ML19351F435

Safety Goal Issue in Historical Perspective, Draft

ML19351F435
Person / Time
Issue date:	12/31/1980
From:	Mazuzan G, Jacqwan Walker NRC OFFICE OF THE SECRETARY (SECY)
To:
Shared Package
ML19351F433	List: ML19351F432 ML19351F435
References
NUDOCS 8101130039
Download: ML19351F435 (25)

Text

.. -

aner O

l THE SAFETY G0AL ISSUE IN HISTORICAL PERSPECTIVE December 1980 L

George T. Mazuzan J. Samuel Walker Historical Office Office of the Secretary U.S. Nuclear Regulatory Commission 8101130o39

THE SAFEIT COAL ISSUE IN HISIURICAL PERSPECTIVE Since the passage of the 1954 Atomic Energy Act, when Congress charged the Atcraic Energy Chrmission with the resconsibility to " provide adequate protection to the health and safety of the public" frun radio-logical hazards, the AE and the NBC have faced the problsn of defining what constitutes " adequate protection." Although the 1954 Act contained many references to the need_ for safeguarding public health, it did not esahliah specific safety goals or delineate standards for judging adequate protection. 'Ibe AEC, therefore, received a very broad mandate but na < laarcut guidance in formulating its safety p%u

. Even after the agency began to issue its regulations, it could not spell out precise safety criteria for evaluating applicaticns for rnv-laar plant licenses.

Largely because of the rapidly changing state of nuclear tedum, logy, but also because of the press of other business, the AT was unable to establish specific generic safety goals in the early years of peaceful atcmic development. Although it recognized the desirability of clarifying the language of the 1954 Act, its own regulations were eaually vague and imprecise.

Nonetheless, the AEC followed a general philosophy in writing its initial regulaticns in the period intnaiately after passage of the 1954 t

Act.

In keeping with the Act's mandate that the Ccrnmission both encourage the developnent of the private nuclear industry and ensure public health l

and safety, the AE sought to frane regulations that were not excessively cautious or rigid. Overly conservative rules, the Ccrimission feared, would ingede the growth of peaceful anplications of atcznic energy at the outset. As Chairman Iawis L. Strauss ocmnented, the Ccrraission's objective was "one of exerting the minimtzn of control" to protect public health and safety so that it would "not impose unne ssary limitations or restrictions upon private participation in the develognent of the atczn's l

civilian uses." The' AE's Division of Civilian Application attm@ed to j

implement that philosophy by balancing safety needs and prcrrotional goals in its early regulations, which dealt with matters such as licensing moedures, safeguarding special nuclear materials, and qmlifications of reactor operators.1 The prob 1sn of defining clearcut safety standards by which to judge

. applications for reactor licenses remained unresolved. '1he regulations simply required that prospective licensees " provide reasonable assur-ance" that their reactors would not endanger public health and safety.

On that basis, the Otxtmission issued its first two construction pennits l

for privately built light-water reactors to Canntnwealth Edison and Con-olidated Edison without arousing protests or irrlimtions of public concern. The question of the AE's ecmnibnent to safety, and implicitly, of its safety goals, flared into a major ocntroversy, however, when it aoproved a construction permit for the Power Reactor Develocrnent Campany (PRDC) to build a fast-breeder reactor in Monroe, Michigan. Supported l

by several utilities and baaaad by the Detroit Edison Chrpany, the PRDC, after informal diamssion with AT staff mcznbers, subnitted an application for a constrtrtion permit in January 1956. Despite its previous crntacts with PRDC officials and the crrrpany's assurances that it could satisfactorily

.s

.- settle technical uncertainties about the fast-breeder before beginning operation, the AT's Advisory Ccmittee on Reactor Safeguards (ACRS) was troubled by the proposed reactor. After a lengthy investigation, the ACRS cancluded that too neny critical questions about the design of the fast breeder reained unanswered to give it an unequivocal endorserrent.

The ACRS tw i:d to the Camission, therefore, that until further u

research was dcne and the pcLential technical problems were fully resolved, "there is insufficient inforretion available at this time to give assurance that the PRDC reactor can be operated... without public bami."

I Nevertheless, the n=iasicn, with Thcxtes E. Murray sha' ply dissenting, voted to issue the ocnstruction permit because the majority believed the PRDC had provided "rmenable assurance that the reactor can* eventually be operated safely."2 The Joint Ccxtmittee on Atmic Energy (JCAE) and the governor of Michigan tried unsuccessfully to obtain copies of the ACRS segu.L, but they knew e2rugh about its contents to be gravely disturbed about the A E's decision to approve the permit application. The growing controversy was further fueled when three ArL-CIO unions with large Irmberships in t

l the Detroit area intervened in the case to ask that the permit be suspended.

The action of the unions resulted in a long series of hearings and court cases that did not end until the U.S. Sucrete Court ruled on the matter l

in 1961. Although the Cburt upheld the AEC's position, it did not l

resolve or even address the issue of safety. Rather, it simply affirmed that the AE had acted according to its own regulations, an:1 that in a

" fast-developing and fast-changing technology," the Oxrmission properly exercised discretionary authority to determine the safety of reactors.

Thus, the PRDC case, despite the considerable controversy and protracted legal proceedings it precipitated, did not produce a clearer definition of safety goals. Nor did it provide guidance for judging other license applications or framing generic safety criteria. The vague goal of protecting public health and safety, without delineation of what that involved in specific terms, continued to be the basis for evaluating proposals for reactors. Subsequent court decisions were no nore helpful in defining safety criteria than was the PRDC case. They affirmed the i

AEC's respcnsibility for determini:q whether a reactor design provided I

adequate safety p yits.cn and for balancing the risks of atcmic power against its benef PmHstically, the courts did not require, nor ~did the AEC attwL to implement, zero-risk safety goals. The Ca mission, Congress, utility i

cmpanies, scientists, engineers and others involved in the develogrnent of atcmic energy recognized that the technology would always pose same I

radiological hazards. The A E sought to minimize the dangers that I

unavoidably accmpanied the peaceful uses of atcnic energy without discouraging the growth of the nuclear industry. The problem for the i

Ccrrmission, of course, was to weigh the risks of nuclear power against l

its benefits.

It addressed that issue in a 1957 analysis, " Theoretical l

Possibilities and Consequences of Pajor Accidents in Large Nuclear Power l

Plants" (WASH-740). The AEC prepared the report for the information of

(

l the JCAE in its M1iharations on legislation to provide liability insurance for nuclear plant owners. NASH-740 concluded that although the con-t sequences of a nuclear acci+nt could be exceedingly grave, the possibility of such an ocetm ence was "excWingly low."

Scientists at Brookhaven National Tahvatory, the authors of the report, nade clear, however, i

l that experien with nuclear reactors was insufficient to " afford a dependable statistical basis for estimating the probability of oo::urrence of serious reactor ami6nts in the future." NASH-740, thercfare, was at best, inprecise, and at worst, inpressionistic in evaluating and quantifying the bamtis of operating nuclear plants. Furtheun.u.=:, it did not address the question of standardized criteria for licens plants and gave no guidance cn the matter of generic safety ' goals.

In the absence of general stardards that could be applied in evaluating applications for nuclear plant licenses, the AT continued to judge the on a case-by-case basis.

