ML20149F826
| ML20149F826 | |
| Person / Time | |
|---|---|
| Issue date: | 05/29/1987 |
| From: | Hulman L NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Beckjord E, Ross D, Speis T NRC |
| Shared Package | |
| ML20149B718 | List: |
| References | |
| FOIA-87-743 NUDOCS 8802170379 | |
| Download: ML20149F826 (44) | |
Text
{{#Wiki_filter:__ [ ...*'**c%), UNITED STATES
'@. NUCLEAR REGULATORY COMMISSION
$ ,I WASHING TON. D. C. 20555
,f W 2 9 $87 MEMORANDUM FOR: Attached List FROM: Lewis G. Hulman, Chief Severe Accident Issues Branch Division of Reactor. Accident Analysis Office of Nucidar Regulatory Research
SUBJECT:
MINUTES OF A MEETING WITH NRC STAFF AND REPRESENTATIVES OF THE ATOMIC INDUSTRIAL FORUM (AIF) ON EMERGENCY PLANNING On May 14,1987 at 1:30 PM a meeting was held in Bethesda Maryland to discuss the status and schedules of industry and NRC efforts on emergency planning. This meeting was requested by AIF. The attendees of the meeting were: Mike Jamgochian, NRC/RES ' Duncan Brewer, Duke Power Tim Tipton, AIF Ed Podolak, NRC/NRR
,Reggie Rodgers, Northeast Utilities Charlie Wike, PA PWR. & Light Jerry Hulman, NRC/RES Len Soffer, NRC/RES Sunil Weerakkody, NUS Corporate Mitzie Solberg, NRC/NRR l The initial discussion focused on the AIF organizational structure. There 1
exists a full comittee on Licensing with two subcomittees. One subcomittee, l headed by R. Rogers, deals with emergency planning issues, and is currently involved with recomendations on revising NUREG-0654, evaluations relating to 8802170379 880204 PDR FOIA SHOLLYO7-743 PDR ' f [' u / _ >
2-tr,s size of the EPZ's and lookir.g at symptom based Emergency Action Levels (LAL's). Mr. Tipton also discussed the 10CDR L EPP.! initiatives presented to NRC in March, 1986 relating to EP2 sizes. The efforts of a sword subcor.r.littee headed by D. Brewer were discussed. These efforts focused on the source term issues as they relate to emergency plannir.g, the validity of a graded response approach, providing FEMA with comments en Guicance Memorandums, and lessons learned from Chernobyl for physicians. In response to an AIF comment that information could not be obtained from Sandia National Laboratory personnel, Jerry Hulman (NRC) stated that he would try to m6ke available for public use the MACCS code which was developed for the 7"e"rtert by Sandia. He also invited AIF input on how to better portray uncertainties involved in NUREG-1150. AIF also nentioned: (1) the National Environmental Studies Project on dose modeling, which is being conducted by NUS, looking at nine locations (three inland, three coastal and three river valley) with various meterological dispersions. (2) the NESP study, just getting started, looking at actual evacuations that have taken place in the U.S. AIF voiced a definite interest in the following areas: 4 h
j ts 3-(1). Codification of a graded response approach in the emercenc3
., ' plar.r.ir.g rcgulations.
(2) The concern that state & local governments and FEP.A have a mindset that evacuation is the orn- protective action, and that e it should be made clear that the EPZ's mean En,ergency Planning Zone rather than Evacuation Planning Zones (3) Clarification of the need to develop symptom based EAL's, event based EAL's, a fission product barrier breach approach, or a combination of the above. (Ed Podolak, (NRR), indicated that a hybrid approach is encouraged, such that a symptom based EAL's (fission product barrier) are backstopped by event based EAL's , not aadressed by a pure symptom based approach. (4) A relaxation of the 15 minute prompt public notification requirement was also discussed by Alf. (5) A desire to have the EPZ sizes more comensurate with any revised insights on reactor risks coming out of new source term and risk research. (6) AIF strongly voiced their concern that a graded response rulemakirig should move forward in addition to the current ongoing rulemaking effort relating to lack of state and local government emergency plans.
-q
- i'r. Pulman then discussed the ongeing i;RC emergency planning activities it 4rb
. included;
- 1. Research assistance in emergency planning licensing retivities.
- 2. Evaluatien cf public corants (approximately 5000) relative to a rulemaking on lack of state and local government emergency plans.
- 3. A reassessment effort on emerg planning, as noted in SECY 86-76, presently being carried cut via a Research contract with 6l;L on Protective Action Decisionmaking. It was also pointed out that NL' REC 0654 would likely have to be revised if the Comission endorsed a graded response concept.
- 4. Licensing activities related to emergency planning at specific reactor sites.
\
bh . - o L'ewisG.kl n Chitf Severe Acci ent. Issues Branch Division.of Reactor Accident Analysis Office of Nuclear Regulatory Research b
- , , . . , ~ - - .
s. g, . . .
#.n ~ N,.
-af y.
J i.
, kdressees:
- E. Beckjord '
D. Ross- l T..Speis. l
~ F. Miraglia
. R. Starostecki R.' l'ouston l F. Congel D. Matthews L. Hulman.
- R M Jaegachian
/.. Soff;r, F. Hebdon l K. Perkins S. McGuire Outside and inside cecting attendees not listed.
Receptionist Desk, Phillips Bldg. POR-Cer. tral files SAI Reading File l l f 1 i'ii-- - -
- n. -
~ h0% TEPFC.T H&
W {} @ F E- G & ibe. N M Y W hp% Cwl\ HOW SHOULO THE IMPROVED XNOWLEDGE AN3 UNDERSTANDING 0F 500RCE TERHS BE USED IN EMERGENCY PLANNING? George D. Sauter, EPRl/NSAC, Palo Alto, CA. The answer to the question which is the subject of this panel discussion was much more apparent before the'Chernobyl accident than it is now. While it is too soon to tell what the impact of that event will be on nuclear power in the U.S. and elsewhere, it is clear that source terms, offsite emergency response planning (0ERP), and containment performance will receive increased attention, both from the regulators and the public. In fact, the impact of Chernobyl on the first two of these three issues is the subject of much discussion at this meeting, it has long been obvious that 0ERP is an issue of high public visibility and interest, and that decisions regarding revisions of OERP . requirements will involve political considerations and public sensitivities, perhaps to a larger degree than will decisions regarding other nuclear power issues. Chernobyl can only exacerbate this situation. Unfortunately, decisions made in the emotional heat of the moment very often turn out to be erronecus in the light of more dispassionate deliberation. Further, we have seen that faulty decisions about OERP requirements, once made, are difficult to reverse. To maximize the likelihood that sound decisions are made, it will be necessary to minimize the emotional influence of Chernobyl on them. This means that the NRC and other authorities should not make, nor should other government officials pressure them to make, hasty decisions before the lessons of Chernobyl can be learned and their relevance to LWRs evaluated. I would therefore recomend that the industry and the NRC cooperatively pursue the following course of action regarding source terms and offsite emergency response planning requirements.
- 1. With all deliberate speed, learn the lessons of Chernobyl; i.e.,
develop the best possible understandirig of what happened and why, the similarities and differences in design features between the Russian plant and U.S. LWRs, and the significece of these findings to risks from potential accidents at U.S. LWRs.
- 2. Keep the public fully informed, in readily understandable terms, as the lessons of Chernobyl are learned and their significance to U.S.
LWRs are evaluated.
- 3. While the lessons of Chernobyl are being learned, continue to solidify our understanding of the source terms from potential accidents at LWRs, of how the risk to public safety is affected by these source terms, and of the most effective emergency response strategies for mitigating these risks. Temper this understanding with the insights from Chernobyl as they become accepted.
g
t
?
4 When it is timely, revise emergency planning requirements to more realistically reflect our improved understanding of the most f probable source terms and corresponding threats to public health and safety. Of course, when it is "timely" to consider revising present OERP requirements depends on a balancirig' of several f actors; e.g., the degree of confidence in the adequacy of the relevant knowledge base, the need for a prompt decision, and the desirability of minimizing the influence of the psychological spillover from Chernobyl. To further minimize this spillover, it will be prudent for the industry not to seek, for the time being, a reduction in the present 10-mile radius of the Plume Exposure Emergency Planning Zone (EPZ) to 2 or 3 miles, even though the technical basis for doing so may appear to be sound. Given the reality of Chernobyl, a better industry strategy would be to seek acceptance by the NRC of a Graded Response (GR) approach. In
. this approach, the 10-mile EPZ radius is retained and evacuation as necessary out to 2 or 3 miles, coupled with sheltering as appropriate out to the 10-mile EPZ boundary, is the primary planning basis. There are several reasons why seeking GR acceptance would be sounder strategy.
