensee of each nuclear power plant is requested to perfonn an individual-plant examination. The plant examination systematically looks for vulnerabilities to severe accidents and cost-effective safety improvements that reduce or eliminate the important vulnerabilities.

This document presents guidance for perfonning and reporting the results of the individual plant examination of external events (IPEEE).

The guidance for reporting the results of.the individual plant examination of internal events (IPE) i= presented in NUREG-1335.

l I

ra i Y wV H us. us sc m r i on s,o...o, r,... ~,..,,..m,. -,,,,,, -.,-, ~,,,.,

it....m u n o m ~ i Unlimited L

Severe Accidents Policy Statement, i t s. -, n -.c. i.*

L Individual Plant Examination, External Events, Vulnerabilities,

<r=,.

IPEEE Guidance l

Unclassi fied

, i-u.nc,,

1 Unclassified 16 NUMtlkH OF PAGkS l'

l-16 PHIGE j

m

~" ai cass uas 7

UNITED STATES '

.' NUCLEAR REGULATORY COMMISSION M ^j,*, "' S ' ^*

E

-m

. WASHINGTON, D.C. 20555-

-Q re m n os. o n OFFICIAL BUSINESS

.h' w.

PENALTY FOR PRIVATE USE, $300

~-

i 1 1A0 SVCb Ik f4 posL1CA110NS j

C-gU l % "g 20sss ox' c

oC n!Mnoso" a

L n

04 m e-m-

N Y fb et

-M[

w

.gj-.

m:

3E

.g6 lat u

E=t in -

5 m-g -.

. m.-

>t OE O<

'e.

M L

'di r-7 6

$~

m 1.L Ei

-C:

d=

Qi

+'

N c

d c:

I t

.s If I

_____--__ _._-__.__-.- _____- --___.__ - - - -- - __ - _ _-