ML20002B103

Preliminary Probabilistic Risk Assessment for Yankee Nuclear Power Station

ML20002B103
Person / Time
Site:	Yankee Rowe, Calvert Cliffs
Issue date:	12/05/1980
From:	YANKEE ATOMIC ELECTRIC CO.
To:
Shared Package
ML20002B092	List: ML20002B091 ML20002B103 ML20002B106 ML20002B111
References
NUDOCS 8012090434
Download: ML20002B103 (29)

Text

,

Enclosuro A i

PRELIMINARY PROBABILISTIC RISi: ASSESSMENT FOR VANKF.F NITCLFAR POkTR STATION ROWE, MASSACHUSETTS December 5, 1980 By Yankee Atomic Electric Company 25 Research Drive Werstboro, MA 01581 80120-905

4[

+

TABLE OF CONTENTS i

'JbEER

+

INTRODUCTION AND.

SUMMARY

...................................... 1 SITE COMP A RISON STUDY......................................... 4 YANKEE PRELIMINARY.PROBABIc'STIO RISK ASSESSMENT..............

6 i

YANKEE' CONSEQUENCES' ANALYSES.................................

10 RISK OF CONTINUED OPERATION WITH SEISMIC' UNCERTAINTY............................................... 14 REFERENCES.................................................. 19 T A BL E S................................................... :.. 2 0 F I GU RE S..................................................... 2 4 e

i s

e s

li fj 5

i a

+

I 3

m

.-,.4

,,,, - _ _ -.,. -. _. _,,, - - _,.., - _. -.... ~ -.

S

'Introduckion and' Summary 4'

n

'In an effort to better quantify the relative safety of the' Yankee plant, Yankee Atomic undertook.several-studies of the site' and the plant. Using the

.I techniques' developed in the Reactor Safety Study (WASH-1400l a preliminary probabilistic risk assessment was performed to provide insights to the following issues:

'l.-

How does.the Yankee plant at the Rowe. site compare in terms of public hea'lth risk to other nuclear plant sites in the United States assuming the failure rates and release categories of WASH-14007 2.

What are the specific core melt probabilities, release fractions, f

and resultiilg public health consequences of the Yankee plant? -

3.'

What level'of additional risk is associated with continued operation a

while plant' improvements and more' detailed analyses are performed in the next two years?

i The conclusions of this work are:

L i

1.

The low power. level;of the Yankee plant and the relatively low 4

-population of the Rowe area limit the earJy consequences of a core-melt accident from 2 to 4 orders of magnitude less than other operating nuclear-plants in' the U.S. assuming the WASH-1400 failure rates and' release categories.

1 i

i i

I 1

m 4

. 1

'S

- 2.

Despite the fact:that-the~. Yankee plant'is in,the populated Northeast, the latent health' consequences of a core melt at Rowe are no greater than for other operaf,ing nuclear plants in the U.S.

. assuming th'e WASH-1400 failure rates and release categories.

3.-

The preliminary'probabilistic risk' assessment study determined that the overall probability of a core melt and release for the Yankee -

~

plant was 1.4 x 10 reactor year compared to WASH-1400's

-5 5 5 x.10 reactor year. Although this study was not exhaustive, the result supports Yankee's inherent advantages of simplicity 'of design, large containment volume, large steam generators, pressurizer, and a larger coolant flow to power ratio.

4.

The early public health gonsequences based on the Yankee specific j.

core melt probabilities, release fractions, population and meteorology are 1 to 2 -orders of magnitude lower than WASH-1400 typical PWR results even if no evacuation is assumed for Yankee.

This is due to a lower core melt probability, lower source term 1

resulting from a lower power level and a smaller local population.

(

5 If realistic evacuation time estimates are assumed for the core melt' sequences, no early fatalities at all.are expected.

6.-

Tne latent or celayed publir health' consequences based on Yankee

. specific core melt probabilities,' release fractions, population and meteorology range from 1 to 2 orders-of magnitude lower than

'1 4:

, ~

4 NASH-lN00 typical PWR re'sults for a composite site.. -These results are consistent.with the lower Yankee source term resulting from sj lower power level as well as the. lower probability of core melt even

- though the population of the Northeast within 500 miles of the Rowe site is relatively large.

7. JA bounding conservative assessment of the additional. risk introduced by a seismic event. causing core failure, leads to -the conclusion that two years of additional work and study atsociated with planned

' plant improvements and fur'ther analysis to better define the seismic design capabilitics of the plant contribute only about 20% of the additional lifetime risk of Yankee if it were operated to 1997 t

8.-

The overall conclusion based on these preliminary analyses is that the Yankee, plant due to its design and location poses less risk to the public than most operating nuclear plants.

4 I

4

\\

4 -

i t

5 ~

~

I s

f Site Comparison Study As a first step in estimating the. risk lof operation or lankee,' a site comparison was;made using the. techniques of the Reactor Safety Study

(WASH-1400) (Ref.1). ' The' CRAC (Raf. b) computer program utilized in that study,' was used to compare the Yankee site with several other nuclear power i

stations inLthis country. The other' stations were chosen to'present a broad'-

spectrum of population densities. Each site presently contains an ' operating

~

power reactor. ~ Table 1. list the sites used for this comparison with a summary of their population data (Ref. 3). The resulting risk curves, defined as either early fatalities and latent cancer fatalities, are shown in Figures 1 and 2.

l i

The curves are.all based' on the fission product release categories, the time of release, and the probabilities of these releases, as calculated in WASH-1400 for pressurized water reactors. Identical 91. weather sequences, based on Yankee on-site data, were used for.all sites..The wind direction frequencies were based on publishe'd results of site specific meteorological data for each nuclear station -(Ref. 4).

Likewise. population data, out te 500 Q

miles was based on 1970 census data for actual site locations.

In all. cases,-

it was assumed that the population.within'10 miles was not evacuated until 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after the radioactive plume has passed, and the population from 10.to 500 miles was not evacuated until 7 days later. The actual power levels of the plants analyzed were input into CRAC to determine core fission product inventories.

s h-

.1

LThe results shown in' Figures 1 and 2 represent,;thereforp, relative risks

. based solely on wind rose weighted-demography and power level. They should-not be interpreted to represent. actual. estimated risk from any of the stations

~

analyzed. The input parameters.used in this site comparison.are sum.marized in Table 0.-

The site crmparison figure ' based on early fatalities shows Yankee to be -

comparable to other -low popul'ation density sites for small consequence events. For events which impact greater than several. tens of persons however, I

the probability for Yankee becomes very much lower than for other sites. This is due to the limited fission p.7oduct inventory of the Yankee core.and the low population density within a few miles of the site. High consequence evt.sts g

j are significantly limited for the Yankee plant because of its size and isolated location. These low probability events are shown to involve 2-4 orders of magnitude fewer persons at Yankee than at other operational sites.

The latent'. cancer fatalities shows Yankee to be comparable to other low 1

population sites. The latent cancers are impacted by high population density areas at large distances from the site. The northeastern section of the country has an above average population density. Despite these population centers at large distances, the total latent cancer risk from Yankee is -no 1

higher than from nuclear plants in-other areas of the country.

i-A 1 -

Yankee Preliminary'Probabilistic Risk Assessment

~

In. order' to quantify the relative safety of the; Yankee plant at the Rowe. site, a preliminary probabilistic risk assessment study (Ref. 5) was performed by Energy Incorporated for Yankee Atomic. Energy, Inc., used the-methodology developed in the. Reactor Safety Study (WASH-1400), to. calculate the

~

probability of accident sequences leading to core melt and associated release fractions. Yankee used these resu'lts in the c +e specific calculation of e

c.onsequences using the CRAC computer code. -

This risk assessment is a preliminary review of the relative safety of -the Yankee plant. The simple fault treos established for primary systems were not on the level of detail normally considered for a full risk assessment. The I

results are, however,. indicative of relative safety and useful:as guidanc'e for policy makers seeking areas for further investigation.

Methodology The preliminary probabilistic risk assessment of Yankee proceeded in the following manner:

All important Yankee plant systems were reviewed and compared to WASH-1400.

The-dominant accident sequences of WASH-1400 were reviewed relative

.to their applicability to Yankee yielding a dominant set of accident sequences for the plant. ~

~

s

^

Simple fault ' wes were' set up for major Yankee: plant systems.

4

~

Key accident sequences important to~ Yankee were identified by reviewing:the dominant accident sequence probabilities.

Core melt and post melt _ phenomenon were examined in detail.

Accident sequences were quantified by calculating overall failure.

probabilities.

Appropriate. source terms from WASH-1400 were applied to the accident sequences and releases were categorized.

If-the Yanke" systems were significantly different from those,found in WASH-1400, system failure rates were recalculated by hand using component failure data found in WASH-1400. The systems specifically examined included:

Electric Power Reactor-Protection System-Main Feedwater Emergency Feedwater

- Safety. Injection Actuation System e

Emergency Coolant Injection Emergency Coolant Recirculation 1

i

? H

A 4

Potential for' operator errors was includea as was~ potential for operator action.to correct. certain failures.,

~

The initiating even's considered were the same as those found in WASH-1400.

. Large LOCA Small-LOCA Small Small LOCA

+

Interfacing LOCA Reactor Vessel. Failure J

Transients - Loss of Electric Power i

i Detailed core melt'and post melt calculations were performed to develop t'

estimates for. release timing and to aid in source term definition in-conjunction with the standard WASH-1400 containment failure modes.

The Yankee plant dces not have an active containment heat removal system nor does it have containment sprays. The Yankee containment heat removal system is a very effective passive steel co'

.nell transferring heat to the environment.

The lack of containment sprays in Yankee resulted in the conservative assumption that the CORRAL (Ref.~ 6) calculations used to estimate the release fractions for a failed WASH-1400 PWR spray' system were valid for Yankee.

It-is clear, however, that due to the larger containment su,rface area.to power

. ratio at Yankee and the fact that the steel containment shell is a very.

effective condensing-surface, t!.e assumption overestimates'the source release

term.

4 The Energy Incorporated study did take-credit ~ for the-new information on the

~

much lower potential for a steam explosion in reactor vessel but it did not i

take ' credit for the more recent ' data by Stratton, Starr, Levenson, et. al.,

. (Re'f.~ 17,. 8, 9). calling for a significant reduction in the iodine and

~

~

particulate source terms available for release from containment.

Due to the fact that Yankee's. containment heat removal system can't fail 'since

- it is passive and that the Opray systems don't exist, certain release.

i categories found in WASH-1400 are not possible for Yankee.

. The seven core melt sequences identified in WASH-1400 are reduced to four for the Yankee plant. PWR 3, PWR 5, and PWR 7 relate. to the functioning of the

_ontainment spray and, thus,-are excluded from Yankee's release categories.

The final release categories, probabilities, source terms and times for release are summarized in Table 3 This information forms the basis for the sit'e specific consequence analysis.

E i

e a

_9_.

Yankee Probabilistic Risk Assessment'Consecuence Analyses The ' release categorie's, and their associated probabilities, as defined by _the 4

study descri~oed:above, were used to calculate cumulative distributions of i

early1 fatalities and latent cancers. These were then compared with the WASH-1400 results as shown in figures 3 and 4.. These curves were developed using the CRAC computer code used in WASH-1400.

4 The weather sequences and wind rose information were taken from the Yankee 9

meteorological monitoring system' This system was designed to meet the requirements of Regulatory Guide 1.23 Ninety-one weather sequences

~

(start-times) were used in this analysis during the period 1/1/78 through 7

12/31/78.

The Yankee site is located in a river valley and its topography is reflected Ein the wind roses, which show-high frequencies of up-and-down-valley winds. -

Tne valley itself makes sharp turns within a few miles of the site, in either direction, and therefore, crosses several wind sectors. To conservatively assess the risk' to the close-in population, the number of people living in the i

valley, within 5 miles of the site, were determined.

These people were then relocated, for. calculational purposes, into one of two predominant wind sectors. The total valley population is, therefore,' assumed to' reside'in high-I frequency wind sectors.

i l

The U. 'S.spopulation'within 500 miles of Yankee Rowe was determined for 1980 i

i.

from the preliminary. estimates of the 1980 Census of Population and Housing,

~

.I i

7 l

U. S.. Department off Commerce, Bureau of Census. For.those. areas within 500 miles which fall in Canada,' the most recent available Canadian government

-popul'ation statistics were for 1978, and' were.used in' conjunction with the

'1980 U. S. estimates...

I.

The populat' ion 'in the study. area was divided into 544 area elements formed by i-the 16 principal compass directions and concentric circles drawn at radii as.

~

i specified for input, to the CRAC code, and' o'terlayed onto maps representing the various-regions lwithin 500 miles.

i 1

}

For towns all or partly'within 10 miles of the site, the 1980 preliminary population estimates indicate that the area population has actually decreased by about 3 6% since 1970. The 1960 population within 10 miles of the site'was distributed into each area element for input to CRAC in accordance with the

{

~

j distribution of buildings as shown on the most current (1971 and 1973) topographical maps of the area.

From 10 to 150 miles the resident population was distributed by the fraction of each town's or county's population which falls within each sector defined by the grid of concentric circles and radial lines was assigned' to that area l element.

i.

From'150 to 500 miles equal area allocation, a' uniform population density on a i

' state-wide basis.was applied for those states which were contained within the 150 to. 500 mile area.

m y

e-e 1

y-g-p e'

=

-rin

' gw ygHe-e

=

-'"r-w--

g--=

g ww n - se

.g-The two Yankee curves shown in Figure 3 represent different. assumptions.

regarding evacuation.cf_ the population within 10 miles of the site. The evacuation curve conservatively assumes that the. 20 mile pc N1ation evacuates radially outward with an effective speed of 5 mph. It furthe assumer.that

~

the start of this evacuation is delayed for 30 minutes following th0 decision to evacuate. A r6cer'. emer'gency planning study (Ref.10) evacuation speeds of 20 mph during good weati.3r and 10 mph during bad weather were possible. The 5 mph value for this analysis was chosen based on the fact that most roads do not lead directly away from the site.

I The no evacuation curve assumes that no evacuation takes place within ten miles until 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after'the radioactive cloud has passed over. It is included here simply to show the effect on the risk analysis of having no 1

emergency plan. It represents an upper bound on the risk from an accident and

]

is useful only for comparative purposes.

In all cases the population beyond 10 miles was assumed to have been exposed to the radioactive cloud and to 7 days of ground plane radiation. Figure 4 shows only one Yankee mirve since the evacuation assumptions used had-very little effect on latent cancer risk. The input parameters for this analys!'

are summarized in Table 4.

J The curves of early fatalities shown in. Figure 3 demonstrate the extremely low level of risk due to' severe accidents.at Yankee. With a conservatively-estimated evacuation the risk of an accident that produces even 1 fatality is,

.~ _

less than 10" per year...This is approximately 4 ordera of magnitude below the WASH-1400 risk estimates. Even assuming no evacuation of.the population, the' risk'from Rowe is almost an order of magnitude below the " average curve" l

'of WASH-1400.

Likewise the risk of latent cancer' fatalities, as shown in Figure 4,.is-significantly lower-for Yankee than for the WASH-1400 average site. The total

~

risk, from both early and ' late fatalities, is estimated at 5 4 x 10 per year for the-population out to 500 miles from Yankee, from these curves. Th s compares with the cancer. fatality. risk from normal background radiation

.t f

exposure of approximately 1200 people per year for this same population of 74 million people.

1 I

i 4

y i

9

.f a.

n w

w v-e*=

Risk of Contin'ued Operation With' Seismic Uncertainty The preliminary probabilistic risk assessment of the Yankee plant performed in 2

accordance with the! methodology of.the Reactor Safety Study-(WASH-1400) indicates that the'. risk to the public from continued operation of' Yankee was several!of orders. of magnitude lower than the typical plant analyzed in L

WASH-1400. At the present time, a seismic probabilistic risk assessment has not yet been performed although one is planned.

In an attempt to provide the decision maker with some guidance as to the level of additional risk a seismic event poses to the public health and safety, the folloeing conservative and bounding analysis is presented to assess the risk of continued operation for the next two years while more conclusive and detailed assessments can be made of the seismic soundness of the structures and equipment of the Yankee plant and the overall seismology of the Rowe area.

T 1

r.

x m.

_ Conservative and Bounding Analysis ASSUME:

4 1

1.

The probability of an earthquake of. causing system degradation is

~0 10 / year.~

~

2..

Assume that regardless of the spectra finally determined, the reactor core and its cooling systems are sufficiently degraded to cause a core melt.

T 3

The containment structure does not-fail..The containment with modifications presently being made will withstand a Regulatory Guide 1.60 spectrum.

THUS:

1 Based on the very conservative assumption that this seismic event at 1

^

the plant site causes a core melt,'the core melt probability is

~

assumed to be 10 / reactor year.

