ML20149F687

Final Draft Rept, Issues,Info Needs & Programs for Improved Emergency Preparedness

ML20149F687
Person / Time
Issue date:	10/31/1984
From:	NRC
To:
Shared Package
ML20149B718	List: ML20149F619 ML20149F626 ML20149F650 ML20149F663 ML20149F678 ML20149F687 ML20149F777 ML20149F792 ML20149F799 ML20149F812 ML20149F826 ML20149F830
References
FOIA-87-743 NUDOCS 8802170316
Download: ML20149F687 (180)

Text

{{#Wiki_filter:'3 % g ISSUES, INFORMATION NEEDS, AND PROGRAMS FOR IMPROVED EMERGENCY PREPAREDNESS Tf. L:. > v 4 DRAFT ~ 8 e For the U.S. Department of Energy d Final Draft ReDort DRAR Prepared by the Working Group on Emergency Preparedness \\ October 1984 8802170316 080204 PDR h LL a -743

~- \\s 5 lr. TABLE OF CONTENTS I h: EAlt i EXECUTIVE

SUMMARY

1 {*.' L 1. INTRODUCTION AND BACKGROUND 7 1.1 Purpose and Importance of Energency Pre-paredness Function 10 1.1.1 offsite Consequences 11 1.1.2 Energency Responses 14 1.1.3 People and Their Institutions 19 1.2 Analyses of a Graded Response 21 1.3 Description of the Energency Preparedness Subfunctions 30 1.3.1 Plan for Emergencies (4.1) 31 1.3.2 Maintain Readiness (4.2) 32 1.3.3 taplement Plan During Ener gencies (4.3) 33 1.3.4 Recover from Emergency (4.4) 35 2. IDENTIFICATION OF ISSUES AND NEEDS IN EMER-GENCY PREPARP"NESS 36 2.1 Discussion of Issues and Information Needs 36 2.1.1 Public Awareness Programs (4.0.A) ,36 2.1.2 Range of Accidents and Planning Distances (4.1.1.A)* 37 2.1.3 Plan Coordination (4.1.2.A) 38 2.1.4 Acconnodating Groups with Special Needs (4.1.2.5) 39 2.1.5 Establishing the Chain of Command (4.1.2.C) 39 2.1.6 Energency Action Level Decision Criteria (4.1.3.A)* 40 2.1.7 Public Protective Action Decision Criteria (4.1.3.5)* 40 2.1.8 Unified Decision Model (4.1.3.C)* 41 2.1.9 Exercises / Training (4.2.1.A) 42 2.1.10 Public Awareness of Their Involve-ment in an Energency Response (4.2.2.A)* 43 2.1.11 Facility Readiness (4.2.3.A) 43 2.1.12 Dose Projection Capability (4.3.1.A)* 44 2.1.13 Adequacy of Accident Progression Prognosis capability (4.3.1.3)* 45 2.1.14 Unified Prognosis Approach (4.3.1.C) 45 2.1.15 Systematized Decision Model (4.3.2.A)* 46 y

~ 1 4 s' l p -a -} TABLE OF CONTENTS (continued) a.

• S.

_1 iX:

2.1.16 Protective Action Effectiveness /.,' ~ Prediction Capability (4.3.2.3)* 46 t ' ', 2.1.17 Onsite Actions (4.3.3.A) 47 2.1.18 Public Response to Instructions i n' '. (4.3.4.A)* 47 .' a 2.1.19 Coordination of Communications Among e i. - Various Organizations (4.3.4.B) 48 2.1.20 Emotional Conflict-of-Interest for Emergency Workers (4.3.4.C) 48 c . '. i 2.1.21 Feedback Provisions (4.3.5.A) 48 e' 2.1.22 Recovery Criteria (4.4.1.A) 49 li, 2.1.23 Role of Public Perception (4.4.2.A) 50 'T 2.2 Prioritization of Issues and Needs 50

,5 s

3. REVIEW OF ONGOING RD&D RELATED TO EMERGENCY PREPAREDNESS ISSUES 54 3.1 Review of Ongoing RD&D Programs 55 4. ADDITIONAL PROGRAM NEEDS 56 4.1 Evaluation of the Extent that Issues and Information Needs are Adequately Addressed by Ongoing RD&D Programs 56 4.2 Information Needs not Adequately Addressed by Ongoing RD&D Programs 56 4.3 Program Recommendations Plans for Reso-I' lution of Residual Information Needs 73 1 APPENDIX - Existing RD&D Program Summaries 75 l l. V. h ~' l-t l vi 1 t

m -- 7 \\% ?o' LIST OF FIGURES ( Fioure g 1-1 Conceptual Framework for the Integrated Approach to L4R Safety 8 1-2 Mean Probability of Early Fatality vs Distance (SST-1) 24 1-3 Mean Probability of Early Fatality vs Distance (1/3 SST-1) 25 1-4 Graded Response for Source Tera SST-1 27 1-5 Graded Response for Source Tera 1/3 SST-1 28 LIST OF TABLES Table 1-1 Nuclear Accidents and Their Associated Responses 17 1-2 Assumptions Employed for Consequence Cal-culations 22 1-3 Two-Mile Mean Early Fatality Risk 25 2-1 Priority Conversion Matrix 52 4-1 Working Group on Energency Preparedness Correlation of Information Needs with RD&D Programs 57 4-2 Summary of Recommended Resolution of High-Priority Information Needs 68 4 l I l l vii

s' -[ ( ZXECUTIVE

SUMMARY

j 1. Major conclusions P* Anelysis of various emergency responses to potential severe [' releases of radioactivity from nuclear power plant accidents shows that a graded response casults in a very low early -tatall'ty risk. For a typical site this graded response would be based on a two-alle evacuation radius. For the '} annular zone between the evacuation radius and 10 miles. sheltering with subsequent relocation would be utilized. Beyond 10 miles, discretionary processes would be employed as is the practice today. Such an approach to severe releases is conservative and warrants further review as source-term technology evolves. For smaller releases where the principal health concern is early radiation illness, discretionary processes such as evacuation and sheltering may be utilized. Discretionary processes need not meet the strict regulatory compliance requirements assigned to actions intended to reduce the early fatality risk. l i' This vastly laproved way of implementing energency planning n does not rest upon reduced source terms. not does it chal-lenge the size (10-mile radius) of the Energency Planning Zone (EpZ). Thus, it is legally ready for implementation now, without source-term or emergency-planning rulemaking hearings. f 2. Purpose. Goals, and Principal Functions l e DOE ettorts under Public Law 96-5,67 have been organized about the conceptual framework shown in Figure 1-1. In this scheme the top level goal is to identify practical generic 1 4

.i / e .a .w .1 j improvements that would enhance the economics. reliability. s .,f-[ and safety of nuclear electric power production, starting [],'{' from this goal. a set of top level functional requirements (E.E was defined along with a hierarchy of required supporting C gy: functions. The top level functions are the following: p, [-Q 1. To maintain the normal operating envelopa (i.e., assure y!' that temperature, flow rate. pressure, etc., are kept Q,. within the design limits) t .le 2. To protect the core and plant against damage in the l '..l. ],A event that plant processes deviate from the normal operating envelope l' ; 3. To contain radionuclides released from the reactor cool-ant system to assure that releases to the environment are kept within acceptable levels for low probability events when the core and plant protection functions are ) not maintained f I' j' 4. To provide adequate emergency preparedness to protect i the health and safety of the public especially for i, those extremely low-probability events when the radio-4 I:, activity containment function is not maintained. 1 l-This is the report of the working group concerned with the f, fourth function, to malatain energency preparedness in the event of an accident such that protection of the public is I[, provided. ,I. pl The energency preparedness goal was divided into four major ..l. subfunctions (;, i I I i 2 e a_w-=wiw-- e --w-, .m

---

,,w-,-.---

-_r..,- y m..

---

w,

,- - - -, - -,ww----m.-

~ t l 1-4 i ( 1. Plan f.or Energencies r The planning function is where the questions of who, what, i t (by) when, where, and how are addressed. The maximum degree N of decisionmaking and preparation should be done during the g. planning phase since there is ample time to think things I through, perform analysis, conduct studies, etc. 2. Maintain Readiness i N i I Maintenance of energency preparedness is the subfunction which ensures that a state of readiness is present at all times. Within the maintenance subfunction is the fostering f of a smooth transition f rom normal to emergency conditions i and public knowledge of and confidence in appropriate ener-gency responess. Readiness applies to emergency workers. l the generet public, and the energency facilities and i equipment. t 3. Implementation of the plan During Energencios The purpose of the implementation subfunction is to carry ( out the plan during an actual emergency. The function includes the collection and processing of information to j make assessments and predictions, decisionmaking, and issu-L ing of energency response and public-protection-action recommendations. Monitoring of the effectiveness of emer-j gency response actions to either validate or reconfigure the j. response to be more effective is part of this function. i 4. Recover from the Energency l f Once the energency situation has subsided and plant condi-l tions have returned to normal (or at least are stable), then decisions have.to be made as to whether any efforts are i i ) 3

9 em y .g g g

• i

,= h l 's%. 4{d. ) h. -required to protect the pubile in the longer term. These

r. :}

W.' recovery actions might, f or e:: ample, include decontamination ..y 'y; of land and property. rN fj[.' This organization of subfunctions and topics enaoles.the emergency activities to be described and treated in an Q) organized, reliable, and effective fashion, !.k This report focuses on emergency-preparedness actions that S are required as a result of core-damage accidents which also ,[ involve at least partial loss of the radionuclide contain-ment function. Energency planning covers a large spectrun d! of accident scenarios, from relatively benign ones requiring k!,, only onsite e'aergency actions to extremely race but severe ' ~ accidents involving a potential or actual release of signif-f,' icant quantitle's of radioactivity into the environment such as to lead to a need for offsite emergency twsponses involv-ing the public. A substantial fraction of. the energency ) + preparedness effort goes towards the less severe accidents ~ because they occur more frequently (e.g., a fire or a valve rupture). However, most of the issues surrounding the emer-5.. gency preparedness function revolve around the low probabil-l:, icy severe accidents and the reduction of their offsite con-sequences. Thus, these accidents are the ones that lead to } nost of the identified issues. '[ offsite radiological consequences can be divided into eco-nomic consequences and health consequences. To a very large degree, offsite economic consequences are independent of energency responses. For these reasons, offsite emergency C5 rianning is largely directed towards reducing health conse-H:.. l quences. D'j

• P, L'.

I l' 4 .. -i - -

s 0% .c i q ( 3.- Issues,and Information Needs ??() The principal functions and subfunctions were examined in f.: detail by the working group to identify issues associated L,I, with the functions and to identify the information needed to [{ resolve the issues. These issues and information needs were (- then prioritized in accordance with the'prioritization scheme developed by the Integrating Conaittee. The issues, y, information needs, and priorities are listed in Table 4-1. Priorities are listed from priority 1 to priority 6, with priority 1 representing those information needs judged to be e' most urgent. 4. Review of ongoing Programs to Detaraine Their Capability to Fulfill Information Needs Existing,- funded light water reactor programs in the United States were reviewed to ascertain how effective these pro- _ grams would be in providing the needed information. Con-sidered were programs sponsored and funded by the Nuclear Regulatory Commission (NRC), the Federal Emergency Manage- ' ment Agency (FEMA), and the Atomic Industrial Yorum (AIF). Information to be supplied by the programs was matched as appropriate with the information needs to estimate the extent to which the needed informatior. would be forthcom-ing. Based on the matching process, a judgement was made as to whether or not existing programs would supply the needed information. With respect to programs, about 13 programs were identified ^ and reviewed for the maintain energency preparedness fune-i tion. There were high-priority information needs (priority 1 or 2) associated with 14 of the issues of which a need for additional information was identified for all but one. These issues and the types of information needed are sun-marized in Table 4-2. 5

i ' i j d.?. [.7 It is important to recognize that recommandations for addi-5 F~ };- tional program activities do not indicate that existing 'i] nuclear electric generating stations are not adequately k[ safe. Rather the recommendations indicate areas where a ~ j further RD&D or engineering activities could be fruitful and cost beneficial with respect to the top level goal (set Q: s... forth in Figure 1-1) of producing electricity in an economic, reliable, and safe way, a N'. ,92 i .f ',5 b 1 1 ) o e e' 4 ,.4 4 4 5 g ,'6 b 6 -a.- ,~g..----. _y_ y 3 y

r 4 . u. + i.?n '( 1. INTRODUCTION AND BACKGROUND

i.

? Public Law 96-567 "Nuclear Safety Research, Development and ..,,l Demonstration Act of 1980," directed the U.S. Department of Yj Energy (DOE) to undertake an accelerated and coordinated Program aimed at developing practical generic improvements to ,V, enhance the capability for safe, reliable, and economical [ operation of light water nuclear reactor power stations. The specific roles of the DOE in this program are to provide the overall coordination required among the DOE, the Nuclear Regulatory Commission (NRC), the electric utilities, and the nuclear supply and construction industry and to manage and direct the department resources for implementing the programs. DOE efforts under Public Law 96-567 have been organized about the conceptual framework shown in Figure 1-1. In this scheme..the top level goal is to identify practical generic laprovements'that would enhance the economics, reliability, an'd s'afety of nuclear electric power production. S'ta r t ing from this goal, a set of. top level functional requirements was defined along with a hierarchy of required supporting functions. The top level functions are the following: 1. To maintain the normal operating envelope (i.e., assure that temperatures, flow rates, pressures, etc., are kept within the design limits) 2. To protect the core and plant against damage in the event that plant processes deviate from the normal operating envelope l i I 7 ,_.,-,-,_w

. -...:.:.;:. i r. 2:..' f.",',?-:'.}. :',:q,.*' /k's.;;:s..?;gj:l;v{!,r,..*8. <.,r ' ',.~' ~ l l ECONOMICAL, RELIABLE, SAFE NUCLEAR POWER PRODUCTION 0.0 as MAINTAIN NORMAL MAINTAIN PLANT MAINTAIN CONTROL MAINTAIN PLANT OPERATIONS PROTECTION OF RADIONUCLIOE EMERGENCY RELEASE PREPAREDNESS 1.0 2.0 3.0 4.0 Figure 1-1. Conceptual Framework for the Integrated Approach to LWR Safety s %d

.a. 4 .f-Lf 4/ I 3. To con.tain radionuclides released from the reactor fu coolant system to assure that releases to the environment [ are kept within' acceptable levels for low-probability /.. events when the core and plant protection functions are j',. not maintained E2t) s'. 4. To provide adequate energency preparedness to protect the health and safety of the public, especially for those extremely low-probability events when the radioactivity containment function is not maintained. These four functions also define the "states" through which an accident must sequentially progress in order for public health and safety consequences to result. 'In order to assure a balanced perspective, the DOE assembled industry working groups to address each of these top level functio's. The assignment of-the working groups was to n provide recommendations to the DOE on the need for additional light water reactor safety research after careful consideration of the following four questions: 1. What are the relevant issues and their priorities? 2. What information is needed to resolve each issue? 3. What of the needed information for each issue will be supplied by existing domestic U.S. light water safety programs? 4. What information will not be obtained by existing programs and what are the priorities of these remaining information needs? l 9

r,, g1 jft h. Those issues that are judged to be of high priority and for Ah which ongoing proge'ans are judged inadequats to provide the )hh needed information can form the basis for future program <k' k recommendations. j? if This is the report of the working group addressing the !$i energency preparedness function of the integrated approach .a.' b conceptual framework presented in Figure 1-1. .. ?' 0;' cU The scope and limitations of the effort described in this g',. report are as follows: .. T. "S 1. The study considers exclusively light water reactors as ' U.,' directed in Public Law 96-567. .C 2. The study primarily focuses on issues and information ,I needs for unlikely accidents involving the loss-of-core-T cooling sequences that progress to core damage. ) d. 1.1 Purpose and taportance of Esercency Preoatedness Function ~ (, Although the probability of a severe accident at a nuclear power plant is very small, and the likelihood of severe [ health effects is such smaller, it is prudent nonetheless to .I, have plans for protecting the public in the vicinity of operating plants. Therefore, the purpose of emergency plan-ning for nuclear power plants is to reduce consequences should an accident occur at one of these plants. While some nuclear accidents can have both onsite and offsite effects, this effort is principally directed at offsite consequences. .,a -U Our. task has been to determine what the present research and i, development needs are, if any, to implement more effective ,Y emergency planning. In order to accomplish this task, it is first necessary to consider the nature of the offsite nuclear } 10

---

r-,_,,

.,.-m_, .,-,-_g._.,. y-,, ,--,m.--v---ec

? s t ' s at GI

• s I

consequene.es, the ability of various energency responses to f.[ reduce these consequences, and people and their institutions (:b required to carry out the responses. ) 'M 1.1.1 offsite Consequences li With regard to the nature of offsite nuclear power plant V' consequences, these can be divided into economic consequences i 9. and health consequences. In most nuclear accidents that have j been analyzed, almost all of the economic consequences would ,' [ be onsite. Although the possibility exists for offsite eco-e, noalc consequences, to a very large degree offsite economic consequences are independent of energency responses. For thwse reasons, of f site emergency planning is directed towards 7 reducing health consequences. It is possible to divide,of'fsite health consequences into two broad categories, early health effects and latent health effects. Various studies have shown that, even for severe hypothetical accidents, nuclear power plant latent health risks are a very small fraction of nonnuclear latent health risks. These studies also show that most of these potential [l radiological latent health effects are due to radiation expo-sure that occurs long after radioactive material was released from an accident. Further, most of this exposure occurs beyond the Energency Planning Zone (EPZ). Because of these factors, latent health effects are only marginally affected by emergency responses. It is therefore ~ the early health effects that offer the most potential for reduction through energency planning. Early health effects from potential nuclear accidents include early radiation illness, from which one recovers, and early l fatalities. In the entire history of nuclear power in the l l l 11 l l

y,. 1 o O di NA IN7 -((i United States, not one offsite early fatality or early injury .. ) ..y has occurred from any nuclear accident. iQ

4

[' A great deal is known about health effects from acute radia-(p} tion exposure, such as the radiation exposure levels required

,.)j;1, to produce an early fatality or an early injury.

It is also y,?ij possible to calculate, assuming a certain amount of radio- "5 active material has been released to the environment, the .' I.] Ilkelihood of causing an early fatality or injury at a parti-

b..

cular distance from a nuclear plant' Analyses have shown 9./ that the early fatality risk, by far the more important health concern, is limited to a'small area near the power

1. [

plant. For example, 95% of the early fatality risk from a .v. ig :, nuclear power plant following an assumed accident at a high- [? population site was calculated to be within 4 miles of the plant, assuming people were exposed to high levels of radia-tion for 24 hours and took no special precautions to protect themselves. Here it was assumed that.70% of the reactor's ) radiolodine, the dominant contributor to early fatalities, j was released to the environment along with other radio-nuclides. b* Another characteristic of the early fatality risk is that it ,I is very sensitive to the amount of radioactive material (par-ticularly radiolodine) that is released during an accident. The smaller the amount of radioactive material released, the smaller and closer to the site the area over which 95% of the i., early fatality risk occurs, stated differently, the number L,' y of potential early fatalities rapidly decreases as well as ,{' the distance over which they might occur as the size of the g, radioactive release decreases. Smaller radioactive releases BI also reduce all other offsite economic and health conse-Oh quences. If only about 1 to 2% of the reactor's original 'Q. Inventory of radionuclides are released during an accident there will not be any early fatalities, even if no emergency ') responses are taken. (It is usually assuned that almost all l 12 b l..

~ 's .s 4 E P.?.' t ?. of the noble gases escape to the environment during a severe Ilc'. accident.) Many recent studies and experiments show that [I almost all potential accidents at nucl, ear power plants lead 5Y to very limited radioiodine and other radionuclide releases,

h far below the level required to produce an early fatality, f.Y,'

At the Three Mile Island nuclear accident the amount of radioactivity released was very small, far below that .~- required to reach the early fatality threshold. .s. How much radioactive material is released during an accident? i :, The answer to this question has been the subject of intense study, especially since the TMI accident. Although various experiments and computer program developments are still -7, underway to improve the state of the art of what is called source-teca technology, it is now possible to draw some preliminary conclusions: 1. Previous source-tera analyses overstated the amount of radioactivity released. 2. For those accident sequences where the radioactive 'l material from a melted-coactor core is kept within the containment for several hours, the eventual release is quite limited and well below that which would cause an early fatality. 4 The second conclusion above has additional inplications. It appears that the bulk of the calculated early fatality risk comes from core-melt-accident sequences where there is a short time between core melt and the release of radioactivity to the envir'onment. Even these care accidents may not be as 'i severe as previously thought. A number of experiments have been conducted recently that slaulated important aspects of one of these prompt release f' sequences and in which the amount of radioactive material 13

~ i 5. 2 ,$j released v.as considerably below the levels that have been f.N historically assigned to such accidents. It is anticipated Ihk that further experiments and analyses will demonstrate that D -even a number of prompt release scenarios would not result in f-[lf early fatalities. The calculated frequency of such prompt $Q releases varies from one plant design to another, but they j$ have been shown to be very low. In one recent probabilistic [J, risk assessment (PRA), the frequency of prompt releases was

• 'fi calculated to be less than one in a million per year of plant

{ operation. /* rg. ?? As stated previously, energency planning is also useful in reducing early radiation illnesses (early injuries). 7= in Oj the case of early fatalities, smaller source terms mean fewer early injuries. The accidents that are most important in 'early injury emergency planning are more frequent (though - y still care) but far less severe than those that dominate early fatality energency planning. Further, this early ) injury class of accidents would likely take longer to evolve, thereby providing more time to implement emergency responses. 1.1.2 Energency Responses Within the EPZ there are two major types of emergency responses to potential nuclear accidents: 1. Evacuation 2. Sheltering followed by relocation e* Evacuation, if laplemented promptly and carried out effi-3; ciently, can move people out of the zone where the risk is "ii the highest. 'High evacuation speeds are not required; a few " l; alles per hour (walking speed) is sufficient to eliminate the risk of an early fatality. The most important consideration 'N \\. 14 ___.,,_m.,y

S ~ ~ .. : r.. \\ s 9 is +n t

• n' ik..

I in evacuat. ion is timina. To be most effective, evacuation 9hf should take place prior to the release of radioactive mate-h.y rial. o. C: Evacuation also has certain drawbacks. If there is confusion f)' or delay, there is the possibility that some evacuees will "4 mistakenly traverse the radioactive plume, thereby increasing 'E' their exposure. Since the early fatality risk decreases ?,) rapidly with distance from the plant while the population to I'.) be evacuated can, in some cases, increase approximately as . }. the square of the evacuated distance, the complexity compared / with the benefits increase dramatically with the distance to f-which people are evacuated. Stated differently, evacuation near the plant is much more beneficial and cost effective than is evacuation a few miles from the plant. Sheltering followed by relocation also is an effective emer-gency response stra'egy 'Because' sheltering is far less coa-

~

t plex than evacuation, it is less costly to maintain and can be implemented by the public with a minimum of reliance on energency personnel and systems. There is also less likeli-hood for errors, such as inadvertent exposure to a radioac-v4 l tive plume. This is because sheltering involves simple, familiar actions with ample time to implement them, such as remaining in one's basement until the radioactive plume has .7 passed and, upon instructions, relocating to a safe area if a particular area is highly contaminated. This relocation should be easily achieved since the contaminated area (plume deposition area) within the EPZ source would be quite narrow the distance from the center of this contaminated area to its edge is likely to be under a mile. The disadvantage to shel- ~ tering is that it does entall some minimum radiation exposure for the public while sheltering and while relocating. Studies have shown that all-evacuation responses (evacua-tion out to the edge of a 10-alle EPZ) and all-sheltering 4 15

m -- -- ; - s .n. r / i ,:S 55 responses.are about equally effective. If evacuation com- ) G-j, d? mences promptly, then the all-evacuation response results in epl. lower overall early fatality risks than all-sheltering. If 'y. evacu' tion is somewhat delayed, an all-sheltering respon'se a k() could be more protective of the public. m v :'. . 'j h,;7 The best emergency response is a synthesis of these two basic responses utilizing the better features of each. Consider d the area ar.ound a nuclear power plant to be divided into ((", three concentric planning zones.* The innermost zone can be

.il a circle whose radius is the evacuation radius, i.e.,

this is k.f.{ the area where evacuation would be utilized. The middle zone ]'j is an annular ring ranging from the evacuation radius to the 3,. ; present 10-mile EPZ radius. Sheltering follo.wed by reloca-L*: tion would be utilized (where appropriate) in this annular I. ' ring. The outermost area, beyond 10 miles, would use discre-tionary measures as is the practice today. ) There are numerous advantages in this graded approach to al'nimizing the early fatality and radiation injury risk: l l 1 1. By limiting the size of the area to be evacuated, the o, full force of the emergency planning system can first be

(,

concentrated on the people in the highest risk area, [ While preserving the benefits of the simplicity of k. sheltering in the lower risk area. E. L, ', 2. Evacuees from the innermost zone are less likely to be r delayed on the roads because people in the middle annular (. ring would have taken shelter, i.e., stayed indoors and off the roads, y k'

• Precise geometric zones are not required.

Rather, zones should account for geographic features, political boundaries and the like as in present emergency planning practice. } b F l.' 16 I L i

s S ( 3. Only a small portion of the middle annular ring would be ji affected by the release of radioactivity. After identi-(9 fying th'e affected area, energency response forces can .'t. concentrate on relocating shelterees out of the area. n.; . j 4. The cost-benefit ratio for nuclear energency planning Ji would be auch more reasonable and more comparable to .y other societal emergency responses. a "N 5. The size of the evacuation radius can be adjusted as -[ source-term technology evolves. 6. Site-related refinements can be implemented to optimize the size and shape of the area to be. evac,uated. It may be valuable to determine the optimum evacuation radius on a site by site basis. At sites where there is a high population near the plant but with good sheltering capa-bility (many basements, brick apartment. houses) then a ~ smaller evacuation area alght result in the lowest over-all early fatality risk. Other sites, such as in a very sandy area, may have few people and few basements. Here the optinua evacuation radius could well be larger than that in the previous example. With regard to early injuries, sheltering and evacuation are also useful in reducing this health risk. For the very unlikely, more severe prompt release accidents, the emergency response dictated by minialzing the early fatality risk would be utilized. For the limited-release accidents that dominate the early injury risk, sheltering and/or evacuation can be used on a discretionary basis. A discretionary approach to this lesser health risk is recommended because of the likeli-hood of having more time to respond to this class of acci-Y dents, because of the smaller release of radioactivity, and because an emergency response framework would already be in t place to respond to the more demanding early fatality class 17

7.- .c ~., - s ,l ^ sii i nq yl of accidents. Discretionary actions are already part of pre- ) i i f?,t sent emergency planning. Because this approach is now I [,i assigned to areas beyond the present 10-mile EPZ where no 19 early fatalities are calculated from a regulatory compliance

j..

~ viewpoint, discre'tionary actions should not have to meet the {.[ rigorous requirements assigned to actions intended to reduce $) the early fatality risk. Y, The table below summarizes the principal characteristics of .t n various postulated nuclear accidents and their associated emergency responses for the early fatality and early injury risks. n 7: Table 1-1 Nuclear Accidents and Their Associated Responses Energency Response } Health Accident Inner Annular Beyond .s '.7 Effect Description Zone Region 10-Mile EPZ Early very unlikely Evacu-Sheltering Discretionary fatal-events, large ation followed by actions ity and prompt relocation release of where neces-radioactivity sary Early Unlikely event, + injury small and Discretionary delayed release actions j <'} of radioactivity i.~l C' *. Although present Energency Planning Zones extend to 10 miles [,;, and smaller EPZs could be justified, no recommendation to -)? reduce the size of the EPZ is given at this time. This very fl conservative size of the EPZ should be reviewed at a future 3, date, particularly after source-term technology has evolved further. l.yj 18 b

l .y y

e 9

Q C f. ( l.1.3 People and Their Institutions y; ','M A' successful response to an energency requires effective, '6 positive actions by numerous offsite organizations, both },; governmental and private. For most accident scenarios, the W. offsite responses are geared toward bringing offsite assis-tance to bear at the site (e.g., fire and rescue).

However,

(( for those ' rare scenarios involving a major radionuclide i release, the re'sponse becomes focused on protecting the .M people living near the plant. In such cases, appropriate I-response by the energency workers (including utility, state. and local government and volunteers), as well as the indi-l i. vidual residents themselves, becomes important. As stated earlier, to minimize the early fatality risk, tin- -l .ing is the most critical factor in carrying out a successful evacuation. By utilizing a graded approach to energency ( planning, a very,important step would'be taken toward making evacuations effective, if they are needed. ~ i Further steps can also be taken. In an accident at a nuclear power plant, it is possible that the focal point of "who is in charge" will be changing during the first few hours. If the public is first warned to take shelter by one authority and then directed to evacuate some time later by a higher authority, this could lead to confusion and, needless expo-l sure to radiation. This concern is particularly true during prompt releases where the time of radioactivity release would likely fall during the time the center of authority is chang-l ing. To minimize potential confusion, all officials who address I tho'public should issue consistent guidance, and decision-making under stressful conditions should be minialted. This calls for emergency plans which are rather prescriptive for ( prompt accidents where public pronouncements would have been

q 19

.___,,,,.,,y.._.,_,___,__,y

..k-L e 'Y' a:...~ lll a k.: 5,r. .(',j agreed upon. Simply stated, everyone should read from the .') 4 .w same scrip't for a given set of accident conditions. g, 4A u' UKj Since time is of the essence, the number of steps and people Q2 involved between recognition of a severe-accident condition hh and the sounding of the General Emergency alara should be $d minimized. The process for making and announcing decisions ' m. ny should be clear and simple. .. ~

• J, ip A key element in ensuring an adequate state of readiness for

'? an energency is the conducting of exercises and the training l5f of emergency response personnel. This would include training 9'. for and exercises of a graded energency response for serious,

N 7

prompt releases. However, most nuclear accidents would not 1ead to serious. releases. The challenge is to include the proper degree of realism and completeness to ensure readiness without overburdening the organizations involved. ) Current practice of including an evacuation scenario in every -i ,?.. exercise may lead to preconditioning of emergency workers'and (,. the public to expect evacuation in all actual emergencies, k For example, the ability to make a valid decision not to eva- ?- cuate during an emergency could be as important to emergency I-preparedness as the ability to evacuate. Drills should be h[ consistent with our state of knowledge about nuclear risks: most nuclear accidents would require no public action, some /, might utilize a sheltering response and, rarely, a full scale fl graded response. In this vein, Protective Action Guidelines L' should be reconsidered utilizing the new guidance provided by c. the International Conaission on Radiological Protection I cI,> E(, (ICRP), along with our present understanding of the nature of nuclear risks, for training, exercises, and actual accidents. $,'. G ?.ja, Public confidence is most important in assuring that the r L.. ['i.- individuals take the appropriate protective action (e.g., go l i to their basement when advised to shelter or depart *in a l e+ 20 l l L

X n~.a J. .3 s. Jf,' ..(y quick and orderly fashion when advised to evacuate). Although

l$

( ~ no one expects every person to follow instructions (i.e., r# some will refuse to leave when evacuation is called for and u Ly some will evacuate if told to shelter), the higher the level m. ~ly of public confidence, the greater will be the conformance to -[.} instructions, and the greater will be the degree of public f,' protection afforded. The building of such confidence begins [] long before any energency arises through the conduct of pub-n; lic awareness programs which introduce the people to what may be expected of them, whom to listen to, what to listen for. s f; In summary, keep the public well informed, keep offsite EF, actions by the public as straightforward as possible, agree i *; upon energency announcements so that the public receives con-sistent statements, simplify the number of steps between recognition of accident conditions and warning the public, l reexamine protective action guidelines, and practice drills that are consistent with the frequency and severity of the nuclear risks involved. 1.2 Analyses of a Graded Response Analyses have been made to determine the characteristics of a representative graded response to a large, prompt release of radioactivity. Results show that an evacuation to a distance of two miles, with sheltering (followed by relocation, if i l needed) from two to ten alles, results in a very low early j fatality risk. l In this analysis, a uniform population density of 100 people per square mile was selected. To adjust the risk numbers for 1 different population densities, the results can be scaled directly. Thus, for 500 people per square alle, the resul-tant risks are five times larger. Various refinements which might be investigated to determine the optinua emergency ( response strategy, such as a site-specific optimization as 21

a s u. s ? ~ .b' ~ ) .~ s l_ s n, s* ]c 5$$ 'y discussed,in Section 1.1.3 of this report, were not analyzed. ) ({} Tne principal input assumptions in these analyses are given %.1 in Table 1-2. N$ ne l'.: These analyses concentrate on determining the mean risk of an .I early fatality as a function of distanco from the point of ',N s;; release. Such a determination was done for two source terms. ,f( SST-1* and 1/3 SST-1 (with no reduction in the noble gases j' released) and results are plotted in Figures 1-2 and 1-3, $7 respectively. In these figures, we are comparing evacuations '.[J from zero to ten miles against sheltering from zero to ten w.

h. -

miles. Several conclusions can be drawn from these figures; m. k' 1. The "all-evacuation response (0 to 10 mil.es) can be -{ better or worse than the all-sheltering (0 to 10 miles) response depending on how long it takes to get people noving. ~ ) l 2. The "all-evacuation with a one-hour delay time" response and the "shelter for four hours, then relocate" response are rather similar.** Additionally, the increment in the .;f mean early fatality risk between a four-hour and an eight-

• l hour sheltering response is not very large.

i 3. At a distance of two miles, even for the larger SST-1 l source tera, almost all of the benefits of evacuation [.,' have already been achieved. At this distance and beyond sheltering results in very low early fatality risks. F-a s E

• See NUREG/CR-2239 for a definition of SST-1.

r5 m/u

• • Delay time is the time after public warning is given.

In 3; all cases, it was assumed that this warning came 0.5 hours \\' prior to the release of radioactive material. j'.*; 1 ) 4 22 I t

z. Ji :. m. s3 Table 1-2 9 (qi Assumptions Employed for Consequence Calculations C-g,'. 3 - 1. 1120 MW(e) PWR. . f. 2. New York City meteorology. Of 3. 10 miles per hour evacuation speeds. 4. Normal activities assumed before evacuation with the ,( '" following shielding factors: l' Cloudshine factor 0.75 ( Groundshine factor 0.33 Inhalation factor 1.00 5. Shielding factors for sheltering: Cloudshine factor 0.50 Groundshine factor 0.08 Inhalation factor 0.50 6. 100 people / square mile. ,4 O e 'J P4 n

1. 6 t

23 4

.v

1., -

s vs .,, si 4;'.\\.. Of 01 ) i.i i i.i.i. SST 1 7.W [yj, 'g 1120 MWe PWR j a.; \\ NYC WEATHER

n

.\\- /., -*- EVACUATION,5-hr DELAY ( f.,, \\" EVACUATION,3 hr DELAY

/. ?

-=- SHELTER 8 hr RELOCATE --- SHELTER 4 hr, RELOCATE y .. + g EVACUATION,1 hr DELAY /" -a- ~ W p \\'\\ n I 0.01 ., \\ \\ g y. \\\\ 7,. 's\\ = u w ~. g , y '\\ + 4 g\\ g \\' O H 5 \\ a \\\\ \\ =\\ } <m O \\ l c 0.001 t 1 1 2 t \\. w E \\ .\\ l:: fi. \\ 4 I I 0.0001 L '- 0 1 2 3 4 5 6 7 l: c. P< DI. STANCE (ml) ll, L..i c.r. c. [ ?, Figure 1-2. Mean Probability of Early Fatality vs Distance (SST-1) l l :- L-I l 24

.~ s ~.. + ,0;. .,'n 0.1 g: ( 3 i,i i e i i i .J,9 1/3 SST.1 1120 MWe PWR t d), NYC WEATHER ,k f. ~ ~ -.- EVACUATION,5.hr DELAY a 11 EVACUATION,3.hr DELAY j\\\\

.g 3

-=- SHELTER 8 hr, RELOCATE = I\\ b l --- SHELTER 4 hr. RELOCATE a }., -.* = EVACUATION,1.hr DELAY -f. g.- 0.01 II .11 1 s t= e L 2 3} u. o q$ 1I b \\ \\ \\. em 1 .iig\\ a O + = I t a: 0.001 \\ a.

\\ =k z

5

\\

I j =\\ lE ( i \\ j .t\\ i f ~: i ; \\. i l I, I ! L I i t f 0.0001 O 1 2 3 4 5 6 7 ' ~n ?.'. DISTANCE (ml) Figure 1-3. Mean Probability of Early Fatality vs Distance t (1/3 SST-1) 25

+ .- ~ .w.

. +

.A r; e (T.' 14 / 'h .Uf A smaller source term (1/3 SST-1) further reduces the mean I* p'. ; early fatality risk and the distance over which it might ,1 4.. lll.1 occur. F7. ,[ Figures 1-2 and 1-3 Plot the mean risk'of an early fatality fij. versus distance assuming that a severe release has occurred. Since various studies indicate that such a release is very unlikely, in the range of once per 100,000 years of plant l 's operation down to about once per million years of operation 4.' (or even less likely), one can get an appreciation of how ib small these mean risk values are by multiplying them by their calculated accident frequency. Table 1-3 shows that at two miles, with an assumed accident frequency of once in 100,000 years, the mean risk of an early fatality per 100 people per square alle is: Table l'-3 } Two-Mile Mean Early Fatality Risk Energency Response SST-1 1/3 SST-1 Evacuation with 1-hour delay 0.0003 + < 0.0001 + 100,000 years 100,000 years Ehelter for 4 hours / relocate 0.0010 t < 0.0001 + j., 100,000 years 100,000 years L.,- Figures 1-2 and 1-3 were for unifora emergency responses, lN i.e., an all-evacuation or all-sheltering response from 0 to ki 10 miles. Figures 1-4 and 1-5 depict a graded response for L'- source terms SST-1 and 1/3 SST-1, respectively. In these 5i latter figures, the area to be evacuated was examined utiliz- ) ing evacuation distances of 1, 2, and 3 alles with sheltering for 4 hours in the remaining area out to the edge of the 'g [..J-l 26 x.

J ~6' g 3 g . (. s. 1;., s 's. 1 1 e i l = ?, < \\ ss l , ~.. s

r t

I y

• 4 a.p,..

a. 1 j-1.0,, .C $$f.1 t.mi EVACUATION. SMELTER SEYONO

• ~ '

~ = 1120 MWe PWR 2 W EVACUAfl0N. SMRTER SEYONO =

= =

NYC wtATHER ..' 'e$ 3 me EVACUATION. SMELTER SEYONO 100 PEOPLE /mes ...= M. 1/3 8871.1.m4 EVACUATION. SMELTER BEYONO -

==

i.;. .' I ~ \\ \\ t 4 W 0.1 ~ Al m' ).. M g I: R.6 g g N g g n gs% O.01 u ____ ~= ~ 9 I 'I 0.001 1 10 100 1000 X. EARLY FATALITIES r S I 5 Figure 1-4. Graded Response for Source Term SST-1 i '.f 27 i 2

3; e o i t b. ' \\'s

4s;

• i 5

~** -.%;f '"[,4' N..? 1.0 ...g ..,,g ...1 1/3 8871 (EXCEPT NOSLE GASES) 11 3 1130 MWe PWR M NYC WEATHER i a; 100 PEOPLE /wa t. ~ [(.,.'. M 0.1 N, .f = g g 8 E f. '). 1 me EVACUATION SHELTER BEYONO - 0.01 U

s.. '

b I E I E .b I i i 1 1 1 t a 1 i i I I E I I 1 10 100 1000 X. EARLY FATALITIES .c i w Figure 1-5. Graded Response for Source Tara 1/3 SST-1 \\ 6 28 ?

s s. L :, a, c ~ m ^<.'e .~ ut kd I.) ( 10-mile EPZ. In all cases, it was assumed that evacuation c 27 would initiate with only a 1-hour delay. k3 f.f Figure 1-3 shows that a graded response is a very effective ,' g e'mergency response. For example, with a release frequency of once per 100,000 years of operation, there is but one chance ,Q 'O in ten million years of exceeding 21 early fatalities

• if

[ evacuation is out to two miles with 4-hour sheltering beyond I. to 10 miles. .y ,j 'y Figure 1-5 demonstrates the enormous sensitivity of the early L' fatality risk to reduced source terms. The dashed lino on f Figure 1-4 is a replot of the data from Figure 1-5. With I such a source term, a one mile evacuation dis,tance results in / very low early, fatality risks. With the 1/3 SST source tera, there is one chance in tan million operating years of exceed-ing 2 early fatalitie's* if evacuation is limited to one mile, with'4 hours sheltering beyond to ten miles. With evacuation out to 2 miles, no early fatalities are calculated. With regard to source terms for large, prompt release sco-nacios, the expected conclusion from various experiaantal and [ analytical programs now or soon to be completed is that the magnitude of radionuclide releases would be a factor of ten or more smaller than the releases used in current analyses, such as SST-1. Based on the above analyses, and even utilizing a SST-1 [ release, a graded response utilizing a two-mile evacuation distance results in a very low early fatality risk and makes rapid evacuations more likely. Because of these principal conclusions and the realization that even these low-risk fig-E- ures are quite conservative, the two-mile graded response is a -? I'

• At 100 people / square mile.

s 29 i

w..-.

.m _a [ i- . *R .:A recommended. As source-term technology evolves, the evacua-I .s .:~ 'S tion distance should be recalculated to reflect this knowl- ,h edge. .V ( >y 1.3 Descriotion of the Esercency Preparedness Subfunctiong o % +.' [O, The Emergency Preparedness functf.on has been divided into

(

four major subfunctions: 5)} [".U, 1. Plan for Emergencies 2. Maintain Readiness } 3. Implement the Plac During an Energency 'J*/ 4. Recover from the Energency ? Each of these can be further divided into subfunctional topics. These subfunctions and topics enable the emergency l activities to take place in an organized, re' liable, and l I effective fashion. Since the purpose of this effort is the 't/' identification of issues, information needs, and RD&D needs, i ~ this particular definition of the subfunctions and topics was heavily influenced by the desire to provide an effective framework for understanding those issues and needs. For T example, we occasionally struggled over whether a particular issue belonged under planning or implementation (or both). The approach was to place it in the subfunction where it fits o best and not to spend time redefining the subfunctions and topics to make it fit better. What was important was that i the issue or need was successfully identified and included in the evaluation process. The working group efforts involved a discovery process, where definition of the subfunctions was 2? the first step. To facilitate understanding by the reader, i I each subfunction has been cross referenced to its identifier f,y used in Table 4-1, and the subfunctions ultimately found to ~ be related to high-priority needs have been flagged with an asterisk (*). J } 30

^ s.. .~.~,- .a..... ~ ,l .. -~

N1

$3

y. :.-

Ali i-1.3.1 Plan for Energencies (4.1) q,5 .c

• h-The planning function is where the questions of who, what,

yj (by) when, where, and how are addressed.

The maximum degree j j,5 of decisionmaking and prope. ration should be done during the i f[ planning phase since there is ample time to think things I/ through, perform analyses, conduct studies, etc. For our ?,) purposes, the planning subfunction was divided into the fol- !.IM loving three topics: i.) [: U..; 1. Establish Planning Bounds (4.1.1)* y;. y,*. Since no emergency plan can possibly account for every sce- [4 nacio, contingency and action, planning bounds must be set { with an eye towards maxialzing the value of planning effort { e. and resources spent. Thus, limitations must exist on the range of accidents and scenarios, the distances.from the e site, and the degree of pectaction for which to plan.

C

^ 2. Establish Logistics and Conaand, Control and Conaunica-tions capabilities (4.1.2) The institutional and organizational details aust be worked out such that everyone with a job to do knows what to do; 7 that everyone who is asked to relocate has a place to go: that everyone involved knows who is in charge. This topic includes a range of subjects, from providing food to the I-F onsite energency response team to negotiating a smooth tran-sition of the energency command among the utility, local gov-

c ernments, state government, etc.

Also included is establish-I ' C' ing agreements with the nonutility energency workers (e.g., C ', fire, police, hospitals) and setting up interagency inter-I .e faces. t

• 5ubfunctions related to high-priority issues and information needs (i.e., priority 1 or 2).

'a 31 1 _.-._..-_,,...__..~_m

---

.,_.,_c-,. - - _ _ _. _ _ _ _.,

'L.

.. u..:.

> j;.. t.(

7. '.I-3.

Estsblish Implementation capabilities / Procedures / Decision .) L.!k. / .,f'y Criterla (4.1.3)* 2 This topic addresses the provision of hardware and'softvare g.i'3 required to carry out the energency response. The creation

• [.

of implementing procedures and the selection of emergency (!! response systems reside here. The Energency Action Levels 2' " and the associated energency classification systen and the protective action decision criteria are developed. A subtle b" activity associated with this topic is deciding how much [ "planning" to leave until an energency has begun versus plan-I ning for everything beforehand. ["- 0-1.3.2 Maintain Readiness (4.2) Maintenance of energency preparedness is the subfunction which ensures that a state of readiness is present at all times. The malatenance subfunction is also responsible for I foste ing a smooth transition from normal to energency condi-tions and public knowledge of and confidence in appropriate emergency responses. Readiness applies to energency workers. the general public, and the energency facilities and equip-ment and is divided accordingly: l. Conduct Exercises. Drills, and Training (4.2.1) This: topic addresses the maintenance and testing of the skim s and effectiveness of the energency workers. 2. Maintain Public Readiness / Awareness (4.2.2)* ll' This topic addresses public information programs for nearby ii residants to ensure that they are aware of the appropriate responses required of them during an emergency. Fostering confidence in the emergency plan by the public is important l 32

4 _\\ ~.;,w c. el i . :/

1

1. '1 v4 av i

S.* sf, in assuring that the plan will be followed in a positive. ( {.N' cala, and ' effective fashion. ?'/.'. V-3. Maintain Facility Readiness (4.2.3) eg .O This topic addresses the means for keeping the emergency [I, response systems, facilities, and equipment ready for an energency. Examples include equipment checklists, periodic [f testing procedures, and monitor calibrations and setpoint "b adjustments. F.i n.f 1.3.3 Implement Plan During Energencies (4.3) t.'; The purpose of the implementation subfunction is to carry out the plan during an actual energency and includes the collec-tion and processing of information to make assessments and predictions. decisionmaking, and issuing of emergency response. and.public protection action recommendations. Implementation also must address monitoring of response effectiveness such that either the actions taken are validated or the response is reconfigured to be more effective. Since time and effee-tiveness,are of the essence, this is where the planning and ( maintenance subfunctions pay off. These concepts have been 6 sorted into five topical areas: [. 1. Evaluate Hazard / Risk (4.3.1)* 5.. As input to emergency response decisionaaking (below). the ) energency manager must obtain an assessment of the current state of the emergency as well as an estimate of the progno-C[ sis (e.g., is the situation getting better or worse?). Thus. information and data regarding the plant conditions, meteor-ology and other factors must be collected and processed in .[.- order to draw conclusions and make predictions. w> 33 m

~..... QY d*d [,'; 2. Implement Decision Model (4.3.2)* ) 4). 21 Once the current conditions and predictions start to coll in, a.- du the emergency managers must then decide upon the appropriate r.<6 7 responses. The planning phase could have considered the kind $[f of scenario which is at hand, and thus many (if not all) of {$ these decisions will have already been prescribed. If not, then his training and experience will be called upon to der-ive the appropriate response. The factore used in decision-making include plant conditions (observables) and trends, 'i weather conditions and forecasts, calculated predictions of T.." accident prognosis and of offsite doses, and the probable 'E outcomes of various protective action alternatives. 3. Implement onsite Actions (4.3.3) This topic refers to carrying out the onsite energency actions appropriate to the current level of urgency. Such actions include notification of utility and governmental personnel, j accountability of plant personnel, first aid and so on. l., 4. Implement Offsite Actions (4.3.4)* [- l ll This topic refers to carrying out the appropriate offsite energency actions. It includes protective action reconnenda-i tions issued to the general public as well as the activities {1 [ carried out by the nonplant emergency workers. l \\.' 5. Verify Effectiveness and Respond Accordingly (4.3.5) i. The effectiveness of the emergency action decisions and their implementation must be monitored to ensure that the expected i, results are, in fact, taking place. If not, the decision (s) r must be reconsidered and possibly differing actions pre-scribed. I 1 34 l ,=,.. - -.

c 3,. -4 .. ~,. .i

o. ;

f;,$ t' 1.3.4 Recover from Energency (4.4) $/ Once the emergency situation has subsided and plant conditions Qm have returned to normal (or at least are stable), then deci-s. x.. ry sions have to be made as to whether any efforts are required fjj; to protect the public in the longer term. These recovery

Jg actions might include decontamination of land and property or even a decision to interdict them.

Recovery is divided into b}:p two topical areas: I'. e'*.,- 1. Establish Recovery Plan (4.4.1) The recovery subfunction is not planned out to the same degree

r, as the implementation subfunction since the exact effect of f!

an energency is highly variable such that any recovery plan will be "accident-specific." Thus, a minimum recovery plan-ning effort is recommended under this subfunction, with the balance to be, performed after the energency has occurred. ?( The detailed decontamination and reoccupation action would be developed at that time. 2. Make Decisions and Implement (4.4.2) hv Once the plans have been laid out above, they would then be carried out here. s 4 ~; f;. N;l. .O o 'i s 35 W ~

.s. h'N .:W ').l /,[,' 2. IDENTIFICATION OF ISSUES AND NEEDS IN e

7 EMERGENCY PREPAREDNESS OY 1"*

?? This section of the report presents a detailed description of @h identifies the information needed to resolve those issues, [ the issues surrounding the emergency preparedness function, .it[ and establishes their priorities. This information was .L-developed by the Working Group and uses the entries of Table 4-1 under the headings of "Issue" and "Information l3 Needs." In order to aid the reader in gaining the proper N perspective on these various issues and information needs. J,{ items which were ultimately assigned a priori,ty of 1 or 2 are j flagged with an asterisk (*) at the end of the heading. The issue identifier used in Table 4-1 is also indicated in each heading. 2.1 Discussion of Issues and Information Needs ) 2.1.1 Public Awareness Programs (4.0.A) There is one issue which cuts across all subfunctional bounds, and that is the issue of public awareness. This relates at a higher level than the awareness of those living in the vicinity of the plant (see item 4.2.2.A. which addresses the response preparedness of residents who could L become involved in an energency response action).

Rather, the issue addresses the arena of public understanding of

,y energency preparedness, their confidence in the nuclear industry, and their receptivity to changes in the ways of Q. handling the emergency preparedness function. In other [,, words, if public confidence is low, improvements in the ener-l.' gency preparedness function may be more difficult to achieve. Tho' associated information need then becomes a review of cut- ] rent public awareness programs to assess their effectiveness and, it improvements are found in order, then to seek out -I those improvements. 36

e.v.L.. . 'o' Ti %f f,p.(d( 2.1.2 Range of A'ccidents and Planning Distances (4.1.1.A)* (if . l.4 Current emergency planning distances (radii) are generally .gj set at 10 miles for acute (plume exposure) pathways and 50 }'. miles for chronic (ingestion exposure) pathways. These dis-PU tances are adjusted in individual plans to account for topo- .:n ('pj logical or other factors where the prescribed circle does not .I')) apply. Short-tera public protective actions, such as evacua-tion and sheltering, are planned within the 10-mile zone. j. Longer-tera dose mitigation actions, such as crop and milk $N condemnation, are contemplated within the 50-mile zone. At k..I issue is whether the criteria for selecting these planning .. ~ distances are appropriate in light of lessons learned from e s i '.. PRAs and the likelihood of demonstrating subs,tantially lower source terms. There may be risk-significant scenarios which are not clearly dealt with, and, conversely, there are plan-ning efforts expended on scenarios which pose an extremely '1 small risk to the public. This issue is further underscored by the lack of a consistent evaluation and planning basis among the many other risks faced by the public. It therefore seems possible that the total "budget" of energency preparedness dollars is not being wisely allocated among the total spectrum of man-made hazards and natural disasters for which planning may afford a reduc-tion of risk. .e-This issue spawned two information needs: 1. An extensive assessment of the current emergency prepared-( ness regulatory requirements and practices is needed in { order to evaluate the impact of (a) insights and knowl-l, . edge gained from recently completed PRAs and (b) reduced radiological release source-term estimates. Should this 37

o 'I. [ assessment reveal areas which should be relaxed, strength- '1 h,{ ened,'or altered in some other way, then corrective modi- / fications of the regulation and/or practice should be laplemented.

g N

2. In the search for an understanding of how nuclear plant gg 's energency preparedness compares with that of other 'a hazards, a comprehensive examination of the energency preparedness planning bases as well as the efforts expended and/or required for these other hazards and emergencies should be conducted. Ways of improving the integrated energency preparedness effort to encompass all hazards in proper perspective should be found and imple-mented. 2.1.3 Plan Coordination (4.1.2.A) The emergency planning zones around a given nuclear plant j intersect numerous political, governmental, and social bound-artes. A typical "plume exposure pathways" planning rone may involve several county and city governments and at least one state government, as well as numerous other entities, such as volunteer fire and rescue squads' and the Red Cross. Within each government, several agencies or departments must be interfaced, such as civil defense, law enforcement, or envi-conmental regulation. In most cases, the various organiza-tions exhibit a great deal of cooperation among each other and with the plant owners. However, there are cases where this coordination was interrupted by a noncooperative entity. The issue at hand is the degree of coordination among organi-rat,lons required to maximize overall energency planning effec-tiveness. The associated information need is an evaluation of current requiations and practices to understand (a) the importance of I coordination in minimizing public risk and (b) how this issue 38

__4 ~ ' ' c.., 4,. ' ,9: . 3' Q u~ 'i? f ( is addressed in nonnuclear arenas. The study should also f:j)[ identify any needed changes brought to light during its I? course as well as document effective strategies for resolving ['[j differences among authorities. (91 j 2.1.4 Acconnodating Groups with Special Needs (4.1.2.3) 1{., A.[ Care must be taken to ensure that energency planning reason-X..; ably considers special population groups, such as children in ,..(j school and those who are handicapped or hospitalized. The h; current methods of acconnodating groups with special needs do ,1,'j not have the benefit of quantitative analyses of the various

l protection strategies available.

Thus, the associated infor- }' mation need is to conduct such a study and to,use it to gen-l erate a base of information and data for developing optimum y strategies for protection of special population groups. 4 .l ( 2.1.5 Establishing the, Chain of' Command (4.1.2.C) l l, During the course of a severe nuclear plant accident, the 1,. command of the energency response will change several times. In addition, there is the potential for more than one person l at a given time to believe that they are in command. This rather delicate area is addressed in current energency plans f and generally works well in exercises. However, for an actual i.' - energency, a change of conaand is likely to take place during [ a crucial point in the accident sequence. For example, con-h,, ' sidering the accident sequences identified in a typical PRA. l' the turnover of command could take place during a release for L.a L* those sequences posing the highest public risk. Thus, the j's ! associated information need is a determination of the rela-ll tionship and importance of the timing of-severe-accident sequences and the timing of change-of-command actions. O 4 4 e 39 l' t

.m r i M ~s (- 2.1.6 Ese,rgency Action Level Decision Criteria (4.1.3.A)* ,,) r. k) The current methods and criteria for establishing the proper ' f' emergency action level (EAL) evolved prior to the completion k of recent PRAs and degraded-core-accident studies. Thus, the 3 adequacy of the current energency classification scheme could be questioned as follows: Are there accident scenarios which would result in a classification that is either too uovere [ (perhaps resulting in unnecessary public res'ponses) or not / severe enough (perhaps resulting in inadequate public .5 response)? The information need associated with this issue is to conduct an evaluation of the present emergency classi-6 fication criteria in light of the knowledge gained in recent FRAs and degraded-core-accident analyses. 2.1.7 Public Protective Action Decision Criteria (4.1.3.B)* The current methods and decision criteria for recommending ) offsite protective actions (e.g., evacuation and sheltering) evolved prior to the completion of recent PRAs and degraded-core-accident studies. In addition, these criteria have never been subjected to a quantitative risk / benefit analysis. Also of special interest is the question of how operator actions to terminate an accident prior to a radiological release should be factored into the protective action decision criteria. These concerns generated the following information needs: I 1. The current Protection Action Guidelines (PAGs) Proposed [ by the Environmental Protection Agency are 1 to 5 rem for whole body exposures and 5 to 25 ren for thyroid expo-sures. These values are well below doses required for production of early health effects, and are in the range of radiation exposures routinely received by many members of our society from occupational and medical sources. 40

~,j,,. t 4.\\ Since protective actions (such as evacuation and shelter-d[}.' ( ing) a're expected should projected doses exceed the PAGs, t further evaluation of the appropriateness of the current (i, PAGs is necessary taking into account the recent ICRP 'h ' recommendations and also PRA risk analyses. ,e '[. g 2. An evaluation should be conducted to determine the reli-j P' ability (likelihood) and effectiveness (completeness) of .e Fr. mitigative operator actions and how such actions should

• ( '

be factored into the protective action decisionmaking process. 3. In order to assess the adequacy of existing irotective action decision criteria, one must also understand the decisionmaking inputs needed. Thus, a determination l .( should be made of the plant and site information needed l by the energency manager to make appropriate energency '2 response decisions. i I-e- r jl 2.1.8 Unifled Decision Model (4.1.3.C)* In light of the stress, the possible confusion, and'the fast -[ timing required during an emergency, it is desirable to make as many protective action decisisns as possible during the planning phase. This would reduce the protective action pro-cess to translating observables into the predetermined protec- [' tive action decisions and then carrying out those actions. This approach is well developed for use in the classification of emergencies, but has been applied to protective actions in j a fairly lialted fashion. The decision criteria and assump-d tions for making protective action decisions often vary among 4 y the organizations responsible for implementing an energency [ 7-response plan during the course of an accident. There are cases where officials within the same state government arrived at and voiced differing decisions, resulting in confusion and public alarm. Thus, the issue le the lack of and need for a j i l 41 l

'e unified (1nteragency and utility) decision model which would ) be developed during the planning phase and merely implemented ,f during the emergency phase. The information needs can then be stated in terms of a feasibility assessment for' developing l such a unified protective hetion decision model on a generic lk basis for application to individual emergency plans. The dk form of the model could also be addressed (e.g., a computer-ized decision aid or a matrix which translates observables into actions). The feasibility study to address this issue g-might include the following elements: 1. 1. Determine whether an adequate spectrum of contingencies can b's accommodated by a comprehensive decision model L. which makes maximum use of predetermined, protective actions as a function of observable conditions in and around the plant. 2. Assess whether there would be enough confidence in the '} nodel so that its outputs would be used during an emer-gency. 2.1.9 Exercises / Training (4.2.1.A) A key element of ensuring an adequate state of readiness for an emergency is the conduct of exercises and training of energency response personnel. The challenge is to include the proper degree of realism and completeness to ensure readiness without overburdening the organizations involved. To date, exercises have not included extensive participation by the general public (and there is no requirement to do so). However, there could be elements of public participation having a useful role, especially if protective actions evolve towards sheltering as a baseline protection strategy.

Thus, an evaluation of public involvement may be a valid information need.

42

-6 D. 1.* t 2.1.10 Public Awareness of Their Involvement in an Energency g Response (4.2.2.A)* A d Public awareness of and confidence in an emergency' plan are essential ingredients to successful public response during an fj,f energency. The public must know what to do, when to do it. (,; and whom to listen to and believe. Thus, the issue at hand

~

is how well current public awareness programs are filling ^ these needs. Two specific information needs were identified: c 1. An evaluation of the methods and activities for maintain-ing public awareness around nuclear facilities could serve to determine those methods and activities which are ,I most effective. Such an approach would sorve as a clear-ing house for positive methods and activities as well as 2 a way to distribute any lessons learned to the energency planning community. 2. The issue of public awareness exists for nonnuclear emer-gencies and, as such, a body of useful information should exist. An evaluation of the public awareness programs ~ for other hazards would serve to identify effective methods, activities, and lessons learned that might bene-fit radiological emergency planning. q, 2.1.11 Facility Readiness (4.2.3.A) Having energency facilities in a continuous state of readiness is necessary for a successful response to an energency. Ensuring the adequacy of the means of maintaining such a readiness thus becomes an issue. The information need is an '/ integrated evaluation of exercises and actual energencies with feedback to the energency planning community. 43 sW

~. = 2.1.12 Dose Projection capability (4.3.1.A)* ) During the course of a radiological emergency, dose projec. ?'" tions provide input used to make protective action ~ decisions ], for less severe accidents where discretionary procedures

• Dose projections, in turn, rely upon knowl-would be used.

-} edge of current and forecasted weather conditions and upon ..] .[ estimates and measurements of radioactivity in the field. ~ The more reliable the projections, the more confidence that ,~' can be placed in the protective action decision. Thus. meteorological prognosis and dose projection evolved as an issue having five information needs: 6 1. The importance of local topology and asso,ciated recircu-lation patterns to plume transport and dose projections should be evaluated to improve current understanding. The outcome would either validate current models or lead to improved modeling techniques for incorporation. ~ 2. The U.S. military, especially the Army, has accumulated a i great deal of information on the behavior of radioactive (or otherwise contaminated) plumes. A review for appli-cable information should be made. 3. The tradeoifs between using dose projections (which can [. be obtained quickly and prior to an actual release) and field, measurements (which may be more accurate but cannot I be taken until after a plume is present) to make protec-tive action decisions are complex. Thus, the information need is stated as an evaluation of the relative serits and optimal mix of these two methods for discretionary situations. 4. Assuming field measurements continue to be an element of protective action decisionmaking, an assessment of the need to improve the capability and reliability of 44 l I

+. ,i ..y ~, e ij ( portable monitors and an assessment of the cost-benefit f.! of usiag field monitors to determine radionuclide [.7,) aixtures are needed. If improvements'are found te be k. needed, then such improvements should be developed. f({ 5. Another method of providing field asasurements is the ); installation of fixed monitors surrounding the site. Aa !,{ evaluation of the value and effectiveness of this approach 9 is needed. .y. (( 2.1.13 Adequacy of Accident Progression Prognosis Capability 2, (4.3.1.5)* During the course of an emergency, estimates,of the prognosis of the accident progression are used to make protective action decisions. The introduction of decisionmaking based on prognosis affords more time to make decisions and a higher degree of confidence in them. However, the methods for determining the accident prognosis, including likelihood of termination pr'ior to a release, are still evolving (as dis-cussed in the reports on the other three functions). The development of advanced computer systems (referred to as "expert systems") and methodologies for providing more accur-ate prognosis estimates are being pursued under these other programs and by the industry. However, an information need still remains here in determining how to integrate these laproved prognosis models into the emergency preparedness 4 ,1 function. r. [; 2.1.14 Unified Prognosis Approach (4.3.1.C) 1^ A number of organizations get involved in the protective i

Y action decisionsaking process. including arriving at dose s

l, ; projections and accident prognoses. There are times when j these organizations use differing models or assumptions in I l I, deriving their predictions. Since such differences can l t l 45

introduce confusion (not to mention loss of public confidence j ~ when repocted in the media), an effort to reconcile and rationalize these differences in models, assumptions, and g, . 'c calculational techniques would seen appropriate. Such an undertaking could also assess the feasibility of creating a fi standardized approach to be used by all organizations., u j: 2.1.15 Systematized Decision Model (4.3.2.A)* i If the trend addressed in paragraph 2.1.8 (issue 4.1.3.C) is realized, then a follow-up issue is whether a systematized decision model is necessary to implement the unified decision approach. Thus, the information need is an assessment of the feasibility of and need for developing a syst,en for implementing the unified decision model (or matrix) discussed under issue 4.1.3.C. 2.1.16 Protective Action Effectiveness Prediction Capability ~} (4.3.2.5),* Methods for quickly calculating the predicted effectiveness of protective action decision alternatives (in teras of, say, dose reduction) are not currently in place. Thus, accurate projected outcomes of those decisions are not available as an l' input to making the decision. The information need is a determination of the feasibility of and the methods for cal-o ).' culating the dose reduction associated with alternative pro-l h., tactive actions. Such methods should be capable of addressing finely defined population groups and contingencies and should provide results in a real time or interactive time frame to ![ enhance decisionmaking. i.

1..

i l l i i 46

4 i; .o 2.1.17 onsite Actions (4.3.3.A) s-I ([;Q The nuclear plant staff is required to take a number of k nonplant-related energency response actions during an emer-dS ^ 'gency. Examples include notification of offsite authorities, jf general, their preparedness for carrying such actions is con-communications, and accountability of plant personnel. In ' c, firmed by exercises, drills, and equipment check lists. How-3 '~-l ever, these tests are anticipated whereas an actual energency h;h is not. Thu's, as an information need, an assessment and f.g feedback / distribution of results of actual energencies would provide useful insights and a crosscheck on onsite operator f." action. T 2.1.18 Public Response to Instructions (4.3.'4.A)* e ,f A protective action decision must be followed by the public in order for it to be truly effective. Accordingly, human I ^ behavior during energencies must be understood to the maxlaus ~ extent possible. It is well known that the percentage of people following instructions is closely linked to their confidence in the decisionmaker. Two primary information needs were identified: .~ 1. In light of an increasing interest in recommending shel-tering (followed by colocation of persons in contaminated areas) as a protective action alternative, an assessment ], of public response to such instructions in the possible l presence of perceived high-exposure and/or radionuclide i-concentrations is advisable. 2. Rumors are a constant source of confusion and insecurity .during an energency. An integrated evaluation of rumor I control methods and procedures. including identification I' - of changes it warranted, would be beneficial, j i 47 i

2.1.19 Coordination of Communications Among Various organi- ) ~~ sa'tions (4.3.4.5) e ~ ll Precise and reliable communications among the various ener-l '. gency response organizations are Laportant during an amer-gency. However, effective communications can be stifled due to hardware incompatabilities or procedural differences. As such, an information need was identified to develop a generic 3

.0 or standardized guidance for promoting effective communica-l' tions.

For example standard radio frequency assignments as I. well as many standard message texts could be developed. 2.1.20 Emotional Conflict-of-Interest for Emergency Workers (4.3.4.C) A large scale emergency response involving offsite protective actions (especially evacuation) requires the involvement of offsite emergency workers, such as police, rescue workers. and bus drivers. However, such workers may feel an emotional conflict (regarding themselves or their families) when called l upon to perform their duties during a high-public risk sco-nacio. Therefore, an information need was identified which would call for an assessment of worker response to directions during high-risk scenarios. 2.1.21 Fee'dback Provisions (4.3.5.A) L Verification that energency response activities are being properly carried out and that they are creating the expected results is necessary to ensure protective action effective-ness. Verification requires feedback from the field as well j as from onsite in order to check that the offsite responses are in concert with current (or predicted) accident condi-tions and to adjust to changes. The specific information i need is an assessment of the feasibility of improving ener-gency response effectiveness through the use of real-time I 48

i u. 'a. feedback loops to measure the effectiveness of energency s response d'ecisions to keep the emergency response activities ,p; [;[ linked to actual (or evolving) accident and meteorological conditions (versus predicted).

3. :

.5, 2.1.22 Recovery Criteria (4.4.1.A) l' (': Recovery and return to acceptable conditions is a phase of energency planning that is left primarily to be performed on N a discretionary basis after the energency. Some broadly 4 defined criteria are in place for determining when decontani-nation is called for, or at what dose levels land can be j reoccupied or should be interdicted; but the specifies would not be developed until accident-specific data is in hand. Generic information needs to help clarify and expedite the development of the specific recovery criteria are as follows: 1. A review of the current criteria in light of recently ( completed PRAs and the prospect of significantly lower source terms. The objective would be to identify and develop any appropriate improvements. 2. A quantitative evaluation of the risks and social costs of residual contamination would provide a bank of infor-nation usable to expedite the postaccident recovery cri-teria development process. 3. Same as 2, above, but for the risks, costs, and benefits of decontamination efforts. 4. A quantitative evaluation of 2 and 3, above, should then be conducted in order to find the optimal relationship .(or provide necessary data to find it for a specific application) among the reoccupation criteria, decontaal-l' nation criteria, and interdiction criteria. This assess-ment should also explicitly address key socio-political 49

t l, issues including impact on nearby critical industries and l h water ' supplies and the economic impact of recovery delays

• ?

on the u'tility and its rate payers.

b..;

f. 2.1.23 Role of Public Perception (4.4.2.A) Public perception of the hazard of reoccupation (and thus N. their responsiveness) might be a key factor in determining the overall economic consequence of a major reactor accident. }' These perceptions could, in turn, affect decisionmakers. The ? information need_that results from the issue of public per-r f-ceptions is an evaluation of long-term mitigative actions taken in other industries. The study should go on to ident- ,h ify how other industry "lessons-learned" could best be applied to nuclear plant energency planning. 2.2 Prioritization of Issues and Needs [ .f. ) The DOE-sponsored Industry Integrati^on Couitaittee developed a methodology for use by working groups in prioritizing issues and information needs. Using this methodology, issues and f information needs are ranked in a uniform way against their l a. i ' importance to safe, reliable, and economical production of P. electric power rather than being ranked against each other. i Using this method, each information n'eed is evaluated against i.. the same four questions, each with a prescribed set of answers. Taken together the four question responses j determine the priority ranking of each information need. Table 2-1 is the key table used to convert the question responses to a priority number ranging from priority 1 I a (greatest priority) to priority 6 (least priority). f i-l- a 50

t

?.

v ( The question, responses, and response codes are the following: >j ouestion 1: iS.. c (0 Will resolution of the issue or satisfaction of the informa- .;} tion need result in a cost-beneficial, significant change '. ~ (either physical or administrative) in plant operation, i.e., will it significantly change plant hardware, administrative or operating procedures, personnel dose, or personnel train-fj". ing? } Probably Yes (Y) Perhaps (M) >~ 7' Probably No (N) Question 2:

• Will resolution of the issue or satisfaction of the information need provide support for a significant change to a licensing co'quirement?

Probably Yes (Y) Pro'bably No (N) Question 3: Will resolution of the issue or satisfaction of the informa-tion need have a significant effect on perception or quan-j. tification of risk, where risk is broadly interpreted to include both health and safety as well as plant economics? Probably yes, and in addition could change i F' extreme consequence estimate (Y1) Probably yes, but would only influence other than extreme consequence estimate (Y2) 51

t

','j Table 2-1 1

Priority Conversion Matriz .t M. Q-1 0-2 Q-3 3 Impact tapact tapact Risk Perception t. Priority Operation Licensing or Quantification

r. :

n 1 Y 2 M Yi 2 N Y1 2 M Y 3 M N Y2 3 N Y Y2/M/N 4 M N M 4 M N N ') 4 N N Y2 5 N N M 6 N N N NOTE: (-) means that the question answer will not change the priority ranking. 9 e a J 52

i b j Perhaps (M) r Probably No (N) '-.a. [.' ouestion 4: .:Y V4 The final priority is obtained by adjusting the preliminary priority on the. basis of the plant class (es) to which the uj issue, information need, or progran element applies. The l1 adjustments are as follows: Applies only to future plants Add 2 5 Applies to plant presently in the application-construction process but not to existing plants Add 1 Applies to existing plants No change 'l in general, all energency preparedness issues apply to existing plants; therefore, question 4 does not affect prioritization. l 9 J' 6 i-i': i j i 53 I ? _ _.... _. ~ _,

i t [ u st-N. 3. REVIEW OF ONGOING RD&D RELATED TO EMERGENCY J PREPAREDNESS ISSUES r': ': 5 This section contains a summary of ongoing RD&D in the United i

• [,

States related to the resolution of the issues and informa-h;'; tion needs identified in Section 2. The extent to which ongoing RD&D activities are expected to provide the informa- ,} tion needed to resolve each issue and those issues for which p ongoing RD&D appears to be insufficient to provide the needed information are both identified. Four points need to be emphasized. 1. The RD&D considered includes only work fu,nded by U.S. sources that is either underway or planned and funded. It does not include work which is contemplated or tenta-tively scheduled. 2. This evaluation obviously'does not account for those RD&D projects whose objectives and goals may change signifi-cantly in the future or those which may be cancelled. 3. Descriptions of the RD&D programs included in this see-tion were obtained from the agencies sponsoring the work in the form of written summaries when such summaries were available. Selected additional information was obtained from personal communications with individuals associated with the programs. In some cases, the available prograa descriptions were vague, and a considerable amount of subjective interpretation was necessary to characterize the program intent and scope. ,j 4. 'It is possible that relevant RD&D programs may exist in the private sector which, because they have not been ,~' publicly disclosed, are not summarized in this section. 54

t 1 l It should be recognized that this evaluation is representa-tive of a[ point in time. Future program data, achievements, completions, and changes are inevitable and shall alter the results and conclusions herein. 3.1 Review of onooina RD&D Procrant f?; Summary descriptions of ongoing RD&D programs related to the issues and information needs identified in Section 2 are con-tained in the appendix. The descriptions include programs sponsored by the Nuclear Regulatory Commission, the National Environmental Sciences Project, and the Federal Emergency Management Agency. With some exceptions, each of the program descriptions in the . appendix contains the following information: Name of program, l program sponsor, performing organization, expected completion date, program objectives, current funding, program summary, and the information needs (as identified in Section 2)- addressed by the program. In some cases, information on per-forming organization and funding was not readily available and so has not been listed. 0 4 4 i [ I i l 1 l 55 l l

I s ?, i f;'s [N 4. ADDITIONAL PROGRAM NEEDS ) ? fi,' 4.1 Evaluation of the Extent that Issues and Information .e Mj Meads are Adecuately Addressed by Onacine RD&D Procrans ['.d .g ?j In order to deduce the need for additional RD&D programs, one o.! must first determine the degree to which the current programs I T *' are addressing the identified issues and information needs. To aid in this process, the existing domestic RD&D programs summarized in the appendix have been included in Table 4-1 ~ and are cross-matched with the issues and information needs. At the intersection of the information need and program, a letter designator appears in the box if the program will supply information relevant to the informatio,n need and will ?- assist in resolution of the related issue. The degree to which the program will provide the information needed for the information need is indicated by the entry of an "H" (high 5, degree of applicability). "M" (medium) or "La (low) at the j cow and column intersection. Where no such entry appears, the RD&D program is not of significance to that information need. 4.2 Information Weeds not Adeauetely Addressed by Oncoina RD&D Proccans The adequacy of existing RD&D programs to resolve each issue and its information needs is summarized in Table 4-2 for s'. priority 1 and 2 information needs. Table 4-2 goes on to [: identify information needs for which additional RD&D is l,'. Judged to be warranted at this time. Note that the need for additional RD&D has been limited to priority 1 and 2 informa-u tion needs where existing RD&D is judged to be either "margi- . nal" or "inadequate." Additional RD&D is not recommended at g-h,. this time for priority 3 through 6 information needs. 9 56

o -,, ' ~, $';~ *,.,, y r PROGRABAS TASLE 4 4 I 40095 810G CDOUP 000 EMENCEseCT PSEP&4ttlesE55 g j g CUSSELATICII OF IIIFOestATIOel teEEDG c ea

m..

w e wiTu a us e Oceans 2.

:n 2

3:3e PRSORITY 3 2..2: e-g ~3 3: -i. si i II e OUESTIONS i' ? ? g FUNCTlON ISSUE INFOnt8ATION NEED g: ga g: 4 4g at 1 2 3 4

• 2

.2 .2 53 53 R-4.. m 5. N = T ieta1e eace.c s.ei . r.ctie.a.ee Emerg.ac, ,etEle . ee lic ..e.e. e..... 4a... e.e......a.e ...g.ae..e4...sem r.. Emes.o.4 m.im. 4.5 Plae for Emergeactee a 4.5.1 4. 4. T T N 3 L L L Estabiloh Baege.f Seelow cetteet eastgency Fleeeleg accidente med plaeolog (EPS segela-tenede eletaeces See tiene sea psactices la estich to plae light of Seeseee leasee4 tree rene ame s.deced seesco-team eetteates. ama laestlir meditlea-l tlose es appsops6 ate 2. M T a .2 N N L Ceepase witie plameleg ettest s egels.4 f or other hasasdees med emesgency elseattene ,mp ~r-e,-

t ) 4 '.. ; I j. ,.,V '.S,* ' c,) ? 4 4h + i t..* l l solutovetstasc soltav ( '.,'.y-g a SA1180143d 90*3EN i. elsettaa toisenatAs g a /Su13011848 10*d83M 4 59139188)$ 1 4811

• ett eA3 4918 3 033 (0*VW34 gr ;-

g C n s a i e i T a f ja N a O l: a 3:} _ l2 y [ 4 3.* 14: W

• . i E

{1 : E 3iiFj

• 1 2*r

..i g 33 31*]! 'I t 2% X I. t g -:r - ni-s p-21:.i: 15 8 2 IgI 5 _22x: s _l r: 2: 2..:. _: -i - 2 :.r E 2ll22 f 7 aJ= =I 0'i 5[-l 3: 3-2 3 g yi:4 = t- ~ g e s

,t.

us g-i i- ._1 ::: ltt =c 2 in

i p.: _

s s 2 :.l:i, -m. .t 1.:1 2,:.::.1 3 a g -E 423 1'8 4.- 3.z:!:: I c:.i:

• i_p 2 : e

=

i -

".12 3.:l l ~;t2.:12 -g

• t':::_

.39 9 59 J

'.'-:,.-~.,-:,v. ^ .,,.f .c 9 4 PROGRAMS m 82

• -8 t

s..

4 Notanno asuur On esencener reermemess

8. " 7

.h .D

• COSSELATICOS OF SIBF09004TIOel teeElpS PR60RITY

.,. m D r.0c. ~.3 o-o 3

e..

k.

e. 3 i

.e. OUESTIONS FUNCTION ISSUE N4 FORMATION NEED 2* g. [ *- 3* d' a* aE 40 1 2 3 4 -2 Va f3 2: E3 C ~ e.a. Sc tl s. l. m v v2 2 m a e.tamil.= na.g.acy f a.al.at. c..... e.t.e s.,ie .ta-4ect.i.. cel-88.. actl ceia.il. ti.. ca,.. t. la f., lencel a.4 ify as slitti..irre. ..e e.al.g me..pelet. redoressDec;- public pre-eles critesta tocalve actless on so 2. m T vt 2 m L m a L L = Ev.; mate elet red.ctlee effectlwene.e of warlo.e prot ective acties alter-9 maatwee 3. Y N T2 1 F L L L Deteamlae reliabl58tr and of f oct Iwomees of altigattwe eyesatet actione as they relate to protectlwe acties d>ctsten.atleg 4. m W T1 2 F M L L L L Determine piaat leforma slee need.4 by Emesgency saamager t o. eke appte-palate e.esgency seepease declelee I

. ),:..:..:... %.,,.:p;-..::e

~;v' : ^*', . ' : 5.~ . *,. '.j %. :.:.;.:. :... u..n.- ~. Y Ma r e .....:.:..; a.u. u:4. m.. n.g.uc - ~ .ve:.. ~.... ~ s. ....5 w. 1 PROGRAMS 3 T**88 4-1 y s. teosEtteG ceOUP Ces EseERGEseCY PSEPARE001EEE es o-w CoetELATICII OF ItsroestATIO91 IsEEDE ,E ,e I a, e g wivu aoso emoceases 3: PRIORITY

.

= a=. - - - t, .1 .= es -

e. B QUESTIONS TT is 32-FUNCTION ISSUE INFORMATION NEED I* l*

IE 05I ~ Et E3 1 2 3 4 t'; e. j 4.1.3 (cent) C. 1. T N N 1 M L Establieb can a unified Assoas femelblitty of Implementa-dectatea medet devoteping untfled ties capa-be developed (lateragency and etti-bilities/Pte. and tople-Ity) decleloa sedelt cedures/Dect-mented7 mattle (see ateo stem Catterla 4.3.2.A) <h 4*2 O Maletale Readinese g 4.2.1 m. 1. N N N 4 L L Ceeduct Techalcal and Aasees esteether there age Esercleest secto-petitt-elemente of public par. Tralalag can adequacy ticipatloa that ebeeld of emetctsees be included la esercises training (esp. proper receipt of abetter laatteset tene) G J

+s - 7. 1 .N' -. ,g.,.. L- !,. ;. ',v :.t..., # : i ? 1' PROGRAMS 3 l TABLE 4 1 ( 9009E 3000 GROUP 001 E8ERCteeCY PSEPAREDesESS 3 CORRELATIOtt OF IDE705MhTIOO8 88EEDS ,g a e 98tTH BD&D PROGEAMS o = w. PRIORITY =

3.;

oo .a u QUESTIONS i '* i" iT FUNCTION ISSUE INFORMATION NEED lE ja l2 1 2 3 .4

e. "

e. 4.2.2 4. 1. M N Y1 2 8. M Maintain Adequacy of Determine meet effective Pubtle Readt-pubtle aware-methods and actlvlties mess /Awarenese mese set malatalalag pubtle awarenese and centteence I 2. M N N 4 K Evaluate effecttwenees ..d ies...e ies...d ftoo pubtle awarenese pro-graGa used in montadle togical EP areas (e.g.. matural gas, paleon con-trol, etc.) 4.2.3 A. R. M N N 4 7 Malatale Adequacy of Evaluate and lacerperate a Factlity facility feedback from emerc sse Readinese readlaces and actual emergencies 9 ? s

,-._.'.s. ', f,Y. -.. - *

• L. v.

.. " I. #, .,' A. - d..,ji.).{;s' * *, 's.h.. :,,'l ~ ~*' ..s. .y. ........,a. c o e PROGRAMS = TABLE 4-t WOSEING GSOUP ON ENERCENCY pBEPABEDNESS "a e a COSSEl.ATION OF INFORMATION NEEDS Og wITH BD&D PROGaAMS -. p ee a. e PRIORITY e ..e. .o .a as o o .e QUESTIONS 0* i! 3.* FUNCTION ISSUE INFORMATION NEED 53 l.; o; aa Em 1 2 3 4 6 um 4.1 laplement Plan Durlag Emergescles 4.3.4 A. 1. M N Yl 2 M N Evaluate Adequacy et Evaluate local teoather Hazard /Elek enwitenmental pregneale capability e y pregaeels relattwo te plume trane capability port and dose protec. (weather / tiene (and improve. It dese) necessary 2. N N N 4 7 Deteralne applicability et U.S. military taoul-edge /R&D en plume bebay les 1. N N M 4 L Assoas relattwo merite et melag plant para-meters we field measure mente for protecatwo actlee declelene M 4

a c,, ;... ;< 9..,..;.. y ;.,.... p g,i ' :.g f . c3 t,,..u.: -; .,.,., ; - -3,., .-r'L

• j,,3

, f ~ 9 4 e ,4 PROGRABAS 3 i TA90E 4-t a e I 0809 5 3000 OSOUp OEI EeStnGEBBCY petPAGEceIESS as as CONSELATICOS OF EIIFOestATIcel IIEEDS 3 WIM SM pSOGSM

• g O8

,g g, PRIORITY

== ~~ a- .= e. .o , e. u k. n .8 .e j QUESTIONS ?! 3" EE "I'El1 0E 00 FUNCTION ISSUE INFORRAATION NEED 1 2 3 4 E 6: R: 53 l-I a.3.1 Scent) A. (coat) 4. n u u 4 L svaluate Adeguacy et Assese need ter leproved sessard/S t ati environmental field moetters to aid la pregneels rapid effette doce capability calculattene (and l (esea t her t develop it necessary) i dese) 1 1 O s. m a m 4 J Deteralee valme/lepact I of eewiremmental radla. [ time asalterlag systees ) ~ S. l. N N VI 2 L L Adequacy et Develop leproved accl. i accident pre-doet progreeslea predic. greseloa ties capability for pregaoele deteratalag emergency capability reopease i-C. l. N N N 4 L j eseed for mal-Evaluate modele/calcula 1 tied progne-tiemal techalques la ele approach order to enaderstand and (to multiple rattomaltre differlag prodlettene predlettene (lacledlag s by different ceasideratloa of ageacteel standard effort) e i G I T' I"

"J p /. r'. m'*. 4* Y e g e. ?. .4, . *l + ' a,*g,,, ) h' s p, s. c ' i, ip r .js I 0.r .e a L e, 0.a P [] ,-.: u a. A.3",a s L L n e , eS : ~.n. 4 ? 2.m e' $ :~.a.* E a. M H . =.m. Nilm. 3 gan S ,E 1 2 4 M f' A ( R 4 G S O N Y R T O 3 M Y N P I IR T M S O E 2 M M N IR U f P Q M M a 1 Y gg/ D ant . eo e E fli e yunw o t optd t cot w al E onon ili t at ra N yl emi l at cl t ou t em i cce c pt S i vend b ut ea rc S N l et ee i rdonr oa E O i d pl s soer ee c N S b aes af r pl t as I D D T i rtl u e ea l o E E so

• t c f eeslt rs E

A af res dsucte dt e RAe e edi eoooeem n i Pt M rdt d nhdidna akc Et e d C. i t r t r cn R semee megantt sae t R Os P M O anell3 rmnvoe abg I o tt t e i it e odr TA Y Ae F sdst t1 rt sm see . tdt Ct enaecei . sem . snyna ec N e 1 1Aasum4 1Dalt art IAf e to lEoe I 4C Fr Re EE t LM D s cea I t e erp f el BE Fu E - ct c 7 etpa o ni) A Os neee ee wt c ameb T0 U eddn on t ee yr cot ac 0 ~ [p 0seu S sdmt t rtt cneoy at nf ane t o/ott o t sn coeictt t P OT S ye u oe eenit uaei.cu I l U Tw I sant ypll tl eti grl t nio. O A l eerpac ot wcl aet oency aL . ett sauee . rclii .dpcnl uct G E Al asussrd 3Patdb AAoa(t mas B G t ~ N oC N I E G O O' I t e t W T nnrr n eoeu e e h' C ml d men eet e

3. t t i et o N

2. t t

.,h ec U 3 pcdo

3. p e t aoer anc F

l 4t DsP 4tOA t e s

s'~ .m, e s..s ,5.-w'*.' o. ; -4**,A,~~.*,3,c',**, e- . <.r...6.' v': \\ ,=N.G.' O *C. . -- a-'. r. f'.:-y *e - ;. w.

• K's ;,".,,,, _

a ..~.... L % r f e I r-l I PROGRAMS 3 J vAsi.E 4.t s, i 9808K1100 CSONr ON ENESCEISCY pSE,AREDNESS M, q g i COSSELATIOct Otr IIIFOWNATIOgg settpS l telTH RD&D pSOCRAftS "U U* FRIORITY 3,. :. 3.3 32; o ~ 90 est se 4. e4 e. QUESTIONS O =>= D ee O. 0 = + .e s. 4. c FUNCTION ISSUE INFORMATION NEED 53 50 5 " i: Em 1 2 3 4 6-

== ma 4.3.4 A. 1. N N T1 2 L M L i s.,i .t r mile A..... sile r..,. on. l a. r.e,. . to

t. dir. cts... la it,ma Acts.

t..ca. cts.no .c.caman.c p.ce.tv.d i alga-.re.e r..itutt..e <e.... .m. iter watie 1 ei e ...ei 1 ~ 2. N N SE 4 L E.. t u t. r r c.. .t er.c.d.:e. cand id..eary 8.es. e.te. tt c... eary3 t S. 1. N N N 4 L C .tcatl.. Dev.t., g..esIc g ldance c..rdinatl.. f er c...iderat l e wenich a .g eatl-wlIt pr et..ffectiv. Lt.. c .lcatl..e (hard-ware /trat l g/pr.ce-d ree) C. 1. N N N 4 N L sole.t Assess key w.rker tl...! re. .e E. directl..e c..ftlet

1. light.f blgh-slak to petic..

.ituatt..e bue driv.ss. .tc. a

i5 W.+ ? :

O,

i i 4dl* s e a py. si u. D. it;, Se13esseleisee seitsy .a 9,,- eA128et*3d 90-335 >'.:(} etoaei .a IX. soltav Aasetaees to-38st y'$ setteelsessoa ~ ..y aetempene20 (0-388 ',~s'd etae 133 seltenaeas 8 /6e13ettegs to-dsam d' 1031 03 e3needz3 d [,Y settelpe8 10 VM34 o.. eleAtovy toestevevu "3 ,si Asse83eet 90*VMEJ soiteta=1s s sets 8 -enteas esissess co vw34 i W g N 4 N o ~. O E 0 m E -a >a ) O w a g g 8k O +- e a a I Ow 2 02: 3:- w

• 1:1.

2 22"*: t4l t' 3;

g!:!j t0 j.

E a s Ig 2 2 i.% 24

• g-

't L.

t. ;

i t.. l 3 33.3.

3

.-3 Es - g

r:::

3 2.3 si:

I ~...t -..st.': -3%.:t.:. o E su o 4. 1" 3 .3 $8 g8

.* r j

s .1 't

• 3 "

EE 2:30 ::

• 4; 3:42

. 2,j 3 3 : 13 cii < u,3 3..1 me.3..%::. 5 m k?' GO E te g O

2"

-[ e = N

1. tt:

3: 2 .ti 43: ?/- 5 4: 2 s 485. Jiu 43 2 - } i ?. N. 66 s t

.s. u_ s ,,..,,.,,. ; ;,. ;,q;e;x-y g;.,.y, s~y-7,p,, ,..,. 3,j. Q PROGRAMS e TAeLE 4-1 3 moesima caour om sm acency receaecomess comers.ariou or mercenation macos 2j: : virN eoso emoceans PASORITY =~ =* a. o. 4 e OUESTIONS ?" TO ?: TT l2 l* la l3 2 FUNCTION ISSUE INFORMATION NEED a .2 1 2 3 4 m 6 6 4.4.1 (coat) A. (cent) 3. N N N L Estabileb Crlteria ter Evalmate slek/ cost-mecovery decentaalma-besettt et decostaal. Planalog Llen. geocca, matlea efterte Criteria patloa er abandeament 4. N .M M $8 1 Evaluate 4.4.1.A.2 and 1 4.4.1.A.3 to find optt-mal balaace and to fac. i ter la secte-pellalcal leemos (e.g.. critical ladestries, water) 4.4.2 A. 1. N N N S L L j Make Docl-sole of pub-twalmate "lessons alone and lle percey-learned" free chemical Implement tien la contaalaattene (e.g.. declelen. Love Canal. FCe com-g making and taelmattene) and thelt a reopenelve. applicatien to meclear mese Er a 6 m. 4 ' e \\

- - ;t. ,... m. :,., z ;ng.g,,~,; q,.3;.yz,g. y y;_, _.; .,j a _ .m

pf.._

e;, _.,.

.g....._ .s. 1 ~ Table 4-2 i Summary of Recommended Resolution of High-Priority Information Needs Ade-quacy of Addi. Time to Existing Prog. Type of Cost Complet,e j Information Need Priority Programs Recom. Study Estimate (months) 4.1.1.A.1 1 NAL Yes Evaluation $100-500K 6-12 Review current [ emergency planning (EP) regulations and practices ~in light I of lessons learned from PRAs and i reduced source-term g estimates. and identify modifica-l tions as appropriate l 4.1.1.A.2 2 M2 Yes Evaluation $100-500K 6-12 Compare with planning effort required for other hazardous and emergency situations 4.1.3.A.1 2 M Yes Evaluation $100-500K 6-12 Evaluate present emer-gency classification criteria l

. ;,, m <..a...<. .,w....r.._._.. n mm em 1 Tab.!e 4-2 Summary of Recommended Resolr' of High-Priority Information Needs e .'nued) Ade-quacy of Addi. Time tp Existing Prog. Type of Cost Complete Information Weed Priority Programs Recom. Study Estimate (months) 4.1.3.B.1 2 M Yes Evaluation $100-50CK 6-12 Evaluate current Protective Action Guidelines (PAGs) and modify as appropriate oe 4.1.3.B.2 2 M Yes Assimilation / < $100K < ~6 Evaluate risk reduction Application effectiveness of various of Many protective action alter-Efforts natives 4.1.3.B.3 1 NA Yes Analysis $100-500K 6-12 Determine reliability and effectiveness of mitigativo operator actions as they relate to protective action decisionmaking 't a 1

4.,,..,.,

e. 7 _.7, q. ... q.. y .., :.Sg.:.g.;;... ; un w c.,., 4.. :, e... Table 4-2 Summary of Recommended Resolution of High-PriUrity Information Needs (continued) 1 1 i (_ ' Ade-quacy of Addi. Time to Existing Prog. Type of Cost Complete Information Need Priority Programa Recom. Study Estimate (months) j l 4.1.3.B.4 2 M Yes Assimilation $400K <6 Determine plant informa-of Existing tion needed by Emergency Efforts Manager to make appro-priate emergency response decision a O r 4.1.3.C.1 1 NA Yes Feasibility 100K 6 Assess feasibility of Study developing unified (interagency and utili-and ty) decision model/ Maybe Full Study > IMM > 12 matrix (see also 4.3.2.A) 4.2.2.A.1 2 M Yes Evaluation $100-500K 6-12 Determine most effective methods and activities for maintaining public awareness and confidence I e %M G. .af - ep +

7, '.;- *_ ~ ~. ~;.. .y :,. ::,.. -

#. t, :g%.'. z!,.:,s.,,;;q,,.,..

,.3.,, ~. ~.. x, 7,,.- . y. , y. ..g :". q, 1 Table 4-2 Summary of Recommended Resolution of High-Priority Information Needs (continued) o Ade-quacy of Addi. Time to Existing' Prog. Type of Cost Complede Information Need Priority Programs Recom. Study Estimate (months) 4.3.1.A.1 2 A3 No Evaluate local weather prognosis capability relative to plume trans-port and dose projec-tions (and improve. if ?! necessary 4.3.1.B.1 2 NA Yes4 Evaluation $100-500K 6-12 Develop improved acci- .l dent progression predic-tion capability for determining emergency response 4.3.2.A.1 1 NA Yes Feasibility 100K 6 Assess feasibility of (see Study and need for developing also system for implementing 4.1.3.C) unified decision model/ r matrix discussed in and Full Study 500K > 12 4.1.3.C Maybe 'd o'

. f,e7,.,,{4.,;",*yig*M, 34 o N."<>gvj* .. -.,,; #,v,i j t ? /4- .. m . p ..s., f,,g /p.p;y. (,:*3 _.y.ik .e",. p 2, '8 Table 4-2 -.'1 + Summary of Recommended Resolution of High-Priority Information Needs (continued) a Ade-quacy of 'Addl. Time to Existing Prog. Type of Cost Complete Information Need Priority Programs Recom. Study Estimate (months) 4J r 4.3.2.B.1 2 M Yes Feasibility 100K 6 7 Determine feasibility Study .f. and methods for calcu-lating dose reductions for various protective action decisions in j real-time / interactive time frame 4.3.4.A.1 2 M Yes Evaluation 100K 6 and Assess public response to directions in light Assimilation of actual or perceived l high-exposure situations (e.g., shelter while plume passes) INA - Not Adequate 2M - Marginal 3A - Adequate 4Maybe in series with RD&D in other functions. e S g Q9' A

. ; s. -, ' s a j L. l J' "l[ ,u (9 ( 4.3 Procram Recommendations Plans for Resolution of Residual ~ Inforaation Needs

.ll,-

i v.. t c'.V Programs have been recommended for resolving those informa-v, tion needs which are of priority 1 or 2 and for which ongoing .r' '( programs are judged not to be adequate to supply the needed ({ information. For information needs of lower priority (prior- ..) ity 3 through 6), progran recommendations have not been made: (', additional program effort on these information needs is not bn recommended at this time. . '.? The progran recommendations appear in the last three columns (,I of Table 4-2. This information includes the type of program I' and estimates for the cost and time required,to perform the 2' program. The cost estimates are not an attempt to allocate future RD&D fundley nor do'ther represent an laplied state-ment of importance or priority. In suancrf, Table 4-2 shows that the majority of high-priority information needs were judged to require new RD&D programs. In large part such judgments were broad on the present scope and directica of the ongoing programs and the need to either integrate their results or to fill gaps in their scope. In order to assure the timely recognition of completed issues, and to assure that new facts which could require changes to 's existing programs are recognized, it will be necessary to implement an activity to monitor the progress and information provided by ongoing programs and to periodically update the information in this report.

a.,

• l 73 G

---m ,,-,y .--_..--_.--,y- .,-7---- c-.--,- p - 9 ,,e e-

Py. ~ g 1." s, ,, a~ .S. s'.> a s Y h 'g*.*,, ,Y. ~,2 ?* Q 6-t ese j f.* Y,')L: ,=f a ge

p. I.

e =

• 'V',

eea* ... e

• War 4

.g .,g o1 .,6 ... -,1 P

• be e

e 4.

• g.

9 e b ) I, h n* )

• 4 6 4*

4 % O e I e e 9 r 4e* 6 $e e. 9

• 4 t'

M S ?..

• 4 s

b .,9q O 74 s

m. =, .m

kI5, v.

[,#'.j ('- APPENDIX - EXISTING RD&D PROGRAM SUMMARIES

. a 3.'i h, Q.!

FEMA-01 [+.,. Comprehensive Hazard Analysis 'tl s '.~ 5.: 20' ~ 30 TyM Procras Name: Development of a Methodology for Comprehensive ] {N Hazard Analysis Sponsor: FEMA .fp. Performino Orcanization: (out for bid) (( Conoletion Date: FY85 (Phase I) Fundinc: Unknown i '. ' f Obiective: Te develop and implement a scientifically sound methodology for analyzing and comparing hazards within specific geo-graphical areas and across geographical arcas for use by the various energency planning jurisdictions. .3 . s.

== Description:== This program will consist of an initial phase to determine J the feasibility of developing a methodology for translating the characteristics of a diverse spectrum of hazards into a common basis for cross-hazard comparison. A literature search will be conducted to locate the most useful sources hazards data, such as frequancy and consequence. Subsequent c Phases are optional but would include methodology refine-I. l-ment, data collection, and eventual implementation in FEMA's j Hazard / Vulnerability / Capability Assessment program. 'A Information Needs Addressed: t/,. j' 4.1.1.A.2--Comparison with planning effort required for I other hazardous and emergency situations. J 75

{ ..v..> 1 r i , #,J 6M

• r:.,w;
• . I FEMA-01 i

f- .7 Comprehensive Hazard Analysis 1,'e-(continued) c99 p S* * '* 'k. a- .s pj'.-l 4.4.2.A.1--Evaluate "lessons learned" from chemical contaal-NI..' nations (e.g., Love Canal, PCB contaminations, etc.) and 5 their application,to nuclear EP.

3...

,h . eg + ri. ?. e y s e 4 ,e a P.. 6 ,s. 4 4 e e. .A,* E '. ..f s iD 4 4 .g

• s 4*,,

3 g U,.

.s w'

..) 1 76 -'t 9

~,.:: u.: ., y ;

1. ![

l I' j.;. {..i h7,', ( FEMA-02 $[ Exercise Eva'Austion & Simulation ~ $P' %?) )$; Procras Name: Exercise Evaluation and Simulation Facility j

w.,

i 5'.' j (EESF) p, .R Sconsor: FEMA .u

2 Performina Oraanization:

Argonne National Laboratories

• ~

v..

Ge: Battelle Pacific Northwest Labs $c ConDietion Date: FY85 Fundina: Unknown .;g 2.s. : ) k. 7,W Objective: 2.' To provide FEMA (federal and regional) as well as state and local authorities with a computer based system to help l eva'luate radiological emergency plans and preparedness. The ( system is designed to plan and evaluate exercises and to assist in the response to actual energencies. DescriDtion: - l. ?. f.. L EESF is a computer based system developed by PNL presently N'I in the implementation phase. EESF will consist of ten 3, regional "work stations" and a central computer facility in H, Washington, DC. EESF incorporates the following computa-lt tional/ display resources: asteorological model, dose model, evacuation model, map information, and exercise information. .c Information Needs Addressed: M '. 4.1.3.B.2--Evaluate risk reduction effectiveness of various protective action alternatives. I

l...

H. 77 I s

e .e~ .A r ,4

a..i

!l.j FEMA-02 } A.' Exercise Evaluation & Simulation II! (continued) L.* q 3d lN (f 4.2.1.A.1--Assess'whether there are elements of public par- {y ticipation that should be included in exercises (especially ] regarding proper receipt of shelter instruction). 95 4.3.1.A.1--Evaluate local weather prognosis capability rela- ~. tive to plume transport and dose projections (and improve if - ~., 5-_ necessary). s:, 4.3.1.C.1--Evaluate models/ calculational techniques in order to understand and rationalize differing pred!'ctions (includ-r. [ ing consideration of standard effort). 4.3.2.B.1--Determine feasibility and methods for calculating }* dose reductions for various protective action decisions in a real-time / interactive time frame. [ '- 4.3.3.A.1--Assess and incorporate feedback from actual emer-gencies. i 4.3.4.B.1--Develop generic guidance for considerations which O' will promote effective communications (hardware / training / ~ procedures. n C 4.3.5.A.1--Assess feasib~ility of using real-time feedback m. loops to account for actual responses and accident condi-tions.

• ?'

I. r I -7 ? 78 5 e w,, n -,.. -., +.

, ; s., ,s y.. c 1,r $k Q:!t i',

4... l (.

FEMA-03

• / ~;

Hazardous Materials Planning %i b.'.} S @r j]. Program Name: Hazardous Materials (HAZMAT) Planning Basis b Sponsor: FEMA >-(,, Performina Oraanization: (out for bid) Conoletion Date: FY85 I.'.' Fundina: Unknown '( }.. .h q. objective: de To de.velop an energency planning basis for (n,onradiological) hazardous mate, rials similar to NUREG-0654. Descriotion: [' This program consists of three parts. First, to investigate the need for a MAZMAT planning bav,ls. Second, a determina-tion of the feasibility of developing an actcal planning basis. Third, a detailed description of the total require-2ents for a HAZMAT planning basis. This program, in con- ,j junction with others, will provide an insight as to relative s.f risk and relative energency planning requirements for HAZMAT ]' and nuclear power plants. t, - Information Needs Addressed: g '4 0 4.1.1.A.2--Comparison with planning effort required for k,' other hazardous and emergency situations. J. [yj 4.4'.2.A.1--Evaluate = lessons learned" from chemical contani- 'Ti nations (e.g., Love Canal, PCB contaminations, etc.) and l-l'~ their application to nuclear EP. ( I 79 a e } 'e _~

.5 f E N Energency Management Analysis j "q FEMA-04

9..

d f Program Wase: Technical Analysis for Energency Management l:j (Task 2) i} Spo nso r.: FEMA ( Performina Oraanization: Oak Ridge National Laboratory p ).,. (ORNL)

e.,

Q] conoletten Dat'e: h* Fundina: $550K Er m f,(i Oblective: ORNL's assessment of research on energency management system requirements and counter measures will identify problems / information requiring new research programs to resolve. Although the principal focus is on nuclear weapons, the five

~

topical areas defined by the program are of interest in 2 power-reactor-energency planning. The five topics are 1. Recovery 2. Shelter 3. Evacuation 4. Emergency food and water supplies g 5. Residual ecological problems of contaminated land I' Descriotion: Available research progran material will be collected, the ) ,'.c findings summarized, and a judgment will be made as to ...j usefulness vis-a-vis energency management information el requirements. Ongoing research will be identified and b.' described and bibliographies will be produced. In addition. l'I \\ 4 80 4.

~- , j,,.

v..

^' j i t l i.~ ..'d -f,} ( FEMA-04 ,",:j Emergency Management Analysis Sj (continued) 3 .r ,f1 the present state of scientific and technical knowledge.will e.b C'd be summarized. f$ ->.T' Information Needs Addressed: L'1 2:-} 4.1.1.A.2--Comparison with planning effort required for ~ other hazardous and energency situations. f.'. 4.1.3.B.2--Evaluate risk reduction effectiveness of various protective action alternatives. l, 4.4.1.A.2--Evaluate risk / social cost of residual contani-

nation, b.

e + w e 6 b e e [ b . 's ".je d 81

i .-----y--- = ~ * +.. ft.d FEMA-05 P.O. ,C Radiation Exposure Control C' ';'i, e y) Plo.a.r_am_MAme : Control of Exposure of the Public to Ionizing 4 Radiation in the Event of Accident or Attack f Sponsor: FEMA lC Performina creanization: National Council on Radiation v: Protection (NCRP) [c[ M Concletion Date: FY85 b Funding: Unknown hl Qh V '. Obiective: l.i Li n r. [ 7-NCRP has produced reports such as "Exposure to Radiation in [- an Emergency" and "Radiological Factors Af f ecting Decision-I making in a Nucler.c Attack". After TMI, the NCRP's focus was shifted to include nuclear plant accidents. The objec-tive of the present program is to consider more completely the approaches to dealing with emergency situations.

== Description:== This broad-based effort is being carried out by an NCRP sub-connittee which was establi::hed to exarine problems of, and advise FEMA on, the radiation hazards from nuclear events. The program was begun in April 1981 with a symposium. The NCRP subcommittee continues to explore knowledge ranging from radiobiology to management roles and responsibilities I, and from human behavior in time of stress to radiation f f*k. detection instrumentation. Specific topical areas around p.' which the kickoff effort was organized are [.GJ \\ 9 8 4 / s2 (

c-n i

p. c (e.

bI.! l,b.$ ( FEMA-05 Hi Radiation Exposure Control h,-f (continued) 4 ~ d.! ,5h 1. Radiobiological basis for establishing criteria for the 7) population j.0f 2. Radiation exposure control procedures ..n. ?;'. 3. Dose and dose rate measurement requirements .).'2 4. Public information and training , 3, ; 5. Roles and responsibilities of federal agencies 4 Information Needs Addressed: 4.1.3.B.1--Evaluate current Protective Action Guidelines (PAGs) and modify as appropriate. . o 4.1.3.B.2--Evaluate risk reduction effectiveness of various protective action alternatives. ~ 4.2.2.A.1--Determine most effective methods and activities for maintaining public awareness and confidence. i 4.3.1.A.3--Assess relative serits of using plant parameters vs field asasurements for protective action decisions. 4.3.1.A.s--Assess need for improved field monitors to aid in t rapid offsite dose calculations (and develop if necessary). t 4.3.4.A 1--Assess public response to directions in light of 2 actual or perceived high-exposure situations (e.g., shelter -f ' while plume passes). ou .s e 1 ( 83

f'n_ n l V.i,.i FEMA-05 ) P.' ~ Ih Radiation Exposure Control Ik (continued) >','1 Gi iP o.- ll ' '. 4.4.1.1.2--Evaluate risk / social cost of residual contamina-c.; tion.

7. 5.,..

4.4.1.A.3--Evaluate risk / cost benefit of decontamination ,*? etforts. p ** 'd,*. g.< -

• Ao go

' :s. e e 4 e e i..* e e a '4 9 l*.' 6 9 8 e 4 D 's P is' Oe ai e .t 'g 4 6w O4 i

3 .v. ,s .o ' .T l d, p.. {, ( FEMA-06 f,;7g social & Behavioral Analysis w gt-cf ' i'! Procran Name: social and Behavioral Analysis in Civil bh' Defense and Energency Management rJJ Sponsor: FEMA Performina Oroanization: University of Pittsburgh (Uni-it; versity Center for Social and

,?

Urban Research) s[ Concletion Date: Unknown Fundina: Unknown l-.. i obleetive: To provide current social and behavioral data and informa-tion relevant to policy and, program planning for emergency management and to improve the ability to analyza social and behavioral factors affecting public perceptions, understand-ing, and acceptance for those aspects of emergency management that depend on public acceptance and cooperation for their effectiveness 1. .l - Descriptioq: This progran consists of eight separately defined, but related, tasks. Some of these tasks are focused on war time [ :j- ^ preparedness (for example, industrial mobilization) and are ,j not of interest to power plant energency planning. Tasks of interest are the following: 4 Data Bank System - A manual file bibliography of material which bears on a variety of social issues in energency planning is being computerized to make information easily accessible to users. It is presently intended to 85 G ---.-------4

.u. s,. .v.3-fj@d FEMA-06 g ) MY, Social & Behavioral Analysis e Il}. (continued). ,f 'E, ; /.t! .o periodically update this file approximately every three l,.fj years. G.O Survey of Eneraency Manacement Officials (EMO) A nation-( 'f. wide survey will be conducted among EMos to determine their fr. view of the credibility and acceptance of crisis relocation. i.L'- General Public Survey - A sample population'will be studied to determine attitudes and perceptions regarding energency 7 -] management issues. This survey will update a 1978 survey 7,,* (and older surveys) with regard to such domestic events as TMI and international events such as the Iranian hostage crisis. ) ~ Sublective Meanina Analysis - This task will attempt to advance the state-of-the-art in the analysis of public perceptions and opinions, particularly the way people think about such things as risk, evacuation, relocation, etc. Information Needs Addressed: 4.2.1.A.1--Assess whether there are elements of public participation that should be included in exercises (especially regarding proper receipt of shelter instruction). 4.2.2.A.1--Determine most effective methods and activities for maintaining public awareness and confidence. l Q '.' 5: i 1 i 86 c ~ ~ - -

/

l Id.1 \\ ( 6.'i... FEMA-06 ) social & Behavioral Analysis j .'.6,' (continued) t... ..q:.

h. '.

,)f.

4.2.2.A.2--Evaluate effectiveness and lessons learned feca f. public awareness programs used in nonradiological EP areas % 4, . ',f (e.g., natural gas, poison control, etc.). k,. 4.3.4.A.1--Assess public response to directions in light of ,p actual or percesived high-exposure situations (e.g., shelter while plume passes). i., j 'c' 4.3.4.A.2--Evaluate rumor control procedures.(and identify improvements, if necessary). 4.3.4.qtl--Assess key worker response to directions in light ( of high-risk situations. l,. 4.4.1.A.4--Evaluate 2 and 3 to find optimal balance and to factor in socio-political issues (e.g., critical industries, water). e l.' r L.., x Ui hh

r..

L :. h. f. 87 m-p,,-e-o --w-o-~~--ww

.e .G.S. e me. -. ~; 1. N 4.' l .fy I,,s' NESP-01 i ,. J sheltering / Evacuation Criteria k L, L tl'I I. 53 Procran Name: Planning Concepts and Decision Criteria for lff Sheltering and Evacuation th:. SDonsor: NESP k Performina Oraanization: Battelle Human Affairs Research Center

t

[i - Concletion Date: 4th Quarter 1984 ?- Fundinct Unknown i( oblective: 2 To produce a two-part generic decision aid'that will serve as the basis for a more detailed, plant-specific aid that . ~, can be tailored by individual utilities and offsite ) authorities.

== Description:== The decision aid will contain: (. 1. A list of planning concepts to be used by utility and offsite emergency plannern that will identify those resources and conditions that must be planned for so that either sheltering or evacuation can be executed. (Presently no such set of criteria exists.) b 2. An implementation tool (either a decision tree or a checklist) of those variables and situations which the implacenter of 4'n ordered protective action must deal with once the action commences. (To avoid overlooking details once the action starts.) e 4 as

7._.

a. _

r s. eg ',j NESP-01 2;j,' g sheltering / Evacuation Criteria [d (continued) I s.3. _.?' ff)f to the ultimate decisionmaker (i.e., governor, county exe-With the above aids, a planner may feel confident in going f,' [; cutive) and making the best recommendation. An Executive '.? Summary will outline those sweeping questions which such a u: decisionmaker must ask himself prior to giving an order to i take action. ,} f. l M* The assumption is made that something has happened (an occur- ) rence at a plant) and that a preliminary decision has been made, based on radiological dose or plant con'ditions, to } recommend one protective action or another. Now, what are the other considerations that come into play? . Plant parameters (conditions) and release parameters (source-tera components) are already in place and are taken as given. Neither must be described in detail, quantified, or ~ ~ analyzed in the study. Comments in these areas may be made [ if it is felt they are useful or pertinent to the study. 7 While the study may be area specific (that is, specific to L; the type of area where a plant is), the study should not be / plant specific in terms of plant conditions. Release levels, deterioration rate of containment, etc., are too s ~' site specific and too broad for this study, and there are too many unknowns to deal with at this point in time (such j, as source-term numbers). ..t The. outcome will be a decision aid which will help planners ,l come up with the best recommendation based on all known or 3 anticipated variables (other than dose or plant conditions) that will work in a real-time situation. 89

..r a,. ' .zu

• • e s

i %1.- ...v ,,.i y; NESP-01 j i sheltering / Evacuation Criteria -s> si,;. (continued) .r n..s

A Ld-Information Needs Addressed:

ds - f.3 i # 4.1.2.B.1*-Develop data base to provide basis for optimum I.h strategy. y? ?. ~*

• s 4.1.2.C.1--Determine relationship and importance of accident r..

timing vs change-of-command timing (scenario dependent). 'h:* 4.1.3.B.2--Evaluate risk reduction effectiveness of various c 1 protective action alternatives. 3 4.1.3.B.3--Determine the reliability and effectiveness of mitigative operator actions as they relate to protective ) action decisionmaking. i. 4.1.3.B.4--Determine plant information needed by emergency h, manager to make appropriate emergency response decision. 4.1.3.C.1--Assess feasibility of developing a unified (inter- [' agency and utility) decision model and/or matrix (see also l" : 4.3.2.A). i 4.3.2.A.1--Assess feasibility of and need for developing a system for implementing the unified decision model/ matrix L discussed in 4.1.3.C. b 4.3.2.B.1--Determine feasibility and methods for calculating ,P dose reductions for various protective action decisions in a ,.[ real time / interactive time frame. ~ c. Y' l 'i to i

e. t 9,* t. I ".. ( NESP-01 ).,- Sheltering / Evacuation Criteria 4 (continued) 9':'t ~ 4.3.4.A.1--Assess pubile response to directions in light of f,5 actual or perceived high-exposure situations (e.g., shelter 7* while plume passes). '*c. .2 Pi 4.3.4.c.1--Assess key worker response to directions in light of high-risk situations. 4.3.5.A.1--Assess feasibility of using real-time feedback loops' to account for actual responses and ace,ident conditions h I e e G t g

• e k

9 e I 91 i

_,.w. i* s . 6' NRC-01 1 ' s' Risk-Based Decisionmaking 6.sY W .$[j Proaraa Name: Improved Methods for Risk in Decisionmaking T(* Soonsor: NRC (FINB2386) h}$' Performina Oraanization: Pacific Northwest Laboratories Concletion Date: 9/84 Fundina: $660K k, obleetive: To develop, demonstrate, and apply risk-based, decision [', methods to high-priority problems at NRC. I

== Description:== ) This is an analysis nethods development and analysis applica-tion task. The principal elements of the program are 1. Development of a risk classification scheme for safety systems components, human actions, and maintenance l,,'. activities to structure and prioritize safety assurance programs.

i..

l. 2. Continue,the development, implementation, and applica-tion of standardized procedures and methods for sys-2 tematica11y performing value impact analyses for requia-c [' tory rules, guides, technical specifications, and safety 1, related issues addressed in the regulatory process. [;- L.- 3. Continue the development and implementation of risk-based j approach to safety research prioritization. I I.'.. [., 92 l

n t l ..2 ?> I. '. L' ( NRC-OL Risk-Based Decisionmaking (? (continued) r.- u 4. Identify changes in regulatory practices needed to

f r.f.

correct regulatory constraints and increase the risk I',' effectiveness of the regulatory process. i? ^ 5. Sponsor a symposium on the application of statistical ( methods to solution or evaluation of U.S. energy 9 problems. Information Needs Addressed: 4.1.1.A.1--Review current Energency Planning (EP) regulations and practices in light of lessons learned from PRAs and reduced source-tera estimates and identify modifications as ( appropriate. 4.1.3.B.2--Evaluate risk reduction effectiveness of various protective action alternatives. e' m 4 's s. 93 i

.. v

a..

.c => .= s .e h( NRC-02 ) 5'c Dispersion Model Evaluation a n. T.

• a+

PO: Proccan Name: Dispersion Model Evaluation )h Boonsor: NRC (FIN B0466) s-n[..-

/

Performine Oraanization: Oak Ridge National Laboratory Concletion Date: FY84 h'- Fundina: $200K Obiective: >4' '.

, r Identify the available atmospheric dispersion models that provide the most accurate estimates of radioaetive concen-

,c trations downwind and develop criteria to implement these models in support of requirements specified in NUREG-0654. ~ Descriotion: 'this task principally involves evaluation of various meteoro-logical dispersion predictive codes. Tasks are as follows: ..c.. a. Selected codes will be evaluated with additional data sets using the pattern recognition objective criteria. Concentrations for 1 hour periods and time-integrated values will be the evaluation cri-teria. In addition, the real time capability of each code will be judged, the sensitivity of each l. code to variability in meteorological data input p will be examined, and cost / benefit tradeoffs for each model type will be performed. e i-94

3 ~- 6 P j _I M.; NRC-02 ( Dispersion Model Evaluation ^ ' u.~ - (continued) ,'.i k:N 1 l 't. .. ? ~ b. Performance criteria for meteorological data, model jf outputs, and compatability with nuclear power plant .d. data collection capabilities will be determined. l h,' c. SRI International ALPHA-1 lidar data will be ana-4'- lyzed to obtain information on the variance of -J. '- concentration values, to determine how ground-level /2 concentrations vary in time and space as a result of plume geometries. d. The variance data will be utilized to evaluate the accuracy of the various codes. I Information Needs Addressed: 4.3.1.A.1--Evaluate local weather prognosis capability relative to plume transport and dose projections (and improve if necessary). e 4 h 95 6

b _. f*e s 4

'i'l; NRC-03

) 4/i Groundwater Contamination ).W. .v.o ':3 Progran Name: Groundwater Contamination w j'j. - SDonsor: NRC (FIN B2454) !);j Performina Creanization: Pacific Northwest Laboratories ? .fi completion Date: FYS4 39 Fundine: $200K (FYS4) $,~' Objective:

• .r;,

The objective of this program is to investigate and evaluate mitigative techniques and examine generic site conditions l. for the control or reduction of groundwater radionuclide ~ contamination resulting from an assumed core-melt accident. The assessment of generic hydrogeologic characteristics would identify those factors that would affect potential offsite releases and possible interdictive options. Descriotion:

f.,.

p" The program will involve selection of two sites for case i-studies, performing the case studies, and evaluating the results. Specific tasks are r 1, l' 1. Design and selection of case studies which would involve "[ the selection of two sites for detailed analysis of various sitigative techniques following a severe nuclear 'i. \\ power plant accident. 2. A case study in which the feasibility and applicability of identified groundwater contaminant mitigative tech-niques will be assessed based on available data from two b existing or proposed nuclear power plant sites. i,,, 96 l

.T. d j P I [ ( NRC-03 Groundwater Contamination (continued) i J.:. 3. Integration of case study results and peer' review to '.N. assimilate the site-specific information into an overall summary of the feasibility and desirability of ground-water interdiction. Information Needs Addressed: 4.4.1.A.2--Evaluate risk / social costs of residual contami-nation. ( e 4 i r e 97

I "... 4

V; NRC-04 j

f;}, g(j NPP Instrumentation Evaluation r. di Proccan Name: Nuclear Power Plant Instrumentation Evaluation k Sponsor: NRC (FIN 6369) ph 57 Performina Orcanization: Idaho National Engineering Labora- ,I tories f,.',; conoletion Date: Not specified h;If Fundina: $850K (FY84 and 85) oblective: Currently, there is no consensus within the l'ndustry with regard to the degree to which Regulatory Guide 1.97 can practicably be implemented. The objective is to provide a ~ basis to evaluate the criteria in Regulatory Guide 1.97 with ) s particular emphasis on range and qualification requirements for instrumentation.

== Description:== The program will include the following tasks: l 1. During an LOCA the environmental conditions for at least a sensor may change drastically causing an indeterminate change in system accuracy. The extent to which the sys-tem accuracy is affected will be investigated for accident-management instrumentation. The approach to be used is to quantify the accuracy changes caused by n environmental effects on sensors, conditioners, and N other equipment subjected to the LOCA environment and then to perform an error analysis on the system. The -l-results of the error analysis will indicate what I I D 90 i ~,

s 4. D E i ej. s ; ',- '12 ( NRC-04 NPP Instrumentation Evaluation og (continued) ,h accuracy and confidence level can be expected from ]. typical commercial nuclear power plant instrumentation under accident conditions. /. 2. Tests have shown that it is possible to obtain indica- ..l tions of inadequate core cooling by measuring coactor-coolant-pump power and cold-leg temperature. LOFT tests have shown that coolant pump power can indicate the state of the primary coolant during certa,in operational modes. In addition, the tests have shown that a plot of pump power vs c'ld leg temperature ("J" plot) can pro-o duce an indication of the type of transient being experienced. In order to determine the usefulness of this method, the LOFT experimental data will be analyzed to obtain the characteristic "J" plots and develop cri-teria to determine the type of plant transient (i.e., heatup, cooldown, LOCA) being experienced. Information Needs Addressed: 4.1.3.B.4--Determine plant informatica needed by Energency Manager to make appropriate energency response decision. 4.3.1.E.1--Develop improved accident-progression prediction capability for determining emergency response. 9 99

~ + '. 8 s' b-e ' N. (y? NRC-05 ) Energency Action Levels P. '. Vi Procran Name: Energency Action Levels j; pponsor: NRC (( Performina Oraanization: Brookhaven National Laboratory f. Concletion Date: FY85

'J Fundina

$400K [ obiectivt: ti. To develop a method of establishing emergency classes

• that I.

can be incorporated by the licensee of nuclear plant into their emergency operating procedures. m

== Description:== Specific accident sequences will be studied to determine those observables which can, be readily identified by the control room operators and utilized in establishing the magnitude of the energency (energency action levels). The relationship to energency classes will be identified. l.., Information Needs Addressed: I f 4.1.1.A.1--Review current Energency Planning (EP) regula-tions and practices in light of lessons learned from PRAs and reduced source-tera estimates and identify modifications g, as appropriate. 4.1.3.A.1--Evaluate present energency classification cri-teria.

• Classes are alert. site emergency, area emergency, general emergency.

100 l 4

<s .I .,,p s v;1, NEC-G3 ,[' Emergency Action Levels Q (continued) i,5 .9n; e ?; 4.1.3.B.2--Evaluate risk reduction effectiveness of various 5 protective action alternatives. 4.1.3.B.3--Determine the reliability and effectiveness of mitigative operator actions as they relate to protective action decisionmaking. 4.1.3.B.4--Determine plant information needed by Energency .( Manager to make appropriate energency respons,e decision. 4.3.5.A.1--Assess feasibility of using real-time feedback loops to account for actual responses and accident conditions. e 4 I i 101 J l l

's

3. -

~ "dI u. .,I ' y. i.1' NRC-06 t /..i Protective Action Decisionsaking h.M )., Procran Name: Protective Action Decisionmaking [') Sponsor: NRC (d Performina Oraanization: Brookhaven National Laboratory Conoletion Date: FYO5 3',- t/l Fundina: $400K s /.9 ?'i; Obiective: s A reevaluation of protective action strategies is required to 1. Identify the factors that are most important in protec- ~ tive action decisionmaking ) 2. Determine how these factors should be incorporated into decisionmaking 3. Develop guidance on protective actions to be recommended for likely combinations of these factors.

== Description:== ^ Protective actions are being reviewed to determine the influence on decisionmaking of factors such as: containment source terms, warning times, and weather.

status,

..I Information Needs Addressed: 's 4.1.1.A.1--Review current Energency Planning (EP) regula-tions and practices in light of lessons learned from PRAs and reduced source-tera estimates and identify modifications y I as appropriate. e + 102

1 ^ L. 1 .~ s. d. 5 l ~. t $.il g NRC-04 Y) protective Action Decisionmaking

..l (continued)

.n f, 4.1.2.B.1--Develop data base to provide basis for optimum ?, strategy.

. e o ',.

4.1.2.c.1--Determine relationship and importance of accident timing vs change-of-command timing (scenario dependent). 4.1.3.B.1--Evaluate current Protective Action Guidelines (PAGs) and modify as appropriate. 4.1.3.B.2--Evaluate risk reduction effectiveness of various protective action alternatives. 4.1.3.3.3--Determine the reliability and effectiveness of mitigative operator actions as they relateIto protective action decisionmaking. 4.1.3.B.4--Determine plant information needed by emergency manager to make appropriate emergency response decision 4.1.3.c.1--Assess feasibility of developing a unified (interagency and utility) decision model and/or matrix (see t also 4.3.2.A). i 4.3.1.B.1--Develop improved accident-progression prediction capability for determining emergency response. 4.1.2.B.1--Determine feasibility and methods for calculating ~ dose reductions for various protective action decisions in a real time / interactive time frame. 103

7 -= s': ,4 t-A - f. h T gx j.s Li ; Y?i ie; .~.$ NRC-04 l )b Protective Action Decisionmaking f,F,. (continued) r 1..

2..:

i,' '.4 ^ 4.3.5.1.1--Assess feasibility of using real-time feedback t.. : loops to account for actual responses and accident condi-A..! t tions. .te ? ?. '

• a,,.

b e Ie 4 e 6-0 t e 4 9 h 94 J j e e e 9 . (< r 6 a 104 k 6

, il...j....

• ^ *~

1s-v 0 5 s NUREG-1082 sh (uyo( TECHNICAL BASES FOR A GRADED RESPONSE IN EMERGENCY PLANNING AND PREPAREDNESS i +' by n Leonard Soffer I James A. Martin, Jr. I Richard P. Grill i Manuscript Completed: November 1984 b-a n____m ,ma, I e 9 I i 6 i 1 1 ? 's 1 / 4 a a ,,,w e+p m e+ A

e-1_ o.. FOREWORD Although the NRC Emergency Planning regulations were significantly upgraded in 1980 as a result of the Three Mile Island accident, this area continues to be a controversial and of ten misunderstood aspect of nuclear safety regulation. While many are aware of the requirement for planning within the 10-mile plume Emergency Planning Zone (EPZ), a significant source of misunderstanding arises from the mis perception that all individuals in this zone are at an essen-tially equal degree of risk, without a clear understanding of the spatial variation of such risk or to what extent potential accident and release timing affect risk. As more recent research has becnme available, therefore, it becomes important to emphasiae in a more explicit' fashion the basis and insights associated with the concept of a "graded" or "phased" approach to emergency preparedness or response. This approach, implicitly recognized within the original planning documents, allows planners to concentrate and focus their efforts and resources on those areas likely to require priority protective actions in an emergency, while also permitting the flexibility of extending such actions, as needed. 1 The information presented in this report provides a valuable basis for re-emphasizing the concept of "graded response", and should be considered as supplementing the insights contained in the original planning documents. It must also be emphasized that the dose calculations presented f;arein rely upon projected radioactivity releases ("source terms") essentially the same as those from the 1975 Reactor Safety Study, WASH-1400, and do not reflect the considerable research ef fort which is underway, but presently incomplete, to reassess such potential accidental releases. Hence, this report contains no new source term information. If the "souNtr term" research effort leads to a technically defensible reduction of accidhnt source terms, as many believe it will, it will make some of the accident consequences discussed herein more pessimistic or conservative than warrantedi When such new source term informa-tion becomes available, consideration will be given to revision of this report, b if the results so warrant. t Frank P. Gillespie, Director Division of Risk Analysis and Operations Office of Nuclear Regulatory Research t 1 j { i l 1 l l 1 I g._. 9 N M "*N -

u._, s ABSTRACT Additional risk studies performed since the current NRC emergency planning regulations were adopted in 1980 were examined.to ascertain the individual variation of risk with distance and time and ttts potential need for protective actions within and beyond the 10-mile plume exposure emergency planning zone (EPZ). These studies used accident source terms essentially the same as those of the Reactor Safety Study since recent research on source term re-assessment is presently incomplete. Timing of release of severe accident sequences as well as time-dose-distance relationships were examined to assess the potential degree of urgency of emergency response, and hence the relative priority required in planning. The results show that both individual risk as well as l the pttential decree of urgency is highest closest to the reactor, declines rapidly within about the first two miles, and then declines more slowly with j increasing distances. The risk does not become zero at 10 miles, however. A graded or phased protective action response scheme which emphasizes priority for those members of the public at greatest risk is proposed and evaluated as a flexible strategy capable of dealing with a spectrum of accidental releases. I i 4 l 1 l. D ! y t 1 l I I i 5 e l, t iii ? i 1 1

m._ +. ls* Table of Cont,ents t ?.39% I t i i Foreword........................................................... i 111 Abstract.......................................................... v l Table of Contents.................................................. vif List of Figures.................................................... ix I List of Tables..................................................... 1 xi l S umma ry a nd C o n c l u s i o n s............................................ 5-

1. 0 Introduction..................................................

1 2.0 Accident Source Terms......................................... 3 3.0 Risk Profiles Within the 10 Mile Plume Exposure 7 i Emergency Planning Zone and Beyond............................ l 3.1 Dose Variation vs. Distance.............................. 7 3.2 Variation in the Risk of Individual Health Effects....... 11 I 3.2.1 General.......................................... 11 3.2.2 Early Fatality.................................... 11 14 3.2.3 Early Injury...................................... 3.2.4 Latent Cancer Fatality............................ 17 3.2.5 Total Latent Health Effects....................... 17 b 3.3 Variation in Accident Timing vs. Distance................ 17 t-3.3.1 Introduction...................................... 17 j 3.3.2 Release Delay Time of Severe Accident Sequences... 17 3.3.3 Time Dose Distance Relationships.................. 19 3.3.4 Conclusions....................................... 28 30 4.0 Protective Action Strategies.................................. 4.1 Selection of a Protective Action Strategy................ 30 4.2 Probabilities of Consequences for Various Protective 36 i Action Strategies........................................ 4 i 4.2.1 Introduction...................................... 36 1 4.2.2 Details of Protective Action Assumptions and Consequened Calculations....................... 36 4.2.3 Results and Giscussion............................ 38 i i i I 59 l 5.0 References.................................................... 1it I I V e m ww ggpee _ __ I

'm List of Ficures Pa21 i 3.1-1 Average plume centerline whole body dose vs. distance, given an SST1 release (no protective action, 1 day ground 8 exposure).................................................. 3.1-2 Average plume centerline whole body dose vs. distance, j given an SST2 release (no protective action, 1 day ground exposure).................................................. 9 3.1-3 Average plume centerline whole body dose vs. distance, j given an SST3 release (no protective action, 1 day ground exposure).................................................. 10 j 3.2-1 Individual risk of early fatality vs. distance given an i SST1 release............................................... 12 3.2-2 Individual risk of early fatality vs. distance given an SST2 release............................................... 13 i 3.2-3 Individual risk of early injury vs. distance given an SST1 release............................................... 16 3.2-4 Individual risk of early injury vs. distance given an I SST2 release............................................... 18 3.3-1 Time-dose-distance relationship given SST-1 release and " 0" s tabil i ty wi th 10 mph wi nd............................. 23 3.3-2 Time-dose-distance relationship given SST-1 release and "F" stability with 5 mph wind.............................. 24 3.3-3 Time-dose-distance relationship given SST-2 release and "D" stability with 10 mph wind............................. 25 3.3-4 Time-dose-distance relationship given SST-2 release and "F" stability with 5 mph wind.............................. 26 3.3-5 Time-dose-distance relationship given $5T-3 release and "F" stability with 5 mph wind.............................. 27 h 4.2-1 Probability of an individual at a given distance receiving t a whole body dose in exc u s of the value shown, conditional l upon occurrence of an SSTA release and shelter plus 4 hrs. 40 and exposure............................................... 4.2-2 Probability of an individual at a given distance receiving a whole body dose in excess of the value shown, conditional upon occurrence of an SST1 release and evacuation at 2 mph i 41 j at time = 0................................................ l 4.2-3 Probability of an individual at a given distance receiving a thyroid dose in excess of the value shown, j conditional upon occurrence of an SST1 release and ) shelter plus 4 hrs, ground exposure........................ 4.2-4 Protability of an individual at a given distance .l receiving a thyroid dose in excess of the value shown, conditional upon occurrence of an SST1 release and evacuation at 2 mph at time of release..................... 4.2-5 Probability of an individual at 3 miles receiving a whole body dose in excess of the value shown, conditional upon occurrence of an SST1 release and 47 i l shelter plus I hr. ground exposure......................... f vii .4 1

_s f e List of Figures (cont.) f*1t 4.2-6 Probability of an individual at 3 miles receiving a thyroid dose in excess of the value shown, conditional upon occurrence of an SST1 release and shelter plus I hr. 48 ground exposure............................................ l 4.2-7 Probability of an individual at a given distance receiving a whole body dose in excess of the value shown, conditional upon occurrence of an SST2 release 49 and shelter plus 4 hrs. ground exposure.................... 4.2-8 Probability of an individual at a given distance receiving a thyroid dose in excess of the value shown, conditional upon occurrence of an SST-2 release and 50 i shelter plus 4 hrs, ground exposure........................ j Probability of an individual at a given distance 4.2-9 receiving a whole body dose in excess of the value 4 shown, conditional upon occurrence of an SST2 release 53 and shelter pl us 8 hrs, ground expost're.................... 4.2-10 Probability of an individual at a given distance receiving a thyroid dose in excess of the value shown, conditional upon occurrence of an SST2 release 54 and shelter plus 8 hrs ground exposure.................... 4.2-11 Probability of an individual at a given distance receiving a whole body dose in excess of the value shown, conditional upon occurrence of an SST2 release 55 and shelter plus 12 hrs, ground exposure................... 4.2-12 Probability of an individual at a given distance receiving a thyroid dose in excess of the value shown, conditional upon occurrence of an SST2 release 56 b and shelter plus 12 hrs. ground exposure................... t: 4.2-13 Probability of an individual at a given distance receiving a whole body dose in excess of the value shown, conditional upon cccurrence of an SST3 release 57 and shelter plus 12 hrs, ground exposure................... 4.2-14 Probability of an individual at a given distance receiving a thyroid dose in excess of the value shown, i l conditional upon occurrence of an SST3 release and 58 j shelter plus 12 hrs ground exposure....................... i I vitt i g

w,........ ~. -.-.~. l LIST OF TABLES 5 4 1 2.1 Accident categories covered by the NRC siting source j terms......................................................... 4 2.2 NRC source terms for siting analysis.......................... 5 3.2 Sensitivity of fatal and injury distances to release magnitude..................................................... 15 3.3-1 Timing of severe releases for Oconee.......................... 20 3.3-2 Timing of severe releases for Calvert Cliffs.................. 20 3.3-3 Timing of severe releases for Grand Gulf...................... 21 3.3-4 Timing of severe releases for Sequoyah........................ 21 4.1 Recommended protective action strategies and rationale........ 32-35 4.2-1 Protection factors used in the calculations................... 37 4.2-2 Key individual whole body dose distribution values, SST-1 and relocation after 4 hours...................................... 42 4.2-3 Key individual whole body distributions values, SST-1 and evacuation at time of release (2 mph)......................... 42 4.2-4 Key individual thyroid dose distribution values, SST-1 and relocation after 4 Hours...................................... 4.2-5 Key individual thyroid dose distribution values, SST-1 and evacuation at time of release (2 mph)......................... 4.2-6 Key individual whole body dose distribution values, SST-2 and relocation after 4 hours ground exposure...................... 51 4.2-7 Key individual thyroid dose distribution values, SST-2 and relocation af ter 4 hours plus 4 hours ground exposure......... 51 b t: 4 I j 1 [ ix t i .i w w y

.,.u. t j.- 1,, s Summary and Conclusions p t A re-assessment of accident risk within the 10 mile plume exposure emergency t planning zone (EPZ) (NUREG-03961, 10 CFR Part 503, NUREG-06544) and the protec-i 7 l tive actions to be taken within this zone has been performed. This re-assessment ~ has been based upon extensive additional risk studies that were not available t. at the time the NRC emergency planning regulations were promulgated in 1980. i (See NUREG-07735, NUREG/CR-16598, NUREG/CR-22397, NUREG/CR-2326s, NUREG/CR-2925' i recent Final Enviconmental Impact Statements for nuclear power plant operating licenses issued since June 1980 (See NUREG-0974, for example) and testimony l before the Commission on reactor risk such as that concerr.ing the Indian Point l { plant in February, 1983). The re-assessment has made use of analyses which have considered a full spectrum j of potential accidents, up to and including a range of core-melt events. g Among these were very severe releases resulting from a core-melt and an early breach of containment directly to the atmosphere. The release characteristics used to describe the range of accidents considered have made use of the "Siting Source Term" (SST) terminology (SST1, SST2, etc.) described in NUREG-07735 and the Sandia siting study (NUREG/CR-2239)7 It must be emphasized that the j re-assessment in this report is based solely upon these existing source terms, 3 g l essentially those of WASH-1400 (NUREG/75/014)2 Although an intensive research effort is underway to re-assess the potential magnitude and timing of accidental l radioactivity releases ("source terms"), this effort is presently incomplete. If the "source term" research effort leads to a technically defensible reduction 4 of accident source terms, as many believe it will, it will make some of the 1 accident consequences discussed herein more pessimistic or conservative than 1 warranted. ? A major objective of this assessment was to determine a protective action strategy capable of dealing with a wide spectrum of accidents. The basic radiation protection objectives associated with such a strategy must be flexible, ~ depending upon the nature of the accident, and must provide a priority ranking of desired objectives, rather than providing a single pre-selected fixed risk 4 objective, or dose criterion, regardless of accident severity. A set of such t-objectives has recently been stated in succinct fashion by the ICRPM. These establish differing priorities in establishing strategies to prevent non-stochastic health effects, i.e., those that can appear in identifiable t individuals above a threshold dose such as early f atalities or early injuries, j i j l and reduce stochastic health effects, i.e., those that exhibit no threshold effect and which can be expected to appear randomly among the exposed population, 3 j These objectives are: (a) "Serious non-stochastic effects should be avoided by the introduction of countermeasures to limit individual dose to levels below the thresholds for these effects. (b) The risk from stochastic effects should be limited by introducing counter-y j I measures which acniese a positive net benefit to the individuals involved. (c),The overall incidence of stochastic effects should be limited as far as reasonably practicable by reducing the collective dose equivalent." i i' r.1 l l \\ I'mu u- -u u num u -ilm e -- u.i .I'=.i, lr l

t._, t The reassessment examined risk as well as dose variation with distance, accident severity, timing considerations (from sequence initiation to estimated time of release), and time after release in which a given dose would be received at a given distance (time-dose-distance relationship). Both individual risk and dose are highest close to the reactor, decline rapidly at first, and then more slowly with increasing distance. At a distance of two miler the average radiation dose from a given release would be about 20 percent of its value at one-half mile. This behavior is the result, primarily, of increased atmospheric dilution or dispersion with distance which is inherent in nature and which reduces the atmospheric concentration, and hence the dose, of any released radioactivity. While the individual dose from a given release would be lower by about a factor of 50 at a distance of ten miles vs. that at a distance of one-half mile (the nearest boundary of typical LWR exlusion areas), the effects become magnified for those health effects where a high threshold dose is required before an effect is observed, i.e. early injuries and fatalities. Hence, for the most severe release without protective actions, the individual risk of early fatality would be reduced by about a factor of more than one hundred at a distance of ten miles vs. its initial level at i one-half mile. Because the risk decreases continuously with distance, however, they do not become zero at a distance of ten miles. Accident timing considerations were also reviewed, based upon additional risk studies performed since WASH-1400 (NUREG U 5/014)2 These indicate that for the most severe releases (SST1), the estimated time from accident sequence initiation to radioactivity release ranges from values significantly less than one hour to times on the order of one day, with times of 2 to 3 hours being typical. These studies also showed that the conditional probability of a severe release sequence being a fast-developing one, that is, where the time from initiaWon to release would be less than 2 hours, ranges from 3 to 30 percent, with a typical value of 15 percent, i Risk studies were also perforsed to assess the time after a release begins for b an individual at a given distance to receive a given dose. Such parametric t } curves, displayed as time-dose-distance relationships, provide insight as to l the relative degree of urgency for response at different distances. For the i most severe core-melt releases (SST1) under adverse meteorological conditions, an individual at a distance of 1 mile would receive a potentially life-threat-ening dose of 200 rem to the whole body about one-half hour after the beginning of the release. In contrast, individuals located 5 and 10 miles away, respec-tively, would require exposure times of 3 and 10 hours, respectively, to I receive the same dose. For less severe releases (SST-2) consisting primarily of noble gases and under adverse meteorological conditions, doses of 50 rem to ? the whole body (taken as the threshold value for early injury) would be received by an individual at 1 mile in about I hour after release, while an individual i l at 4 miles would require 8 to 10 hours to receive the same dose. These results show that individuals at close-in distances within the plume EPZ not only would receive higher doses, but that they would do so within a shorter period of time, as well. i 1 I Based upon these results, the major conclusion emerging from this study is that the 10-mile plume EPZ represents a region not only where risk varies significantly in magnitude with distance but where the need for protective 6 xii I -@%-h_.. g

a o, actions at close-in distances would be a greater degree of urgency (given a v declaration of a General Emergency) as well, bereuse of accident timing consid-erations. Consequently, emergency planning as w.;i as response strategies should appropriately reflect such a variation in the magnitude and timing of risk with distance. I 1 It is recommended, based on these risk considerations that emergency planning and protective actions to be taken should be graded or phased within the EPZ with respect to both distance and timing. I i Since core-melt releases ("source terms") can vary significantly in severity, but the magnitude of a given release cannot be readily predicted prior to its j actual occurrence, primary attention or priority should be given to the early initiation of pre planned actions in the event of a core melt accident to reduce the risk within the highest risk portion of the EPZ. This is judged to be that region within about the first 2 miles of an LWR, This difficulty in determining the ultimate severity of an accident once plant l conditions have gone so far beyond the design basis that core degradation is indicated, plus the possibility that this highly unlikely accident sequence 3 I may also be fast developing, led to a primary recommendation in NUREG-0654% i This recommendation was, and is, that nuclear power plant emergency plans l develop predetermined indicators of core degradation precursors, called Emer-I gency Action Levels (EAL's). When these are exceeded, a General Emergency should be declared. This declaration, in turn, should initiate preplanned emergency actions, The preplanned action of choice should be a precautionary evacuation of the j highest risk region to be commenced prior to any plant release. The evacuation of the first two miles should be planned with the objective of being accomplished within about 2 hours. The capability to carry out additional protective actions, including evacuation, D as necessary, should be retained throughout the remainder of the plume EPZ, s t but should receive lower priority. To enhance this capability, members of the public located beyond 2 miles should be advised to seek available shelter and to stay informed of additional developments. It should also be recognized 1 that protective actions might be warranted at distances beyond ten miles, in t unusual circumstances. At the time of an accident, decisionmakers should, of course, take into consid-eration acttal local conditions that may make an evacuation temporarily infeasible, such as impassable road conditions due to adverse weather, and that might present '4 greater risk to the general public, if attempted, than radiological i conditions warrant. Under such conditions, authorities should encourage sheltering and should also consider the auxiliary benefits provided by ad hoc respiratory protective measures. I A distance of about 2 miles has significance for the following reasons: 'For many core-melt accidents (those without a direct release to the atmosphere), projected doses would not exceed the Protective Action Guide 1 (PAG)" levels beyond this distance; xiii I,

b...

-....-n...-n.. o' For most core-melt accidents (those without early failure of containment), projected doses would be unlikely to result in early injuries beyond this distance; Early precautionary evacuation by people within this distance, and immediate sheltering beyond, would provide a large reduction in risk to the highest risk group. It would also provide additional time to assess the developing situation further and to decide upon and implement any required protective actions beyond this distance. It should be specifically noted that an early evacuation within two miles would not be sufficient to avoid life-threatening doses beyond this distance under adverse meteorological dispersal conditions, for the most severe core-melt accidents (those involving prompt failure of containment) and that actions beyond two miles should be necessary in such cares. During such an event it is expected that emergency planning authorities would recommend immediate evacuation, from shelter, of people within about rive miles in a downwind direction even though a release were taking place. A time frame of about 2 hours has significance for the following reasons: Since about.'% to 2% of all core-melt accidents (or 10% to 20% of the most @nces) are estimated to result in severe releases that will also be severe fast-developing (warning time less than 2 hours), an evacuation time of 2 hours for individuals within the first two miles can provide 4 high, but not absolute, degree of assurance that life-threatening doses would not be received in the event of a core-melt. For this reason, the recommended planning goal I of 2 hours should not be construed to mean that individuals at risk could not be expected to eva Eate more quickly or that they should not be encouraged to do so. In other words, it is strongly recommended that the 2 hour evacuation goal be used as criteria for evaluating the level of preparedness only, and should not be used during an actual accident, as that might lead to an unwar-ranted lack of urgency, b For many core-melt accidents, response times well in excess of this value g wouldbeavailablebeforeprojecteddoseswouldbeexpectedtoexceedEPA Protective Action Guide (PAG) 1 levels. j i For most core-melt accidents, an evacuation within this time would avoid j doses capable of producing early injuries in the population. l For the worst core-melt accidents, warning time (prior to releases) of 2 hours or more are predicted for most (about 80% to 90%) severe accident sequences. Based upon the above results of the risk studies, a two-step protective action strategy is proposed as follows: 4 IN THE EVENT OF A PLANT CONDITION WHERE A MOLTF.N OR DEGRADED CORE CONDITION IS OBSERVED OR PROJECTED, OR WHERE ADEQUAi. CORE COOLING CANNOT BE REASONABLY ) ASSU, RED, OR UPON THE DECLARATION OF GENERAL EMERGENCY: 4 1. . REC 0 MEND I MEDIATE PRECAUTIONARY EVACUATION BY EVERYONE WITHIN ABOUT I TWO MILES. THIS EVALUATION TO BE ACCOMPLISHED WITHIN ABOUT TWO HOURS OR LESS WHERE AT ALL FEASIBLE. i xiv 5 i ~

U = _ EVERYONE WITHIN THE REMAINDER OF THE TEN MILE EPZ SHOULD BE ADVISED TO SEEK AVAILABLE SHELTER AND REMAIN INDOORS UNTIL FURTHER NOTICE. 2. ACCIDENT ASSESSMENT SH6ULD CONTINUE, WITH HONITORING OF BOTH PLANT AND FIELD CONDITIONS. AS THE ASSESSMENT CLARIFIES THE SITUATION, FURTHER ACTIONS SHOULD BE TAKEN, SUCH AS INFORMING THE SHELTERED POPULATION PROMPTLY AS TO THE NEED FOR CONTINUING TO SHELTER OR RECOMMENDING THAT THEY PREPARE TO EVACUATE. 4 The potential benefits of this strategy was evaluated by examining individual consequence estimates (as measured in whole body and thyroid doses) for a spectrum of postulated core melt events. For most core-melt releases, an early evacuation out to 2 miles and sheltering beyond this distance (in a one or two story wood-frame house without a basement) followed by a relocation of people before they accumulate the equivalent of 4 hours of ground exposure results ir, no early fatalities and very low risk of early injuries or long term health effects. For the most severe releases postulated, an early evacuation 1 out to 2 miles, followed shortly (within about I hour after an actual major release) by evacuation out to 5 miles in the downwind sectors only, with shel- 'I tering elsewhere ano reloaation from contaminated areas within 4 hours of ground exposure, would n:Jid early fatalities and provide a low risk of early I injuries or long-term health effects. A distinction should be noted throughout this report in the discussion of evacuation stategies reconnended oy the authors, between the early precautionary evacuation of the first two miles and the best tvacation strategy to be followed when an actual severe release is expected or in progress. The first, early, precautionary evacuation snou H be an automatic, preplanned response to the severely degraded plant conditions that wou!d tr.gger declaration of a General Emergency. The second, subsequent strategy that might involve evacuation should be followed only after the plant situation has clarified, and as a more accurate assessment of the potential magnitude of the threat can be made. The first strategy'would provide for a major reduction in individual risks of severe non-ctochntic and stochastic health effects. It also would clear the b way for the'seco6d phase of pr9 planned but more flexible protective action g decisions by authorities on the scene that consider actual conditions at the time. ~ f It is concluded that the proposed strategy can be very effective in meeting I the basic radiation protection objectives and is flexible enough to accommodate a complete spectrum of core-melt eventi. It can be effective because a) those j at greatest risk would be given the most immediata (early) attention, b) emergency response resources can beysed'most effectively by concentrating i response in a.greded, or phased manner and c) its simplicity can lead to a 4 high confidence that it can 0; well understood and that it would work. }k 4 i I. 4 e i f xv 5 3 gj I Y,J. - _ J-

n,,, _ s. 1.0 Introduction The NRC Emergency Planning regulations were significantly upgraded in 1980, largely as a result of the Three Mile Island accident. These regulations required that emergency planning be instituted around every nuclear power plant in the U.S. for a plume exposure pathway planning zone having a radius of about ten miles, and an ingestion exposure pathway p14nning zone, having a radius of about 50 miles (10 CFR Part 50, Appendix E)3 The bases for these recommendations we e given in a report by a joint NRC/ EPA Task Force entitled "Planning Basis For the Development of State and Local Government Radiological Emergency Response Plans in Support of Light Water Reactor Nuclear Power Plants", NUREG-0396, published December,1978.1 Further elaboration was also published in "Criteria for Preparation and Evaluation of Radiological Emer-gency Response Plans and Preparedness in Support of Nuclear Power Plants", 3 NUREG-0654/ FEMA-REP-1, Rev. 1, published November, 1980.4 Theroleofemergncyplanninghasbecomeincreasinglycontroversialsincethe accident at TMI. In practice, virtually all of the controversy has centered around the 10 mile plume exposure pathway zone (EPZ), and may have arisen largely because of misperceptions regarding the nature of accident risk within this zone. These misperceptions, simply stated, are that: (1) All individuals within the 10 mile plume EPZ, regardless of their loca-tion, are exposed to an essentially equal degree of risk, and (2) An expeditious evacuation of this entire area is the only effective protective action to be planned or carried out. 21 that a 10 mile planning radius is simply In addition, some have claimed inadequate and should be extended outwards, oerhaps to 20 miles. D It is clear from a perusal of the reference documents leading to the regula-tion that there wrs a clear awareness of the significant spatial variation in I risk (see NUREG-03961, page 118, for example), as well as an awareness (see NUREG-06544, page 9, for example), that actions should be concentrated first j in the near or close-in regions and extended outward, as necessary. There was i also a clear recognition that protective actions might be required beyond 10 l miles, in unusual circumstances (see NUREG-03961, pp. 16 and I45, for example). Since the Emergency Planning regulations have been promulgated, a number of addi-j tional risk studies and licensing assessments have been performed.5'S'7's.s,to 1 j As a result, it was felt that a re-assessment of the nature and timing of j accident risk within the plume EPZ, using the results of these recent studies, ] might provide additional insight for emergency planning. ] lt should be emphasized that only existing accident source inms, essentially { the same as those used in the Reactor Safety Study, WASH-140J,2 have been J employed in this study. Although an intense research effort is underway to l re-a,ssess these accidental radioactivit/ releases, this effort is presently 1 j Q" W - =

incomplete. If the "source term" research effort leads to a technically defensible reduction of accident source terms, as many believe it will, it will make some of the accident consequences discussed in this report unduly j j pessimistic and more conservative than warranted. t In the sections that follow, this report discusses risk as well as dose varia-tion with distance, projected accident timing considerations from sequence initiation to estimated time of release, and the time-dose-distance relation-ships that would exist after a given release. From these observations, a j, proposed protective action strategy is derived.and examined which is intended to place priority where risk is highest while still pro.iding for a flexible j response to a wide range of possible accidents. This is referred to as the concept of "graded" response. It should be emphasized that the material and conclusions presented in this report should be viewed as supplementing, and not replacing, those given in NUREG-03961 and NUREG-06544 i i r l 'D [ i i f n -1 2 1 1 1

w a L._.h .:. w .o 3 2.0 Accident Source Terms Since a major objective of this report is to examine and assess individual accidentriskwithinthe10-mileplumeEPZinscaedetail,asuitablegtarting j point for examining such accident risk is needed. Although NUREG-0396 examined risks from both design-basis as well as severe accidents, this study will focus only upon potentially severe accidents since the consequences of such events, although less probable than the design-basis accidents, are generaly conceded to dominate the risk as well as to have a "reach" which extends a greater distance. l The starting point for assessing consequences from severe accident sequences j are accident "source terms." These comprise a set of values representing the i possible quantities and types of radioactive materials that could be released j to the environment from a sequence or group of similar possible accident sequences. Tabulations of source terms, in addition to the data mentioned I above, may also contain additional information such as the probability of the j release, the estimated warning time prior to release, the release duration and the energy content. Sets of source terms for severe accident sequences were first given in the j Reactor Safety Study, WASH-14002, which provided a tabulation of severe acci-l dent releases ranging from PWR 1 to PWR 9 for a Pressur' zed Water Reactor (Surry) and ranging from BWR 1 to BWR 5 for a Boiling Water Reactor (Peach 1 Bottom). As additional Probabilistic Risk Assessment (PRA) studies were i performed after WASH-1400, additional plants and severe accident sequences ] were investigated. These developments"are discussed in more detail in NUREG-0773s which also detailed a suggested generic set of source terms that would be suitable for providing insight for siting and emergency planning studies. These have been referred to as the "siting source terms" (SST), and b range from the very severe SST-1 release to the relatively benign SST 5. t' Table 2.1 provides a delineation and qualitative distinction between these 1 releases, while Table 2.2 provides data on the quantities and types of l isotopes released as well as the release duration. The siting source terms (SST's) were applied in the Sandia Siting Study (NUREG/CR-2239)7 to examine the consequences of such releases at each of r' U.S. sites. The insights gained from this study are of value for emergen,j planning, as well. Accident source terms in the form of the siting source terms (SST) were used i for this study both because they incorporate the research effort on additional plants and postulated accident sequences beyond those of WASH-1400. The SST grouping are somewhat easier to conceptualize than those of WASH-1400. At the same time, the reader should bear in mind that the SST's are highly synthetic and that different severe accident sequences, each of which might be reason-ably categorized as leading to an SST 1 release, for example, might neverthe-less. differ in some degree from the values in Table 2.2. It should be noted that the postulated consequence models in this report are conditional upon a o I 3 I li -~ " ^ ' - - -e ---2

f....r..

Table 2.1. Accident categories

• covered by the NRC siting source terms i

} l SST1 Severe, early direct breach of containment. Core Melt. Essentially 3 involves loss of all installed safety features. Propulsion of large fractio 1s of the core inventory of radionuclides into the atmosphere. Similar to PWR-2 of the Reactor Safety Study. SST2 Containment fails to isolate. Core Melt. Containment fission product release mitigating systems (e.g., sprays, suppression pool, fan coolers) operate to reduce the release. Small fractions l of the core inventory released to the atmosphere. Similar to PWR-5 of the Reactor Safety Study. SST3 Containment fails to isolate by late basemat melt-through. Core Melt. All other :elease mitigation systems function as designed. j Very small fractions of the core inventory leak into the atmos-phere. Long-term leakage of radionuclides into the hydrosphere. SST4 Containment systems operate but in a somewhat degraded mode. Limited to moderate core damage. Containment is not breached. 3 Similar to PWR-9 of the Reactor Safety Study. I SSTS Containment functions as designed. No failures of engineered l safety features beyong those postulated pursuant to 10 CFR f Part 100 (siting criteria) are assumed. The most severe acci-j dent in this group includes substantial core melt.

• The accident at THI-2 March 28, 1979 does not fit neatly into any of these categories. Although severe core damage was experienced, such as in SST3, b

the operation of the containment safety features severely restricted t. releases to the environment. Offsite consequences falls into the SST4 or SSTS category. I i i t i d a 4 t

- s.-. Aw .a v.~. -1+ --.-_ _L -...._. ~)' 2' ] > . n. cr -. f. Table 2.2 NRC Source Terms for Siting Analysis i i. Reisase. Characteristics

• Source Term SST1 SST2 SST3 SST4 SSTS Accident Type Core Melt Core Melt Core Melt Gap Release Gap Release l~

Containment Failure Mode Overpressure H Explosion i 2 E or Loss of Isolation Containment Leakage Large targe EE/ day 1%/ day 0.EK/ day Time of Release (hr) 1.5 3 1 0.5 0.5 l-Release Duration (hr) 2 2 4 1 1 Warning Time (hr) 0.5 1 0.5 Release Height (meters) 10 10 10 10 10 Release Energy 0 0 0 0 0 Inventory Release Fractions Xe-Kr Group

1. 0 0.9 6 x 10 3 3 x 10 s 3 x 10 7 3 x 10 3 2 x 10 4 1 x 10 7 1 x 10 s I Group 0.45,

Cs-Rb Group 0.67 9 x 10 3 1 x 10 5 6 x 10 7 6 x 10 s I Te-Sb Group 0.64 3 x 10 2 2 x 10 5 1 x 10 8 1 x 10 13~ ( 4 l Ba-Sr Group 0.07 1 x 10 3 1 x 10 8 1 x 1011 1 x 10 12 i t Ru Group 0.05 2 x 10 8 2 x 10 8 0 0 La Group 9 x 10 3 3 x 10 4 1 x 10 s 0 0 - a t t 3 t As defined in the Reactor Safety Study [2]. 't a. l

severe accident occurring. This is a highly unlikely event. Core melt events are estimated to have a probability of occurrence of once in every 10,000 - reactor years. A core melt resulting in a severe release (SST-1) would be even more unlikely with a probability estimated as once in every 100,000 reactor years. Only the accidents involving core melt releases, i.e., SST-1, SST-2 and SST-3, are discussed in this report as they are the only ones with the potential for severe consequences. The probability of a core melt resulting in a relatively benign SST-3 release is estimated to be about 70%. About 20% are estimated to result in SST-2 releases and 10% or less would be estimated to be the most severe SST-1 variety. As a practical matter, it may be difficult or even impossible for a reactor operator to predict releases for an accident that has gone well beyond the design basis of the plant, particularly during the early stages of the accident. It is the staff intent that nuclear power plant emergency plans require that the operator declare a "General Radiological This Emergency" at that point rather than waiting more precise assessment. prudent declaration would activate immediate emergency response procedures that would minimize possible consequences of even the most severe accident I category. This has important implications for selection of an appropriate protective action strategy, as will be discussed later in this report. Finally, it must also be emphasized that the accident source terms used in this study do not reflect the considerable research effort which is underway, but presently incomplete, to reassess the potential for such accidental releases. If such a source term research effort leads to the conclusion that severe accident source terms have generally been overestimated, it will make some of the accident consequences (in particular, those for the SST1 release) more pessimistic or conservative than warranted. b Y i 6 i ED O e 4 - M - _

• • N8"***'*"N***-

' N*'S ply M M 9ge . - - ~.,, - -,, - +,,,. -,,., - -, - - -., -.. r

I i 3.0 Individual Risk Profile Within the 10-Mile Plume Exposure EPZ and Beyond This section discusses the magnitude and variation of individual (conditional) risk that exists within the 10-mile plume exposure Emergency Planning Zone l (EPZ), as well as beyond it. Section 3.1 discusses the variation of dose as a function of distance. Sections 3.2 and 3.3 discuss the variation of health j effects and time-dose-distance relationships, respectively. It is emphasized { that only risks conditional on accidents occurring will be discussed. Abso-lute risks are not discussed or estimated. 3.1 Dose Variation with Distance 1 j The variation of projected dose with distance, given a severe accidental I release, has been analyzed and reported in NUREG/CR-22397 Reference 7 dis-played the mean whole body dose vs. distance to an individual,* given an SST1, SST2 or SST3 release, on a log-log plot. For clarity these have been replotted on a linear scale, as Figures 3.1-1, 3.1-2 and 3.1-3. Note that f these projected doses assume no emergency response. Although the absolute doses from these three different release categories differ by about a factor of one thousand, they all display the same general variation with distance; in fact, the natural effects of atmospheric dilution and dispersion would make the effects of almost any release vary in the same relative way. From an initial value at a distance of 0.5 mile (taken as the typ' cal exclu-1 sion boundary distance), these curves show that the individual dose drops off sharply with distance at first, then the slope continues to decrease more slowly. At a distance of two miles, for example, the dose is about 20% of the value at 0.5 mile, while at a distance of ten miles, the mean whole body dose

i is about 2% of its initial value.

This illustrates the important point that, although the projected dose varies continuously with distance, most of the decrease over the 10 mile plume exposure emergancy planning zone (EPA) occurs very close-in to the reactor, with about 80% of the decrease from 0.5 mile to g 10 miles occurring within the first 1.5 miles. It is also important to recognize that projected dose levels may still be 'l significant in absolute terms at distances of 10 miles and beyond, even though j there has been a large decrease with distance. For example, at a distance of J 10 miles the whole body doses for an SST release is about equal to the pro-posed Environmental Protection Agency (EPA) Protective Action Guide (PAG)11 1 1

levels, i.e., 1 to 5 rem whole body and 5 to 25 rem thyroid.

For an SST1 } release, the while body dose is in excess of the upper PAG levels out to about 1 40 miles. This was well understood at the time the 10 mile EPA was proposed in NUREG-0356 - l

• These doses apply to an individual directly downwind in the plume centerline.

J i ? l I j 7 l 1

1 --1. ) . re. o* I a - t 1 I I I I I I I I 8000. - i I, i. s000. 2 k m oc 1 i N 4000. l, o o k m O 3000. r I a O { l I i 3 1, 2000. 1' 1000.- f I I I l I I i l 0 1 2 3 4 5 6 7 8 9 10 L i I I DISTANCE, MILES Average Plume Centerline While Body Dose vs. Distance, Given Figure 3.1-1 an SST1 Pelease (No Protective Action, 1 Day Ground Exposure)

. _ u _.u._. -

• r* cr s

i i M~ l I I I I I I I I l \\ l 200 i l 180 l i i 1 l I 160 2 l i 140 l m m U) 8 120 I oo 100 m I i w o m 4 13 l n t a i 1 t 20 t i I I I I I l I i i O 1 2 3 4 5 6 7 8 9 10 DISTANCE. MILES Average Plume Centerline Whole Body Dose vs. Distance, Given Figure 3.1-2 an SST2 Release (No Protective Action,1 Day Ground Exposure) l

.a c. u.~

---...uw..-u__.

e, . r* 7 l e i . L ] 1.4 g g g l l l l l l l 1.2 - 1.0 2 m II" N 0.8 oo e* o a f O

• 0.6 3

0 I I 3 I 0.4 t i ) f. 6 I I r 0.2 i F t i I I I I I I I I O 1 2 3 4 5 6 7 8 9 10 DISTANCE. MILES Average Plume Centerline Whole Body Dose vs. Distance, Given an SST3 Release (No Protective Action. 1 Day Ground Exposure) Figure 3.1-3

- - ~. - - -. 3.2 Variation of Health Effects with Distance 3.2.1 General This section discusses the distance variation of some of the major radiation-induced health effects such as those for early fatality, early injury and latent cancer fatality, given various accidental releases. The intent is to examine distance relationships primarily for those non-stochastic effects having the highest priority or urgency for planning and response assumptions. It is not intended to be a complete discussion of all health effects associ-ated with radiation exposure and has knowingly omitted discussion of a number of these, such as genetic defects, sterility, fetal development, or mental retardation in utero which may manifest themselves. In addition, the discus-sion of health effects has also implicitly assumed dose-health effect relation-ships that are generally attributable to a normal, healthy adult population. The reader should be aware that there are also sensitive elements in the population such as pregnant women, the sick and the elderly where such relation-ships may not be applicable. For a further discussion of health effects, the reader is referred to References 2, 13, 22 and 23. I' i It is also important to note that there may exist other adverse effects of accidents which would require later post-accident responses. In particular, t deposition of particulate activity may cause significant land contamination requiring possible decontamination efforts and possibly involving interdiction of buildings and land and impoundment of milk and other crops. Such countermeasures may extend well beyond the ten mile plume exposure EPZ and into the 50-mile ingestion pathway EPZ. Depending upon the severity of the i accident and the countermeasures taken, significant numbers of latent cancer fatalities and other health effects may be predicted to occur as a result of i long-term habitatin of relatively low-level contaminated areas. Increased j efforts, and costs, of decontamination would result in a reduction of such i latent cancer fatalities. Such a cost vs. health-effects trade-off has been discussed in References 2, 7 and 24. These aspects of post accident countermeasures D are not discussed within this report, however, because the societal decisions

i required and the costs they impose vary from case to case and cannot be readily planned prior to an actual release, and because the relatively low level of j

contamination at greater distances would generally ease any urgency requirements j and so permit sufficient time for such decisions to be made on an ad hoc basis. l Ij 3.2.2 Early Fatality The individual risk of early fatality vs. distance, given SST1 and SST2 releases, is shown in Figures 3.2-1 and 3.2-2, respectively*. The risk of early fatality from an SST3 release is zero for all distances, since the doses are not j sufficient to produce this effect. l 1 I i

• Notes:

1) Data taken from Ref. 7 and replotted on linear scale. These effects apply to the average individual at this distance, rather than the maximally j

exposed individual, whoe risk would be about one order of magnitude greater. 1

2) The "No Evacuation" plot assumes that no protective measures are taken.

j

3) All "Evacuation" scenarios assume that this protective measure extends to l

10 miles. I l l 11 i i i l \\ 1 - - - -. ~.

1i 1!, i [ , s} t' ,1; :1l,! 0 1 I 9 I H H H P P P ne M M M v i 0 0 0 G 1 1 1 e T T T 8 c A A A n I a Y Y Y t A A A s L L L i E E E D D D D s R R R I 7 v H H H I y 5 3 1 t i R R R l E E E a T T T t N F F F a F O A A A 6 y IT N N N 1 l A O O O r S U I I I a T T T E E C A A A L A I f U U U V M o E C C C A A A O V V V ? 5 E y t C 1 i N E E E N l ~ A i b T ~ A O O S bea } I os D ra Pe 4 l I l l e aR u d1 iT vS iS dnn r 3 I a t l tr e 1 2 3 2 e l r l u g i F + 1 l 0 2 0 6 6 4 2 1 1 0 0 0 0 ga<64u. bm4** o b'_s<mOen- ~" t1 I 4

~ 0.012 y y y 0.010 Ns 0.008 i 1 J cc4m 0.006 mo Ns iii j 0.004 O b cc t.: ct i j 0.002 l 1 I I i l t l l 0 1 2 3 4 5 l DISTANCE, MILES Figure 3.2-2 Individual Probability of Early Fatality vs. Distance Given an SST2 Release i 13 1 'l ij

e.... j Figure 3.2-1 displays the risk of early fatality for four differing protective actions given an SST1 release. It clear that the risk of early fatality is strongly dependent upon distance, the protective action taken, and the time I evacuation is initiated. Even where no protective action is taken ("no evacua- } tion" case) and an individual is presumed to remain exposed to the plume j during its passage plus 24 hours of exposure from ground contamination, the risk decreases by about a factor of over 100 over the 10 mile EPZ, with most i of the decrease in risk in the first few miles. When evacuation to 10 miles j is included in the calculations, the decrease in acute fatality risk with distance becomes even more pronounced. Prompt evacuation (1 hour delay followed by evacuation at 10 mph) red,uces the probability of early fatalities by a factor of over 4.5 at 0.5 miles and almost eliminates this risk beyond 1.5 miles even for the SST-1 accident case. Figure 3.2-1 also provides valuable insight on the degree of urgency with which protective actions need to be taken at various distances. Because of I increased dispersion as well as travel time, protective actions at shorter distances must be implemented more swif tly to significantly reduce risks. For example, to assure an indivdual risk of early fatality of no greater than 4 0.005, given an SST1 release, an individual at a distance of 1 mile from the reactor needs to evacuate with a delay time of 1 hour, whereas individuals at distances of 2.5 miles and 4 miles can delay evacuation for 3 hours and 5 hours, respectively, with the same degree of risk. Figure 3.2-2 shows the risk of early fatality vs. distance for an SST2 release with no protective actions taken. Clearly, the risk is much lower for an SST2 release and becomes very small beyond about 1 mile. It is also important to recognize that the risk of early fatality can extend beyond 10 miles, in unusual circumstances. Table 3.2 shows that, when no protective actions are taken, early fatalities could occur beyond 8 miles in about 10% of SST1 releases and could occur beyond 12 miles in about 1% of such releases. Realistically, these percentages would be much lower since it is D highly unlikely that no protective actions would be taken, even beyond the plume exposure EPZ, given a severe release at a nuclear power plant in the - l; i U.S. I j 3.2.3 Early Injury Early injuries, which would be manifested by symptoms such as nausea and loss I of appetite, are generally presumed to commence at whole body doses of above } about 50 rem.** The individual risk of early injury as a function of dis-j tance, given an SST1 release, is shown in Figure 3.2-3*. Although the risk for this effect varies both with distance and the protective action taken, reduction with distance"is not as abrupt as for early fatality in Figure 3.2-1. For example, for the no evacuation scenario, the reduction in indivi-l dual early injury risk between 0.5 miles and 2 miles is about a factor of 1.5. For early fatality it is about a factor of 2.5. When protective actions are included, the reduction in risk of early injuries is greater, dropping by a factor of 16 between 0.5 and 2 miles. As in the case of early fatality, at

• Note:

Data taken from Ref. 7 and replotted on linear scale.

• • Note:

For a fuller discussion of health effects, the reader is referred to References 2, 13, 22 and 23. l 14 q M

-L . c.. t. m-e re. C7 Table 3.2 Sensitivity of Fatal and Injury Distances to Release Magnitude NEW YORK CITY METEOROLOGY, 3412 MW(t) PWR, AND NO EMERGENCY RESPONSE ASSUMPTIONS: 3

• SOURCE: NUREG/CR-2239, NOV. 1982 FATAL DISTANCE (MI)

INJURY DISTANCE (MI) ACCIDENT RELEASE

• MEAN 90%

99% MEAN 90% 99% SST1 3.9 8.0 12 11 20 35 /~ 1/2 SST1 2.5 5.0 10 7.0 12 20 1/10 SST1 0.9 2.0 2.2 2.8 5.0 10 j 1/20 SST1 0.5 1.2 2.0 1.9 3.5 7.0 l 1/100 SSTI -0 <1 1.0 0.9 2.0 4.0 lt Ul .1 t i i r

• Release fractions reduced for all isotopes except noble gases.

i I

j(! Ifi. l l lI t 0 W1 9 H H H n P P P e v M M M i G 0 0 0 1 1 1 e T T T 8 cn A A A a t Y Y Y s A A A i L L L D E E E D D D s 7 v R R R I H H H yr 5 3 1 u j R R R n E E E I T T T N F F F y O A A A l 6 r i IT N N N a I A O O O S E E U I I I f T T T L o C A A A I U U U My A V C C C t E i E A A A 5 C l O V V V N b i I N E E E A a T be A O O S os { ra I Pe D l l e I 4 aR u I dI iT vS iS dnn I a 1C h 3 I e r. 3 I -2 3 er 2 ug l i F I 1 y 0 0 s s 2 1 o. O. o. 0 3c3}_ >e5 u.O Nga<*oE a ym ^ iI 1 l l t!l l! I-

• l,I' l

\\ ll

i.... distances less than 2 miles the only protective action strategy that would be highly effective in reducing the risk of early injury is early evacuation. The individual risk of early injury without protective action, given an SST2 release is shown in Figure 3.2-4*. The risk reduction with distance is seen to be very large and occurs primarily within about the first 2 miles. Risk decreases from a value of about 10 2 at a distance of 1 mile, given an SST2 release, to a value of 10 8 at 2 miles, and to about a value of 10 4 at 2.5 miles. 3.2.4 Latent Cancer Fatality The individual risk of latent cancer fatality as a function of distance was also examined. It was determined that the SST1 release dominates the risk of 7 latent cancer fatality being about one order of magnitude greater than that for an SST2 release. The reduction in individual risk between 0.5 miles to 10 ' miles was about a factor of 20, given an SST1 release. From 0.5 to 2 miles the risk reduct~ ion was about a factor of two. Since the studies were only for the case of an evacuation with an average delay time of about three hours, the effect of different protective action strategies was not clear. Unlike the risk of early fatality or early injury, there is no sharp dropoff in risk of latent cancer beyond some distance, but the risk drops off monotonically and smoothly with distance. It is clear, however, that although monitonically decreasing with distance, individual latent cancer fatality risk extends out to large distances. Section 3.3 Variation in Accident Consequence Timing vs. Distance 3.3.1 Introduction 4 This section discusses the relationship of the time of release of radio-activity following an accident in a r'eactor and time-dose-distance relation-b ships after this release has occurred. Discussions combined with graphic t: presentation allows conclusions to be drawn about how these factors affect variation of potential accident consequences and emergency planning. J 3.3.2 Time of Release of Severe Accident Sequences 1 The time from accident sequence initiation to a release f radioactivity is l important because it provides insight into the degree of urgency with which protective actions, such as evacuation or sheltering, need to be initiated to provide an effective response. A major study since the Reactor Safety Study (WASH-1400): which has investi-l gated potential severe accident sequences, including their tilling, is the l Reactor Safety Study Methodology Applications Program (RSSMAP).s This study, using the methodology employed by the Reactor Safety Study, examined risk dominant accident sequences for four specific light-water reactor (LWR) l plants. These plants were Sequoyah Unit 1, a prsssurized water reactor (PWR) with,an ice condenser containment and emmploying a Westinghouse-designec' nuclear steam supply system (NSSS); Oconee Unit 3, a PWR employing a Babcock and Wilcox NSSS in a large dry containment; Calvert Cliffs Unit 2, a PWR with 17 l l gy m es>u4eW e g * *m M -- 48-

J 0.014 [ l l I 0.012 l 0.010 3 t = 3 i- .z d 0.008 <w u. O N3 0.006 Ei< m O oc b Q. t 0.004 4 0.002 l t I i l l l l 0 1 2 3 4 5 DISTANCE, MILES j Figure 3.2-4 Individual Probability of Early Injury vs. Distance Given an SST2 Release 18 . ~

l a Combustion Engineering NSS$ in a large dry containment; and Grand Gulf Unit 1, employing a General Electric designed boiling water reactor (BWR) NSSS and a GE Mark III pressure-suppression containment. The RSSMAP study analyzed the i probabilit of occurrence of key core-melt sequences and also made use of the MARCH and CORRAL codes, developed from the Reactor Safety Study, to examine the timing of these sequences as well as the magnitude and type of radioactive l fission products released. The results of the RSSMAP study have been pub-lished in four volumes as NUREG/CR-16598 from February 1981 to May 1982. Results on the timing of severe releases for each of the four plants are shown in the accompanying tables (3.3-1, 3.3-2, 3.3-3, 3.3-4). The tables list for each plant the probability of core-melt sequences which would lead to a severe radioactivity release (taken as equivalent to an SST1 release, using the nomenclature of NUREG/CR-22397, or a PWR 1, 2, 3 or a BWR 1 or 2 release, using the terminology of WASH-1400)2 Each table also lists the dominant accident sequences for each plant as estimated by reference 7, the estimated annual probability of ocentrance for each sequence, the time after the sequence commences when the release would begin, and the relatve contribution of each sequence to the total probability of a severe release. From these tables, it can be seen that the time when release begins can range from values significantly less than one hour to times on the order of one day with a large fraction of the sequences estimated to have release delay times of about two to three hours. Also of iterest is the fraction of sequences for which the time of release is less than two hours after accident initiation. These sequences are considered to be fast-developing. The fast-developing sequences were estimated to constitute about 3 percent of the total sequences for the Calvert Cliffs plant and range to a high of about 30 percent for the Oconee plant, with the two other plants having values of about 15 percent. Conclusions Estimates of the release delay time for severe accident sequences ranges b from values significantly less than one hour to times on the order of one t' day, with two to three hours considered to be typical. i Estimates of the probability of a severe sequence being a fast-developing one, that is, where the release delay time is less than two hours, ranges from 3 percent to 30 percent, with about 15 percent considered to be typical. 3.3.3 Time-Dose-Distance Relationships In addition to the time when a release of a severe accident sequence begins, discussed in the previous section, one must also assess the rate at which an individual would accumulate a given radiation dose after a release commences, assuming a fixed location and release characteristics; in other words, the time required to receive a specified dose at a given distance. This type of information is often displayed in a curve or a set of curves known as a time-dose-distance relationship since it shows time plotted along the ordinate vs. the observer's distance along the absicissa, with values of constant dose a as a parameter. This information was not available from the published litera-ture'for core-melt releases, and was therefore specifically developed for this study. 19 s y M

• 1 e~*N*P

_- _ = Table 3.3-1 TIMING OF SEVERE RELEASES FOR OCONEE PROBABILITY OF POSTULATED CORE-MELT SEQUENCES GIVING RISE TO SST1 RELEASES CONDITIONAL SEVERE EVENT TIME OF RELEASE RELEASE SEQUENCE PROB /YR (MIN.) PROBABILITY TMKU y 3.9 x 10 8 o 60 min. 0.11 V 4.0 x 10.s 62 min. 0.11 SD y 2.4 x 10 8 '79 min. 0.066 TM00 y 7.5 x 10 7 79 min. 0.020 TMLU y 2.7 x 10 8 202 min. 0.074 TMQ-H y 5.5 x 10 8 > 202 min. 0.15 TMW-FH y 2.5 x 10 8 > 202 min. 0.069 SH y 5.0 x 10 8 > 202 min. 0.137 i SFH y 2.1 x 10.s > 202 min. 0.058 i TMLUD y 7.4 x 10.s > 202 min. 0.20 3.6 x 10.s/R-YR** SOURCE: NUREG/CR-1659 Table 3.3-2 TIMING OF SEVERE RELEASES FOR CALVERT CLIFFS PROBABILITY OF POSTULATED CORE-MELT SEQUENCES GIVING RISE TO SST1 RELEASES CONDITIONAL SEVERE EVENT TIME OF RELEASE RELEASE SEQUENCE PROB /YR (MIN.) PROBABILITY D. t TXML y, 6 4.2 x 10.s < 60 min. 0.028 THQ-D 6, y 2.0 x 10 5 120 min. 0.014 j TMQ-H y, 6 2.3 x 10 5 120 min. 0.016 1 THQ-FH y6 8.4 x 10 5 120 min. 0.057 i TML y, 6 1.21 x 10 8 149 min. 0.82 i TMLOO1 6 1.0 x 10 4 319 min. 0.068 j 1.5 x 10 8/R/YR** SOURCE: NUREG/CR-1659 l 2 f

• Probability of SST-1 release given accident sequence.
  • Summation assumes sequences are independent. 3.6 x 10 5/R-YR = 3.6 events per 100,000 reactor years.

] 1 1 20 j l

.g .a j Table 3.3-3 TIMING OF SEVERE RELEASES FOR GRAND GULF j PROBABILITY OF POSTULATED CORE-MELT SEQUENCES GIVING RISE TO SST1 RELEASES CONDITIONAL l SEVERE f EVENT TIME OF RELEASE RELEASE SEQUENCE PROB /YR (MIN.) PROBABILITY

• j TC-6 5.4 x 10.s 90 min.

0.16 TPQI-6 5.3 x 10 8 1190 min. 0.16 i TQW-6 1.8 x 10 5 1340 min. 0.54 4 SI-6 4.6 x 10 8

• 1400 min.

0.14 3.3 x 10 5/R-YR** -1 i SOURCE: NUREG/CR-1659 Nble3.3-4 TIMING OF SEVERE RELEASES FOR SEQUOYAH PROBABILITY OF POSTULATED CORE-HELT SEQUENCES GIVING RISE TO SST1 RELEASES CONDITIONAL I SEVERE EVENT TIME OF RELEASE RELEASE SEQUENCE PROB /YR (MIN.) PROBABILITY

• V 5 x 10 8 38 min.

0.14 5 H-6 2 x 10 5 110 min. 0.56 2 5: HF-6 5 x 10 8 197 min. 0.14 S HF-6, 6 3 x 10.s 219 min. 0.085 t TML-6 3 x 10.s 238 min. 0.085 4 b 3.6 x 10 5/R-YR** g: SOURCE: NUREG/CR-1659 ~

• Probability of SST-1 release given accident sequence.
  • Summation assumes sequences independent. 3.6 x 10 5/R-YR = 3.6 events per 100,000 years.

t 1 i i 21 I l +

o Sandia National Laboratories was requested to perform an analysis to determine the cumulative whole-body dose vs. distance with the time from the beginning of the release shown as a parameter. This information was generated for each of three core-melt releases, namely, SST1, SST2 and SST3, and for two constant weather conditions for each release: 1) a Pasquill "0" stabilit, with a 10 mph windspeed, which corresponds approximately to a 50 percentile or "average" atmospheric dispersion condition; and 2) : Pasquill "F" stability with a 5 mph windspeed, which corresponds to about the worst to be 10 to 15 percent in terms of atmospheric dispersion. Parametric curves of dose vs. distance were developed for times after release of 0.5, 10., 1.5, 2, 3, 5 and 10 hours. The data were cross plotted to show time from the beginning of the release vs. distance with dose as a parameter. These curves, the time-dose-distance relationships, are presented in Figures 3.3-1 through 3.3-5. Only one curve is shown for the SST3 release, since the reach of this release, in terms of distance, does not extend beyond about 2 miles, even for the lower level whole body PAG (1 rem) and for adverse meteorology. An examination of the time-aose-distance relationships shown in Figures 3.3-1 to 3.3-5 indicate that, for a given dose, each of the curves displays a similar general characteristic in that the time to accumulate that dose increases mono-tonically with distance, with the slope of the curve also increasing with dis-tance. At some particular distance the slope of the curve becomes essentially vertical, The significance of that distance is that, beyond it, it becomes virtually impossible to receive a dose in excess of the given amount, since essentially an infinite time would be required to receive it. With these general characteristics in mind, it becomes possible to reach some conclusions after examination of these figures. From an examination of Figure 3.3-5, for an SST3 release, it is clear that at distances beyond about 2 miles, times well in excess of 12 hours would be required before an individual would receive a dise in excess of the lower D level PAG (1 rem whole body), for the assumed adverse weather conditions. g From Figures 3.3-3 and 3.3-4, for an SST2 release it can be seen that at distances beyond about 1 mile, very long accumulation times would be required i before an individual would receive a life-threatening dose (200 rem or more, { whole body). At distances of about 2 miles and beyond, an exposure time of 2 hours or more from the beginning of the release would be required to produce 4 a dose sufficient to produce early injuries (50 rem or more, whole body). For the most severe core-melt releases (SST1) Figures 3.3-1 and 3.3-2 show j that poeple within 2 miles can receive life-threatening doses within short time periods after the beginning of a release (time periods of I hour or j less). Consequently, individuals in these locations would be at the greatest i degree of risk and urgent protective actions, particularly evacuation, should be commended as early as practicable when a General Emergency is declared to maximize the available time prior to the occurrence of a release. For these most severe core-melt releases, the degree of urgency for people located from 2 to 5 miles away is somewhat eased. Figures 3.3-1 and 3.3-2 show that for average weather conditions, the time required to receive a 22 k i 'e*

...-~_...__.-__u.. s. -__._..m..

• YC. Q*

l I I I I I I I I I I '3 D = 200 Rom 12 D = 100 Rom ASSUMPTIONS D = 50 Rom 11 SST1 Release. ~ us 1120 MWe Reactor m 2 Hr. Release Duration 3 10 Conetont Weather:

• • D" Stability, w

S 10 mph Windspeed vs4 Shielding Factor: "l Norrnal Activity g E Whole Body Dese Shown u. o 7 oz E s z 5 = 5 4 m i 2 o 4 e L = s 3 P 2 D = 5 Rom _ i< j i I i l l I I I I l 0 1 2 3 4 5 6 7 8 9 10 t i DISTANCE. MILES Figure 3.3-1 TIME-DOSE-DISTANCE RELATIONSHIP I t

F .1[>! l;i ll; i i l' i; ~ ~ 0 l l 1 m m m m o o 1 9 I R R oR o 0 0 R 0 0 0 2 1 5 5 = = = = D D D D a i I P 7 I I H I S NO I TAL ER I 6 I E d S CN e E A e L T I p S M e I d D n 5 E -E i I C I W S N O h A D p T E m S M I I 5 D T r y t o y i I 4 t t v I c i l it 2 a i e b c R a A 3 t n l e S a w 3 v W n " mo e n M o F r h r ". o i S 3 u e I l t 0 a r N g e 2 r e i

r s

F 1 u h S 1 o o D t N D a t ,e e e c O s s a y WF d I a a T e e t g o P l l e e n n B 2 M R R a I i t d e U l 1 . s l S T n e o r h S S H o h i A S 2 C S W I 1 2 1 0 9 3 7 8 5 4 3 2 1 0 1 1 1 m$oI Jm<",wE u.O g EO"" 2oxs. *2 r a a l I 4 l ,i! !l! l< )iiiI

_ a = ~.;_ _c._.___ _ -.... _ _ = - '__l- -_m__ ~ i ,, y , 13 g l l l l l l l ~ 12 11 D = 5 Rom D = 100 Rom 58 t f E D = 25 Rom D = 10 Rom a 10 o ASSUMPTIONS m 9 y SST2 R'esesse. m D = 50 Rem 1120 MWe Reactor g a E 2 Hr. Release Duration u. ~ o 7 Constant Weather: 3 c

• • D" Stability,10 mph Z

Windepeed u ~ u z 6 i' Z Shielding Factor: 5 Normal Activity j us 5 f Whole Body Dose Shown 2o 4 1 E i as y 3 P 2 1 D = 1 Ran i I I I I I I I i{ O 1 2 3 4 5 6 7 8 9 10 DISTANCE, MILES Figure 3.3-3 TIME-DOSE-DISTANCE RELATIONSHIP L l

__x____ ~ s

• ? C' f

g g g l l I I I I I

• 13 '

j 12 1 11

• p $

D = 50 Rem 13 D = 25 Rom l 3o I ~ ASSUMPTIONS g 3 y SST2 Release, m D = 200 Rem D = 1n0 Rem 1120 MWe Reactor I -.s S a m 3 2 Hr. Release Durettors g m l Constant Weather: i u. o 7 i "F" Ste-bluty,5 mph e Windspeed j E Shielding Factor: 6 g y Normal Activity 5 Whole Body Dose Shown m 5 m y D = 10 Rom _ o 4 e u. ] 3 P 2 1 l l l l 1 l I I I o 1 2 3 4 5 6 7 8 9 10 DISTANCE, MILES Figure 3.3-4 TIME-DOSE-DISTANCE RELATIONSHIP + r

- m 6te c7 ~ 13 I I I I I I I I I I I 12 - !k 11 I D = 1 Rom e 3 10 l 0 I l J 9 us4* 3 mx o 7 e E_ SST3 Release.1120 MWe Reactor ASSUMPTIONS u ~ ~ S 5 4 Hr. Release Duration a Constant Weather: "F" Stability,5 mph Windspeed yo 4 Shielding Factor: Normsl Activity m u. 3 Whole Body Dose Shown P 2 1 1 I I I I I I i 1 l l 0 1 2 3 4 5 6 7 8 9 10 DISTANCE. MILES Figure 3.3-5 TIME-DOSE-DISTANCE RELATIONSHIP t

Y J Ty, ( p,;- s, .,s m sa ( n, life-threatoning dose is from 3 hours to well beyond 12 hours after the beginning c,f a release; while fo'r ejderfy weather conditions an exposure time of 1 to 3 hors after the beginqjpg vf relege would t,e required. 7 Fur thsse san most,,q"ere corciaelt releases,'the degree of urgency for people ley.ated,from 5 tj 10 miles away would be eased even more. For average I weather corgliofu, times well beyond che dty_are rvquired to receive life j thre6ening doses;" while for adversu weather conditions exposure times ranging from '3 go,1d huu'rs aftcr,the beginnlag of release would be required. 3 j 3.3.4 NConcluJfo'iis < ~ " s An examination ' f,the time of relek for seve're accident sequences as well as o the tire requited for an indtiric4W t'o accumulate a given dose at various distanca).Q/s to th.e followingr,i;enclusions,:., s,1.,,. e 3 t Thid<tisat vdJ'elearidklayltjees for severe' a'ecliQrt sequences, that is, j the de,itietWen accldcut, gittation and when 4he. rehase would commence, o J rangci fpom s'ignificantly 'lepp Gan one bpur 'to delay times on the order [ of one,' day; 2 to 3'nours are'donsidertd'typir.a1 s s s ,3 The perbability ofsa, sever'e, sequence afso'being's fast-developing one, that 46,'ebere the time of reh ase is.ius than 2 hours, ranges from 3 perces to 30 perc4nt; witf(dbout 15 perqtat considered to be typical. ~% 7 i Fer a majority of core-@t veleases (SST3) im11vid'uals located beyond ebaut 2 miles from the r?ac. tor sould require ah exposure time after the tsginning of the rely n of w Q1 in excesl of, p hoyrs e.efore receiving a i case that ceeds ine icwer :ltwwl PAG 'value.(2 ~res whole body).

Hence, very eery' pro.'.ective actions'bLyond ab@ut 2.iiles do not appear to be

. arriate& for these svents S a-w i ,.w inW viduals located at about 'fer most4.o[m-nelterelesbes (SST2 and SST3), ' time af ter the beginning of i ' 2 ml1 $.and beyond would require an extasure D I- / release of at l' east 2 hours Isefore'recelirine a dose likely to produce ,early injuries (apprairaately 50 rem whule bo@). 3 c N. s, Forthemostseverecore-meltraleases(SST1),jndividualswithin2 miles i can receive life-threatening doses (200 rer 9P more, whole body) within j tinic, periods of 1 hour or less 'after (he beginning of"a release. Indivi-ij dudis in these locat. ions would be atsthe greatest depree of risk, given i thi$ tygs of eclease, and an early precautionary avace, tion" should be cemened as early as practicable when a general %2erpency is declared to maximize the daQable time py_jor to tor, cevurrwr.ce cla release. Since s about 1 tc 2 percent of all core-melt relesses'(or,about 10 to 20 percent l of $371 releases) are also far.t-deyeloping, an eracuatica time of up to 2 hours for pecplewithin 2 mileslan Druvi&&"high degree of assurance that life-threatening doses would be'stvo Wed in the event of even the N most sovery core-melt accidenys. ( YFarthese same severe core-s'elt relea ps., the degree of risk and, there- 'Mre the br2ency for evar,uation by iudividuals located f rom 2 to 5 miles For away wquld begomewhat less, than,for individuals within 2 miles. i. s . e s / 28 l s 1 s ,/ W wpgG o.

.O e a t average weather conditions, the time required to receive a life-threatening dose would be from 3 hours to well beyond 12 hours after the beginning of a release. Unlikely adverse weather conditions, however, could reduce this time to 1 to 3 hours in a narrow downwind sector. During these releases, the early risk of life-threatening doses for individuals located from 5 to 10 miles away would be much less. For average weather conditions, exposure times well beyond 1 day would be required; while for adverse weather conditions exposure times ranging from 3 to 10 hours after the beginning of release would be required, b D i 1 I i e An e'vacuation befors a release would be precautionary since at such a time the release magnitude would not be known. Indeed, a release might never 5 occur after the precautionary evacuation. 29 j I I i i

n r, E 4.0 Protective Action Strategies j 4.1 Seiection of a Protective Action Str,at gy Q i l Based on the preceding discussion of the variation of risk with distance as l well as considerations of release delay time, it is possible to devise a simple, reasonable, protective action strategy that is capable of meeting the i f basic radiation protection objectives, given a core melt accident, for the full spectrum of potential consequer,ces, from the relatively more probable small-scale accidental releases to the less likely large-scale releases. Because of the risk variation and timing considerations noted earlier, a distance of about 2 miles and a time frame of about 2 hours has significance for the implementation of early prompt evacuation, where at all possible, for all core-melt events. Consequently, the authors recommend that the 10-mile plume exposure EPZ be divided into 2 zones. The first or inner zone would extend from the reactor out to about 2 miles. A design objective or goal would be to establish the capability for essentially all persons wit.hin this area to evacuate it within about 2 hours. The second, uuter, zone would be the remainder of the 10 mile emergency planning zone. The size (about 2 miles radius) of the inner prompt action area is to 5e based primarily upon the following considerations: For many core-melt accidents, projected doses would not exceed the Protective Action Guide (PAG)12 levels beyond this distance; For most core-melt accidents, projected doses would be unlikely to result in early injuries beyond this distance; Pronpt evacuation by the population within this distance and immediate sheltering beyond would provide a large reduction in risk to the highest risk group and would tit.o provide additional time to assess the developing D situation further to decide upon and implement required additional E protective actions, as necessary, beyond this distance. The design objective to have the capability for evacuation by essentially all persons within the inner prompt action zone within approximately 2 hours is based upon the following considerations: For many core-melt accidents, a response time of about a day or more would be available before projected doses are expected to exceed the Protective Action Guide (PAG) levels. For most core-melt accidents, an evacuation within this time would be unlikely to result in doses capable of producing early injuries. For the worst core-melt accidents, warning times (prior to release) of 2 hours or more are predicted for about 80% to 90% of severe accident sequences. Since about 1% to 2% of all core-melt accidents (or 10% to 20% of SST1 releases) i are estimated to result in severe releases that would also be fast-developing 30 i w_ ~ '

y. (warning times less than 2 hours), an evacuation time of 2 hours for individuals can provide a high, but not absolute, degree of assurance that life-threatening l doses would not be received in the event of a core-melt. Realistically, it / would be expected that the pre-instructed populations living close to a reactor could and would evacuate the 1.5 mile distance from the edge of the exclusion I 2one to 2 miles in significantly less than 2 hours. This early, prompt, rapid movement should be encouraged by planners.

• 1 I

This protective action strategy and its rationale are illustrated in detail in the accompanying Table 4.1. I The uncomplicated, understandable, recommendations to be public that result from this strategy and rationale can be summarized as follows:

SUMMARY

RECOMMENDATIONS j IN THE EVENT OF A PLANT CONDITION WHERE A MOLTEN OR DEGRADED CORE CONDITION IS OBSERVED OR PROJECTED, OR WHERE ADEQUATE CORE COOLING CANNOT BE REASONABLY ASSURED, OR UPON THE DECLARATION OF GENERAL EMERGENCY: i 1. RECOMMEND IMMEDIATE PRECAUTIONARY EVACUATION BY EVERYONE WITHIN ABOUT TVO MILES. THIS EVACUATION TO BE ACCOMPLISHED WITHIN ABOUT TWO HOURS OR LESS WHERE AT ALL FEASIBLE. EVERYONE WITHIN THE REMAINDER OF THE TEN MILE EPZ SHOULD BE ADVISED TO SEEK AVAILABLE SHELTER AND REMAIN INDOORS UNTIL FURTHER NOTICE. 2. ACCIDENT AiSESSMENT SHOULD CONTINUE, WITH MONITORING OF BOTH PLANT AND FIELD CONDITI'f45. AS THE ASSESSMENT CLARIFIES THE SITUATION, FURTHER ACTIONS SHOULu dE TAKEN, SUCH AS INFORMING THE SHELTERED POPULATION PROMPTLY AS TO THE NEED FOR CONTINUING TO SHELTER, OR RECOMMENDING THAT THEY PREPARE TO EVACUATE. D (NOTES: 1) At the time of an accident, decisionmakers should, of course, take Y into consideration actual local conditions that may make evacuation temporarily infeasible (for example, impassable roadways due to severe snowfall or floods) and that might present a greater risk to the general public, if attempted, I than potential radiological conditions warrant. Under such conditions authori-ties should encourage sheltering and should also consider the auxiliary benefits of ad hoc respiratory protection (breathing through handkerchiefs, pantyhose, } etc.) by those elements of the public potentially affected. 4

2) Since early evacuation within two miles would not be sufficient to avoid i

i life-threatening doses beyond this distance for the most severe core-melt releases (those involving prompt failure of containment) under adverse meteoro-logical dispersal conditions, prompt actions beyond two miles would be necessary in such cases. In such a situation, it is expected that emergency authorities would recommend immediate evacuation, from shelter, of people within about five miles in a downwind direction even though a release were taking place. l i 31 1 1

- -. _.. -... -.=

• e n. 0" L.

TABLE 4.1 RECONNENDED PROTECTIVE ACTION STRATEGIES AND RATIONALE .. i DISTANCE INPACT AND TINING ACTION (S) TO BE PLANNED AND RATIONALE 0-2 miles SST3 (70% of core melt accidents) ACTION (S): I. Prompt evacuation (within 2 hours) exceed lower level PAG (1 ree) upon declaration of a general emergency in a 2-12 hours after release complete 2 mile circle. (Expected to be carried out for all core-melt events, SST1, 2 or 3.) SST2 (20% of core melt accidents) II. All persons instructed to take best shelter Average dispersion:* available, preferably that provides facilities for receiving further instructions, within the complete circle from 2 to 10 alles. 4 RATIONALE: For most core-melt events, have about no life-threatening doses 2 hours from accident initiation to beginning of release. Difficult to distinguish between SST1, 2, injury producing doses 1.5-8 hours or 3 at early stages of event. Want to start people after release moving promptly as a precaution prior to release. This buys time to assess situation at greater - Poor dispersion:** distances. Even though this evacuation will probably be only precautionary, it should start life-threatening doses 2-8 hours as quickly as possible and proceed rapidly in u, after release the event the accident develops quickly and an early release results. injury producing doses 0.5-2 hours after release SST1 (10% of core melt accidents)

• Average dispersion taken as Pasquill "D" stability with 10 mph windspeed (50 percentile Average dispersion:*

conditions). life-threatening doses 0.5-3 hours

• • Poor dispersion taken as Pasquill "F" stability after release with 5 mph windspeed (90 percentile conditions).

injury producing doses 0.25-1 hour after release

e rn. e-l RECOMMENDED PROTECTIVE ACTION STRATEGIES AND RATIONALE (Continued) . I. TABLE 4.1 ACTION (S) TO BE PLANNED AND RATIONALE DISTANCE IMPACT AND TIMING Poor dispersion:** l life-threatening doses 0.25-1 hour after release injury producing doses 0.1-0.5 hours I after release 2-5 miles SST3 (70% of core melt accidents) ACTION (S): I does not exceed lower level PAG 1. Immediate sheltering upon declaraton of ( general emergency. (Expected to be carried i SST2 (20% of core melt accidents) out for all core-melt events, SST1, 2, or 3.) Average dispersion:* II. Evacuation, as necessary, for down-l wind sectors after assessment of situation. [ no injury producing doses (Expected to be carried out for 30% of core-i melt events, SSTI and 2.) exceeds upper level PAG (5 rem) 1-3 hours after release u> RATIONALE: Have additional time to assess Poor dispersion:** situation. no life-threatening doses For SST3, no additional protective actions injury producing doses 2-12 hours needed. after release For SST1 and 2, have about I hour after release, SSTI (10% of core melt accidents) or 3 hours after accident initiation, before serious effects experienced. For poor dispersion Average dispersion:* conditions, for rapidly developing accidents or actual releases, shorter response time is necessary. life-threatening doses 3-12 hours after release s i

_. _ _.. ~ -~ .__._2.~___. e rs tr TABLE 4.1 REC 0m ENDED PROTECTIVE ACTION STRATEGIES AND RATIONALE (Continued) I" DISTANCE INPACT AND TINING ACTION (S) TO BE PLANNED AND RATIONALE injury producing doses 1-3 hours { after release I Poor dispersion:** Ilfe-threatening doses 1-3 hours i after release injury producing doses 0.5-1.2 hours after release 5-10 mt? _ SST3 (70% of core melt accidents) ACTION (5): k does not exceed lower level PAG I. Immediate sheltering upon declaraton of general emergency (expected to be carried SST2 (20% of core melt accidents) out for all core-melt events, SST1, 2, or 3.) Average dispersion:* II. Delayed evacuation, as necessary, for down-wind sectors after assessment of situation w* no injury producing doses (expected to be carried out for 10% of core-melt events, SSTI). exceeds upper level PAG (5 rea) 3-12 hours after release Poor dispersion:** RATIONALE: Have additional time to assess situation. no injury producing doses For SST3, no additional protective action exceed upper level PAG (5 rem) needed. 1-3 hours after release i For SST2, sheltering provides significant dose { 'SST1 (10% of core melt accidents) savings. Average dispersion:* For SST1, have about 3 hours after release, or f 5 hours after accident initiation, before serious q no life-threatening doses effects experienced.

7. a-..--m-.---- s t1. G" e 8 TABLE 4.1 RECOMMENDED PROTECTIVE ACTION STRATEGIES AND RATIONALE (Continued) DISTANCE IMPACT AND TIMING ACTION (S) TO BE PLANNED AND RATIONALE injury producing doses 3-11 hours For poor dispersion conditions shorter response after release time is necessary Poor dispersion:** life-threatening doses 3-10 hours after release injury producing doses 1-3 hours after release f a l t I i l. i i

n

4. 2 Probabilities of Consequences for Various Protective Action Strategies 4.2.1 Introduction Previous sections have suggested that a protective action strategy which incorporates provisions for early precautionary evacuation within about 2 miles, with immediate sheltering, possibly followed by relocation beyond 2 miles can be very effective in consequence mitigation for a spectrum of severe accidents.

In order to test this as well as other alternatives, the calculated consequences for these and other response strategies were examined using the CRAC code.s A* number of calculations were performed to estimate the relative worth of various shelter, evacuation and relocation emergency response strategies for a spectrum of accidental release scenarios. Use of the CRAC code provided a means to investigate the chance or probability of accident consequences conditional on ~ the accident and emergency response assumptions, t Three accident source terms (SST1, SST2 and SST3) were used for the calculations. I Please refer to Section 2.0 for additional discussion of these accident source l terms. 4.2.2 Details of Protective Action Assumptions and Consequence Calculations Emergency response assumptions included combinations of evacuation, shelter and relocation from shelters after plume (puff) passage. Here, evacuation means the early movement by people, preferrably before plume exposure. Relo-cation means later movement from contaminated areas after plume passage. Results for an evacuation speed of two miles per hour (2 mph) are presented. Estimates of consequences for slower speeds would produce results close to or coincident with the shelter / relocation results. For early evacuation speeds faster than 10 mph, consequences in the evacuation area would be rare because evacuees would outrace most releases (wind speeds average 5 to 10 mph), i.e., the results are predictable. An assumed evacuation speed of 10 mph would produce optimistic results in that the evacuation duration would be very D. short, i.e., only one hour or less for evacuations within the entire 10 mile I EPZ. A speed of 2 mph cortcsponds to an evacuation duration of one hour for persons within 2 miles and two hours for persons within four miles. Thus, a 2 mph evacuation speed analysis provides neither unduly conservative, nor unduly optimistic overall results, even though actual individual and group evacuation speeds would be expected to be much more rapid. Further, results l are presented for evacuations which occur commencing at the beginning of a 1 release. Results for evacuations commencing much later would again approach I the shelter / relocation results. Results for evacuations commencing much earlier or proceeding more rapidly are also predictable in that the analysis j would much lower doses received by fewer people and few, if any, health effects. l A distinction should be noted throughout this report in the discussion of evacuation stategies recommended by the authors, between the early precaution-ary evacuation of the first two miles and the best evacuation strategy to be followed when an actual severe release is expected or in progress. The first, early, precautionary evacuation should be an automatic, preplanned response to the severely degraded plant conditions that would trigger declaration of a General Emergency. The second, subsequent, strategy that might involve evacu-i ation should be followed only after the plant situation has clarified, and as l 8 36 i i 9 }

y a more accure assessment of the potential magnitude of the threat can be made, j The first strategy would provide for a major reduction in individual risks of 1 severe non-stochastic and stochastic health effects. It also would clear the j way for the second phase of preplanned but more flexibile protective action decisions by authorities on the scene that consider actual conditions at the time. l Protection factors for cloud gamma, inhalation and ground shine pathways, used i for the calculations, are realistic for evacuees in automobiles and somewhat } conservative for people in single wood frame houses without basements. These are listed in Table 4.2-1. It was recognized that sheltering protection factors would be greater in other types of buildings, but that single wood frame dwellings are abundant and are common in the near vicinity of reactor j sites. Thus, the calculations pertain to near worst case sheltering. Since j the protection factors are relatively large (poorer protection), the risk i estimates made would not be grossly underestimated for persons without shelter. i TABLE 4.2-1 PROTECTION FACTORS USED IN i THE CALCULATIONS i ] Cloud Ground j Gama Inhalation Shine Autos 1. 1. 0.5 Residences 0.9 0.75 0.43 Slightly different protection factors were used for evacuees and persons in shelters. Very little or no protection was provided for either group for exposures to external gamma rays from the passing plume (puff). No inhalation protection was provided to early evacuees. An inhalation protection factor of Q 0.75 (twenty-five percent dose reduction) was used for shelterees in recognition i of the fact that air turnover rates in small dwellings average about one per hour, which would reduce doses inside the shelter (WASH-1400, App VI).2 Conservative protection factors for ground shine (gamma rays emitted by radio-nuclides deposited on the ground) were used for small vehicles and small t isolated dwellings with contaminated walls. H,ts y ] It is noted, however, that the potential exists for much better protection 3 factors for most people. Future risk studies should include the most realistic factors available, where they can be justified. Since the purpose here is to ? investigate the relative worth of various evacuation and sheltering schemes, and since timing of protective actions is the most important* parameter, by 3 far, the higher protection factors used should not warp the relative perspec-4 tives to a major degree. The most pronounced effect of better protection factors would be in the reduction of the estimates of early fatalities, because J' of the high threshold of dose required to induce this consequence. Relocation from contaminated areas after plume (puff) passage was assumed to occur after 4 hours within ten miles and after 8 hours outside ten miles, unle'ss noted otherwise. There is no experience on which to base these esti-l 1 37 l 4

f.. f e j. mates. They were chosen as reasonable estimates of the time it would require i for environmental monitoring teams to locate areas of "hot spots" and notify t people to leave. Identification of areas in which ground level dose rates l exceed, say,1 R/hr (i.e., "hot spots") would be a simple matter. Although these highly contaminated areas could be long, they should be narrowl s,17,. such that relocation times would be short and the dose accumulated while relocating would be a small fraction of the total dose. The conservative shielding factors used for the calculations (see above) account, in~part, for the small dose that would be accumulated while outside, relocating. This method of accounting for outside exposures during relocation was necessary because the CRAC code does not provide a separate calculation of the dose i while relocating. Further, the way the code is structured, all persons outside the 10 mile EPZ were relocated after eight hours, whereas in an actual emergency response situation only persons in actual "hot spots" would have to relocate. These code restrictions do not affect the principal conclusions of this study. s was used for the calculations. A s1; htly modified version of the CRAC2 code i This trsion provides for three emergency response zones rather than the two zones rovided in earlier versions. In this version, evacuation is modeled in s' the first zone, nearest the release point. The radius of this zone is an option of the user. In the second and third zones people are assumed to be i always exposed to the full plume. In these two zones the user may vary both j the time of exposure to ground shine from deposited radioactiv4y and protec-tion factors for cloud (external) gamma, inhalation and ground shine pathways. i i The washout coefficient for rainfall in CRAC2 was changed from 10 8/sec per mm/hr of rain (for unstable meteorology sequences) to 10 4/sec per am/hr of rainfall to realistically model measurements of this parameter.18 No other 1 revisions of the CRAC2 code wer,e made for these calculations. The weather sequences used for these calculations is the New York City National Weather Service data set which is available with the CRAC2 code. This selection was somewhat arbitrary, but it has been used in the recent past for numerous consequence calculations performed for the NRC, which have been widely reported. 4 D. Thus, there is a tie point for previous consequence calculations. l ? It is noted that the average ("expected" in the statistical sense) consequence estimates are dominated by normal daytime and nighttime wind speeds of 5-10 4 mph, which are the norm across the U.S. Low probability "peak" consequence estimates are associated with rare weather sequences, which may be different in different climates and places. Thus, the selection of different weather .) tapes would lead to significantly different estimates of peak consequences, 1 j both in terms of mt.gnitude and probability. In brief, there is much greater ,j confidence in the average result than in the peak of the calculated distri-bution of consequences, regardless of which weather tape is used for the j estimates. l ]i It is noteworthy, also, in this regard, that the New York City weather tape includes rainfall some eight percent of the time. Rainfall has been associ-ated very strongly with peak consequence estimates. Thus, the peak consequence estimates should be on the worst case side of the uncertainty bound. p 4 p j 38 1 .I d y o 1 4

. ~ -- l j 1 Section 4.2.3 Results and Discussion The results of the calculations have been plotted as whole body and thyroid _l dose probability distributions for an indiviudal; that is, the probability of an individual eceiving a given dose vs. dose, conditional upon the occurrence l of an SST1, an ;ST2 or an SST3 release. The whole body dose distributions for j an individual located at various distances between one mile and ten miles from l the reactor, given an SST1 release, and assuming sheltering in a one story wood-frame house without a basement and relocating after exposure to the plume j plus 4 hours of ground contamination, are shown in Figure 4.2-1. Whole body dose distributions for an individual assumed to evacuate at an effective speed of 2 miles per hour starting at the time of the release are shown in Figure 4.2-2. i For convenience, key values from these figures are also tabulated in Tables i 4.2-2 and 4.2-3, for sheltering and evacuation, respectively. Comparison of the dose distributions in these figures indicates that an evacuation of persons located within 10 miles, commencing at the time of release and at an effective speed of 2 miles per hour, results in lower individual doses at all distances, both in terms of mean or expected values, as well as lower i l values at comparable levels of probability, than sheltering in a one story ( wood-frame house and relocating af ter 4 hours. However, it is worth noting j that the differences in dose, and more importantly, differences in acute health effects, is greatest close-in to the reactor and diminishes with increasing distance. At a distance of 1 mile, for example, the dose differences are such that acute fatality would be expected for the sheltering followed by relocation strategy, but early fatalities would not be expected for early evacuation. However at distances of about 5 miles and beyond, the dose differences are j such that acute fatality would be highly unlikely for either response. Since the first priority in a release of this magnitude would be to save lives, it I is clear that for individuals located beyond about 5 miles either evacuation or sheltering followed by relocation within about 4 hours would be about equally effective in this regard, even for the most severe releases, b For individuals located at closer distances, and especially within about the ! E first 2 miles, it is clear that evacuation would be far more beneficial than {' sheltering. However, it must again be noted that even the evacuation strategy examined here cannot provide assurance of no acute health effects. Since the time-dose-distance relationships discussed in section 3.3.3 showed that the j dose is accumulated very rapidly (of the order of hundreds of rem per hour) to i individuals very close to the reactor for an SST1 release once it begins, it is clear that to be effective, an evacuation of close-in individuals should j commence prior to any release, and should be as rapid as practicable. The calculated thyroid dose distributions

• for an SST1 release, assuming 1

sheltering and relocation after 4 hours exposure to ground contamination cre 4 i "An artifact of the CRAC code significantly influences the thyroid dose calcu-l ati or.s. In esssence, shelterees are exposed to the full plume in CRAC. In reality, exposures to only part of the plume, e.g., the leading edge, would be expected for a long duration release. In this sense, the thyroid dose { estimates are unrealistically high, for early evacuees. i t 39 i m I

~ ~ ~ - -. _=. _ . n o-k i i i i i 111] 1 1 1 111] I i i i 1 : 1' L t

• 1 story mod-fram Muse, no hsenent 1.0 i

L i n n t t t i C =f I m i 4 m 10 7 5 3 2 1 Mile o l 10 i .i,. cc n. t I i i Arrows indicate mean values j r i. l, .1 \\t 1 10 2_ t i e i ii1t! 8 I I I18 8 8 { 10 10 19 1M; 2 i WHOLE BODY DOSE, Rem Figure 4.2-1 Probability of an Individual at a Given Distance Receiving a Whole Body Dose in Excess of the Value I Shown, Conditional Upon Occurrence of an SSTI Release and Shelter

• Plus 4 Hrs. Ground Erposure i

... =..-..._ - _ 1- .ro o-i i i ising i i i i i s sl i i i 1 14 - i c 1.0 }' 4 t f I ~ f j D i 3 m4 M O l ~ cc 10-3 10 7 5 3 2 1 Mile n. i i. i Arrows indicate mean values i 10-2 a aai l s anl a n aa 2 3 10 10 10 10 WHOLE BODY DOSE, Rem I Figure 4.2-2 Probability of an Individual at a Given Distance Receiving a Whole Body Dose in Excess of the Value Shown, Conditional Upon Occurrence of an SST1 Relcase and Evacuation at 2 MPit at Time - 0

'.-.,.--.--..-..:x. ^ l Table 4.2-2 l 1 KEY INDIVIDUAL DOSE DISTRIBUTION VALUES 1 CONDITIONAL UPON SST1 RELEASE SHELTERING

• AND RELOCATION AFTER 4 HOURS GROUND EXPOSURE DISTANCE WHOLE BODY DOSE (REM)

(MILES) MEAN DOSE 90 PERCENTILE 95 PERCENTILE 99 PERCENTLE 1 550. 1500 2100 3000 2

250, 700 1000 1700 3

140. 400 600 1000 5 70. 180 250 300 7 40. 110 150 220 10 25. 75 100 150 "Sheltering in 1 story wood-frame house, without be.sement i I Table 4.2-3 KEY INDIVIDUAL DOSE DISTRIBUTION VALUES t j CONDITIONAL UPON SST1 RELEASE AND EVACUATION AT 2 MPH STARTING AT TIME OF RELEASE DISTANCE WHOLE BODY DOSE (REM) a (MILES) b MEAN DOSE 90 PERCENTILE 95 PERCENTILE 99 PERCENTLE 1 125. 350 500 700 2 80. 250 350 500 3 57. 160 220 300 1 5 34. 100 140 200 7 25. 75 100 150 1 10 15. 40 55 90 l i 4 42 g

• ww see 9 W W

4 m_ l.- l shown in Figure 4.2-3, while those for evacuation at 2 miles per hour starting at the time of release are shown in Figure 4.2-4. Again, for convenience, key i values from these figures are tabulated in Tables 4.2-4 and 4.2-5, for shel-tering and evacuation, respectively. Comparison of the thyroid dose distribu-t tions reveals once again that evacuation results in the lower thyroid doses, but that differences in acute thyroid effects, such as ablation of the thyroid (highly likely at doses in excess of about 3500 rem) become small for indi-viduals located beyond about 5 miles. In order to ascertain whether a shorter relocation time than 4 hours might prove effective in mitigating acute effects for non-evacuees at distances less than 5 miles for the most severe releases, a calculation was performed where an individual at 3 miles was assumed to be sheltered and then relocated after being exposed to the plume plus 1 hour of ground contamination. The whole body and thyroid dose distributions are shown in Figures 4.2-5 and 4.2-6, respectively. These results indicate that although both whole body and thyroid doses are reduced somewhat, they are still higher than the doses where evacuation was performed. Since sheltering followed by relocation after only 1 hour of ground exposure is virtually indistinguishable from evacuation, it is concluded that evacuation would be the preferred strategy for individuals located at distances less than 5 miles, in the event of an SST1 release. For an actual SST1 release, consequently, the major objective would be to avoid early fatality. For individuals located from 0 to 5 miles from the reactor, the preferred response would be an early evacuation, which should commence prior to any release. For individuals located from 5 to 10 miles, either evacuation or sheltering followed by relocation within about 4 hours would be about equally effective. Since doses in excess of the PAG levels can extend beyond 10 miles, protective actions may be necessary for members of the public at distances beyond 10 miles in the event of a release of this magnitude. However, as shown by Figure 4.2-1, early evacuation beyond ten miles would not be warranted, since sheltering in a 1 story wood-frame house without a basement followed by relocation after 4 hours of ground exposure essentially precludes i ! I early fatality and makes early injuries unlikely as well. J i We now consider the relatively smaller SST2 release. From Figures 3.2-2 and 3.2-4, it is clear that the risk of early fatality to an individual shows a very rapid decrease with distance, with a value of about 10 4 at a distance of 1 mile. Similarly, the risk of early injury also shows a rapid falloff with distance, although it displays a greater reach in terms of distance. Figure 3.2-4 l shows that the risk of early injury, given an SST2 release, is very low beyond about 2 miles with a value of about 10 8 at this distance. Based on these individual risk variations, a prompt evacuation is considered to be the preferred response for individuals located within about 2 miles from the reactor, in the event of an SST2 release. To examine the impact to individuals beyond 2 miles, calculations were performed i to examine both evacuation and shelter by individuals beyond 2 miles. The whole body and thryoid dose distributions for an individual located at various j distances from 1 to 10 miles, given an SST2 release, and assuming shelter in a 1 one story wood-frame house without a basement and relocating after exposure to the plume plus 4 hours of ground contamination, are shown in Figure 4.2-7 and 4.2-8. For convenience, key values from these figures are also tabulated in Tables 4.2-6 and 4.2-7, for whole body and thyroid doses, respectively. i 43 I l

-_e c' ~ " ~ ~ ~~~ _ __ m - -1 I I I I I I[ t I I i 1 11Il 1 I I I I I,1 ~ i

• 1 Story wood-frame house, no basement l'

1.0 I I I ~ l ~ m J L m i E F i s. m I, - O 10 7 5 3 2 1 Mile E 10~1 7 i t 1 Arrows indicate mean values t i ; l l l I i a i 10-2 i i e i i il i e i kin s i e i

n!

2 3 10 10 1@ 1 THYRQlD DOSE, Rem I Figure 4.2-3' Probability of an Individual at a Given Distance Receiving a Thyroid Dose in Excess of the Value Shown, Conditional l?pon 0ccurrence of an SSTI Release and Shelter

• Plus 4 Hrs. Grour.J Exposurg

.re o- , - ]' .__.m ll i i i 1 i i:l 1 i i iiI i i i i i i ) l - 1.0 d k d b J L jb J i l 2 m m 10 7 5 3 2 1 Mile vu O 10-i cc , s I n. 1 I 4 I Arrows indicate mean values 1 t 1 l' i i e i el i i i iei l t i i i eiit 10-2 a i i 10 10 10" 10 2 3 j ~ THYROID DOSE, Rem Figure 4.2-4 Probability of an Individual at a Given Distence Receiving a Thyroid Dose in Excess of the Value Shoun, Conditional Upon Occurrence of an SST1 Release and Evacuation at 2 MPH j at Time of Release j ~

.- :..~.., _ ___._ 4 i Table 4.2-4 ? KEY INDIVIOVAL DOSE DISTRIBUTION VALUES CONDITIONAL UPON SST1 RELEASE, SHELTERING

• AND RELOCATION AFTER 4 HRS. GROUND EXPOSURE 1

O! STANCE THYROIO DOSE (REM) (MILES) MEAN 90 PERCENTILE 95 PERCENTILE 99 PERCENTILE l 1 14,200. 30,000 42,000 70,000 2 6,000. 15,000 23,000 30,000 3 3,300. 8,000 13,000 16,000 5 1,500. 4,100 6,500 10,000 7 880. 2,200 3,800 7,000 10

510, 1,300 2,300 4,200 4

• Sheltering in 1 story wood-frame house, without basement.

Table 4.2-5 XEY INDIVIOUAL DOSE DISTRIBUTION VALUES CONDITIONAL UPON SST1 RELEASE AND EVACUATION AT 2 MPH STARTING AT TIME OF RELEASE J DISTANCE THYROIO DOSE (REM) (MILES) MEAN 90 PERCERTILF 95 PERCENTILE 99 PERCENTILE 1 4,900, 11,000 14,000 20,000 b 2 2,800. 6,500 10,000 13,000 t 3 1,900. 4,600 6,800 9,000 i 5 1,000, 2,600 4,200 6,500 7

660, 1,600 2,500 4,200 10 360.

900 1,400 2,000 t 9

l 1

i e 46 4 r

I___,.__. t .ro o-l i y i i i iiy a i I i isu] I y I I I III l '1.0 T

• 1 story wood-frame house, no basement d b i

D 35 i MEAN DOSE 1 O m = 85 Rem } o 10-1 oc A l [ i r i 10-2 e t i i ieinl 1 i i i etil a a e i aans 1 2 3 10 10 10 10' l DOSE, Rem i Figure 4.2-5 Probability of an Individual at 3 Miles Receiving a Whole Body Dose in Excess of the Value Shown, Conditional Upon Occurrence of an SSTI Release and Shelter

• Plus 1 ifr.

Grcund Exposure r

... _ _ _ _ _ _ _.. - _ _ _ _ _ _. = - .. m.1 - u :._ _ m._ ' i. .r. o-b e i i e iev-i i i i iiy a i i i i iiy ,t 1.0

• 1 story wood-frame house, no basement I

~ f ( i ._.a m 4 m @ 10-1 MEAN DOSE n. = 3240 Rem i i j [ i t i i i i i a a i i i e il e a aeie1 a aaaa i 10 2 2 3 10 10 10' 1 DOSE, Rom Figure 4.2-6 Probability of an Individual at 3 Miles Receiving a Thyroid Dose in Excess of the Yalue Shown, Conditional Upon Occurrence of an SSTI Release and Shelter

• Plus 1 Hr. Ground Exposure k

- ~ ' " ~ ^ ^ ~ ' ~ ~ ~ ~. ~..... - --a -w----'-~- 3- . re v ',I l l i I I I i 1 5I] I I I I I I I I-g i a

I 1 ig

' 1.0 w

• 1 story wood-frame house, no basement 2

l 1 i ) n n m N_. m4 S 10 7 5 3 2 1 Mile 10-1 a- ~ A l Arrows indicate mean values l 1 i e i e i n iil i i a i i is e i i i ses i 10-2 j 1.0 10 10 gg 2 i DOSE Ram [ Figure 4.2-7 Probability of an Individual at* a Given D[ stance Receiving a Whole Body Dose in Execss of f the Value Shown, Conditional Upon Occurrence of an SST2 Pelease and Shelter

• Plus 4 l' s.

Groun'J Exposure j . ~.

t l . re. v -. '[ i .y t I I I I I IIIj I I I I I I IIl I I I I I III

• 1 story wood-frame house, no basement i,

i 1.0 ~_ ! l i 1 { o n n a C__ t 1 u t. J. se m e 4 10 7 5 3 2 1 Mile _._j m g 10-1 1 m f I i -i _I, ~ Arrows indicate mean values ~ t iil i i i il a i i ei 10 2 i e i i i 2 3 1.0 10 10 10 i THYROID DOSE. Rem Probability of an Individual at a Given Distance P.eceiving a Thyroid Dose in Excen of the Figure 4.2-8 ? Value Shc m, Condttional Upon Occurrence of en SST2 I!elease and Shelter

• Plus 4 Hm. Ground Exposure a

T l -_.__..a..- 4, Table 4.2-6 2 i KEY INDIVIOUAL DOSE DISTRIBUTION VALUES 1 CONDITIONAL UPON SST2 RELEASE, SHELTERING

• AND RELOCATION AFTER 4 HOURS GROUND EXPOSURE DISTANCE THYROID DOSE (REM)

(MILES) MEAN 90 PERCENTILE 95 PERCENTILE 99 PERCENTILE 1 2 l 3 11. 30. 45, 70. 5 6. 16. 24. 32. 7 3.9 11, 17. 23. 10 2.5 7.5 12. 16.

• Sheltering in 1 story wood-frame house, without basement.

i l Table 4.2-7 KEY INDIVIOUAL DOSE DISTRIBUTION VALUES CONDITIONAL UPON SST2 RELEASE, SHELTERING" AND RELOCATION AFTER 4 HOURS GROUND EXPOSURE DISTANCE THYROIO DOSE (REM) (MILES) d MEAN 90 PERCENTILE 95 PERCENTILE 99 PERCENTILE 1 iI. 2 3 54. 120. 190 280 5 25. 65. 90 110 i i 7 15. 35. 55 75 i 10 9. 23 35 50 l l .l 1 i l -l l l 51 i l

j a,- Comparison of dose distributions indicate that an evacuation (figures not shown), comencing at the time of release and at an effective speed of 2 miles per hour, results in lower individual doses at all distances, than sheltering i in a one story wood-frame house and relocating after 4 hours. However, the differences in dose between these two strategies are such that acute health I effects such as early injuries are unlikely for either strategy for the SST2 l release. Both strategies result in individuals receiving doses above the PAG levels but below the level where any early injuries would be expected. However, 3 I sensitive elements of the population (e.g., the fetus) might be affected by { doses at these levels. Hence, while early evacuation beyond 2 miles for an l SST2 release may provide no reduction in acute health risks for the general population, there could be such benefits for special elements of the population. It was also decided to evaluate the consequences of a delay in relocation, given an SST2 release. Consequently, the dose distributions were obtained for j individuals beyond 2 miles who, given an SST2 release, were in shelters and then relocated af ter* 8 hours and 12, hours of ground exposure as well. The j whole body and thyroid dose distributions for sheltering with relocation after 8 hours of ground exposure, are shown in Figures 4.2-9 and 4.2-10, respec-tively. The corresponding dose distributions for relocation after 12 hours are shown in Figure 4.2-11 and Figure 4.2-12. From these curves it can be seen that there is a relatively small penalty, in terms of dose increase, if relocation is not accomplished within 4 hours. For example, for an individual located at 3 miles the mean or expected whole body dose is increased from 11 rem to 13 rem if relocation is delayed from 4 to 8 hours, and increased to 15 rem for relocation after 12 hours. Similar percentage increases are noted at the other distances, as well. To summarize for an SST2 release, the major objective of emergency response would be to avoid early fatalities and early injuries. For individuals located within 2 miles from the reactor, the preferred response would be a prompt evacuation, which should commence prior to any release. For individuals located from 2 to 10 miles, either evacuation or sheltering followed by relocation fb within about 4 hours would be about equally effective. However, there would i be a relatively small increase in dose, of no significance for early health i effects, if relocation were to be delayed up to 12 hours. For sensitive elements of the population (for example, pregnant women), evacuation out to ) about 5 miles may be advisable, however. Since doses would not be expected to i exceed the upper PAG 1evels beyond 10 miles, evacuation for members of the { public beyond this distance would not be expected to be warranted for a release of this magnitude. l Finally, the SST3 release was also examined. From Figure 3.1-3, it is clear i that the doses would not be expected to exceed the PAG" levels at distances j beyond about 2 miles, for this release. To evaluate this, the whole body and thyroid dose distributions for individuals located beyond 2 miles, were calcu-l lated assuming sheltering in a one story wood-frame house without a basement and relocating after exposure to the plume plus 12 hours of ground contami-i i nation. The results are shown in Figures 4.2-13 and 4.2-14. From these curves it is clear that the doses are well below the PAG values for this response strategy. 52 1 i =

7 ._-_1 = - . n. cr ^ I I I I I III) i I I I IIIIl l l I I I I II i* '1.0 t = ! 1 4

• 1 story wood-frame house, no basement f

i k 1, m m C 2 i m i 0 4 l 10 7 5 3 Miles .i cc l i n. i } i f - Arrows indicate mean values g;ll g i g g g ggl g g g ggggg [ 10-2 g i g i ; 2 3, 1.0 10 10 '10 i. DOSE, Rom Figure 4.2-9 Probability of an Individual at a Given Distance Receivir.g a Whole Body Dose in Excess er tt:e Value Shown, Conditional Uron Occurrence of an SST2 Release end Shelter

• Plus P, firs.

Ground Exposure 'l

r__. er*C7 'l I I I I I I 31l l 1 I I I 1 11) 1 1 I I I III a 1.0 t- ~ j

• 1 story wood-frame house, no basement

~ l 6t t 4 j 1 l g m =' i i u, m e l m 4 o 10 7 5 3 Miles 1 e 10-i a i g Arrows indicate mean values i. j 10-2 g g i giil l l l l l l[g i n1 2 3 1.0 10 10 10 THYROID DOSE, Rem Figure 4.2-10 Probability of an Individual at a Given Distance Receiving a Thyroid Dose in Excess of the Value Shown Conditional Upo: Occurrence of an SST2 Release and Shelter

• Plus 8 Ifrs.

j Ground Exposure

_a ~ .re cr ~ o {" I su 1 I I I I I I3l I I I I i 11Il j l 1.0 n 1 !y i 'f

• 1 story wood-frame house, no basement I

i f n lj N i i 3 ,l I m l 4 w 10 7 5 3 Miles v m O 10-1 i! m ~ A ~ I 1 l .j 4 j j i Arrows indicate mean values l, ~ ~ i 1 l \\ l 1 i si l i i i e insa l n iil l 10-2 a i i 2 3 1.0 10 10 19; DOSE, Rem l y Figure 4.2-11 Probability of an Individual at a Given Distance Receiving a Wole Body Dose in Er: cts nf the value shown, Conditional Upon Occurrence of an 55T2 Release and Shelter

• Plus 12 Ilts.

1 Ground Exposure l

.: u -

i

~ ^ 6 m.F Ij I I I I I III f* I I I I f I I i y g g y g yyIg i i 1.0

• 1 stcry wood-frame house, no basement

, a. ; -, sg. . x,. s t t t s + s = i m we 4 ~1 10 7 5 3 Miles m 10 a l s. f s. '~2 I Arrows indicate mean values 1 } j l 10-2 i e i i i i iil i t i n iiI 8 8 I I ' ' '8 7 2 1.0 10 10 1% THYROID DOSE, Rem Figure 4.2-12 Probability of an Individual at a Given Distance Receiving a Thyroid Dose in Excess of the Value Shown, Conditional Upon Occurrence of an SST2 Release and Shelter

• Plus 12 lirs.

Ground Exposure } j

4' e rs. 7 s j I I I I I I I Il I I I I I III] I I I I I I I I-I ' 1.0 9 _~ l ~

• 1 story wood-frame house, no basement.

l. jl a t a 6 h 3 M v 4 w m 10 7 5 3 Miles O 10-1 E i I i Arrows indicate mean values i 4 ( [ 10-2 e i i a i i ii1 e i i l an [ 10-2 10 1.0 10 DOSE, Rem Figure 4.2-13 Probability of an Individual at a Given Distance Receiving a Whole Body Dose in Excess of the Value Shown, Conditional Upon Occurrence of an SST3 Release and Shelter

• Plus 12 !!rs.

i Ground Exposure { 4 1

l l,y n . i. n an.cr s a I I I I IIlll I I I I IIII l-1 1 I I I III)

• 1 story wood-frame house, no basement 1.0

l i

u 7 j. n h d 6 3, 3 m wa 4 b-m 10 7 5 3 Miles 0 10-1 a l' Arrows indicate mean values i i

i l

, i i. i l i i e,,,i j iiiil i i 10-2 i i i i l I 10-2 10-1 1.0 10 THYROID DOSE, Rem Figure 4.2-14 Probability of an Individual at a Given Distance Receiving a Thyroid Dose in Excess of the Value Shown, Conditions 1 Upon Occurrence of an SST3 Release end Shelter

• Plus 12 Ifrs.

't Ground Exposure

9 ...-.~. a.- .. - - - - - - - - - - - - - - - - - - ~ - - - --- - ~ ~- ,..u~.._...-.~

5.0 REFERENCES

1. U.S. NRC and U.S. EPA, "Planning Basis for the Development of State and Local Government Radiological Emergency Response Plans in Support of Light Water Nuclear Pcwer Plants," NUREG-0396 and EPA 520/1-78-016. U.S. Nuclear Regulatory Commission, December 1978. 2. U.S. Nuclear Regulatory Commission, "Reactor Safety Study, An Assessment of Accident Risks in U.S. Comercial Nuclear Power Plants," WASH-1400, NUREG-75/014, October 1975. 3. Code of Federal Regulations, Part 50 "Domestic Licensing of Production and Utilization Facilities," January 1,1984. 4. U.S. Nuclear Regulatory Commission, "Criteria for Preparation and Evaluation of Radiological Emergency Response Plans and Preparedness in Support of Nuclear Power Plants," NUREG-065% (FEMA-REP-1), January 1980. S. R. M. Blond, M. Taylor, T. Margulies, M. Cunningham, P. Baranowsky, R. Denning, and P. Cybulskis, "The Development of Severe Reactor Accident Source Terms: 1957-1981," U.S. Nuclear Regulatory Commission, NUREG-0773, November 1982. 6. NUREG/CR-1659, "Reactor Safety Study Methodology Application Program," Parts 1-4, Jan. 1982. (RSSMAP) 7. D. C. Aldrich, et al., "Technical Guidance for Siting Criteria Development," NUREG/CR-2239, U.S. Nuclear Regulatory Comission, December 1982. 1 j 8. "Calculations of Reactor Accident Consequences Version 2 CRAC2 Computer i Code. Users Guide." Ritchie, L. T.; Johnson, J.D.; Blond, R.M. Sandia Laboratories, NUREG/CR-2326, April 1983. k .t~ 9. "In-Plant Considerations for Optimal Offsite Response to Reactor Accidents," Burke, R. P.; Heising, C. D. Massachusetts Institute of Technology, Cambridge, MA. Aldrich, D. C. .Sandia Laboratories, NUREG/CR-2925, March, 1983. 10. "Final Environmental Statement related to the operation of Limerick Generating Station, Units 1 and 2, Occket Nos. 50-352 and 50-353," U.S. Nuclear Regulatory Comnission, NUREG-0974, April 1984. 11. "U.S. Nuclear Regulatory Commission 1984 Policy and Planning Guidance," HUREG-0885, Issue 3. i 12. U.S. Environmental Protection Agency, "Manual of Protective Action Guides and Protective Actions for Nuclear Incidents," EPA-520/1-75-001 (September 1975), Revised June 1980. 13. ,I. B. Wall, P. E. McGrath, S. S. Yaniv, H. W. Church, R. M. Blond, J. R. Wayland. "Overview of the Reactor Safety Study Consequence Model," U.S. i Nuclear Regulatory Commission, NUREG-0340, October 1977. a 59 I f I .. ~..

I ...t. .~..._._m_u...__._.._____________ t s s 'y J 14. B. Laurisden and P. H. Jensen, "Shielding Factors for Vehicles to Gamma Radiations from Activity Deposited on Structures and Ground Surfaces." Health Physics, v. 45, m. 6, Dec.1983. 15. Z. G. Burson and A. E. Profio, "Structure Shielding in Reactor Accidents," Health Physics, v. 33, n. 4, September 1977. 16. T. S. Marjulies and J. A. Martin, Jr. "Dose Calculations for Severe LWR l Accident Scenarios." NUREG-1062. May 1984. 17. J. A. Martin, Jr. Dose While Traveling Under Well Established Plumes. Health Physics, v. 32, April 1977. 18. Protection of the Public in the Event of Major Reactor Accidents: Principals for planning. ICRP/84/C4-5/2, Adopted by the Main Committee in May 1984. 19. "The Calculation of Wet Deposition for Radioactive Plumes," H.D. Brenk and K. J. Vogt, Nuclear Safety, Vol. 22, No. 3, May-June 1981 l 20. "Nuclear Power in an Age of Uncertainty" (Washington, D.C. : U.S. Congress, 1 Office of Technology Assessment, OTA-E-216, February 1984). \\ J 21. "Basis for Selection of Emergency Planning Zones for the Shoreham Nuclear Power Plant, Suffolk County, New York," F.C. Finlayson, E. P. Radford. 22. "In Utero Exposure to A-Bomb Radiation and Mental Retardation; A Reassessment." Masanori Otake, William J. Schull, The British Journal of Radiology. 57.409-414, 1984. 23. "The Effects on Populations of Exposure to Low Levels of Ionizing i Radiation:1980" (BEIR III), National Academy of Sciences,1980. 24. "Objectives of Emergency Response and the Potential Benefits of t -l-Evacuation and Shelter," James A. Martin, Jr., Proceedings of Topical i Symposium of the Health Physics Society, January, 1985. 4 't i i l l i l l ?~ 1 l 60 i l 1 1 I ~ ~ ~ -* ~ ______}}