ML20009C082

Proposed Factual & Legal Findings Re Remanded Seismic Issues & Supporting Argument.Aslab Should Find Chinnery Probabilistic Hypothesis Factually Valid.Intervenor Motions to Strike Should Be Granted.W/Certificate of Svc

ML20009C082
Person / Time
Site:	Seabrook
Issue date:	07/15/1981
From:	Bernabei L, Jordan W HARMON & WEISS, NEW ENGLAND COALITION ON NUCLEAR POLLUTION
To:
References
NUDOCS 8107200256
Download: ML20009C082 (56)

Text

'

r , .

C) g "

• UNITED STATES OF AMERICA N2. { 7.

Ir _n'.-

NUCLEAR REGULATORY COMMISSION V, u s. m%Dros, BEFORE THE ATOMIC SAFETY AND LICENSING APPEAL hfS

)

In the Matter of )

)

'PUBLIC SERVICE COMPANY OF ) Docket Nos. 50-443 NEW IIAMPS!! IRE , et al . ) 50-444 (Seabrook Station, Units 1 and 2) V

/l- DOCKruED 's emre .

-4p,

!UL161981 *

~

M-ctre C;; a t' r e Nrchry c"CO2 3

NEW ENGLAND COALITION ON NUCLEAR POLLUTION 6 - h:h PROPOSED FACTUAL AND LEGAL FINDINGS ON &

REMANDED SEISMIC ISSUES AND SUPPORTING ARGUMENT '%$/

t William S. Jordan, III Lynne Bernabei IIARMON & WEISS 1725 Eye Street , N .W .

[

Suite 506

j. Washington, D.C. 20006 (202) 833-9070 >

l DATED: July 15, 1981 6' I l

8107200256 810715 gDRADOCK 05000443 PDR l

F TABLE OF CONTE!:TS Page I. Introduction and Statement of the Case................ 1 11 In the Absence of a Pr obabil is *. i c Justification to the Contrary, the Record Requires an SSE of MMI XII..............................................- 4 III. The " Chi nnery Methodology " for Determining Earthquake Return Periods ir Factually Valid, and the loplicant and Staff Have Not Met Their Burde n of Proof With Respect to Determination of the SSE............................................ 9 A. The Standard f or Determining " Factual Validity ".... 10 D. Dr . Chinnery's Methodology for Estimating Earthquake Retarn Periods is Fundamentally Sound and is the Only Method That Provides a Rational Basis for the Choice of a Safe Shutdown Earthquake............................... 12 (1) The Factual Validity or Dr. Ch:- ane ry 's Methodology........................... ..... 12 (a) Linearity............................... 13 (b) Uniform Slope of 0.57............... ... 17 3ci Upper P.,und............................. 20 (d) Extrapolation Beyond the Hi s t o r i c a l R e c o rd . . . . . . . . . . . . . . . . . . . . . . . 23 (2) General Criticisms of the Chinnery Methodology................................. 27 (a) Dr. Chi nnery addrcsses and admits uncertaintics in i.is methodology more forthrightly and to a far greater degree than either the Applicant or the Staff................. 27 (b) Dr. Chinnery fully considers geology to the e x te n t possible......... 31 (c) Dr . Chinnery's probabilities for the Bostor-New Hampshire zone must be assumed 'o ar ly to the Seab roo k site............. ..................... 33

(d) T he record does not support the argument that less damage wculd occur to a nuc lear plant sited on bedrock than on alluvium........... 34 (c) The Staff and the Applicant must undertake a thorough reevaluation of the seismic design of Seabrook cased on the return periods esti-mated by Dr . Chinnery................. 36 IV. The Staff's Methodolc,gy For Correla ting Vibratory Ground Motion With the SSE Is Scientifically Invalid and Has ResulteG in a Design Response Spectrum With a High Probability of Being Exceeded... 40 V. The Pending Motions to Strike Shauld be Granted...... 46 A. Testimony Discussing the Impact of the Chinnery Hypothesis is Beyond the Scope of this Proceeding............................... 47 B. Appendi x A Requires that Earthquake Probabilities for the Province de Assumed at the Site and Prohibits At temp ts to Reduce those Pr ob ab i l i t ie s . . . . . . . . . . . 48 C. Mr . Kni j nt 's Tes timony Concerni ng Conservatisms Inherent in the Desigt Basis is Irrelevant to Determining What the Design Basis should Be...... 50

oNCLUS;0N..................................... .......... 51

gs UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING APPEAL BOARD In the Matter of

)

PUBLIC SERVICE COMP ANY OF ) D;cket 'os.

, SJ-443 NEW HAMPSHIRE, et al. ) 50-444

)

(Seabrook Station, Units 1 and 2) )

)

NEW ENGLAND COALITION ON MUCLEAR POLLUTION PROPOSED FACTUAL AND LEGAL FINDINGS ON REMANDED SEISMIC ISSUES AND SUPFsRTING ARGUMENT

.. Introduction and Statement of the Case Af te r a myriad of decisions in this proceeding too numerous to detail here, the App ea l Board now has before it a new f actual record on which it must judge two narrow issues remanded to it by the Commission. The relevant prior rulings are the Decluion of this Board and related dissent and supplemental opinien, Public Service Co. of New Hampshire (Seabrook Station, Units 1 and 2), ALAB-422, 6 NRC 33, 54-65 (1977), and ALAB-561, 10 NRC 410-435, 436-a - 436-n (1980), and the Commission's decision to remand on seismic issues, Public Service Co. of Neuf Hampshire (Seabrook Station, Units 1 and 2), CLI-8 0 -3 3, 12 NRC 295 (1980).

In the former, the Appeal Board re j ected Dr. Chinnery's methodo-logy for predicting the return periods of earthquakes in the Seabrook area as technically de ficient and incons is ten t with 10 CFR Part 100, Appendi x A , and accepted the Applicant's and

m Staff's posit ion tha t the Safe Shut down 1,ar t hquake ("SSE")

should simply be the earthquake of highest intensity in the hi s to r i c al record, Modi fied Mercalli I.donsity ("MMI") VIII.

The Appeal Board also a cce p te d the Stat f's methodology for cor re 1 a t i ng vibratory ground motion W1th the SSE and approved a response spectrum anchor point of 0.25g. In a vigorous dissent, Mr . Farrar argued that the majority had improperly placed the burden of proof on the intervenors and had ignored the mandate to take a conservative approach to the interpre-tation and applic ation of' Appendix A.

The Commission essentially agreed with Mr. Farrar in ruling that a probabilis tic approach to estimating earthquake return per iods - as a basis for determining the SSE is consis ten t with App end i x A . In so doing, the Commission emphasize:d both the cur ren t lack of unders tanding of earthquake science, including the lack of a theoretical basis for the Staff's choice of SSE, ar need for conservatism in applying what knowledge and unders tanding are available. Af ter re ject ing the geological conditions on which the Board based its decision as "inconuistent with the intent of Appendix A to provide a conservative approach to determining the SSE in the absence of a theore t: cal basis for such a de termin a tion , " the Commission stated, As this yet early state in earthquake s c ien c e ,

we are not prepared to dismiss an empirical relation on the basis of failure to satisfy criteria, which although they may appear reaso nable , imply a greater understanding of the relation between geology , seismology, and earthquakes than is actually available.

4

• h * * * *

• Accord ing l y , in view of the ne"d for conservat1sm in this arei, we f2nd Dr. Chinnery's methodology is nct inconsistent w i t h A pp en u i a 7. .

Id . at 297.-1/-

F1th consers a t 1sm as the wat chword , the Commtaston remanded the c as e to this Board to consider .he f ollowing narrow i ssues :

1. The " factual validity of Dr. Chinnery's hypothesis," and

2. The " con sis tency of Appendi x A and s ta f f's methodology for correlating vibratory ground motion with the SSE."

Id. at 297-98. The Appeal Beard further emphasazed the narrow scope of the proceeding in its Memorandum and Order of February 12, 1981, in which it ruled that uncer t a inties and flaws in competing methodologies are not relevant to determining the validity of Dr. Chinnery 's me thodology . The issue with respect to the Chinnery methodology, t he re f ore , is solely whether, in the aosolute sense, it has scientific validity.

'. n compli an ce w i t h t he Commi' tca's remand, the Boari held an evidentiary hearing on April 6 -9 , 1981. Pursuant to agreement of the par tie s , the Applic an t filed its ma.n brief on May 8, 1981, the Staff f iled its main brief and any responses to the Applicant's arguments on June 16, 1981, and the New England Coalition on Nuclear Pollution ( NECNP ) is now filing its main brief and responses to the Applicant and Staff. The Applicant and Staff h ave the opportunity to f ile final briefs, limitad to responses 1/ Since much has been made of the fact that the Commission's Temand was a split decision, we note that Mr. Ahearne agreed with Mr. Farrar and the majority on these points. Id . at 299.

4_

to NECNP 's arau ents, by August 10, l ') H 1.

This brief wi 1 1 demons t rat e t hat :

1. The record requires an SSE of ?D!I XII, er at least MMI X, unless a probabili st i c theory such as that proposed by Dr. Chinner y est ablishen that a lower intensity is acceptable.

2. The "C hin le ry methodolo Jy " is valid under the standard by which ' validi ty "

must be de t er mi n e d .

3. T he Staff's methodology for correlating vibratory ground motior wit h the SSE is scientifically invalid.

!. A complete re-analysis of the seismic design of Seabrook is required based on the Chinnery methodology and the high probability of exceeding the approved design response spectrum.

5. The pending motions to strike should all be granted.

II , In the Abs en ce of a Prob abi l is t i c Jus t i f icat.1 on to tt:e Contrary, the Record Requires an SSE of !C1T XII.

Des p it e the narrow scope of talu proceeding, th" evidence adduced at the hearing establishes that the SSE for Seabrook must be i n cr e a s ed considerably regardless of the validity of the o

Chlnner y methodology . Indeed, the Chinnery methodology or some other probabilistic 7pproach is the only basis on which to establish an SSE le: s than MMI XII (or MMI X If a non-conservative approach is taken to the evidence).

According to th' App l i c a n t ' s w i t n e s .s , "r. !!al t , there is a

patial correl a t ion between l a rge earthquakes and geologic struc-2/ In light of its public interest responstbilltles, the board

,must consider al l of t he. implications of this evidence regard-lesa of the parti ular terms of the Commission'> re na nd . Duke Power Q . (Catawba Nuclear Station, Units 1 and 2), LBP 35 , aRCI-75/6,

p. 626 , 054-655 (1975). That is particularly true here, where the Applicant volunt arily presented the evidence in 'guen t ion for its own pu r pos e s . The Applicant cannot argue that the Board may consider the evidence to the extent that it is favoracle, but must ignore it to t h .s extcat that it is unfavorable.

-S-tures known as "in t r u s i ons " or " plutons , " .thich are associa+ed with l a*r g e trending faults. (liol t , Tr. p. 381). This is ' rue both in New Engl a nd , where the la;ge eart briuakes can be related to the Ossipee complex in the White Mountains or to a pluton neat Cape Ann ( flo l t , Tr. o. 381-183), and in the rest of the Eastern United States. In particular, he f inds the theory vall1 for the areas of New Madrid and Charles ton , South Carolina, which have experienced earthquakes of MMI XII and MMI X, respectively, in recent history. ( flolt , Tr. p. 448-450, Direct t o 4 t i m o n ,> ,

Figtres 4 and 5). tie also contends t h a t. at least the New

!!ad r i d intrusions are similar to those in New England.

( flolt , Tr. p. 405, 1. 4-8 ) . Mr. Ilol t 's view s supported, at leas t in part, by two United States Geological Survey s tudies th.t relate large eart hy 2akes to plutons both in the Mississippt embayment and in New England. ( !!al t , Tr. p. 431).

Although Mr, llol t and the Applicant did not take the narrow position that the New Eng land plu cons controlled large earthquakes in t. h e a re a and thereby argue for an even smaller SSE (flolt, 'i c .

p. 440, 1. 11-23), Mr. lloi t firmly believen that the large earth-quakes and the plutons are related. Indcod, he believes that the evidence is strong enough to allow him virt ually to predict the occurrence of the MMI XII New Madrid event, assumang that it had not already occurred. (llol t , Tr. p. 447-450, esp. p. 450, 1. 10-16).

Witn re s p ec t to the Seabrook area, in particular, Mr. 1101 t testified that that the pluton that correlates with the Cape Ann eart hquake is approximately 15 milo . from seabrook, (110 1: , il .

( -

p. 390-391, 420, 1. 13-16). lie also te'; t i l led that the pluton i tt el f is about 10 ki l omot e rs wide and t hat 1t could generate an eart hquake 15 niles away. ("olt, Tr. p.'461). Therefore, the pluton could generate an ear t hquako of !Ott XII int ensi t y on the Seabrook site. Mr. !hlt attempted to mitigate the impact of this in formation by noting t hat the eart hquake -

plu ton rela t ionship rel a t es to the existence of trending f au l ts , which in the case of Cape Ann do no: trend in the direc-t.lon of Seabrook. (llol t , Tr. p. 461). However, there is absolutely no evidence that those faults have been act.ive in any Way in recent geologic history, even when the 1755 Cape Ann >arthquake occurred.

l'ur t her more , no historical evidence indreates that earthquakes corra.la ted wi t h t he pluton have occurred only along tne trending fau1ts.

The only conservative conclusion t h a t. can oc reached from thts evidence is that the Cape Ann plut on is ca p ab l e of generating an eart hquake of MMI XII, or at leact MMI X (as a t. Charleston), a t.

Seabrook. There is no evidence that o t. r unders tand ing of the ph en om en o n i s great enough to allow the conclusion that the plu-ton cannot generate ca rt hquake s as large as thos- Jenerated by sim.lar plutons, or (Nat it cannot generate such an earthquake at s'abrook. Undoub ted l y , the Applicant and the Staff will argue that the soil-bedrock dis tinc tion woul d l e ..ien the intensity of any earthquake occurring at Seabrook. We discuss the weakness of that arg umen t bel ow at p. 34-36, but it suffices in this context to note that none of the evidence on that point allows the reduction of an earthquake from MMI XII or MMI X to MMI VIII.

m I n t he se circumstances, A[,p en d i x A r e q u i r r'< that the SSE be esi abl fs h ut at MMI XII, or at least >D;I X, unless a probabilistic theory such as Dr. Ch a n nery 's provides a just i f icat ion for a louer value. Under SIII(c) of Append i x A, t he Safe Shutdown Eartnqoske is that earthquake which is based upon an evaluation of the maxtmum earthquake potential considertng the regional and local gt-ology and seismology cind speci fic characteris tic s of local subsurface mat e ri al . It is that earthquake which produces the maximum vibratory ground mot ton for which certain structures, sys tem s , and components arr designed to remain functional.

( emph as i s supplied). On its face, Append ix A does not allow anything less t h an the maximum earthquake potential to be chosen, regardless of the rarity or return period of the event.

Recognizing "the limited geophysical and g"'c logic al I nf orma-tion available to date concerning f aults and earthquake occurrence a nd e f f ec t , " (91), Appendix A dictates that the effort to determine the SSE hegins with the iden t i f ic at tor of the historic earthquake of max imum in te ns i ty co rre l e.ted wi t h tv c tonic structures and provinces surrounding the sit e . However, the maximum historic earthquake related to either a tectonte s truc ture or a tectonic pro'inco is only the minimum "alue that may be consi d' red for the SSE. Where "ge o log ic al and seismological data warrant, the Safe Shutdown Earthquake shall be larger" than the maximum h2storic eartFquake. (emphasis supplied) S V( a )( 1 )' i ) and (iv). Once the correct m ag n i t u de or inte nsity of the SSE is determined, the elfect on the site is evaluated based on the tect oni c structure or province to which the SSE 18 re l a ted ,

le ~urrent Seabrook SSE is based oa the maximum hist.oric carthquake of MMI VIII which occured at Cape Ann. It is assumed

s

-U-to occur at the s .i t e since at the _ime the SSE was initially deter-m1 Sed,'the Cape At n earthquake was not cor relat ed wit.h a pluton or in t ru sive structu e. Since there was apparently no indication at the time that such a correlation would indicate a hinnel intensity SSE, this was the uonservative approach.

Now, however, there is ancontroverted evidence in the record from Mr. liolt that the Cape Ann earthquake correlates with a pluton which could generate an earthquake at the Seabrook site and that the pluton is similar to others that have gener ated MMI XII and MMI X earthquakes in recent years. Section V(a)(1)(i) of Appendix A requires that such earthquakes be as s u n<ed to occur at the site since the pluton or intrusive structure is capable of generating such an earthquake at the site. More importantly, the dic tates of Appendix A clearly

. require that

[T]he procedures in paragraphs (a)(1)(1) through (a)(1)(iii) of chis section shall be applied in a conservative manner.

Th is is precisely the point. the Commission emphas ized in remanding the case to the Appeal Board. A conservative approach to this ev i denc e requires a dete rmination that the " max imuu earthquake pot en t i al " for the Seabrook site, based on available geologic ev iae nc e , is a MMI XII.

While this is the inescapable conclusion based on Appendix A and the available geologic evidence, NECNP re cogni z es that despite t he max imum earthquake potential for the site, the likelihood of a MMI XII may be so remote that it should not be considered as the SSE. Ilowever , one must find a sound basis to determine that the likelihood cf the max imum potenti al earthquake is so low that the possibility of i ts occurrence need not be included in any seismic analysis. Only a probab i l is tic analysis of

(. w. r r--s y : & -- t~ ~

-9 the sort propos d by Drs. Chinnery and Tr1iunac provides a rational scientific basis for making this judgment. At this poinc, the Chinnery methodology may provide a bas is for choosing a MMI X, or possibly a MMI IX, depend ing upon the analysis of vibratory ground motion for those intensttles. ,

Dr. Tri f unac's approach may eventually justify a lower SSE, but by his own testimony it is clear that he has not yet undertaken an analyJis sufficient to determine whether his results are acce p tably conserva t ive . (Trifunac, Direct Testimony, p. 10).

III. The "Chinnery Methodology" for Determining Earthquake Return Periods is Factually Valid, and the Applicant and Staff Have Not Met Their Burden of Proof Wi th Respec t to Determinatton of the SSE.

The f irs t ques tion that the Commission remanded to the Appeal Board for consideration is the " factual val idity of Dr. Chinnery's hypot hesi s. " That hypothesis concerna the es timat ten of earthquake return p er i ods , carticularly for earthquakes of large intensities that are not cont a t ned wt t hin the hist orical record of an area, or that have occurred so rarely in human history that their oc c ur r e nc es do not provide data adequate to estimate return p er iods .

Dr. Ch 2n nery 's hypothesia is not and dood not purport to be, a comp 1ete methodology for determining the seismic design of nuclear power plants. In terms of Appendix A, it is a methodology that p er m i t s a rational evaluation of the likelihood of occurrence of earthquakes equal to or larger than his tor ic earthquakes o f greatest intensity, and it provides se ismo log ic al data that may warrant a choice of SSE larger than the largest historic eart hquake . In this c as e , NECNP helieven that it warrant s a choice of SSE at MMI ]X or MMI X.

The Chinner y .,e t hodology has been discussed thus far in te r ras a ,

of tour "assump ti a n , " linearity, uniform slope, lack of upper bound, and extrapolat ion. We will adhere to that approach, al-though t he evidence demons tra tes that none is strictly an assump-I tion, and the first two in particular are based on empirical ob-serve:i ons of available data. Bef or e d is c uss i ng the "assum p ti ons "

and the issue of factual validity, however, we ftrat address the standard by which f actual validity must be judged. Due to the limitations on the scope of this hearing, and the conservative appproach mandated by Appendix A and by our minimal understanding of earthquJe science, the " validity" threshold is extremely low.

Finally, we will addr ess gener al criticisms made by the Applicant and Staff.

