ML20246B440

Technical Basis for Offsite Emergency Planning in Us

ML20246B440
Person / Time
Issue date:	06/05/1989
From:	Congel F Office of Nuclear Reactor Regulation
To:
Shared Package
ML20246B415	List: ML20246B411 ML20246B440 ML20246B459
References
NUDOCS 8907070337
Download: ML20246B440 (10)

Text

. _ - _ _ - _ _ .

I L; Ibn .c ..,

L 's i

• ~

l l

TECHNICAL BASIS-FOR OFFSITE EMERGENCY PLANNING IN THE UNITED STATES Dr. Frank J. Congel-Director Division of Radiation Protection and Emergency Preparedness US Nuclear Regulatory Commission Washington DC, USA ABSTRACT The technical bases for offsite emergency' preparedness in the United States are summarized, with emphasis on recent

_. developments. Dominant considerations are the kinds of accidents that must be addressed and the potential releases of radioactivity associated with each accident type. There have been important new developments in both areas but the impact on emergency- preparedness are less than might have been expected.

No major changes in emergency preparedness requirements are forseen. The new developments do . place added emphasis on early initiation of evacuation for some c6re melt sequences, especially for people within a few miles of the plant.

RtSUMt

= "

Les bases techniques des plans particuliers d' intervention hors site aux Etats-Unis sont prdsentdes ici, en insistant sur les ddveloppements rdcents. Les points principaux sont les types d' accidents qui doivent stre pris en compte et les rejets radioactifs potentiels associds & chacun d'eux. Il y a en 1 d'importants ddveloppements rdcents dans ces deux domaines mais leur impact sur les plans d' intervention a 6ts moindre que ce qui stait attendu. Aucun changement important concernant ces plans d' intervention n'est prdvu. Les rdcents ddveloppements portaient surout, en cas d' accident avec fusion du coeur, sur le ddelenchement rapide de l'dvacuation des populations dans un rayon de quelgpes miles autour de la centrale.

i G907070337 890605 FDR ORG NRRB

- - - - - _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ . _ _ i

l

\

l

?

Background

The possibility of accidents with serious offsite consequences was recognized when the U.S. nuclear power program was initiated. Methods for assessing the potential hazards were addressed at the 1955 Geneva Conference.[1] Our first generic study of nuclear power plant hazards was published in 1957 and for several years it guided our thinking about potential offsite consequences.[2]

) To control the perceived potential hazards, a " defense in depth" approach was l

(

_J taken in the siting, [3] design, construction and operation of nuclear power ]

plants. Emergency preparedness was seen as the last line of defense and was a part of the regulatory requirements almost from the beginning. The early l emergency plans were rudimentary, however, and failures in some early plants 4 suggested that emergency preparedness might merit more attention. " Protective Action Guides" were developed [4] and the Reactor Safety Study [5] showad that proper protective action could significantly reduce the off-site consequences of a major accident. Emergency planning requirements were detailed in the first Standard Review Plan [6] and the " planning bases" for emergency response plans were developed well before the accident at Three Mile Island (TMI).[7]

The detailed regulatory requirements of Appendix E [8] to 10 CFR Part 50 (Emergency Planning) were issued shortly af ter that accident, as were the criteria for plan evaluation.[9] Additional progress is being made, at least partially in response to the accident at Chernobyl. This additional progress includes guidance on instrumentation [10,11] end other aspects of coping with and recovering from an accident. Attentiori also is being paid to accidents at facilities other than power reactors.[12]

The U.S. NRC has an extensive and comprehensive emergency preparedness program.

The purpose of this paper is to offer an evaluation of the technical bases for the program and the reasons for the positions we have taken.

Objectives

=

The basic objective of an emergency preparedness program is to provide additional protection for the local populace, beyond the protection provided

.by plant's safety measures. In an effort to define the extent of the pro-tection to be provided, " Protective Action Guides" (PAGs) were developed for nuclear power plants [4]. (Earlier PAGs had been developed with fallout from atmospheric weapons tests in mind.) [13 & 14] The PAG values of I to 5 rems (cSv) to the whole body and 5 to 25 rems (cSv) to the thyroid of a member of the public are well below the levels of discernible ill effects. These dose values are not limits but are presented as potential doses at which action is initiated so that they can be mitigated or avoided. Thus, emergency prepara-tions are designed to facilitate performing protective actions wherever such i doses can be avoided, recognizing that it may not always be possible to keep doses below the PAGs.

b f*

Just before the accident at TMI-2 in 1979, a joint NRC/ EPA task force decided l that the emergency plans should provide for:

(1) Substantial reduction in early severe health effects (injury or death) in the event of a worst-case core melt accident, and (2) Dose savings for a spectrum of accidents [7].

These are basic radiation protection objectives for emergency response and the plans should reflect them.

The Task Force recommended establishment of two emergency planning zones of 10 mile and 50 mile radii around a nuclear power plant to provide reasonable assurance that these objectives could be met for a spectrum of accidents and radionuclides releases (source terms). Af ter TMI-2, those zones were estab-lished by regulation.

Source Term

~

Determining the timines , quantity, and composition of the radionuclides mixture that might be released from the reactor core and from the containment has been considered the most difficult aspect of determining potential offsite conse- I quences of an accident. Fission product release experiments were initiated in l 1942 and by the time the NRC's siting regulation,10 CFR Part 100 [3], was j written in 1960, it was clear that the potential releases were strongly depen- i dent on accident conditions and that those' conditions could not be delineated accurately in advance. For licensing purposes, a standard release to the reactor containment (100 percent noble gases, 25 percent halogens and 1 percent others) was assumed and this assumption strongly influenced emergency prepared-ness before THI-2. For licensing purposes, the containment is assumed to remain substantially intact.

The " Reactor Safety Study" [5] developed 14 different source terms and used

- them in assessing reactor risks. Most of these source terms to the environment were smaller than the release to containment assumed in licensing analyses.

They differed with plant characteristics and accident conditions.

After TMI, a serious effort was made to reassess the source term assumptions

[15]. A sophisticated set of computer codes. was developed to address each step in the release process and, to the extent practicable, the source term code package was checked by comparing its results to the results of tests.

The NRC staff believes this code package represents a substantial improvement of the " Reactor Safety Study" [5] source tems. While the new source terms were often smaller than those of the " Reactor Safety Study," this was not always the case. The newer sturdy did not justify a major reduction of the conservative licensing source tem or of emergency planning requirements.

Accident Conditionss In a system as complex as a nuclear power plant, the number of hypothetical accidents is very large. In assessing safety and in preparing to cope with

k I t

potential accidents, it is necessary to select and evaluate a systematically defined set of accidents. For licensin defined in the " Standard Review Plan" [6]g purposes, aHistorically, is analyzed. small set ofaccidents accidents more severe that those of the " Standard Review Plan" were categorized as

" Class 9" and were not reviewed in the safety (licensing) review. For the licensing evaluations of the site and the containment systems , however, a release of radioactivity to containment is assumed that corresponds to the potential release from a " Class 9" accident. For the next generation of nuclear power plants (advanced reactors), Class 9 accident potential is being considered explicitly as part of the design basis of the plant.

In the U.S., " conservation" and " defense-in-depth" are the fundamentals of licensing, whereas " realism" is the keyword for risk analyses. The licensing system of accident analysis was developed to provide reasonable assurance that the plants could be operated without undue risk to the public. It was not intended to be a framework for determining the magnitude of the risk or for optimizing the safety measures. The " Reactor Safety Study" was a major step in that direction but it was limited. Consequently, a new study was under-taken; it was designed to reflect (1) the new understanding of the

- source term [15], (2) new knowledge about reactor accidents, and (3)improve-potential ments in reactors resulting from the TMI experience. The new study has been updated and published in draft form for comment as the " Reactor Risk Reference Document."[16]

Containment For emergency planning, the containment is 'the most conspicuous reactor safety feature and one of the most important in protecting the public. No matter what the accident or the in-containment source tenn, if the containment anain-tains its integrity, the public is protected. This fact has long been evident. ,

The new accident study [16] also showed that integrity need not be maintained I indefinitely; late failures such as those from overpressure or basemat melt-through are much less likely to cause offsite injury, let alone fatalities.

Late failures also would provide more time for emergency response measures to l r be implemented.

~

One important virtue of the new reactor accident study [14] is that it offers . _ .

l detailed analyses of possible containment failures. This is a complex l analysis. The Sequoyah containment event tree, for example, has 49 top events. I Further work is needed to reduce the uncertainties associated with the pro- l babilistics of containment failure but an important step has been taken, j 4

The possibility of direct heating of the containment atmosphere by the sudden l release of molten fuel is a particularly significant subject for further study {

because it tends to cause early failure and because the probability of occur-  !

rence is highly uncertain, l l

I i

i ,

I t

9 Atmospheric Transport For radioactive material to cause injury, it must be transported from the core i to where the people are. This is one of the areas of great uncertainty in evaluating reactor hazards. The new reactor accident study [16] did include a serious analysis of the behavior of aerosols in the containment atmosphere, which is important for the possible particulate source terms. There also are statistical analyses including the probability of various meteorological condi-tions. There remains the likelihood that our assessment of risk and our pro-visions for emergencies could be improved by additional work in this area.

One of the lessons from the Chercobyl accident is that plume rise can be of critical importance. Ignoring the fac: that hot gas tends to rise may be an important conservatism for dose modelin1 But counting on aieme rise would not be a prudent planning basis. Wind shifts during atmosphan, fransport are virtually ignored. We also may be overestimating the doses frora radioactive materials ieposited on the ground. Onsite deposition is also virtually ignored. Further studies in this area are being considered and we may benefit from the work being done by the CEC in this area.

Biological Effects of Radiation In 1956, the U.S. National Academy of Sciences announced that ionizing radiation was the best understood environmental hazard and radiobiological knowledge has advanced greatly in the last 30 years. Still, there is great uncertainty about the consequences of low. levels of radiation exposure. We base our estima tes on " prudent assumptions" but recognize that the risk estimates may change with time. To support the new reactor accident study, the NRC supported an extensive study of the health effects of radiation exposure

[17]. The NRC report is slightly more conservative than the BEIR report [18]

in the area of low doses and it addresses the consequences of high doses in a manner that is convenient for use in risk assessments. Its conclusions, how-ever, are not so different from our previous models that plans and conclusions

, need to be changed. Improvements in this area are not expected. _

Value of Protective Actions There are a number of things that can be done to. protect people from a release of radioactive material. These consist of:

1. Evacuation,

2. Shelter?ng,

3. Potassium Iodide Medication,

4. Decontamination,

5. Medical Treatment, and

6. Control of contaminated food and drink.

The relative importance and effectiveness of each of these protective actions would depend on the situation, the objective, and the timing of the response.

.- .. 1 Without question, evacuation can be effective if accomplished in time.

Evacuation before the release can essentially eliminate exposure. Evacuation  ;

after a major release would be of lesser benefit but the benefit can be sub- i stantial. All available evidence indicates that evacuation is easy and i effective in most instances in the United States, although it may be traumatic psychologically. Large numbers of people regularly move into and out of even l

l densely populated areas. There are exceptions, of course. Evacuation may be l difficult in a snow stonn and some people, such as prison inmates and invalids, may pose special problems (which are be recognized in the plans). Even so, evacuation is a first consideration. )

l Sheltering may be less effective than evacuation. Most people are indoors )

most of the time anyway. The value depends on the kind of shelter available {

and the kind of release. Sheltering, however, is relatively easy and could be  !

the only recourse in some instances. Shelter can be effective in larger j cities, particularly in countries more densely populated than the USA, where i expeditious evacuation may not be feasible.

Potassium iodide can reduce the dose to the thyroid from radiofodine. That j is the only potential benefit and it could have undesirable effects, such as

~

misleading people about their need to take other precautions. Potassium l fodide is a limited option.  !

Decontamination and medical treatment should be provided as nn@d but they should be used on a case by case basis. In any event, they a not early actions. j Emergency planning Zones i In the U.S., system, there are two emergency planning zones.[19] The area within about 10 miles of the plant is referred to as the "Pluine Exposure Emergency l Planning Zone" (EPZ). The EPZ is the area of primary concern regarding poten-tial exposures to the plume. Beyond the EPZ, out to about 50 miles, is the )

area where the principal concern is potentially contaminated foodstuffs.

~

These distances were selected somewhat arbitrarily and were deliberately j left open to accommodate local conditions. The principal reasons for the 10 j mile plume EPZ are: l l

1. Calculated doses from design basis accidents are below the PAGs beyond I about 10 miles; j i

Calculated doses from most core melt accidents are below the PAGs beyond 2.

about 10 miles; and l

3. For the most severe accidents, calculated doses generally are below lethal i levels at this distance. {

l It is also recognized that protective actions could be extended beyond 10 miles if conditions warrant and much more time would be available for emergency response beyond these distances. Another important consideration is that 10 miles is a practicable distance for planning in the United States, a condition that may not be valid in more densely populated countries.

