ML20084K813

Testimony of Gf Daebeler,S Levine,Mi Goldman,Er Schmidt & Gd Kaiser Re Severe Accident Risk Contentions.Certificate of Svc Encl.Related Correspondence

ML20084K813
Person / Time
Site:	Limerick
Issue date:	05/11/1984
From:	Daebeler G, Goldman M, Kaiser G, Levine S, Schmidt E NUS CORP., PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC
To:
References
NUDOCS 8405140198
Download: ML20084K813 (84)

Text

%. '

RELATED CCRRES'PONomC UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION Before the Atomic Safety and Licensing Board DCCKETED U!iliRC In the Matter of )

Philadelphia Electric Company ) Docket Nos. 50-352 04 Ntl7 J J P2 :20 50-353

rr 7 cc (Limerick Generating Station, ) r;t - -7

~~3 Units 1 and 2)

TESTIMONY OF G.F. DAEBELER, S. LEVINE, M.I. GOLDMAN, E.R. SCHMIDT, AND G.D. KAISER, RELATING TO SEVERE ACCIDENT RISK CONTENTIONS OVERVIEW G.P. Daebeler 1. Applicant's evaluation of severe accidents to fulfill S. Levine the Nuclear Regulatory Commission's Statement of Interim E.R. Schmidt Policy on Severe Accidents (June 13, 1980: 45FR40101) is G.D. Kaiser contained in its Severe Accident Risk Assessment (SARA) for the Limerick Generating Station (LGS), which provides realistic estimates of public risk (SARA, p. 2-1) . The results of the NRC's consideration of the potential en-vironmental impacts of possible, but low frequency, accidents at the Limerick Generating Station are pre-sented in the NRC's Final Environmental Statement (FES),

NUREG-0974 (March 1984). Although the public impacts presented in NUREG-0974 are somewhat higher than those presented in SARA, the dif ferences between them lie within the range of uncertainties of such analyses. Thus the applicant agrees with the Staff's conclusion (FES p.

5-126) that potential accident risks from LGS are expected to be a small fraction of the risks the general public incurs from other sources.

1 8405140198 840511 PDR ADOCK 05000352 """

l 1)S03

G.F. Daebe10r 2. In tddrc3cing tha contintionn of Lincrick Ecology Action S. Levine (" LEA") and the City of Philadelphia (" City") , which are E.R. Schmidt discussed below, Applicant has used the models and s

G.D. Kaiser results developed in SARS, together with some additional sensivity studies which have been carried out using the SARA models. This is a reasonable approach because, for the purposes involved here, the differences in the results of the analyses of Applicant and NRC Staff are not significant enough to affect the conclusions regarding the contentions. Thus, conclusions drawn from SARA and these sensitivity studies can be used to confirm the adequacy of the NRC's FES as challenged by the contentions of LEA and the City.

REVIEW OF CONSEQUENCE ANALYSIS G.F. Daebeler 3. Since the contentions primarily involve the consequence S. Levine analysis portion of a risk assessment, it is worthwhile E.R. Schmidt to initially establish an understanding of consequence G.D. Kaiser analysis and how it is generally accomplished.

Consequence analysis is the determination of the impacts on public health and offsite costs due to the accidental release of radioactivity from a nuclear plant. Inputs to the consequence analysis are the frequencies of accident sequences (as determined by a plant specific systems analysis), and the magnitude and the timing of accidental releases of radionuclides associated with each accident 2

requsnca (es dstsreined by contcinment, tecidsnt procosa e and source term analyses). The consequence analysis .

includes the calculation of the environmental transport of' radionuclides, the calculation of radiation doses (taking into account the effect of various emergency response actions) and the determination of the resulting health effects and economic consequences.

1

- G.F. Daebeler 4. CRAC2 is the computer code used in LGS-SARA to perform l

S. Levine the consequence analysis. The environmental transport, E.R. Schmidt radiation exposure and health effects, and economic con-i.

I G.D. Kaiser sequence analyses have been performed by-CRAC2 for a r i series of meteorological sequences selected from site l I

specific meteorological records for a period of five l years. The frequency of the radionuclide release is l I

combined with the probability of the various l L

meteorological sequences to produce a frequency [

distribution for the predicted consequences. This  ;

distribution is discussed further in Paragraph 11.

f I

S. Levine 5. CRAC2 performs its calculations utilizing a grid, which i f

E.R. Schmidt is divided into sectors corresponding to the various wind .

G.D. Kaiser directions and into convenient radial rings. Downwind l 2 .

atmospheric and ground deposition concentrations of the radionuclides are calculated within each element of the grid for each meteorological sequence. The code assumes  !

l that the wind direction is invariant once the release has I

[

occurred; however, the effects of changes in wind speed, ,

i i

+

3 ,

i  ;

- L r

L

- - - - - - --n-..,,-,-- ,-n. . , . .,w.--m---,- . . . . , , - - - - - ,-.,,n-,- --,~.n,. , - , - - - . - .,n- .. , , .--,-.n-.

j i

I weather stability and rainfall as a function of time are  !

considered.

l S. Levine 6. The code calculates the accumulated radiation dose for i E.R. Schiedt people in the region around the plant after exposure to l

G.D. Kaiser both the passing plume and to radioactive material deposited on the ground. The population exposed is i

determined by the population in each element of the grid l

representing the region around the plant out to a 9 distance of 500 miles. The normal population in the i

region affected is modified by emergency response f i actions. Both immediate and longer term emergency '

response actions are modeled in the CRAC2 computer j i

, program. Immediate actions such as evacuation or a  !

combination of evacuation and sheltering of the people living near the site are assumed to be taken to reduce I early exposure to the passing radioactive cloud and

fission products deposited on the ground. Actions are ,

also assumed to be taken to reduce the longer term [

accumulation of radiation dose due to exposure to radioactive material deposited on the ground or the  !

ingestion of contaminated food. I t

S. Levine 7. The evacuation and sheltering model employed in CRAC2 is i G.D. Kaiser a revised implementation of the CRAC model developed for ,

use in the Reactor Safety Study (Ref. 1, Appl. Exh. ).

The revised model was developed for the NRC by Sandia ,

Corporation personnel (Ref. 2, Appl. Exh. ) and is .

[

4 I i t a 1

intended to provide a more realistic description of the +

public risks involved and more realistic modeling of l their modification through emergency response strategies  ;

i

^

than existed in the CRAC Model used in the Reactor Safety

. Study. The CRAC2 evacuation model includes three basic' l

zones of emergency response: 1) an evacuation zone i (nearest the site), 2) a zone beyond the evacuation zone, sometimes referred to as a "special sheltering zone," in i which people may be instructed to remain indoors, or may be assumed to continue their normal daily activities until directed to relocate, and 3) a zone beyond the evacuation and special sheltering zones in which people are assumed to remain unless predictions of accumulated radiation dose exceed prescribed limits, in which case t they are modeled to be relocated after some period of time.  ;

S. Levine 8. The timing of emergency response activities, that is the E.R. Schmidt time when actual movement of evacuees begins relative to G.D. Kaiser the release of radioactive material to the atmosphere, is divided into two, components which are inputs to the CRAC2 codes

a. A warning time, which is defined as the interval of time from the declaration of a Site General Emergency to the time when radionculides first emerge into the i

environment. The time of declaration of a General Emergency is defined by the plant's Emergency Plan 5

l 1

4 0

,.-,w -,,--,----w--,-----..+- , , , ,--.,.--,.---v..- .-- - c - - . ---

Implementing Procedures. The warning time is a function of the physical processes of the accident sequences with contribute to the source term.

b. The evacuation delay time which is defined as the time between notification of off-site agencies and the ,

beginning of evacuation (that is movement of people). '

S. Levine 9. Once the evacuation begins, the CRAC2 model evacuees are t

G.D. Kaiser assumed to physically move radially out of the area at a constant speed. The emergency response model assembles estimates of the radiation dose accumulated by evacuees during each phase of the emergency response; i.e., prior 7

- i to actual movement and during the actual evacuation  ;

process, includes assessments of dose for people that are 9

given special sheltering instructions, and finally y estimates dose for people outside the special sheltering

, Zone.

S. Levine 10. The CRAC2 computer model tracks the transport and deposi-G.D. Kaiser . tion of airborne releases over great distances from the plant. The predicted ground contamination levels are combined with ingestion pathway models within the CRAC2

' code to derive estimates of the extent of milk and crop l contamination, the requirements for the decontamination of land and structures, and the potential need to relocate people for extended periods. Using dose 6

criteria for the ingestion pathways, the chronic dose models of CRAC2 estimate possible interdiction requirements for agricultural products. The economic impacts of the interdiction of foodstuffs and the denial of land use are estimated using specific land use information for the area around the Limerick Generating Station.

PRESENTATION OF RESULTS OF CONSEQUENCE ANALYSIS G.F. Daebeler 11. The results of the offsite consequence analysis are S. Levine primarily presented in the form of complementary cumula-E.R. Schmidt tive distribution functions (CCDFs) for measures of the G.D. Kaiser various health effects--such as early fatalities, early injuries, latent cancer-fatalities--and measures of the offsite costs of accidents. CCDFs are plots of the frequency with which a consequence of a given magnitude is equaled or exceeded. Examples may be found in, for example, Figures 1 through 16 of Suplement 3 of SARA. In order to accomodate the wide range of frequencies and

. consequences predicted, logarithmic scales are generally-used for both axes. The CCDFs themselves can be used to depict risk; alternatively, the area under the CCDF, which represents the expected value, may also be used as a simpler depiction of risk. The area under the CCDP has a relatively simple interpretation; for example, for early fatalities its inverse is the average predicted 7

c.

L L

interval between the occurrence of an early fatality in [

the population surrounding the LGS due to accidental

' releases of radioactive material to the atmosphere.

r G.F. Daebeler 12. As noted in both the FES and SARA, the numerical results 4

S.-Levine of a Probabilistic Risk Assessment (PRA) are subject to E.R. Schmidt uncertainty. The uncertainty arises for a variety of ,

G.D. Kaiser reasons: ' estimates of the likelihood of the accident sequences leading to release of radioactivity are uncertain because the estimates of the parameter values l for the models used to evaluate these likelihoods are  !

uncertain; estimates of the consequences of a release are uncertain because of uncertainties in the modeling of the ,

release and its environmental consequences. These fundamental uncertainties arige from both an inadequate data base associated with the rare event nature of reactor accidents, and from uncertainties in the modeling i of the physical progression of accidents. Because of f

these and other uncertainties, SARA presents results, not as a single CCDF,, but as a family of CCDFs. A range of results, including a lower and an upper estimate are presented. The lower and upper estimates are not ,

absolute bounds but define the range in which there is a large degree of assurance that the actual result would lie. The uncertainty characterized by these different estimates takes into account the uncertainty in frequencies of accident sequences as well as uncertainty in consequence magnitudes.

8  ;

G.F. Daebeler 13. The uncertainties in the results of a PRA are large. It S. Levine is stated in the FES that the risk estimates could be E.R. Schmidt "too low by a factor of 40 or too high by a factor of G.D. Kaiser 400." Typically, the area under the upper estimate CCDFs in SARA are on the order of a factor of one hundred greater than the area under the lower estimate CCDFs.

Any comparison of the results of sensitivity studies, or of other PRAs must be made with this large range of uncertainty in mind. If the uncertainty ranges of two estimates are large and overlap to a large extent then the two results cannot be regarded as being significantly different. Thus, for instance, changes of a factor of I two in estimates of public risk are insignificant in view  ;

of the large range of uncertainty.

P i

a f

t t

a i

i i

k 9

g - - - - - - . - - - . . , , . . - . . . . , _ ,n.- ,, -

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

CONTENTION DES-1 (LEA)

"The DES' severe accident consequence modeling I assumes the relocation of the public from con-taminated areas beyond the 10 mile plume exposure EPZ. (DES, Supp. 1, pp 5-21 to 5-22) . Such an assumption in Limerick's case is implausible and without foundation in fact."

G.F. Daebeler 14. This testimony addresses the possibility and need for S. Levine protective actions such as the relocation of people E.R. Schmidt beyond the designated plume exposure Emergency Planning G.D. Kaiser Zone (EPZ) . For the purpose of this testimony, the EPZ is defined as a circle with a radius of ten miles, centered on the Limerick Generating Station (LGS). The l following will be discussed; a) prior NRC guidance with regard to such actions and the manner in which they are taken into account in SARAp b) the results of sensitivity studies for different relocation and sheltering assump-tions; and c) a discussion about previous experience with t

evacuation of large numbers of people. Based upon these I

considerations it is concluded that the assumptions made in SARA about the treatment of individuals outside the EPZ are reasonable and appropriate, and further that the FES assumptions are equally reasonable.  !

