ML20212P714

Paper Entitled, Sensitivity of Collective Dose (Person-REM) to Interdiction Criteria,Population Distributions & Source Terms, Presented at Sixth Intl Ans/European Nuclear Soc Topical Meeting on Thermal Reactor Safety on 860202-06

ML20212P714
Person / Time
Issue date:	02/02/1986
From:	Martin J NRC
To:
Shared Package
ML20212P709	List: IA-86-849, Paper Entitled, Effectiveness of Early Evacuation of Small Areas,Shelter & Relocation in Reducing Severe Accident Consequences, Presented at Fifth Intl Meeting on Thermal Nuclear Reactor Safety on 840909-13.Related Info Encl ML20212P706 ML20212P714 ML20212P743 ML20212P770 ML20212P823
References
FOIA-86-849 NUDOCS 8703160213
Download: ML20212P714 (9)

Text

_ _ _ _ _ _ _ . - . -

! Praprint for Proc:edings of the Sixth Intsrnational ANS/ ENS Tcpical M:eting  ;

' ~

on Th2rmal Rractor Safety. February 2-6, 1986, San Diego, CA l

SENSITIVITY OF COLLECTIVE DOSE (PERSON-REM) TO INTERDICTION CRITERIA, '

POPULATION DISTRIBUTIONS AND SOURCE TERMS 3

James A. Martin, Jr.

3 U.S. Nuclear Regulatory Comission a Washington, D.C. 20555 INTRODUCTION The proposed Nuclear Regulatory Comission safety goal l includes a cost / benefit guideline of $1000 per person-rem averted as one consideration for safety improvements. How person-rems are to be computed is not stated. For most

' probglistic risk assessments perfomed in the United States, the CRAC2  ;

code or an analogous model has been used for accident consequence assess-ments. It is worth examining results using this model to detemine the parame-ters and submodels that influence the person-rem calculation.

I 3

' For severe accidents, CRAC2 trades person-rem against property damage costs, using a land interdiction projected dose criterion. For purely noble gas releases, the collective dose is calculated for the first pass of the full release. For more severe releases which include sizable particulate release fractions, notably cesium, property is either decontaminated to an acceptable level, or its use is interdicted.; the sum of the costs of these options is the property damage cost estimate. People are allowed to reside in areas for which the projected dose is less than a user selected value, esuelly 25 rem in thirty years (after decontamination, if any). Person-rem is accumulated predominantly in the non-interdicted areas.

Radiation protection criteria for reentry of contaminated areas has 4 not been

. developed in the U.S. However, ICRP protective action guidance states,that an ALARA concept should apply. This paper will discuss the results of an inves-

. tigation of the sensitivity of person-res estimates using various source tems, a few selected LWR sites in the U.S., and various interdiction criteria.

CALCULATIONS

! The equations used for the calculations are those embodied in the CRAC2 code, l with two exceptions: A 50 mile radius was used rather than the standard 500 he Safety Goal. Also, the washout mile radius for coejficient in the code, rainfall to confom was set to 10'to g/sec per m/hr of rainfall rate fvs 10~ in CRAC2) to confom better to the most recently published infomat' on .

Rainfall can enhance property damage and radiation exposures.

The input data to the model (CRAC2) was comprised of the population dis-tribution and land use data in the standard CMC 2 data sets for the Indian Point. Zion, and Sequoyah sites, the source terms in Table 1, and the New York City meteorological sequences, also in the standard CRAC2 data sets. Earlier studies have shown that person-rem results 6 are insensitive to the choice of weather sequences across the United States 8703160213 870312 -

PDR FOIA CURRAN 86-849 PDR 8_ L_:_ .

m

'.o .

2 i

The accident source tems were chosen to span a range of release fractions and l mixesofradionu:1gdes. Calculations were made for the NPC's SST1 and SST2 accident s9enarios , and the TC sequence presented in Table 4.13 of NUREG-0956 . To span a useful range of release fractions of particulates (non-noble gases, more precisely), release fractions both ten times higher and ten times lower than those in SST2 were used, except for noble gases, which were not changed. Noteworthy in these groupings is the relatively low Cs/Ba ratio in the severe TC sequence, i .

It has been known for some time that for source terms of this ilk person-rem is accumulated over long distances (20-200 miles) and long times (decades). Evac-

~

untion scenarios within ten to twenty miles have little effect on person-rem estimates. Fgr these calculations, the so-called "sumary" evacuation in NUREG/CR-2239 was used (in essence, a three liour delay and 10 mph evacuation spreed,withintenmiles). Shielding and sheltering assisiptions used were also I, as described in NUREG/CR-2239, except that the ground shielding yond ten a miles for people in dwellings was set more realistically to 0.33 vs 0.08 in NUREG/CR-2239). All of the calculations were made for a power leve of 3320 themal megawatts.

This completes the description of the input data and the equations used for the calculations.

RESULTS Sumary results of the calculations are displayed in Tables 2, 3 and 4. Mean whole body person-rem and mean interdiction costs are displayed for various

' source tems, land interdiction criteria and population distributions (sites).

Person-rem is pertinent because it is the parameter of interest in backfit considerations. Interdiction costs are displayed because this represents the l . value of the land and property that would be abandoned because of long-tem contamination. Comparison of these would be necessary under ALARA. The inter-diction criteria used in this study were chosen on the basis of results from a few trail runs.

These results can be viewed from a number of perspectives, some of which will be discussed. Results for the non-interdiction case are displayed in Figure 1, wherein average dose is shown vs the release fraction of Cs of the various source terms. Mean person-rem Ts almost directly proportional to Cs release fraction over almost a three decade range. In this Figure, mean person-rem has been divided by total populations within fifty miles of the three sites to obtain average dose.

i Average doses for Zion and Indian Point are heavily influenced by large popu-1ations at thirty to fifty miles from these sites, i.e., Chicago and New York City, respectively. For the Sequoyah site the population distribution is more uniform. Thus, the average dose for Seouoyah is somewhat higher than for Zion and Indian Point. The simplicity of the proportional relationship is intrigu-ing, none the less, because it indicates that, to first order, person-rems is

1 - .

I

~

3 proportional to curies released, tgtal population and power level. This has been observed for routine releases , also.

Realistically, should one of these accidents actually occur, interdiction of some areas would occur, especially for the larger source tems. Interdiction would significantly affect property, people and radiation exposures. Several perspectives on this aspect of the matter can be obtained from the results in the Tables. In Figure 2, mean collective dose within 50 miles vs Cs release fraction is displayed for the Sequoyah site, for three interdiclTon levels.

Man-rem ranges over nearly two order of magnitude for the more severe source terms (TC, SSTI), depending on the interdiction level used. Moreover, average person-rem is highest for the TC sequence, even though the Cs release fraction is less than that for SST1 (0.14 vs 0.67). This is due to a combination of at least two factors. First, the higE Ba release in the TC sequence contributes

' significantly to the whole body dose. Second, a more subtle effect of the

- interdiction criterion effects the result. It has been well known amongst accident. consequence practioners that if the Cs release fraction is lowered

- slightly from that in SST1, person-rem increases! This is because CRAC2 then allows somewhat more people to reside in contaminated areas, thereby increasing person-rem, somewhat. Clearly, this cannot continue as the Cs fraction is reduced more and more, and person-rem eventually declines. This should be pursued further for source terms with even lower Cs/Ba and Cs/La ratios.

The effect of interdiction criteria on average dose is shown in Figure 3, for the five source terms studied, also for the Sequoyah site. The interesting perspective here is that person-rem is 'almost linearly related to interdiction level for this relatively uniform population density site. The slight curva-ture of the points in this Figure are more pronounced in similar plots of the analogous data for Zion (and Indian Point), as illustrated in Figure 4, but linear fits to the points could be acceptable.

The ALARA perspective is best illustrated in Figures 5 and 6. Here, person-rem vs interdiction cost is displayed. The points displayed are labeled as to Tnterdiction criterion (dose in 30 years). These results suggest that an interdiction criterion of 10 rem (in 30 years) would be a better choice under ALARA than the 25 rem most often used, i.e., below 10 rem estimated costs increase dramatically with only relatively small decreases in person-rem as a benefit.

DISCUSSION l

Person-rem estimates are dependent upon a number of variables. For assessments I made pursuant to NRC's Safety Goals for backfit considerations, the estimated benefit (person-rem saved) could well range over several orders of magnitude depending on the assumptions used for the calculation. In it's Policy and PlanningGuidance,theCommissgnhasstatedthatemergencyplanningshouldbe I

based on realistic assumptions . Realistically, some interdiction of land would occur in the wake of a severe source term. The ICRP reconspends that an ALARA approach be planned. Results of this study suggest that a projected dose of 10 rem in thirty years should be an acceptable interdiction criterion on an

,._,-,.r --

.g., .___,,.,,,-,___-...n- - - , _ - , , , , , . - , ,._..,_,.,,._,-_.w.-- , e, .w,_,.,,.,., . , - , , . _ . . , _ _

l

. .. l 4

ALARA basis for most source tems; the often used criterion of 25 rem /30 years

results in a CRAC2 person-rem estimate almost twice as large in some cases.

However, for Safety Goal applications, person-rem could be calculated using a number of options: e.g., doses for only the plume exposure phase, or only external ganna dose, and with or without emergency response assumptions. The method of choice is a policy matter directly related to choice of the cost figure (e.g., $1000) and the objectives of the backfit clause in the proposed Safety Goal.

Lastly, results using a wider range of Cs/Ba/La release fractions should be investigated. However, CRAC2 may not be an appropriate model for very low Cs/Ba and Cs/La source terms. In CRAC2 interdiction is basgd on only ten -

nuclides and only the ground shine (external gamma) pathway . The NRC is sponsoring the development of a model (MACCS) to replace CRAC2, in which all nuclides and three pathways (ground shine, inhalation of resuspended material, and ingestion) will be considered for intervention. Once developed in 1986, MACCS should be used to perfom ALARA studies. This emphasizes the main con-clusion of this study--for Safety Goal purposes the method for calculating person-rem must be specified. The information resulting from this study may be of some help in this regard.

References

1. safety Goals for Nuclear Power Plant Operation. Rev 1 for casument.

' NUREG-0880. May 1983.

2. Calculations of Reactor Accident Consequences Version 2--CRAC2: Computer Code. Users Guide. EURES/CR-2326 February 1983. sandia National Laboratories.

3. CRAC2 Model Description. NUREG/CR-2552. March 1984, sandia National Laboratories.

4. Protection of the Public in the Event of Major itadistic Accidents: Prin-ciples for Planning. ICRP 40. May 1984. Internattonal Ceumissfon on Radiological Protection.

5. James A. Martin Jr.. Objectives of Emergency Response and the Potential

. Benefits of Evacuation and shelter. Proceedings of the Eighteenth Mid-Year Topical symposium of the Health Physics society. Col. sprint. CD.

January a-10.1985. James A. Martin, Jr. Effectiveness of Early Evac-untion af ses11 Areas, shelter and Relocation in Reducting severe Accident Consequences. Trans. Am. Nucl. soc., v. 46, pp. 71-72. June 1984.

6. Technical Guidance for siting Criteria Development. NuREG/CR-2239 December 1982. sandia National Laboratories.

7. Reassessment of the Technical Bases for Estimating source Terus.

NUREG-0956. July 1985 draft report for comment.

8. overview of the Reactor safety study Consequence Model. NUREG-0340.

October 1977.

9. J. A. Martin, Jr. C. B. Nelson and P. Cany. Calculations of Doses.

, Population Doses and Potential Health Effects Due to Atmospheric Releases of Radioncildes from U.s. Nuclear Power Reactors During 1971. Radiation Data and Reports. June 1974 and August 1974

10. Nuclear Regulatory commission Policy and Planning Guidance 1984 and 1985 NUREG-0085. Issues 3 and 4. respectively. January 1984 and February 1985.

1

5

% 44 Table 1. Somete Toms TC $$T1 1055T2 $$T2 0.155T2 telease Ttee 1.5 1.5 3. 3. 3.

(hrs) (Acclernt lnttletten to start of release)

^

Seteese Deretten 10. R. I. 2. 2.

(hrs)

Iserntag Ttee for 0.5 0.5 1. 1. I.

tvecuatten(hrs hefererelease)

EttA$t FRACTite5**

Re. Er 1.0 1.0 0.9 0.9 0.9 1 0.18 8.45 0.03 0.003 0.0003 ts,ab 6.14 0.87 0.00 0.009 0.0009 Te. Sb 9.40 9.64 9.3 0.03 8.003 ,

0.39 0.07 0.01 0.001 0.0001

, Os. Er I m. .tc. e.cor e.05 e.ar e.cor 0.000r 0.009 0.003 0.0003 9.00003 to. etc 9.010

• • e55 Grouptags, he tecludes tie. N. Oh. Tc. Le troup tecludes fore earths est treasurentes. .

• zero releau energy; 10 meter reiseu height--eil seurte terus.

Table 2. Results Secenry INDIAN POINT Mean Interdiction 50 Mile - Mean Level (Oose Source Whole-Body Int'erdiction in 30 res.) Term Person-Rom Cost (5) 2.(

.- No Inter- TC 0.

.. . diction $5T1 6.9 ) O.

1055T2 1. ( 0.

- .- 55T2 1.1 I 7) 0.

0.155T2 1.4d6) O.

~

50 TC 1.2 f 8;i 1.7 l'8; l 55T1 -

1.3 i ',8 l 1.0 ll9p

. 1055T2 5.9 i 1.1 L8 l (lli 5.(6)

~

55T2 8.6

. 0.155T2 1.4 1, ll 6. (2) 25 TC 9.0 ;I 3.8 I; ll

~

. 55T1 7.4 J 3.5

. 1055T2 4.6 p 2.5 i( p i 55T2 8.1 1.6 0.155T2 1.3 h l

2.5(,lLI l

10 TC 5.5 l'7;l 1.1 91 55T1 4.1 L7 l 1.4 ((10!i 1055T2 2.7 l'7'i 6.6 8'l 1572 7.3 461 5.1 7h 0.155T2 1.3I,6h 1. ( )

1 TC 2.2 ll 1 ' 3.1 ll 5 '

55T1 2.1 I I 5.8 i l 1055T2 7.6 d h 2. d j

$5T2 3. I'6 6.5 (8 1 0.155T2 1.d6 5.1(7)

Note:

10,650,000 people within 50 miles.

1 i

l i

Tehle 3. Rosetts Sommery Table 4. Roselts Sammery l tl0N SEQUOTAN

• Meen g Neon i Interdittlen 50 Atle Meen 1sterdictfen- Se M11e Meen

]

Level (gese Seerte Whole-Sedy laterdiction Lees)(gese Seerte Whole-Gedy laterdictfen Cost z,

to 30 yrs.) Teve _ Persen-Ree in 30 yrt.)

h Persen-See Cost l

.Re Inter. Tt 7.31 7 9. As later. TC 2.1 ll7'l 0. .

diction $5T1 f.5 I ;8? t 8. ggetten $5T1 6.41 7'l 9 1055T2 3.71,7 i S. 195STt 9.4 i,6 '

O.

SSTt 3.9 i 6 i 0.

• SSTt 1.9 l'8 '

O.

0.155T2 5.515.1 9. 9.155T2 1.5 ll5,i S.

90 TC 3.3 '" 3.2(8? 30 ft 7.8 L'6'l 1.6 ITl SST1 3.0 f7h S.3 t 3, '

SST1 5.9 l 61 f.7LSL e 1055T2 1.4 I: 7 l 7.5 l'7, '

195572 3.5 I 6' i 1.1 I 7' '

$$T2 f.4 i 6l 1 5.11 6 .

$$T2 8.4 l l'1 9.51 4?

9.155T2 4.91;5? 5.5 l'4 l 0.155T2 1.5 i 5! i 9.7 12,i 25 ft 2.3 i 7 l 2.4 FSS is TC 5.4 fel 6.41 ;e?7 SST1 1.8 l'71 1.11 9 i SST1 3.81;6h 5.5 I i 10$$T2 7'l 1.6 I SJ 1055T2 f.3 1,61 4.11 7 '

SSit 1.1i i'S' 2.2 1.21 7 i SSTt 7.3 I 5'i 5.71 ,5l 1 0.1$$T2 4.7 l'l 1 2.8 L5,i 9.155T2 1.5 ll5 1 4.5 I,3 l

3g TC 3.4 ll71 5.9 {SI' 3g TC 3.3 l'6'l 2.7 I I '

SST1 1.8 i Ph 3.48, '

SST1 f.6 i;6' i 1.2 ' ?

105ST2 6.4 l 6 i 3.7 I;8, '

195512 1.3 i 61 1.7 i ?

$$T2 1.8 I l6 I a 3.31 7 i 5572 5.5 L5h 3.5 l'6 I 0.155T2 4.4 LSD 1.8 L,6 '

0.135T2 1.3 flh 2.8 f81 i TC

$$T1 5.5 ll61 5.71,61 7.2 (9 1.4 (19l l

3 ft 1.0 Col 1.9 l'S'i 2.1 f91 2.21 9, '

? SSTI 3055Tt' *2.(6) 4.4 ll9? 305ST2 5.5 l l' i 1.5L9,

$5T2 7.7 ll51 3.7 L8, '

$$72 1.6 l' S' ' 1.71.8 i S.13572 3.41 SD 3.31,7, '

O.1SSTt 1.s 3) 3.5 ll6,1 Rote -

.~ Roter 6.970.000 people within 50 stles. 691.000 people within 90 stles.

l

7 0

• e El.F E.

85t

........ i........  : .*

33b m... ..

~ o  :

. I'4s t 1

)

s, 35 o D 4 r 4 1 k E

- j)~3.4 J . 5

. 3*.

r

~

2sb5 r [ a 41 8 g.g3s3

"  : 2 E

i I . s-ltt,3

. Eea. -t

. oa4 -

Ti g 8 a.e. . . . A !a.... .. . _

• 1... . . . .

$ ) D D

• mmuseo amnemo una g

c 55

........ . , . . . . . . . . .- jg

: eit,,

lg

. 34

. in 28 -

=

D s, 3 l E04OI

, .s

: s-

15

- J of

. $g g.

.e.

0

. yh D

. g Ei e:

......... ......... ......... . . . . . . . g ~

I *

• 5

- --. .. g

=

_ s -.

r .- _ ,,c-_- _.._. .-. , - _, . - . - _ _ _,,-_y- _ , _ . _ , . , . - , . _ . _ _ . . . - _ , - . _ - , . - _ , _ . _

E 8

i e i i i i i i 3

o l.

e 1-

.i W

, gl 'i ils Rt I et

{

4 y i g . E b

.3 l

{}

a l

,- I t I e e a e t g ,e e e. e e e e se , e 45584f1851088 W IIIIt&AA4 3000 39VIISAW

. . 1 4

5 5

3*

I i e i i e iiiii,,, s

. p

_ . p< = wi g

E -

k b

.5I l 5

=[ ((

n . .

y jg, l

gj e'.

ff i

e

- -, me.m - - .-.- y----. -.-,,- - - - - -m-- -, ,- --,%-,- - - - - - . - - - - . . . - - . , ~ - - - - - - - - - - , , -.,,-e.,-- - - - - - ~ . _ _ _ _ _ . _ _ _ _ _ _ - - - - - -

.. - .m. -- .

r.

9 s . . .

• • u i s a s a u l

gf M 1A -

EEva

)

g ItN= WrTgnoscTeost LEVEL -

)

enansas vasi I ggL - -

1. as.i esi y u tr "

es -

nu a

I as, u k -

. - N Sesovorm aa -na -

aan "

, u . ,

e t I E A I

'e i a a 4 e e *

, essa annosenos costenn seen c

Figure 5. Average thele body dose within 50 miles interdiction cost, for the $$72 seerte and ti n e j sites. ,

i w , . . i a a a *

~

WL i

W e usv:

l IS am.a m mesenomLawet messasvam '

l e -

e .* -

mm am a so ei.th

" '" d.

s r som sopAse i G 8 I I "

t s ,a u u se u u u is sa as as e

asam annosenomeastenn Figure 6, Average whole body dose within 50 sites 3 seen interdiction east, for the TC source tem and three sites.

r--e.--- .-, , ,- ---- , ,,,----.---,c,,,w,--..-,----------.-.--.v