ML19241B230
| ML19241B230 | |
| Person / Time | |
|---|---|
| Site: | 07002623 |
| Issue date: | 06/04/1979 |
| From: | Garrick B AFFILIATION NOT ASSIGNED |
| To: | |
| Shared Package | |
| ML19241B218 | List: |
| References | |
| NUDOCS 7907130242 | |
| Download: ML19241B230 (11) | |
Text
e to p
occxcts usnic s
S-2 JUN 51373 > O UNITED STATES OF AMERICA
,,,, grp NUCLEAR REGULATORY COMMISSION {
wg'"""
't
/ggu In the mat *.er of DUKE POWER COMPANY
)
License Amendment for Transportation and
)
Docket No.
)
70-2623 Storage of Oconee Spent Fuel at McGuire
)
Nuclear Station TESTIMONY OF DR. B. JOHN GARRICK My name is B. John Garrick. I am a la -nsed nuclear engineer with 27 yea rs of experience in technical raanagement, systems analysis Iam and design, risk and safety analysis, and reliability engineering.
presently Vice President of Pickard, Lowe and Garrick, Inc., of Washington, D. C., and Irvine, California. My offic e address is California 92715.
2070 Business Center Drive, Suite 125, Irvine, Pickard, Lowe and Garrick, Inc., is a consulting engineering firm Prior experience includes a primarily in the field of nuclear power.
Director and Senior Vice President of Holmes & Narver, Inc., and President cf that engineering and construction firm's Nuclear & Systems Sciencea Group; Physicist and member of the first Reactor Hazards Evaluation staff. U. 3. Atomic Energy Commission; and Physicist, Phillips Petroleum Company at the National Reactor Testing Station in My training includes a Ph. D. and M. S. in Nuclear Engineering Idaho.
from the University of California, Los Angeles. I am a graduate of the Oak Ridge School of Reactor Technology and hold a B. S. in Physics from 5
>6 u,l ',,t
\\y n,
J 7 9 07130,2 t/2 1
B righam Young University. My Ph. D. thesis was one of the first theses on the application of fault tree analysis
- to assess nuclear power plant I have authored approximately seventy papers and reports on risks.
I have taught nuclear power, reliability, safety, and technology.
numerous courses on reliability, risk, and safety analysis, primarily as
.rs and short courses at UCLA and more recently for the special sem Electric Power Research Institute, the U. S. Department of Energy, and several electric utilities.
My involvement in safety analysis of shipping containers began in 1952 at the National Reactor Testing Station in Idaho where I was a member of a team to redesign the final product shipping containers for the Idaho Chemical Processing Plant. My area of work was nuclear criti-cality and accident analysis. My experience has included numerous safety and ris,k analysis assignments:
Project director of a comprehensive risk e.nalysis of i large operating nuclear power plant.
- Expert witness in numerous hearings involving the transport of spent nuclear fuel and radioactive wastes.
Fault tree analysis is a technique of system safety as sessment based on:
(1) a graphical display of events and failures representing possible failure paths to an undesired event (the top event) such as the release of radio-activity and (2) a mathematical analysis of the resulting failure logic to compute the probability of the top event.
M9
\\38 ca n
\\h 2
U. S. representative on International Atomic Energy Agency missions to Pakistan and South Korea on nuclear plant siting and safety.
U. S. representative on USAEC exhibit, Reactor Safety Research and Development, Second International Conference on the Peaceful Uses of Atomic Energy, Geneva, Switzerland.
Principal investigator on several studies sponsored by the USAEC in the areas of reliability and safety.
- Pr; iipal investigator in the preparation of a " Risk Model for the Transport of Hazardous Materials" for the U. S. Army.
- Project director of a risk study for the State of California on power plant siting to the year 2000.
- Project director of a study for the U. S. Environmental Protec-tion Agency, " Transportation Accident Risks in the Nuclear Power Industry 1975-2020."
- Advocate and prime mover for many years in developing and gaining acceptance of the general idea of applying more quanti-tative probabilistic methods to decision questicns involving risk, safety, and reliability.
APPLICATION OF RISK ANALYSIS TO THE TRANSPORT OF SPENT NUCLEAR FUEL In my judgment the risk to the public from the shipment of spent nuclear fuel is extremely small both in terms of the likelihood of a release
< /
3 307
.l)Y 0 )I 3
t
My conclusion is of radioactive material and as to its consequences.
dio-based in general on the excellent safety record of shipments of ra d
extreme active material and the testing progrnm of such casks un er f the shipment of spent
.oads and specifically on my risk analysis o fuel by truck from Oconee to the McGuire Noclear Station.
My analysis made use of several risk studies (References f radioactive materials.
and 3) that have been performed on the transport o (1) the In particular, the analysis performed was primarily based on:
dies into a combining of selected results of several transportation stu f generic risk suitable form for conveying risk; and (2) specialization o studies to the Oconee/McGuire shipping conditions.
In my analysis, I adopted as an expression of risk the frequency that N or more people would receive D or more dose pe In the conduct of the risk shipment of fuel from Oconee to McGuire.
l d
analysis it was important to identify the specific steps invo ve.
Underlying Figure 1 is a These steps are presented as Figure 1.
risk analysis methodology which is outlined as follows.
To begin with, we divided the proposcd rcute between Oco McGuire into sections, and placed each s ection into one of three c From tabulated statistical data, rural, urban, or suburban.
gories:
we then obtained:
(1)
=~ (accident rate on road type r,)*
accidents per kilometer or 4
T*
4
\\ a -
i
l MM -
1 OBTAIN TRUCK ACCIDENT RATES 3
2 REVIEW F AILURE AN ALYSIS SPECI AllZE ACCIDENT OF SHIPPING CASKS SUBJECT R ATES TO OCONEE/McGUIRE TO TRANSPORTATION ACCIDENTS ROAD CONDI flONS i
4 DETERMINE RELEASE QUANTITIES AND PROBABILITIES FOR DIFFERENT ACCIDENT SCENAR'OS I
6 5
^
OBTAIN POPULATION DATA FOR SHIPPING ROUTE OCONEE/McGUIRE ROUTE 7
AGGREGATE ACCIDENT RATES, ROAD TYPES, RELEASE PROBABILITIES, WEATHER CONDITIONS AND POPULATICN INTO A DESCRIPTION OF TRANSPORT RISK 8
PRESENT RISK AS A FRECUENCY VS D AM AGE CURVE (SU RF ACE)
IN TERMS OF DOSE LEVELS TO PEOPLE PER SHIPMENT Figure 1. Sequence of Risk Analysis Tasks 0Y qq'15%
From Reference 2, which is an 2xtanrive study of the consequences of accidents on shipping casks, we obtained:
probability of :adioactive release \\
(2)
I.
p(R) s of type R, given an accident
/
R is a discrete variable which ranges over eight release cate-Each release category (or accident " case") is defined by gories.
specific quantities of various isotopes released.
The dispersion of isotopes, once released, depends of course on the weather at the time. The degree of contamination is measured by a dose parameter 2, which can be related to actual dose, D, using the actual quantities of isotope released. Thus (3)
D = D(2, R).
Let
[ area in square miles contaminated to degree g
\\ 2 or greater under weather condition, w j If we take f, as the fraction of time that each weather condi-tion is experienced in the vicinity of the Oconee/McGuire route, then for a given 2 let p) a f
F(O', a )
w Aaa where the summation is over all weather conditions w, having the property that (6)
A(2,w) aa
-)
.- rm,
\\
i 6
3 e,g.
Thus F(2, a ) is the probability, given a release, that a square miles or more will be contaminated to level G',
or greater.
Now if h; is the populatic;n density in the vicinity of the release, then we may also give F(2,a) the interpretation:
[the probability, given a release in road type r,)
F(2,a) that h o or more people will experience dose i.
(7)
=
r
{
parameter 2 or more
}
Turnir.g this around slightly, by defining N = ha (8) r and
/
i FN?r,9 P (N,2)
I (9)
=
1
)
we have
[the probability, given a release in road \\
type r, that N or more people will I
(10)
P (N,2)
=
(experience dose parameter 2 or more /
Now if we multiply this probability by the probability p(R) of release type R, we obtain:
the probability, given an accident in road r, p
. g)gg) _
that there will be a release of type R, and
, 11)
(
r as a result. N or more people wi'.1 experi-ence dose parameter 2 or more Recall now that we can convert the dose parameter 2 to an actual dose us;ng Equation (3).
We thus write:
P (N, R, D) s P (N, O') p(R)
(12) where D = D(G, R).
ee R
c.F
)b n \\ *s 7
Thus the probability, giv en an accident in road r, that there will be a P,(N, R, D) =
release of type R, and as a result N or more people will get dose D or more If we now multply this by 6, the accident rate per kilometer on r
road type r, and by L., the number of kilometers of type r in the shipping route, we obtain:
6 L P (D, N, R)
@ (D, N, R)
=
= [ frequency, in occurrences per shipment, of N or
\\
more people receiving dos e D or more as a result of (an accident in road type r with release category R.(13)/
Summing over release categories, we obtain:
8
[6 L P (D, N, R)
C (D, N)
=
r r r r g=1
= [ frequency, in occurrences per shipment, of N or )
more people receiving dose D or more as a result (of an accident in road type r.
/
Similarly summing over road types we have 3
8
[ [ e,L P,(D, N, R)
@(D, N)
=
r=1 R=i ^ ~ '
= / requency, occurrences per shipment, of N or more f
people receiving dose D or more.
The quantity @(D, N) as a function of D and N is the final expression of risk. It is shown tabulated in Table 1 for D expressed as whole body dose.
MLA4 v
W 8
T All L E 1.
FREQU ENCY OF N OR MORE PEOPLE RECEIVING D OR GREATER WilOI.E IlODY DOSE FROM ALL C ATEGORIES ON ANY TYPE OF ROAD Do n c, p Number of I)cople, N MREM 1
!0 100 1,000 10,000 100,000 1,000,000 1
1.16x10-4 5.45x10-5 2.77x10-5 1.13x10-5 4.51x10-6 1.68x10-6 4.78x10-5 5.79x10-5 3.86x10-5 1.79x10-5 5.72x10-6 1.90x10-6 3.82x10-7 0.
10 4.77x10-5 2.86x10-5 9.51x10-6 2.86x10-6 4.78x10-7 0.
O.
50 3.76x10-5 1.89x10-5 4.22x10-6 1.37x10-6 0,
o, o,
100 2.76x10-5 9.92x10 6 1.65x10-6 3.82x10-7 0.
O.
O.
500 1.97x10-5 4.89x10-6 5.48x10-7 0.
O.
O.
O.
e 1,000 1.12x10-5 2.Ilx10-6 0.
O.
O.
O.
O.
5,000 3.96x10-6 5.48x10-7 0.
O.
O.
O.
O.
10,000 1.23x!O-6 0.
O.
O.
O.
O.
O.
50J00 0.
O.
O.
O.
O.
O.
O.
U}
=
c T'
r0
-S*
O
Table 1 is our final risk surface. It enables us to look at the frequency of occurrence of essentially any combination of numbers of people and dos e levels. For example, the dose due to natural back-ground radiation is approximately 100 millirems per year to the whole body. Suppose then we want to know the frequency of one person or more receiving 1 year's worth of background dose as a result of these shipments. We observe from Table 1 that the frequency is 2. 76 x 10-5 per shipment. For 300 shipments between Oconee and McGuire, the frequency per year is 2
(2. 76 x 10-5) (3 x 10 ) =.8.28 x 10-3
- i. e., the frequency of one or more people receiving 100 millirems or more as a result of 300 shipments is 8. 28 x 10-3 Thus, it will happen less often than once in 100 years that anyone would receive a dose from a transportation accident equal to what every member of the entire population receives each year from natural background. As another example of the use of the table, note that the threshold for obs ervable radiation sickness in man is about 50 rems = 5 x 104 millir ems. From the table, the like1Ecod that even one person would receive 50 rems or more is below our roundoff, i. e., below about 1 x 10-7 per shipment.
So, we are able to conclude that the risk of transporting spent fuel from Oconee to McGuire is indeed extremely small.
'kh 10 r<d
4 R EFERENC ES 1.
" Final Environmental Statement on the Transportation of Radioactive Material by Air and Other Modes, " :!CREG 0173, December 1977, U. S Nuclear Regulatory Commission.
2.
"An Assessment of the Risk of Transporting Jpent Nuclear Fuel by Truck," PNL-2588, November 1973, Pacific Northwest Laboratory.
3.
" Environmental Survey of Transportation of Radioactive Materials to and from Nuclear Power Plants," WASH-1238, December 1972, U. S. Atomic Energy Commission.4 1Oil 11}}