A. The Standard for Det e r mi n i ng " Factual Validity.

The Commission was clear in ruling t h i t, " t h e use of Dr. Ch ir.n er y 's hypothesis concerning earthquake return pm1D)ds ir consistent with Appendix A. Ilowe ver , it provided no cri t eria by which to judge the "f ac tu al validity" of that hyp ot h es i s . Rather, it simply s ta ted that " greater e xplora tion on the record is required,* Public Service Co. of New ilamps h ir e (Seabrook Station, Units 1 and 2), CLI-8 0-3 3, 12 NRC 295, 297 ( 19 8C ' From that s ta teme nt , we cannot determine whether the Commission believed the record to be inadequate to support Dr. Chinnery's hypotbrsis, inadequate to refute it, or simply inadequate to reach a juugment. In this circumstance, the standard by which Dr. Chinnery's methodology is t o be judged must be garnered f rom the more general language of the Commission's decision, the l angu ag e of Appendix A, the c ur re nt state of earth-quake science, and the particular status of this case.

If there were a platonic ideal at seismic methodology agj i ns t, whi c h t he Channery hypothesis could be compared, one could Judge ghether particular uncertaint tes or assumptions are acceptable or, instead whether they render the hypothesis

ci ent i f ically invalid. No suc h ide.C e is t

, . Rather, we must work with the minimal information and understanding that we have, which is the primary reason the Commission emphasizes the need for conservatism in addressing these questions. Accordingly, NECNP believes that the Ch inner y hypothesis cannot be judged in the abstract, but must be compa red to other approaches, all of which are subject to significant uncertainities. Since the Chinnery hypothesis re f lec ts unc er t a : nt ies that are no gre., tar than those involved in approaches taken by the Staff and Applicant, and since the Chinnery hypothesis alone provides a rational scientific basi- ior determining seismic risks, it is undoubtedly "compar a t i vely " valid . Howe v +:r , the Board has ruled that this type of comparison is outside the scope of this proc eed ing ,

and the standard for decision must be found elsewhere.

Firs t, the Board is making only a "t hres hold" decis ion on t he scienti fic validity of the Chinnery hypothesis and not a dec is ion about the acceptability of the Chinnery hypothesis in lieu of other theories. Tnere f ore , t he evidence thus far presented need not met this far more . e .ze r e test. To re j ec t the Chinnery hypothesis, the Board must find that :t is virtually impossible for Dr. Chinnery to be correc t.

Second, the Board must look to Appendix A's mandate of con-

~ .

servatism. If Dr. Chinnery's hypothesis is plausible, a conserva-tive approach der ands that it be considered valid and compared with al tern a ti ve approaches, which NECNP contends (but has been precluded from demonstrating ) lack any sound scient i f ic bas is for choice of the SSE.

Th ird , the Board must consider the <quali f ic at ions of Dr .

Chinnery and the sources in which his hypothesis has been published. There has been absolutely no challenge to Dr.

Chinnery's qualifications, and his hypothesis has b een publ is h ed in the regular scientific literature, which is sub-Ject to peer review in being selected f or p ub l ic a t i on . In the absence of any absolute standard, or any comparative standard ?f validity, the Board mus t weigh signliicantly the evidence that Dr. Ch in ner y 's peers consider his hypothesis to be valid and valuable for the purpose of publication.

While this alone does not demons tr a te that Dr. Chip ery's methodology should be adopted for Seabrook, it has s tro ng bearing on any threshold determination of its factual vailidity.

B. Dr. Chinnery's Methodology for Estimating Earth-l quake Return Peri ds is Fundamentally Sound and is the Only Method That Provides a Rational Basis for the Choice of a safe Shutdown Earthquake.

(1) The Factual Validity of Dr. Ch [,r.tiery 's Methodology.

Dr. Chinnery's methodology consists of the following four elements, each of which is addressed in turn below:

a. The empirical observation that earthquake frequency and intensity appear to have a linear relationship.

9

P

b. The empirical observa t ion t hat the linear relationship of frequency and i n t e ns i t y

. . app ears to have an essentially uniform slope in the Ea ste r n , Uni t ed dt a te s , and p er haps in other areas, including Ca l i f o r ni a .

c. The lack of any basis f or exc luding the possibility that an eart hquake of MMI XII can occur in the Seabrook area at some

" fu ture time.

d. The extrapolation from the historical data on the linear curve of return periods of large earthquates do not appear in the historical reco.d.

(a) Linearity The first element of Dr. Chinnery 's approach is the examina-tion of earthquake frequency versus intensity data for three areas of the country, which he undertook in two papers. The first was a 19 73 study in Earthquake Notes which concerred earthquake s ta tis tic s in New England (NECNP Ex. 1), and the second was his 1979 paper published in the Bulletin of the Seismological Society of America, in which he com pared frequency-intensity data from the Sout heas ter n United States, the Central Mississippi Valley, and Southern New England (NECNP Ex. 2). lie concluded that the data demons tr a ted a remarkably linear relationship between frequency and intensity. His conclusion is consistent with the view of the vast majority of seismologis t s that a linear relationship exists between f requency and magnitude and with the opinion reached by several inve s tiga tors that such a relationship exists between magnitude and intensity. (Chinnery , Direc t testimony, p. 8-11, Tr. p. 197, 1. 1-22).

The Applicant and Staff criticized Dr. Chinnery's use of the data, especially his failure to adopt the more complex relation-

r i l

_14 l shi p indic ated by the plots of earthquakes r a ng ing from MMI I to M$I IIf , and the fact that he did not revise h is plot based on recent downgrading of MMI VII earthquakes in the Smit h cat alogue.

However, Dr. Chinnery ef f ectively deal t with these points in his direct te s timony . Fi rs t , he explai ned that a relatively recent time period must be chosen f or the lower intensities, and that even then the data are probably not complete since the lower intensities cannot be f el t clearly, and probably have not all been reported. (Channery, Direct testimony, p. 8, Tr. p. 59, 155-56).

Indeed, even Mr . !!olt I ndic ated that it is necessary to adopt a rec en t time per iod wi t h respec t to lower intensities. ( lio l t ,

Tr. p. 373). Secon1, Dr. Chinnery explained that the number of MMI VII events in the data base was probably unusually high, so that if some or all of these events were dow ng r a ded , the plot would become more, rather than less, consistent with the linear hypothesis. (Chinnery, Direc t te s t im ony , p. 8, Tr. p. 133, 1.

16-20, p. 13". 1. 1-7).

Dr . Chinnery concludes, re s pec t i ng the Bos t on-New flampshire zone, that the MMI IV data are probably also in-complete. Therefore, he is left with only two reliabl.e data points, one for a MMI V and one for a MMI VI. As a result, he does not rely on those points to establish linearity for the zone, b.t on the fact that linearity is established elsewhere and is consis te nt with what is known of the New England data.

(Chinnery, Direc t testimony, p. 8).

Mr. Ilolt also at temp ts to attack the linearity hypothesis by arguing that Dr. Chin 6ery i ncorrec tl y res t ric ted the time

are consistent with linearity, and provide one of the best linear fits Dr. Chinnery has seen for New England. (Chinnery, rebuttal testimony, p. 7, Tr. p. 110-111). Much the same is true of the Bloom and Erdman datu, which the Staff cites to i nd ic a te a non-linear relationship. Again, these data are consistent with .

linearity, and t' o abse nc e of a j us ti f ic at io n for a more complex rela t ionship dic tates that linearity be accepted. (Chinnery, Rebuttal testimony, p. 6, n. 1, Reiter, Tr. p. 510-512). 3/

liowever , vas tly more important than the App l ic an t and St af f 's a t tempt to criticize particular pieces of data, is the fact that t he scien ti fic community has accepted the linear hypothesis. This is true not only of those whom Dr. Chinnery cites, but of Dr . Tri funac as well, who tes ti fied that reliance on linearity "is a very typic al daily process used by hundreds of seismologists world wide and I meant to say that all of us do this and I just wanted to emphas i z e t h is . " (Trifunac, Tr. p. 751,

1. 22-24, 776 ). I nde ed , eve n Dr . Jackson accepted linearity up to a po2nt, although he would truncate the li ne . 4 / (Jack >n, Tr. p. 573).

3

_/ The Staff also a ttempts to create an argumerit agai ns t linearity based on the fact that the line will begin to curve downward slightly at the highes t in te ns it ies as it approaches the upper bound, if there is one. (Staff Brief at p. 6). liowe ver , the change is very slight, and it has virtually no e f f ec t even at the intensity immediately below the upper bound. (Chinnery, Tr. p. 154 -i55). The more significant point to be drawn from the fact that the line should begin to curve near the upper bound is that there is no evidence in the data to indicate the exis tence of such a curve. Th is lack of a downward curve may well ind ic ate that the data do not yet begin to approach the upper bound. This supports Dr. Chinnery ' a argument that the upper bound in New England is at least MMI X, if not greater.

4/ Dr Jackson's differing approach is relevant only to the separate, issues of "u pp er bound " and " ext rapol at ion . "

1

- 1.7 -

(b) Uniform Slope of 0.57 i

The second element of the Chinnery methodology is the empirical observation that the linear frequency-intensity curve appears to have ; tihi form slope for the Eastern United States, and perhaps for the country as a whol e. Dr. Chinnety demonstrated this phenomenon in his 19 79 paper (NECNP Ex. 2),

in which he found that slope values in the Eastern U.S. lie in i the range of 0.54 to 0.60 and are consistent with his preferred

~

slope of 0.5 7. (Chinnery, Direct testimony, p. 11). This app ea rs to be true not only in the Eas t, but in California as well (Chinnery, Tr. p. 186-lf', NECNP Ex. 2, p. 765 ),

although various areas of the East and West are quite dissimilar geologically and tec tonic al ly . Nonetheless, the data appcar to demonstrate such a relationship. (Chinnery , Tr . p.

186-187, 19 7, 1. 1-22).

Again, this hypot h es is is widety accepted. For exampl<,

McGuire, on whom the Staff relied, proposes a uniform slope for the Eastern United States (Chinnery, Tr. p. 275-277, Reiter, Tr.

p. 513-514), and even Dr. Reiter admits that the adoption of a uniform slope m:2y be valid, although he believes it to be subj ec t to substantial uncertainty. (Reiter, Direct testimony, p. 6).

I nde ed , the issue of uniform slope is perhaps the least hotly contested th this case. However, the nature of the StafE's criticisms raise serious questions about whether the Staff was a t tem pti ng to f ul fill its mandate to assure a complete record for decision, or whether it was simply picking and choosing information for its own convenience to ensure that the Board would have to

r adopt the Staff's position rather than make an i nde pende nt assess-ment. Jn particular, Dr. Reiter cites slope values of 0.24 to 0.76 without giving any indication of the weaknees of the claims at the extremes of that range. While he does later adntt that the bulk of slope values fall betwc.en 0.4 and 0.6, he apparently has done no analysis of, and has provided the Board with no inf ormation on whether the narrower range is considered the scientifically significant range and is the one with which the Board should concern itself. As a result, he conveniently, but dec ep ti vely , leaves an impression of f ar greater uncertainty in the range of slope values than actually exists. This by itself is a serious shortcoming. However, it is compounded by the Staf f 's selective use of the McGuire article. Dr.

Reiter apparentli found it useful to do an i nde penden t determin a tion of the impact of varying s l op es , as discussed by McGuire, to use as a tool against Dr. Channery. However, at aa time did Dr . Reiter under take any s tudy of the bes t method to determine the slope; nor did he undertake an i nde pen de n t evaluation of McGui re 's pre ferred slope. (Reiter, Direct te s timony , p . 5 C , Tr . p . 513-416 ) . That analysis might well have provided important in f orma tion to the Board, demonstrating that while there is a r a nge o f sl opes and while variations in slopes may produce signi ficantly different results, there is also rela ti ve agre eme nt on what the slope should be. Dr. Reiter 's f ailure to address that point raises serious questions concerning his credibility and that of the Staff as a whole. Indeed, it indi c ates that the Staf f h as not f ulfilled its responsibilities in this case to act to inc re ase public knowledge and to serve the public

interest.

Unf ortunately , the Staf f's gamesmanship wit h t he evidence does not stop there. Rather, Dr. Reiter claims that Dr. Chinnery's estimates of return periods for earthquakes not contained in the his toric al record are based on only two data points at low inten-si t ie s. (Reitet, Rebuttal tes timony , p. 4). This is a gross distortion of the Chinnery methodology. Far from bas ing his return period esti nates on two data points, Dr. Chinnery relies on data from three different areas of the Eastern United States. His 1979 paper refers as well to consis tent slope findings by many other authors. Unlike the Staff, which narrowed its focus to a smal.1 area of New England and, in reality , relied solely on the maximum his torical earthquake, Dr. Chinr.ery reached out to the entire seismological community and found signi ficant support. It is simply unf air and inaccurate to characterize his conclusions as an extrapolation from t.wo data points. To the contrary, it is significant that the only two reliable data points in the Bos ton-New Hampshire zone are 5/

completely consis te nt with the pos tulated uni form slope.- '

Since the uniform slope and the reliability of the two data points are i ndependen t conclusions, the fact that they are per f ec tly cons is te nt argues very strongly in favor of this hypothesis, as does the fact that the occurrence cf the Cape Ann earthquake, which was not in the data base, is not incon-sis tent with the slope of 0.57. (Chinnery Tr., p. 265-267).

Finally , whether the range is considered to be from 0.24

_3/ We note that while two seems a small number of data poi t.t s ,

the universe of potentially reliable data points extends, at a m ax im um , to only four aata points, MMI IV to VII.

m to J 76 ; 0.4 to 0.6 ; or 0.54 to 0 . 6 , the uniform slope of 0.57 proposed by Dr. Chinnery leads to among the lowest probabilities and longest return periods for large earthquakes. ( Ho l t , Tr, p.

376 , Reiter, Tr. p. 517). If the contrary were true, and Dr.

Chinnery were arguing the v xtreme case for high probabilities, the Staff and the Applicant would have a strong a rg u.T.en t against him. Hosever, from a large range of slopes proposed, by a large number of scientists, Dr. Chinnery proposes one of the lowest probabilities of occurrence of large earthqaakes, yet his results are appare ntly still higher than the probab311 ties asserted by 6/

the Applicant or the Staff. Dr. Chin ner y 's position can hardly be characterized as extreme in preposi ng this uniform slope, and the mandate for conservatis.n requires that it be considered f actually valid.

(c) Upper Bourn As previously discussed, the tes timony of Mr . Holt considera-bly s trengthened Dr. Chinnery 's assertion that t here is no basis for deciding that an earthquake of MMI XII, or at least MMI X, cannot occur in the Seabrook area. Based on Mr. Holt 's tes timony ,

it appears that a MMI XII can be generated at the Seabrook site itself. Eve n i n th e abse nc e of that evidence, however, there is no basis for adopting an upper bound of less than MMI XII.

As Dr. Chinnery has discussed in detail, it is not possible :v dem ons tr a te the exis tence of an upper bound, and, in particular, t here is no geological basis for such a showing. (Chinnery, Dire c t te s tim ony , p. 12 -13, Tr. 101, l. 17-22). Indeed, 6 Of course, they avoid a probabi lis tic approach by simply

_/

adopting the largest historical earthquake, although they have no idea what its probability of recurrence may be.

Dr. Jackson agrees with him on the latter point. ( Ja ck so r. ,

birec t tes timony , p. 6 ) . - 7/ Mr. Ilol t attempts to counter with the argument that there is no geologic evidence of large earthquakes in the area, (liolt, Di rec t testimony , p . 4-5),

bu t he is easily and conclusively ref uted by Dr . Chlnnery's explana tion t hat glaciers would have obliterated prior evidence and that the existing thin overburden is such that tracea of an ancient large aarthquake would be very difficult to recognize if they exis t at all. (C h i nner y , Rebuttal testimony, p. 13).

Perhaps the most telling evidence in the record is the opinion of a portion of the scientific community that extends beyond the advocates in this c as e . As reflected in Dr. Chinnery's direc t tes timony , the range of estimates of maximum possible earthquakes in the Cape Ann region shows that of ten experts so lic i ted , five gave "Best Estimates" grea ter than MMI VIII, while three gave "Best Estimates" of MMI X or g re a te r . When the "High Es tima tes " are also considered, eih t of ten are h:Jher than MMI VIII, and five of ten are MMI X <>r gre a ter . Dr. Chinnery admits that he is at the high end, which is not a result of deciding that a MMI XII may occur but of an inability to preculde an MMI XII. Mr . !Io l t , by contras t, is the lowest. (Chinnery, Direct tes timony , p. 12 -13. Tr. p. 69-72, liol t , Tr. p. 388).

As Dr . Chin nery sugge sts, the Only reasonable conservative con-clusion thet can be drawn from this evidence is that the upper bound, if one exists at all, is at leas t MMI X, if not h igher .

7

_/ Dr. Jackson does cite some s tud i es of this issue in areas where surf ace rupture has occurred, but that is not the case in New Eng l a nd .

The Staff, by contrast, never really addresses this issue.

Instead, it contents itself with leaving the impression that Er. Chinnery is unreasonable because he would have all structures built to the larges t possible earthquake , and because he does not leave room for engineering j udgme nt to adop t a lower value.

(Reiter, Direc t tes timony , p. 6 -8 , Jackson Direct tes timony ,

p. 4-5). Again, the Staf f grossly dis torts Dr. Chinnery's r,sition. Par from the Staff's suggestion, Dr. Chinnery offers a methodology on which ennineering j udg men ts can be rationally bas ed , as opposed to a blind acceptance of the maximum historic earthquake in a region where geology provides virtually no usef ul inf o* ma tion . Ultimately, as he must, Dr. Jackson admits that Dr. Chinnary does not propose that all nuclear power plants be built te the upper bound earthquake. (Jackson, Tr. p. 536).

Finally , whatever the upper bound may actually be , there is simply no basis for bel.ieving that it is within the his tor ical record. ' C h i nn er y , Rebu t tal testimony, p. 4, Tr. p. 17-21)

Given the short his tor ic al record, it is extremely unlikely on its face that the largest possible event has occurred within that pe r i o d . It is this obvious fact that apparently leads Dr. Trifunac to conclude that any effort to identi fy an upper bound earthquake results "from a naive expectation that the past seismicity, geology and tec tonics in the area can somehow be employed to compute the largest credible site intensity." (Trifunac, Direct cestimony,

p. 3). That is precisely Dr. Chinnery's point. This may also be the reason that the Staff has f ailed to calculate the upper bound

m earthquake for tec tonic provinces (Reiter, Tr. p. 529, 1. 8-11, 91, is 5-6), although that is the Sta t f 's responsibility in a t tem p ting to determine maximum earthquake potential under App end i x A.

(d) Extrapolation Beyond the liia torical Record Extrapolating beyond the historical record is at onc e the mos t difficult and the mos t important task in attempting to determine the seismic risk at the site. The reason, obviously, is that it is difficult, i f not imposn1ble, to determine the accuracy of es timated return per: ods . At the same time, however, it is essential to make the attempt s i nc e , as the Staff admits, there may well be earthquakes in New England with ret urn per iods longer taan the his tori.c period of 300 years (Reiter, Tr. p. 500,

1. 1 -6 ) , and such earthquakes may not appear in the his torical record. (Reiter, Tr. p. 502, 1. 8-12). If what Dr . Reiter says is true, and there is no reason to believe it is not, if a MMI IX earthquake has a return period of 500 years, or even 1500 or 5000 years, one could well occur during the life time of the Seabrook facility. Yet the NRC Staf f has no method for predicting the return period for earthquakes (Reiter, Tr. p. 501, 1. 9-11),

i and it has not und'rtaken a study of predictive methods other than Dr. Chinnery's to deter.nine whether they are valid or not.

(Reitor, Tr. p. 503, 1. 5-24, ]. 506, 1. 13 - 508, 1. 15). Not surprisingly, therefore, the Staff does .o t know the probability of exceeding the maximum his tor. cal earthquake in a given province.