In considering proposals on their individual nerits, the Cmmission relied heavily on detailed analysis of each application by the regulatory staff and the ATS. Although the Ca mission acknowledged the desirability of clearrut criteria, " capable of cm-paratively autmatic application," the state of the technology made it

[

a perplexing, if not unsurnountable, challenge. Development of atcmic energy was proceeding so rapidly and in so many directions that soecific safety criteria were likely to becme obsolete as soon as they were finalized.

Since there were no standardized designs for power reactors, it was extrcmely difficult to draw up standardized safety goals.5 i

In only two areas, site criteria and radiation protection standards, did the ADC au.up. to arrimlate specific safety goals.

In both cases, however, its efforts failed to resolve fully the issues and raised con-siderable controversy.

In 1959, the Cmmission published for public l

cmment a " general statment" of factors that had to be considered in I

the selection of sites for nuclear reactors. They included population density and meteorological, hydrological, and geological conditions.

The proposed rule, which the AE viewed as a "first step toward developnent l

and publication of ccuplete site criteria," evoked, in the words of one camissioner, "a good Mal of unfavorable reaction." Even when the Cmmission published its site criteria in the Code of Federal Regulations l

in 1962, it emphasized that they were intended only as an " interim guide" because " insufficient experience has been a m = Hated to permit the writing of detailad standards."6 In prcmulgating radiation protection standards, the AE could draw on greater experience and scientific investigation than was available on other safety matters.

But even in that area, many uncertainties and disagreements among experts existed. In fornulating radiation standards, as in evaluating the safety features of proposed reactors, the Cmmission had to balance risks against benefits and act an problmatical information.

When it issued radiation standards for the protection of workers in t

j

l t 1

i l

l nuclear facilities and the general public in 1957, the AEX' followed the rectrmendations for maxinun pennissible doses of various radioisotopes developed by the National O:mnittee on Radiai-ion Protection (NCRP). The TRP was a quasi-official group of experts which had considered the i

effects of radiation and published findings an tolerable levels of exposure since 1929. Despite those efforts and the respect they received frcm scientists in the field, many uncertainties remained about the l

biological and genetic inplications of the growing use of atanic energy l

and the valiaity of the standards reammended by the ICRP.

i The issue became highly cantroversial in the mid-1950s when fallout l

fran nuclear banb tests by the United States, the Soviet Union and Cceat Britain exposed, actually or potentially, the entire world pooulation, not just nuclear workers or those living near atanic installations, to i

radiation. The AEC insisted that the test explosians did not threaten l

public health and safety to any significant degr and that the levels of fallout were well within permissible limits. Sane respected scientists disputed that position, Incver, and accused the Catmission of deliberately playing down the dangers of radioactive fallout. Corgressional hearings and public debate failed to resolve the issue, largely because even experts in the field acknowledged that understanding of the effects of radiation was still limited. They pointed out that scien could not provide definite answers on what constituted a safe dose of radiation.

Since exposure to even small concentrations of radiation posed at least sme risk, the determination of a tolerable dose was necessarily a political decisian based on risk / benefit analysis. 'Ibe AEC, again l

following the lead of the ICRP, revised its regulations to lower the l

permissible levels of exposure to radiation in 1959. But controversy, i

uncertainty, ard apprehension about the effects of radiation were not stilled. Even in an. area that was nore rmaily quantifiable and where knowledge and experience were greater than in other aspects of nuclear technology, the AEC found it impossible to establish clear and urquestion-able safety criteria.7 l

Throughout the early years of peaceful atanic developnent, the Catmissicn maintained that its primary concern was public safety. In 1958, for example, it declared:

"There can be no doubt that public safety is the first, last, and a permanent consideration in any decision on the issuance of a construction permit or a license to operate a nuclear facility." 'Ihe AEC's performance in matters such as the PRDC case and radiation protection standards, however, raised questions in nany minds about the agency's cratmitment to safety. Its regulatory responsibilities were generally subordinated to prmotional activities and nuclear weapons develognent. As of 1960, the ammissioners were l

devoting less than one-forth of their time to regulatory issues. Other demands on their attention, energy, and resources precluded a greater effort to define nore clearly the agency's safety goals. The ammissioners were frequently insulated fran the regulatory process because they relied heavily on the findings of a hearing examiner in uncontested

I

' l I

cases.

Even when they did play a direct role in regulatory decisicms, they did rot clarify safety criteria beyond the ceneral objective of providing adequate protection of public health and safety. Nor did they l

provide much specific guidance to the AEC staff on regulatory policy.

Although the crmmissioners constantly had to halance the risks of atcnic l

energy against its benefits, either implicitly or explicitly, in their deliberations, they did not develop overall guidelines for doing so.

l The void left by the otrmissioners on those matters might have been fillM by the Advisory Ccumlittee on Reactor Safeguards, but it was too occupied with reviewing individual license applications to develop generic safety criteria or make reammendations on general regulatory policy.8 The Joint Ccumittee on Atcmic Energy carefully and often critically l

examined the AEC's regulatory activities. In 1957, after the controversy generated by the construction permit granted the PRDC, the Joint Ccnmittee made a study of AT licensing pnacedures. One issue it considered was the creation of a separate agency to handle only regulatory matters.

The Ccomittee detemined, however, that dividing the AEC was not a practical course of action, though the agency did m9 scme organizational changes to give its regulatory staff somewhat greater auis scrf. The JCAE again porvlered the possibility of a separate agency when it scrutinized l

the AT's regulatory process in 1961. One advantage that was cited for i

dividing regulatory and promotional functions was that a new agency I

could devote rare attention to spelling out nere precise safety goals.

[

Once again, the idea of an independent agency was rejected, though, as in 1957, the AEC reorganized its regulatory framework. It established the post of Director of Regulation, who for the first time, became directly responsible to the ammissioners. The new organizational mt-l up, while reducing the isolaticn of the crrmissioners frcm regulatory l

problems, did not produce any significant changes in policies or priorities l

or any new efforts to better define safety criteria.9 The delineation of safety goals remained a desirable but unattainable objective. The AEC maintained that after nuclear technology further progressed and became nore developed, it would be possible to establish

. clear criteria for jtriging reactor applications and evaluating the performance of plants in operation. Ccnrissioner Ioren K. Olson ob-served in 1960:

"We expect that reactor technology.... site criteria, and other factors affecting safety questions will scm be better defined."10 The JCAE and spokesmen for the nuclear industry agreed that standardized gnW1%es would be beneficial for bcath regulatory and prtmotional purposes. But in the period 1954-61, when atmic technology was rapidly changing and the Ccumission was often preoccupied with other matters, I

the articulation of precise safety goals, Irwever desirable in theory, j

proved inpossible to achieve.

By 1962, the AT was in a position, as a result of the Power Demon-stration Reactor P w mu, to be optimistic about the future economic prospects for power reactors, particularly light water mcxlels.

In a key

l

\\

l

. 1962 report to the President, the Ccmnission conclMM that nuclear power was "on the threshold of econcmic ccurpetitiveness" and that relatively redest financial assistance by the govement would ensure that goal.ll Thirteen manths later, in December 1963, the Jersey Central Power and Light Ctruoany announ d its purchase of a 515 MWe boiling water reactor frcm General Electric to be constructed at Oyster Creek.

This facility was the first of thirteen " turnkey" contracts negotiated between 1963 and 1967 by GE: and Westinghouse with various utilities.

Under the terms, the vendor agreed to construct a amplete generating facility for a price that could change only according to certain inflationary indices. The turnkey develo; ment reflected the 1962 Am teru.L, indicating I

hath vendors' and utilities' opHmi=m over the cx2ncetitiveness of nuclear power with fossil-fueled plants.

In addition to the turnkey systems, utilities ordered twenty-seven non-turnkey plants during the same period.

Thus the Oyster Creek development was a milestone that began the " Great Bandwagon Market" for light water reactors in the United States.12 The optimistic 1962 repet did rot overlook the' potential safety i

problems that would be encountered in the rapid expansion of power i

reactors. Yet it gave little enphasis to them. Stating that continued l

research was needed on the "rossible spread of fission products in case l

they do escape frcm the reactor and its containment vessel," the report I

cited continued reliance on containment and on sites remaved frcm population centers. The report particularly noted that operating data l

was not yet available to determine many safety questions. Reliance on conservative subjective safety evaluations, made on a case-by-case basis, would be eliminated once enough operating experience permitted new statistical evaluations that would lead to solutions of generic problems.13 In other words, the report implied that AT regulators would have difficulty coping with the fast technological evolution of reactors.

The underlying prenise of the A E regulatory staff's mission was to provide adequate protection of the public health and safety. 'Ihe 1961 PRDC opinion and the strong statements made in the case by the A E that public safety was the "first, last, and permanent consideration" indicated that the agency did not condone any approach giving equal weight to

-other, non-safety factors.14 '1b the contrary, however, the 1962 report to the President cited eaancnic considerations in several instances in discussing reactor safety and limnsing. Ebr example, in a brief staterrent on reactor containment, the report stated that the " efficient design of containment vessels must be studied and exploited with a view to decreasing I

costs." In regard to the siting of reactors, the wru.L explicitly i

mentioned how remote siting "aMa both to transmission systen costs and to expensive power losses in the lines." At that tirre, prulence dictated "that large reactor installations be fairly far rerroved frcm population centers," but econcaic inplications made different alternatives appealing.

The report conclMM: "With adequate technical improvenents and the l

accumulation of satisfactory experience, it should be possible to gradually re::ove many of the siting restrictions in force today."15 l

i l

l l

l

. The agency recognized too, the problems caused by basing the licensing process an a case-by-case review. The regrL affinted that standardization of reactor designs would reduce costs. It noted that once " sufficient data are available," the agency would be able "to evaluate the econcmic inpact of special safety features and hence address ourselves to steps

[in the linensing process] to minimize their costs."16 So while' risk / cost l

analysis was not explicitly a part of the safety determinations cn reactor designs, the regu.L dertonstrated how pressure frca the develop-nent side of the AE affected the regulatory operaticm in meeting its broad safety mandate of adequate protection for the public health and safety. The regulatory staff did not work in a vacutn.

It was subjected l

to pressures frcm within the AK, frcm the utility and vendor industries, and the Joint Ccmnittee cn Atomic Energy, all of which wanted rapid developnent and licensing of nuclear facilities. 'Ihe regulators had no intention of constraining nuclear power's connercial use. While they continued to require what safety neasures were available, and to apply them as conservatively as possible short of disrupting the camercializa-tion process, they still lacked clear safety standards on which to base their judgments.

The little work done an quantifying the risks associated with nuclear power had been sunnarized in NASH-740, but it did not attstpt to define acceptable risk. The AEC's definitions of regulatory objectives continued to be subjective in the early 1960s.

In 1961, Clifford Beck, l

then an assistant director in the ADC's Division of Licensing and Regula-l rid the JCAE that the AEC must reduce the liklihood of an accident

" lowest practical level" and " minimize hazardous consequences."

l w tter Zinn, former head of Argonne Laboratory, suggested to the JCAE l

that the goal was to " reduce the probability of serious hazard to the public to a low enough value so that the risk is ocarparable to other risks which are found acceptable in our society."17 These definitions acknowledged and assumed certain risks in nuclear power technology.

i Given the statEWaf-the-art and operating experience with reactors at that time, a nere precise delineation of the acceptable level of risk was unattainable. But with the growing prospects for greater crmrerciali-zation, the AT realized that a clearer definition of acceptable risks and safety standards, backed by regulations, was necessary.

To assist in the review process, the regulatory staff evolved four types or categories of measurestent to apply to the reactor applicaticns.

In regulators' terminology, standards were definitive recuirements specifying procedures for obtaining and confirming performance require-ments. While scue standards had been issued (10 CFR 20 cn general radiation protection; 10 CFR 50 cn licensing procedures and general plant design requirements, and 10 CFR 100 cn site considerations), by w hensive standards the mid-1960s nost regulators agreed that nere e

could not be agreed upon until the rate of technological change stabilized.18 Codes constituted special types of standards that detailed design require-ments for ccuponents or systens. There were usually drawn up by the professional societies (for instance, the American Society of Mechanical Engineers already had a boiler and pressure vessel code). Both standards

. and codes could be rigorously applied. The two other categories were later developed to b mdle the large number of ambiguous situations.

Criteria were general pdormance objectives against which to capare an applicant's proposed design. Guides gave suggested ways to meet possible safety issues expressed in the design criteria.

Reactor manu#ah had pressed the AEC for design standards since the early 1960s. Although the regulatory staff did not have the information necessary for drafting such standards, by the mid-1960s it produced a set of general criteria as interim aids (for exanple, the i

General Design Criteria for Nuclear Power Plants, a Guide for Organization and Contents of Safety Analysis Peports and a Reactor Site Criteria).19 While not inviolate like standards and codes, the criteria reflected positions that were acceptable to the regulatory staff. The regulators scrutinized reactor designs %ss in the review process if applicants caplied with the criteria. But obviously, the AEC was still feeling its way during those early camercialization years.

In hearings on reactor licensing and regulation in 1967, Beck, by then prm oted to Deputy Director of Regulation, outlined for the JCAE the basic goals of regulatory review and the major censideratiors affecting reactor licensing.

The overriding safety objective in design and operation of nuclear reactors is to assure that the fission products produced within the fuel do in fact reain confined at all tines. The important

(

question in all reactor safety analyses and evaluations, and in the evaluation of safety sites is: will the fission products reain within the structures? Or will scrne accident occur which would release and distribute a significent portion of the fission products into the er.hmeut? Between the two possible answers to these questions lies the difference between security and the possibility of serious danger.

Thus the possibility of accidents, what kinds of accidents, and what their consequences would be became matters of major importance l

in the safety analysis and evaluation of reactor facilities.

Beck cited the uncertainty of the experts:

The fact is, no one is in a position to dertonstrate that a reactor i

accident with consequent escape of fission products to the environment will never happen. Although it is expected that fuel failures will occur that will release substantial quantities of fission products into the primary system, the probability of a major fission products escape frmt a reactor facility built to present standards is certainly very low; no one really expects such an accident, but, no one is in j

r. position to say with full certainty that this will not occur.20 Beck could speak frcrn his own recent experien He had been chairman of the steering group assigned in 1964 to direct an update of NASH-740 on the probability and consequences of such an accident.

. Essentially ccmpleted in draft form in 1965, the report, in com-parison with NASH-740, greatly increased the estimte of deaths, injuries, and propq damage prevkvwl if a major accidental release of fission prodtets occurred. But the new sttriy, based on a meager data base that its authors admitted would be ac pted only by a " fringe m mhar of the' statistical cmmunity" was not fully cmpleted nor publicly released until 1973.21 Like the original 1957 sttriy, the updated version of WASH-740 showed the air's inability to quantify or express with precision the risks inherent in nuclear technology. The Am, therefore, continued to use the broad safety standard of providing adequate protection based on the limited operating experience of the previous ten years and the case-by-case engineering jtrigments the regulatory staff and' the ACRS made.

Although the regulators were unable to define quantitatively the level of risk expressed by the probability of an accident, they were sanguine that using their established procedure, licensed reactors would be safe. In 1967 testinuny, Beck underscored the " vast difference" between the " potential hazards" to health and safety and " actual dancer."22 Spanning that gap were the safety systems that had been developed to that time. Three lines of defense were used:

(1) cuality in design, construction, and operation of the basic reactor system including such itsas as containment, the primary coolant system, and sheathing of fuel pins in stainless steel or zirconium cladding, (2) accident prevention systems including redundancy in controls and shutdown devices, independent stergency power, and emergency cooling systems, and (3) consecuence-limiting systems such as building sprays and washdown systens, and internal filter-collection devices.

'Ib insure that these systems were effective, the regulators applied their own evaluation of reactor safety system designs throuch required

(

safety evaluation @d.s, independent reviews by outside experts including the ACRS and periodic inspections during construction and actual operation.25 This rather informal, case-by-case review, used frcxn the beginning of the licensing process, created continuing problems.

While the 1963 Oyster Creek plant heralded the enset of expanded I

camercialization, a large increase in orders for new plants did not I

occur for three more years. The sivh1 surge in applications for con-struction permits in 1966-67, however, threatened to engulf the A E regulators.

It allowed then little w.u.Lanity to develop standards to measure clearly the level of adequate safety.

The period 1967-69 was critical in the history of nuclear safety regulation.

If the regulatory staff had possessed the power to control events during that time, it might have taken action that averted later problems. But because of the convergence of several trends over which the regulators had little or no influence, it later became evident that they constantly fought a rear-guard action to gain sway over the situation.

. The nuclear irdustry believed that nuclear technology, lack of ooerating experience notwithstanding, was well in hand by the early 1960s. 'Ibe industry pushed hard, for example, for permission to under-take urban and seismic area siting, all the while arguing that nuclear power posed no appreciable risk to society. Teading nuclear industrialist Chauncy Starr told a meeting of the Amrican Ntriear Society in 1965:

" Safety is a relative matter and I believe we have reached a point in the denonstrated safety of nuclear power to say nuclear power is safe, period." Buoyed by such arguments, the industry jtmped on the " Great Bandwagon Market" and greatly increased both the nunber of applications t

l for construction permits and the power densities of the new reactors.

The nunber of construction permit applications went frtun five in 1966 to 14 in 1967 to 21 in 1968 to 25 in 1969. Of even greater inpviLance was the proposed power outputs of the new reactors. The average power capacityofreactorsorderedin1965was650MWe;ig41966, 800 MWe; in 1967, 850 MWe; in 1968, 900 M4e; in 1969,1025 MWe.

The significanoe of this trend was apparent in altering the risks inposed by nuclear power.

If, as nest experts hgreed, the risk was the probability of an accident multiplied by its consequences, then as the number of new reactors increased, so did the protability of a reactor accident.

Likewise, as the reactor power size increased, so did its inventory of fission products, thus magnifying the consequences of any accident.

Another significant occurrence of the late 1960s had an indirect bearing on the question of reactor safety goals. Passage of the National Environmental Policy Act in 1969 and the Water Quality Inprovement Act in 1970, coupled with the Calvert Cliffs court decision in 1971, placed an expanded workload on a regulatory staff already overbtnxiened with problems in the radiological area.2$ These landmark events greatly affected the regulators' work and it was a situation over which they had little ocntrol.

Within the AEC and at the Joint Ccx:mittee on Atomic Energy, the develo;xnent side of the technology dczninated. Most comnissioners, with the exception of John G. Palfrey (1962-66) and James T. Ramey (1962-73),

i showed little interest in regulatory matters. While the JCAE constantly reviewed the agency's prop am of regulation and licensing, one is struck by a reading of the extensive hearings held in 1951, 1967, and 1971 with both the Ccmnittee's and the ocmnissioners' overriding concern with the trappings of the licensing process (public hearings, multiple reviews, reorganization, etc.) and the means by which it could be streamlined to bring reactors on line nore quickly. JCAE mernbers dutifully listened to the concerns of the regulators and ACRS members when they testified about problems concerning standards, the fast evolving technology, and nore safety research premrams, but greater enphasis always seemed to be on licensing procedures.26 A gcod exanple of the predczninance of develvpud. within the ABC during this critical period lay in the area of reactor safety research.

The agency regulators depended on safety research that came under the control of the Division of Reactor Development and Technology. Lack of direct control over safety research p vgams that would be responsive

. to the regulatory staff's needs posed a serious probim. And when the AEC named Milton Shaw director of Reactor Developrent in 1964, he crrtmitted the division to developnent of the liquid metal fast braMer reactor as the next level of technology to the light water family. Shaw's priorities caused the light water safety research ge in the division to crrnpete I:cre keenly than ever before with developnental fast reactor research.

Without the data that could only come frts the national hhnratories, the regulators c. uld at best only hope to establish ricre definitive safety standards. Alticugh the regulatory staff and the ACRS consistently prr*1M the AEC to speed up and better fund the safety research program, the developtental research trend continued until 1973. At that time the agency created a new Division of Reactor Safety Research unde.r Herbert Kouts that underscored the iww. Lance of safg research by giving it separate funding and an identity of its own.

By then, however, the smling up of reactors crrning on line, and the additional workload imposed by the new envi-e:ntal legislation had pushed the regulators into a position of " catch up" frcm which they only started recovering.

in the mid to late 1970s.

The regulator's lack of independent control over safety research does not mean that safety research was totally ignored. Frczn the beginning of the Power Denonstration Program in 1955, safety had been a part of the research effort enccrupassing all aspects of reactor develognent:

fuels, materials, physics and otrnponents. Several programs dealt with simulated reactor accident tests. For example, the BORAX experiments, dating from 1953, tested in part the boiling water reactor's steam generation as a cutoff in case of a core meltdown. For several years a Special Power Excursion Reactor Test (SPERI) series investigated reactor kinetic behavior with various fuels, moderator and coolant types, and conditions. More recent safety studies started in 1963 with the Ioss of Fluid Test Facility (IDFr) and in 1964 with the Power Burst Facility (PBP), the forrter designed to test loss of primary coolant and the latter conducted research on fuel behavior under simulated accident conditions.28 Criticism, then, should not be directed at the types of l

safety research, but rather at the a:tount and the managment of its I

direction.

All these trends came together at a critical time and raise the question about what might have happened if the regulators had more control over their destiny. What if, for exanple, the regulatory staff and the ACRS had been able to convince the Ccmnission to attempt to slow the develognent of the technology because of the lack of stardardization and operating experience and to place a higher priority on reactor safety research. Certainly the question includes econcunic as well as safety implications. Perhaps, given the history and nature of the AEC's develognent activities prior to the 1967-69 period, it would not have reduced the growth to a slower rate. But the historical record gives no indication that the Ccmnission seriously made such an attempt.

The Ccmmission agreed publicly that the highest level of reactor safety was necessary if the product was to be sold successfully and it

. constantly ertphasized the good overall safety of the nuclear industry.

Pressure for the rapid chveloptent of the reactor industry, however, overshadowed the pressure to understand fully the safety inplications cf large scale camercial use before it was inplernented. The rapid increase in the nmber and size of reactors in the late 1960s and the continuing changes in the technology still nede specification of safety goals unattainable. While the Ca mission acknowledged their desirability, these various pressures precluded a concentrated effort to nore clearly define them.

l Two indications of the difficulty in establishing definitive safety goals during this period were exmplified with the start by the ACRS of a list of " asterisked" or generic it es, later called unresolved safety issues, and the controversy over the errergency core cooling syste (ECCS). The unresolved safety issues list began in the 1967 ACRS review of Browns Ferry 1 and 2.

These two boiling water reactors represented a larger power level increment over BKRs previously approved for con-struction. Moreover, their projected power output was considerably greater than any operational BWRs.

In a letter to the Ca mission in March 1967, the ACRS indicated reservations on specific safety issues that the cxxtmittee identified as not only particular to the Browns Ferry reactors but that were "of significance for all large water cooled power reactors, and warrant careful attention." In subsequent letters of June and July on Vernant Yankee and the Oconee PhRs, the ommittee repeated its reservations and added scme new ones. This procedure developed as a standard approach in all ACRS review letters.29 In addition, the ACRS began to list separately the unresolved issues and asked the regulatory staff for p w ess reports on their status. Through collaboration between the ACRS and the staff, by the fall of 1970, regulatory positions on sme iterns were covered by the issuance of new safety or regulatory guides. But the list continued to grow, for even as sme matters were resolved new ones arose, " partly as a result of operating experience, partly frm changes in design approach, and partly frm the surfacing or resurfacing of questions about existing criteria, designs, etc."30 The unresolved generic items posed a difficult problem in arriving at a definition of safety goals.

In answering the question of "how safe is safe enough," the regulators had to balance undue risk to public health and safety against the knowledge that a number of significant generic issues remained pending.

In nest of those cases, the staff relied on engineering judgment in lieu of detailed analysis and resolution.

Additionally, the question of standardization and backfitting had to be considered. When a particular item was resolved, it would apply to current and all new reactor applications, but a backfit decision on operating reactors had to be made. In the late 1960s and early 1970s, the AEC made such judgments on a case by case basis and only after examining alternatives.

l

. In cimilar fashion, the ET.S chbate illustrated the difficulty of arriving at an acceptable safety goal when technical issues had not been clearly resolved. Early -escial reactors, such as Indian Point I and Dresden I had very limited HIS capabilities. Containnent and to a lesser degree, siting away frczn population centers were the main lines of defense against release.of fission products in the event of a cnre meltdown. As larger units were proposed in the mid-1960s, both the regulatory staff and the ACRS became concerned that they had to apply assunptions and experimental data frczn early 100 MWe reactors to nuch i

larger units.31 In late 1965 the ACRS suggested to the Ccumissicm that sczne pro-visions be made in future reactor designs against the "unlikely accident" of a "swMan major pressure vessel failure leading to breaching the con-tainment." The ecmmittee recognized the "very small probability" of a pressure vessel rupture, but suggested that "the orderly growth of the industry, with cono:xnitant increase in number, size, power level, and proximity of nuclear power reactors to large population centers will

... make desireable, even prudent, incorporating... the design approaches... recrzmended." Among the approaches it advocated was to

" provide adequate core cooling or flooding."32 Underlying this was the ACRS concern that previous reliance on containment and siting were not adequate for the new reactors being ord: red.

The Indian Point II and Dresden III reactor applications under-scored this issue. The ACRS's 1966 review of the 800+ MWe Westinghouse (PWR) and General Electric (BWR) reactors focused on possible large scale core neltdown that might breach the containment and the engineered safeguards necessary to prevent a loss of coolant. A wide range of opinion on the problem existed in the crmmittee, which eventually arrived at a consensus that it could approve both reactors, "on the basis of greatly inproved emergency core cooling systems and much l

greater emphasis on primary system integrity to reduce the probability of a IIX%." While the regulatory staff acknowledged that further study was warranted on both the EOCS and on problems associated with core meltdown, the ACRS decided to write a general letter to the &nmission making a strong reaxmendation on the rapid developnent and future implementation of engineered safeguards including an inporved HIS. The conmiss' ion convinced the ACRS to postpone such a letter pending the report of a task force it appointed to study the problem further.33 Consequently the Ccumission appointed in 1967 a ocmmittee under William Ergen of the Oak Ridge National Tahnratory to review the IOCA and assess the adequacy of the available protective systems. The Ergen report reaxnended sczne improvements in the ECCS and pivevsed a major research effort to resolve uncertainties in the functioning of the systen. Moreover, it noted that an improved ECCS might not always work, a:x1 while a partial core neltdown probably would not breach containment, a couplete core neltdown would.34

i

. So the question of what level of uncertainty was acceptable retained unanswered. The ACRS had sphasized, and the Ergen crmnittee verified, that a couplete core neltdcwn would breach r.he containment. It was essential, therefore, to prevent the loss of coolant, but data was not available to prove that the EECS would work.35 The Ccmnission chose to accept this higher level of uncertainty and continued to license large power level reactors.

It relied on the past safety record of the industry as evidence to the public that everything was well in hand. But over the course of the next few years, other events again noved the EXIS controversy to the forefront.

Resolution of the uncertainty of the EDCS depended on expanded research on core neltdcwn that resalted from loss of coolant. But as noted above, the AEC Division of Reactor Develcp ent and Technology, responsible for safety research, did not vigorously pursue such a pr%ma in spite of Asus recomendations fran the regulatory staff and the ACRS. In the meantime, much of the attention and energy of the regulators were redirected to the new enviromental issues posed by NEPA. Then in 1970-71, a renewed concern about ECCS arose.

In March 1970, the AEC published its " Water Reactor Safety Prwom Plan" (NASH l

1146) that identified many unsettled safety questions including several that related to the EDCS.

In response, the Ccmnission appointed a task force under Stephen H. Hanauer to reevaluate the ECCS problem. The study group designed and supervised a series of semiscale tests which showed that much of the ECCS make-up cooling water would be forced out of the pressure vessel during blowdown follcwing a cnld leg break. Subsequently, controversy raged among researchers as to wnether the non-representative size and design of the test facility led to those consequences. But in any case, the disturbing results of the tests were unexpected.36 The semiscale tests alarmed the AEC enough to issue energency Interim Acceptance Criteria on the ECCS in June 1971. In response to growing demands by intervenors, the Camission ordered a separate public l

l rule-making on the criteria. That hearing, which ran frcrn January 1972 l

to July 1973, showed the depth of disagreenent over the adequacy of the criteria.

Industry thought it too conservative; envituw_ntalists, by then backed by scientific expertise, believed them too weak. 'Ibe regulators who wrote the criteria, and found thernselves caught in the miM1e, were shaken by the hearings. In D-br 1973, the AEC issued new criteria that were substantially nore stringent. Nonetheless, sufficient test data in several important areas of the ECCS obviously were not available to prove or to disprove the conservative assunptions of the new criteria. Although research on the EDCS was given greater l

priority after 1973 (when a separate safety research office was formed),

l critics continued to the present to disparage the effort for its failure to provide the basis for defensible regulatory decisions. David Okrent, longtire ACRS iner, for example, recently wrote that the "NRC safety research program on IOCA-ECCS, rather than ' quantifying' the safety margins in the existing IOCA-EDCS, is being used to reduce these margins where the Staff believes such reduction is necessary."37 I

. Even before the IICS hearings were held, James Schlesinger, who became cMi-n in August 1971, discussed the status of nuclear power plant safety with several mernbers of the JCAE.

In October 1971, Senator John O. Pastore, cMiman of the JCAE, suggested that the AEC undertake a study in prnhahaliatic assessment "which would be jast as valuable in providing the industry and public a basic understanding of our status and objectives in this critical field as the 1962 Report [ Report to the President on Civilian Nix: lear Power) was to the formulation of the civilian nuclear power p w ou."38 1

l As a result, the air prodtx:ed in 1973 "The Safety of Nuclear Power l

Reactors (Light-Water Cooled) and Related Facilities" (NASH-1250). While it showed that the AEC was attsupting to deal in-house with' this critical issue, WASH-1250 did not provide the quantitative assessment of risk Senator Pastore had suggested. A year before the final draft of WASH-1250 was issued, however, the AEC did begin a major outside study on light-water reactor risks in U.S. certmercial nuclear plants. MIT Prcfessor Norman Rasmussen served as study director and fomer AIC staff matber Saul levine as the full time staff director for the study.

The first draft of the paper (NASH-1400) was issued for omment in August 1974; the NRC issued the final report in October 1975. Extensive crxtments were received on the draft et and these were sunmarized in the final 6ocument.39 The study's objective was to rake a realistic estimate of public risk involved in potential " class 9" accidents (accidents that could lead to radiological consequences outside plant boundaries that would exceed A!r regulctions). Research focused on large pressurized and boiling water reactors. The authors used m3thodological technioues called event trees and fault trees. These had been developed over the previous ten years by the Depart 20ent of Defense and the National Aero-nautics and Space Administration to predict the effect of failures of small wuponents in large, ocxtplex systems. An event tree first defined an initial failure within the nuclear plant. It then examined the l

course of events that follow as determined by the operation or failure of various provided systems that prevent oore neltdown and release of radioactivity. A fault tree defined an undesired event identified in the event tree accident path. Then it determined how the system could fail by.means that would produce that event. Using these techniques, thousands of possible sequences in reactor failures were assessed for their occurrence probability.40 Tha basic conclusion of this amplex study stwted that risks to the public from potential nuclear accidents were small ocznpared to other forms of risk in a amplex technological society. Nonnuclear accidents that the study ampared included fires, explosions, toxic chemical releases, dam failures, airplane crashes, earthauakes, tornadoes, and hurricanes.41 Even as the reactor safety study was being prepared, other events highlighted the growing concern cr er the safety of nuclear plants. The beginning of a national energy crisis in late 1973 gave inpetus to an

. idea dating frcm 1971 to create a consolidated Deputient of Natural Resources that would incorporate the nuclear developnent activities of the AT.

Eventually, this resulted in the Energy Reorganization Act of 1974, which established the Nuclear Regulatory Ccmnission as a separate, independent regulatory agency.

In effect, the new law underscored a twenty year old assunption that when cmnercial nuclear power reached naturity, the AEC's regulatory functions would be reorganized in an independent agency. The NRC began operations in January 1975 under the legislative mandate of the 1954 Act to protect the public health and safety.42 Before the NRC could consider its overall goals, it was faced with scute significant unanticipated reactor operational probles. Iess than two weeks into the life of the agency, hairline cracks in safety-systen pipes were discovered at several operating plants. The Ccrimission ordered inspection of twenty-three similar reactors and undertook an intensified study of the causes. Then, in March 1975, a major electrical fire at the Browns Ferry nuclear station in Alabama p the NBC to detail tM accident's implications for other facilities. 3 These events trade it obvious that the agency would not have a " honeymoon" period to sit back and address broader safety goals.

while the Browns Ferry fire and the hairline cracks raised nore l

public concern over the safety of nuclear pcur, the agency's release of the final draft of MSH-1400 in the fall of 1975 seemed, at least initially, to put the controversy to rest. The atcrnic industry hailed the report as affirming nuclear power's safety. The Ccximission did not formally endorse ESH-1400, but its press release arvmpanying the publication described it as a "rmlistic assessment..., providing an objective a d meaningful estimate of the present risks associated with the operation of present day light water reactors in the United States." At the sane time, Chairman William Anders stated that the " final report is a scedly based and inpressive work.... Its overall conclusion is that the risk attached to the operation of nuclear power plants is very low i

cmpared with other natural and man-made risks."44 A wide spectrun of the scientific, technical, legal, and public interest crxmunities studied the reactor safety recort. Although nest u.mmerded it for advancing the quantitative methodology of nuclear risk assessment, the study also aroused considerable criticism. Meanwhile, the agency cautiously began in-house work on the application of the l

study's methodology and insights on a variety of related nuclear matters.45 In early 1976, three General Electric engineers and an NRC project manager resigned their pocitions and charged the industry and the agency with condoning unsafe conditions and following inadecuate regulatory priorities. Those charges pmted the JCAE to investigate the broad area of nuclear reactor safety. In testinony, Ccanissioner Victor Gilinsky suggested to the JCAE that the cuestion of defining safety l

I 1

[

. had always been skirted and that perhaps the time was agr upriate to address the issue. Gilinsky noted that for a ntmber of reasons, nostly econcmic, industry growth had slowed. Therefore, it appeared to be an opcetune Incnent in which to " consolidate previous efforts." Congressman Mike McCormack pursued Gilinsky's ocmment by asking how one could measure safety. Gilinsky replied that the recently published reactor safety study was one way to measure safety: "It is kind of a report card on the safety of reactor systems as they are regulated now."

He added:

"But beyond that, I think one might consider an explicit safety goal, which we do not have now. I think it would be useful to try to state such a goal." Asked if he thought it was possible to be precise in establishing goals, Gilinksy replied: "I think we will never get it to the point where it is nerely an arithtetical calculaticn. However, I think we can reduce further the element of judgment in safety decisions."46 Asked to discuss the issue further through corresponden, Gilinsky wrote that the question of reactor safety was a pressing public concern that could not be rertoved by the paradoxical fact that "we have no experience with large nuclear accidents to confirm our estimates of probable public risk." Consequently, regulatory safety decisions had to

" rely heavily on calculation and technical judgment. Many people - in goverrcent and in the public - are reluctant to accept regulatory decisions... that are based on 'judg: tent. ' Any replacement or reinforement of judgment with nore explicit quantitative criteria may go a long way towards enhancing public confidence."47 Gilinsky saw the reactor safety study as a step in that direction, although nore work would still be needed.

" Society," he wrote, "must weigh the alternatives and decide the ultimate ouestion of what aan-stitutes an acceptable risk in the use of nuclear power." He believed the agency must nove strongly to devise "sme kind of cuantitative standard... which the public can ac pt or reject."48 Gilinsky admitted the task of developing a safety goal would be difficult and could not be accmplished in a short tine. In March 1977, Chairman Marcus Rowden, responding to a state. tent on reactor licensing developed by the Atcnic Industrial Forum, surtmed up the on-going problem of defining acceptable risk.

There is much merit in your view that it would be useful to have established levels of acceptable risks for use not only in the licensing of nuclear power plants, but also in many other fields of activity of our society, such as the major energy alternatives.

NRC's frequently stated position, that any analysis of nuclear risks must be considered in the context of risks posed by other energy sources, is consistent with this view. Hcuew.r, the techniques for defining what are acceptable levels do not currently exist, nor is there now any established means for develooing a public consensus on the level of risk that is considered to

. be acceptable. A halancing of quantified benefits and risks would be needed as part of the process for establishing such acceptable levels of risk. Although the Reactor Safety Study provides a quantification of potential accident risks, it does not attetpt to quantify the benefits to be derived frcxn the use of nuclear power plants.

'Ib our knowledge, nethods to quantify benefits to be derived frcm an activity have not been developed and such develop-mnt would appear to be extraordinarily difficult. Even if benefits could be quantified, the halancing of such benefits against risk would represent a further area of difficult develop-ment.

Nonetheless, we believe that it would be useful to proceed with the inprovement of techniques for cuantifiying risks, as discussed above, with the recognition that developnent'of a means of quantifying acce;*ah414ty of risks would take a nutber of years and would involve a broad-scale effort that should properly draw ontheresourcesandexpertisegfotherfederalagencies, private organizations, and the public.4 Continued public misgivings about the reactor safety study, and in particular its " executive sumary," caused Congreessnan Morris Udall, Chairman of the House Ccmnittee on Interior mi Insular Affairs (one of several cxmmittees that replaced the JCAE after its dissolution in 1977) to ask the NRC to review the report's approach and conclusions. The agency appointed a Risk Assessment Review Group in July 1977 (informally called the Lewis Ccmnittee after its chairman, Harold Lewis) to review WASH-1400.

It Key.u. Led to the Ccmnission in the fall of 1978. 'Ihe Iewis Ccmnittee praised the study's general nethodology and ecognized its contribution to the assessment of the risks of nuclear power. But the Iewis report questioned the executive sumary, the peer review process used in producing the final report, and the calculations in the body of the report.

Its cczments on the executive sumary were particularly negative.

It was "a poor description of the contents of the 1e

.u.L, r

should nto be portrayed as such, and has lent itself to misuse in the discussion of reactor risks." The ccmnittee thcaght that one who read only the executive sumary "may be left with a misplaced confidence in the validity of the risk estimates and a nore favorable impression of reactor risks in omparison with other risks as warranted by the study."50 The review group we.u.L puuyled the Nuclear Regulatory Ccrimission in January 1979 to issue a policy statement withdrawing any past endorsenent of the executive sumary of the report, agreed with the Iewis Comnittee that the peer review process in NASH-1400 was inadequate, accepted the review group's conclusions that NASH-1400's absolute values of riske should not be used uncritically, and regarded as unreliable the Rasmussen study's nuTerical estimate of the overall risk of reactor accidents.51 On the eve of the March 28, 1979 accident at Three Mile Island, therefore, the risks of nuclear power remained a subject of controversy and the future of the enbattled industry uncertain. WASH-1400 attempted to establish a quantified risk acceptance criterion, but it appeared

4

. that as many questions were raised by the effort as were resolved. Ebr exanple, how should risks be weighed against benefits? If people are opposed to risk, how should very high runsequence but low probability anHaants be assessed? Ibw could risks and benefits be weighed when one segment of the population received the benefits while another segnent was subjected to the risks? How can the present generation count or discount risks that may be imposed on future generations? Answers to these questicns might well be impossible to quantify. The limitations of risk analysis were unde clear in the disclaimer in the conclusicn of I

WASH-1400: "It will take a:nsideration by a brnaMar wf=nt of society than that involved in this study to determine what level of rnv laar power plant risks should be acceptable."52 The accident at Three Mile Island, if viewed in the historical context outlined above, appears to provide a new era that could do much to revise a 25 year trend. During the period 1954-1979, the regulators seldcm had time to step back and objectively take stock in defining their safety goals and objectives. While they operated under the broad statutory obligation of regulating in the interest of public health and safety, there were many developments that prevented them fran demarcating nore definitive goals. The regulators initially were a part of an agency whose primary focus was developental; their workload increased with the beginning of the " Great Bandwagon Market" in the early 1960s; in the later years of the decade they were faced with the scaling up of reactor power levels without a corresponding research prwimu to study reactor safety features; in 1970 and thereafter they assumed an additional workload in the broader environnental area. Throughout the entire period, the regulators were pressured by industry, the JCAE and the Ccmnission to streamline the licensing process to bring nore reactors on line faster.

l These trends placed the regulators in a position of constantly attenpting to catch up with the changing technology. Beginning in the 1967-69 period, misgivings about the state of the technology were registered by scue n=hars of the regulatory staff and the ACRS. These were exemplified by such developments as lack of operating experience, lack of standardized plans, unresolved safety issues, particular concern over the E%rS, and lack of a vigorous safety research pr@tmu. But because the regulators really had little control over the pace of the techrology, they were forced to seek solutions to these problems even as they licensed nere and nore plants.

1 Public concern in the early 1970s, especially highlighted by interventions in licensing cases and by the controversial II:CS hearing, brought pressure frcan the JCAE on the AEC to cruantify the risks inherent in the technology.

In that way, nere definitive safety goals might result. WASH-1400 attspted to do that and although it remained con-troversial, it provided a methodological basis fran which to move toward such a goal.

1

. The creation of a separate regulatory agency in 1975 renoved some of the pressures from the regulators that burdened them under the AEC.

But many old problems remained, such as continuing work on operating reactors and unresolved safety issues as sell as work on new reactors still in various review stages of the licensing process. Only a limited ancunt of tire could be devoted to articulating overall safety goals.

Nonetheless, a new trend can be detected. The decline in reactor orders and increased enphasis on gaining experience with operating reactors slowly provided a context wherein nere effort might be exoended on the safety goal task.

Three Mile Island appears as a capstone to that trendi It has provided greater incentive and, with a slowdown in licensing operations, a greater eggommity for the agency to focus on delineating soecific safety goals. No longer is the licensing of new facilities the first order of business -(even though the pressure re: rains and the agency continues to issue licenses). Unlike the 1967-71 period when the rega-lators sv_;e governed by the pace of the technology, they are freer now to step back to review and capitalize on the experience of the past and apply it to the establishment of definitive goals.

l O

l l

t t

n u m ES 1.

George T. Mazur.an, " Regulations and Regulatory Organization, 1954-1956" (historical paper in preparation).

2.

Harold P. Green, " Safety Detenninations in Nuclear Pcuer Licensing:

A Critical View," Notre Dame Iayer 43 (June 1968): 635-36, reprinted in Joint Ccmrittee on Atcxric Energy, Hearincs on AEC Licensing Pro dure and Related Incislation, 92nd Cong., 1st Sess.,

1971, pp. 1502-1526; William H. Bcrran and Ice M. Hyde:ran, The Atanic Enerav Ccmnission and Regulatina Nuclear Facilities, printed in JCAE, Print, Incroving The AEC Regulatory Process, II, 87th Cong., 1st Sess., 1961, p. 494; George T. Mazuzan, " Nuclear Power Safety: The Enrico Fermi Case,1955-1956" (historical paper in preparation).

3.

George T. Mazuzan, "Taken to Court" (historical paper in preparation);

Berran and Hydeman, Regulating Nuclear Facilities, pp. 494-95; Leonard Bickwit to Ccmrissioners, " Adequate Protection of the Public Health and Safety," October 18, 1979, pp. 3-9.

4.

U.S. Atanic Energy Comnission, " Theoretical Possibilities and Consecuences of Major Accidents in Large Nuclear Power Plants" (h7&l-740), March 1957, pp.1-3; U.S. House of. Representatives, Report No. 435 (to accxx:pany H.R. 7383), 85th Cong.,1st Sess.,

1957, pp. 1-2.

5.

JCAE, "A Study of AT Procedures and Organization in the Licensing of Peactor Facilities," 85th Cong.,1st Sess.,1957, pp. 5, 28, 6.

AT Press Release, fty 22, 1959; JCAE, Incroving the AEC Reculatory Process, p. 579; 10 CFR 100.1.

7.

J. Samuel Walker " Nuclear Safety and The States, 1954-1959" (historical paper in preparation).

8.

Berran and Hydeman, Reculating Nuclear Facilities, pp. 4R9-496; Harold P. Green and Alan Rosenthal, Government of the Atan (New York: Atherton Press,1963), p. 76.

9.

George T. Mazuzan and Roger R. Trask, "An Outline History of Nuclear Regulation and Licensing, 1946-1978" (IGC Historical Office,1978), pp. 45-53.

10. JCAE, IncIoving the AEC Reculatorv Process, II, pp. 559, 579.

11. AT, Civilian Nuclear Power - A Recort to the President - 1962, reprinted in JCAE, Nuclear Power Econcznics - 1962 through 1967, 90th Cong., 2nd SesE,1968, p.134.

12.

Irvin C. Bupp and Jean-Claude Derian, Light Water: How the Nuclear Dream Dissolved (New York: Basic Books, 1978, p. 42; Pobert Perrv et al., Develoonent and Camercialization of the Light Water Bea,. tor, 1946-1976 (Santa ibnica: Rand Corporation, 1977), pp. 28-29.

13. AT, Civilian Nuclear Power - 1962, p. 154.

14. Metro, Bickwit to Cmmissioners, " Adequate Protection of the Health and Safety of the Public," Oct. 18, 1979, p. 7.

15. AEC, Civilian Nuclear Power - 1962, p.154, 16.

Ibid., p. 160.

17. JCAE, Hearings on Radiation Safety and Reculation, 87th Cong.,

1st Sess., 1961, p. 30, 13.

18. JCAE, Hearings on Limnsing and Reculation of Nuclear Reactors, 90th Cong.,1st Sess.,1967, Part I, p. 92; JCAE, Hearings on AEC Licensing Procedure and Related Legislation, 92nd Cong.,1st Sess.,

1971, pp. 97-98.

19.

Ibid., pp. 695-96.

20. JCAE, Hearings cm Li nsina and Regulation, 1967, p. 62.

21. Freedcrn of Information Act file relating to the Reexamination of NASH-740, Public Document Roam; Elizabeth S. Rolph, Nuclear Power and the Public Safety (Iexington, Mass.: D. C. Heath, 1979),

p. 49; John G. Fuller, We Alnest Icst Detroit (New York: RaMar's Digest Press, 1975), pp. 129-81.

22. JCAE, Hearings on Licensing and Regulation, 1967, pp. 62.

23.

Ibid., pp. 62-63.

24.

R. L. Ashley, ed., Nuclear Power Reactor Siting, Proceedings, Feb.16-18, 1965, American Nuclear Society, CONF 65-0201, AEC, Division of Technical Information, p. 6; Bolph, Nuclear Power, p. 79; JCAE, Hearinas on AEC Licensing Procedure and Related Legislation, 1971, pp. 480-82.

25. Roger R. Trask, "The Calvert Cliffs Decision,NEPA and Nuclear Power Power Plant Licensing, 1969-1972" (historical paper in preparation);

l JCAE, Hearings on AEC Licensing Procedure and Related Legislation, 1971, p.483.

26. Rolph, Nuclear Power, pp. 74-75; JCAE, Inproving the AEC Regulatory Process,1961; Hearing on AEC Regulatory Problens, 87th Cong.,

2nd Sess.,1962; JCAE, Hearinos on Licensing and Regulaticn,1967; JCAE, Hearings on AEC Licensing Procedure and Related Iagislation, 1971, p. 483.

27. Bupp and Derian, Light Water, p. 50-55; Rolph, Nuclear Power, pp. 74-75; David Okrent, "On the History of the Evolution of Light Water Reactor Safety in the United States" (unpublished manuscript),

pp. 6-102-147.

28.

Ibid.

29.

Ibid., 6-1-5; ACRS letter to AEC, Mar. 14, 1967.

30. ACRS Ietters, " Status of Generic Itsns Relating to Light-Water Reactors," Dec. 18,1972, Feb.14,1974, Mar.12,1975, Apr.16, 1976, Feb. 24,1977, Nov.15,1977; NURH3-0410, "NRC Propam for the Resolution of Generic Issues Related to Nuclear Power Plants,"

Jan. 1, 1978; Okrent, " Light Water Reactor Safety, p. 6-38.

31. Rolph, Nuclear Power, p. 86.

32. ACRS Ietter, " Reactor Pressure Vessels," Nov. 24, 1965.

33. Okrent, " Light Water Reactor Safety, pp.2-182-274, 6-48-57.

34.

W. K. Ergen et al., Diurgmcy Core Cooling, Report of Advisory Task Force on Power Reactor Coolim to the ABC, Oct.1967.

35.

It is signifimnt to point out that both the ACRS and the regulatory staff considered the IOCA as the Itost probable source of core meltdown during this period. Other accident sources were recognized, but it would not be until later that they would be etphasized as important contributors as the IOCA. Okrent, " Light Water Reactor Safety," p. 2-184.

36. WASH-ll46, Water Reactor Safety Program Plan, Feb.1970; Rolph, Nuclear Power, p. 92; Okrent, " Light Water Peactor Safety,"

pp. 6-52, 6-62.

37.

Rolph, Nuclear Power, pp. 92-94, 114, 142-43; Okrent, " Light Water Reactor Safety," pp. 6-96-97.

38. John O. Pastore to James R. Schlesinger, Oct. 7, 1971 w/ enclosure

" Status of Nuclear Power Plant Safety and Associated Safety Research."

39. WASH-1400 (NUREi-75/014), " Reactor Safety Study: An Assessment of Risks in the U.S. Ccumercial Nuclear Power Plants," Oct.1975.

40.

Ibid.

41.

Ibid.

42. Mazuzan and Trask, " Outline History," pp. 80-92.

43. NBC Historical Office, "The U.S. Ntx: lear Regulatory Ccenission:

l A Brief History" (historical paper in preparation).

44.

"NRC Statement cm Risk Assessment and the Reactor Safety Study Report (WASH-1400) in Light of the Risk Assessment Review Group Report," Jan. 18, 1979.

1 1

I

45. NRC Annual Reports,1976, pp.191-92; 1977, pp.147,180-81; 1978,

.p. 212.

46. JCAE, Hearirns on Investigation of Charges Relatina to Nuclear Reactor Safety, 94th Cong., 2nd Sess., 1976, vol. I, pp. 295, 306-307.

47. Victor Gilinsky to George Murphy, Apr. 28, 1976, in ibid., Vol. II, pp. 1743-44.

48.

Ibid., pp. 1744-45.

49. Marcus A. Rowden to Clyde A. Lilly, Jr., Mar. 31, 1977.

50.

"NIC Statement on Risk Assessment...," Jan. 18, 1979; NRC Annual Reoort, 1978, p. 213.

51.

"NRC Stat m ent on Risk Assessment...," Jan. 18, 1979.

52. WASH-1400, p. 202.

1 i

l l

1 I

1 l

,