G.F. Daebeler 15. In its guidance on emergency planning, the NRC has S. Levine discussed the possibility of protective actions for E.R. Schmidt persons residing beyond the EPZ. For example, G.D. Kaiser NUREG-0396, " Planning Basis for the Development of State 10

i and Local Government Radiological Emergency Response Plans in Support of Light Water Nuclear Power Plants"

.(Ref. 3, Appl. Exh. ) states on page 16 that, for distances exceeding 10 miles, " actions could be taken on an ad hoc basis using the same considerations that went into the initial action determinations." Fur thermore, NUREG-0654 " Criteria for Preparation of Emergency Response Plan and Preparedness in Support of Nuclear Power Plants" (Ref. 4, Appl. Exh. ) states on page 9 I

that " detailed planning within 10 miles would provide a substantial base for expansion of response efforts in the f event that this proved necessary." Thus, an assumption [

that individuals living beyond the EPZ could be relocated is not unreasonable.

G.F. Daebeler 16. If there should be a significant accidental release of S. Levine radioactive, materials from the Limerick Generating E.R. Schmidt Station, the radioactive plume and any radioactive G.D. Kaiser materials deposited on the ground would be closely monitored. Over a period of time, there could be areas f within which radiation doses arising from exposure to gamma rays emitted by these airborne and deposited l fission products ("cloudshine" and "groundshine"  ;

respectively), together with the doses arising from the d

inhalation of airborne fission products, could attain i

levels that are sufficient to cause early health effects.

l These areas would be defined by measurement and would be 11 f i

G finite in extent. Experience with the CRAC2 code, which calculates the consequences of the accidental release of ,

radioactive material to the environment, shows that, for most source terms, the probability of occurrence of early health effects among the population outside the' EPZ is calculated to be very small. For larger, lower probability source terms, the initial radiation dose arising from cloudshine and the inhalation of fission products, together with that arising from groundshine, would result in early health effects beyond the EPZ during a period of many hours after passage of the t radioactive plume. The responsible authorities would be closely monitoring the plume and fission products deposited on the ground and would be aware that there would be areas within which individuals would be in danger of receiving radiation doses that could cause early health effects. In order to simulate ad hoc measures that would be taken to protect people living in such areas, it was assumed in SARA that people beyond the EPZ, in the special sheltering zone that is described in Paragraph 7 above and which extends from 10 to 25 miles, would continue their normal activities for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> af ter the passage of the plume and then be rapidly relocated.

This assumption, used in SARA, is more conservative than that used by, for example the authors of " Examination of Offsite Emergency Protective Measures for Core Melt Accidents" (Ref. 5,, Appl. Exh. ) who assume 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> of ,

I 12

sheltering in basements, followed by rapid relocation.

Six hours was chosen by these authors as "a practical lower limit for effective exposure time." It is  !

pertinent to note that, in this context , " normal activities" means thatfa shielding factor of 0.3, i

characteristic of that for people who spend most of their  :

time indoors, is used for groundshine (see SARA p.10-12). A shielding factor is simply the ratio between the radiation dose that is accumulated by a person to that i

which could be accumulated by a person standing in the  !

9 open on a flat surface. It thus represents the ,

effectiveness of surface features or structures in [

attenuating gamma rays. Thus, the assumption of " normal activity" during the period of 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> discussed above means no more than that people spend most of their time indoors. " Rapid relocation" means moving away quickly enough so that there is only a small increment in the radiation dose received while traveling.

4 G.F. Daebeler 17. It should be stressed that this assumption of normal S. Levine activity followed by rapid relocation is not the only E.R. Schmidt course of action open to the authorities which could lead G.D. Kaiser to comparable or greater dose reduction factors. For example, asking people to shelter in their basements for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, followed by rapid relocation, would also result in a reduction in radiation dose. Sheltering in schools or large buildings would afford even greater protection, i i

13

. - < - - - .-..,.-,_--.--.,m-_,, , , , , . 4 ,-, --- - - - ---,v,.1--------r ---w

, n ,--,w,,,,--- < wy r-,,,w,,.

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

I

• k L

see SARA Table 10-4. Thus, the assumption of normal activity for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> followed by a rapid relocation, is ,

merely a calculational convenience that simulates the re-duction in radiation dose which could be achieved by a i i number of possible options open to the authorities.

I 4

S. Levine 18. In order to illustrate this point and to determine the l G.D. Kaiser effects of alternative modeling assumptions concerning t shielding and relocation, a series of sensitivity studies

, has been carried out using CRAC2. Estimates of the f public risk of early fatality (internal initiating events  ;

only) have been calculated for several different cases as i

?

shown in Table 1. In each of these cases, people within 10 miles are assumed to participate in the base case j evacuation model as summarized in SARA, Tables 10-8 and i 10-9, and people beyond 25 miles continue their normal activities and then relocate 7 days after the passage of t

the plume, unless the predicted radiation dose to the j bone marrow from groundshine exceeds 200 rem over 7 days,  ;

i in which case they relocate after one day. The results given in Table 1 are areas under complementary cumulative I distribution functions (CCDrs), as defined in Paragraph 11. From Table 1 it may be seen that the l

results are insensitive (within a factor of two or less)

I to a variety of assumptions. The last case in the table  ;

I is very conservative because it assumes that people l between 15 and 25 miles are not relocated for 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.

e L

14 f

B

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

.\-

.o

\Q7 .

,r-

-- ', [ ' s Ii '% g - si \ t ,

i t.

r

, ,- G.F. Daebeler 19. 'It is also ps:tinent to note that, although SARA p. 10-11 s sO- s

n. % .

' yx U4w

- , 0 . IS. Levine can[oettakevito imply that everybody within the special E.R. Schmidt shelteCaTJ;.ne relocates at the same time, say, 1 g, ,;\% , 3 -

d YN.D. G Kaiser >$2 hourd, this is merely a calculational convenience.

'yi~ ,

,i s

p e\L

• s,. 'he same, .public risk of early fatality would be predicted r g 1, w s

b- tv .

, ;T '

if, fo't'eimnplet it were, assumed that only people within it  ; contaminated hot spots' were relocated within 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />,

\ i

. with phased withdrawal of people in other, less highly w,

e'. contaminated areas in which doses build up more slowly

'\., + 1,.-

S g' : t towards thresholds for early effects. Thus, although the

~

-(f, 's,~

, ,\ s . SARA ard ,tbe FES models relocate large numbers of people

• '- .. L " '

,t i y, .

,,%, e .

5' s " at the sE3c time, it should be stressed that this is s g \ '

merely\.N, a calculational i 1 ,

convenience and in no way implies ,

<. 3 i that this is what wo'tild' actually happen, or that it would s, ,s

? , .

t. . -

b'e necessary. co avoid early health effects, in the highly i

<% cs i 3 ,

i 5 '

g [ improbable event pf a large, accidental release of [

s ..

ci s hg',radioactive material from' LGS.v s t

4 \- s,, .%

y s

~'

[

'N ,T G.W. Daebeler 20. The existence of lar e numbers of twople in the "SE" i g"s 3 '\ , i i \

(

ihe S) vine' 'and "ESE" sectors'between 10 ;a'nd ,

25 miles from LGS does

s. s . g

\ not present unprecedented pecblems for relocating ht.R. ::c' Eidt

,A s q( 6,.D.

?3 Kaiser people as asserted ia the contention.

s. >

1. - )i ,q k).\; '

o  ; .

C.F. Daebeler 21. First, it s, extremely unlikely that it would be i G.D. Kaiser I nec3ssar;,' t'o relocate all of these people. Of the j

g p' '

s k t

, 689, W pecple' projected to be ir Sector SE, between

\

(

3 ,, s % , . I

, . g ,

A- \  !

i  ! i 15 ss ,

y ,

---

r- c- - .W*W--,-,-----.  % .m, . - . - - , - - - - - - -..-.,-w--....-e, . - , . -

10 and 25 miles, only 117,000 are within 20 miles, and of the 505,011 people projected to be in Sector ESE between A

10 and 25 miles, only 169,000 are within 20 miles.

(These are projections for the year 2000, see SARA

p. 10-33). Even if no protective actions are taken for 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />, the likelihood that there will be persons in Sectors SE or ESE beyond 20 miles receiving large doses f

?

(interpreted here as a 200 rem bone marrow dose j accumulated in 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />) is estimated to be approximately l l

1 in 750 million per year. The chance that there will be ,

i .

people beyond 20 miles in Sectors SE and ESE who would i

. accumulate potential radiation doses over 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> sufficient to lead to any clinically detectable early offect at all (25 rem whole body dose) is estimated to be approximately 1 in 16 million per year. Therefore, the ,

probability that relocation of these large numbers of people would be required in the time used in SARA and in the FES (12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />) is very small indeed. i i

, S. Levine 22. Second, evacuation of large numbers of people have in l G.D. Kaiser fact taken place quite expeditiously, see " Evacuation -

i Risks--An Evaluation" by Hans and Sell (Ref. 6, Appl.

Exh. ), page 42 and Appendix B. Baton Rouge, Louisiana, I population 150,000, was almost totally evacuated in two  ;

s hours after a decision was made to evacuate the city fol-  :

lowing an accident involving a chlorine barge. Wilkes 1 16

" j., J n e t

Barre, Pennsylvania, population 75,000, was efficiently e'sacuated to a level of 96% in one hour because of a ,

t

, t flood warning. Downtown Portland, Oregon, with a i

population of 100,000 was evacuated in one hour during a civil defense test exercise. One of the largest recent

.r public evacuations occurred in Canada. Late in the eve'ning of November 10, 1979, a freight train transporting both flammable and toxic raterials derailed in downtown Mississauga, Ontario, Canada's ninth largest  ;

i city. During the next 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, 216,000 people were -

t evacuated from homes and hospitals in a 50 square mile ,

area around the accident site. See "Mississauga Evacuates: A Report on the Closing of Canada's Ninth Largest City", (Ref. 7, Appl. Exh. ). The contention is j therefore incorrect in its assertion that there is no ,

L precedent for the ad hoc relocation of large numbers of l P

people.

G.F. Daebeler 23. It is therefore concluded that the SARA treatment of ,

S. Levine population relocation outside the 10 mile plume exposure E.R. Schmidt emergency planning zone, as shown by the additional cases G.D. Kaiser ' calculated and reported above, presents a reasonable  !

estimate of the risks to this population. The relocation of sizable populations is not unprecedented, and there is ample time to implement plausible protective actions for the very unlikely cases where such actions would be con-sidered appropriate by the responsible authorities. ,

17 ,

CONTENTION DES-2 (LEA) and CITY-14A DES-2 (LEA) ,

"The DES' severe accident consequence modeling uses an assumption of a uniform two-hour evacua-tion delay time in its emergency response model.

(DES, Supp.1, pp 5-21 to 5-22) . This assumption understates the likely delay time for a high population density site such as Limerick. This understatement of delay time results in an understatement of Limerick's risk, because accident consequence calculations are sensitive to evacuation time delay assumptions.

CITY-14A "The DES does not accurately reflect either the  :

median or upper estimates of the radiological effects which could result from an accident at Limerick because several key input assumptions  !

associated with human activity after a severe accident are not realistic. t i

(a) The base case average evacuation time of l 2.5 mph is based on an 1980 study which is ,

now inaccurate. See also Statement of

• Issues of the Commonwealth of Pennsylvania with Respect to Offsite Emergency i Planning, January 30, 1984."

G.F. Daebeler 24. This testimony first examines contention DES-2's implied S. Levine assertion that, the higher the population density, the E.R. Schmidt greater the delay time, and shows that the available data G.D. Kaiser do not support this view. Second, the basis for the delay ,

l l

• times and evacuation speeds utilized in SARA are discussed and shown to be appropriate. However, this testimony does not solely rely on the above. It is also demonstra-ted that the results obtained using the SARA evacuation f

l model are relatively close to those presented in the FES.

i 18 l

Therefore, the FES assumptions of a two hour delay and a 2.5 mph evacuation speed produce reasonable results.

Furthermore, the effects of varying the evacuation assumptions utilized in SARA are presented. The results show that the sensitivity of the risk to plausible changes in evacuation assumptions in the pessimistic direction are relatively small.

i S. Levine 25. In order to clarify the following discussion, it is E.R. Schmidt pertinent to note that the total evacuation time, G.D. Kaiser measured from the declaration of a General Emergency, is made up of several components. CRAC2 divides it into a delay time, followed by a travel time which depends on a i

constant effective evacuation speed. This effective evacuation speed is an average over the evacuees' complete journey and does not necessarily represent the actual speed in any part of the evacuation route. Both of these components affect the result and to a certain i

l extent a shorter delay time can be compensated for by a l

slower speed.

I S. Levine 26. Contention DES-2 states that the likely delay time is E.R. Schmidt understated because of the higher population density G.D. Kaiser around LGS. However, data from actual evacuations presented in the Hans and Sell report (Ref. 6, Appl.

Exh. ) and reproduced here as Figure 1 demonstrate that the time required to completely evacuate an affected area appears to decrease as the population density increases.

19 1

i

The reasons given for this by Hans and Sell are as follows:

1. The evacuation times reported include the time to warn the population, the time for them to prepare to evacuate, and the time required to move the >

population out of the affected areas. The time necessary to notify the population that evacuation is required may lengthen as the popu-lation density decreases because of increasing distances between persons and the fact that more individual contacts have to be made.

2. More time is required to prepare farms for evacuation than for residential dwellings.

3. Road networks generally decrease as the population density decreases; therefore, more time may be required for evacuation in rural areas because of limitations in the road network.

S. Levine 27. The above three points suggest that evacuation delay E.R. Schmidt times may decrease as population density increases, and G.D. Kaiser that the total evacuation time may decrease somewhat as population density increases. Cases examined by Hans and Sell did not have a prompt notification system similar to that required for nuclear power plants. Such systems 20

. . s

[

would decrease notification time and thus tend to minimize the effect of Item 1. Thus caution is necessary i

in extrapolating the Hans and Sell results to nuclear power plant sites. However, it is fair to draw the conclusion that there is no one to one correspondence between population density and delay time and/or total ,

evacuation time. Thus high population density per se is not a sufficient reason for increasing the delt.y time.

I S. Levine 28. In the basis for Contention DES-2 (LEA) a argument is i

E.R. Schmidt given suggesting that the Sandia Generic Study, (Ref 2, G.D. Kaiser Appl. Exh. ) which gives a mean delay time of three hours, implies a longer delay time than three hours for i

Limerick Generating Station, presumably because of its higher-than-average population density. However, as is shown in Figure 1, the Hans and Sell report (upon which the Sandia Generic Study is based) contains examples of evacuation from areas with population densities that greatly exceed that for LGS, which is about 700 persons l l

l per square mile within 10 miles. Thus, it would not be appropriate to increase the delay time simply because the population density around Limerick Generating Station is i i higher than for average nuclear plant sites without further justification.

i 21

_ _-_)

i G.F. Daebeler 29. The above argunents give general, qualitative reasons for [

S. Levine believing that the evacuation delay time does not

(

E.R. Schmidt necessarily increase as the population density increases. I G.D. Kaiser However, this testimony now continues with a demonstra-tion that, even if plausible increases in evacuation 9

delay times and/or clear times beyond those already  !

considered in SARA or the FES are made, there is no I significant increase in the predicted public risk. The clear time is simply the delay time plus the time to l

[

evacuate the EPZ at the average evacuation speed.

G.D. Kaiser 30. In order to examine the effects of changes in delay times and evacuation speeds on the final risk results, sensitivity analyses have been performed using various models and various values for the delay time and evacuation speed parameters. These studies used the CRAC2 code and made use of the SARA source terms (SARA, Tables 12-7 and 12-8). Initially the Sandia (SARA)

! evacuation model, was compared with the FES " Evac-reloc" model. The SARA evacuation model is summarized in SARA Tables 10-8 and 10-9 and incorporates the results of the Sandia Study (Ref. 2, Appl. Exh. ) explicitly with delay l times weighted as follows; 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> (30%); 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> (40%);

and 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> (30%) . The FES Evac-Reloc model (FES, Table 5.11f) makes use of a single two hour delay time with an effective evacuation speed of 2.5 mph. The results of these. calculations are summarized as estimates of 22

e the public risk of early latality (area under CCDF. see paragraph 11 above) in Table 2, rows 1 and 2. It can be i

seen that the FES model produces risk estimates that do not differ greatly from those in the Sandia model, even though the delay times and evacuation speeds are different in the two models.

S. Levine 31. Table 2 also contains examples of a number of sensitivity '

E.R. Schmidt studies in which evacuation clear times vary from 4 G.D. Kaiser through 13 hours1.50463e-4 days <br />0.00361 hours <br />2.149471e-5 weeks <br />4.9465e-6 months <br /> (Cases 4 through 8) and case 3 in which the SARA delay times (1, 3 and 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />) are combined with a 2.5 mph evacuation speed. None of the results varies from that of the SARA base case by more than about a  ;

factor of 2 and they are all within a factor of 3 of the result for the FES evac-reloc model. These are small changes when compared wth the range of uncertainties that apply to estimates of public risk, see paragraph 12. It is concluded that, within the range of delay times and evacuation speeds covered in Table 1, the predictions of public risk do not differ significantly when the evacuation speed is varied from 2.5 to 10 mph. Thus, Contention CITY 14A part (a) does not present arguments that would lead to significant changes in results. It is ,

1 also concluded that the use of a two hour delay time in the FES (Contention DES-2) does not lead to a significant understatment of Limerick's risk.

23

CONTENTION DES-3 (LEA)

"The DES' severe accident consequence modeling fails to account for the probability that a portion ,

of the population will fail to take protective action despite planning and instructions, thus understating the actual consequences of a severe  ;

accident at Limerick."

G.F. Daebeler 32. This testimony will show that the omission of a non-S. Levine evacuating fraction of the population surrounding the E.R. Schmidt plant from the consequence analysis has a relatively ,

G.D. Kaiser small effect on the results, well within the estimated uncertainties. The incorporation of this omission in the analysis does not change the overall conclusions.

S. Levine 33. It was advanced in the basis for the contention that G.D. Kaiser an EPA evacuation study by Hans and Sell, " Evacuation Risks - An Evaluation" (Ref. 6, Appl. Exh. ) estimates that a fraction of the population ranging from six percent to fifty percent will not evacuate in response to instructions during an emergency. However, the attached l

l copy of page 48 of the Hans and Sell report ,

( (Attachment 1) makes it clear ..nat approximately six i

percent of the population of interest in the cases studied refused to evacuate. The fifty percent figure is from a separate report, quoted by Hans and Sell, by Harry r

E. Moore et al., "Before the Wind - A Study of the l l

Response to Hurricane Carla" (Ref. 8,, Appl. Exh. ).

This report is a study of the response to Hurricane Carla, which came ashore in Calhoun County, Texas f

24

Y on September 11, 1961. The Moore study considers the evacuation behavior of people not only in Calhoun Co>aty itself but also in Chambers County, Texas; the cities of a

Baytown and Galveston, which are approximately 100 miles '

to the northeast, and in Cameron Jarish, Louisiana, which  :

l' is about 200 miles from where the storm came ashore.

I G.F. Daebeler 34. The inclusion of people who live at such great distances  !

S. Levine from the eye of the hurricane suggests the need for }

E.R. Schmidt caution in extrapolating the observed evacuation per-G.D. Kaiser centages for Hurricane Carla to the case of a General i Emergency declared at Limerick Generating Station.

I Furthermore, a survey carried out af ter the hurricane i

revealed that 63 percent of the more than 1500 people i

questioned in the affected area were not at any time advised or ordered to evacuate (Ref. 8,, Appl, Exh. ,

i Table 2.14, p. 31) . Even in Calhoun County, which bore the full brunt of the storm, 64 percent of the re- l spondents stated that they were neither advised nor ordered to evacuate. It is therefore not at all I surprising that evacuation in this instance was far from l r

(

complete. Furthermore, since a prompt notification I system is required for nuclear plants, this case does not [

apply to a nuclear emergency. .

l I: ,

f 25 P

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

, , . . - - . , , - --,n - . ,.,-.,

G.F. Daebeler 35. Thus, it is concluded that population non-response S. Levine percentages as large as fifty percent, as observed in the E.R. Schmidt case of Hurricane Carla, are not applicable to postulated G.D. Kaiser emergencies at Limerick.

G.D. Kaiser 36. It is pertinent to note, that in Cameron Parish, Louisiana, evacuation was more than 96 percent effective (Ref. 8,, Appl. Exh. , Table 2.17, p. 33). This level of <

evacuation was possible in part because the same area had been severely affected by Hurricane Audrey four years earlier, and a publicized evacuation plan was in place.

In fairness, it should be noted that the effectiveness of the evacuation cannot be entirely attributed to the existence of a plan because nearly 40 percent of those questioned said that they were not aware of its existence (Ref. 8,, Appl. Exh. , Table 2.17, p. 33). However, the fact that a large percentage of the population was aware of the plan and therefore evacuated must surely have l encouraged the rest of the people to evacuate.

t S. Levine 37. As concluded above, fifty percent is too large a non-  ;

l >

( E.R. Schmidt evacuating fraction of the potentially affected popula-G.D. Kaiser tion to use in models for postulated LGS emergencies.

This conclusion can be reinforced by considering the basis for the Sandia generic evacuation model (Ref. 2, l

l l

Appl. Exh. ) which was specifically developed for I

I' nuclear power plant studies and was used in SARA. This 26

Sandia model' explicitly excludes natural disasters such as hurricanes as being inappropriate because of r

differences in warning times, evacuation movements and the character of the emergency. Transportation accidents were used to develop descriptive models for reactor ,

accidents. The description of the Sandia model states that Civil Defense personnel observed five percent as the fraction of nonparticipating people in actual evacuations (Ref. 2, Appl. Exh. , p. 13). Hans and Sell consider six percent to be an appropriate value (Ref. 6, Appl.

Exh. , p. 48). In the analyses reported below, six percent was selected as the fraction of nonparticipating people in the modeled evacuations.

S. Levine 38. The base case CRAC2 computer runs used to generate the E.R. Schmidt SARA results were modified to include a six percent non-i G.D. Kaiser participating fraction of the population in the evacua- i tion and relocation model. This means that 6% of the population out to 25 miles does not take the protective actions which the remainder of the affected population is assumed to take, i.e., evacuation within 10 miles of LGS or normal activities for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> after plume passage with subsequent relocation for people between 10 and 25 miles from the plant. The individuals who did not ,

participate were assumed to remain outdoors for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after the declaration of a General Emergency and then to rapidly relocate. This is the equivalent of exposures 27 ,

f

that would be accumulated in over two days of " normal activities" following plume passage. The predicted public risk of early fatalities as represented by the i

area under the CCDF (see paragraph 11) increases by 494, which for calculations of this type represents a small increase. This 49% applies to the sum of the contributions from internal and seismic initiators. This t

calculation demonstrates that conservatively including a non-evacuating fraction of the population would not be ,

expected to increase the public risk by a large amount, .

and that other uncertainties discussed in SARA, such as uncertainties in source terms, are much more significant.

4 9

+

t 28

CONTENTION DES-4A "The DES Supplement fails to adequately disclose or consider:

(1) Total latent health effects due to both initial and chronic radiation exposure, other than those resulting in fatalities, including genetic effects, non-fatal cancers, spontaneous abortions, and steril-ity (see, e.g. , BEIR I-III) ;

(2) The total land area in which crops will be interdicted; (3) The total land area in which milk will be interdicted; (6) The quantification of the cost of medical treatment of health effects; (8) The population within the land areas to be interdicted."

G.F. Daebeler 39. This testimony will show that (a) estimates of the public S. Levine risk of latent health effects other than those resulting M.I. Goldman in fatalities can readily be obtained from estimates of E.R. Schmidt risk that are already presented in SARA or the FES by the G.D. Kaiser use of simple multiplication factors; (b) several of the items listed in the contention--the total land area within which crops would be interdicted, the total land area 1

within which milk would be interdicted, and the population l

within interdicted land areas--have been explicitly con-sidered in both SARA and the FES in determining the off-site economic costs of reactor accidents; and (c) current estimates of the cost of medical treatment of health effects show that they do not have a large impact on the L total offsite costs of a reactor accident.

I I

l 29 l

ITEM (1) NON-FATAL LATENT HEALTH EFFECTS G.F. Daebeler 40. At the outset, it should be noted that SARA results (and S. Levine the FES as well) include consideration of initial and M.I. Goldman chronic radiation exposure in determining health effects.

E.R. Schmidt Additionally, latent health effects other than those G.D. Kaiser resulting in cancer fatalities are generally not included in the numerical results of risk assessments; however, they can be estimated from the available information.

These health effects include non-fatal cancers, genetic effects, spontaneous abortions and temporary or permanent sterility.

M.I. Goldman 41. In general, mortality rates from cancer expressed as frac-G.D. Kaiser tions of the incidence rates differ widely depending upon e

cancer type and site. For example, most thyroid cancers are well differentiated, slow growing, and treatable, and exhibit a mortality rate much lower than other forms of cancer. The Reactor Safety Study (RSS) assumed a 10 per-cent mortality rate for malignant thyroid cancers (Ref. l., Appl. Exh.

, p. 9-26) and this is what was used ,

in SARA. This is probably an overestimate as other studies suggest lower mortality rates. For example, NRPB M78, " HEALTH-MARC; the Health Effects Module in the Methodology for Assessing the Radiological Consequences of Accidental Releases" (Ref. 9,, Appl. Exh. , Table 1) suggests a 5% thyroid cancer mortality rate. The 1977 30 j

UNSCEAR Report, " Sources and Effects of Ionizing Radiation" (Ref. 10_, Appl. Exh. ) in Annex G, Paragraph 149, indicatec a 3% fatality risk over 25 years. (Note that adeption of either of these lower rates would reduce the number of fatal thyroid cancers '

predicted in SARA. It would not increase the number of non-fatal thyroid cancers predicted below.) A distribution of 9 non-fatal thyroid cancers for every fatal thyroid cancer would, therefore, seem reasonable based on the conservative fatality estimates of SARA.

For other cancers (excluding thyroid cancer) the  ;

mortality fraction is higher; approximately 50 percent of all observable soft tissue cancers would result in l fatality, see NRPB M78, Table 1 (Ref. 9, Appl. Exh. )

I and the BEIR-III report, "The Effects on Populations of Exposures to Low Levels of Ionizing Radiation" Table V-15 (Ref. 11, Appl. Exh. ). ,

M.I. Goldman 42. From the risk results of SARA, the point estimate public G.D. Kaiser risk of latent cancer fatality, excluding thyroid i

cancers, is 0.033 per reactor year and the public risk of thyroid cancer fatality is estimated to be 0.0064 per reactor year. Note that the SARA results are for total latent fatalities over a thirty year period as opposed to the latent fatalities "per year per year" given in the RSS. Applying the mortality fractions above to the fatal cancer risks yields a combined risk for all non-fatal latent cancers of 0.091 per reactor year, thyroid cancers comprising more than half the total. Thus, non-fatal 31 r

cancers are approximately 2 to 2.5 times greater in number than the reported fatal latent cancer fatalities.

This figure of 0.091 non-fatal latent cancers per reactor year, together with the 0.04 fatalities per year (0.033 + 0.0064) may be compared to the expected number of cancer fatalities per year from all causes in the population around LGS out to 50 miles, about 20,000 cases per year (see SARA p.12-14) .

M.I. Goldman 43. Genetic effects manifest themselves in the descendants of G.D. Kaiser exposed individuals and consist of both gene mutations and chromosome disorders in offspring. Genetic effects evaluated in the RSS (Ref.1, Appl. Exh. , Sect. 9.4 and Appendix I) included autosomal dominant disorders (single gene), multifactorial disorders (multiple gene pairs) and chromosomal disorders which result in a live birth. The RSS did not include the consideration of spontaneous abortions as these effects fell outside the RSS defini-tion of genetic risks, i.e., genetically caused disorders in live births capable of being transmitted to the next generation. The RSS used the BEIR I report "The Effects of Exposure to Low Levels of Ionizing Radiation" (Ref. 12, Appl. Exh. ) to estimate potential genetic disorders subsequent to external and internal radiation exposure.

Most recent estimates of genetic effects in the BEIR III report (Ref. 11, Appl. Exh. ) are based on new data that

' have become available since 1972, but they are not significantly different from the 1972 estimates. Genetic 32

effects were not estimated in SARA or the FES, but can be derived from the published results of public latent-cancer fatality risks. Current estimates suggest that the total number of genetic effects is of the same order t

as the number of latent-cancer fatalites (NRPB M78, ,

Table 1 (Ref. 9, Appl. Exh. )); UNSCEAR 1982 Report,

" Ionizing Radiation, Sources and Biological Effects" Annex I, Table 40 (Ref. 13, Appl. Exh. ); BEIR III, Table IV-2 (Ref. 11, Appl. Exh. ). Thus there wodld be about 0.04 genetic defects per year in the population surrounding LGS, which compares with an expected 6,000 cases per year from other causes in the population out to 50 miles, see RSS Table VI 9-10 (Ref. 7, Appl.

Exh. ).

M.I. Goldman 44. Spontaneous abortion estimates can be derived from G.D. Kaiser Tables VI 9-11, and 9-12 of the RSS (Ref. 1, Appl.

Exh. ) and they are on the order of 25 to 45% of total  ;

genetic effects estimated for live births.

M.I. Goldman 45. Sterility consequence effects are generally not reported

( G.D. Kaiser in risk assessments as they are viewed as subordinate to more serious radiation effects, such as acute fatality or ,

r early radiation illnesses. In the human male, radiation doses beginning above 10 rads and extending to 400 rads produce temporary sterility (either a decrease or an l

absence of sperm count, beginning 6 to 7 weeks after j exposure) lasting a few months to several years, see the l 33 l

BEIR I Report, p. 499 (Ref. 12, Appl. Exh. ). Doses delivered to the whole body at the upper end of this range would likely result in a fatality; however, localized exposure at this level would not be sufficient to produce permanent sterility in males. Impairment of fertility can result from doses to the female ovaries on the order of 300-400 rads, see the BSIR I Report, p. 499 (Ref. 12, Appl. Exp. ). Doses in this range may produce permanent sterility in women approaching menopause but only temporary sterility in younger women.

Again, an acute whole body exposure at this level would likely be lethal and, hence, the permanent sterility effect is less important. If the doses to the ovaries are fractionalized (i.e., delivered in smaller increments over a period of a few weeks) a dose of 1000 to 2000 rads may be required to reach the permanent sterility endpoint for exposed females.

ITEMS (2), (3), AND (8) - CONTAMINATION OF LAND l

l G.F. Daebeler 46. The deposition of radioactive material from the plume may S'. Levine result in unacceptable levels of contamination in milk r E.R. Schmidt produced by cows grazing on contaminated pastures, in the i

G.D. Kaiser contamination of crops, and in excessive radiation doses delivered to people by gamma rays emitted by deposited l

fission products (groundshine) if no protective counter-measures are applied. Note that the areas within which

• l 34

-. . =. _ _ - ..

O e i

these three effects would occur are not the same; different criteria are used to define them as described below, and they are of different sizes as may be seen from Figures VI 11-6 and VI 11-7 of the RSS (Ref.1, Appl. Exh. ). The contaminated areas could be easily identified by emergency response personnel after the accident and controls on both the ingestion pathways and access to highly contaminated areas could be put into i effect. The CRAC and CRAC2 computer codes are capable of estimating the different areas affected by the contamina-I tion effects described above, and are routinely used to estimate the associated costs.

S. Levine 47. It should be stressed that the principal impact of these G.D. Kaiser three kinds of contamination is economic; they would not contribute to health effects because interdiction would l be employed to prevent people from accumulating doses.

t i To put these impacts in perspective, Table 3 has been  ;

i prepared from CRAC2 runs that were carried out to obtain estimates of the public risk of economic cost for SARA (as represented by the area under the corresponding CCDF, see paragraph 11). Table 3 shows the relative contribu-tions of crop interdiction, milk interdiction and relocation of people from highly contaminated areas to the economic risk. Together, they make up less tisan 5%

l of the predicted costs, which are dominated by the cost of decontamination. As noted in SARA, Supplement 3, 35

Table 1, an estimate of the offsite economic risk l associated with LGS is $15,600 per year, in 1980 dollars which could be conservatively compared with an annual cost for the area within 50 miles of LGS of $13,000,000 per year for automobile accidents and $5,500,000 per year for fires in.1970 dollars. Thus, the total offsite economic risk associated with LGS is small and the P contamination effects mentioned in the contention contribute only a small fraction of this risk. Each of these effects is discussed in more detail below.  :

1 S. Levine 48. The interdiction model employed in the CRAC and CRAC2 G.D. Kaiser models is based on the concept of limiting chronic (long term) doses to acceptable levels. The limiting dose criteria used in the RSS, SARA and the FES are summarized in the attached Table 4. In CRAC and CRAC2, these dose ;

criteria are translated into corresponding contamination levels (curies per unit area) using dosimetric estimates for the grass-cow-man, the soil-root-crop-man pathways or the buildup of radiaticn dose due to radiation by gamma rays emitted by deposited fission products. The CRAC or CRAC2 computer models employ these contamination level i

criteria to derive the land areas subject to interdiction.

S. Levine 49. As noted above, the total area in which crops may have to be interdicted is routinely considered in CRAC2 and CRAC 36

O #

runs and is factored into the calculation of economic risks. The predicted frequency with which areas of various sizes would be contaminated above the levels set  ;

i for crop interdiction in Table 4, using the source terms in SARA Tables 12-7 and 12-8, are shown in Table 5.

These areas are those within which levels of deposited strontium and/or cesium are calculated to lie above the number of curies per square meter that correspond to the dose limits of 2 rem to the bone marrow or whole body as  !

described in Table 1. That is, urban areas are included, 1 as are bodies of water. Thus, the actual area of [

farmland in which crops may have to be interdicted is t smaller than given in Table 5. For example, tne fraction of farmland in Pennsylvania is 0.28 (SARA Table 10-11) so j that on average only about 30% of any contaminated area .

in Pennsylvania would actually contain farmland on which there could be crops. This factor is taken into account in the calculation of economic costs. ,

l t

(

50. It is also pertinent to note that assumption of invariant wind direction contained in CRAC2 may lead to the predic-tion that areas contaminated above limits set for inges-tion pathways (this applies to both crop interdiction and l

l milk interdiction) extend unrealistically far downwind.

This argument has been put forward by the NRC's Task Force on Emergency Planning in NUREG-0396, P. I  !

(Ref. 3,, Appl. Exh. ); "for the ingestion pathways (due  !

37 ,

l l

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

~. . t to the airborne releases and under Class 9 accident conditions), the downwind range within which significant contamination could occur would generally be limited to r

about 50 miles from a pvwer plant, because of wind shifts during releases and travel periods. There may also be i

l conversion of iodine in the atmosphere (for long time j periods) to chemical forms which do not readily enter the i

> ingestion pathway. Additionally, much of the particulate i materials in a cloud would have been deposited on the ground within 50 miles." ,

S. Levine 51. The results provided in Table 3 demonstrate the -

G.D. Kaiser principal contributor to economic risk is the cost of decontaminating land. This is why the total land area within which crops are interdicted is generally not

- explicitly presented as part of the results of a PRA. It should be noted that crop interdiction is expected to last one year (FES, p. 5-106) . f The total land area in which milk should be considered

13. Levine 52.

G. D. Kaiser for interdiction is also provided routinely in both CRAC w and CRAC2 and is therefore used in calculating the i

I: '

contribution to economic risk in both the FES and SARA.

f The predicted fre-quency with which area of various sizes j will be contaminated above the levels set for milk interdiction in Table 4 are shown in Table 6. This area -

is that within which levels of deposited iodine, p

38

--v- -, . - - - - , _ , , , - - ,, ,, , , - - - , . . , , - - , , - - - -w - - , - - e ,,. - , , - , , - , . . . - - - - - - - - -

strontium and/or cesium lie above the number of curies per square meter that correspond to the dose limits of I 3.3 rem to the bone marrow or whole body or 10 rem to the e

thyroid as described in Table 4. That is, as in the case  !

of crop interdiction, urban areas and bodies of water are  ;

included, and this is taken into account in calculations f

of cost. As can be seen from Table 3, the interdiction  !

of milk products is not a dominant contributor to economic risks. It should be noted that the time for  ;

milk interdiction is 2-3 months (FES, p. 5-106) .

S. Levine 53. The population within the land areas to be interdicted 1

G. D. Kaiser because of exposure to gamma rays emitted by deposited l fission products is also routinely considered in both CRAC and CRAC2 and is a centributor to the offsite economic risk in both the FES and SARA. As can be seen from Table 3, relocation costs are a relatively small contributor to total economic risk. ,

?

l S. Levine 54. In CRAC and CRAC2, it is assumed that a person will be l f

G.D. Kaiser relocated if, by remaining in the contaminated area,  ;

groundshine would lead to an accumulated radiation dose t

ef 25 rem in 30 years, see Table 4, even after the maximum feasible decontamination has been carried out.

The probability that various numbers of people will have l to be relocated for long periods is shown in Table 7.

l l

l i i

i 39 l

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

. _ ~ -

ITEM 6 THE COST OF MEDICAL TREATMENT OF HEALTH EFFECTS f

S. Levine 55. The quantification of the cost of medical treatment of G.D. Kaiser health effects is generally not considered in probabi-listic risk assessment. However, the FES does refer to a generic calculation (Nieves et al., 1982) which shows

! that the cost of medical treatment of health effects is l t "a fraction" of the costs of other offsite consequences

[

of reactor accidents (FES, p. 5-102) . In SARA, an

! estimate of the contribution to the offsite economic risk of health effects is given on p. 12-18; i

$1900/ reactor year. This compares with S6000/ reactor year for the median economic risk due to other offsite l

consequences (SARA Table 12-9), an increase of about a third (Note that these results would not change signi-l l

ficantly if they were recalculated using the revisions l

! described in SARA supplement 3) . These scoping calcula- 3 i

' tions show that there would be a small increase in the calculated offsite economic risk if the cost of health l l effects were included. This conclusion is supported by a i recent study; " Estimates of the Financial Consequences of .

Nuclear Power Reactor Accidents", by D. Strip, SAND 82-ll10 (NUREG/CR-2723) (Ref. 16, Appl. Exh. ). ,

I I Figure 11 of this reference shows that, for Reactor sites l

in the USA, estimates of the ratio of the cost of health effects to total offsite costs vary from 5% to 25%.

40 l

l I

. 1 I

CONTENTION DES-4B 6 "By treating some environmental costs in a CCDP i format and treeting other quantifiable costs in a  ;

non-quantitative subjective manner, the DES format obscures the total impact of severe ,

accidents at Limerick."

l l

G.F. Daebeler 56. Both FES and SARA attempt to treat the significant t

S. Levine environmental consequences and costs and to disclose j E.R. Schmidt information relating to the impacts of hypothetical low f G.D. Kaiser probability, and in some cases high consequence accidents, at the Limerick Generating Station. Both I

utilize sophisticated computer modeling to accomplish i

-this. l i

G.F. Daebeler 57. The output of such analyses is presented in terms of pre-t S. Levine dicted risk which is an accepted method of treating both '

E.R. Schmidt the frequencies of occurrence and the consequences of G.D. Kaiser potential accidents and giving appropriate consideration to each of these factors. While not all aspects of the j

i l

subject are at this point in time amenable to a fully ,

i rigorous probabilistic treatment, both SARA and the FES j l

i have treated them to the current state-of-the-art in risk ,

assessment in order to provide full disclosure. One must look at the entire discussion, both its quantitative and j I

( qualitative aspect, in the FES and SARA in order to

! understand the risks associated with the operation of  ;

Limerick Generating Station. Finally, any attempt to l t

isolate a discussion of the consequences of hypothetical l 5

l 41 l t-

, ~ - - - , , . - , , - - , , - , , - , , , , -_----,-,n- ,,,m,-,~,,,-., -- _ , - , , , , , , .,,,.-,,.,,.n,-,,,. , -,,,, + , - - -

f accidents from its associated probability or to examine only " worst case" consequences would distort the picture ,

l and would provide neither a rational disclosure nor any basis for decisionmaking. Applicant asserts that its treatment f severe accidents and that of the Staff, are f

entirely adequate to fulfill Connaission requirements.

h e

t i

l I

42 I . . - _ - - . - ,_ ,

_ 3 __ _

. .' L\ l

~

, n

\.

b g$

% \ i vi TQ ,- l s  %! e CONTiNTION CITY-13 f l

9

" CITY-13 Consequences to the citizens of

?hiladelphia in terms,of dose-distance relation-g ships are not presented in the DES analysis, nor, ,

in fact are such consequences for any area. The  ;

absence of this explicit data makes it impossible l

.for this Ccamission to,g.ccurately ascertain the likelihood of the public receiving doses in excess of Protective Action Guide (" PAG") levels,  !

or in excess of some.'other unacceptable level of '

societal risk, at,'for example the 21 miles which is the distance a plume would have to travel to reach the City of Philadelphia. Computer analysis by the Cityihas developed preliminary ,

specific dose-distance consequence data for the high density Philadelphia area.* These findings raise serious questions about the adequacy of the ,

DES. [

Under these values, should there be a severe accident at Limerick with the winds moving toward i

! the SE Sector, the chance of citizens of Philadelphia receiving a whole-body dose of 5  ;

l rems at the City boundary 21 miles down wind from i Limerick is 70%; the chance of a 30 rem dose is 40%. (At the eastern boundary of the City on the  :

l Delawarc River, some 30 miles from the plant, the l public has a 55% chance of receiving a 5 rem dose j and 15% chance of 30 rems). In 50% of such  ;

i severe accident releases, given wind direction toward Philadelphia, the total exposure within .i the SE Sector in the 20-30 mile range could reach j 10.5 million person-rems. This could result in as many as 8,400 latent induced cancers including 4,200 latent cancer fatalities."

! "*For purposes of this presentation source terms from the DES case II-T/WW were used. This ,

i sequence is 1/100,000. The ingestion pathway assumptions as to no protective action as t developed in NUREG-0396, were also used for these  ;

purposes. This analysis is not in all respects i one that would be presented, for example, in >

testimony. It is a limited analysis made under constraint of the filing deadline for the sole purpose of presenting some dose-distance data and some high density population data to the Board to demonstrate the seriousness of the City's contentions."

43

L G.F. Daebeler 58. This contention asserts that the risk to people in the S. Levine City of Philadelphia has not been separately presented E.R. Schmidt and presumably could be large. Initially, this testimony [

G. D. Kaiser addresses the possible consequences inside the City of Philadelphia by presenting a family of dose-distance curves and various specific probability-consequence rela-tionships for persons within the City of Philadelphia.

Using these relationships, it is concluded that the level of societal and individual risk arising from severe but t low probability accidental releases from LGS is extremely ,

i small for people within the City of Philadelphia.

Finally, some perspective is given as to the specific u t

example cited in the contention and it is also concluded i

that, when viewed in the context of risk, this example in fact supports the general conclusion that the risk is

• extremely small.

t G.F. Daebeler 59. SARA (and the NRC's FES) take into account the population i

l S. Levine of Philadelphia in calculating the societal risk associa-I E.R. Schmidt ted with Limerick Generating Station. This is accom-G. D. Kaiser plished by using the CRAC2 (and CRAC) codes which utilize

! actual site meteorological data and the population distribution around the facility, projected to the year 2000. Thus, the risk shown in these two documents takes into account the projected population in the City of Philadelphia. It is not necessary in terms of l

44 m

disclosing the environmental risk to prepare dose-distance curves for the sectors which include the City.

These are merely another way of presenting the data.

The end product of dose-distance curves does not consider the effect of these doses on the population.

This information is already presented in SARA and the FES.

S. Levine 60. Nevertheless, in order to respond to the contention, E.R. Schmidt dose-distance curves have been developed for the two G.D. Kaiser sectors (SE and ESE) which encompass the City. These are presented in Figure 2 for whole body dose and Figure 3 for thyroid dose. These were calculated by using the CRAC2 code in a way that is similar to that employed in SARA for the calculation of CCDFs, except that an intermediate result (i.e., dose vs. distance curves) has been selected.

S. Levine 61. CRAC2 was run with the source terms that are summarized E.R. Schmidt in SARA Tables 12-7 and 12-8 and was used to calculate l G.D. Kaiser the individual whole body dose and individual thyroid dose as a function of distance, using assumptions as in SARA, with the following major exceptions; (a) the radia-tion dose from inhaled radionuclides was integrated to one year; (b) people between 0 and 50 miles continued their normal activities for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The above assump-l tions were made in order to be consistent with the way in. .

t 45

- _- . - - - - . - . , __ , , , , , , . - , ,,r,-. , ,,7n-m,.,-.- .-,, _v-

' which dose-distance curves are calculated in NUREG-0396 l

(Ref. 3, Appl. Exh ; see caption to Figure I-11, p. I-38).

G.D. Kaiser 62. In order to provide further perspective, additional I results pertinent to the risk to the citizens of Philadelphia have been calculated and are presented in .

Table 8. These include the predicted frequency with which there might be one or more early fatalities in Philadelphia, one or more persons with bone marrow doses ,

in excess of 200 rem, or one or more persons with whole-

• body doses in excess of 25 rem. These results have been calculated with CRAC2 using the methods described in Chapter 10 and Appendix F of SARA, with the major exception that it was assumed that no emergency countermeasures would be taken for 43 hours4.976852e-4 days <br />0.0119 hours <br />7.109788e-5 weeks <br />1.63615e-5 months <br />. Thus, this is even more conservative than the NUREG-0396 type results presented in Figure 2. Even so, the predicted risks are extremely small.

G.F. Daebeler C3. The contention gives the results of what are termed " pre-S. Levine liminary dose-distance consequence data". The figures E.R. Schmidt provided for the conditional probability of exceeding ,

G.D. Kaiser various levels of dose at various distances, given the l DES source term II-T/WW, do not appear to be unreason- '

able, although they hsve act' been checked by independent CRAC2 calculations. However, the way in which the re-suits are presented does not give useful insight. A more helpful perspective is provided by factoring in the 46

frequency of occurrence of II-T/WW, which, from FES

p. 5-77 is 1/500,000+, and the probability with which the '

wind blows towards Philadelphia (0.272 in the CRAC2 runs for sectors SE and CSE) . The results, using the condi-tional probabilities from the contention, are given in Table 9. Thus, using frequencies provided in the FES and conditional probabilities provided in the contention, the probability of exceeding 5 rem or 30 rem in the City of Philadelphia is calculated to be very small. These doses -

would not lead to clinically detectable early effects (see FES at p. 5-66) .

S. Levine 64. There is an additional error in the footnote to the con-G.D. Kaiser tention, which states that the ingestion p athway assump-tions as to no protective action were useu as developed in NUREG-0396. However, the doses presented in the con-tention are whole-body doses. As is clear from NUREG-0396 (Ref. 3, Appl. Exh.  ; p. E-38 and p. I-46) ,

the ingestion pathway does not figure in the calculations of whole body dose, which is made up of three parts, ,

(a) inhalation component; (b) cloudshine from the passing plume; (c) groundshine accumulated for one day.

+There is an error in the footnote to CITY-13, where the frequency of II-T/WW is incorrectly quoted as 1/100,000/per reactor year. This has been recognized by the City. The frequency, magnitude and other characteristics of II-T/WW have not been changed from the  ;

DES to the FES.

47

S. Levine C5. The conversion from 10.5 million person-rem to 4,200 G.D. Kaiser latent cancer fatalities given in the contention implies a dose-response relationship of about 400 fatalities per million man-rem. This is towards the upper end of the range of uncertainty on this relationship (SARA

p. 10-25). In CRAC2, the corresponding point estimate relationship is about 168 fatalities per million man-rem (SARA p. 10-25) . Furthermore, since 10.5 million person-rem spread over the population of Philadelphia corresponds to about 5 rem per person, the central estimate applies, and a further reduction by a factor of five is in order (SARA p.10-15) . Thus, the 10.5 million person-rem would lead to about 400 fatalities. These would be spread over a period of about 30 years, at a rate of about 13 per year. This compares with a death rate due to cancer from all causes of about 3,000 per year in a city of the size of Philadelphia. It is noted that the central estimate as applied in CRAC2 is consistent with the linear-l quadratic dose-response relationship of the BEIR-III report (see PRA Procedures Guide, (Ref. 12, Appl.

Exh. , p. 9-54). Furthermore the results quoted above (i.e., 400 latent fatalities) must be associated with their frequency of occurrence, which is 1/500,000 (prob-ability of source term) times 0.27 (wind direction) times

.5 (accounts for the less favorable diffusion condi-tions--see the contention) which sequals 3x10-7, i.e.,

approximately one chance in three million.

48 I .

- - + . - - , , - . - _

- G.F. Daebeler - Thus, it may be seen that the predicted societal and S. Levine individual risks within the City of Philadelphia are very E.R. Schmidt small indeed.

G.D. Kaiser h

i

~

i t

0

. f l -

l..

1-t l ~.

e 9

e

• 49

?

9 e

n +- ,v.. ---n,m , .., -.,n. - - .-.e, ,ee.-- ,, , , ,,-,. ,.n., n, -- .,,, - ,- -., ,, ,--m,

CONTENTION CITY-14 (b) and (e)

"The DES does not accurately reflect either the median or upper estimates of the radiological effects which could result from an accident at Limerick because several key input assumptions associated with human activity after a severe accident are not realistic.

(b) Not included in the base case is the known phenomenon that as evacuees approach the City outskirts, their speeds would reduce, backups would occur and consequences due to trapped evacuees would increase.

(e) The DES does not separately portray the health consequences of an accident under a bad weather scenario. Many weather scenarios, including theoretically bad weather conditions, are averaged together."

S. Levine 67. As described previously, the CRAC2 code assumes that the E.R. Schmidt wind direction is invariant and that the evacuating G.D. Kaiser population moves radially. These are assumptions made for calculational convenience and do not reflect the actual planned or expected situation. These assumptions ,

in fact result in a conservative treatment of the expected interaction of evacuees in the downwind sector and the radioactive plume after an accidental release of radioactive material from the Limerick Generating Station. If the wind were to blow towards Sectors SE or ESE, (or any other sectors), those implementing the Disaster Operations Plan, Annex E, would advise evacuation routes which would avoid the radioactive plume. Thus, in the particular situstion, it is unlikely that there would be massive numbers of people driving towards Philadelphia. Rather, 50

the authorities (and the people themselves) would attempt to evacuate people laterally or crosswind. This is accounted for in the FES because the accumulation of r radiation dose ceases when people have moved 15 miles downwind of the reactor (FES p. 5-81) . SARA' conservatively assumes that evacuees travel radially to a distance of 20 miles from the facility, at which time they cease to accumulate radiation dose. (See SARA i

Table 10-8).

S. Levine 68. Even if large numbers of people were to drive towards E.R. Schmidt Philaaelphia and become " trapped" on the outskirts of the G.D. Kaiser City, the results given in the response to City-13 show that there is a very small probability that additional significant radiation doses would be accumulated even if

- the delays on the outskirts of the City were a number of hours. (See paragraphs 60, 61 and 62 above.)

S. Levine 69. Furthermore, the 2.5 mph evacuation speed assumed in the E.R. Schmidt FES is already low. It is an average over the total G.D.

Kaiser distance travelled by the evacuees and would account for some time spent moving slowly in congested condi-tions as well as time spent moving more rapidly on relatively clear roads. Thus, such phenomena as a slowing down effect on the outskirts of the city are already taken into account in the low evacuation speed used in the FES.

51

. e ,

9 S. Levine 70. With regard to Item (e) the contention states that many ,

E.R. Schmidt weather scenarios, including theoretically bad weather G.D. Kaiser conditions are averaged together. In SARA, the meteoro-

. logical conditions were taken into account according to ,

I the importance sampling procedure described in SARA Sec-tion F2.4.1; see also Section 3.1 of NUREG/CR-2552 (Ref. 16, Appl. Exh. ). The weather sequences are first j sorted into 29 " bins" which span the range of weather l

conditions that might be encountered by a plume of radio-l active material. These bins include " bad" weather condi-tions (i.e. , conditions that would maximize the number of health effects that would occur in the surrounding population). CRAC2 ensures that weather sequences are ,

i sampled from each bin as it calculates CCDFs. The most \

i severe consequences that arise are generally explicilty displayed at the right hand end or "tIail" of each CCDF.

S. Levine 71. To the extent that this contention may be interpreted as l l

E.R. Schmidt discussing severe weather as it affects evacuation, the l '

G.D. Kaiser data bece for the Sandia generic model used in SARA is l .

i derived from US evacuation experience which contains a rain case, a fog case, and a snow case (see Reactor I Safety Study (Ref. l.,

Appl. Exh. ) , Table VI J-1) and c

thus the impact of at least some examples of bad weather  ;

2 on evacuation is included in the selection of delay times in the generic model.

4 52 i

i

G.F. Daebeler 72. While it could be postulated that extremely adverse S. Levine weather could coincide with the occurrence of a E.R. Schmidt significant release, and result in a bounding (or " worst G.D. Kaiser case") calculation for adverse weather, such singularly unique events are really not significant when the purpose is to determine the overall risk due to the operation of a nuclear power plant. They are low probability occurrences and they would'have to be associated with diffusion t

conditions which would cauee large consequences and would affect only the " tails" (i.e. high consequence / low probability end) of the CCDFS, and not the overall risk.

G.F. Daebeler 73. The FES presents separately the health consequences of S. Levine accidenta under the "early reloc." emergency response E.R. Schmidt mode which is intended to cover " adverse site conditions G.D. Kaiser that would cause long delays before evacuation" (FES

p. 5-80). Table M.la of the FES shows that, even if such e

long delays were to occur all of the time, the public risk of early fatality from causes other than severe earthquakes, would increase by a factor of four. Such an assumption is, of course, highly unrealistic. If severe earthquakes are included, Table M.la of the FES shows a very small change in the predicted public risk of early fatality. Thus, this bounding analysis shows that the increase in public risk that would follow from an explicit inclusion of bad weather scenarios would be very r

small.

53

S. Levine 74. In order to provide additional insight on the effect of ,

E.R. Schmidt bad weather, a sensitivity study has been performed which G.D. Kaiser arbitrarily assumes that 4% of the time (i.e., a total of i about 15 complete days or 360 hours0.00417 days <br />0.1 hours <br />5.952381e-4 weeks <br />1.3698e-4 months <br />) the evacuation is i

adversely affected by bad weather to such an extent that l I

the assumptions for evacuation speeds for the seismic evacuation case are appropriate (3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> delay and 1 mph evacuation speed). This is quite conservative in that the evacuation parameters already take a range of weather conditions into account and because the adverse weather conditions would have to be exceptionally bad and therefore infrequent to csuse a reduction of the effective evacuation speed to 1 mph. The methodology used to calculate the risk is the same as in SARA, apart from the change in evacuation assumptions described above. These calculations show that the explicit inclusion of slowing of evacuation due to bad weather, in addition to that already included in the model data base, would have only a very small (less than 5% increase) effect on the results.

t t

54

[

REFERENCES  ;

1. USNRC (U.S. Nuclear Regulatory Commission) , 1975. " Calculation of

. Reactor Accident Consequences," Appendix VI of Reactor Safety Study--An [

Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants,  ;

WASH-1400 (NUREG-75/014) , Washington, D.C.  !

2. Aldrich, D. C., R. M. Blond, and R. B. Jones, 1978. A Model of Public f

Evacuation for Atmospheric Radiological Releases, SAND 78-0092, Sandia

, National Laboratories, Albuquerque, N.M.

, i

3. USNRC (U.S. Nuclear Regulatory Commission) ,1978. Planning Basis for i the Development of State and Local Government Radiological Emergency [

Response Plans in Support of Light Water Nuclear Power Plants, A report prepared by a U.S. Nuclear Regulatory Commission and U.S. Environmental Protection Agency Task Force on Emergency Planning, NUREG-0396 7 (EPA 520/1-78-016), Washington, D.C.  ;

. l

4. USNRC (U.S. Nuclear Regulatory Commission) , 1981. Criteria for Prepara- i tion and Evaluation of Radiological Emergency Response Plans and Preparedness in Support of Nuclear Power Plants, NUREG-0654, Washington, f D.C.

I

5. ,Aldrich, D. C., P. E. McGrath, D. M. Ericson, Jr., R. B. Jones, and N. C. I Rasmussen, " Examination of Offsite Emergency Protective Measures for [

. Core Melt Accidents," paper presented at ANS Topical Meeting on Prob-abilistic Analysis of Reactor Safety, May 8-10, 1978, Newport Beach,  ;

Cal. I i 6. Hans, J. M., Jr., and T. C. Sell, 1974. Evacuation Risks--An ]

Evaluation, EPA-520/6-74-002, U.S. Environmental Protection Agency, Las '

) Vegas, Nov.

7. Hilbert, G. D., F. M. Quinn, and J. H. Berkley, 1981. Mississauga

. Evacuates: A Report on the Closing of Canada's Ninth Largest City, j NUS-3614 (Prepared for the Philadelphia Electric Company and Others),  !

NUS Corporation, Gaithersburg, Md.

8. Moore, H. E., F. L. Bates, M. V. Layman, and V. J. Parenton, 1964.

Before the Wind--A Study of the Response to Hurricane Carla, National

%cademy of Sciences--National Research Council Publication 1095, ,

Washington, D.C.

9. Hemming, C. R., 1982. HEALTH-MARC: The Health Effects Module in the  ;

Methodology for Assessing the Radiological Consequences of Accidental l Releases, NRPB-M78, National Radiological Protection Board, Chilton, j Didcot, Oxon; U.K. l t

10. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation), 1977. Sources and Effects of Ionizing Radiation, United Nations, N.Y.

i i

_ ~. __

11. BEIR III; NAS-NRC (National Academy of Sciences--National Research Council) , 1980. The Effects on Populations of Exposures to Low Levels ,

of Ionizing Radiation, Advisory Committee on the Biological Effects of -

Ionizing Radiation, Washington, D.C.

12. BEIR I; NAS-NRC (National Academy of Sciences--National Research -

Council), 1972. The Effects of Exposure to Low Levels of Ionizing Radiation, Advisory Committee on the Biological Effects of Ionizing ,

Radiation, Washington, D.C.

13. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic '

Radiation), 1982. Ionizing Radiation, Sources and Biological Effects, United '

Nations, N.Y.

14. Strip, D. R., 1982. Estimates of the Financial Consequences of Nuclear Power Reactor Accidents, NUREG/CR-2723 (SAND 82-1110) , Sandia National  ;

Laboratories, Albuquerque, N.M.

15. USNRC (U.S. Nuclear Regulatory Commission) , 1983. PRA Procedures Guide, ,

NUREG/CR-2300, Washington, D.C.

16. Ritchie, L. T., D. J. Alpert, R. P. Burke, J. D. Johnson, R. M. Ostmeyer, D. C. Aldrich, and R. M. Blond, 1984. CRAC2 Model Description, NUREG/CR-2552 (SAND 82-0342) , Sandia National Laboratories, Albuquerque, N.M.

r i I i

i l

l l  !

t I

M F

L I

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

4 Table 1 Description of Sensitivity Studies Performed

• To Examine the Effect of Different Relocation and Sheltering Assumptions Between Ten and Twenty-five Miles Resultsa CASE 1 . Base case as in SARA consisting of 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> normal 4.52-5 activity after plume passage followed by rapid relocation.

CASE 2 People in the 10-25 mi.le range shelter in basements 4.32-5 for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> af ter plume passage and then relocate '

rapidly.

CASE 3 People in the 10-25 mile range shelter in basements 4.86-5 for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after receivita a warning of a release and then move out at 2.5 mph.

CASE 4 People in the 10-25 mile range shelter in basements 4.32-5 for 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after plume passage and then relocate rapidly.

CASE 5 People from 10-15 miles delay for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> with 9.32-5

" normal activity"-type shielding factors after ,

receiving a warning, then evacuate at 2.5 mph. .

People from 15-25 miles delay for 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> with

" normal activity" and then relocate rapidly, aThese are areas under the CCDF for early fatalityb.seeSensitivity for internal initiators only. 4.52-5 = 4.52 x 10-Paragraph 11 abovefor studies the seismic case would be expected to show similarly small variations.

w 57

Table 2 Sensitivity of Predicted Public Risk of Early Fatality To Changes in Delay Time and/or Effective Evacuation Speed Delay Time Speed Cleara Publieb Case Description (Hr) (MPH) Time (Hr) Risk 1 SARA base case 1, 3 or 5 10 2, 4 or 6 4.5-Sc 2 FES evac-reloc 2 2.5 6 3.5-5 3 SARA base case 1, 3 or 5 2.5 5, 7 or 9 6.3-5 modified--2.5 mph evacuation speed 4 3 10 4 4.1-5 5 5 10 6 8.8-5 6 3 2.5 7 6.6-5 7 5 2.5 9 1.0-4 8 Very pessimistic 3 1 13 9.9-5 evacuation speed for internally initiated '

events aTime from declaration of a General Emergency until last person leaves EPZ.

bPublic risk of early fatality as depicted by the area under the CCDF (see Paragraph 11). Due to internal initiators only (the Sandia Generic model and the FES evac-reloc model apply to internal initiators only). Units are fatalities per reactor year.

c4.5-5 = 4.5 x 10-5 ,

l 58

l i

Table 3 ,

Contributions of Various Components of the CRAC2 Economic Risk Model Economic Risk Percentage

($1980 per of Total Category Reactor Year) Economic Risk i

t Crop Interdiction 320 2.0%

Milk Interdiction 67 0.4%  !

Relocation Expenses 380 2.4%

Cost of Decontamination 12,100 77.1%

Others (all remaining 2,850 18.1%

sources) .

Total Economic Risk 15,600 l

r i

b l

I l

k l-59 l .

i-I -

w l

Table 4 Dose Criteria Used in CRAC and CRAC2 to Define Interdiction Requirements a l Exposure Pathway Dose :

I 25.0 rem to the whole body External irradiation:

in 30 years.

l Ingestion via milk:

Strontium 3.3 rem to the bone marrow l in first year Cesium 3.3 rem to the whole body Iodine 10.0 rem to the thyroid Ingestion via "other" pathways:  ;

Strontium 2.0 rem to the bone marrow ,

in first year Cesium 2.0 rem to the whole body i

aThis table is taken from Table VI 11-6 of the RSS (Ref. 1, Appl.

