ML20125B603

Discusses Draft Rept Siting Nuclear Power Plants. Disagrees That Risk Assessment Methodology Should Not Be Based on Arithmetic Formula.Forwards Comments

ML20125B603
Person / Time
Issue date:	12/04/1979
From:	Muller D Office of Nuclear Reactor Regulation
To:	Yellin J MASSACHUSETTS INSTITUTE OF TECHNOLOGY, CAMBRIDGE
Shared Package
ML20125B605	List: ML20125B603 ML20125B608
References
NUDOCS 7912190292
Download: ML20125B603 (16)

Text

._ .__ _ _ _ . ._ . _ . _ _ _ _ _ _ _ . , _ . .. _ __. .

y'

~

, iva npe  ;

UNITFu S FATES

[W 4'Iogo. NUCLEAR REGULAlORY COMMISSION WASHING TON, D. C. 20555

${h:)dQ,$

y.....

DEC 4 1979 D

PP]D % n n n n Mr. Joel- Yellin I

MM Q' L h' $

6

, t Associate Professor Center for International Studies Massachusetts Institute of Technology Cambridpe, Massachusetts 02139 .

Dear Professor Yellin:

As requested by Mr. Harold R. Denton, I have revicued your draft report,

" Siting Nuclear Plants." We appreciate very much the opportunity to review your thoughtful analysis of remote siting policy issues. In general, we find agreement with the main thesis of your paper as to the desirability of developing more explicit policy on remote siting as one of the potentially useful approaches in reducing the adverse conscquences of a class 9 accident. Indeed, certain goals of the Reporttof the Siting Policy _Ta._sk Force _ (which Mr. Denton has already sent to you) express the need to encourage more isolated siting of nuclear power plants and identify criteria important to such policy formulations.

However, we find disagreement with the central philosophy of your paper that risk assessment methodology should not be based on the " arithmetic formula" (p.10 of your report) of multiplying probabilities by consequences. ' tor do we agree with your statement (p. 2) that "the potential consequences of a proposed technology deserve greater emphasis in risk assessment than do the associated probabilities.

From the lessons learned in the Three Mile Island accident, we feel that greater emphasis is needed on both parts of the risk assessment formulation. Indeed, substantive recommendations were made by the Kemeny Conmission report on the TMI accident which would require, to the " maximum extent feasible," the location of new power plants in areas remote from concentrations of population, but also a number of other measures designed to reduce the probability of class 9 accidents and their consequences. The NRC, in the enclosed report,* has expressed general agreement With these substantive recommendations.

• Preliminary Analysis and Views of the Nuclear Regulatory Commission on the

. Recommendations of the President's Commission on the Accident at Three Mile Island, NUREG-0632, November 9, 1979.

90013158 7912190 99Q%

a Dm Jum D r,}'%*

• J uuu 1 Azru l

_, DEC 4 1979 Mr. Joel Yellin Enclosed are some additional comments on your report prepared by several staff members. These comments provide more specific views on your approaches and analyses of issuds which it is hoped will be of some benefit to you in your further efforts in preparing a final draft.

In our future staff deliberations in the development of siting policy which ,

would explo,a the tradeoffs in benefits and costs in achieving greater site  !

isolation, ..e would seek your comments at such time as public inputs are i invited.

l

e are ricased to have had the opportunity to review your report and look forward to receiving copies of the final draf t.

Sincerely,

,z . s;vt. / /V!s' Daniel R. Muller, Acting Director Division of Site Safety and l Environmental Analysis l

Enclosures:

As stated i

l l

90013159 l

I

DEC 4 1979 SOME SELECTIVE COMMENTS ON THE REPORT BY LELLIN AND JOSK0W ON " SITING NUCLEAR POWER PLANTS" 4 1.. A. major thesis of the authors is that there should be explicit consideration of large accidents in reactor siting and that the NRC staff appears to be -

opposed to major changes in siting policy (see footnote No.135). The authors should be aware that a recent NRC staff tast, force has recommended (Report of the Siting Policy Task Force, NUREG-0625, liugust 1979) major changes in siting policy to accomplish the following goals (p. iii):

(i) To strengthen siting as a factor in defense in depth by establish-ing requirements for site approval that are independent of plant design consideration. The present policy of permitting plant design features to compensate for unfavorable site characteristics has resulted in improved designs but has tended to deemphasize site isolation.

