ML20055B465

Supplemental Comments on Des.Statement Should Be Revised Due to Issuance of ORNL Accident Sequence Precursor Study, NUREG/CR-2497.New Info Cannot Be Adequately Treated by Altering Fes

ML20055B465
Person / Time
Site:	Seabrook
Issue date:	07/20/1982
From:	Curran D, Jordan W HARMON & WEISS, NEW ENGLAND COALITION ON NUCLEAR POLLUTION
To:
References
NUDOCS 8207220393
Download: ML20055B465 (5)

Text

<

' "' 000 METED USHRC jd JW.20 PS:09 0FFTP CF SECRETAkV CC:;Eit !G & SERACF BRANCH SUPPLEMENTAL COMMENTS OF THE NEW ENGLAND COALITION ON NUCLEAR POLLUTION ON THE DRAFT ENVIRONMENTAL STATEMENT RELATED TO THE OPERATION OF SEABROOK STATION, UNITS 1 AND 2 DOCKET NOS. 50-443 AND 50-444 PUBLIC SERVICE COMPANY OF NEW HAMPSHIRE, ET AL.

July 20, 1982 c,oo2 8207220393 820720 PDR ADOCK 05000443 D PDR

NECNP filed comments on the Draft Environmental Statement for the Sedrook nuclear f acility on July 6, 1982.

Since then, the Oak Ridge National Laboratory has published a study which significantly affects predictions of the probability of severe accidents at nuclear plants. Precursors to Potential Severe Core Damage Accidents : 1969-1979, a Status Report. NUR EG /CR-249 7 (June, 1982). The Accident Sequence Precursor study (ASP) assesses the probability of a core melt accident at between 1.7 x 10-3 and'4.5 x 10~ per reactor year, which is two orders of magnitude greater than the predictions of the Reactor Safety Study (RSS), upon which the Seabrook DES is based. For example, the ASP predicts 1 accident in 200 to 600 reactor years, in comparison with the RSS, which predicts 1 in 20,000. Unlike the theory-based RSS, the ASP study is based on actual events documented in Licensee Event Reports ( LER s ) between 1969 and 1979.

Of these LERs, 169 were found to be precursors to severe core melt accidents. Id. at xii . Approximately 38% of the significant precursors involved human error. Id.

The probability of serious reactor accidents could easily be even greater than that predicted by the Acciden t Sequence Precursor study. It is possible that many more core melt precursors than identified actually occurred, since the study was restricted to consideration of LER s and did not include other event indicators such as PNO 's

2 (Preliminary Notification of Occurence ) or PNSs (Prelimin ar y Notificiation of Safeguards Events ) In addition, the study _

did not assess the probability of accidents other than core melts, such as core degradation, which could also result in large radiation releases.

The Seabrook DES 's assessment of accident risk is founded on a "rebaselined " version of the Reactor Safety Study. DES at E-1. We have already commented that this discussion is inadequate in that it identifies only 4 accident sequences without explaining or justifying that choice. NECNP Comments of 7/6/82 at 4. NECNP submits that the DES discussion of radiological risk should be redrafted to include the following :

1. A discussion of severe care melt probabilty as O

explained in the ORNL study , and as specifically applied to Seabrook.

2. A discussion of the probability of accidents including other than severe core melt as may be extrapolated from the ASP and other studies.

3. Identification of accident sequences critical to the Seabrook facility, and not just to nuclear plants in general. The identification of critical scenarios must t ake into account individual plant design and site characteristics.

For example, critical accident scenarios for -Indian Point have been identified as fire and seismic events, and for l

Three Mile Island as inter f acing system LOCAs.

l

! 4. An evaluation of the contribution of human error

to accident risk, in light of the f act that 38% of the core l melt precursors identified in the ASP were caused by

! human error.

I e

3 Although the ASP study is subject to statistical uncertainty, it has nevertheless drastically changed the bounds of uncertainty in determining the risk of accidents at nuclear power plants. Because of its empirical rather than theoretical foundation, it appears to present a much more reliable diagnosis of accident probability than the RSS. The National Environmental Policy Act requires an agency to either disclose the complete range cf uncertainty or to consider only the uppermost bounds of risk in assessing the costs of a project. NRDC v. NRC, No. 74-1586, D .C .

Circuit, April 27, 1982, slip op. at 51 . Consistent with that obligation, the NRC must include the ASP assessment of core melt probability and its implications in the Seabrook DES.

Because of the extreme significance of this new information in assessing the cost of the Seabrook reactor, it cannot be adequately treated by altering the Final Environmental Statement.

The Draft Environmental Jtatement should be revised and recirculated so that the public may have su f ficient opportunity to participate in the evaluation of the costs and benefits n

of operating the Seabrook reactor.

J

l 4

Respectfully submitted, l

l

!<b M William S dan, Ig '

&& 0w-Diane Curran HARMON & WEISS 1725 I Street, N .W .

Suite 506 Washington, D .C . 20006 (202) 833-9070

)

l

!