Exh. ).

t

(

L' i

l f

l.

1 i i

i l

l i

l l f

60 i

i

'"able 5 Predicted Frequencies with Which Areas of Various Sizes Are Contaminated Above the Levels Set for Interdiction of Crops Frequencya Centaminated AreaD (per reactor year) (square miles) 10-5 300 10-6 4,000 10-7 7,000 10-8 18,000 10-9 25,000 aThe cumulative frequency in the correspond-ing CCDF, see Paragraph 11.

DAs noted in the text (Paragraphs 49 and 50),

this exceeds the actual area of farms lying within the contaminated area because it con-tains urban areas, mountains and water, and because of conservations in the CRAC2 analysis i,

i e

L l

l l

I l

I 1

61 l

i w-. , , ,e- - - . . - , . v. _ . , - ~ . - - - - , .- , , .. , - - - - - - - - ,

.. o Table 6 Predicted Frequencies with Which Areas of Various Sizes Are Contaminated Above the Levles Set for Interdiction of Milk Frequencya Contaminated Areab (per reactor year) (square miles) ,

10-5 7,000 '

10-6 20,000 10-7 25,000 10-8 30,000  ;

aThe cumulative frequency in the corresponding ,

CCDF, see Paragraph 11.

bAs noted in the text (Paragraphs 49-50), this  ;

exceeds the actual area of dairy farms lying l I

within the contaminated areas because it con-tains urban areas, mountains and water, and because of conservations in the CRAC2 analysis &

i 1

I, t

i 62  :

t I

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

.,.. -- .i

. o Table 7 Predicted Frequency with which Various Numbers of People Need to be Relocated for Long Periods Because of Contamination of the Ground by Deposited Fission Products Frequencya Number of People (per reactor year) Relocated 10-5 0 10-6 20,000 g 10-7 100,000 10-8 500,000 10"9 1,000,000 aInterpreted as a cumulat.ive frequency as on the corresponding CCDF (see Paragraph 11).

63

o .

Table 8 Predicted Frequency of Occurrence of Various Radiological Effects in the City of Philadelphia Chance that there will be One in 3 billion per one or more early fatalities year in Philadelphia Chance that there will be One in 750 million per one or more persons in Phila- year delphia requiring hospital treatment (bone marrow dose 200 rem)

Chance that there will be One in 16 million per one or more persons in Phila- year ,

delphia with whole body dose in excess of 30 rem I

i I

i r

l I

f 1

i

[

I I

64 F

, ,w, +-----s.m_ ,,. --r _a__. - - -%--, .- w o* y ,,.- -, --7 ve-,----, y.- y---,.,,-9 w __

..wr

Table 9 Predicted Frequency with Which various Dose Levels Are Exceeded in the City of Philadelphia, Using Information From the Contention Predicted Frequency with Which Dose Level is Exceeded at Given Distance in the City of Philadelphia Dose Level Distance (per reactor year) 5 rem 21 miles 0.7 x 0.272 x 1/500,000

= 3.8 x 10-7 .

= One chance in 2 1/2 million 30-rem 21 miles 0.4 x 0.272 x 1/500,000

= 2.2 x 10-7

= One chance in 5 million 5 rem 30 miles 0.55 x 0.272 x 1/500,000

= 3 x 10-7

= One chance in 3 million 0.15 x 0.272 x 1/500,000  !

30 rem 30 miles

= 8.1 x 10-8 '

= One chance in 12 million 4 l

h

?

l t

t I

s t

65  ;

l

Figure 1 Population Density vs. Evacuation Time em '

em ett e EVEt4T hohneER

• l

.me _ A 2 -

e.e l

. Sie . t

= ,

.. y

. _ e. 4

?

em em ,

l ear I

I i l

• " i

. _ A i

k en i en  !

l l eu ,

l -

6 l

l ", .  :

1 Tene tcouwed to evxu.de afketed .46.1 (hourd From Hans and Sell (Ref. 4_, Appl. Exh. x_)

I i

I j 0-4 . . . . . . . . ....e

.....l ..l- _-

- t

~

. i n

1 W 10 4  :

c E -

W -

I Q. - .

v

_ 5 REM 1 REN _ r u

m .

o .

o .

> g p-e a _ -  !

a -

co .

w _ 30 REN -

J _

o _

I _

3 .

i e

z H

O -  !

w 10-7 -

50 REM _

w _

o _

x -

W - ,

W - -

o .

I O i 2

w -

3 1 0-' -

w m

200 REM

- \ -

)

t t t t tt t t t t II It t '

t t t tt1

.g t .I i'

10' 10' 10 10' l t

DISTANCE CMILES)

Figure 2 Dose-Distance Relationship for hhole Body f

l Y

i 1 O.4 . . . . . . . . . . . . . .c

...l .

n E 10 4 -- -

w _ _

p. -

g .. .

w . .

Q.

v - -

d 25 REN O

Q o 1 0-* _ _

g .

O m . .

3 r . .

M . . . . 5 REM -

o z

g Q

w d 10-7  ? -

x w

b. 300 REN o . ,

l

> l u

z - -

w c

W 1 0-* - -

e 1

i t

' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' l ' ' ' ' ' ' ' '

10-'  !

2 tO' 10' 10 10' ,

DISTANCE (MILES)

Figure 3 Dose-Distance Relationship For Thyroid  !

P l

~^

1 ATTACHMENT 1 l PAGE 48, Hans and Sell, Evacuation Risks - An Evaluation, ,

EPA-520/6-74-002, U. S. Environmental Protection Agency, l June 1974 l Based upon Dr. Dynes' response to the specific question '

cf behavior to radiation versus other threats, cerroborated l by tha research (40) that reveals the true behavior of people during a disaster as opposed to the panic conception, there i is no reason to believe or assume that the risk or injury or l death should be any higher due to an evacuation than the normal accident or injury rate.

. . . one fact is borne out by various data of past  ;

disasters: the freedom to escape from threat of death or i injury has a calming effect on the population." ,17 ) ,

Motivation to evacuate l

In many cases, even when presented with a grave threa*

people refuse to evacuate (16,21,28,40). Many reasons have been j given both by persons who have not evacuated (17,23), and persons l

conducting the evacuation as to this reluctance to leave. To some degree, it is the individual's impressions and .4nterpre-tation of tha.aeriousness of the situation based on eht official or unofficial information he/she receives. An individsal evaluation is made..and a positive or negative action elicited.

It cannot be taken for granted that an official order to evacu-  ;

ate will be followed, even if it is a mandatory rather than a  ;

voluntary order. Results of this study indicate that approxi-  !

mately six percent of the total population refused to evacuate. l Other reports indicate this figure can run higher than 50  ;

percent (23) . j

. There is no reason to believe that because the disaster I agent is radiation rather than some other agent, thatRather, is, in  ;

itself, will provide suffici'ent motivation to leave.

the opposite viewpoint should be taken--people will be hesitant I to leave. Cognizance should be given in the planning stage

• to this problem and appropriate thought given to its remedy.

Warning systems and communication systems between evacuee-avacuator, evacuator-evacuater, and evacuator-news media- i i

' population play a significant role in the emergency and/or

evacuation process (11,12,12). It is not only important that

pretested, workable systems be available, but that an under- .

standing of peoples' response and behavior to warning systems l

! be recognized and be advantageously used. [

Emergency plans f

There have been many documents published on emergency and [

~

l disaster planning (44-48), some of which are listed in the bibliography. It was Eot the intent of this report to go into l i

l f

i

P e

PROFESSIONAL QUALIFICATIONS George F. Daebeler Supervising Engineer, Environmental Branch Philadelphia Electric Company My name is George Daebeler. My business address is 2301 Market Street, -

Philadelphia, Pennsylvania, 19101. I am in charge of the Environmental Branch of the Nuclear and Environmental Section of the Mechanical Engineering Division of the Engineering and Research Department. In this position, I ,

l supervise engineers and other professional personi.el responsible for environmental monitoring, radioactive effluent monitoring systems, and l probabilistic risk assessment associated with the Limerick Generating Station. ,

I received a Bachelor of Science degree in Mechanical Engineering from Rennsselaer Polytechnic Institute in 1962 and a Master of Science degree in Nuclear Engineering from Pennsylvania State University in 1966.

I joined Philadelphia Electric in June, 1966 and was assigned to an organization which 1 ster became the Nuclear Section of the Mechagical Engineering Divisi66. My work in this group included responsibility for nuclear fuel, various plant systems, and licensing activities.

In January,1973, I became the head of the Safety and Licensing Branch and in November of 1975, I became a Senior Engineer and head of the Nuclear Steam Supply Branch where I supervised engineers responsible for nuclear i reactor and safety system, including those associated with Limerick. In June, 1982, I became head of the Environmental Branch.

1  !

4 4

I am a registered Professional Engineer in Pennsylvania and a member of the American Society of Mechanical Engineers and the American Nuclear Society.

4 t F

s l

I 4

L i

f 4

i t

I t'

i  !

f l

i i

l O

I r

l t

i

)

+

[

I 6

?

. I l

f l

t l

2 f

+

p

)

e .w--. a , , .---r----r. ---,--..r--,-----,--- ,e,ca-..,,,. n---,. ,,n-..-- ~ ~ - - - - - , -, - - - - . -- _ w ,- n - --

E PROFESSIONAL QUALIFICATIONS MORTON I. GOLDMAN Senior Vice President - Technical Director NUS Corporation My name is Morton I. Goldman. I am Senior Vice President and Technical Director of NUS Corporation, 910 Clopper Road, Gaithersburg, Maryland, 20878.

I have served in this capacity since January 1982. Prior to this assignment, I had been Senior Vice President, Environmental Systems Group in which I was responsible for all site evaluations, safety analyses, waste management system evaluations, health effects analyses, and environmental programs conducted by this group. This included the evaluation of site and environmental safety factors for about 50 nuclear and fossil-fueled plants in this country and abroad.

I graduated from New York University in 1948 with the degree of Bachelor of Science in Civil Engineering. In 1950, I received a Master of Science degree in Sanitary. Engineering; in 1958, a Master of Science degree in Nuclear Engineering; and in 1960, a Doctor of Science degree, all from the Massachusetts Institute of Technology.

From 1948 to 1949, I was a Research and Teaching Assistant at the Sanitary Engineering Research Laboratory, New York University, conducting research on water coagulation and assisting in teaching sanitary chemistry and sanitary biology laboratory courses.

From 1949 to 1950, I was a Research Assistnat at the Radioactivity Research Laboratory, Sanitary Engineering Department, Massachusetts Institute 1

. o of Technology, conductbg original research on removal of radionuclides from water by standard water treatment techniques.

From 1950 to 1956, I was on loan to the Oak Ridge National Laboratory as Chief of Soils and Engineering Section, Waste Disposal Research Activities.

In this position I conducted and supervised research on disposal of radioac-tive wastes at Oak Ridge National Laboratory.

From 1956 to 1959, I was assigned to Massachusetts Institute of Technology as Project Leader for the Radioactive Waste Disposal Project of the Sanitary Engineering Department, and in training in the Nuclear Engineering Department. In the former capacity, I initiated and supervised research on novel methods of disposal of high activity fission product waste materials.

I also served as secretary to the MIT Reactor Safeguards Committee.

From 1959 to 1961, I was designated as Nuclear Installation Consultant with the Divsion of Radiological Health in Washington, D.C. In this capacity I provided technical consultation and assistance to state and federal agencies on health and safety problems associated with nuclear installations.

As part of my responsibility, I served on a working group responsible for the Radioactivity Section of the USPHS Drinking Water Standards (1960).

Since 1961,' I have been with NUS corporation and active in the environ-mental and safety activities described earlier. I was elected Vice President and General Manager, Environmental Safeguards Division in January 1966, and Senior Vice President, Environmental Systems Group in February 1973.

2

. - e i

In 1968, I served as U.S. representative to and chairman of an IAEA expert I i

panel on Radioactive Waste Management at Nuclear Power Plants, resulting in IAEA Safety Series No. 28 of that title. From 1972 to 1975, I served as con-sultant to and witness for the Consolidated Utility Group in the ABC/NRC rule-making hearing on "as low as practicable" radioactive waste discharge standards. From 1977 to 1982, I served as consultant to and witness for the Utility Group particpants in the GESMO rulemaking hearing, and from 1978 to 1981 was consultant and witness in the NRC's Consolidated radon hearing originating in the Perkins proceeding.

I am the author and coauthor of a number of papers on radiation and public health, nuclear safety and siting, and radioactive waste management. I I am a member of the American Society of Civil Engineers, and the American Nuclear Society. I am a licensed Professional Engineer in the  !

states of New York, Maryland,' California, South Carolina, Arizona, and the District of Columbia; and a Diplomat of the American Academy of Environmental Engineers in radiation hygiene and hazard control. I am a member and former i chairman of the ASCE Technical Committee on Nuclear Effects, and a member of the Nuclear Energy Committee, ASCE. Other activities include Chairman, Atomic Industrial Forum Ad Hoc Committee on De Minimis Concept in Radiation  ;

Protection, and Radiological Aspects of the Clean Air Act. I am also a member of Steering Group, AIF Committee on Environment; member, Committee on l Nuclear Standards, ASCE; member, Standards Committee ANS-2 on Site Evaluation.

3

PROFESSIONAL QUALIFICATIONS SAUL LEVINE Vice President and Consulting Group Executive NUS Corporation My name is Saul Levine, and I am Vice President and Consulting Group Executive, NUS Corporation, 910 Clopper Road, Gaithersburg, Maryland 20878.

NUS Corporation is an internationally known consulting company in the field of energy and has some 1300 employees. My organization is responsible for performing nuclear power plant safety analyses, probabilistic risk assessments and reliability analyses, providing quality assurance services, supplying environmental services, and assisting NUS clients in reactor licensing. .

I have been involved with the application of nuclear energy for nearly 30 years. I hold a Bachelor of Science degree from the U.S. Naval Academy and two degrees from the Massachusetts Institute of Technology: Bachelor of Science in electronics engineering and a Master of Science in nuclear engineering. After serving in the U. S. Submarine Service from 1945 to 1954, I reported, from 1955 to 1958, to Admiral Rickover as Project Officer for the U.S.S. Enterprise, the world's first nuclear powered aircraft carrier. In this position, I was responsible for directing all technical, financial, production, and administrative aspects of the reactor plant prototypes and the production plants for the U.S.S. Enterprise. From 1958 to 1962, I worked in the U. S. Navy's Special Projects Office, which was responsible for l'roducing the submarine based 1

Polaris Missile System. I managed the design, integration, installation, testing, and performance evaluation of the Polaris Missile Submarine Navigation System. ,

From 1962 through the end of 1979, I was with the U. S. Atomic Energy

'Cosmission (ABC) and its successor, the U.S. Nuclear Regulatory Commission (NRC). During those years, I was Assistant Director for Reactor Technology; Assistant Director of the Division of Environmental Affairs; Project Staff ,

Director for the Reactor Safety Study (WASH-1400) (1), which represented the [

I first comprehensive evaluation of the likelihood and consequences of nuclear power plant accidents; Assistant Director, Division of Reactor Safety f Research; Deputy Director, Office of Nuclear Regulatory Research; and  ;

i Director, Office of Nuclear Regulatory Research.  !

• i In 1980 I joined NUS Corporation as Vice President and Consulting Group Executive. In this capacity I have been closely associated with work l

performed by NUS' Consulting Division in the area of probabilistic risk assessment. This group has performed several PRAs concerning plants such as l Limerick, Susquehanna, Shoreham, and Ringhals 2. Many other smaller PRA tasks have also been performed such as mini-PRAs on a number of reactors and the review of PRAs done by others. In particular I have performed a technical management overview function for both the Limerick PRA and the Severe Accident' Risk Assessment (SARA).

2

PROFESSIONAL QUALIFICATIONS E. ROBERT SCHMIDT Director, Systems Analysis NUS Corporation My name is E. Robert Schmidt. My business address is 910 Clopper Road, Gaithersburg, Maryland 20878. I am Director of the Systems Analysis Group of the Consulting Division and as such am responsible for directing all systems analysis consulting services associated with nuclear and nonnuclear technology, including radiological and nonradiological accident analysis, thermal-hydraulic and heat transfer analysis, and risk assessment and probabilistic safety analysis.

I received a Bachelor of Science degree in Mechanical Engineering from the University of Missouri in 1958 and a Master of Science degree in Nuclear Engineering from the same institution in 1959. After graduation I worked for General Electric for one year. I then worked for Internuclear Company from 1960 to 1963. During that time I developed design criteria and analyzed in-pile loops of the experimental gas-cooled reactor at Oak Ridge National Laboratory and participated in the design of several small reactors.

I have been with NUS Coporation since 1963 and during that the time I have been involved in all facets of the design, operation, and analysis of nuclear power plants. I was onsite startup consultant to the Governement of India, the Japan Atomic Power Company, and the Toyko Electric Power Company for the startup of four BWR units.

1

I have directed a vast amount of licensing and safety analysis work and have participated in many special nuclear technologies studies. Some of the most significant include a study of steam cycle conditions for a prototype large breeder reactor, safety analysis report review for foreign licensing authorities and domestic utilities, industrial and aircraft impact hazards analysis, containment and subcompartment temperature and pressure analyses, and the design and safety analysis of several spent fuel shipping casks.

Prior to my current position, I was Manager of the Reliability and Risk i

Assessment Department. I performed and directed risk assessments, degraded  !

core accident evaluations, safety goal analyses, and detailed assessments of the probabilities and consequences of accidents involving hazardous material l transport near a nuclear power station. I was also involved in a study of i

aircraft impact probabilities which included providing hearing board  ;

testimony.

Most recently I have been responsible for directing the Kuosheng, ,

Susquehanna, and Ringhals 2 risk assessments. I also directed the Limerick external event risk assessment, and with Mr. Saul Levine, provided the tech-nical monitoring of the Limerick inplant failure risk study. I also managed limited scope, mini-PRAs for six nuclear power plants.

l I an a Registered Professional Engineer in the District of Columbia. I l

an a member of the American Nuclear Society, the American Society of i Mechanical Engineers, and the Society for Risk Analysis.

l 2

o . .

PROFESSIONAL QUALIFICATIONS GEOFFREY D. RAISER Manager, consequence Assessment Department NUS Corporation My name is Geoffrey D. Raiser. My business address is 910 Clopper Road, Gaithersburg, Maryland 20878. I am manager of the Consequence Assessment Department. In that position, I am responsible for managing projects relat-ing to the consequences of accidental releases of radioactive, toxic, and flammable chemicals.

I received a Bachelors of Science degree in Physics from Cambridge University (UK) in 19643 a Master of Science degree in Physics from Cambridge in 1967; and a Doctor of Science in Elementary Particle Physics, also from Cambridge University in 1968. Subsequently, I had postdoctoral research fel-lowships in theoretical particle physics at the Cavendish Laboratory at Cambridge and the University of Miami. I held a temporary lectureship an applied mathematics at the University of Durham (UK) during the academic year 1970/71 and served as a senior Research Associate in theoretical particle physics at the Daresbury Nuclear Physics Laboratory, Warrington, UK, from 1971 to 1974.

From 1974 to 1980 I worked at the United Kingdom Atomic Energy Authority's safety and Reliability Directorate (SRD) in the Environmental and Fission Product Group. In 1976, I was apointed Head of Physics and led a group which grew to include 10 people involved in the development of methods with which to predict the consequences of the accidental release of radiotoxic, chemically toxic, and flameable materials to the environment. During my time at SRD,1 1

developed the nuclear consequence modeling code TIRION, which was widely used in the United Kingdom and abroad in applications to reactors, reprocessing plant, nuclear shipping, and the transport of plutonium by road, rail, and sea. The most important application of TIRION was at the Windscale Inquiry into the building of a reprocessing plant for oxide fuel. I also participated in and/or managed multidisciplinary projects relevant to the safety and environmental impact of advanced technologies, including par-ticipation in the well-known Canvey Island Study.

I was a frequent speaker at seminars and international conferences, and participated as a lecturer at courses arranged by the United Kingdom Atomic Energy Authority. I chaired several international working groups on conse-owwe analysis.

In 1981, I joined NUS Corporation and in 1982, became Manager of the Consequence Assessment Department. Since that time I have been involved in many significant projects. I provided overall technical management for the phenomenological and consequence analysis portions of the Susquehanna Probabilistic Risk Assessment, and for the consequence analysis and ttansportation accident analys.fs for Limerick. I have recently been managing the Phase 2 probabilistic safety study for the Swedish State Power Board's Ringhals 2 plant, the purpose of which is to develop source terms for severe accidents. I.an also responsible for the consequence analysis for the Industry Degraded Core Rulemaking Program. I have managed " mini-PRAs" for the Pelo Verde and Hope Creek Nuclear Generating Stations and have written Chapter 7 of the environmental reports for Hope Creek and Limerick. I was a 2

. .a founder member, and also as author and co-editor, of the comunittee on the Safety of Nuclear Insta11abious International Benchmark Comparison of Conse-quence Modeling Codes.-

i 1

i I

?

?

L f

I i

l 3

. .o UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION In the Matter of )

)

Philadelphia Electric Company ) Docket Nos. 50-352

) 50-353 (Limerick Generating Station, )

Unita 1 and 2) ) ,

C_ERTIFICATE OF SERVICE I hereby certify that copies of " Testimony of G.F.

Daebeler, S. Levine, M.I. Goldman, E.R. Schmidt, and G.D.

Kaiser, Relating to Severe Accident Risk Contentions" in the captioned matter have been served upon the following by Federal Express and hand delivery as indicated or by deposit in the United States mail this lith day of May, 1984:

• Lawrence Brenner, Esq. (2) Atomic Safety and Licensing Atomic Safety and Licensing Appeal Panel

• Board U.S. Nuclear Regulatory U.S. Nuclear Regulatory Commission Commission . Washington, D.C. 20555 Washington, D.C. 20555 Docketing and Service Section

• Dr. Richard F. Cole Office of the Secretary Atomic Safety and U.S. Nuclear Regulatory Licensing Board Commission U.S. Nuclear Regulatory Washington, D.C. 20555 Commission Washington, D.C. 20555

• Ann P. Hodgdon, Esq.

Counsel for NRC Staff Office

• Dr. Peter A. Morris of the Executive Atomic Safety and Legal Director Licensing Board U.S. Nuclear Regulatory U.S. Nuclear Regulatory Commission Commission Washington, D.C. 20555 Washington, D.C. 20555

• Hand Delivery May 11, 1984

. e .6 Atomic Safety and Licensing Angus Love, Esq.

Board Panel 107 East Main Street L U.S. Nuclear Regulatory Norristown, PA 19401 Commission Washington, D.C. 20555 Robert J. Sugarman, Esq.

Sugarman, Denworth &

Philadelphia Electric Company Hellegers ATTN: Edward G. Bauer, Jr. 16th Floor, Center Plaza 71ce President & 101 North Broad Street General Counsel Philadelphia, PA 19107 i 2301 Market Street 1 Philadelphia, PA 19101 Director, Pennsylvania Emergency Management Agency Mr. Frank R. Romano Basement, Transportation 61 Forest Avenue and Safety Building '

Ambler, Pennsylvania 19002 Harrisburg, PA 17120 Mr. Robert L. Anthony

• Martha W. Bush, Esq.

Friends of the Earth of .K,athryn S. Lewis, Esq.

the Delaware Valley City of Philadelphia 106 Vernon Lane, Box 186 Municipal Services Bldg.

Moylan, Pennsylvania 19065 15th and JFK Blvd.

Limerick Ecology Action P.O. Box 761 762 Queen Street Spence W. Perry, Esq.

Pottstown, PA 19464 Associate General Counsel Federal Emergency

!

• Charles W. Elliott, Esq. Management Agency

. Brose and Postwistilo 500 C Street, S.W., Rm. 840

! 1101 Building Washington, DC 20472 lith & Northampton Streets .

Easton, PA 18042 Thomas Gerusky, Director Bureau of Radiation

, Zori G. Ferkin, Esq. Protection l Assistant Counsel . Department of Environmental Commonwealth of Pennsylvania Resources Governor's Energy Council 5th Floor, Fulton Bank Bldg.

1625 N. Front Street Third.and Locust Streets Harrisburg, PA 17102 Harrisburg, PA 17120 Jay M. Gutierrez, Esq.

U.S. Nuclear Regulatory Commission P.O. Box 47 Sanatoga, PA 19464

• Federal Express

(. .

James Wiggins Senior Resident inspector U.S. Nuclear Regt.latory

. .. Commaission P.O. Box 47 Sanatoga, PA 19464 Timothy R.S. Campbell Director Department of Emergency Services-14 East Biddle Street West Chester, PA 19380

$s '

sa -

Robert M. Rad.er e

e