- 1. Changing the EPZ size is more likely to be interpreted by the public as reducing the level of public protection and hence more strongly opposed. -
- 2. The 10-mile EPZ size is explicitly called for in pertinent sections ,
of 10CFR50; thus changing it will require rulemaking with its increased visibility and numerous opportunities for delay in reaching a decision. On the other hand, there is nothing in 10CRF50 which rules out the GR approach. Thus, it could be adopted by administrative action; no rulemaking is required.
- 3. Because GR can be implemented more quickly than a reduced EPZ size, it should afford more immediate relief to plants such as Shoreham and Seabrook.
- 4. Adopting the GR planning basis does not preclude later reduction of the EPZ size. In fact, GR is a natural first step toward reduced EPZ size; demonstration of the viability of the GR planning basis will make later reduction of the EPZ size easier to accomplish.
l l S. Calculations by both the NRC staff and by an EPRI contractor have shown that for the most likely scenarios for offsite releases of radioactivity, GR will give a greater degree of public protection against prompt health effects than will evacuation to the 10-mile EPZ boundary. This is because the evacuation of fewer people allows those nearest the plant (and thus most at risk) who are evacuated to i be moved more rapidly. l l One. caveat to this recomended strategy must be acknowledged. Some in l the NRC staff have stated the opinion that rulemaking will be required / to shift to the GR planning basis even though nothing in the present (, l L
s b* . regulations rules it out, because the precedent of the present planning basis is so well. established. If this opinion prevails. the industry can argue that, because the GR approach provides as much or more protection than the present planning basis, the NRC should make the shift to GR immediately effective, with any rulemaking hearing to follow, If this argument fails, it is unlikely that either -the. industry or the NRC will want two serial rulemakings for revising OERP requirements. In this case, the industry may have to choose between a rulemaking on Graded Response with its more rapid but less extensive-revision of emergency. planning requirements, or bypassing GR and seeking by rule a reduction in EPZ size which, particularly in the light of Chernobyl, the industry is less likely to be successful in obtaining. 4005456 e 4006NS6
o xo W 4 I./ TECHNICAL BASES FOR GRADED RESPONSE STRATEGY - USING NEW SEVERE ACCIDENT SOURCE TERMS T. Margulies and S. McGuire U.S. Nuclear Regulatory Commission Washington, D.C. 20555 W. T. Pratt and A. Tingle Brookhaven National Laboratory Upton, New York 11973 ABSTRACT A planning basis has been developed for offsite opergency response during the initial phase of a nuclear power plant accident which uses new technical infor-mation and insights contained in NUREG-1150 and related documents concerning severe accident phenomena, uncertainties, and potential offsite consequences and risks to the public. This state-of-the-art information is used along with realistic assumptions regarding protective actions to examine a planning zone concept and protective action strategy that takes into account the same overall objectives as those used by the NRC/ EPA Task Force in NUREG-0396. The technical analysis supports three zones as a basis for protective actions that would result in substantial dose savings in the environs of nuclear facilities in the event of a serious power reactor accident. The criteria used to define g Y these zones and various calculated results that support this planning basis are discussed. Currently, the USNRC's Emergency Response Center guidance concentrates on early phase, initial protective action in the first zone (2-3 miles distance from the reactor) within a ten-mile planning zone. l
b s' ic INTRODUCTION This paper presents a technical basis for a time-phased, or graded response strategy for protecting the public in the event of accidental Previousradislogical supporting releases analyses 3 from large commercial light water reactors.for this approach were based o independent of plant design since they were developed for siting analysisThe new technical purposes) and did not explicitly address uncertainties.2information base ment," NUREG-1150 (February 1987) concerning risks to the public, uncertainties in predicting severe accident phenomena and source terms, and potential con-sequences are used to examine the efficacy of different protective actions in the vicinity of the power plant. This paper is arranged as follows. First, some background information on the Next, off-guidance used to develop the planning basis in NUREG-0396 is given. Finally, a discussio
- site radiological dose predictions are presented.
what has been learned from the calculated results that is important to emergency planning and response and a protective , action strategy is proposed. b. t 9 1 NUREG TECH BASES GRADED RESP t__ _ _ _ _ _ . _ . _ . _ __ . _ _ . _ _ . _ _ _ _ _ _ _ _ _ _
4 PR]NCIPAL OBJECTIVES The principal cbjectives of accident response planning and decision making are discussed in NUREG-0396.3 In particular, the joint NRC/ EPA Task Fprce decided that they are to provide: (1) dose savings for a spectrum of accidents (not the source terrs from a single accident sequence, but the time frames for radiologica4 release characteristics from a range of accident severities); and (2) substantial reduction in early severe health effects (injury or death) in the event of a worst case core melt accident. The criteria for preparation and evaluation of radiological emergency response plans are given in NUREG-0654 (NRC/ FEMA 1980).* Iri brief, emergency planning zones (EPZ's) are to be established around nuclear plants corresponding to
- short term plume exposure and longer term incestion exposure pathways. The generic emergency planning zone for airborne exposure has a radius of approxi-mately ten miles (16 km) while that for ingestiop exposure from contaminated food and water has a radius of about fifty miles (80 km).2 The size of the plume exposure pathway was based on four considerations as outlined in NUREG-0396:
(1) Projected doses from baseline design accidents would not exceed Protective Action Guideline (PAG) levels outside the zone. (2) Projected doses frem most core melt accidents would not exceed PAG 1evels outside the zone.
+
b (3) For the worst core melt sequences, imme#iate life threatening doses would t generally not occur outside the zone. (4) Detailed planning within 10 miles would provide a substantial base for expansion of response efforts if this proved necessary. 2A categorization of three time-frames, or time phases of an accident response has been proposed as a means to usefully distinguish between different-pathways of exposure to the public, protective actions and decision making ("Protection of the Public in the event of Major Radiations: Principles for Planning," ICRP 40, Vol. 14, No. 2, 1984). The "early phase" comprises the time-frame starting from the threat of a serious release to several hours after the commencement of that release. The "intermediate phase" covers that period of time from the first few hours to a few days after the onset of the accident. Thar "recovery" phase covers the period of time where decisions are made corycerning the return to normal living conditions. This last_ phase may extend for a prolonged period and involve decision making involving social, economic and technical inputs. In comparison, the "plume exposure" pathway considera-tions of NUREG-0396 pertain primarily to the "early phase" of an accident response whereas the "ingestion exposure" pathway category overlaps the inter-mediate and recovery phases. 04/20/87 2 , NUREG TECH BASES GRADED RESP
i' . IMERGENCY PLANNING ZONE CRITERIA Three generic protective action zones have been developed taking into account the state-of-the-art information on potential consequences, timing and release characteristics of a spectrum of accidents for representative readtor/ containment types. The three zone sizes are primarily based on the following criteria and are discussed more fully below in terms of a protecttwe action - l strategy. See Tables 1 and 2. l l Table 1 Criteria for zone size Name Whole Body Oose Criteria Early Evacuation Zone Probability > 200 Rem ~ 0 Emergency Planning Zone Probability > 50 Rem ~ 0 j . Extended Emergency Planning Zone Probability > 5 Rem - 0 Table 2 Early health effects from exposures from released radioactive material Health Effects Whole Body Dose Early Injuries 2 50-100 Rem Early Fatalities ,s 2 200-600 Rem At 50 rems within a week, deaths, vomiting, or diarrhea would not generally b occur. Most people would experience no symptoms at a 50-rem dose, but a few people exposed to that dose could experience the milder symptoms of acute radiation syndrome, specifically loss of appetite or nausea. 2A 200-rem dose is a dose at which the major symptoms of acute radiation syndrome, specifically vomiting and diarrhea, are fairly likely. Deaths due to acute radiation syndrome have not been observed in humans exposed to doses of 200 rems or less. 8 Preliminary information on the impacts of the accident at Chernobyl indicates that fifty-three patients with estimated doses of 200-400 rads resu11ed in one fatality. All these expcsures occurred onsite, and none of the patients had thermal skin burns. I 04/20/87 3 . NUREG TECH BASES GRADE 0 RESP
. , . , N ,,
- t Early Evacuation Zone (EEZ)
Currently, NRC regulations (10 CFR Part 50, Appendix E, Section III.D) allow the utility 15 minutes to ,otify9 State and local officials after d,eclaring an emergency. The officials must then consider protective actions. fiostoften the decision rests with the State governor. In general, it can b4 expected that they would want information on plant status and dose projections based on calculations involving real-time meteorological data and measuremehts of radio-activity in containment. This decision making has a potential to be time con-suming and slowed or even paralyzed by uncertainties about the source term and containment perforchnce since many severe accidental releases are not along monitored pathways. The new approach would require utilities to immediately and automatically warn the public within.2 or 3 miles to evacuate based upon specific plant indicators (i.e., Emergency Action Levels). This simplifies the decision making and
. speeds the warning to the public.2 The NRC cannot predict whether the public after being warned will participate in an evacuation, but can regulate'the con-tent and time-of-notification of a message regarding public protective action.2 Thus the people who could suffer the most serious effects and who must att quickest to effectively protect themselves would be warned earlier than under the existing regulations.
Emergency Planning 7ene (EPZ) Beyond the EEZ out to approximately 10 miles, the recommendation to the public would be to prepare to shelter and to evacuate if necessary. The preparation done at this stage would speed a subsequent evacuation if it is necessary. This allows these additional people sufficient time to evacuate even if they do not start until after a release has started. (Good sheltering in large b buildings may be preferable to evacuation under certain circumstances such as 5 during severe storm or heavy snow situations.) Because of the limited size of this second phase evacuation, downwind only if the wind persists while a tehtse occurs, it could generally be accomplished ' without support from the local or State governments although that support would be desirable. E' forts would also be made to immediately locate highly contami-nated areas and promptly relocate people from "hot spots" (i.e. , within , approximately 4-6 hours) if a release occurs. , Extended Emeroency Plannino Zon_e (EEP2) As a means to further minimin public radiation exposure in the early response phast_,ty a nuclear accident, cheltering would be recommended to reduce T,y ~ , 2 At present the HRC's Emergency Response Center guidance consists of recommend-ing for the initial response an early, precautionary evacuation in the area within 2 to 3 miles from the plant (NUREG-1210).S .- alt was reported that Soviet protective actions during the Chernobyl e'mergency
, involving evacuation of 90,000 inhabitants was initially difficult since many peasants refused to abandon their animals, so an evacuation of animals was ordered to convince,these peasants to leave.7 04/20/87 4 .
NUREG TECH BASES GRADED RESP
s .( L. :. .i. >% .r jy. u,> ...-C:l,
'T s . I' i . . (c,
- -- 7 L w . 77 .s external exposure and inhalation from the plum E Aaynoc respTratory protection .
for the public outside evacuated areas would aIso be beneficial. 2 Although the_ ingestion pathways can be of concern at considerable , distances from the release point, this is not a major contribution to early pealth effects considered in the plume exposure pathway. 'For the ingestion pathway, early protective actions are designed to minimize subsequent contemination of milk or other foods (e.g., remove cows from pasture and put them o's stored foods). The Food and Drug Administration specifies criteria and actions to be taken regarding vegetables. , i b 1 .
- 04/20/87 5 NUREG TECH 8ASES GRADED RESP
. q; / y 4, s, e N'
N- '[,, s' s' 'a ;
'~ ,
s N
- w .
j
'x ;
.~ s l
DOSE cal.CULAH0NS N _ v - s x x I Dose calculations were made with.th'r MELCOR Accident k nsecuence Code System (MACCS)8 and source terms developed for the major new risk Study published in draf t f orr:o as NUREG-1150 "Reactor Risk Reference Document,'"~ (Febetfary 1987).
' 4-s These so'urce , terms characterize, f oi example ~, thejiming, duratiod.and magni-tude of rOease for 'a set of renesentative plants. 'These , characteristics of the release .substantially infloance the potsntial offsite r'adiation doses and early fetalities and injuries sind thus the needed offsits emeigency measures.
First, w= <ill discuss a design basis accioent Itakage'sceni-io (involving no containment Nilure): , Then'we present sorre relevant results' for the extremely severe at:im nts. The status of,the sourcf term research and specific assump-tions used in Oc raciological oose crict0atyns are present3d in Appendix A. s % MACCS is a new co.1e dueloped at Sandje National Laboratory and is intended to
- be used insttu of CPM 28 for~the PJrpose of performing consequence calcula-tient. . The omncipal difference.C hom,CRAC2 are that W,CCS uses a multi puff atmospheric dinpersion model and a more.recent,!y,develcped draft set of health eff:tet modc1s. The code calcubtes dose from three pathways: external whcle body dose from the plume as it pe3ses over ("clou1 shine"), external whole body dose from radiurraclides deposited on the ground ("ground shine"), and internal dose to various organs from inhaled radionuclides.
DesiaL Bash Containment Leakaoe Scenaric f or calculation 61 purposes ("r.tttica blackout" scenario at the Surry plant was chosen with the' intent of providing high" consequences for leakage, t'1 Thq calculated dese results assune a "worst" (no rain case) weather condition b of category F ttability and a 1 m/sec windspeed. A single uniform puff release of two hour duration was assumed The release occurs at ground level, with no plume buoyancy, And with no buildir.g wake effect. A dry deposition velocity of 0.2 cm/see was assumed for all elements except the noble gases. Shielding ' factors for' individuals out of doors and a breathing rate which is three times the normal bresthing rate were assumed. The plots-(Figures 1 and 2) show total committed dose from all pathways as a function of both distance and time after the beginning of the release. Because of the relatively large release of noble gases and the low deposition velor.ity, For example, the cloudshins' ar.1 inhalation doses dominate bone marrow doses. the cumulatin bone marrow doses rit 5.5 km,1 five hours after the release began are: 2.3E-3 Sv2 cloudshine, 1.6E-5 Sv groundshine, and 1.9E-3 Sv committed inhalation dose. Thus, af ter passage of the plum 2, there is very little further exposure from groundshine. Thyroid doses are, of course, dominated by inhalation. Oosts for other leakage rates may be obtained by scaling the y-axis. . < These resuMs can be compared to Protective Action Guidelines PAGs) picaulgated by the Environmental Protection Agency (EPA) which represent trigger leuls (rather than dose limits) for initiating a radiological 11 KM = 0.62 mi 83 SV = 100 REM - 04/20/87 6 NUREG TECH BASES GRADED RESP
1 l l
\
, =
~
10 _ , , y i ,
~
~
I 0.1%/ day leak rate -
- s 1 -
=
- Time from beginning of the release (hours)
. E.
j 0.1 = .
~
24 0.5 ": 0.01 =- 5 g 1.5 2' 3 , ~ f ~ -
- 1'
- 2.5 3 r , ~
l ~ b
. I I i 1 8 O.001' 12 4 6 8 10 6 0 2-Distance (km)
Figure 1 Total bone marrow dose . . Notes: (1) Calculations made using MACCS code assuming a representative "worst" (no-rain case) meteorology. (2)A single, uniform puff of 2 hour release duration was assumed. Release
- occurs at ground level with no plume buoyancy.
(3) Individuals were assumed out-of-doors with no special shielding and three times the normal creathing rate were assumed. 7 NUREG TECH BASES GRADED RESP 04/20/87
~ -
I = 1 I I I - I = 0 2 0.1%/ day leak rate 2 10 =- =
= =
~ ;
Time from beginning - of the release (hours) , E
$1 =- ,
g
- 24 -
- s -
- 0.G
- 5 ,
~
~
1.5 21 ,= 0.1,=- 1
- 3 :
- 2.5' -
6 I i i I I Y 0.01 4' 6 8 10 12 O 2
. l I
Stance (km) l Figure 2 Total thyroid dose -.
.lotes:
(1) Calculations .made using MACCS code assuming a representative "worst" (no-rain case) meteoro1cgy. (2)A single, uniform puff of 2 hour release duration was assumed. Release *
. occurs at ground level with no plume buoyancy.
(3)lndividuals were assumed out-of-doors with no special shielding and three , times the normal breathing rate were assumed, 04/20/87 8 NUREG TECH BASES GRADED RESP ,
- , - -- w
' * , l I
l J assessment of dose savings, and for making a recommendation to the public . regarding protection action.10 See Table 3. These guideline dose levels ao not represent acceptable doses, but rather are projected doses chosen based on - a balancing of averted radiological risk (by taking a protective A.ction such as evacuation) and the risk (and cost) associated with taking that agtion using realistic accident consequence assumptions: L L These results show that the EPA PAGs of 5 Rem (whole body dose) arid 25 Rem (thyroid dose) are only exceeded near the reactor, within approximately a half-mile and one mile distance, respectively. It is noted that the NRC/ EPA Task Force accepted the principle that "acceptable values for emergency doses to the public under the actual conditions of a nuclear accident cannot be predetermined. The emergency actions taken in any individual case must be based on the actual conditions that exist and are projected at the time of an accident. For very serious accidents,
- predetermined protective actions would be taken if projected doses, at any place and time during an actual accident, appeared to be at, or above, the applicable proposed Protective Action Guides 5 (PAGs), based on information readily available in the reactor control room, i'.e. , at predetermined emergency action levels. Of course, ad hoc actions, based on plant or environmental measurements, could be taken at any time."