(ii) To take into consideration in siting the risk associated with accidents beyond the design basis (Class 9) by establishing populatica density and distribution criteria. Plant design improvemnts have reduced the probability and consequences of design basis accidents, but there remains the residual risk frca accidents not considered in the design basis. Al though this risk cannot be completely reduced to zero, it can be significantly reduced by %ctive siting.

(iii) To require that sites selected will minimize the risk from energy genara tion. The selected sites should be anung the best available in the region where new generating capacity is needed. Siting requirements should be stringent enough to limit the residual risk of reactor operation but not so stringent as to eliminate the nuclear option from large regions of the country. This is bec use energy generation from any source has its associated risk, with risks from some energy sources being greater than that of the nuclear' option.

Consequently, we expect, if the recomendations of the Siting Policy Task Force .

I are implemented, that this aspect will address the major thesis of the authors.

2. The authors have, in addition, proposed that reactors be sited at distances of 50 to 150 miles from major population centers based upon:

(i) the authors' estimates of reactor consequences which indicate that such distances minimize long-term consequences in the event of an accident.

(ii) the authors' estimate of transmission costs which indicate that nuclear plants are not uneconomical at such distances, and (iii) the authors' belief that water' availability is no problem. ,

90013160  :

i D^"JD o o Ju 7W sJu d 1 j/L l

x DEC 4 1979 A potentially controversial aspect of the autflors' analysis is the estimation of reactor consequences. The authors' have rejected existing conventional consequence models as representing " simulations" and have chosen, instead, to rely upon their own " empirical" model based on certain data. The authors have not, however, compared and contrasted the results from their model with c the conventional ohes. Without a detailed and in-depth review of the bases for each competing model (which we also have not done) it is not possible to arrive at an objective evaluation that their proposed siting option is justified over other risk assessment mdels and cost-benefit methodologies.

3. The statement on page 26 that "the staff plans to recommend to the Commission, in accordance with an NRC-EPA report, that the consequences of major accidents not receive attention in emergency planning" mischaracterizes both the report referred to (NUREG-0396) and staff actions. Recognizing it is appropriate and prudent for emergency planning guidance to take into consideration the principal characteristics of a spectrum of design basis and core melt accidents, the Commission has recently endorsed this NRC/ EPA task force report. (See encived MRC Policy Statement as published in the Federal Register, Oct. 23, 1979). This policy statement endorses the establishment of two Emergency Plming 70nes (EPZs) around each nuclear power reactor. The EPZ for airborne espmure has a radius of about 10 miles; the EPZ for contaminated food has a radius of cbout 50 miles. These distances are considered large enough to provide a response base which would support activity outside the planning rone shculd this ever be needed.

4. ha ronments on the models and underlying assumptions made by the authors in Chapter 2 are as follows:

(i) The release assumptions are not based on any empirical evidence as claimed, and are the same type of extreme assumptions. made in 1957 i in the WASH-740 analysis. 1 (ii) The biological response models are reasonable, but less precise-and less empirical than WASH-1400 models.

(iii) The atmospheric dispersion and deposition models are old and outdated, and for distances beyond a few tens of kilometr.rs, less empirical than the models now in use (e.g., UASH-1400).

(iv) In general, the assumptions (e.g., source terms) and methodology (e.g.,

meteorology) used are so simplistic that any relationship between these results and those of any other study would seem to be purely coincidental.

(v) The deletion of release probability in defining a reasonable source term results in unrealistic estimates of risk.

D* ]D ]D vYg -

90013161 o o Ju o Ju S. lRLo

.rodeal neSi s$r /._vol.~.mmm u. No. 206 / Tenday, a 12:

__..c ,,,,_ ., - --mcaober e- e- - 23. we'/ Notices -c==~

beted et Bethesda. Md. th.4 trth day of NUREG-@e. EPA 5:0/1-?S-016. dated a opriate and pn.>de.nt inc ec:.c gscy Oc ier 19:e Decernber 1978 Single copies of the p ing gudance to take into r.:L.* M.1. azo. report can be obtained by writing lo the considerstbo the principal Achef"brmari ArwicSa ery r cad Director. Division of Technical characterstics (such as nuclides Info ~ nation and Document Control. released and d.istacces likJy to be Lrersb S.42rr/P.:nct invc!ved) of a cpectrum of desya hads irm .3._ s r.- is.ve c ; Nuc! eat Fer.:!atory Commission, Washington D C. 2D555.The task force and core melt accidents. While the g;.uscch r w i

\ report w as published for public Comrnicaion recognizes that the

\ .

comment in the Federal Register on guidance enay have signiScant responas (Dociet I:o. 51.'1] December 15.1978 and the comment irrpacts for many local juri. dictions. Lt period w as extended to May 15.199 to believes that implementa* ion of the Pcci'ic Gas Electric Co.!( vmboldt allow additional cornments resAng to "ty F:wcr Pdnt Unit No Fecittry fr m the accident at Three Mile Island.

guidance is nevethelets ceeded,anning Operating t.icey'e Ho./DN'7'- A syn psis f the comments received trnprove emergency and preparednesa around r%nuc pense gat power Feconst. tut;on o Deard and the task force consideration of these reactora.

Edw ard 1 uton. %. as Chattman of comments is available from the The Cominission is direction its stas the Mom:c Safet) . ,d icensing Dcard Assistant Director for Eme gency to incorporate the planning basis f:t the abose procer@jng Mr. Luton has Preparedness. Office of State Progares, guidance into existing documents used t antfer ed to anotF/r, federal agency. U.S. Nuclear Regulator) Comtcussion. in the evaluation of state an local Ac:ordeg!). Ro deriwi Lazo. Esq Washington. D C. 0555 emergency response plans to de extent

". hose address a torafSafet3 and practicable. The NRC has reca::'Jy f it.-s.r g D.u td .nel t.1 S Lclear FI8""I"3 83535 published and Advance Notice of Va'or) Co assion, b/ashington. The major recommendation of the Proposed Rule =aking concernicg D C 2 55.is ppointed Chprman of report is that two Emergency Planning cd6tional regulations on emerpacy this S a d. onst:tuhon c Zones (EPZs) should be establahed plans. 44 FR 41484 Tuesday. Ju y 17, in th.3 rnan ; r a in mordq(he e with Scard around light w ater nuclear power plants 1979 Additional Fuidance will be

! :/,1 of .e Cornmission s 'u of The EPZ for airborne exposste has a provided fo!!owing this rulemaking This practice. .s amended radius of about to mi!es. the EPZ for additional guidance can be expected to tN f e sethese M tr.a t tr. , u of contaminated food has a ridus of about consider how local con 6tions such as octs mp 50 miles Predete mined protective demography. land use and meteorology N' M tun. acticn p!ans are neaded for the EPZs can influence tF# size an shape of the The exact s:re and shape of each EPZ EPZs and to addren other issues. such

.4rt [c3: .-cr 4neue9 ud w:!! be decided by emergency planning 4 off c:als after they consider the specific as evacuaticn planning.

Specific imp.ementation dates for full L.b 3

ego.v Ext:r Anc.

e + 2 3 m m

,k,4 33, ,,,w., cond.tions at each site.These 6 stances implementation of the task force

/ _

cre considered large enough to provide a recommendations and any others that res;onse base which would suppcrt are deseloped will be established as P!.nning Basis for Emergency activity cutside the planning zone part of the ongoinF rulemaking effort. ,

Tiesponses to fluclear pas.et Reccior shou:d this eser be needed The Cornmission also expects the staff '

AccM nts The report also provides p!anning to assist state and local governtnents in bash gWana in the form of a tange of Improsing their emergency response  ;

t.c o.cy: % clear Reg 4, tors es 6 which emergency Ume capabiliues at exist.ng sites in the i Comm:s sion. response offic:,als should be prepared to immeiate future. i acne,r NRC f4 tic) Statement implement protective action. The report )

in6 cates that. dependmg on such Dated at %.as W m D C *.Msisd ds) of ,

Purpose OctoWr m facto s as the spec:fic sequence of For the Nucl ear Rep!atory Cammission.

j This is a statcrnent of polics with events danng an accident which results regard to an Envitenmental P'r ctection in the re! ease of ra6cactivity to the S'" *3 I " ..

S'CT 'T C# 'h' C#""

Agenc) (epa) and Nuc! rat Rega!ato y atmoshpere and the prn aihng truos - r m =.",'*^ ei

)

Comm:ssien (VRC) tuk force report en met eorologm al cond. tier.s. protective pidance for uw in state and local action may be required from perhaps eec co:c we ra6o!cpca! emergency response plans one.b!f hour to oce day after the at nuc! ear pow er plants Back ground .

intiation of the accident. Dev,elopment and penode tutmg of proa.u n for k.RWhthd6mnd d

Avabbility rapid notification of emergency The NRC received a request from the response ofhcials is encouraged since The . iclear Regulatory Comms5n Conference of Rad.at:on Centrol the time available for action is strong!y staff has . 4ased a topical r i on the Prep.m Deectors an organizauen of affected by the time consumed in charactenst.o oflightnin: ' es The State oftcia!s. to "make a determination notification, repcrt was prd;qred byp National of the rnost sen ere accident basis for The chemical and physical Hurricane and cunrental whd. ra6clercal emergency rnponse characteristics of those radentac! ides MeteorolcSy Labobr> of the Nati:nst plans should be developed by o'! site which contnbute most significant!y to Oceanic and At 4pb agencies."In response, an EPA and NRC human exposure are presented. Administratio rthe h C.

uk force was establisbed which REC /Cn-1rr4 NRC Poa.f cy The.arepoyve,sment of Flashbn.

Initial sity and"An pr p4~d a report entitled " Planning Bois for the Des elopment of State and NRC cone'.:rs in and endo ses for me Local Cos emment Ra 6elogical the guidance contained in the tr.sk forte Peak C#ent IJgM..cg FlashesCharacteristicsk to Ground in uth Emergency Response P ans in Sapport of report. In endorsing this guidance the F[lda." pre sents information

!;ght Water & clear " -' 0:a nts." Com:rnsics recu,nizes that it is excerning the frequency and intens. - .

.ad WV o ~

90013162

&a a

S. On page 2 of the fntroduction, the authors."suggest that the potential consequences of a proposed technology deserve greater emphasis in risk' assessment than do the associated probabilities." Their principal reason 1 for this unorthodox view of risk assessment is that there are special circumstances surrounding the nuclear industry in which major damage may

, be caused to the fabric of society itself. The social costs of a Class 9 r

accident would, of course, be severe and should be given due weight as consequences along with economic costs and personal or individual costs.

However, it is very objectionable to ignore or deemphasize the probabilities associated with these consequences for the following reasons:

(1) The overwhelming body of literature dealing with decisionmaking un-der uncertainty treats risk as the product of consequences times i

their associated probabilities which are then aggregated and weighed.

against the expected benefits from the proposed action.

.(ii) A line-by-line comparison of risks, costs and benefits of the proposed action with its realistic alternatives is made possible only when the associated probabilities of the consequences are properly included intsmuch as alternatives will ususally have significantly different probabilities associated with the various consequences.

(iii) A deemphasis of the probabilities associated with grave social an'd economic consequences would deprive the very logic for deciding on continued improvements in safety technology (including human factors) that would reduce probability of those consequences associated with Class 9 accidents or, indeed, design basis accidents and those levels of radioactive emissions associated with normal plant operations.  !

(iv) Even if certain kinds of social consequences of a Class 9 accident l were drastically' reduced by locating plants in sparsely populated and remote reg _ ions, there would still remain the prospect of severe economic ccnsequences for the censurers of electricity and financial 90013163.

o f.0 4 T:) '

consequences for investors. The probability of these consequences needs to be given appropriate weight in regulatory decisions concerning the siting, design and operatio,n of nuclear plants.

(v) The implicit notion that the problems of public acceptance of the-nuclear option would be solved by giving greater emphasis in risk assessment to certain social consequences than to the associated  ;

probabilities is open to serious question. The notion that rural lives and the fabric of rural society are less important to the potentially affected parties than for their urban cousins -- which is implicit in a deemphasis of attention to probabilitics associated with a Class 9 accident -- runs its own risks of public acceptance for remote siting strategies: An appreciation of this concern is reflected in the current oppositientolow and high level radioactive waste

, repositories proposed for siting or expansion in lightly populated rural areas. Only through a balanced consideration of consequences and their associated probabilities can a logical defense be made of the reasonability of regulatory decisions affecting choices among sites and technological options as it affects different segments of society. ,

6. On pages 3 and 4 of the Introduction, the authors state that the " emphasis on individual risks _ underlies regulatory efforts to make nuclear power safer" and that the " essence of social life, relations amor,g individuals, plays no role in these risk assessments." The basis for such assertions was not dealt with in this report. Since regulatory policy decisions have been and continue to be based on both technical considerations and subjective judgments related to consequences, it is patently presumptive to assume regulatory decisionmakers have made safety-related decisions in complete ignorance or ascribing no weight to societal consequences other than individual risks. Indeed, the 90013164

1979

., , DEC 4 cocronplace use of " margins. of conservation" in safety-related decisions by the NRC provides, by the same token, a tradeoff of sorts for societal consequences that have not been, quantified or that resist quantification. Thus far, with relatively limited exceptio,ns, it has been regulatory policy not to provide a

~

formalized cost-benefit analysis or value-impact assessments as related to Class 9 accidents. The reasons for this policy are quite complex. However, there are two exceptions of especial note which deal with a formal' analysis of the consequences of Class 9 accidents:

(i) Final Environmental Statement Related to Manufacture of Floating Nuclear Power Plants by Offshore Power Systems, NUREG-0502,  ;

Occket No. STN 50-437, Part III, December 1978. See also the cc.c.panion study, Final Liquid Pathway Generic Study Report, ,

NUREG-0440, February 1978.

(ii) Anticipated Transients Without Scram for Light Water Reactors, HUREG-0460, Vol. 2, Appendix XII (Value-Impact Analysis),

April 1978. I It should be noted that analytical efforts to assess the economic and societal

{

consequences of Class 9 accidents as reflected in the above studies are pioneering and that both inhouse efforts and contract research are being devoted to the  !

improvement of the methodologies and the assessment of risks, costs and benefits.

It is possible, in view of experience gained with the TMI accident, that insufficient attention has been given in these studies to impacts affecting the

" Fabric of society" as well as other impact issues. See the enclosed paper by Miller Spangler on " Methodological Concepts and Issues Involving Psychic Costs .

r and Benefits, Risk Perception, and Risk Aversion in the Use Management of Alternative ,

Energy Sources," which appears in Volume II of Changing Energy Use Futures ,

(Pergamon Press,1979).

90013165

_, -. --y A

7 ,

-T- CEO 4 197 1 ,

7. On page 7 the authors propose, among other items, setting " threshold"  !

consequences that trigger special regulation of particularly hazardous industrial activities. It would be helpful to call their attention, in this regard, to the description of NRC policy and practices pertaining to the Section on " Hazardous Activities in Plant Vicinity" on p. 20 of the Report of the Siting Policy Task Force (NUREG-0625). ,

8. On page 7, in their alternative approach to nuclear siting, the authors propose, among other features, " placing a heavy burden of proof on those who propose a particular site to show, primarily through the use of empirical data, that the chosen policy objectives will be met." Again, on page 9,in recognition that "neither precision nor certainty in tha 2chievement of safety goals is possible," their proposals " require instead that particularly hazardous industrial activities meet a high qualitative )

standard of proof that explicitly stated safety objectives will be met."

This proposed treatment is deserving of a number of comments:

(i) The concept of a " burden of proof:" regardless of on whom the burden is placed, is a sterile and scientifically objectionable phraseology when directed to impact predictiers that are basically and patently unprovable. Since the future cannot be proven ex ante, nor even all of the potential causal factors reliably identified -- much less parfectly and convincingly evaluated for all parties -- professional or expert judgment is indispensible in dealing with such matters.

(ii) Nor is the primary use of empirical data in the formulation of such judgments necessarily an adequate approach in estimating or forecasting consequences of stated proposals since: some empirical data that would be relevant to such judgment may escape the attention of the estimator; gaps in empirical data may exist; or inappropriate weights may be assigned to the causative force of such empirical data upon which primary reliance is made.

90013166

- ,..r-2 4- , v.,y

4 , LEC 4 (iii) The history of technological progress suggests that increased benefits at reduced costs ar~e made available t6 society by " learning through doing." That is to say, it is often the empirical data made available through the implementation of policy decisions or project - related decisions which provide an improved basis for subsequent rounds of decisions regarding technological changes or implementation measures.

Unless society is willing to proceed in certain technological develop-ments in the absence of proof, the potential benefits of that technology will be denied to society. Much of great benefit in improving nuclear safety is likely to result from the experience gained in the Three ,

Mile Island accident but these expected benefits are still not a provable matter but must be dealt with by competent professional judpent.

(iv) As for the notion that " qualitative standards of proof" should actually be used to determine that " explicitly stated objectives" have, ex post, been met, this could encounter considerable public controversy percisely because of the imprecision of quatitative s tr.adards or because of a basic lack of understanding by the lay public of the explicitly stated (technical) objectives or else lack of agreement over their desirability.  !

9. On pages 7 and 8, the authors describe four siting alternatives which, as characterized by the table on page 21, encompass the following ranges of distance from urban areas: (i) less than 25 miles, (ii) greater than 25 miles (but presumably less than 150 miles), (iii) minimum of 150-200 miles (but no maximum stated), and (iv) hundreds of miles. Aside from the overlapping nature of the latter two categories, in regard to the future development of renote siting policy it would appear that the second of these distance categories (namely, 25 to 150 miles) will rcove to be of greatest strategic importance.

Accordingly, such a range of distances would appear much too gross for policy 90013167

,g, OR ' 'JS considerations and perhaps should be divided into 25-mile sub-intervals up to a distance from urban areas of at least 100 miles. Nor should a parametric analysis of siting alternatives in regard to safety impacts be limited only to distances from urban areas of a stated minimum size but due consideration also should be given to wind-rose patterns, present and projected population densities within concentric rings and sectors in closest proximity to the plant site, as well a* transportation ease or difficulty of emergency evacuation.

10. On page 10,the authors call attention to the diversity of views in society l l

between (i) those who feel the number of potential fatalities can be tolerated if their probability is sufficiently low, and that it is more

" cost-effective" to use engineered safety features and emergency planning to reduce probabilities to " acceptable" levels than to site reactors further from population centers, (ii) those who will assert that while we do not live in a risk-less society, industrial activities which involve the possibility of catastrophic reactor accidents are simply unacceptable, and (iii) those who will insist that given the mere possibility of large numbers of fatalities, a restrictive siting policy is in order only if it can be implemented at a "reasonsble" cost, without unduly delaying socially valuable technological development. The authors go on to say that "the relative merits of these views cannot fairly and effectively be evaluated through a mechanical cost-benefit analysis" and, moreover, that their alternative procedure l

" recognizes the inherent differences in values underlying the nuclear debate..."

The authors recognition of the diversity of values and differences of views in safety, the need to evaluate them fairly and effectively, and that this can't be accomplished by a cost-benefit analysis that is reduced solely to mathematical formula are basically valid observations in their general form.

However, the specifics of their approach such as the scope of impact issues 90013168

DEC 4 1979 considered (or not considered) in evaluating siting policy options and.  ;

' the neglect of probabilities associated with consequences, the gross manner by which are characterized the vario'us segments of society having diverse views on the acceptability of nuclear risks in relation to other costs and benef,its including other technological options, and the lack of attention to ethical principles of decir,ionmaking based on criteria of what is fair and equitable (not only regarding the present generation but to further generations as well) are highly questionable aspects. For example, the authors fail to discuss how regulatory decisions should, if this is at all practicable, reconcile the views of those who find the possibility of catastrophic accidents of nuclear plants unacceptable (regardless of whether they are remotely sited or not) with the views of those who also are concerned about catastrophic nuclear accidents but prefer to make up

- their minds about the acceptability of the nuclear option by weighing such  ;

I considerations as: safety and health in the light of proposed siting and  ;

l technological designs and operating procedures as well as confidence in regulatory practices; a wide range of environmental, economic and social l aspects; and their related concerns for similar risks, costs and benefits i

of other fuels for generating electricity both for their own welfare and that of their children and grandchildren. There are, indeed, significant segments of our society who are concerned about morbidity and deaths j l

associated with ccal-fueled electricity generation and such potential  ;

catastrophies as war over oil supply failures and the greenhouse effect due to the combustion of fossil fuels that could alter clitnate and inundate l l

coastal cities with rising ocean levels. "Catas trcphy", of course, can be viewed at the personal or individual level as well as the social level. I i

Unchecked inflation of energy prices along with induced inflationary effects j on the economy as a whole can be fewed as nothing short of catastrophic by numerous retired persons or other low-income groups. The problem of determining differences of values in society affecting energy policy is a cost difficult one and subject to the dynamics of changes in both values 90013169

( _

]

and the levels 'and kinds of information which feed them, , Provocative

-t questions have been raised by Aaron Wildavsky in his article, "No Risk

' Is the Highest Risk of All" (American Scientist, Jan.-Feb.1979):  ;

Is it our environment or ourselves that have changed? Would people like us have~ had this sort of concern in the past?

Imagine our reaction if most of modern technology were being introduced _today. Anyone aware of the ambience of our times must be sensitive to the strong possibility that many risks, such as endless automotive engine explosions, would be post-ualted that need never occur, or, if they did, would be found bearable. Wouldn't airliners crash into skyscrapers that would fall on others, killing tens 'of thousands? Who could prove othentise? Even today there are risks from numberous small dams far exceeding those from nuclear reactors. Why is the one feared and not the other? 'Is it  ;

just that we are used to the old or are some of us looking

> differently at essentially the same sorts of experience?

Our society is obviously engaged in some serious soul-searching regarding energy options and policies not only as a result of the TMI accident but also due to the recent Iranian developments and the unsettled, or uncertain,  !

outlook for energy options with which to replace oil imports. Semehow these concerns need to be factored into a cost-benefit analysis and this will undoubtedly include numerous "non-mechanistic" elements in whatever approach is decided upon to make the benefits of the nuclear option rore socially acceptable.

11. On page 33 the authors take the position that "further nuclear development l

appears desirable and is probably inevitable." They did not, however, isolate l 1

the considerations which lend support to these views. Yet, the specific char-acter of the net benefits of nuclear power which would make it desirable and inevitable are of potential importance to the evaluation of technological options and the development of siting policy including the exploration of equity con- ,

siderations regarding the diversity of values of different segments of the public and the appraisal of tradeoff strategies or mitigative measures.

12. In the Su. mary, the authors affirm that their analysis " suggests that siting 50 to 150 miles from city centers would reduce the consequences of major 90013170

DEC l W3 accidents and would not damage the competitive position of nuclear power vis-a-vis conventional energy sources. Siting at hundreds of miles from cities would, however, make nuclear reactors uneconomic in comparison to coal-fired power plants." The basis for their economic comparisons in arriving at these conclusions appear t1 be limited to Chapter 3 (Costs of Electrical Power Transmission) and Chapter 4 (Siting Restrictions and Access to Cooling Water). In the narrow confines of this cost analysis, they conclude on page 39 that siting plants 50 miles further from load centers would (in 1977 dollars) increase nuclear-electricity generating costs by less than one mill /kWh and that siting plants 150 miles further would increase the cost of nuclear generation from 1 to 2 mills /kWh.

However, a number of dollar costs as well as environmental costs have been excluded from their analysis which should be part of a cost-benefit (or value-impact) comparison of siting policy alternatives:

(i) There are good reasons to believe that the dollar costs associated with the capital requi- nuclear plant itself will increase substantially with site. naving greater remoteness from urban centers, (e.g., recruitment of less skilled or less productive workers at more remote siting plus costs of delay in construction schedules due to lags in the recruitment and training of workers when limited numbers of already skilled workers can be induced to work at remote locations).

(ii) By the same token, socioeconomic stresses on community services, housing, etc. in the vicinity of the site can generally be expected to be greater for more ramote siting than, say, sites within a more comfortable commuting distance (say, 50 to 75 miles) from urban centers.

900l317i

- ._. ~ ^

~ m -

' ^ -

_ 13'- i

(

f (iii) The dollar costs' of transporting heavy reactor components to remote sites' can generally be expected to be substantially greater involving east penalties of millions of dollars, not to mention the potentiality for costs of delay when suitable highways need to be built, cleararices on overpasses increased, etc.

(iv) Potcntial costs of construction delay may materialize due to the exercise of eminent domain or public controversy and court actions associated with longer transmission corridors possibly crossing nature preserves,public parks,and recreational or other land uses of unique value.

i (v) Electric systems reliability would diminish significantly with longer transmission routes since a large majority of electrical outages are associated with transmission failures rather than forced plant shutdowns due to equip-ment failures. This could entail social costs as well as ,

higher operating costs. i (vi) The environmental impacts would be greater due to longer ,

transmission corridors. This would involve disruptions to terrestrial and aquatic biota, risks of soil erosion, lost agricultural productivity, and aesthetic impairments to certain kinds of related land or water uses.

(vii) The possibility of higher water costs for these regions ,

where the least-cost available water sources are, say, within 25 to 50 miles of urban areas and more remote locations have streams with relatively limited low-flow volumes thus requiring low-flow augmentation reservoirs to be added with costs running into the tens of millions dollars plus the environmental disturbances of such reservoirs.

(viii) The desire in some regions to avoid increased thermal loading on some streams or rivers due to conflicts with water quality improvement programs to restore or enhance certain fishery or water recreational uses. For example, such consid-erations caused the State of New Hampshire to prefer the use of seawater as opposed to inland streams for the proposed Seabrook nucicar plant.

13. The two-and-one-half page chapter on " Siting Restrictions and Access to Cooling Water" has some other deficiencies or defects besides the ones cited above.

On page 44 the authors state that unlike fossil-fired electric generators, nuclear plants reject substantially all their waste heat to circulating cooling water. While this is a true statement for nuclear plants and non-steam fossil plants (oil or gas), it is not true for coal-steam electric plants which are the principal options to baseload nuclear plants. Although thermal loadings of coal-fired plants on cooling waters are significantly less than for nuclear plants of equal capacity, they are. nonetheless substantial. On page 45 it is said 90013172

E0 4

• , that, if once-through cooling is abandoned, water availability cannot be a strong constraint oh siting. This is not necessarily true. Evaporative water losses in the use of cooling towers, for example, are substantially greater than for once-through cooling. The cumulative impact of such losses in water availability by numerous cooling towers located on the same stream or river is of increasing concern in a number of regions. Aside from the reduced quantity of water available due to such evaporative losses, there can be significant adverse impacts on water quality due to higher salinities (i.e. , total dissolved solids). On page 46 it is said that, in the Southwest, nuclear energy does not appear to be a cost-effective alternative to coal-fired plants. While the authors do not identify what parameters or factors should be part of the cost-effectiveness analysis, the NRC cost-benefit analysis for the proposed Palo Verde nuclear units west of Phoenix (see the FES) indicates that nuclear plants can be cost-effective options in the Southwest, although the dollar cost comparison with the coal option is appreciably narrowed so that dollar cost considers-tions are less controlling than perhaps other factors in this choice.

14. In Chapter 5 (page 63) the authors state that two " intermediate siting policies" (their Regions B anc C in Table 1) are particularly attractive and can be implemented at costs which do not significantly alter the relative economics of nuc1 car and coal-based electricity generation.

Uhather we agree with this conclusion or not, any perception by utilities of increased cost elements with rerote siting such as the transmission costs outlined by the authors and the caitted cost considerations cited under item 12 above plus the increased uncertainties over these cost elements-not to mention their concern over regulatory delays regarding the additional environmental impacts-may have profound influence on the outcome of their nuclear versus coal decisions. Thus, if the nuclear option is desirable in the national interest as the authors assert, it would be helpful for an independent research effort such as theirs(with no obvious axe to grind)to explore how utilities or other investors parceive the relative merits of remote siting policy options in terms of business risks associatad with the nuclear and coal options for generating electrici ty. Likewise, where States play active roles in power plant siting, some exploration of their assessments of the comparative penalties and benefits of remote siting options might also be useful.

90013173