(Jackson, Tr. p. 540-541). This is particularly importan*. for 4

Seabrook, since the SSE Is based on the implicit premise that the probability of exceeding precisely that earthquake is low

- . . - - - - - - ~ . . . , ,, . - -

-24 .

enough to allow it t o be chosen as the SSE. (Jackson, Direc t testimony, p. 5-6). The Applicant adds nothing to the analysis other than insis ting that a MMI VIII cannot be e xceeded , al-though its own geologic evidence indicates that a MMI XII can occ ur at the site.

We are left, then, in a vacuum, but with empirical observa-tions of linearity and uniform slope, and the inability to esta-blish an upper bound earthquake. These l ead directly to t he p go;;1-tion that we can extrapolate from the his torical record. Although we clearly cannot prove that the extrapolation is cerrect, it app ea rs logical, and, even the applicant admits that no physical law precludes it. (Holt, Tr . p . 434-435). Indeed, Dr. Trifunac is more specific in s ta ting that there is no physics that would require a change at h ig her intensities in what is otherwise a straight line f unc tion . Whil e it may be that we do not yet unders tand the physical law that dictates the s traight line relationship throughout the frequency-intensi ty range , it is always the cas e that the observation is made before the physical law is determined. To require a physic al law at this stage of minimal knowledge oi earthquake science would be to ignore the dic ta tes of the Commission's remand and of Appendix A.

Although we cannot expand the h is tor ic al record to demonstrate the accuracy of the extrapolation , we can work with available data not relied upon in determining the linear function in order to dutermine whether the data are consistent with the es timates .

That is precisely what Dr. Chinnery has done i n p lo t ti ng t h e Cape Ann earthquake (Chinnery , Rebut tal tes timony , Figure 3).

The Applicant argues that the presence of only one MMI VIII in England and Scotland in the his torical record, when l l

the Chinnery methodology would have predicted a higher inten-s ity eve nt in the time period, demonstrates that the extrapola-tion is i nc o rrec t . (Holt, Tr. 443-444). That proposition is 4

the pures t speculation from Mr. Holt, who provided no detailed plots of British earthquakes, did not reconcile foreign and American earthquake scales, did no work himself, and recalled his convenient conclusion vaguely from articles that he chose not to identify. That testimony simply carries no weight.

Both the Applicant and the Staff leap with glee on the fact that rec ent ten-year data in New England do not appear to be consis te nt with Dr. Chinnery's estimated return period from MMI's I-IV. However, Dr. Chinnery explained in his own direct tes timony , and on cross-examination, that the recent i ns trumen t al seismicity can contribute little to the assessment of seismic cisk, precisely because it is rec ent . He further suggested the rec en t data, and even data for the las t century, may well be an anomalous cycle of low seismicity, particularly since it does not reflect the seismicity of the Cape Ann area, which is h is tor i-i cally es t ablished. (Chinnery, Direct testimony, p. 2, 4, Tr. p.

320-321, 326-326, NECNP Ex. 1, p . 89) At this point the only tes timony on the subject is that the instrumental data would not change Chinnery's conclusions in the abse nce of f ur ther in f orma tion .

,. ., - , - , , - - - - - - . - . - . , - - , - - . , ,----,--e- , .e- - - - - - , . -y . . - - .

- - . ~ - . _ . .

Finally, we return to the mos t telling points, which are that the data plots clearly de:ianstrate a recurrence relationship (Chinnery, Tr., p. 197, 1. 1-22), that Dr.

Chinnery's method of extrapolation is widely accepted in the scientific community (Trifunac, Tr., p. 751, 1. 21-25), and that this is the only rational basis for estimating earthquake return perisds and seismic risk that has been presented to the Board la this case. Since there is no physic al law that prohibits the relationship, the record 4

requires that the Board at least recognize it as a valid hypothesis.

(

-2 7-(2) General Criticisms of the Chinnery Methodology In addition to the points discussed above, the Applicant and the Staff leveled a number of other criticisms against l

the Chinnery hypothesis as applied to the estimation of cl.e

)

SSE fcz Seabrook. Each criticism reflects a shallow, and often deceptive, treatment of the underlying facts. Each is invalid.

(a) Dr. Chinnery addresses and admits uncertainties in his methodology more forthrightly and to a far greater degree than either the Applicant or the Staff.

Undoubtedly the greatest attempt on the part of the Staff and the Applicant to warp reality to salt their purposes is in the claim that Dr. Chinnery does not adequately take into account the uncertainties involved in his methodology. To the contrary, recognition of the need to address the uncertainties inherent in the consideration or prediction of seismic risk is at the very core of the Chinnery approach. Far from ignoring uncertainties, he discusses them throughout his work, and he recognizes the need to follow fundamental scientific principles such as Occam's Razor in attempting to draw valid conclusions in light of those recognized uncertainties. By contrast, the Staff and the Applicant utterly fail to address the multitudinous uncertainties involved in their approaches, or to attempt to account for them in reaching their conclusions.

Dr. Chinnery begins with the proposition that none of the available information-seismological, geological, or geophysical-is ad- ite to allow us to make precise decisions or predictions.

n ___ - _ _ _ _ __

He then examinea data sets for three geographic arecs, which produce remarkably consistent results, and notes that these results are consistent with work done in other areas of the country, and with widely accepted hypotheses of linearity and uniform slope.- 8 / -

Based on this analysis, he provides not a single falsely precise figure, but an empirical observation of linearity and a range of slopes from 0.54 to 0.6, of which he believes 0.57 to be the best for purposes of analysis, given the consistency of the data. (Chinnery, Direct testimony, p. 11 Tr. p. 302-303). Indeed, he specifically states that his slope of 0.57 is subject to an error of +3, a point that the Staff ignores. (NECNp, Ex. 1, p. 97). In addition, he demonstrates the consist;ncy of his approach with even wider error bounds, which would still produce nearly the same results. (Chinnery, Direct testimony, p. 14, Figure 4). He als; recognizes the 8 / The Staf f suggests that Dr. Chinnery chose the three areas for study "so that the study would yield certain pre-conceived results." (Staff Brief at 21, n. 9). This charge is nothing short of shocking and grossly irresponsible. It is a personal attack on Dr. Chinnery in an area that is crucial to his reputntion as a scientist, yet the Staff can provide absolutely nu evidence to support its charge of bias. Rather, the Ste.ff has breached the boundaries of acceptable argument by creating a slander from the most tenuous of theories. If anything, Dr. Chinnery is the only witness who had no personal stake in arriving at a particular conclusion. He is not opposed to Feabrook, and he did his analysis before he had any idea that he would testify in this proceeding. There is absolutely no reason to believe that he was doing anything more than attempting to analyze the data in manner that would allow increased understanding of seismic phenomena.

By coatrast, Mr. Holt is exceedingly well paid to produce theories that favor the Applicant, and the Staff's witnesses are bound to defend the Staff's position. Whether or not they have exhibited actual bias, they have the motive to do so, while Dr. Chinnery does not. In light of its public responsibilities, it is incumbent on the Staff to withdraw this thoroughly unfcunded charge and to apologize to Dr. Chinnery.

range of possible upper bound carthquakes, and argues that it munt be assumed that an MMI XII can occur in the area because no significant evidence exists to the contrary. Even so, this does not lead him to rigid adherence to that value as an absolute, but rather to the conclusion that the upper bound is at l ea s t- MMI X, if not higner. (Chinnery, Direct testimony, p. 12-13).

More importantly, Dr. Chinnery clearly explains that the specific numbers that result from his probability analysis cannot be considered accurate in themselves, but represent a range or order of maanitude. This approach, by itselt, indicates the likely uncertainty. (NECNP Ex. 2, p. 769, Chinnery, Tr. p. 92-93).

The significant point is that a probability approach is an inherently more rational method of approaching an area in which significant uncertainties exist. It serves to give some feeling for possible rror, while other approaches give no indication of what the error might be and provide a false impression of precision.

(Chinnery, Rebuttal testimony, p. 15, Tr. p. 120, 1. 1-6, Trifunac, Direct testimony, p. 1, Tr. p. 732, 776).

The contrast between Dr. Chinnery's openness, which is highly unusual in an adjudicatory proceeding, and the lack of any effort on the part of the Staff or the Applicant to reflect, or even admit uncertainties in their approaches is striking.

The Staff claims the authors of Appendix A believed that the likelihood of the maximum historical earthquake is sufficiently low that it can be accepted as the ESE. (Jackson , Direct testimony, p. 5-6). However, the Staff does not know the probability of exceeding the maximum historic earthquake s

-30 dt Seabrook (Jackson, Tr. p. 538-539), and it admits that the probability can vary from provin:c to province. (Jackson, Tr. p. 340-541). Indeed, the Staff has made no effort to determine how best to calculate slope values for particular regions.

Moreover, it has not calculated the parameters of seismic risk for Seabrook, and it has no method for predicting return periods of earthquakes, including those that it chooses as the SSE. (Reiter, Tr. p. 513, 531, 501, Jackson, Tr. p. 514). Finally, the Staff is forced to admit that it follows a mechanistic process leading to results for which it is impossible to determine the ancertainties, (Reiter, Tr. p. 601-603).

, The Applicant is in much the same situation as the Staff.

While !!r. Holt does not even know whether the data from the Charleston earthquake on which he relies are from aftershocks or indecendent events, he is prepared to develop a radical theory basou on those data. (Holt, Tr. p. 170, Direct testimony, p. 2).

In addition, when given an opportuni ty to provide "High", " Low" and '

.est Estimat es" for the largest possible earthquake in the Seabrook area, he designates only one value, and then admits to no error, althc, ugh that value is the lowest of all of the experts participating in the study. (Holt, Tr., p. 387). Unlike Dr. Chinnery's designation of a tiMI XII, which is based on a recognition of uncertainties and his inability to be precise, Mr. Holt's choice reflects the view that a single value can be chosen because there are no uncertainties. Mr. Holt's result simply is not credible. -

J

~__m

. The reality is that only Dr. Chinnery (and Dr. Tri funac as well) addresses and admits uncertainties. The Staff and the Applicant ignore them and attempt to achieve a false precision that is comforting to the mechanics of regulation and to use in adjadicatory argument, but that does not comport with the truth.

(b) Dr. Chinnery fully con side rs geology to the extent pa s sible .

Dr. Chinnery has been criticized for failing to consider geology in his analysis, or to use the Applicant's term, for failing to admit that geology has a place in the analysis. Again, the truth is ignored.

Dr. Chinnery virtually initiates his testimony with geology, and discusses our inability as yet to develop a detailed correlation of seismology and geology in New England, and the lack of geological evidence of an upper bound to earthquakes in New England. He also notes geological evidence of large earthquakes is meaningless in attempting to determine an upper bound. Since we do not understand earthquake mechanisms in New En9 1and, he adds, geology is of little use to our analysis. (Chinnery, Direct testimony, p. 1, Rebuttal testimony, p. 3, 1. 10-13, Tr. p. 27, 101, 107, 300). In addition, he specifically notes, as discussed earlier, that the uniform slope hypothesis appears to hold true regardless of geological variations. (Chinnery, Tr. 197-188, 197). With the exception of Mr. Holt's pluton theory, which would establish an SSE of fiMI XII, neither the Staff nor the Applicant challenges t!.e proposition that we do not undarstand the seismology-geology relationship in New England.

. _Nonetheless, the geology argument hinges on the proposition that geology can provide useful information for this analysis.

Apparently, the Applicant and Staff believe that if a larger I.

earthquake were possible, they would find geologic evidence i

to that effect. But even they must admit that is not the case.

I Although the MMI X earthquake that occurred at Charleston left I

geological evidence, Mr. Holt testified that there is no geologic evidence of prior large earthquakes in the area. That means that before the MMI X earthquake occurred at Charleston, Mr. Holt and the Staff would have argued that it could not occur.

(IIolt , Tr. p. 405-406). Similarly, Dr. Jackson, despite valiant attempts to evade the question, is unable to point to geologic evidence in New England of the MM1 VIII that the Staff adopted as the maximum historic earthquake for the area. (Jackson, Tr. p. 609-619).

l The geology argument is put to rest by the fact that the Staff adopted the maximum historical-earthquake as the.SSE (Peiter, Tr. p. 525) without the use of geological evidence, 1

I and did not increase or otherwise alter the SSE by using geology.

This in absolutely consistent with Dr. Chinnery's explanation that we do not know how the geology of New England relates to earthquake generation. That being the case, the Staff could not use geology to increase the SSE. It is ludicrous to suggest, however, that this lack of knowledge in ene area {

means that valid predictions cannot be made using another method, particularly when that method seems applicable regardless of geology.

1

r (c) Dr. Chinnery's probabilities for the Boston-New Hampshire zone must be assumed to apply to the Seabrook site.

The Applicant and the Staff argue that whatever Dr. Chinnery't probabilities may be, they cannot be used for Seabrook because they apply to the entire Boston-New Hampshire region, not to the site itself. This is incorrect as a matter of law since Appendix A-requires that once the highest intensity earthquake is determined, it must be assumed to occur at the site. It is also basically incorrect as a matter of fact, as Dr. Chinnery has demonstrated.

The process for determining the Safe Shutdown Earthquake under Appendix A is set out at SV(a) (1) (i)-(v) . The first step is to determine the maximum historic earthquake. The second step is to determine whether geological or seismological information indicate that a higher intensity should be chosen than that for the maximum historic earthquake. The third step, if the maximum historic earthquake cannot be l

ccrrelated with a tectonic structure or a different tcetonic p2cvince, is to assume that the earthquake (as increased by geological or seismological data) occurs at the site. The only function of the Chinnery methodology is to help determine

, how much the intensity should be raised under the second i

step. The third step of assuming occurrence at the site

! remaine the same, consistent with the conservative mandate of Appendix A.

L f

l l

f I

l. .

Even if Appendix A did not require assuming the probability at the site, however, it would make little difference. As Dr. Chinnery testified, the occurrence of an MMI X anywhere in the Boston-New Hampshire zone v;ould emise an MMI IX over the entire zone, so that the probability of occurrence of an MMI IX at the site is at least equal to the probability of occurrence of an MMI X anywhere in the area, which is 10-4 (Chinnery, Tr. p. 298-200). Of course, since there is alco an independent probability for the occurrence of an MMI IX itself within the province, the actual probability of occurrence of an MMI IX at Seabrook is greater than 10 -4 In the absence of a sophisticated calculation of MMI IX probability at the site, which would be substantially more complex than the simple division performed by Mr. Holt (Chinnery, Tr. p. 285-286), it is both reasonable and conservative to accept as valid a probability of 10-3, (or slightly less) for the occurrence of an MMI IX at the Seabrook site.

(d) The record does not support the argument that less damage would occur to a nuclear plant sited on bedrock than on alluvium.

Much of the thrust of Mr. Holt's attack on Dr. Chinnery is to the effect that Dr. Chinnery's data derive largely from earthquakes in alluvium, while equivalent intensity earthquakes would cause ?ess damage at Seabrook since the plant is built on bedrock. In fact, Mr. Holt's evidence is extremely sparse and anecdotal, and he fails to provide any information relevant to the siting of a nuclear reactor.

l

-35, Mr. Holt's argument is based hlmost entirely on brief descriptive accounts of damage in different areas as a result of a' Canadian earthquake in the St. Lawrence valley.

His favorite example alleged to make his case is that a grain elevator on alluvium was severely damaged, while the Chateau Frontenac, standing nearby on bedrock, was not affected. '(Hol t , Direct testimony, Appendix A). While Mr.

Holt attempted an the spur of the moment to dismiss the possible effects of focusing which might explain the difference, in fact he has done no careful study of the question, and he was l unable without prompting to remember the relative locations j of the faults ard structures involved. (Holt, Tr. p. 352, 410-412). More importantly, he provided no information with which the Board can judge whether either the Chateau Frontenac 1

or the grain elevator is structurally comparable to a nuclear i

power plant. Since a nuclear power plant is a stiff structure that reacts like a low building, it is likely that the ,

Chateau Frontenac does not have the same resonating frequency as Seabrook would have, and therefore would not be subject

to the same damage. Without that sort of development, Mr.

l Holt's testimony on this point is irrelevant.

The soil-bedrock argument becomes particularly weak l when it is recognized that nuclear plants are subject to the i high frequency end of the specturm. As Dr. Trifunac testified, i

we can expect higher frequencies, and higher accelerations

, associated with.those frequencies, on bedrock than on 4

soil. (Trifunac, Tr. p. 783-784). .That being the case, the evidence indicates that greater damage can be expected to a nuclear power plant on the Seabrook site than on an alluvium site. Therefore, we should be more, rather than less concerned with the hazard. The most reasonable conclusion at this point is that the evidence is simply too weak to reach any useful conclusions about how the character of the site affects the seismic hazard or risk. .

(e) The Staff and the Applicant must undertake a thorough reevaluation of the seismic design of Seabrook based on the return periods estimated by Dr. Chinnery.

Ranging outside the scope of the proceeding as established by the Commission,- /9 ' both the Staf f and the Applicant argue that even if the Chinnery hypothesis is valid, it will have no effect on the seismic design of Seabrook. (Reiter, Direct testimony, p. ll-16, Jackson, Direct testimony, p. 9-12, Knight, Direct testimony, Holt, Direct testimony, p. 5-7). These arguments fail because they have not been adequately developed on this record with sp2cific reference i Seabrook, and because they ignore the principle of defense-in-depth that is central to the c ommission's approach to protecting the public health and safety. However, we see a glimmer of hope in these approaches, at least as discussed by the l

Staff, in that they represent an effort at reasoned scientific analysis, as opposed to blind rellance on the maximum historic earthquake and the scientifically baseless contention that the probabili ty of such an earthquake is low enough to 9 See argument at p. 47-48, below.

/

- 3 7-

) Justify its choi ce as SSE. As suc h , these approaches should be pursued in tull at the conclusion of this phase of the SeabrooA proceeding.

The Staff's position on this point is based on f our propos t-tions :

1. Probabilities of earthquake recurrence are subject to such uncertainties that they cannot be used directly in deter-mining a maximum intensity earthquake on which to base the choice of a re-sponse spectrum anchor point .

2. If the Chinnery hypothesis is to be used, it must be in cor pora ted into a general probacilistic methodology such as that developed in the Site Specific Spectra Pro g r am (SSSP).

3. The SSSP method results in a response spectrum that falls below the Seabrook spectrum as now established.

4. In any event, there are such conservatisms in plant design and cons truction that the plant will be safe even if the design basis earthquake is e xceeded . l 0 /

At this point, these arguments are not sufficient to establish the SSE or the response spectrum for Seabrook, particularly in light of Dr . Tr i f u nac 's demonstration of the high probability of exceeding the existing response spectrum.

We would generally agree wit h point 1, but only if it is ma he in combination with point 2 and in connection with rejection of the mechanistic and scientifically unfounded approach of deriving an anchor point from the maximum historic earthquake. The method that was used to determine the SSE and design response spectrum f or Seabrook was simply to adopt the maximum his toric carthquake and scal e the R.G. 1.60 10/ Points 1 diid 2 are in Reiter, Dir e ct Tastimony, p. 11-15; point 3 is in Reiter, Direct testimony, p. 15-16 and Jackson ,

Dir e ct testimony, p. 9-12; point 4 is Knight, Di r e c t testimony.

Holt's a rgume nt is similar to point 3, but will not be discussed separately, particularly since it completely ails to reflect uncertainties, and was developed by Mr. Holt alone, while the Staff's approach involved canvassing a reasonably broad spectiam of scientific opinion.

l 1

e , j l response spectrum to an anchor poiht based on the intensity of that earthquake. To the extent that is the method relied upon to justify the Seabrook design, it is far more rational to choose the maximum earthquake based upon the probabilities

established by Dr. Chinnery than on the size of the largest earthquake in an extremely short historic record, That is l particularly true in New England, where geology offers no

~

guidance in determining whether a larger earthquake should be chosen. We certainly recognize the uncertainties involved

in moving from the chosen intensity to ground motion, but if the Staff method of doing so is considered acceptable, the Chinnery method is superior, and must be adopted. It would then be up to the Board and the Commission as a matter of

' regulatory policy to establish the acceptable probability level for the earthquake whose intensity governs seismic design. A probability of 10~ would be an absolute maximum given the " incredibility" goal of 10- and the uncertainties inherent in the remainder of the design process. At Seabrook, s

[ therefore, an earthquake of MMI X must be used as the source i

! of the anchor point under the Staff's method of determining i

ground motion.

If point 2 were added to point 1, and the method used by the Staff to establish the SSE for Seabrook were rejected, i

4 not only would we agree, we would strongly urge the Board

, and the Commission to pursue a general probabilistic methodology such as that used in the SSSP. Similar to the Trifunac i

1

-ve.- w --,----w - n.--,,--,,- - - - . , - - - - - , -,e -w - -

~---c ~--, ~ ~ - ~ -

l

= ,

approach, it is attractive in that,it incorporates probabilities to provide some rational basis for the design, and it takes

into account a broad range of scientific opinion rather than relying primarily on the Staff and Applicant. This

. would undoubtedly represent progress; we urge that a -

a

. reanalysis of Seabrook be undertaken using this approach.

The Staff suggests that this analysis has already been done by Lawrence Livermore Laboratories using the SSSP

! information and that the results demonstrate the acceptc.bility of the Seabrook response spectrum as previously determined.

(Reiter, Direct testimony, p. 15-16). While a preliminary reading has apparently been developed, no " report" detailing I how the work was done or allowing either the parties or the Board to examine the validity of the result has been presented.

The Board simply does not have before it information adequate ,

to support the Seabrook design spectrum based on the SSSP, particularly when the use of an MMI X earthquake to determine the anchor point, as would be donc under the current Staff i

methodology, would clearly result in a design response spectrum exceeding the one previously established. In f

addition, since, as discussed below, Dr. Trifunac indicates a relatively high probability of exceeding the current response spectrum, the seemingly contradictory SSSP results

! must be examined very carefully to determine whether conservatism

)

i- and uncertainties are adequately considered.

I-i A

1 i

, - - - , , . , . . . , , - , _ - - . , - - . _ ~ - , . _ _ - , ~ - - . - . - . . . .. .. , _ _ . - . . .- -- , . - - - - - - .m.- - --.

-40 ..

j - .

1

Finally, the Staff makes mugh of the substantial conservatism inherent in the design of Seabrook and argues that the plant i will be safe even if the design basis is exceeded. (Knight, Direct testimony). This ignores the entire concept of ,

defense-in-depth. It may well be true that nuclear plants are overdesigned such that in all cases they can withstand earthquakes greater than the design basis. If that is the case, however, the reason is that the uncertainties of

, seismology and of nuclear technology demand caution. The i

governing principle of defense-in-depth is that these conservatisms exist once the design basis is determined. However, they may not be considered in determining the design basis itself.

Stated otherwise, whether the design basis earthquake is a MMI IV or a MMI X, the conservatisms identified by Mr.

Knight are assumed in the design of the plant. (Knight, Tr.

p. 687, 1.25-688, 1. 3). Only after the design basis is

determined dc these conservatisms come into play. In this i

j case, they must be ignored until the design basis intensity i or response spectrum is arrived at based on seismic information.

I IV. The Staff's Methodology For Correlating Vibratory Ground Motion With the SSE Is Scientifically Invalid and Has Resulted In A Design Response Spectrum With ,

A High Probability Of Being Exceeded.  :

1

! The second issue .cmanded to the Appeal Board for consideration is the question of whether the Staff's methodology l

for correlating vibratory ground motion is consistent with Appendix A, with particular reference to "the relation i

l t

between the mean of the maximum ground accelerations and maximum effective ground acceleration." 12 NRC at 298. The record establishes that the Staff's methodology is scientifically invalid, and therefore inconsistent with Appendix A, and that there is an unacceptably high probability of exceeding the response spectrum in the frequency range of interest during the life of the plant.

The Staff's methodology has been" adequately described in its Brief (Staff Brief at 29-31) , and consists of the following:

1. Choice of the SSE,

2. Using the trend of the means of Trifunac and Brady (1975) to provide a peak acceleration anchor point for the design response spectrum,

3. Use of the R.G. 1.60 response spectrum to determine the maximum level of response to l ground motion that is assumed for the purpose of the design.

The issue of " maximum effective ground acceleration" appears to have arisen with respect to the second step as a result of the 1

tact that SVI(a) (1) of Appendix A states that the vibratory ground motion of the SSE shall be defined by response spectra corresponding to the maximum vibratory accelerations at the elevations of the foundations of the nuclear power plant structures.

(emphasis supplied). The Staff approaches this requirement by arguing that the highest possible accelerations have no effect for variona reasons, and that the trend of the means of Trifunac and-Brady provides a reasonable approximation of

-n._ v w -- - - - - --

1 the maximum effective acceleration; or what the Staff terms the maximum acceleration "of engineering significance."

(Reiter, Direct testimony, p. 20-22). The Staff also attempts a comparison of its own R.G. 1.60 spectrum anchored at 0.25g with a purportedly directly estimated mean plus one sigma  ;

response spectrum. (Reiter, Direct testimony, p. 23-25).

Dr. Trifunac's testimony establishes that the Staff's methodology is invalid and that significantly more analysis is required to determine the acceptability of the Seabrook design response spectrum. Dr. Trifunac's difficulties with the Staff's methodology-begin with the fact that its mechanistic step-by-step process yields a falsely precise result, when our knowledge of the ,

area does not allow such precision. (Trifunac, Direct testimony,

p. 1). This leads Dr. Trifunac.to conclude that the Staff's approach should be rejected altogether in favor of a probabilistic analysis that takes uncertainties into account. (Trifunac, Direct testimony, p. 3-5).

However, recognizing that the Staff approach may have some utility if R.G. 1.60 is scaled correctly, Dr. Trifunac addressed that methodology as well. Referring to the choice of l anchor point, Dr. Trifunac explained that the spectra being anchored were developed based cn the use of actual peak accelerations, so that the use of some lower value is " totally inacceptable." (Trif unac , Tr. p. 733-737). He added that

-neither the concept of "9ffective peak acceleration," nor the Staf f's formulation of " maximum acceleration of engineering s

-43 ,

signif,icance" permitted the use of a lower anchor point because the terms are essentially meaningless, avoid the physical basis of the problem, and allow unwarranted freedom for expert judgment. (Trifunac, Direct testimony, p. 4).JJ/

Indeed, Dr. Trifunac rejected the use of the trend of the means of the data in his own 1975 study as a basis for determining the anchor point. (Trifunac, Tr. p. 797-798).

In addition to challenging the choice of anchor point, Dr. Trifunac strongly criticized the use of R.G. 1.60 itself because it does not have the correct shape. Rather than representing the correct average plus one standard deviation, this spectrum is normalized and is much closer to an average acceleration at the short period end, which is the end of the spectrum of interest for nuclear power plants. (Trifunac, Direct testimony, p. 4-5, Tr. 740-741). The result is that the spectrum is less conservative in the area relevant to reactor design, and that lack of conservatism is componded by improper scaling to a value below the actual peak acceleration on which the R.G. 1.60 spectrum was based.

The staff argues in support of its use of the trend of the means that the absolute maximum actual accelerations clearly do not affect the facility and should not be considered.

Ilowever, even that argument is weakened by the fact that the Staff did not have any sound theoretical basis for its 11/ Remarkably, the Staff agrees that the " effective acceleration" concepts are essentially meaningless. (Knight, Tr. p. 670, Jackson, Tr. 694).

approach and simply used the Trifunac and Brady data because they were the only data available at the time. (Reiter, Tr.

p. 647, 1. 12-15). However, even if we assume that the Staff is correct and the actual peaks should not be used, the conclusion we must reach is not that some lower. number 4

chosen virtually at random should be employed, but that the attempt to scale R.G. 1.60 should not be allowed if it t

cannot be done consistent with the basis on which the spectrum I

was orginally developed.

! Instead, as Dr. Trifunae argues, we must deal directly i with the distribution of ground motion that is involved in

) seismic risk. Remarkably, Dr. Reiter agrees on his crucial

point. (Reiter, Tr. p. 693-694). When Dr. Trifunac performed a direct estimation of the design spectrum using a probability distribution, he found that there is a 5% probability of exceeding the Seabrook response spectrum in the frequency range of interest during the life of the plant if the maximum

! possible earthquake in the area is MMI XII. If the maximum possible earthquake in the area is MMI X or MMI VIII, the probability of exceedence is less than a few percent.

(Trifunac, Direct testimony, p. 5-9, Pigure 3, Tr. p. (753-

. 755). In fact, when Dr. Trifunac incorporates pessimistic assumptions into his analysis, as he believes is necessary to account for uncertainties and gaps in our knowledge, the result is that the Seabrook spectrum has a 5% chance of

}' ,

Deing exceeded if the maximum possible earthquake is MMI J

t

_, . __.._.,.~.....m,-_. ,__ m,,,,,... _ . , , , . . _ , , _ , _ _ _ _ . . , _ . . _ _ _ _ _ _ . _ . . . .

-45 ,

V111, and substantially higher probability if the maximum is

! higher. (Trifunac, Direct testimony, fieure 4). These results lead Dr. Trifunac to conclude that the Seabrook design spectrum "may" be acceptable, but that he would have to do more detailed calculations to determine whether his j assumptions are sufficiently pessimistic. (Trifunac, Direct testimony, p. 10).

The fundamental conclusion from this record is that the Staff methodology is invalid and that further analysis is required to determine an acceptable response spectrum for Seabrook. It does appear that the Staff's R.G. 1.60 spectrum anchored to 0.25g may be in the correct neighborhood, but given the scientifically unsound approach to its development, this seems to be little more than an accident. Nuclear plants are not designed by accident and must not be licensed without a sound ccientific and rational basis for judgement that the public health and safety vill be adequately protected.

! Dr. Trifunac provides a method for making that judgment.

l His own conclusion, based on what he admits to be cursoty analysis Liiat may not be adequately pessimistic, is that there is approximately a 5% chance of exceeding the Seabrook design response spectrum during the life of the plant, which I

translates into a probability of 5 x 10 -2 of an earthquake exceeding ~the design basis of the plant.12,/ We do not know i 12/ Even the probabilities assuming a gaximum earthquake of -

i MMI X or VIII are approximately 1 x 10 , a significant probability in the context of determining the design basis of a nuclear power plant.

i l

l l

the basis for Dr. Trifunac's suggestions that this initial analysis indicates tnat 5tabrook's response spectrum may be acceptable, and he provides none. Certainly that is inconsistent with the concept of defense-in-depth, which requires a substantially lower probability in order to achieve an overall likelihood of no greater than 10 -7 of exceeding the design basis of the plant. In any case, Dr. Trifunac believes that further analysis is necessary to determine whether his figures are correct. We certainly agree. At this point, the record requires the conclusion that the present design-spectrum is based on a methodology that has no scientific basis and that Dr. Trifunac's approach may provide an acceptable means of judging its adequacy. However, Dr. Trifunac's analysis must be carried out in full in order to reach a valid conclusion. In addition, the Board.and the Commission must address the question of the acceptable level of probability that the response spectrum will be exceeded. Since it constitutes the design basis of the plant, NECNP asserts that the probability must be no greater than 10-7 of an accident, and substantial additional complex analysis is required to determine the point at which that probability is

! reached.

V. The Pending Motions to Strike Should Be Granted When the Applicant and Staff testimony were offered f

during the hearing, NECNP objected to various portions on three separate grounds. In order to expedite the hearing, however, NECNP I

, -. . . , . , , . , . . , - _ , . , _ . _ . . - _ . , _ . - . . . . . -. _ ,, . _ - , , _ , _ . , . . - .-- - .. ,. . m,, . . . - . _ - - - .-- -_

-47, agreed that the Board should withhold decision pending post-hearing briefs. NECNP now renews those objections and motions to strike.

A. Testimony Discussing the Impact of the Chinnery Hypotheses Is Beyond the Scope of this Proceeding.

The first area in which NECNP objected to Staff testimony is their efforts to address the impact of the Chinnery methodology, if it is assumed to be valid. The objection and motion to strike applies to Question and Answer 9 at page 11 of Dr. Reiter's Direct ,

testimony and to Question and Answer 5 at page 9 of Dr. Jackson's Direct testimony. (The discussion of the objections begins at Tr. p. 467.)

The essence of the objection is that this testimony is beyond the narrow scope of the proceeding as remanded by the Commission and interpreted by the Appeal Board. There are but t.io narrow questions here, (1) the validity of the Chinnery hypothesis, and (2) tro consistency of Staff's methodology of correlating vibratory ground motion. G.+ e the validity of the Chinnery hypothesis is established, the fi sc question is resolved, and the second question clearly does not extend to this testinony.

NECNP originally believed that the scope of the proceeding i extended beyond-the validity of the Chinnery hypothesis, at least to the extent that it encompassed the question of whether

! the Chinnery hypothesis was superior to methods used by the Staff or the Applicant. However, the Staff itself successfully refuted this notion in its arguments to the Board on NECNP's discovery request. It is truly remarkable that the Staff

_ - - - - - . . . , . ..-_ _ .- _ . ,.._,, _ _ . _ _ . ~ , _ _ _ . . . . _ _ - - _ . . . _ . . . _ , ~ . - _ . _ _

shoukd at once restrict the scope of the hearing as it applies to NECNP and then expand the scope for its own purposes. ,

It is doubly remarkable that the Staff should then question whether fundamental fairness requires the exclusion of its testimony. If that question is to be reached, it must be judged from the point of view of NECNP, which is the affected party. NECNP finds itself in the position of '

having asserted that the comparative merits of var.ious approaches to determining the SSE are relevant to judging the validity of the Chinnery hypothesis. The Board may have viewed the Commission's remand more narroely, but at least NECNP's argument was related to attempting --; make a judgment on the validity of the Chinnery hypcthesis. The Staff's testimony objected to here is in no way related to attempting to answer that question. Rather, it assumes the answer and adds to the hearing an additional issue which the Commission did not discuss and of which NECNP had no previous notice. To allow the testimony to stand would constitute a fundamental violation of the NECNP's procedural rights.

B. Appendix A Requires that Earthquake Probabilities for the Province be Assumed at the Site and Prohibits Attempts to Reduce those Probabilities.

NECNP objected to portions of Mr. Holt's testimony (pages 1 and 4 of Direct testimony, see Tr. p. 333 for discussion) and Dr. Reiter's testimony (Question and Answer 2 of Rebuttal testimony, see Tr. p. 488 for discussion, and Tr. p. 518, 1. 7-12) on the ground that they represent attempts to avoid the require-

ments of Appendix A by determining,carthquake probabilities for particular sites rather than for the tectonic province as a whole. Those portions of their testimony must be stricken for that reason.

As we have previously discussed, p . 3 3 -34, Dr. Chinnery's hypothesis provides a means for making a rational choice of the maximum earthquake intensity as required by SV(a) (1) (1) of I Appendix A. In effect, it provides a' scientifically valid basis for deciding whether the intensity of the maximum historic earthquake or some larger intensity should be chosen for the purpose of' analysis under that section. Whichever method is used, however, subsection (ii) requires that where the earth ~

t quakes cannot be correlated with particular structures within the province, they must be assumed to occur at the site. That requirement applies to the maximum historic earthquake and to any larger earthquake that may be chosen-based on geological or seismological data.

This requirement reflects and is ecsential to the conservatism

! of Appendix A. In adopting the regulation, the Commission i

decided that our knowledge of earthquake science is such that unless we can specifically relate a likely earthquake to a

) particular structure, we must assume that it will occur at the site. Just as the unknown probability of the maximum historic i

earthquake was assumed at the site, so identified probabilities of larger. earthquakes must be assumed there as well.

,, JC . Mr.: Knight's Testimony Concerning Conservatisms Inherent In the Design Basis is Irrelevant To Determining What The Design Basis Should Be.

NECNP's final objection is to the testimony of Mr. Knight, ,

which relates almost entirely to the extent of conservatism in the design of the plant, which is the same as the inherent in any nuclear power plant. Again, this testimony is beyond the' scope of the proceeding. The question here is not what consc Jatisms should be adopted once the design basis is established, but how to establish the design basis .itself.

Any conservatisms that are inherent in the desi 3 a are assumed as part of the concept of defense-in depth and may not be considered in deciding what the basis of the design should be.

The Staff virtually admits that Mr. Knight's testimony is beyond the scope of the hearing when it indicates that his testimony was filed because the Staff did not know whether Dr. Chinnery or Dr. Trifunac would challenge the adequancy or conservatism of the design itself. (Staf f Brief at 44) of course, the Staff had no reason to believe that either witness wor.13 challenge the conservatism of the design since that issue is not within the scope of the proceeding. Neither the Commission nor the Board mentioned anything like it in their prior. rulings, and again, NECNP would be severely prejudiced by having this testimony admitted without adequate prior notice.

CONCLUSION For the reasons stated above, NECMP urges the Appeal Board to find that Dr. Chinnery's probabilistic hypothesis is f actual.ly valid, that the Staff's method of correlating vibratory ground motion with the SSE is scientifically invalid and, therefore, inconsistent with Appendix A, and that NECNP's motions to strike should be granted. These findings will then require a thorough' reanalysis of the Scabrook seismic design basis, taking into account both Dr. Chinnery's hypothesis and Dr. Trifunac's methodology tor estimating the probability of exceeding the established re3ponse spectrum.

Respectfully submitted,

- /

William S. Jordan,III L

'ynne Bernabei -

Harmon & Weiss -'

3725 I Street, N.W.

Suite 506 Washington, D.C. 20006 (202) 833-9070 DATED: July 15, 1981

UNITED STATES OF AMI:ltICA

- .o a NUCLEAR REGULATOitY COMMISSION BEi' ORE Tile ATOMIC SAFETY AND LICENSING APPfC L BOARD

)

In the Matter of )

)

PUBLIC 3ERVICE COMPANY OF ) Docket Nos. 50-443 NEW 4 A M P S II I R E , et al. ) 50-444

)

(Seabrook Station, Units 1 )

and 2) )

~

)

CERTIFICATE OF SERVICE I hereby certify that a copy of "N ew Eng l a nd Coalition on Nuclear Pollu tion Pro posed Fac tual and Leg al Findings on Remanded Seismic issues and Supportina Argument " was mailed first class pos tage-prepaid this 15th o., of July, 1981 to the following:

Alan S. Rosenthal, Chairman Dr. John H. Buck Atomic Safety & Licensing Atomic Gafety & Licensing Appeal Board Appeal Board U.S. Nuclear Regulatory U.S. Nuclear Regulatory Commission Commission Washington, D.C. 20555 Washington, D.C. 20555 Frank Wright, Esquire Assistant Attorney General Assistant Attorney General Environmental Protection Division Environmental Protection Office of the Attorney General Division State House Annex, Room 208 Office of the Attorney General Concord, New Hampshire 03301 One Ashburton Place Boston, Massachusetts 02108 Thomas G. Dignan, Jr., Esquire Ropes & Gray Robert A. Backus, Esquir+ 225 Franklin Street O'Neill, Backus, Spielman , & Little Boston, Massachusetts 02210 116 Lowell Street -

Manchester, New Ilampshire 03101 Docketing and Service Section U.S. Nuclear Regulatory Commission Roy Lessy, Esquire Washington, D.C. 20555 Office of Executive Legal Di. rector U.S. Olclear Regulatory Commission Washington, D.C. 20555

= -

i i + . .

• i Dr. W. Reed Johnson D. Pierre G. Cameron, Jr., Eng Atomic Safety & Licensing -

General Counsel Appeal Board Public Service Company of U.S. Nuclear Regulatory Commission New Ilampshire

, Washington, D.C. 20555 1000 Elm Street Manchester, Nil 03105 Ms. Elizabeth H. Weinhold 3 Godfrey' Avenue Atomic Safety & Licensing l Hampton, New Hampshire 03842 Board Panel U.S. Nuclear Itegulatory Commission -

~~

Washington, D.C. 20555-l 4

)

. ' 7-;. . _,

j' William -S'.' Jordan , III o

4 i

j 1 8 t

t i

i e

i i

s I