ML20207Q109
| ML20207Q109 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 01/23/1987 |
| From: | Perlis B NRC OFFICE OF THE GENERAL COUNSEL (OGC) |
| To: | Matthews D, Noonan V NRC OFFICE OF INSPECTION & ENFORCEMENT (IE), Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20207Q102 | List: |
| References | |
| FOIA-87-10, RTR-NUREG-0396, RTR-NUREG-396 NUDOCS 8701230029 | |
| Download: ML20207Q109 (4) | |
Text
--
l I
MEMO FOR:
Dave Matthews, I&E Vince Noonan, NRR FROM:
Bob Perlis, OGC
SUBJECT:
SEABROOK EPZ EXEMPTION REQUEST I
On December 18th, PSNH submitted a petition for a waiver (pursuant to e
10 CFR 52.758) of the requirement for a 10-mile EPZ at Seabrook.
The petition is based on a number of probabilistic studies PSNH has had prepared, and it requests a one mile EPZ.
The Staff review of PSNH's request will have to follow two tracks.
First, the Staff will have to-assess the technical wotk embodied in PSNH's studies to determine whether the risk assessments propounded by PSNH are reasonably accurate.
Second, under 52.758, we must ascertain whether application of the 10 mile EPZ requirement under the circumstances presented at Seabrook would serve the purpose (s) for which the requirement was adopted.
This necessitates a staff determination on, among other things, why 10 miles was chosen back in NUREG-0396.
PSNH has identified the following four rationales it believes were behind the choice of a 10-mile EPZ:
(1) for design basis accidents, thyroid and whole body doses fall off sharply at 10 miles such that lower level PAG's would be exceeded at 10 miles in only 30% of 129 nuclear units analyzed and higher level PAG's would not be exceeded at any unit at 10 miles; (2) for less severe Class 9 accidents, WASH-1400 showed that doses would not exceed PAGs at 10 miles; (3) for more serious core-melt accidents, protective actions out to 10 miles would result in significant savings of early injuries and deaths; and (4) planning for 10 miles would provide
- a basis for any necessary actions to be taken beyond 10 miles.
I have drawn up the following (non-exhaustive) list of questions that we are likely going to need to address in order to respond to the waiver request.
As you can see, they raise a number of policy and/or regulatory interpretation questions; a number of different Staff offices may want to have some input into the responses (as may FEMA).
I think we need to start considering at least some of these questions very soon.
1.
The most basic question: is PSNH correct in its interpretation of why 10 miles was chosen in NUREG-03967 Are there four rationales correct, and are they exhaustive (or are there other rationales that need be considered)?
In this restard, it might be helpful to go back to the individuals who participated in writing 0396 to get their views.
2.
Do we want to focus on the historic origins of the 10-mile rule, or should the staff focus instead on the policy considerations of the rule today without delving into history?
It may not be possible to identify a coherent rationale for 10 miles from NUREG-0396, and to the extent that an interpretation today remains consistent with 0396, we may wish to consider whether we want to concentrate on current policy considerations.
If we decide against the historical approach, we need to articulate the policy reasons for setting the EPZ (in the normal case) at 10 miles.
A p o3 jj 9 7- 0/6 8701230029 870116 g,
n I
. policy interpretation will have to include (for this case) either an adoption or rejection of PSNH's four rationales.
3.
What role does risk play in the process of determining EPZ size?
Insofar as risk is considered in the derivation (and perpetuation) of the 10 mile EPZ, does one focus on absolute risk or risk conditional on core melt?
In other words, can a plant take credit 'for accident prevention 'in the emergency planning sphere? Also, can a plant take credit for design features, such as containment, which mitigate the consequences of a core melt, cr must we look at emergency planning requirements only after there has been an unplanned release?
I haven't had an opportunity to read PSNH's full submittal; if they rely on risk conditional on core melt (which I believe they do), the only question we may have to answer here is whether risk conditional on core melt is an appropriate consideration.
If NUREG-0396 used a spectrum of accidents, must a waiver petition 4.
(and any challenges thereto) rely on the same spectrum?
If not, what accidents should be looked at in establishing EPZ size?
5.
Did NUREG-0396 consider in its risk analysis a spectrum of plant designs, or only one particular plant? How plant specific was the original determination of 10 miles, and how plant specific should the rule be?
6.
What risks are we protecting against?