Since the containment is capable of withstanding the postulated

seismic event, the containment fe ure modes revert to the

~

probabilistic failure sequences'h=aociated with random core melt sequences. Thus, the: containment-failure probabilities are those.

associated with' WASH-1400 failure modes for the key accident sequences. Namely, core' melts causing 1 Yankee specific containment failure _due to:

, 1

~

'Rt2cter Vassol Sts:s Explosion 10-3

-3

~10

.Fallure to Isolate

~

~

. Hydrogen' burn ~

10

~

10 Overpressure

-Vessel Mel't-through 10"

~

The. major public health concern is the relatively. rapid release of fission products.,.This consideration excludes the vessel. melt-through phenomenon which takes on the order of 50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br /> for Yankee.

Thus, we are then left with a containment failure ' probability of 10 to 10 Let us assume 10-1

-3

~

~

since there is some probability of a seismic event causing damage to piping

. systems leaving the containment.

Thus, the probability of a seismic induced event causing a core malt leading to a relatively rapid escape of fission products to the environment is:

Probability of Seismic Probability of Probability of

.Frobability of.

Release

= Seismic Event X

Core Mcit X. Release of Fission Products 10-3fyr 'X 1.0 X 10-1

=

Probability of Seismic 10-4/ reactor year j

Release

=

i Based on these conservative assumptions and using the analyses of R. Mattson

'as developed in.the August 8,.1980 NRC memorandum to D. Eisenht,t (Reference 11). the additio'nal risk for two years of operation is small in comparison to the lifetime risk calculated in UASH-1400 as 5 5 x 10' -per~

r-reactor year.

q l

i

g..

For i two year. period the prot 3bility of a core melt and release is:

~

+ seismic core Core melt probability = random core melt (for.2 years}

probability (WASH-1400) melt. probability

= 2 years X 5 5 x 10-5 + 10-4 Core melt probability.= 3 x 10-E (for 2 years)

T1e risk of Y nkee operation to the ead of its operating license in 1997 at an

-5 L

1-assumed WASH-1400 risk value of 5 5 x 10 / reactor year (Yankee's actual

-6

-preliminary value was 1.4 x 10. / reactor year):

1 Yankee Lifetime Risk 17 years x 5 5 x 10-5/ reactor year j

of Core Melt and' Release

=

Yankee Lifetime Risk 4

9 4 x 10-4 of Core Melt anc Release

=

If Yankee operated for-the next 2 years while definitive detailed studies _are made to assess the true seismic resistance of the plant, the risk of core melt and release is:

' Yankee Lifetime' Core Melt and Release Peebability including 2 year interim

= Core Melt Probability + Core Melt Probability Operation for 2 years for 15 years (Plant fixes if necessary to reduce probability to WASH-1400 value)

= 3 x 10-4/ reactor yr + 15 x 5 5 x 10-5 reactor yr

/

Yankee Lifetime Core

= 11.2 x 10-4 Melt and Release Probability-including 2 year interim. operation.

Thus, should we make these conservative assumptions, the additional risk for two years of operation amounts to about 20% of the remaining lifetime risk of Yankee operation to ~1997 (11.2-9 4/9 4). '

-~

..me

!If 'we~ compare the1additionali2-years operation 5to' ttie typical' operating' life of

."a.new large platit. analyzed in. WASH-1400, the total 'lifetic t risk of. operations is 75

-4' 40l years'x 5 5'x 10

/ reactor year or.22 x 10. ' The additional 2 years.of Yankee operation _is 105 of'this; total' individual plant risk.

L-l

[1 L

l I-i-

(.

l l

l -~

l f

I-I t

~ i

4 i

REFERENCESE

~ Reactor Safety Study,f"An Assessment of' Accident Risk in U. S.lCommercia'l 1.

Nuclear: Power Plants".. U. lS. Nuclear Regulatory Commission, WASH-1400, NUREG-75/014 41975).

" Overview oftthe Reactor Safety Study Consequence Model", U. S. Nuclear-2.:

Regulatory: Commission, NUREG-0340, October,1977

" Report'oflthe Task Force.on Interim Operation of Indian Point", U. S.

3 Nuclear Regulatory Commission, SECY-80-283, June 12,1980.

" Correlations Between Wind Flow and Population Location at 67 Light Water

~

4. -

Nuclear Power Plant Sites", J. L. Sprung, G. P. Steck, A. W.' Frazier, SAND-78-0557, October, 1978.

" Preliminary WASH-1400 Probabilistic Risk Assessment of Yankee Rohc 5

Atomic Plant", Energy Incorporated, December, 1980.

Appendix VII,'" Release of Radioactivity in Reactor Accidents", Reactor 6.

Safety Study, U. S. Nuclear Regulatory Commission, WASH-1400, October, 1975 W. R. Stratten, A. P. Malinauskas, D. O. Campbell, Letter to NRC Chairman 7

-Ahearne, Autist. 14,.1980.

8.

Stratton, Malinauskas,-Campbell, " Testimony to the Commissioners on Iodine Source-Terms",' November 18, 1960.

C. Starr, M. Levenson, I. Wall, " Realistic Estimates of _ the Consequences -

9 of Nuclear Accidents", Nuclear Regulatory Commission Briefing, Novem;,r 18, 1980.

. Yankee Atomic Electric Company, Letter to B. Grimes, U. S. Nuclear 10.

Regulatory Commission, July 25,- 1980, WYR-80-86 Memorandum from R. J. Mattson to D. G. Eisenhut, U. S. Nuclear Regulatory.

11.

Commission, " Analyses and Recommendations Related to Plante Without Seismically Qualifiec-Auxiliary Feedwater Systems", August 8, 1980.

A

+

'l

-l

TABLE 1 SITI

)MPARISON STUDY Power 0-10 Mile 0-50 Mile Site Level _(MWt)

Population Populttion Calvert Cliffs 2535

-16,800 2,310,000 Farley 24S7 9,500 321,000 Indian Point 2895 218,000 17,470,000 Prairie Island.

1590 19,400 2,060,000 Surry 2364 66,630 1,550,000 Yankee 600 21,800 1,540,000 i

1 l

.. ~.

o TABLE-2 SITE COMPARISON STUDY.

4 INPUT PARAMETERS-V t

i r

Para' meter -

Site Comparison Study

i l-Core' Power Level ~

Plant Specific.

Fission Product Belease Fractions.

k SH-1400

{

- Release Category Probabilities-

-WASH-1400 Release Times WASH-1400 Weather Sequences Yankee 1978-All Sites

~

Wind Direction Plant Specific

Population Plant Specific-1970 Evacuation Delay 0

Evacuation Speed 0 mph 4

9 I

e

[

7 l- '

i

.L

_ ~ --

W TABLE 3 s3 gl..

PRELIMINARY PROBABILISTIC RISK ASSESSMENT Qt

SUMMARY

OF ACCIDENTS INVOLVING Tile CORE D

=

N,td YANKEE ATOMIC Q

(Plant' Specific) b-Duration 18arning

. Elevation Containment Probability Time of of Time for of Energy Freetton of core inventory peleased selease per solease Anlease tvecuation. Delease pelease category teacter=Tr (He#

(Hr)

(Mr)

(Meteral (10 Btu /Hr) 3Ie = K r Org. I I

^ Co-Ab Te-Sb Pa. Sr po La -

6 YR.1 t,,4 10*1 20 0.5 1.0 30 100 0.9 6m 10*3 0.7 0.4 '

n.4 n.ng

' n.'s

't..

1m~ 3

'(1):

Iy' I

1R 2 t.7 s 10*#

2.0 0.5 1.0 0

35 0.9 7m 10~3 0.7 0.5 0.3 0.06

'O.02 4 m 10*3 (2) 10 6 54.0 0.5 54.0 0

0 0.9

'6x 10*3 0.7 0.4 0.3 0.05 0.02 3 m 10 N"

g.19 a

5. H s 10'8 4.5 2.5 0.8 0

0.2 0.6 2x 10*I 0.09 0.04 0.03' 5 m'10*3 3 m 10*3 4m to *

'(4)

Notes:

in WASil-1400.

Not calculated - assumed same source term as PWR 1

.(1) calculated - assumed same, source term as PWR 2 in WASil-1400.

(2) Not in Wash-1400.

Not calculated assumed same source term as PWR 1 (3)

Not calculated - assumed same source term as PWR 4' in WASil-1400.

(4)

. -.. ~...

d' TABLE 4 PRELIMINARY PROBABILISTIC RISK 1 ASSESSMENT I

YANKEE SITE SPECIFIC ' STUDY i-INPUT PARAMETER'COMPARISION TO WASH-1400 Parameter -

WASH-1400 Yankee Specific Study-1

- Core Power Level-3200 Mwt 600 MWt

~

-Fission Product Release Fractions Table VI 2-1

. Table 3

-(WASH-1400) t.

- Release-Category Probabilities Table VI-2-1 Table 3 Release Times.

. Table VI-2-1 Table 3 Weather Sequences Composite Site Yankee 1978 2

- Wind Direction Composite Site

-Yankee 1978 Population Composite Site

' Yankee 1980 1

Evacuation Delay

.0 minutes 30 minutes Evacuation Speed 30% 0' mph 0 and 5 mph 40% 1.2 mph 30% 7 mph 1

4 i

e '

FIGURE 1 PRELIMINARY PROBABILISTIC RISK ASSESSMENT SITE COMPARISON FOR EARLY FATALITIES 5

(WASH-1400 Failure Rates Used For All Plants - Site Specific Power, Population, Wind Rose 10

rr,.,

.ii -

i i..

~

i gan p

4

/

10

[

Prairie Island Y

g4 alvdrt Cliffs r

E P

3

s 2

>. 10 f

',f d

5 Yankee g

~_

=

l.

E E

EARLY CONSEQUENCES OF A CORE MELT ARE 10 LIMITED AND ARE 2 TO 4 ORDERS OF~

x MAGNITUDE LESS THAN OTHER TYPICAL

~

OPERATING PLANTS.

1

-5

-6

-7

-8

-9

-10 10 10 10 10 10 10 PROBABILITY PER REACTOR YEAR > X Note: Analysis based on site specific demograph, wind rose and power level.

FIGURE 2 PRELIMIRART PROBASILISTIC RISK ASSESSIE?C SITE C MPARISOU

' FOR IATEUT FATALITIES 5 (WASH-1400 Failure Rates for All. Plants - Site Specific Power, Population, Wind Rose) _

10,,,,,,,

.10.

yt&

~

c:

0 3

/

x 10 ce 4

5

~.

3 m

i CO g

CM i

8 4 + [c* v ' p 'e " #

4 i

=

f B

sV

/

I LATENT HEALTH CONSEQUENCES 10 0F A CORE MELT AT ROWE ARE TYPICAL OF

~

I OTHER OPERATING PLANTS.

[

~

J bau La a J 1

1.....

i O

10 10 10-6 10-10~

10;

~

~

PROBABILITY PER REACTOR YEAR >,X

]

Note: Analysis based on site specific demograph, wind rose and power level. -

~. -

m f6-

~

IGURE 3-a.

PRELIMINARY PROBABILIST7 AISK ASSESSMENT FOR EARLY FATALITIES.

(PLANT SPECIFIC FA! LURE RATES USED FOR' YANKEE) 5 10 '. Tiill i i i i ll'FT T i.i ti t i i.s i

stiii 1_

i

i s i i i

iiii i i i i

~

4 10 5

.5

~

WASH-ll400,, Average,, Plant.

EARLY PUBLIC llEALTil CONSEQUENCES FOR' YANKEE:

With Evacuation.g

+

.10'3, T u)

-1 to.2 orders of magnitude lower d

I even for NO EVACUATION

~

h f

NO FATALITIES' WITil EVACUATION i

2-J.

i 3

cx:

d' 2

. _...... ~.

E YANKEE Without Evacuation I-f

[, _.

i J

e 10 I

/

YANKEE

/

e' With Evacuationm '. -

i

/

t

/

No

~~

' s, '

~

/

FATALITIES -

/

I I

1 s i i i_il t i

i t i_t. L.1._ L _;.___t LLt.l L _i._ i l_t1J.J t. l. L. -

.# 11.1.L1.1._A 11iI t i I i 10- f' 10-7 10-8 10-9'

10-10 10-1l' 10.

PROBABILITY PER REACTOR YEAR ? x

.a FIGURE 4 a

PRELIMINARY PROBABILISTIC RISK AISESSMENT~

.FOR LATENT FATALITIES (Plant Specific Failure Rates used for Yankee) 5 10 104 E

-w LLJ c.

0. 103 b

g 1

Q L' ASH-lh00 " Average" Plant

,{

u

,/

e w

M 5.

7 E

2

/

5 10 f_

w

/

5

/

x

/

YANKEE

/

/

/

/

/

10

/

/

/

/

LATENT PUBLIC HEALTH CONSEQUENCES

[

/

1 to 2 ORDERS OF MAGNITUDE L0h'ER

/

FOR YANKEE.

)

I i

1 f

10-4 10-5...

10-6 10-7 10-8 10-9 PROBABILITY PER REACTOR YEAR 2 X l

i i

.