Extremely 3evere Early Release Scenario Calculations were made using the MACCS code and the radioactive source terms for large, early releases from core melt accidents. The meteorological data for each site are used to generate probabilistic dose information. It is pessimistically assumed that the wind does not shift direction throughout the accident. Radioactive concentrations within the puff-releases are decreased by b both wet and dry deposition mechanisms. The public's radiation exposure can be t estimated from these radioactivity levels. Dose calculations have been made for a range of source term estimates by sampling hourly sequences of meteorological data throughout a year. Results are given in Tables 4 through 11 in terms of the conditional probability (i.e. , 4 given an early, extremely severe release) of exceeding various red bone marrow dose levels. The source terms used in the calculations are presented in Appendix A. The range of values shown in the tables reflect the estimated source term uncertainties. These results indicate, for example in Table 4, that the chance of exceeding a 200 rem whole borty individual dose is pelatively small, and limited to within 2-3 miles assuming people evacuate at the start of release. These results are based on realistic assumptions regarding emergency response 11 (a 10 mph evacuation speed and, a 6-hour duration of exposure). Resultr, for a slower evacuation speed are included. These results generally support the distance of 2-3 miles for the size of the EEZ. Furthermore, one can observe the importance of when a general emergency is declared and when ev'acuation movement begins. The probabilities of exceeding 50 rem (red bone marrow dose) given ear.ly and late releases from core melt for the reference plants are given in Tables 6 through 10. The sensitivity of these results to the assumption of delay time before start of evacuation movement is also shown. These results indicate that 04/20/87 9 . NUREG TECH BASES GRADED RESP
a.*. cf
*6 Table 3 EPG's recommended protective actions to reduce whole-body and k
thyroid dose from exposure from exposure to a gaseous plume 80
$ Projected Dose (rem)
Whole body Thyroid Recommended actions
- Comments Population
<1 <5 No planned protective actionsb Previously recommended State may issue an advisory to seek shelter and protective actions may await further instructions be considered or Monitor environmental radiation levels terminated
<1 to <5 5 to <25 Seek shelter as a minimum If constraints exist, Consider evacuation. Evacuate unless constraints special consideration make it impractical should be given for Monitor environmental radiation levels evacuation of children Control access and pregnant women U S and above 25 and above Conduct mandatory evacuation Seeking shelter would be Monitor environmental radiation levels and adjust an alternative if area for mandatory evacuation based on these levels evacuation were not Control access ,,,
immediately possible z Emergency team workers SE E 25 125 Control exposure of emergency team members to these Although respirators and
-4 except for lifesaving missions. (Appropriate stable iodine should he
$ controls for emergency workers include time used where effective to limitations, respirators, and stable iodine.) control dose to
[
> emergency team workers.
N 75 Control exposure of emergency team members performing thyrgi# dose,may not be [ , lifesaving missions to this level. (Control of time a limiting factor for
- = ,
of exposure will be most effective.) lifesaving missions 5 m "These actions are recommended fo? planning purposes. Protective action decisions at the time of the incident must take existing conditions into consideration.
}"
At the time of the incident, officials may implement low-impact protective actions in keeping with the principle of maintaining radiation exposures as low as reasonably achievable.
, e_ cr Table 4 Probability of exceeding 200 rem red bone marrow dose E . ,
given early releases from a core melt accident (expressed in percent) g s 0 1. Start evacuation at time of release
, Surry Peach Bottom Sequoyah Grand Gulf Zion Distance 10, mph 2.7 mph 10 mph 6.7 mph 10 mph 2.7 mph 10 mph 4.5 mph 10 mph 2.5 mph 1 0 0-17 0 0 0 0 0 0 0 0-7 2 0 0-2 0 0 0 0 0 0 0 0 -
3 0 0 0 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 0 0 0 10 0 0 0 0 0 0 0 0 0 - 0 U
- 2. Start evacuation one hour after time of release Surry Peach Bottom Sequoyah [ Grand Gulf Zion Distance 10 mph 2.7 mph 10 mph 6.7 mph 10 mph 2.7 mph 10 mph 4.5 mph 10 mph 2.5 mph 6-72 32-75 0 0 0 0-10 0 10-20 0 29-52 h 1 2 . 0-12 0-22 0 0 0 0 0 0 0 0-15 cm 3 i 0-2 0-5 0 0 0 0 0 0 0 0-5 U S 0 0 0 0 0 0 0 0 0 . et' -r
'S 0 ,' 0" 0 0 0 g 10 - 0 O O O O G A 4
~
3
, e. ct i
Table 5 Probability of exceeding 50 rem red bone' morrow dose given l early release from a core melt accident (expressed in percent) k o" g 1. Start evacuation at time of release. w Surry Peach Bottom Sequoyah Grand Gulf ' Zion 10 mph 2.7 aph 10 mph 6.7 mph 10 mph 2.7 mph 10 mph 4.5 mph 10 mph 2.5 mph Distance , 0 5-40 0 0 0 0 0 0 0' 25-50 1
0 0 0 0 0 0 5-25 -
2 0 0-30 0 3 0 0-15 0 0 0 0 0 0 0 3-10 0 0 0 0 0 0 0 0 0 0-3 . 5 0 0 0 0 0 0 0 0 0 10 0 M 2. Start evacuation I hour after release. i Surry Peach Botton Sequoyah ,' Grand Gulf Zion l 2.7 aph
- 10 mph 4.5 mph 10 mph 2.5 mph Olstance 10 mph 2.7 aph 10 mph 6.7 mph 10 mph l 2 78-85 80-90 5.5-36 7-36 40-68 10-70 0 11-22 55-67 60-65 l 'g 1
- o 8-55 35-60 0-1 0.1-6 0 0-25 0 0 12-49 50-55 l .g 2 ,
2 3 0-25 10-40 0-5 0-6 0 0 0 0 0-7 20-50 0-0.3 0-20 0 0-0.1 0 0 0 0 0 ,,, ,}-10 h 5 E 10 0 0 . 0 0 0 0 0 0 0 0 B i B R
'O
Table 6a: Probability of exceeding 50 rem red bone marrow dose (expressed %) at various distances given early releases from a core melt accident Plant: Surry Distance Evacuation Starting: (, from 1 Hour 0.5 Hour Large - Reactor After At Before Basement Building Normal (railes) Release Release Release Shelter Shelter Activities 1 78-85 0 0 80-87 66-80 85-92 2 8-55 0 0 66-80 25-51 79-85
. 3 0-25 0 0 40-66 12-24 65-79 5 0-1 O O 12-22 0 19-62
' 's 10 0 0 0 0 0 0-1 20 0 0 0 0 0 0 Table 6b: Probability of exceeding 50 rem red bone marrow dose given late releases from a core melt accident (expressed in percent)
Plant: Surry 5 Distance Evacuation Starting: from 1 Hour 0.5 Hour Large Reactor After At Before Basement Building Normal (miles) Release Release Release Shelter Shelter Activities 1 0-13 0 0 11-48 0-22 21-79 2 0 0 0 0 0 0-39 3 0 0 0 0 0 0-2 5 0 0 0 0 0 0 10 0 0 0 0 0 0 20 0 0 0 0 0 0 04/20/87 13 NUREG TECH BASES GRADE 0 RESP
Table 7a: Probability of exceeding 50 rem red bone marrow dose given early releases from a core melt accident e (expressed in percent) , C Plant: Zion (' Evacuation Starting: . 0.5 Hour Large Distance 1 Hour Basement Building Normal After At Before from Reactor Release Release Release Shelter Shelter Activities (miles) 0 68-78 53-67 80 1 55-67 0 0 53-60 20-52 60-69 2 12-49 0 0 0 34-53 0-24 55-60
. 3 0-7 0-1 31-53 5 0 0 0 12,30 0 0-0.2 0 1-14 10 0 0 0 0 0 0 20 0 0 Table 7b: Probability of exceeding 50 rem red bone marrow dose given late releases from a core melt accident
'.. (expressed in percent) 5 Plant: Zion Distance Evacuation Starting: Large from 1 Hour 0.5 Hour At Before Basement Building Normal Reactor After Release Release Shelter Shelter Activities (miles) Release 0 15-65 1-48 48-74 1 9-55 0 0 0-45 0-12 12-56 2 0-12 0 0 17 0 0-53 3 0 0 0 0 0 0-24 5 0 0 O O O O O
l'0 O O. 0 0 0 0 20 0 14 NUREG TECH BASES GRADED RESP 04/20/87
Table Ba: Probability of exceeding 50 rem red bone marrow dose given early releases from a core melt accident (expressed in percent) Plant: Sequoyah { Evacuation Starting: ; Distance 1 Hour 0.5 Hour Large L from Reactor After At Before Basement Building h,ormal (miles) Release Release Release Shelter Shelter Activities 1 40-66 0 0 68-100 11-98 98-100 2 0 0 0 11-98 0-36 18-99 3 0 0 0 0-43 0 8-98
. 5 0 0 0 0 0 0-30 10 0 0 0 0 0 0 s'
20 0 0 0 0 0 0 Table 8b: Probability of exceeding 50 rem red bone marrow dose given late releases from a core melt accident (expressed in percent) ti Evacuation Starting: b Distance 1 Hour 0.5 Hour Large i from Reactor After At Before Basement Building Normal (miles) Release Release Release Shelter Shelter Activities 1 0 0 0 11 0 11-68 2 0 0 0 0 0 0 3 0 0 0 0 0 0 5 0 0 0 0 0 0. 10 0 0 0 0 0 0 20 0 0 0 0 0 0 9
- ? .
04/20/87 15 NUREG TECH BASES GRADED RESP i
.; . . ,e Table 9a: Probability of exceeding 50 rem red bone marrow dose given early releases from a core melt accident (expressed in percent)
Plant: Grand Gulf I Evacuation Starting: !. i Distance 1 Hour 0.5 Hour Large - from Reactor After At Before Basement Building Normal (miles) Release Release Release Shelter Shelter Activities 1 0 0 0 52-83 27-74 76-87 i 2 0 0 0 23-73 4-40 46-78 3 0 0 0 10-52 0-19 23-74 5 0 0 0 0-28 0-1 5-52 10 0 0 0 0' O 0-15 20 0 0 0 0 0 0 Table 9b: Probability of exceeding 50 rem red bone marrow dose given late releases from a core melt accident (expressed in percent) 5 Plant: Grand Gulf
- h. Evacuation Starting:
Distance 1 Hour 0.5 Hour Large from Reactor After At Before Basement Building Normal (miles) Release Release Release Shelter Shelter Activities 1 7 0 0 0-53 0-22 2-79 2 0 0 0 0-21 0 0-55 3 0 0 0 0-7 0 0-33. . 5 0 0 0 0 0 0-10 10 0 0 0 0 0 0 20 , ,0 0 0 0 0 0 4 l 04/20/87 16 NUREG TECH BASES GRADED RESP
Table 10a: P.robability of exceeding 50 rem red bone marrow dose given early releases from a core melt accident ,' (expressed in percent) Plant: Peach Bottom Evacuation Starting: 1 Hour 0.5 Hour Large f Distance from Reactor After At Defore Basement Building Mormal (miles) Release Release , Release Shelter Shelter Activities 1 5-36 0 0 44-40 23-17 49-57 2 0-1 0 0 21-23 0-6 36-39 3 0-5 0 0 5-13 0 18-34
. 5 0 0 0 0-7 0 0-21 10 0 0 0 0 ,, 0 0-4 20 0 0 0 0 0 0 Table 10b: Probability of exceeding 50 rem red bone marrow dose given late releases from a core melt accident (expressed in percent)
Plant: Peach Bottom t Evacuation Starting: Distance 1 Hout 0.5 Hour Large from Reactor After At Before Basement Building Normal
- h. Shelter Shelter Activities (miles) Release Release Release 1 0-27 0 0 3-27 0-13 7-36 2 0 0 0 0-15 0-1 0-22 3 0 0 0 0-4 0 0-10 5 0 0 0 0 0 0 ,,
10 0 0 0 0 0 0 20 0 0 0 0 0 0 s, 9 e e I 04/20/87 17 NUREG TECH BASES GRADED RESP
l an early, precautionary evacuation within 2-3 miles of the reactor site can *! provide substantial dose savings given a major accidental release such that l 50 Rem (Red Bone Marrom) dose levels (or less) can be achieved if the public is adequately warned and motivated to take protective action. Finally, calculational results are shown in Tables 11 and 12 for tioth the conditional probabilities of exceeding 5 Rem (red bone marrow dose) at two miles distance from the reactor, and the probabilities, respective 1y. A com-parison is shown in Table 12 using information from NUREG-0396, which is based on WASH-1400 data.11 It is observed that the risk of exceeding 5 Rem at ten miles which was previously chosen as acceptable (for representative worst case accidents) is comparable to the value at 2 miles using the new source terms and protective action assumptions. This risk perspective suggests a 2-3 mile choice for the radius of the EEZ. Furthermore, the conditional dose informa-tion (Table 11) supports an extended emergency planning zone size of approxi-mately twenty miles. A perspective on the relative importance of each radionuclide group to the mean red bone marrow dose at approximately one mile from the reactor is given in Table 13. Table 14 presents the relative importance of each pathway as a contribution to pxpected doses received from each radionuclide group. Table 11 Probability of exceeding 5 rem red bone marrow dose (expressed in %) given early releases from a core melt accident at a light water reactor (basement sheltering) Distance sg (miles) Surry Zion Sequoyah Grand Gulf Peach Bottom f 1 95-98 87-94 99.5-100 90-99 73-85 2 90-92 69-93 98-100 82-87 55-75 3 80-88 67-75 98-100 79-82 47-63 5 79 56-69 35-98 47-76 33-49 10 26-62 43-57 0-16 7-54 14-33 20 0 0-8 0 0-3 0'- 4' F 04/20/87 18 NUREG TECH BASES GRADE 0 RESP
. , . =.c Table $2 Probability > 5 rem (red bone marrow) at 2 miles distance from reactor ,2 3
[ basement sheltering) Reactor ( Name Probability (per year)3,2 ( Surry 2.3 x 10 5 - 2.4 x 10 5~ Peach Bottom 5.5 x 10 8 - 7.5 x 10 5 Sequoyah 1.08 x 10 4 - 1.09 x 10 5 Grand Gulf 2.1 x 10 5 - 2.26 x 10 5 Zion 1. x 10 5 - 1.4 x 10 5 PWR 1-5 NUREG-0396: Probability >5rematl'0'$iles=1.5x105 per yr (Wash-1400)
- 1. Assumes probability of early release given core melt is unity.
- 2. Range of values correspond to upper and lower ends of the range of mean source terms used in the calculation.
b T Table 13 Relative importance (in percent) of radionuclide to expected red bone mar, row dose given early large radioactive releases-Radionuclide Group Sr Ru La 8a Ce Xe I Cs Te 28.8 16 2.4 3.6 13.2 1.9 Percentage of 4.2 26.9 3 Expected Dose . g h (~ ., -t k e a ,i , _ : %.,. _Q
.f- y, ~6 u . M . . +y ( >: > k . -, .}. q .-< _) -
_ 4 . oh (% pn y s 19 NUREG TECH BASES GRADED RESP 04/20/87 .
f . L ' ', , [ l 1
.l l
Table 14 Pathway importance (in percent) for each radionuclide '!
, group given early large radioactive releases Radionuclide Group
- Pathway Te Sr Ru La Ce Xe 1 Cs ( Ba Cloud 99.7 69.8 49.5 24.8 9.4 60.2 23.2 7 13.8 Ground 0 25.9 18.4 31.4 11.9 ,23.9 29 9.5 20.6 Inhalation 0. 3 4.3 32.1 43.8 78.7 15.9 47.8 83.5 65.6
- 1. Calculations,'Mai:ie'using MACCS for an individual at one mile from the reactorassumingnormalactivity(sixhoursofexposure);n;r-t:-tS eriease>.x L
%/ 7o y l
-b t
i b 9 . 9 5 l ! 1 i 1 i t 04/20/87 20 , NUREG TECH 8ASES GRADED RESP
+ - - e ,m - , .- s w r V-"'-
;..".g-CONCLUSIONS i The present role of the U.S. Nuclear Regulatory Commission in a severe accident is one of men'toring the reactor licensee to assure that they are recommending appropriate protective actions to offsite officials. It is recogn42ed that as many predetermined action levels and protective action decisions (5 possible shouldbemadebeforeemergencysituationsoccur,therebyminimiz(ngthenumber of decisions that would have to be made in an actual emergency.23 h.
A set of three generic protective action Zones is proposed along with a protec-tive action strategy which would result in substantial dose (and correspondingly , health effect) savings to the public in the event of a serious nuclear power accident. The intent is to generally avoid early fatalities and injuries and be consistent with the planning guidance objectives of NUREG-0396. In summary, these consequence results and observations corroborate earlier predictions 12 that the greatest dose reductions occur within the first two to
- three miles and indicate that substantial dose savings can be achieved to generally avoid early fatalities and injuries. It is observed tnat to be most effective evacuation should take place before or immediately at the time of a major release to the atmosphere.8,23 i s b
t 9 0
. -~ ,
21 NUREG TECH BASES GRADED RESP 04/20/87
;.../
REFERENCES
- 1. L. Sof fer, J. A. Martin, Jr. and R. P. Grill, "An Examination of a Graded Response Strategy in Emergency Planning and Preparedness," Proceedings of the International Topical Meeting on Probabilistic Safety Me(hods and Applications, February 24- March 1,1985, San Francisco, CA, 'olume 2, page 128-1 to 128-10.
b
- 2. R. Blond, M. Taylor, T. Margulies, M. Cunningham, P. Baronowsiy, R.
Denning, P. Cybulskis, "The Development of Severe Reactor Accident Source Terms: 1957-1981," U.S. Nuclear Regulatory Commission, NUREG-0773,1982.
- 3. "Planning Basis for the Development of State and Local Government Radio-logical Emergency Response Plans in Support of Light Water Nuclear Po.er Plants," U.S. Nuclear Regulatory Commission and U.S. Environmental Protection Agency, NUREG-0396, 1978.
. 4. "Criteria for Preparation and Evaluation of Radiological Emergency Response Plans and Preparedness in Support of Nuclear Power Plants,"
NUREG-0654, U.S. Nuclear Regulatory Commiss,jon Federal Emergency Management Agency, 1980.
- 5. J. S. Evans, D. W. Moeller, and D. W. Cooper, "Health Effects Model for Nuclear Power Plant Accident Consequences Analysis," Harvard School of Public Health, NUREG/CR-4214, July 1985.
- 6. T. J. McKenna, J. A. Martin, Jr. , C. W. Miller, L. M. Hively, R. W.
Sharpe, J. G. Glitter, R. M. Watkins, "Pilot Program: NRC Severe Reactor Accident Incident Response Training Manual," Vol. 1-5, NUREG-1210, U.S. Nuclear Regulatory Commission, 1987. 4t 7. "Report on the Accident at the Chernobyl Nuclear Power Station," U.S. Nuclear Regulatory Commission, NUREG-1250.
- 8. D. I. Chanin, L. T. Ritchie and D. J. Alpert, "Melcor Accident Consequence '
Code System (MACCS) User's Guide," Sandia National Laboratories, Albuquerque, New Mexico, 1986.
- 9. L. T. Ritchie, D. J. Alpert, R. P. Burke, J. D. Johnson, R. M. Ostaeyer, D. C. Aldrich, and R. M. Blond, "CRAC2 Model Description," U.S. Nuclear i
Regulatory Commission, NUREG/CR-2552,1984. .. ! 10. "Manual of Protective Action Guides and Protective Actions for Nuclear Incidents," Environmental Protection Agency, EPA-520/1-75-001, June 1980.
- 11. W. F. Witzig and J. K. Shillen, "Evaluation of Protective Action Risks,"
U.S. Nuclear Regulatory Commission, NUREG/CR-4726, Draft, April 1986, 12'. T. S. Margulies and J. A. Martin, Jr. . "Dose Calculations for Severe LWR
- Accident Scenarios," U.S. Nuclear Regulatory Commission, 1984.
- 13. J. A. Martin, Jr. , "Emergency Response Decision in the Planning frocess -
l 1 Predetermined Conditions for Predetermined Actions," in CRC Handbook of l l 04/20/87 22 NUREG TECH BASES GRADED RESP
Management of Radi^ation Protection Programs, CRC Press,.Inc., Boca Raton, ,.. Florida, p. 143, 1986.
- 14. "Reactor Safety Study - An Assessment of Accident Risks in U.S. Com.Tercial Nuclear Power Plants," U.S. Nuclear Regulatory Commission, W85H-1400 (NUREG-74/014), 1975. ,
-15. "Radionuclide Release Under Specific LWR Accident Conditions, Batte11e, '
Columbus Laboratories, Columbus, Ohio, BMI-2104, Vol. 1, Vols'. Il-VIII.- Darft, July 1983 - July 1986. * ; G 1 , . D ' 5 I 4* e
# e 04/20/87 23 ,
NUREG TECH BASES GRADED RESP
/,.,*, ,
APPENDIX A: SOURCE TERM STATUS AND CALCULATIONAL ASSUMPTIONS Over the past several years, a major research effort has been under way to i develop an improved understanding of severe accidents and to provi,de a technical basis to support regulatory decisions. Current plans f7 the ; completion and extension of this research in support of ongoing r(gulatory ' actions are provided in NUREG-0900, Revision 1, Nuclear Power Plarts Severe Accident Research Plan (April 1986) and research plans addressing {sajor are of uncertainty are discussed in SECY-86-369 ("Plan to Address Source Term Technical Uncertainty Areas," December 1986). A key product of this Severe Accident Research Program is a reassessment of the severe accident source term technology. A source term is defined as the quantity, timing, and characteristics, such as chemical forms and particle sizes, of the release of radioactive material to the environment following a postulated severe reactor accident. Source term technology is employed for a
. variety of regulatory applications, including plant siting evaluation, emergency planning, and evaluation of performance of engineered safety features such as containment isolation and containment spray additives, qualification of safety-related electrical equipment for performance under accident conditions, environmental impact statements, post-accident monitoring requirements, and criteria for re-entry of a plant after an accident. In addition, an understanding and quantitative assessment of source terms is necessary for conducting probabilistic risk assessments, which are emerging as a significant part of the regulatory decision process. New information and insights on radioactive source terms may have an impact on rules, guides, and other regulatory practices in the aforementioned areas through implementation of the NRC Severe Accident Policy Statement.
- Early research on severe accidents has pointed out the need to integrate the h" analysis of complex severe accident phenomena to obtain realistic estimates of source terms. Research programs initiated following the accident at Three Mile Island resulted in a systematic, mechanistic approach to source term analysis i
that has been developed by Battelle Columbus Laboratories. This analysis is l encoded in the Source Term Code Package'and described in NUREG-0956, "Reassess-ment of the Technical Bases for Estimating Source Terms." (July 1986). The principal elements of the models which predict the source term include: in-vessel release from fuel, transport in the reactor coolant system, ex-vessel release from fuel, and transport in the containment and secondary buildings. These new analytical tools for estimating accident source terms for nuclear l l power plants have been exposed to extensive peer review and represent'an inte-grated mechanistic framework for NRC evaluation of source terms in regulatory applications. When this effort was initiated, there was an expectation among many in the nuclear community that a correct treatment of the physical and chemical behavior of fission product release and transport would show a reduction of l several orders of magnitude in calculated source terms, except for noble gases, ;
; compared with the Reactor Safety Study H (RSS). Such a result would have made it easy to develop new generic source terms, and such reduced source tarms would have translated directly into reductions in estimated risk without th i
need to reevaluate other areas of risk assessment (event frequencies, conta e L ment performance, and offsite consequences). Such clear-cut reductions in source terms have not been found, however, and the self-consistent method of 04/20/87 24 - NUREG TECH BASES GRADED RESP
. ;.,*. / evaluating source term has demonstrated a high degree of plant-specific 3 variation. A preliminary examination of the current models and source term rgsults to date indicate that the simpler Wash-1400 models overestimate-some of trp radio-nuclide releases to the environment in many accident scenarios, far example, current models predict significant retention of radionuclides in (pe reactor coolant system. However, the Wash-1400 analyses do not overestimale the source terms for all accident scenarios. Furthermore, for any particular scenario, it is possible for the release fraction of one chemical group (e.g., halogens) to decrease while the release fraction associated with another group (lanthanides) may increase. The uncertainties in the source terms are quite large and typi-cally encompass the Wash-1400 results. Uncertainties need to be considered for each regulatory application under consideration, due to their sensitivity to plant, accident type and phenomenology assumptions.
. The Source Term Code Package has been used in a major new risk study published in draf t form as NUREG-1150 Reactor Risk Reference Document," (February 1987).
The purpose of this study is to analyze the risktdue to severe accidents in selected nuclear power plants of representative reactor and containment designs. (Refer to Table A-1). Table A-1 NUREG-1150 reference plants Reactor Reactor Type Design Power Containment Name (Designer) MW(t) MW(e) Type Surry PWR (West.) 2441 788 Subatmospheric t i b Peach Bottom BWR (GE) 3293 1065 Mark I t Sequoyah PWR (West.) 3423 1148 Ice Condenser Grand Gulf BWR (GE) 3833 1250 Mark III Zion PWR (West.) 3250 1040 Large Dry The new effort has built upon NUREG-1050, the "Probabilistic Risk Assessment Reference Document" (September 1984). NUREG-1050 discusses the state-of-the-art of probabilistic risk analysis, provides summary information from existing PRAs, and gives general insights on the use of methods and results of PRAs in regulation. This latest effort provides the next step by documenting current PRA information on these selected reference plants, using state-of-the-art methodologies developing plant-specific insights about the dominant contrib-uters to severe < ore damage frequency and risk, and providing an explicit anal-ysis of their associated uncertainties. See Figure I which displays core melt frequency results for each reactor examinea. NUREG-1150 also.provides specific insights on the use of such information in making various types of regulatory decisions, considering the results obtained and the nature and magnitude of uncertainties. 04/20/87 25 NUREG TECH BASES GRADED RESP
.. o-
. .~
2 - - S 1E-3F REQUENCY, PER REACTOR YEAR
~
' - ~ 1 ~~4 -
~
, 1E5 7 _ _
~
~
~
l . 1 h ' T : .a o : M . a .
~
l 5 -
, m , ,
1E-7 , ggg gg gm mw 5 . R o E 4 Figure A-1 Comparison of severe core damage frequencies of reference plants (WIREG-1150)
,e .,
(1) Design Basis Leakage Scenario: , for a representative design basis accident leakage source term for emergency response considerations, and calculational purposes, a "station bl,ackout" scenario at the Surry plant 35 was chosen with the intent of genertlly envelop-ing the source term magnitude in the containment (TMLB'c scenariol for an early leakage scenario. The accident sequence assumes that,no sprays ate available to attenuate the radioactivity and provides high consequences of Tiakage. As a result, the pressure in the containment rises in the accident, prov' ding a driving force for leakage, and only natural deposition mechanisms are available in the containment for fission product removal. For this sequence, it is esti-mated that core melting begins at approximately 118 minutes (essentially the beginning of fission product release), core collapse occurs at 147 minutes, bottom head failure occurs at 157 minutes, core-concrete attack begins at about 290 minutes (the beginning of vaporization release), and containment failure occurs at 738 minutes. The approach taken was to explicitly account for a constant leakage of 0.1% per day in a revised set of NAVA2 calculations for the integrated release fraction of fission products. The following assumptions and approximations were made in the analyses of integrated leakage values for a two hour release duration from the beginning of cort collapse when the first signi-ficant release of fission products occurs:2 (1) Airborne masses of fission products associated with WASH-1400 groups 2 to 7 were obtained from the NAVA code run at 37 time intervals up to the time of containment failure; (2) One hundred percent of the noble gases were assumed to be released from fuel instantaneously at the start of core melting; and s :- (3) a 1100 MWe reactor was assumed. The following release fractions were used in the analysis:
- f. La ,
Cs Te Sr Ru Xe/Kr 1 9.1 x 10 4 7.1 x 10 8 6.9 x 10 8 1.1 x 10.s 9.86 x 10 7 2.6 x 10 7 1.4 x 1 (2) Extremely Severe. Early Release Scenarios: J Here, we briefly discuss information and a number of conclusions concerning extremely severe accidents, especially early releases, which are important to emergency preparedness, that have been drawn from the analysis prepared for NUREG-1150. Containment Perfo_rmance The information from the reference plant studies indicates that there is still co,nsiderable uncertainty with regard to predicting whether or not containments
~
2NAUA calculates natural processes of agglomeration and deposition in' contain-ment that may involve condensing steam atmospheres (NUREG-0956). eSource Term calculations were provided by R. Denning at Battelle Columbus l Laboratories. 27 NUREG TECH BASES GRADED RESP ! 04/20/87 .
, / , ,'
will fail early. Uncertainty ranges for early releases from a core melt acci- ; dent are given in Figure A-2. The early release is an important consideration > in determining the potential consequences of a severe accident. This timing of radioactivity release is important because it subs ntially affects the magnitude of release (the earlier the release, the la ger the source term) and it affects the time available to take emergency grotective actions in the vicinity of the plant (i.e., within a few miles). ',7hese rela-tively high probabilities of early containment failure (conditional on core melt) suggest that early warnings to the public to take t:ffective protective action would be prudent if there is a cort damage accident. Source Term Characteristics Release of radioactive material to the environment during most severe accidents (particularly those resulting in early containment failure) is expected to
, occur in two distinct phases. The first release is of short duration, usually occurs before there is significant core concrete interaction, and consists of all of the noble gases the more volatile fission product species together with significant fractions of the more (i.e.,,901atile species, such as Cs!
and Te). The second release occurs after the core materials have melted ' through the reactor pressure vessel and are interacting with the concrete cavity. This release usually takes place over a period of several hours. Refer to Figures A-3 and A*4 for a selection of representative source terms for core meltdown, early release scenarios taken from NUREG-1150 which indicate estimates of the rangers of uncertainty of the source term magnitude by radionuclide group. The separation of the fission product release into these two phases is an insight that appears to be important to emergency preparedness.
- 6.
- b. Predicting the magnitude of the first release is important for accidents in ,
- which the containment fails early because it poses the immediate health threat to the population surrounding the plant and thus it affects the area in which early protective action should be taken to provide the greatest reduction in early health effects. Most of the late fission product release would normally occur after emergency actions have already been taken in anticipation of, or in response to the early release. Thus, the magnitude of the late release is of ,
lass importance to emergency preparedness because the necessary protective actions should already have been taken. A selection of source term assumptions for the dose calculations for. the extremely severe accidents presented in the text are summarized in Table A-2. 1 i ) O 9 i . h 04/20/87 28 . NUREG TECH BASES GRADED RESP
= . l .,* <s
.e-W. 1 J
.I t oc mia ciases witn ti% a west twaogon
, _ mt ati%G -
s og - " e o ti%g e h ! - un ti i=n ovc= -
< C IC i.
{ -- we cierci h -
= me as.g o 3
2 CK -
- E -
2 m g eu - 2 - C -
~
~ ~ -
U - - CW - 5 5 _ ,=, _
<oc - -
- _ 9 3 ex = - _ i -
t s = r s -
. " - = [ _~
E 0 70 - as s - t= 5 = = a 6 < - - g e ,. - z ~ i 3 E
- 5 I e W .
w PlaCH BOT 10V CRMD
$URRv rioN st ovot AH GVt' Range of early failure probability for.each of the analyzed plants
.. +3*-5 6, a.
t.
- s
+ v..'.e Dee:: : Nc ~s:-
+ H08tr$ 4 1'."O 3 4 g; .
,o 4.
+
- +
++ M
+ + + +
+ 46+ + _ _ + + ++ + +
it f ,4tH M } . + 44+4+ 4. .d+ .D+ _ + + +. . M
*; . . . .s 4 ...
Ey : c .s - + - t- .+: :v - Figure A-2. Conditional probability of early containment failure 29 NUREG TECH BASES GRADED RESP 04/20/87 .
e er ::
.. ~.
2 s i E E f g 1EO _ nvironmental Release
- x o i
-O
+ n 1E-1 r o +
~
X i
- x o 4
1E-2 7 i :
- + x o
1E-3 r os -
*a8 -
Iladlonoclide Group -{-
- Reactor Safety Study i ... . i ,n,i -
se-4 - j MARCH / CORRAL $ U'" 4 tellectum
~
+ Source Term Code Package s ..;i . st..ti-
=
?, 1E-S r a s r.t hcai-o M -
l NUREG-1150 [ Bin 1)
' ' '"a 'd $ a " 'de s 9
"g
'tE-6 y } } } [ j 0:
g ; Radionuclide Group
?;
G A Figure A-3 Comparison of results for station blackout scenarios at Surry
. . cr ., .
=
g . . 3 1EO E nvirormental o Release R
=
xo X 0
- x
+ 4
+
- 1E--1 =- x +
3- 0 x .
+ +
1E-2 =- o 5 x 1E-3 p 2 Itadionuclide Group y x Reactor Safety Study n ,~. santo. r iodi-o MARCH / CORRAL 3 ceslw 1EM 2" [ + SOwce Term Code Package a teit.rio. -
- 5 bartum, stronti,um
~
'a~'- '
x y NUREG-1150 F ianthseldes. actinides g 1E-5 g nx :
- I I
f f
"p a f
6 7 3 4 5 M 2
- Radionuclide Group 8
G A T, Figure A-4 Comparison of results for interfacing systems LOCA (without scrubbing) I _ _ _ _ _ - _ - _
-m O
l=' e s. Q'
.^
1 Table A-2 IRIRfG-1150 source teros fee representative. tate retcases i Release Heat Release Fraction 8 by Isotope Group Plant Meme $ewrte Release Release 4 5 6 7 duratioso (Mcal 1 2 3 Ter=8 phase time Te Ra-5r Ris la
/sec) Kr-Ne 1 Cs-Rb (br): (br) -
.69 .62 02? O .14 0 3.5 .32 2.8 1.0 Sorry Migh 1 0 0 .54 .32 0 ' t.2 3.8 8.01 034 0 2
.15 047 .02 0 00 % 0
.290
~
2.0 2.45 1. 0 ^c tow 1 0 0 .0095 .0078 0 .n012 2 4.4 1.19 0 0 44 .093 076 .037 .0031 .01n .0013 MIgh 1 6.5 46 .9 0 .0072 Segueysh
.0'33 .56 47 43 .24 .37 2 7. 8 5.99 0078 0036 .0012 . G=9E4 00012 4.2 .93 .22 .72 .021 tow 1
.28 .0092 0 0 0 000092 0 2 5.4 6.06 .041 Meet Release Fraction 3 by Isotope Group Plant Name Source Release Selease Release 6 n 9 1 2 3 4 5 /
Termi phase time duratiese (Mcal Ce na
/sec) Kr-Me 1 Cs-Rb Te Ba-Sr Ro la (br)8 (hr)
.0'14 .011 3.4 2.0 0 1.0 1.7 2. 7 .97 .045 Grand Guir Nigh 1 6.8
%(5 6) (-9) (-10) (-4)
.16 .33 43 03 .018 .069 .32 2 8.6 4.5 0.32 .028 .15 1.7 .97 .034 .0091 .0075 3-1 2.4 (1.6- 0 6.n tow 5.5 2.6
(-6) (-9) -10) (-5)
.06 .026 .0023 1.4 7. 7 0012 .016 2 7.2 4.4 .28 .28 .1
(-4) (-4)
.602 608 .131 .205 .0548 .009 .146 1.8 5.0 1.5 .974 .725 Ziese Mi@
.065 .344 .116 .0% .04n * .011 .117
- 2. 4 5. 0 1. 2 .924 .178 tow 0.6 0.022 0.012 3. 0 0 0 3.2 Peach Bottom ~ Migh 1' e 3.4 2. 4 -
(4) 0.89 ' (-67."~ " ' (-4) g 0.18 0.44 0.68 9.2 0.014 0.26 .o49 3.4 13.62 6.4 0.11 0.13 2 ' (-6) 2 3.9 o o o 2.6 tow 1 2.3 0.75 0.49 n.65 3 .- (-n) (-6) (-6) (-7) (-10) s.s 0.35 0.019 0.15 0.074 2.2 3.n 3.n 2 3.0 14.00 0.11 0.2 (-1) (-41 (-11 (-7)
O ye* to G" '
.e We 4
Table A-2 (Cont.) fB M C-1150 source teens for representatie , early releases Plant teame Release Selease Release Meet Release Fraction 3 by Isotope Group phase time duration (Mcal 1 2 3 4 5 6 / e (h) (h) /sec) Kr-Ne I Cs Te $r Ru ta Ce Sorry Nigh 10.2 2.71 [0.033 1.0 .097 .04 .05 .0063 .0023 00045 tow
- 11.4 3.65 0.021 1.0 .0036 .00046 .00061 .00n06 00n03 00n006 5egueweh Migh 1 6.9 0.1 3.8 .97 .011 055 .00012 0 0 0 2 35.0 1.7 .19 .03 .25 2 .12 .055 .0061 0014 tow 1 6.1 2.17 .15 .91 .0062 .0016 .002 .00027 00n014 .00001)
P 2 8.9 7.8 .047 0 .023 0073 .0017 .0011 .00023 .00n001)
. Sevrce Release Release Release Meet Release Fract ton' by Isotope Cem, Tere phase time duration (Mcal/ 1 2 3 4 5 6 / a 9 sec) Kr-We 1 Cs Te Sr-Ba Ru ta Ce Ra Zfon Nigh 9.8 2.7 9.46 1.0 48 .504 .299 .075 .004 .005 .001 .023 tow 7.9 1.8 0.06 .926 .006 .01 .005 .0005 0 0 0 .00n5 Proc 9t Settee Migh 1 2.3 .008 30.0 .9 .012 .0042 .021 3.3 .0s7 3.3 3.3 3.3
(-4) (-4) e(-4) (-4) 2* 3.1 13.62 2. 6 .999 .16 .07 .1 .03 2.9 .0018 .coln .016
. H 4) a i tow 1 6.7 0.01 65.0 .87 2. 0 1.4 1.5 F. 0 1.3 1.1 8.4 I.2
(-4) (-4) (-4) (-6) (-n) (-7) (-1) (-s) 2 6.7 10.30 0.16 .13 .044 .0036 .0031 T. 3 3.5 1. 3 2. s 2.4 (-6) (-p) (-1) (-1) (-5) _.r- _ _ . , y _ - ,
o'.v' ty. Cf ' : C_ - 1 Table A-2 (Cont.) NUREG-1150 soerce teries for representative early releases Release Heat Release Fractions by Isotope Group F? ant h Re3 ease Release 4 5 6 / 8 time (oration (Mcal 1 2 3 phase Te Sr Ru La Ce
/sec) Kr-Me 1 Cs (h)
.. 2.
(h)
.38 .45 .0026 .0014 , 1.8 5.5 9.0 .001 Grand Gulf High 16.8 4.2 0.3 1.0
(-5) (-5) (-5) 0.2 .% .045 .13 0013 .001 7. 4 5.0 8.4 7. 3 tow 13.4 4.2 (-5) (-4) (-5) (-5) [ 9 4
.O t
t
f; ' the D' APPENDIX 8: PROTECTIVE ACTION MODEL ASSUMPTIONS . Calculations are shown in Tables 5 and 6 to illustrate the effectiveness of various protective actions for radioactive releases at the refereqce plants. The assumptions used to model the different protective actions art outlined below. One alternative called "normal activity" assumes no prote4tive actions are taken. Instead,' people are assumed to go about their normal 4ctivities spending their time as follows: home, 69.2%; school or work, 19.6T; commuting, 5%; and outdoors, 6.2% (from Table VI 11=10 of the RSS). Shielding factors for gamma cloud sources are 0.75, 0.2,<1.0 and 1.0 respectively (from Table VI 11-7 of the RSS). The composite gamma cloud shielding factor is 0.67. Shielding factors for groundshine are 0.3, 0.02, 0.5, and 0.7 respectively (Table VI 11-9 of the RSS). The composite shielding factor for groundshine during normal activity is thus 0.32. No credit is given for reduction in inhalation exposures.
. Another alternative protective action considered is "sheltering" in a home basement. The cloud gamma shielding factor was taken to be 0.5, an average of the values for masonry and wood frame houses from Table VI 11-8 of the RSS.
The shielding factor for groundshine was taken t'o be 0.05 (Table VI 11-8 of the RSS). For the inhalation protection factor, it is assumed that people would close windows and stop ventilation. Assuming this would reduce ventilation rates to 0.5 air changes per hour, the protection factor for a 1-hour release would be 0.2 and for a 10-hour release would be 0.8 (from Equation VI 11-1 of the RSS). The sheltering effectiveness of a 1arae building assumes: 0.2 for the cloud gamma shielding factor, 0.01 for the ground shielding factor, and 0.1 for the inhalation protection factor. The other alternative is "evacuation." The effectiveness of evacuation was calculated with differing delay times. It was assumed that the minimum delay would be 0.5 hours within 2 miles of the plant. It would apply to people who
- q. are readily mobile and who could, if so motivated, jump in a car and leave. In other words, it applies to most of the people most of the time. The radial speed of evacuation is assumed to be 10 mph, a speed generally achieved in i
large evacuations that hve occurrcd. This is faster than the speeds used in ' NUREG-1150. We use 10 mph because it is appropriate for the phased response strategy developed in this report, whereas the speeds in NUREG-1150 were based on the assumption of simultaneous evacuation of the entire 10-mile planning zone. Studies of actual evacuations have found little evidence of evacuation speeds slower than 10 mph, particularly for prolonged periods, except for queueing while awaiting reentry into evacuated areas after the emergency is over.20 Evacuation risks were also computed for 'a delay of I hour a'fter start of release and slower evacuation speeds. The protection factors for evacuation are as follows. Prior to the evacuation the factors for normal activity are used. These are: 0.67 for the cloud shielding factor 0.32 fo- the ground shielding factor, and 1.0 for the inhala-tien protection factor. The assumption is that people's activities would be more typical for normal activities than sheltering. They would move throughout th*eir houses packing belongings and would from time to time go outsidi, for example, to load their cars. After evacuation starts, the protection factors would be those for automobiles: 1.0 for the cloud shielding factor (WASH-1400, Table VI 11-7) 0.5 for the ground shielding factor (WASH-1400, Table VI 11-9), and 1.0 for the inhalation protection factor (no protection assumed). 35 NUREO TECH BASES GRADED RESP 04/20/87 , ,}}