Do we look at all the PAG's, or just early fatality doses? Whole body, thyroid, or both?
7.
What role does the ability to take ad hoc measures beyond the EPZ play, and what are the parameters within which such measures need be effective? In evaluating this aspect of PSNH's request, do we look at the ability to take ad hoc measures only out to 3 miles, or 11+ miles?
How are such ad hoc measures affected by time components of accident development?
In other words, if PSNH could show that its containment would slow down releases, could this make up for a weakness in the EPZ planning base because of the small base? This may be narrowed when we see the actual terms of the submittal.
8.
What role do dose savings play in a proposed shrinking of the zone?
Assuming that expanding the EPZ at Seabrook beyond 1 or 2 miles would result in some dose savings in the event of certain accidents, how does one make a determination that various dose savings are or are not significant?
9.
How heavily can we rely on PRA's and risk assessments in determining the size of the EPZ?
Hasn't the Commission taken a position against relying on PRA's for IIcensing purposes?
How does the safety goal fit into a licensing decision like this one?
10.
If we can rely on PRA's in establishing an EPZ size, what level of certainty do we need to have?
Ilow much confidence do we need to have in the specific numbers generated; how much uncertainty can be tolerated?
l
a i
~
l 11.
What individuals, branches, offices, and other agencies do we need to get involved in answering these questions? This question needs to be addressed as soon as possible; undoubtedly the answer will differ for the different questions.
--e d
W
J...a Dr s, L ihn.t ti on Feti ti on -
Work Needed to Respone,o
- Ann iba r.odu o PF case
(,p p J e_g u e s t ' f o r 1 mile should
.be oranted J) IJa n.1 to acree or.fisagree with App list of underlying purposes (4 r4 t i os.cl e).
This i t.co) ves review of f orti a l background (Rul e, State of Consids, Statements by lir:C i t. Congre'ss Heing, Licensing Cases);
0396 and its surrounding bned wround s whel 0396 people thought (ii we don't call them other parties ut11); h e.u we br/u used the rule: how we have used 0396; where we want to come out.;o. t hu lu n.ile quet, L i or e in guieeral ; where we want to come out on 10 mile question in Seabrool:.
- 2) fie s2 e w rd Apps aruuments I
rf)f, D (!! 0;
't ! C TEC :iJICAL (CALCULATIONAL) h AT l ON,' LLi.,
e) Dores ileei
.,ppe oc t h in iact ]eed to FULL REVIEW the ult 2malc conclusions t h e y-say it OF THE REQUEST AND 1 o o<!w taf UNDERLYING DOCUMENTS b) Do wo sqree that the trethod they used LIUE INDIAN POINT is the r i c,b l... A r.ocil appr opi i a lti corr ect ?
(Not full review but assessment of whether the r+thodol oov i s %1id) c) 00
'M v o n-*.t Rt the input appears to bo e r..i..;. l e.e w.d - n-a l i,h t stuff, t,n s e,d_
s pn q epr.yf ul t mo;no:
r.r o thr, using i I..:
..: i r 1
.. i i-2s 2t suii1clont f or what t h e-v nro v :s i ng at for/
filo t soll r e-- v i 4 u of se cur ecy o+
the date but & % c t s s y....en t of whether they at e using the tiqr.t n-ta h n tes said at l oo k e, complete enouob?
d) Do t he t or.c l ut.1 or e s. look reanonable besmi on t.ho n.e thodol ogy and input?
bo i hc.v i ooi.
r.. %.n i t,1 c hesed on our o poriu.co and m:pertise? Do the c a.i t ul e t i or e l c onc l u s i or n, lead to the conclusions about the waiver that App re, thm ejn? h t do we think the calculalicnal cosiclusions say about waiving rule?
e) Are the conclusions Seabrook or generic?
- 3) For rationale 4 Again question is:
Dow it look reasonoble and correct to us?
Suro loci.n like we will need fella input
- 4) li we agr ee that a waiver of 10 miles in warrented, what standard should be unett for judgement in place of the ten t..i l es ? The.pplicant is asking not only gy. Q-.d/O for a wsi ver of 10 mi1es but acceptance of a standard based on risk equivalent Sg, to.the 03Te> curve.
IS that the standard the Staff recommrends that the Commi ssion adopt?
I Thi s pr oliel.1 y has the most Policy involvement]
- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ - -.