ML20206J246

Paper Entitled, Rational Approach to Emergency Planning Zone Definition, Presented at 860922-23 ANS First Regional Conference in Pittsburgh,Pa.Related Documents Encl

ML20206J246
Person / Time
Site:	Seabrook
Issue date:	09/22/1986
From:	Hazzan M, Warman E STONE & WEBSTER ENGINEERING CORP.
To:
Shared Package
ML20205K769	List: ML20205K763 ML20206J224 ML20206J246 ML20206J276 ML20206J291 ML20206J329 ML20206J395 ML20206J464 ML20206J485
References
FOIA-87-7 TP-86-86, NUDOCS 8704160052
Download: ML20206J246 (43)

Text

'

- - ,i. Q iv i g ; . 4- j p-.I.j , 7 g Qeger.y.v~  ; m A RATIONAL APPROACH

.s TO EMERGENCY PLANNING .

ZONE DEFINITION .

by M.J. HAZZAN E.A. WARMAN STONE & WEBSTER -

l' ENGINEERING CORPORATION

[ '.

1

.. ')q -

~. d i$

Presented at  ?!

AMERICAN NUCLEAR SOCIETY .f.

FIRST REGIONAL CONFERENCE ,2 m

l Pitt'sburgh, Pennsylvania l September 22-23,1986 j;j m

4 A d

< fo:rg 7 :

. E/9 ':

' ~

TP 86-86 ,.

NX/EMR g W%

%t.k W '

V , hi.A -

870A160052 870409 STONE & WEBSTER ENGINEERING CORPORATION sfictthe77 b PDR CHERRY HILL OPERATIONS CENTER CHERRY HILL, NEW JERSEY

A RATIONAL APPROACH TO EMERGENCY PLANNING ZONE DEFINITION Michael J. Nazzan Edward A. Warman Stone & Webster Engineering Corporation Cherry Hill Operations Center Cherry Hill, New Jersey 08034

. (609) 482-3000 INTRODUCTION AND

SUMMARY

As part of the IDCOR-85 program, Stone & large number of source terms; and 3) does not Webster Engineering Corporation conducted a readily provide visibility of the importance of ,

reassessment 1 of emergency planning requirements accident sequences wt.ich dominate exposure risk.

in list.t of progress in the estimation of nuclear Additional discussion of the NUREG-0396 approach power plant severe core damage accident source is included in the following Background terms. (IDCOR is the Industry Degraded Core discussion.

Rulesaking Program sponsored by the nuclear TAstt :

industry and operated under the auspices of the Atomic Industrial Forum, Inc.) Source terms ExAnetts or sotact rtans define the timing, composition, and magnitude of severe core casas Accident source Terms releases of radioactive material to the environ- ,

ment. The reassessment, which concentrated on the technical bases for the size of the plume t,rac w wto core

,i,,,ed wn%

Environment exposure pathway Emergency Planning Zone (EPZ), ass an t.nos locon was undertaken to determine what changes in the Put-2 surrv zien EPZ a;equirements, if any, are appropriate as a 'C'*"*'Y " ^- Tyswe result of the vast increase in the knowledge and Time selease, hr 2.5 2.5 12 understanding of accident source terms for both xenon-Krypton o.90 o.as 1.00 the Pressurized Water Reactors (PWRs) and Boiling  !* din'* Bro *ia' o fo o of o 0017 Water 3.eactors (BWRs)s-s since the existing C" '""-n un i d i um o.so o.oss o. coir l . Tellurium-Antimony 0.30 o.o55 0.00002 requirements were established, sarium-strontium o.06 o.oi .o.0000:

7t.e exi, sting requirements?-s were based on Ruthenaus o.02 0.001) so ooool the results of a United States government inter. Lanthanum o.co2 a.oooir -

agency task force study, as reported in a docu-i

' ment stabered NUREG-0396.8 Although a spectrum Based on this and other observations dis-of accidents' was included in that study, the cussed later, an approach was developed and offsite risks were dominated by the severe core proposed1 as an alternative to NUREG-0396. The damage accident release categories reported in alternative approach is embodied in an exposure the Reactor Safety Study, also referred to as risk guideline as exprsssed by criteria for

. WASH-It00.lo Present IDCOR2 and NRC3 source prompt and latent exposure risks for any individ-terms a re substantially lower than those reported ual in the general public.

in WASbl400, with few exceptions, as discussed The exposure risk guideline, as discossed in

! 1ater in this paper. Table 1 presents some this paper under the heading Proposed Approach, example s of changes in source terms. Within the is based on the premise that emergency planrdng context of this paper, risk is defined as the is not necessary beyond a specified distance from product of the radiological conse'quences (e.g., the poin,t of release if the probability of,either dose)t.imestheprobabilityo(occurrenceofthe a prompt or latent fatality for an individual at conseqtences, per reactor year'. or beyond that distance is extremely low, even if In light of substantial changes in accident protective actions, such as evacuation, were not

, source terms, the underlying bases for the taken for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> following a postulated acci-l existir.3 EPZ requirements, as documented in dent. (Note: Protective actions not requiring NUREG-0396,' were reviewed. Based on this review Preplanning would be expected to be taken on an it w.; determined that the NUREG-0396 approach: ad hoc basis beyond the radius of the plume

1) is somewhat subjective and very difficult to exposure pathway EPZ.if protective action guides interpret quantitatively; 2) is ur. wieldy in its were exceeded in an actual emergency. Protective applica tion, and would be especis.11y so with a Action Guides (PAGs) have been established by the I

.,. . . .s..,.. A

for any such accidsnts. Beyond this

' Environmental Protection Agency in terms of generic distance it was concluded

• dosage units, e.g., I rem (lower level) and 5 rem The that actions could be taken on an ac j

• (upper level) for whole body exposure.) hoc basis using the same considera-guideline is expressed in terms of criteria which tioni that went into the initial are based on less than one chance in one million action determinations."

per reactor year of incurring either a prompt or The calculated consequences of Design Basis latent fatality. If the calculated risks exceed Accidents (DBAs) were considered for most of the the criteria at a given distance, then emergency then-licensed nuclear power plants. Tne source planning is recommended within that distance from terms for the DBA analyses are based on Regulato-the point of release. ry Guides 1.3 and 1.4, which assume that 100 per-Application of the proposed guideline with cent of the noble gas inventory of the core and past and present source terms is discussed later 25 percent of the iodine inventory are immedi-in this paper. stely available for release from the containment.

The principal conclusion of the reassessment The composition of the 25 percent of the iodine is that the utilization of the.new, more realis- available for release is defined as 91 percent tic source terms cad result in the reduction of elemebtal, 4 percent methyl or inorganic, and

- the plume exposure pathway EPZ from the current 5 percent particulate. The leakage from the radius of approximately 10 miles to a radius of containment is assumed to occur at design basis approximately 2 miles. Typically, this trans- leak rates and the effects of engineered safety lates into a reduction in the numbers of persons systems, such as filters, are included in the DBA included in the emergency planning from tens of analyses. Figure I-8 of Appendix I of NUREG-thousands to fewer than one thousand, and much 0396, which has been redrawn on a linear scale Thus, less than that for a number of sites. and is presented herein as Figure 1, is one of emergency planning resources can be much more the principal figures in NUREG-0396 with respect focused. to suassarizing the results of the review of DBA The reassessment of EPZ requirements summa- analyses.

rized heuein illustrates the importance of low probability, high-risk accident sequences, and 500 conversely the lesser importance of less severe It mamovewommesmsom accidents with respect to emergency planning.

also illustrates that concentration on a rela- 480 -

tively few higher risk sequences results in quantification of a very large fraction of the total exposure risk for any individual in the 400 -

general public. This observation is important both in appraising new source term information and developing realistic emergency planning 350 - 107. CURVE. HIGHist 107. OF requirements. 67 Sfft CALCULATIONS g

3 aco -

BACKGROUND Although the existin's requirement for a g plume exposure pathway EPZ radius of approximate- g 250 ly 10 miles is based on an analysis of a spectrum e of accidents, as reported in NUREG-0396, acci- "

• dents at the higher consequence portion of the 300 . p SM Cmt. MAN W 67 ACNAL W CALCMAtlONS spectrum can be observed to be dominant with regard to emergency planning considerations.

In addressing potential accidents which y, y

should be considered in developing emergency response requirements, NUREG-0396 states (Page

• ~

The Task Force agreed that emergency response plans should be useful for responding to any accident that would produce offsite doses in excess of the PAGs. This would include the more , , , , ,

severe design basis accidents and the 2 4 a e to 12 i o

accident spectrum analyzed in the gMSTANCE. Nults RSS. After reviewing consequences associated with these types of FIGURE 1 accidents, it was the consensus of NUREG.0396 CALCULATION the Task Force that emergency plans 2 HR WHOtt 500Y DOSE FOR UCEN$1NGCALCULATION OF could be based upon a generic dis- D8A/LOCA AT 2 HOUR $ ASSUMING 5 PERCENTILE METEOROL tance out to which predetermined AND STRAIGHT UNE PLUME TR AJECTORY actions would provide dose savings 2

Figure I-11 of Appendix I to NUREG-0396 damage accidents using WASH-1400 source terms.

. (reproduced here as Figure 2) was a primary This is also true for many of the more recent figure in the development of the rationale for source terms as well, although the higher dose recommending a planning basis incorporating a curves drop off substantially with distance.

radius of about 10 miles for the plume exposure In expressing caution in the use of PAG pathway EPZ. The data in that figure were based doses in establishing requirements for emergency on then-current, best available source terms from planning, NUREG-0396 stated (Page 4):

' the Reactor Safety Study (WASH-1400).10 The data "The nature of PAGs is such that they were presented in the form of a graph of the cannot be used to assure that a given conditional probability of exceeding whole body exposure to individuals in the doses as a function of distance using the WASH- population te prevented. In any 1400 release categories as inputs. The data wer particular response situation, a conditional on a core melt frequency of 5 x 10** range of doses may be experienced, per reactor year. principally depending on the distance from the point of release. Some of these doses may be well in excess of e

the PAG 1evels and clearly warrant the initiation of any feasible protective actions. This does n_ot I

=

mean, however, that doses above M PAG 1evels can be prevented or that .

l] emeraency response plans should have g ,, ',

,,,, g their objective preventina doses g above M levels." (Emphasis adoed)

= -

" Nonetheless, the reference to the PAG doses in the discussion of the dose-distance relation-ll se esa ships for severe core damage accidents is one of 3

8 -

the difficult aspects of attempting to use the F=

• NUREG-0396 approach in the reassessment of

$3l ** g-: emergency planning requirements with new source i

term information. As discussed in the introduc-l  : tion, a number of other difficulties in the application of the NUREG-0396 approach were

, encountered in reviewing that document for soo sem-- potential reassessment of EPZ requirements with aam - i- 1 - - ..I . T new source terms. It was determined that an alternative approach to NUREG-0396 was in order.

The approach which was developed is summarized in FIGURE 2 this paper and is discussed in detail in IDCOR CONDITIONAL PROSAeluTV OF EXCEEDING WHOLE SCOY DOsg Technical Report 85.4 (5.3).1 Before proceeding vensus otsTANCE. pnosA88UTIES ARE CONDITIONAL ON A CONE to discuss the proposed alternative approach, two MELT ACC10ENT(5 : 10 8). additional background issues are briefly discussed.

The first issue relates to the relative With regard to Figure I-11, NUREG-0396 contribution to exposure risk from a spectrum of coted: releases. Figures 3 and 4, respectively, depict "Given a core melt accident, there is the relative exposure risks for the nine PVR and about a 70 percent chance of exceed- five BWR release categories reported in WASH-ing the PAG doses at 2 miles from a 1400. These data illustrate that approximately power plant, a 40 percent chance at 99 percent of the PWR whole body exposure risk is 5 miles, and a 30 percent chance at represented by the PWR-1, -2, and -3 release - -

10 miles. That is, the probability categories. The same is true for BWRs, where the of exceeding PAG doses at 10 miles,is BWR-1, -2, and -3 release categories contribute .

1.5 m 10-s per reactor year (one ' ' approximately 99 percent of the wh:ile body chance in 50,000 per reactor year)' exposure risk. From an exposure risk perspec-from the Reactor Safety Study analy- tive, these Yelatively low probability, high sis." It also noted, in referece to potential consequence releases dominate,the r.isk, the 200-ree curve depicted in the and the contributions to risk from the much more figure, "As can be seen from Fig- probable, low potential consequence releases have ute I-11, core melt accidents can be little risk significance. Since the objective of severe, but the probability of large emergency planning is dose savings, studies in doses drops off substantially at support of emergency planning requirements can about 10 miles from the reactor." and should concentrate on the high exposure risk As can be observed from Figure 2, the 1- and releases, rather than on the more probable, low 5-res PAG dose curves do not drop off substan- exposure risk events. Such concentration is tially within tens of miles for the severe core reflected in the approach proposed herein.

3 stone & WeOSTER

to Furthermore, it was observed in the present ews.: study that a few severe core damage accident sequences contribute essentially all of the exposure risk within the high release categories.

- Thus, concentration on such sequences is appro-

priate in assessing emergency planning require-ei g ments. Application of the exposure risk criteria discussed in this paper is predicated on this observation.

The second additional background issue deals withthemethodologyusedinthegresentanaly-l sei sis. The CRAC2 computer program 1 was utilized 5 in conjunction with a representative set of l3 " -

• meteorological data. Figure 5 depicts the mean acute whole body dose as a function of distance w ^ = a. : with-a baseline source term (discussed below) for

• four sets of annual meteorological regional data

[ ,, "**'

from Miami, New York, Chicago, and Phoenix,

- g respectively. These data were obtained from

• wea Sandia National Laboratories ** and were utilized in the Sandia Siting Study.15 The variation of

! the mean doses is observed to be within a factor I

aseen of two for all four regional weather conditions.

Therefore, the Miami data were used as represen-tative data in the present study, and offsite doses were calculated with the representative meteorology with numerous past (WASH-1400) and

,,,,, M present (IDCOR and NRC) source terms.

REL8 Alt CAT 5cotY g FIGURE 3 RELATIVE CONTRIBUTION TO WHOLE SODY EXPOSURE RISK I FROM WASH.1400 PWR RELEASE CATEGORIES E 3

o set to o s

own.:

h m

= ......

3 ='. . . . . ..

E Ot5f ANCE { Mil l 3 e meteosotoor i

ciecaco e tweenna e 2 ansame e saw toes .

0.01 FIGURE 5 u

i . MEAN ACUTE WHOLE SODY DOSE. SASELINE (NORM AL ACTIVITY) 3 The offsite dose calculations were performed 5 e w s.4 i

E by: 1) running the CRAC2 code with a baseline I source tern comprised of release f ractions of 1.0 0.00, for the noble gases and 0.01 of each of the other fission product groups in a manner such that the

' mean acute whole body dose contributions for each of the fission product groups were calculated l separately by CRAC2; and 2) multiplying the resultant offsite dose contributions from each swa.5 fission product group by the release fraction for

o. coot J

that group in each of the source terms under investigation. Table 2 illustrates an example RELEASE CATE00EY calculation. (Note: The effects between the FIGURE a 2-hour release time for the paseline analysis and RELATIVE CONTRIBUTION TO WHOLE SODY EXPOSURE RISK the 2.5-hour release time for the specifled FROM WASH 1400 gwR RELEASE CATEGORIES source term are also reflected in the results.)

4

the National Academy of Sciences a prompt fatality for an individu-Committee on the Biological Effects of al exposed to a dose equal to or Ionizing Radiation (frequently referred greater than 500 rem. (Note: All to as the BIER-III report).18 The doses in excess of 500 rem are thus assumed to result in prompt table presents the results of a rela- fatality for emergency planning tive risk projection model calculation which indicates that a single exposure purposes.)

of 10 rads (essentially 10 res) is -- In the dose range of 200 to expected to result in an increased 500 ree, a line segment is estab-mortality rate of 2,255 per 1,000,000 exposed persons. Using this informa- lished to reflect the dose-tion, the increased mortality risk for response function in this I

the lifetime of an exposed individual

• potentially life-threatening exposure region.

. is: ,,

= 2.255 x 10" - Figure 7 is a reproduction of Figure VI 9-1

- 1'000 000 x 10 per rem from-WASH-1400.2 It depicts the probability of prompt mortality (within 60 days) as a function

-^

With the above-stated premise that of acute whole body exposure with minimal, emergency planning should be undertaken supportive, and heroic medical treatment, respec-l at a given distance if the mortality tively. The minimal treatment curve from this risk exceeds 1.0 x 10-s per reactor figure was considered in developing the shape of I the line segment depicted in the 150- to 500-rem year, the allowable frequency of an event that would result in a mean dose dose region of Figure B-1.

of I rem at a given distance per reactor year is: ,,,,, , . , , , . , . , . ,

sw memwm 1.0 x 10-6 = 4.43 x 10,3 per MINIMAL ==w 2,255 x 10-, reactor year 99,9 -TREATMENT An exposure risk of 4.43 x 10-8 res per reactor year would result in an in- M ~

~

creased lifetime risk of latent mortal- '8 ~

ity of 1.0 x 10-s yr ~1 A constant ,

exposure risk of this magnitude is E 95 -

expressed as a straight line with a 45' o ~

slope passing through 4.43 x 10'3 psr , 90 -

SUPPORTIVE reactor year at a dose of I rem, as * -

depicted in Figure 6. z a0 -

TREATMENT 2 70 -

• It is recognized that a constant 60 -

exposure risk does not imply constant 50 -

health effects risk. However, the b 40 -

. magnitude of the exposure risk line was lll 30 -

( derived from the linear-quadratic $ -

dase-response model which takes the $ 20 -

l nonlinearity into account. In light of g -

the potential variations in dose- , 10 -

,g Z response models and the age and sex 5 -

distributions of the potentially D exposed individuals, a line of constant 5

"" 2 - HEROIC -

l I exposure risk, the magnitude of which ,

1 TREATMENT was based on the linear-quadratic 0.5 -

dose-response model, was deemed to be -

adequate for establishing the line 0.2 -

0.1 segment for latent exposure risk for 0.05

~

emergency planning purposes.

0.01

' ' 'I ' ' ' ' ' ' ' ' '

• Prompt Exposure Risk Criterion 0 200 400 600 000 1000 1200 1400

, The prompt exposure risk criterion DOSE, RADS includes two line segwents, as follows:

-- The criterion for exposures above ESTIMATED DOSE RESP NSE CURVES WITH MINIM AL SMME, OR HEROC MEMAWMM e se t a rob a of one in one million of incurring 6

Th2 critsria ussd to exprssa the exp:sure In ordar to rsprsstnt this safsty goal in risk guideline have been reduced by a factor of Figure 8, the following calculations were per-ten to account for uncertainties in the analysis formed. Table V-23 of the report by the Commit-of exposure risk for individual accident sequenc- tee on thesaBiological Effects of Ionizing

• es, as depicted by the lower line in Figure 6. Radiations reports a normal expectation of Thus, these criteria correspond to one chance in 163,800 cancer deaths during the lifetime of ten million (1 x 10~7) per reactor year of 1,000,000 persons from all natural and manmade encountering either a prompt or a latent fatali- causes. That same table reports a relative risk ty, with nominal calculated exposure risks, i.e., projection model calculation of the excess number without increasing the calculated exposure risks of deaths, using the linear quadratic model, of by a f actor of ten to account for uncertainties. 2,255 resulting from a single exposure of one Although the exposure risk guideline was not millica persons to 10 rads (essentially 10 rem),

developed from the NRC safety goals,17 it is as previously discussed.

instructive to see how the guideline compares The societal risk safety goal can be ex-with the safety goals. There are two numerical pressed by applying this information as follows:

safety goals: one for societal risks (based on latent cancer mortality risks); and one for Individual Normal Cancer Lifetisie ,

individual risks (based on prompt fatality Mortality Probability ~

considerations). These goals can be compared with the exposure risk guideline, as presented in 163,800 Figure 8 and discussed subsequently. 1,000,000 = 0'164 The societal risk safety goal states:

"The risk to the population in the resulting in the numerical expres u lMof the' area near a nuclear power plant of

• safety goal of one-tenth of one percent (0.1%) of cancer fatalities that might result the sum of cancer fatality risks from all other from nuclear power plant operation causes, as follows:

should not exceed one-tenth of one percent (0.1%) of the sum of cancer fatality risks resulting from all Societal Risk Safety Goal = 0.001 x other causes." 0.164 = 1.64 x 10

The safety goal can be expressed in units of probability per reactor year by combining the f slocistu a:sa sadtv ooat ' *** "'*

f

~

exposues mism outosuus # "Ub """"

(essentially 10 rem) is expected to result

• ', lN

' g/tartur eiss esoion in an increase in cancer-related mortality

/

V  ! of 2,255 per 1,000,000 exposed persons:

\ smseosNCY PlaNeeNo N escommsmoso son.aisus Excess Cancer 2

[ "d*" " " '""' Mortality per .' 2,255 _

Rem per ~

ge d Individual 1,000,000 x 10 rem C

I l l j

- exPosute Stst oulosures peomptaiss esoio" 2,255 x 103 r - = 2.255 x 10'*

1 x 10' l 8 i

( Is"c"oYms"moso ros? The societal risk safety goal can be repre- ~

! 2 -

assas m exesss sented by the following equatton:

i os ouiosuN

- ourosums enoucio sv ( ' $

eactoa os reps to account S.G. = XxDxLxf .

,, e _ ros unaeramnas

\l Where:

! X= 2.255 x 10-* excess cancer mor-l -

• j/

- memouai siss sarity som tality per rem per individual j j .D = Dose (rem)

..l -

/ ./. L= Ndmber of Reactor Years per i io tm , , .

Plant (assumed to be 40) we.ots soor ooss. asm f= Frequency of Dose (D) per l

Reactor Year

! FIGURE e COMPARISON OF EXPOSURE RISK GU10EtlNE AND INTERIM Solving the above equation for f at a dose SAFETY GOALS of I rem:

1 1

l 7

STONE & WEBsTE85

~

f*

S.G.

• 1.64 x 10-9

• As mentienzd earlier, the proposed exposure XxDxL 2,255 x 10-4 x 1 x 40 risk guideline was not developed from the NRC safety goals. The above discussion is merely 1.82 x 10-2 yr -3 included to indicate how the proposed guideline compares with the safety goals.

where f is the rate of occurrence of an exposure The concept of displaying exposure risk as a of I rem per reactor year without exceeding the plot of dose as a function of probability of societal risk safety goal. occurrence is not new. Farmer proposed the same In the context of Figure 8, the societal idea in 1967 s and more recently Xinchin,18 risk safety goal is depicted as a line of con- Birkhofer,20 Levine.21 and Uchida22 have pub-stant risk (i.e., decreasing probability with lished work which would provide similar approach-increasing dose) which intersects a probability es. Although these articles address different

. of 1.82 x 10-a per reactor year at a dose of aspects of exposure risk, the concepts are quite <

1 ren. It is a facto,r of ~4. greater than the similar to the emergency planning zone definition proposed exposure risk guideline in the latent exposure risk guideline proposed herein.

risk region and a factor of 40 greater than the f

, proposed latent exposure risk criterion, which APPLICATION OF GUIDELINES WITH PAST AND PRESENT has been reduced by a factor of ten to account SOURCE TERMS

~' for uncertainties in the risk analyses. .

The individual risk safety goal states: Summary of Analyses "The risk to an average individual in the vicinity of a nuclear power plant Analyses of exposure risks were performed of prompt fatalities that might with past (WASH-1400) and present (IDCOR and NRC) '

result from reactor accidents should source terms with their associated probabilities net exceed one-tenth of one percent of occurrence. The result of these analyses are (0.1%) of the sum of prompt fatality compared with the proposed exposure risk guide-risks resulting from other accidents line in this section. The source terms and to which members of the U.S. popula- associated probabilities are reported in IDCOR tion are generally exposed."

~

Technical Report 85.4(5.3).1 This safety goal can be expressed numerically A reference reactor power level of 3,412 MWt as follows: was used for all cases to allow direct comparison of effects of changes in source terms on a Since there are approximately 100,000 consistent power level basis. The release accidental deaths per year in a national fractions for specific radionuclides, for each popula, tion of approximately 200,000,000,1" source term, were multiplied by the end of core the normal accident risk has been inter- life inventory for a reactor at the reference preted as follows: power level. The doses for the specific plants included in this study can be determined by 100,000 200,000,000 = 5 x 10"* per year multiplying the results reported herein by the following ratios:

Therefore, the numerical expression of the individual' risk safety goal may be expressed as: Power Ratio Plant MWt Plant MWt/3.412 MWt Individual Risk Safety Goal = 0.001 x 5 x 10-*

= 5 x 10-7 per year Surry 2,441 0.72 Zion 3,238 0.95 In depicting this safety goal in Figure 8 Sequoyah 3,579 1.05 doses above the 500 rem were assumed to result in Peach Bottom 3,293 0.97 prompt fatalities and were assigned the 5 x 10-7 Grand Gulf 3,833 1.12 probability asscciated with the individual risk safety goal (i.e., a horizontal line at 5 x Both acute whole body doses and committed 10~7 yr-a for doses equal to or greater than whole body doses were computed. The committed '

500 ren). The proposed prompt exposure risk whole body doses include the acute dose and the criterion is a factor of five lower than the lifetime dose commitment from inhaled radionu-individual prompt risk safety goal. clides. The acute doses are compared with the It should be noted that the individual risk prompt exposure risk criterion and the committed safety goal is proposed to be applied within doses are compared with the latent exposure risk I mile, whereas the exposure criteria proposed in criterion.

l this report are intended to be applied at any As discussed in the Background (see Fig-l distance from a nuclear power plant. As a result ures 3 and 4), the whole body exposure risk from of the rapid decrease of dose as a function of the WASH-1400 release categories is dominated by l distance, the two prompt risk line segments, from the low probability, high-dose release catego-l 150 to 500 ren, and above 500 rem, respectively, ries. Of the nine PWR release categories,.

I are only important within a few miles, even with release categories PWP-1, -2, and -3 comprise the WASH-1400 source terms. At a distance of 10 essentially all of the PWR esposure risk.

miles, calculated mean doses are in the latent Similarly, release categories BWR-1, -2, and -3 risk rar.ge, even with WASH-1400 source terms. Comprise essentially all of the BWR exposure a

~

J risk. An analysis of the complementary cumula- the prompt exposure risk cri_terion, discussed tive distribution functions of whole body expo- earlier. In Figure 10, the calculated committed sure risk with the WASH-1400 release categories exposure risks are compared with the latent

, when combined with the WASH-1400 core melt exposure risk criterion, probability of 5 x 10-s yr-8, indicates that ~95 ,,.

percent'of the exposure risk is accounted for by E j release categories with probabilities of occur.  : e i mn tats a.

. rence less than 7 x 10** and a x 10** per reactor -

• 8'"*" '"*"

e s suur v year, respectively, for the WASH-1400 PWR and BWR j release categories.

ii ** " " '8C '

Furthermore, the low probability, high-dose  :

release categories are dominated by a few acci- - '8"""*"'***

dent sequences. The PWR-2 and -3 release catego- ~

'8""**"*"''

ries are dominated by the TMLB'-6 , TMLB'-y, V 3 m.s c E and S C-8 sequences. The BWR-2 Ind -3 release

• E lu '

categorieI are doeinsted by the TW-y', TV-y and 8

~

' *3 TCy sequences. Thus, it is appropriate to 3 l e. ' e s concentrate analyses of exposure risk on the low I e* ,

probability, high-dose sequences. Most of the g E '

en effort in the recent reassessment of severe et  ;

~

l accident source terms has reflected such cor.~.

tration. This is also the case in this reassess-l* ,,'osisa .,r Conie supos,ues ose o ment of emergency planning requirements, the f "', i results of which are summarized below. Table 3  :

provides a definition of the above dominant ,-

accident sequences.

taMa s b Ilminert ACCIMot MM MF13f72W - '

et .. i R'IA' , i

' frene tent fallseed by fattare of tae stees generettee relief g le 103 50 3 Be*

mtem and pam navnesee ornam, festere of e ilung f=d.

enu, ese esereemry of ten a pews etuu -3 se fellee- ACuts wMots soor pose. esa see a suuse unues coeus-set fulere se au to en art,testutaas esorymewe tremenent, bresca. FIGURE 9 tis.s . y its.s' esta eenument future aw to byereses weases. MEAN ACUTE WHott 800Y EXPOSURE RISK AT to MILES WITH WASH-1400 SOURCE TERMS v tema of tae meter eeelaat pe e.e. boommesy taCes) er a mten seenwue u tas scas eeuses of emuammeet. valotae '0 -8 -

ebsta senad teelete the bred Instsee fram tan reester woesen 8e asamed to feil. The breet lecettee autende resuaemmet -

any er est met be towered by e enter peel, dependsee eyee te*

daendeel pleet efetas and be .

u-t ur-tm u ne=ti, eld,t,ag beenf emd.asseettees. The ese- .

g sC*4 - -

e of ea.st!e.at.

laCa (ee 2 &a. egnaveleet diameter) u.e f.allseed _by a lose

c. ,e-et f.aere u o t.t e,re,. ..ri,.,et u., ,e bre..ei.  :

I ,,,,, ,,o.

.wer,reeee .

, e, e, .

CRittBioN M ~

' l

. .... ,r t ..1- b. . es.,lete 1 o ., .li .ee.. ..t re

.1 u,eunity. ca.ut.-t ini-e se e- u -r,r.oen e.

e tu reluu a noemed u b, pen tan reuur wandtse. . -

o - x TW

• 1 TW ella tae everpressere breach of testaamment essemed to re* O ime u, e tu nutu wateias a tw m n. =

e, e, Tc . y freesteet fellemed by teos of resetae stateous cepettlaty (ee- a 60

er taetpeted trees *ent estaset spaa

• aTW). Ceetesament feat. E e

3 are se essemed to be the y ende es everpreesere release. ,,- . .

TC

• Weet/to TC ewest esta aperater ettaen to oost tae weteelt a&repete j

, spee had preseers. 51$ s tes wad u

t. <wm t.e, oui o. u.. m.e. . un nu. i e teen / gee.u.peratore uoment

,.weil fem. en m-well flee late las estus11 ese tpf seppreestee poet entti rr.

as i tae reseter I

=

., su.evims.a.  :-

't IU80Y IMO *I mus-set bmea ame a sw dryme!! ens u non t,apue- -

tore frem the ette comesete tatsraettaa. -

e n sueer y TC(g)

• Sant/ TC esent I

• es sumeY s2C 8e to este ayerator tana Lawsotery settee hefere lese ts restore of CW flee. Oe,sadessete ereses e eterage xt PlaCM sonomb tw.y*'

me the en-11 a .lo. . mu.s as me,stsee mmeto 90'8

=-

boun. Tw unimal cue f1- premu ==== *,mell  :

x naCM sonom tw.r t-smt= e. sets u me-ed te fut etur en=11 muse x PlaCMso"omfC I Figures 9 and 10 present the results of ...

analyses of prompt and latent exposure risk, ' 'o 58 '88 **

respectively, at a distance of 10 miles with the 'C""*"'C"'*"""

four dominant PWR and three dominant BWR accident FIGURE 10 -

sequences from WASH-1400. In Figure 9, the MEAN WHOLE SODY 00SE COMMITMENT EXPOSURE RISK AT 10 calculated acute exposure risks are compared with mites WITH WASH 1400 SOURCE TERMS 9

._e..s... A

It can be observed from these figures that Although the risks from the individual sequences

, the mean exposure risk at 10 miles is in the from WASH-1400 are within the proposed latent latent risk region. As discussed in the IDCOR risk criterion at a distance of 10 miles, the Technical Report, 95th percentile doses would summation of the fractional contributions to risk result in exposure risks approaching the prompt exceed the criterion. It should be noted, risk region at 10 miles with WASH-1400 source however, that the summation of risks at a dis-terms. tance of 10 miles with WASH-1400 source terms is Figure 11 presents the reruits of analyses within the comparative criterion derived from the of prompt exposure risk at a distance of 2 miles proposed interim societal risk safety goals.

with WASH-1400 source terms. At this distance, Figures 12 and 14 presents a graphical depiction the risks are observed to be an the prompt risk of WASH-1400 data in relation to the interim region. Therefore, data for the latent risk safety goals. Establishment of a plume exposure region, comparable to those presented in Fig- pathway EPZ of "about 10 miles" with WASH-1400 ure 10 for a distance of 10 miles, are not source terms in NUREG-0396 is observed to be

- presented. The risks from individual accident roughly consistent with the exposure risk guide-sequences exceed the prompt risk criterion, and line approach proposed in this report.

the summation of the fractional contributions to overall risk far exceeds the prompt risk criteri- ".

on at this distance with WASH-1400 source terms. j % ,, ,,,f ,,,, ,,,,,, ,, ,,, g,,,,, ,,,,

lo*3 -

N . s ..

Notti Pwt.e F 4.

I ~

a9 (185m

~. SUID gwg,$ i gt,

*., 1. N, IPo,$,Utt .el5E,,y.o.

N ELINE

=

N ilmstoENCY PLANNINo l 08CommiNotD fot tilKs IN IICll5 of 10 ** -

oWIDtuht REDUCID sf $ACfot of flN

l f o ACCOUNT fot UNCIlf AINTill

~

l X I io > =

~

T

=

N/ \

-  : i t .. U.. ,,,. oo,,,uN, PROMPT EXPO $Utt E. -

3 Peompt aisa esosom

_8  : Ris CanttiON x, j

...._i . . . _

- ..Comm,No.....

..* ., . s

etsalIN RECIS$

3 4 _ 8

,, = = - or oUiosuht t

• - 8 , 6 g  : s.

g ,,.

VN Q i,o.r .

r  ;  : 7

1. 3 SuttY Tmit'.3. .

,,,,,,,g ,g ,,,, ,,,,,, n o g g

~

e2 SuttY Tutt'.7 , ,, ..! , ..I / .

83 SUSEY V e to soo i poc to ooo E #a SUttY I2E.Se wuote s00T oost CommifulNT. GIm

~

X g PEACM BOTTOM TW.T

~

X2 PtACH BOTTOM TW.Y COMPARISON OF MEAN WHOLE BOOY DOSE COMMITMENT AT to

~

X3 PtACH SOTTOM TC. 7 MILES WITH WASH 1400 SOURCE TERMS WITH EXPOSURE RISK

' GUIDELINE AND INTERIM SAFETY GOALS to '

io not 1o 3 sod

( ACUTE WHOtt tooY cost. Etm Figures 13 and 14 present the results of I analyses of prompt and latent exposure risks, respectively, at a distance of 2 males with IDCOR MEAN ACUTE WHOLE BODY X $ E RISK AT 2 MILES WITH source terms. At this distance, the risks are WASH 1a00 SOURCE TERMS observed to be in the latent risk region. The As discussed in the IDCOR Technical Summary risks f rom several of the individual sequences

(

Report, the summation of exposure risks at a approach the latent exposure risk craterion. As distance of 10 miles with the important WASH-1400 discussed in the IDCOR Technical Report, the accident sequences results in an overall risk summation of the f ractional contributions to i

l 1.58 times the latent risk criterion for the PVR overall risk at a distance of 2 miles with IDCOR j source terms and 2.12 for the BWR source terms. source terms is as follows: '

l 10

s Summatica cf Expssure *-' s ,,,,,,,,,,,,,

• . Risks Compared with

., rion e.t. . ,

the Latent Exposure _

• moa w Sequences Criterion Risk *

.,ssowo,aas ta.s.

,  ; . uovo... ,

Zion Sequences 0.0.  : es mouovaa s imos a.

Sequoyah Sequences 0.13  : '8 Peach Bottom Sequences 1.72 * *"*" ['*[

' ~~

y.* ,

Grand Guli S quences 0.41 3 5 ** e. - ,'

These data indicate that the criterion is g *, l $,,,,,,,,,,,,,,,,

u not exceeded at a distance of 2 miles with the .

= +; canision '

IDCOR PVR (Zion and Sequoyah) and the BWR .

E '8 -

H.

Hark III (Grand Gulf) source terms. Although the j ,8 l 8*

risks for individual BWR Mark I (Peach Botton) g ,,.,

sceident sequences are within the criterion, when E E I' ainacu sano= rc.ve=re,e the fractional contributions to the overall risk  : samacasonomic.*re are susumed, the criterion is exceeded at this  : Ye '

, ,'*" *," ,',*,,* 7' "

distance. This is a result of the contributions ..e ,

1. naen seno. tw. re from the TC(q)-Vent /0 and TC-Vent /tB sequences. E g a.=4cm onom,ovv.,e These sequences are presently being reanalyzed by . l ',,"*

, *j ',';Fl.

IDCOR, and are discussed in more detail as -

+ osano ouu touv. r- .

follows. See Table 3 for a definition of the .'

l ,

TC(q)-Vent /0 and TC-Vent /te sequences. ' * "' "' **

It is concluded that a plume exposure ""*'*"'*"*"""'"'""

pathway EPZ with a radius of about 2 miles is FIGURE 14 appropriate with IDCOR source terms, as contrast- MEAN WHOLE sODY DOSE COMMITMENT EXPOSURE RISK AT 2 MILES WITH IOCOR SOURCE TERMS ed with a radius of about 10 miles with WASH-1400 source terms. Figures 15 and 16,present the results of so-3 :

analyses of prompt and latent exposures risks,

, , respectively, at a distance of 2 miles with

{

Xg PEACH SOffom TC. VENT /rf NUREG-0956 source terms. At this distance, some X PIACH 80ff0m fC(e).yf of the risks are in the prompt risk region while X PtACH sOff0m fC(g). VENT /s

[ others are in the latent risk region. The prompt X4 PEACH Boff 0m TC. rf exposure risk for two NUREG-0956 analyses - the to

• Xs PEACM SOff0m Tw T8-

=

- Surry V unsubmerged, and the Peach Bottom TC-y' X PE ACM SOffom f 0VW. e f

! sequences - exceed the prompt exposure risk i

1

+3 o. ANO oWLF TC. 7' criterion.

Xp

. oaANO QULF TW.T ,

+3 osapo cutF70uV.r *8 5 er ic'I .

• e, sweet f ate .3.

+y

_ , [ e, sween tme a e.

g [ X, sa

• e, sues, ,=. .. -

e, swee, ,

g PROMPT I i

e, sues, v : sus.esoso.

=t _ 'tXPOSUtl fl8K CRITIBION -

e, sveer sio r E o-' '

= 2

,,C"'*"',

I

e. m: n E nac..orio.i.-

E es 3 "

4 - #4  :

E E E -

24 ~

5 ,

I g - +: Xs I.

$ gg*F -- , g *

'e r Peompf It'oswel ^

{, g g sess casfletow

- h{ , ., ziOu i m e.S, g

y e, , ,,

- y, .ir>0=fm.-s 9 ZION V m'

,,.. _ si siivoYAu imts.s. .;  ;

52 SEOUOYAH V -

[- es seOuCYAH s Heroes 3, -

e, 84 380UOYAM $2 MF(Dobla ** i: -

• e s EOUOYAM s *

$ 3HF(DOF A [ , i *a 10-9 ' ' ' ' ' ' ' ' ' ' ' '

I 30 102 som see ,, i Acuff WHOLI SODY 00st, atm .

i

. .. ie*

acu's wuota soor ooss esa FIGURE 13 FIGURE 15 MEAN ACUTE WHOLE SODY EXPOSURE RISK AT 2 MILES WITH MEAN ACUTE WHOtt 800Y EXPOSURE RISK AT 2 MILES WITH IOCOR SOURCE TERMS MUREG.0954 SOURCE TERMS

,,. . e . . e . . ,. . A

so 3 It is concluded that a plume exposure

g E

pathway EPZ with a radius of about 2 miles is

• speeY Tmts'.8 e appropriate with present NRC source terms, with

. [ es suest imts. A the exception of the above two analyses which as suaay ists.e require further investigation.

88 ** : e4 suesy v es suety v (suomsaoto) Discussion of Exceptions f

_ e spatY se c. r  ;

x iesACM SOff 0M fC.y- IDCOR Analysis of Peach Botton TC-Vent /tB 1 X Puew nottom tw.v' I .s E As noted above, although no risk from any I  : individual sequence analyzed in IDCOR exceeds the

• 8

. e proposed criteria at a distance of 2 miles, the LAtsNT EXPCsues summation *of risks for the BWR Mark I (Peach

' $ Botton) sequences exceeds the latent risk crite-alsa CarttaioN 3,, -4 rion/due primarily to the contribution from the g E

• '  : - TC-Vent /tB sequence.

The drywell overtemperature failure mode (t)

-.g

( _

a r, is currently under further investigation by IDCOR

,8 and appears to be considerably less likely than I '8 #

previously thought. Given that the drywell E 5 overtemperature failure would not occur eveh without ex-vessel core debris cooling with the wetwell vented, the TC-Vent /tB sequence will not

't occur, and there would be only a slight increase We  :: .,

in the probability of releases characteristic of i

e s

the present TC(q)-Vent /8. As a result, it is expected that the summation of the fractional contributions to risk will meet the exposure risk criterion at a distance of 2 miles with IDCOR source terms when the current IDCOR reanalysis of 8 'o not nos to4 the TC sequences is completed.

WHOLt BOLT DOst CommlTMINT, rem NUREG-0956 Analysis of Peach Bottom TC-nGURE16 r Sequence MEAN WHOLE SODY DOSE COMMITMENT EXPOSURE RISK AT 2 MILES WITH NUREG.0054 SOURCE TERMS The analysis of the TC-y' sequence for Peach Bcttom as reported in NUREG-0956 (obtained from The summation of the fractional contribu- BMI-2164)23 is qualitatively much different from tions to overall risk at a distance of 2 miles those performed by IDCOR and other investigators.

with NUREG-0956 source terms is as follows: The BMI-2104 analysis of the thermal hydraulics and fission product behavior within the reactor Summation of Exposure pressure vessel (RPV) is substantially different Risks Compared with from the IDCOR analysis. The BMI-2104 methodol-l Sequences Exposure Risk Criterion osy results in very little retention within the

'- RPV at the time of vessel meltthrough, whereas Prompt I.a t ent industry analyses result in very high retention.

With respect to the analysis of RPV melt-Surry Sequences 10.47 1.41 through, the industry studies are based on the relatively rapid failure of individual or multi-Surry Sequences 0.47 0.08 ple penetrations in the lower head, whereas the Excluding BMI analysis is based on meltthrough of the thick V-Unsubmer:ed vessel head itself.

The post-seltthrough core / concrete interac-Peach Bottom 3.07 tion phenomena are the subject of ongoing inves-Sequences tigation. However, at should be noted that the behavior of the molten corium is very much Peach Botton 0.07 affected by the rate and mode of ejection from Sequences the RPV. The release of tellurium and barium-l Excluding TC y' strontium during this phase of the sequences has j

been identified as the major consequence. Analys-The NUREG-0956 data for the Surry V.unsub- es of this issue should emphasize: 1) the rate merged and Peach Bottom TC y' sequences consti- of corium ejectin ; 2) the corium temperature vs.

tute exceptions to the proposed plume exposure time; and 3) behavior of the corium on the concrete pedestal, including *the production of pathway EPZ radius of 2 miles and require further analysis, as discussed subsequently. aerosols and the release of fission products.

12 1

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

.' i The NUREG-0956 analyses will be superseded e Although located in larger structures,

~

bytheanalgsespresentlybeingperformedfor flood protected compartments are NUREG-1150.

• employed for ECCS systems and the low pressure pipe break location may be NUREG-0956 Analysis of Surry V Unsubme_r j ed flooded for other plants as well as Sequence Surry, but with additional enclosed structures to facilitate fission The probabilities associated with the product retention. Additionally, net-Surry V unsubmerged sequences reported in positive suction head considerations NUREG-0956 were based on WASH-1400. In Appen- for the ECCS pumps require that the dix V of WASH-1400, a discussion is presented pump suctions (and hence the pump which addresses the effect of check valve testing rooms) are well below grade where the

.on the probability of occurrence of the V se- RWST is generally located. This quence at Surry. Specifically, WASH-1400 stated: provides a passive gravity flow condi-

. ". . . yer rly testing' could yle.1d about tion to supply water to help flood the an order-of-magnitude decrease -, break location.

compared to the average probability of failure for the Low Pressure e: At some plants, the piping code classi-Safety Injection System (LPIS) check fication change from high to low valves. If more frequent testing pressure occurs in containment with were to be performed, there would be points of highest stress near the point a further decrease in the failure of transition, thus making the ECCS probability..." WASH-1400 went on to rupture sequence one additional in-state, "In summary, at least an . containment pipe break sequence, and order-of-maAnitude dgeresse in the not a containment bypass sequence.

failure probability for the LPIS check valves can be obtained by a

• The system arrangement included in the reasonable testing program." NUREG-0956 analysis is unique to Surry, Such a te rting program was initiated as a e.g., two check valves and one motor-result of an NRC-IE actification and is in effect operated valve which is locked open.

at Surry and other PWR plants. Thus, the proba- The systems analysis aspects and the -

bility for this sequence should be reduced by an probabilities of this sequence, includ-order of magnitude for Surry. This would remove ing the thermal hydraulics reported in this sequenc'e from the exception status. NUREG-0956, are limited to Surry.

The Surry V unsubmerged sequence analysis reported in NUREG-0956 essentially represents a In summary, the Surry V sequence analysis as hypothetical case of release of fission products reported in NUREG-0956 ceases to be an exception via a small contiguous structure with postulated relative to meeting the ;roposed exposure risk direct release to the environment. At Surry, the criteria at a distance of 2 miles when the above postulated break in the low pressure portion of considerations are included. It should be noted the safety injection system is most likely to that the IDCOR V sequence analyses discussed in occur at a location which is submerged by 3. feet this report are well within the proposed guide-of water in a compartment in the small (10,000 line for emergency planning within an EPZ radius cubic feet) safeguards building.2s The flooding of 2 miles. Additional analyses of the Peach results from the break effluent and the drainage Botton TC sequence are in progress in support of of the contents of the Refueling Water Storage NUREG-1150 and are expected ta replace the Tank. The depth of the flooding is limited by the NUREG-0956 data for this and other sequences.

height of a large opening to the adjacent struc- The analysis of exposure risk with past and ture. The NUREG-0956 analysis assumes that the present source terms illustrates the substantial pipe break occurs outside the flood region with reduction in calculated risk resulting from the the WASH-1400 probability of 4 x 10-s yr -1 new improved knowledge and understanding. This apportioned'at 3 x 10-s yr -1 and 1 x 10-s y,-1, reduction is evidenced here by the substantial respectively, for the probability of submerged decrease in the distance at which risks compara-and unsubmerged pipe break locations in the ble to those based on WASH-1400 technology are ,

volume of the building used in that analysis is calculated based on present technology. Although not apparent in either NUREG-0956 or BMI-2104, a few exceptions have been identified and dis-but may be 250,000 cubic feet (obtained from cussed, the resolution of the current differences BMI-2104). which result in these exceptions is under way.

The Surry V unsubmerged analysis is not Emphasis on resolving these curgent excep-applicable to other PWRs for the following tions should not obscure the larger perception of reasons: the substantial reduction'in emergency planning requirements, which is supported by the totality

• The most probable low pressure piping of the newly available data.

break location at most PWR plants is in large auxiliary buildings, ranging in .

free volume from nearly one million to several million cubic feet.

13 t

evows & WassTan

.c ..

?

REFERENCES 14. Effect of S:urce Term Compasition on Offsite

. Doses, P. Karahalios, and R. Gardner, SWEC

, 1. IDCOR Technical Report 85.4 (5.3), "Reas- TP 84-56, ANS Topical Meeting on Fission sessment of Emergency Planning Requirements Product Behavior and Source Tern Research, With Present Source Terms," Stone & Webster Snowbird Utah, July 1984.

Engineering Corporation, February 1986.

15. " Technical Guidance for Siting Criteria

2. IDCOR Technical Sumary Report, " Nuclear Development," NUREG/CR-2339, Sandia National Power Plant Response to Severe Accidents," Laboratories, November 1982 (Also referred Atomic Industrial Forum, Inc., November to as the Sandia Siting Study).

1984.

16. "The Effects on Populations of Exposure to

3. " Reassessment of the Technical Bases for Low Levels of Ionizing Radiations" (BEIR Estimating Souree Terms." NUREG-0956 Draft III)., National Academy of Science / National Report for Comment, U.S. Nuclear Regulatory Research Council, Washington, DC, 1980.

Commission, July 1985. ,

. 17. "10CFR50, Safety Goals for the Operation of

4. Report of the American Nuclear Society Nuclear Power Plants Policy Statement, Special Comusittee on Source Terms, September Federal Register, Vol. 51, No. 149, p 28044, 1984. August 4, 1986.

5. Report to the American Physical Society of 18. " Reactor Safety Siting: A Proposed Risk the Study Group on Radionuclide Release from Criterion," F. R. Farmer, Nuclear Saf et y, t

Severe Accidents at Nuclear Power Plants, Vol. 8, 1967.

Reviews of Modern Physics, Vol. 57, No. 3, Part II, July 1985. 19. " Design Criteria, Ccacepts, and Features Important to Safety and Licensing,"

6. " Nuclear Reactor Accident Source Terms," G. Kinchin, Proceedings of International Report by an NEA Group of Experts, Nuclear . Meeting on Fast Reactor Safety, Seattle, WA, Energy Agency, OECD, March 1986. August 1979.

7. 10CTR50.47(c)(2), Code of Federal Regula- 20. " Consideration of a Proposed Rationale for tions. Title 10 Part 50'. Quantification of Safety Goals,"

A. Birkhofer and A. Johns, GRSabH, Nat. Rep.

8. 10CTR50, Appendix E. Code of Federal Regula- of Germany, Proceedings of International tions Title 10 Part 50. Meeting on Thermal Nuclear Reactor Safety, Vol. 1, February 1983.

9. " Planning Basis for the Development of State and Local Government Radiological Emergency 21. " Safety Goals for Nuriear Power Plant Response Plans in Support of Light Water Regulation," L. Levine and F. T. Stetson, Nuclear Power Plants," NUREG-0396 (EPA Progress in Nuclear Energy, Vol. 17, No. 2, 520/1-78-016), December 1978. 1986.

10. " Reactor Safety Study - An Ass'estment of 22. " Current Nuclear Safety Issues in Japan,"

Accident Risks in U.S. Commercial Nuclear Hideo Uchida, Nuclear Safety Commission, Power Plants," WASH-1400 (NUREG-75/014), Japan, Proceedings of Second International October 1975. Topical Meeting on Nuclear Power Plant Thermal Hydraulic and Operations, April

11. " Assumptions Used for Evaluating the Poten- 1986, Tokyo, Japan.

tial Radiological Consequences of a Loss of Coolant Accident for Boiling Water Reac- 23. "Radionuclide Release Under Specific LWR tors," Regulatory 1.3, U.S. Atomic Energy Accident Conditions," BM1-2104, Battelle Commission, Revision 2, June 1974 Columbus Laboratories, July 1984.

12. " Assumptions Used for Evaluating the Poten- 24. " Nuclear Power Plant Risks and Regulatory tial Radiological Consequences of a Loss of Applications," U.S. Nuclear Regulatory Coolant Accident for Pressurized Water Commission, NUREG-1150, U.S. NRC (to be Reactors," Regulatory Guide 1.4, U.S. published).

Atomic Energy Cosesission. Revision 2. June 1974. 25. "The V-Sequence: An Engineering Viewpoint,"

A. Drozd, F. A. Elia, Jr. , J. E. Metcalf.

13. " Calculation of Reactor Accident Consequenc- ANS Topical Meeting of Fission Product es Version 2. CRAC2: Computer Code," NUREG/ Behavior and Source Term Research. Snowbird, CR-2326, SANDB1 - 1994, February 1983. Utah, July 1984 14

.. 3 .

I l

l i

i NRC STAFF PRESENTATION TO TE ACRS

_. ON THE REVIEW PLAN FOR THE SEABROOK EERGENCY PLAfMING SENSITIVITY STUDY SCOPE AND FOCUS OF STAFF REVIEW S. LONG (NRR) -

COMPARISON OF PLG-Of456 WITH NUREG-0396 D. f%TmEWS (IE)

PETHODOLOGYANDJTATUSOFREVIEW:

SOURCE TERMS T. PRATT (BNL)

RISK ANALYSIS CONTAINfft.T STRUCTURAL INTEGRITY C, HOFMAYER (BNL)

CONTAINPENT BYPASS R. YOUNGBLOOD (Bht)

INTERFACING SYSTEM LOCA l

l fef f)- 3'7- Y 5 / 10 1

i PURPOSE OF EVIEW TO IDENTIFY AND REVIEW TE PORTIONS OF TE STUDY THAT ARE M)ST SENSITIVE WITH ESPECT TO TE STUDY'S PRINCIPAL CONCLUSIONS IM)IVIDUAL RISK OF EARLY FATALITY AT SEABROOK IS WITHIN SAFELY G0AL 1 MILE EVACUATION AT SEABROOK PROVIDES SIMILAR RISK 0F EARLY FATALITIES TO TE WASH-Il400 RESULTS WITH 25 MILE EVACUATION 2 PROBABILITIES OF SPECIFIC RADIOLOGICAL EXPOSUE LEVELS AT 1 MILE FROM SEABROOK AE LESS THAN THE CORRESPONDING PROBABILITIES SHOWN AT 10 MILES IN NUEG-0396 1

, - . _ - _ a m A h h 4

9 BASES FOR CO WARISONS WASH-IL100 SOURCE TERM ETHODOLOG EARLY FATALITIES WHOLE BODY DOSES 4

e O

t

?

e i

MAJOR CONTRIBUTORS TO RISK DIFFERENT CONTRIBUTORS FOR DIFFERENT RISK COMPARISONS:

- PROBABILITY OF EARLY FATALITIES, GIVEN EVACUATION, APPEARS TO HAVE CONTRIBUTIONS FROM SEVERAL RELEASE CATAGORIES AND EVENT INITIATORS, EVENT V WAS DOMINANT IN ORIGINAL PSA, DOSE VS DISTANCE CURVES (N0 EVACUATION) ARE DOMINATED BY SINGLE REl. EASE CATAGORY AND SEISMIC EVENT INITIATORS.

THESE CURVES WERE NOT PRODUCED IN ORIGINAL PSA,

^ ' ' _ _ _ _ _ - - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .

.S FOCUS OF STAFF REVIEW EFFORT FACTORS SHOWN BY CURRENT STUDY TO BE IPPORTANT FOR RISK MITIGATION AREAS WHERE SIGNIFICANT RISK REDUCTION OCCURRED BENEEN UPDATE AND ORIGINAL PSA o

• 1

4 %%

l PLAKf DESIGN FEATURES SIGNIFICAKT TO REVIEW CONTAINENT STRUCTURE RHR VAULT o

e

\

l

O g g-e P0DELING FEATURES SIGNIFICANT TO REVIEW l

CONTAIPPENT RESPONSE AT HIGH PRESSURE CONTAIPPENT BYPASS ASSlWTIONS INTERFACING SYSTEMS LOCAS COPFLETtNESS CHECK VALVE FAILURE DATA SOURCE TERM REDUCTION FROM SCRUBBING OPERATOR RECOVERY CREDITS EVENT V STATION BLACK 0llT 6

h unsi i

l I

.. 10 MILE EMERGENCY PLANNING ZONE '

-~ -

UPON 4 PRI'NCIPAL CdNSIDERATI0 tis '-

1) DBA DOSES LESS THAN PAGs EEYOND 10 MILES

- 2) MOSTCOPEMELTACCID$NTDOSESLESSTHAN -

PAGs EEYOND 10 MILES .

3) FOR WORST ACCIDENTS -- PROMPT DEATHS

- GEI'ERALLY WITHIN 10 MILES

4) 10 MILES - BASE AREA EXPAND AD-HOC IF NECESSARY .

O E

l h==-w...i ie c-

meeama.emm..sh-4.- im.w.mma.asAW--8+-- 8- - *- #-a--' * - -+-a-.**& 4 m%am--at_-._.=4--.==_____._A  %,.g4 .-_4.g%ua ,mm m,_,. _A m. .4 ,

l O

O 2

m PROBABILITY OF EXCEEDING WHOLE BODY DOSES GIVEN A CORE MELT ACCIDENT E ,

o m '

o r=

@ e' - 4

~

i i i i i i sit i i i i si il i i i ti e in >

O O

[l>m O m2 ,

O4 '

~ CW dO, a>

1

! o-r 5 -

gO ,

,, C 9

m n t- r-8 m)a z 2 m 2 -

2 m =/=- 05 g -

o m mz

$ E Y MW F -

g m

- rI C

-  : Em 8

mE mm M2

- TO Om

- cn 2 CO m<

W r-

~

g" z-

- . z g i O

N O

2 m

l ., . .

9 ,

j  :  :

l l

CORE MELT ACCIDENTS CALCULATIONS i EMERGENCY PLANNING ZONE l

OF ABOUT 10 MILES FDR PLUME EXPOSUR l .,.

About 30% chance of exceeding s

PAG doses at 10 miles

! Less than one chance in 100 of l ..,.,

exceeding life threatening do'ses

l beyond about 10 miles i

l e

9

• I $

OM/6WL- % LredL....%lM.4o - .

e .

.~

2 .... . -. - . .. . . .

~

. ph - <- . . . .. M F . C ... . . ..

. WGre.9'd l b . ..... . . =E 1? h> i .

. .. AU.Y . operdon.s. .

E p.m- sel h a m aia . .

u.te v.uua) '

3m Mood -

aVH Y.. . .

2.N'?/ A. . - . . _ .. . $NW^2-C. ..- .  :

. &. . . BAGenz .- . . . . . . . . - . .

. yes .

. .x nf . _ - _ _ _ = n / P w e M ..M. ._ - _

.. . QW/ .. $/ U$ .. . .. ... . . .- . ... ?L f.O -. N f.

j%e m,//e . . . .

. B nA.. . . . - .- . . . .

f4 .h ou> d?u! . .

. BALL.-

k2 d  %.

NHX=Enfweekf

. . . . ..._l f . . . . . . _ . . . . . . . - . . . . . .. . . .

hWekw b.o& ._ .

.Th.f ihpW xand . R.  % dyq R,y . ... .R iv i .

do,m L wu. .

5 l @gu azo b' 34]J L.c s s~kay Nso ysec. z9 m 3ob Gigor$ .NAet. LsceJsm)q

$ e e. . . .. . . . -- . .

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

. . . . . . . . . . . _ . _ . . . _ . . . _ . . . . . . . _ . _ _.. i _ _ _ _

Fo2.// 7 6//

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

9 v . . .e . . ,s 6

-- ---e . . . . . - . ,...4...e . , _ . . _ . , , , , , , , , , , . . -. ..

V Noonu l

{ Ae.hn  :

f'*M"'"1

,, gi d /ncAvclok.

do Q nau vuv) '

uu.ucs g gec/ fo goucen A-use cudadeo , yqc

. 3 h u.uc noka- ma o6lcarch . '. utilparas e Q otyed :. un. up*.poxL WA mi p th ihn wa (stowp by ekuJ-kkJ a p6y w ped 74

% "'6"'

la Creuk 9 0 "'*/7 -

e a

^5? pm. .

fed hk uuL o p dy - Aq a Dont p m W L mvd y a x p a .>: s fb1/I-S7-7 Sh

I

/

(bcarid: C ,<c n T c e c x -

fcopcm. *lb Woed& ~cd J cv>1 a d . d64 Aaal Mfd

'fcai CcwmA rn3 ~ .o_3 rem *4 f2 L +rn ecwet eez. c ru n.m e r e ANC.V - a loco A MM d awa _

q cum 7% ru aca tccen) hi e sn _

% b ~~-&.nAmv

'(

[',( [ ~ ' '

('t. e' y

p. i t,o a

,,j t i T {'7 I '

a 2 . . ~1

(' ) 5' ,]

l .'< r -

Qr. .

S.w 2. o.r d u. J.e.+ f . nu-t .

~

d =

)( ) f, () b , I g ,

s ex !- .

s ,JJ] '

,,o,,u* l n - '.' -

,('

l csp.;y c ( .. l ' . L !:s 1,' < 1 n

. ..i L

^-

3 34-87-7

m) _

y . ..

2/3 i

LABS. /JuecoloirA *) Ac.43 y$ 93n.9 - /V6C 7mIq wth IrJot to 70 e pu.euwA h pu>u.fa*w e /L wor g CW4 inmau>r gunn

/N Se hAett . ,o a & p g co vr w - A k b s Rh oawo uauy mA wal b- e Ackbg cHL.

bcutend uxu dou.

9

' gfl>J enci dc% [ Ltex.ernem I __ _

V W h c.9(cd /o @t i

O /

I fo2n - Tr7 -7

'/1

me,m,,q wu . , = ;= r,. - . . -. 3 _ . . .

wwn =nourJ usark. u,J> AcLS o x_. G e > w 4 n u e a e a A #

mqo d u -% d li, al du_.

.. c?3d _ _

to w - t/:03 A9cna2u < CuJU h M) ochCC. .

/}OulA) to d>-oo uo d* hwg q #cerud

.y (w' M on uneDay g unQco.LL <Qu% O< ^ead W Q a l W , b d 7)a ~ <l e ~ u org M stu.Jh M vAe/ uri de) w d g udec6 h

y & Ao m euc gd ae picu todo. ..

Sf duu uAd ux_uAukn)cdum'

.urtdu) <.m pt fluuYah-Ww j /30-430(430)

L l

> V c.

l tu o 930  % heck Acro - d.ucao z/

.9/er . decod /Otfia*cw t .Auupwe 4 E

&cmhoq/ap pud.AA e=%4 soa.am O'k caum u a .a - d a ner ou d 6 tu-nek

• • "lY teu .

chaaa

/vn/) inuunnaganahu.uzh da)I,4 e<ed - Le Aish

- . , oo 2

u

,, F . ~ . _ _ . _ , . ,

RESPONSE TO QUESTIONS LOW POWER LICENSE 1.

While the rules don't specifically address whether such a motion could be filed directly with the Commission, 2.718(1) does give the Commiscion the authority to certify matters to it, 2.764 indicates the Commission reserves the power to step in at an earlier time in a proceedL*1g, and 2.772 establishes the procedural mechanism by which the Commission could handle such a motion. When combined with the fact that the Licensing Board is acting as the Commission's designJrte. I would think a motion could be filed before the Commission at any time in a proceeding. Such a motion would not have to go to the Appeal Board in ,

the first instance. The showing that would have to be made for the Commission to tlie Commission; agrant (or even consider) the motion is at the discretion of

[ tougher question here is the type of showing that could be made given the facts of this case.

2.

The Staff review of the issues for hearing could be done in no more than two weeks. Considering that a motion would require responses and the hearing is only two months away, I doubt that more than a few weeks could be saved (unless the motion is not to change the schedule but to remove the need for a hearing). Considering the limited nature of the hearing, the Staff could easily support a tightening of the schedule; considering the limited value of a schedule change, I'm not sure the Staff should in this instance support the precedent of the Commission reaching down and injecting itself into the case on such a purely piocedural matter (future policy considerations?). Supporting a motion to eliminate the

' hearing would be much tougher to support; I think this avenue is more trouble than its worth (you'd. have to show that there are no factual issues and beinvery dispute, andto difficult that wod!'d both require a detailed round of papers show).

I don't know how the Staff feels about i

early Commission accomplish very much.

involvement generally; in this case, a motion would not i

EXEMPTION REQUEST 1.

My sense from the technical staff is that an exemption request would take at least three months to review. Technical staff could better answer this question incoming. Butafter they've had an opportunity to closely examine the

! of time may we,ll depend on depth of review;at Indian Point,Length review lack of Thoroughness may of the PR l

i save time, but could hurt at hearing as well as in future (if similar

, requests come in).

\

! 2.

( issues. An exemption request would raise difficult technical and policy I

i the depthThe time of the needed staff review.for preparation before hearing depends upon A rough guess would say that at least six

! months will be necessary from the date of submission of the request until the Staff could be ready for a hearing. Close consultation with the technical staff is necessary here.

Six months seems to be a minimum time; Indian Point took much longer (a year for review, an additional six I

Foza 7 i 5/6

months for testimony). Any gains from decreasing review and testimony time must be balanced against harm of going forward without adequate review and/or testimony. This must be examined very carefully; cutting down the review and preparation time, if it means in any way doing a less thorough review, may leave the Staff vulnerable at a hearing. I can't give a more detailed estimate, but we ought not at this time to be ,

promising that we can be ready to go to hearing in an unrealistically &

short period of time.

3. Generally, a hearing starts approximately six months after contentions are first filed. In this hearing, it is unclear whether contentions would be required (there were none for the Shoreham low-power hearing) or if they would be filed before the Staff completed ,

its review. Contentions would definitely be advisable as a means of narrowing the issues. Tightening of the schedule would either mean foregoing summary disposition (possible) or curtailing discovery (also

/ possible, but likely to generate a great deal of controversy).

k

4. At Indian Point, a PRA hearing took approximately three months. A rough guess would say that this hearing should take a comparable amount of time.

5. Under 52.758(d), the Commission has the discretion to grant a waiver with or without a hearing. Technically , the Commission could grant a waiver without a hearing.

6. Given the nature of the ongoing proceeding, in my opinion an exemption could not be issued under 550.12 without providing intervenors with the right to a hearing.

7. The nature of a hearing 5Iould depend on the Commission. Under

$2.758(b), the proponent of a waiver must demonstrate that "special circumstances with respect to the subject matter of the particular proceeding are such that application of the rule or regulation ... would not serve the purposes for which it was adopted." The waiver petition must be accompanied by an affidavit that must, inter alla, " set forth with paricularity the special circumstances alleged to justITy the waiver or

• exception requested." Other parties may file written responses to the petition. If the Licensing Board finds that a prima facic showing for a waiver has been made, the Doard certifies the matter to the Commission for final determination on whether the regulation should be waived in a given proceeding. Under 52.758(d), the Commission could order further filings, it could rule on the basis of the papers filed before the Licensing Board, or it could " direct such further proceedings as it deems appropriate to aid its determination." Thus if an evidentiary hearing were required, the Commission would have the discretion to determine the nature and scope of the hearing.

Under $50.12, the Commission may grant exemptions from regulations if the Commission finds that the exemption is " authorized by law, will not present an undue risk to the public health and safety, and are consistent with the common defense and security." In addition, 550.12(a)(2) states that the Commission will not consider he issuance of an exemption unless

"special circumstances" are present. The regulation then defines "special circumstances" in some detail. While a party, in my opinion, has a right to a hearing when an exemption is being considered for a regulation involved in a hearing, it is not clear whether the hearing would encompass the entire scope of $50.12 (the consideration of special circumstances is arguably at the Commission's discretion) or just the "no undue risk" standard in $50.12(a)(1). In the first instance, it would probably be up to the Commission or Licensing Board to determine the scope of a hearing under $50.12.

8. The positive side to this approach is that it might yigid a license eventually. The negative side is that it would require that the applicants .

carry a burden of proof on three fronts (realism, exemption, and utility plan), it would require the Staff to shoulder the burden of litigating on three fronts, and it might undercut the rationale for an exemption (if the

/ grant of the exemption is as safe as the utility will probably have to argue, why make the exemption temporary; if the exemption is not that safe, why should it be granted at all).

OFFSITE EMERGENCY PLANNING

1. Notification requirements are set out in Part 50, Appendix E, Section

, D, and in 550.47(b)(5). The. regulations are not clear on whether the governor of one state may perform the notification function in another state. However, assuming that the governor can physiqally and *2nder state laws provide notification to the citizens of another state, there should not be a problem here because ISO.47(c)(1) provides applicants i

j with an opportunity to demonstrate that plan deficiencies are not significant or that adequate compensating measures have been taken.

Under the facts assumed in tys question, an applicant should have no trouble in satisfying $50.47(c)(1)'.

2., Under 44 CFR 5350.7, only a State may request formal FEMA review of radiological emergency respons plans . FEMA's regulations do not appear to allow a local government to bypass the State and submit its plan directly to FEMA. Ilowever, FEMA's M.O.U. with the NRC allows the NRC to submit plans to FEMA for its review; presumably, a local <

1 government could submit its plan to NRC and NRC could request that

' FEMA review the plan (this was the mechanism through which the LILCO ,

plan was reviewed for Shoreham).

3. Although a utility could probably not submit a plan

• directly to FEMA (see the answer to # 2 above), the, NRC could. This is what happened at Shoreham; the NRC esked FEMA to review the LILCO plan and FEMA complied.

! 4. Yes.

! can prove Again, this may result in some literal deficiencies, but if you that the deputized citizens can adequately perform their emergency response functions, the reasonable assurance finding could be made.

I

- 4_ t I

t

5. Same answer as to # 4 above. Legal authority and factual proof f would be problems, but if these hurdles are passed, a reasonable l assurance finding could be made.

I

+

e

• t 6

'9

~l.

o G

e 8

f-i e

f s.

O e

9

,--w. - , - - , , - - , . , , - - - -----,-m---w.- , , - - - - - - - ~ - - - -,m-e. - -,

1 SEABROOK BRIEFING

1. Issues to be resolved before the issuance of a low power license:  ;

\

a. Envira=---tal analification of acuinment (the applicants have not I cctisfied the reqdirements of GDC ,4 that all equipment important to safety be  !

cnvironmentally qualified because the Applicants have failed to specify the time duration over which the equipment was qualified)

b. Emergency classification scheme (the applicant's emergency plans do l not contain an adequate emergency.. classification scheme as required by 10 CFR 50.47.)( emergency action levels are: notification of unusual event, alert, cite area emergency and general emergency)

c. . Control room design (the'Seabrook Station control room design does not comply with general design criteria 19 through 22. Specifically is the dssign adequate -- is sufficient information displayed and is the information adequately displayed from a human factors standpoint.)

(The Board has now entered an order setting thes'e matters down for hearing secetime.between September 29 and October 10, 1986;.the hearing should last cbout four days. To date, no party has expressed an interest in litigating emargency c14ssifications. At present, the staff. review of the control room ic not complete, but should be finished within about two weeks. Applibants h:ve also filed a motion seeking summary disposition of this issue.)

2. Tentative request for an exemption from thu 10 mile EPZ requirement.

' ~

(The applicant has presented a PRA to the staff for review which could nupport an exemption request. The staff's review should take about 3 months. ,

) -

3. Offsite emergency planning. (New Hampshire is now redrafting its off site plon -- this should be complete by August 25 1986. All hearinga have been cancelled at the request of FEMA and the other parties. The Board has not set a new hearing date for offsite emergency planning, but FEMA had asked that the h9arings start on October 20, 1986. Also pending before the Board are Applicant's motion for sammary disposition of approximately 20 of the 35 cdmitted offsite emergency planning.)

4. There are..pover,al motions pending before the Board requesting that the Board not authorize the issuance of a low power license in light of the uncertainty now surrounding offsite emergency planning. We have opposed all ,

motions. ,

.- _ - - - _ _ _ _ = _ _ _ - _ _ - - _ _ - . - - . _ _ - _ - . _ _ __

.. 9. .

b H Yf W$ $Y0 h [ h y) &[Np f G ,

by /SN#.

.a L t 1;, ,w D1,'s wh,; I ~-n s In% 1aLu.#eA f , ,..!!;

du hc c>kOC $ t h , r] e ,A c lcif Ls tt i C fe v" 5 p

(?SNH u the All J 19tc.

c fe r/f- ? ? 7 N/7

- =* ,

ABOUT THE RESPECTIVE ROLES OF RMEPS AND THE SENSITIVITY STUDY A concern has developed about the relative amount of emphasis being

{ given by the NRC Staff and their contractors to the two reports that were

~

submitted which address the risk ef fects of energency planning. These two reports are The Risk Management and Emergency Planning Study (RMEPS) and The Emergency Planning Sensitivity Study (SSEPSS). We first report reflects the best and most complete picture currently available on the risk levels assessed for Seabrook Station and the most accurate determination we can make of the extent to which these risk levels can be reduced through evacuation and sheltering. The second report is only a sensitivity study that was specifically designed to isolate one and only one important factor: source terms. The

• purpose of performing the sensitivity study was to show that the conclusions

}

of RMEPS are robust with respect to source term uncertainties. '

, 'k SM o ivity study do not corres nd with The numerical results the sen

) either a realistic or conservative ssessment of risk levels. I RMEPS the g

Seabrook PSA team gave betueen 1 -20% weight to a conservative et of source

' ter# which are comparible ;o th se developed in BMI-2104 and 9 -80% weight

to source term comparible to theife, developed either specifical y for Seabrook 5 using MAAP or those for in the IDCOR program. The PSA te currently placeszero%ight of credibility on WASH-1400 Source Term Technology. [In 19 m the same PSA tednYlaced 2% wieght on these source terms, however, state of knowledge today strongly supports our position that this methodology greatly over states and distorts the current perspective on risk levels]. An overall comparison of the two studies is presented in Table 1. As seen in this table, RMEPS has a complete set of the attributes listed with one exception, whereas the sensitivity study, according to its design, has only 3. The only aspect of i the sensitivity study not included in RMEPS is the use of WASH-1800 Source Term Technology.

As stated by the NRC Staff- A-_. d ::::::i -: .' _ i ; :- f .---_-

'  ::: Q .;in :h::, they have been directed to base their revi on the results j of the sensitivity study. We understand that by so doing, re is some means a of gesolving the emergency planning issues for Seabrook St ion in a timely f ashion without pre-empting the degreaded core rulemakini process,which C ult'imately put the NRC blessing on revised source terms ilz-a-viythe ones used in WASH-1400. The ability to decouple the resolution of the immediate licensing issue for Seabrook Station from the more long term effort in the degraded core rulemaking process was in fact, invisioned as a primary motivation for performing the sensitivity study. However, there is a concern that a pre-occupation with this study at the expense of the RMEPS study will lead to a ,

number of problems not the least of which is a distorted picture of risk levels and a misleading perspective for judging the importance of any unresolved issues.

In view of the above considerations, the following conclusions are reached.

$. f ecisions 38e by NHY to improve plant safety

1. Any c re bt, should be* based on the risk perspectives provided by RMEPS rather than the sensitivity study.

i l

i - -_ - - - . ____ - __

ABOUT THE RESPECTIVE ROLES OF RMEPS AND THE SENSITIVITY STUDY

) e **.Y

2. While it is recognized that the NRC Staf cannot pre-empt ,

the degraded core rulemaking process b passing Janeist-j udgement in either the conservative r realistic source i terms used in RMEPS, the staf f ;' - _ _ ' :; recognise RMEPS on the risk benefits as the bestatand of evacuation mostStati Seabrook complete -(--

currently available.

/*! 5 J

3. Any additional sensitivity studies should be made with respect to the base case in RMEPS and not simply " stacked on top of" the assumptions built into the sensitivity 0 study. Such stacking of sensitivities would be arbitrary, wit hout technical scrit, and not particularly relevant '

to the true safety characteristic of Seabrook Station.

ie I

i

l TABLE 1 ATTRIBUTES RMEPS SSEPSS 6

f h'

Includes Assessment of Yes No Full Actual Risk Levels Quantifies Risk With Best Estimate Assumptions on

- Plant Event Sequences Yes Yes ,

Frequencies

- Source Teras Yes No

- Site Model Yes Yes L

i f Quantifies Risk With

. Conservative and Credible Assumption on

{

- Plant Event Sequence Yes No Frequencies

- Source Tere Yes No

- Site Model Yes No Yes No Quantifies Uncertanities Includes Analyses Based on Souce Terms From IDCOR (MAAP) Yes No

' - BMI 2104 Yes No h --

WASH-1400 ( rr ) No* Yes

/

t-O R A A L

• So M of the RMEPS source terms were developed from WASH-1400 methodology and then " corrected" or modified to account for more realistic source term -

assumptions.

t

j- 1.s

.g llr ) p,w f 4 - 14 s s A

., gp l$ s :. ..

g gf." g Io The NRC is concerned that containment leakage may be greater than O1.' Technical Specification limits (in the event of a design base acci-dent) because of penetrations being inadvertently lef t open to atmosphere.

o Technical Specification 4.6.1.1.a requires verification that al I penetrations not capable of being closed by operable containment automatic isolation valves and required to be closed during accident conditions are closed by valves, blind flanges, or de-activated automatic valves secured in their positions. The verification must be performed at least once per 31 days except for valves, blind flanges, or deactivated automatic valves lo-cated inside containment and which are locked, sealed, or other-wise secured in the closed position. These penetrations must be verified closed during each cold shutdown not more of ten than once per 92 days. Note that the Technical Specification is en- .

titled containment integrity and the LCO is applicable in tiodes 1, 2, 3, and 4.

o In addition to the above surveillance requirements, Technical Specification.3.6.3 requires that with one or more containment isolation valves inoperable (for maintenance, repair, replace-ment, etc.), at least one isolation valve must be operable in each affected penetration that is open ands

a. restore the inoperable valve (s) to operable status within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />; or

b. isolate each af fected penetration within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> by use of at least one deactivated automatic valve secured in the isolated position; or

c. isolate each af fected penetration within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> by use of at least one closed manual valve or blind flange; or

, d. be in at least hot standby within the next 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and in cold shutdown within the f ollowing 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />.

[o Technical Specification 3.6.1.3 requires that both doors for each .

f containment airlock be closed except when the airlock is being used for normal transit entry and exit through the containment, then at least one airlock door must be closed. The LCO for this Technical Specification is applicable in Modes 1, 2, 3, and 4.

We contend that the above Technical Specification requirements pro-vide adequate controls to ensure the following:

1. Containment penetrations required to be isolated in the event of an accident will not be inadvertently left open.

Fo16-97-7 C/7

1. l

. l l

2. In the event a non-automatic penetration isolation device (e.g.,

manual valve, blind flange, or reactivated automatic valve) must be opened when containme.nt integrity is required (Modes I, 2,1, and 4), adequate provisions are implemented to ensure that the penetration will be isolated should an accident occur.

& THE F~[h3If6)LJ7~,' QF. Q:> s 71 50 0 J.S c-%.3 rni s.)e,rA T T14M m > d i ic>c:-t 3 Cs !b Mr. 6(2 E/ALO R Tic)d . H 6(A7E9 C .S AS IF' u)E H AJ.E AI. C THE PP.OJ t Gio.JJ Tc> /WRf W TM/S /2[MiRLnddT Exc.QW TH C. f.>r + J 7 0 A<c3 G R A.rl. Tc>n d D. A JC e,iu (J.

MthtE 6410 D C y ll l - N / f.

d e

9

?

6 D

l

-( ) What do Technical Specifications require for containment pressure?

o Technical Specification 3.6.1.4 requires that primary containment internal pressure be maintained between 14.6 and 16.2 psia during Hodes 1, 2, 3, and 4.

( ) Would we be able to maintain Technical Specification secondary con-t ai nment requirements without primary containment integrity?

o No. A functional requirement of secondary. containment is to maintain a negative pressure of at least 0.25" W.C. in all secondary containment areas / cubicles. If a blind flange term-insting in secondary containment was lef t of f or a non-bypass penetration that does not have an ef fective water seal was not isolated following a LOCA, the regative pressure could not possibly be maintained in secondary containment.

( ) Do all blind flanges isolating containment penetrations terminate in secondary containment?

o No. A blind flange serves as a pressure boundary on the outside doors of the equipment and personnel watches. The flange is re-moved to leak test its respective airlock. However, a DCR is to be impicmented which will provide a tap line of f the flanged

, connection complete with its own isolation valve. Once this DCR j is implemented, the blind flange will no longer be removed to perf orm the leak test. At this time, there are no other blind flanges which serve as isolation devices located outside of containment.

e

{6) Do all penetrations without blind flanges have isolation valves, or valves and check valves?

o Ref erence Section 6.2.4 of the FSAR for a detailed description of the containment isolation systcm.

o In general, all piping lines penetrating containment which are not flanged of f are designed with redundant isolation valves (one inside containment and one outside containment). The valves are either automatic or locked closed. A check valve

/ is only deemed acceptable as an automatic valve when located f inside containment. Exceptions to this design are as follows.

i

- Lines which form closed systems inside containment and do not communicate with the reactor coolant system or containment atmosphere in accordance with CDC 57 are designed with one isolation valve located outside con-tainment.

- Certain small instrument lines are isolated by a sealed bellows arrangement located inside containment and a di aphragm arrangement out side containment. The lines are sealed fluid tubes.

d

. .- l 1

- - The normal RHR suction lines are not designed with an iso- )

lation valve outside containment.

- The inlet lines to the hydrogen analyzers are designed with 2 manual isolation valves. However, the isolation valve inside containment is locked open to ensure system avail-ability following a LOCA.

- Only one isolation valve is provided outside containment on the lines from the containment sump to the suction of the RHR and CBS pumps.

l .

i e

f a

9

/

i

TECHNICAL ASSOCIATES TECHNICAL CONSUL TANTS ON ENERGYO THE ENVIRONMENT Dale G Bradenbaugh 1723 Hamdton Avenue-Sw*e

• Richard B Hubbard San Jose. CaMornia 95*25 Gregory C Monor Phone (408) 266 2116 4

.M December 1986 FRf100M OF lHf0HMA160n Mr. Donnie H. Grimsley, Director ACT REQUEST Division of Rules and Records / 3 S' 7 Office of Administration r ~G f /

U.S. Nuclear Regulatory Commission j _$

Washington, D.C. 20555 /" #

RE: FOIA Request for Records Related to NRC Staff and Brookhaven National Laboratory Review of PLG-0432, PLG-0465, and NTS 1589.01

Dear Mr. Grimsley:

Pursuant to the Freedom of Information Act, please make available for public inspection and copying at the Commission's Washington, D.C . , Public Document Room, records in the following categories:

A. All records in the possession and/or control of the U.S. Nuclear Regulatory Commission which review, critique, comment upon, or otherwise discuss the Seabrook Station Risk Management and Emergency Planning Study (PLG-0432, December 1985), the Seabrook Station Emergency Planning Sensitivity Study (PLG-0465, April 1986), and Seismic Fragilities of Structures and Components at the Seabrook Generating Station, Units 1 and 2 (NTS 1589.01, Rev. 1, NTS Engineering, June 1986).

B. All records in the possession and/or control of the U.S. Nuclear Regulatory Commission which discuss the scope of work, funding, and manpower levels (including man-hour estimates and staffing levels) related to any reviews of the three named studies referenced in "A" above.

C. All records in the possession and/or control of the U.S. Nuclear Regulatory Commission which discuss the validity of the results and/or which estimate the uncertainty in the results of the three named studies referenced in "A" above.

qfgD

Z1 Mr. Donnie H. Grimsley ~2- JT December 1986 If there are any questions concerning this request, please contact me at (408) 266-2716. Your prompt attention to this request will be appreciated. If the nature of your document search is such that it appears that the 10-day response time provided for in the NRC's regulations cannot be met, please provide such partial responses as can be made as documents subject to this request become available for release.

7 Sincerely, b

f Steven

(.- C. Sholly Associate consultant i

d