1

_ l

1

• 1 E .) '

Impact of the New Analyses on Emergency Preparedness The new accident analyses [16] ' represent' substantial' advancement in several

~

areas.. The implications for emergency preparedness, however, are less than

. might have been expected. Perhaps the most important conclusion is that large early releases given a core melt still .seem possible. Thus, provisions must I be made :for coping with such releases, so the scope of emergency preparedness requirements cannot be reduced in the U.S. at this time.

~

A second very important conclusion is that the times and magnitudes of releases can not be _ predicted by the operators with confidence in real time. After ';

over .20. years of study, there remains considerable uncertainty and - contro-versy among experts regarding the timing and magnitude of source terms..  ;

There is important positive information. First, there should be at least two hours warning before a major release, so early emergency response could prevent-fatalities ' regardless of the source' term. The new analysis describes core melt accidents as unmistakable by the time core damage starts. With the cool-ing provided by natural circulation, melt-through and a substantial radioactive

- release to the environment should be at leas't two hours away. This time can be used effectively by people if they are provided early warning.

Secondly, early evacuations within only about two to three miles of a plant can substantially reduce the conditional risk of early fatality and injury, regardless of the source term.

This' is consistent with the perspective presented in the NRC/ EPA Task Force report [7] that risks decrease markedly within two to three miles --'and slowly beyond this distance. Fortunately, much more time would be available for radiological ' monitoring teams to identify hot spots beyond this distance --

where relocation from. shelters might be prudent or necessary.

- The overall implications of these studies for emergency preparedness are:

C 1. Don't delay an evacuation within two to three miles if a core melt acciden't in indicated.

2. Monitor for hot spots as soon as possible after a release.

It is important to remember that core melt accidents are not expected. Thus,  !

a protective action scheme based on initiating early precautionary evacuations for a core melt sequence, but not otherwise, would accomplish as several things simultaneously:

• Planned evacuations should be very rare because they are warranted only by ,

core melt accidents. j

• Prudent, precautionary evacuation would be initiated early, for cause (e.g., core melt).

I l

_____m _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

1 1 c, '

.c l'

- 8'-

• Early evacuation within -two to three miles considerably reduces risks (by-a factor of 10 or more).

Hot spots beyond two to'three miles can be readily identified after a release and relocation from them can. be accomplished in a more leisurely 4

manner to accomplish a dose savings objective.

l' Thus, we are confident that the basic radiation protection objectives of the plans would be met if a severe accident were to occur.

1

=

1

_ _ _ - _ _ _ _ - - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ l

9 REFERENCES

1. Parker, H. M..and J. W. Healey, " Environmental Effects of a Major Reactor Disaster", Proceedings of the International Conference on the Peaceful Uses of Atomic Enerav. New York: United Nations, 1956

2. Beck, C. K.,et al., " Theoretical Possibilities and Consequences of Major Accidents in Large Nuclear Power Plants," Atomic Energy Commission Report WASH-740, March 1957

3. " Reactor Site Criteria," Code of Federal Regulations, Title 10 Part 100, Issued April 12, 1962, Amended 1966, 1973, 1975, 1977 and 1984

4. " Manual of Protective Action Guides and Protective Actions for Nuclear Incidents," Environmental Protection Agency Report EPA-520/1-75-001, DRAFT, September 1975, Revised June 1980 and June 1986

5. Rasmuscen, N., et al., " Reactor Safety Study: An l

Assessment of Accident Risks in U. S. Commercial Nuclear Power Plants," Nuclear Regulatory Commission Report WASH-1400, October 1975

6. " Emergency Planning," Section 13.3, " Standard Review Plan," NRC Report NUREG-0800, 1975, Rev. 2, July 1981

7. Collins, H. H., et al., " Planning Basis for the Development of State and Local Government Radiological Emegency Response Plans in Support of Light Water Nuclear Power Plants," NRCa Report NUREG-0369, December 1978

8. "Fmergency Planning and Preparedness for Production and

' Utilization Facilities," Appendix E to Code of Federal Regulations, Title 10, Part 50, Issued August 19, 1980, Amended 1981, 1982, 1984, 1986 and 1987

9. " Criteria for Preparation and Evaluation of '

Radiological Emergency Response Plans And Preparedness in Support of Nuclear Power Plants," NRC Report NUREG-0654 (also FEMA-REP-1), Rev. 1, October 1980; Supplement 1, November 1987

10. " Guidance on Offsite Emergency Radiation Measurement Systems Phase 1: Airborne Release," Federal Emergency Management Agency Report FEMA REP-2, Rev. 1, July 1987 i

d

_ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ - _ _ _ _ . _ _ . . _