ML19351F149
| ML19351F149 | |
| Person / Time | |
|---|---|
| Site: | La Crosse File:Dairyland Power Cooperative icon.png |
| Issue date: | 12/05/1980 |
| From: | Reiter L Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19351F147 | List: |
| References | |
| ISSUANCES-SC, NUDOCS 8012290640 | |
| Download: ML19351F149 (7) | |
Text
12/5/80 UNITED STATES OF AMERICA O
NUCLEAR REGULATORY COP 911SSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 1
2 In the Matter of 3
DAIRYLAND POWER COOPERATIVE Docket No. 50-409-SC (La Crosse Boiling Water Reactor)
(Order to Show Cause) 4 5
PREPARED TESTIMONY OF LEON REITER IN RESPONSE TO BOARD QUESTIONS 6
7 Q.
Please state your name and business address. ~
8 A.
My name is Leon Reiter and my business address is the U.S. Nuclear g
Regulatory Comission, Washington, D.C. 20555.
10 Q.
Would you state your position with the NRC and your professional 11 experience?
12 A.
I am currently a Seismologist in the Geosciences Branch, Office of 13 Nuclear Reactor Regulation. A statement of my professional experience 14 and qualifications is attached to this testimony.
15 Q.
Please state the purpose of your testimony.
16 A.
The purpose of my testimony is to respond to questions 1, 2, 3 and 5 17 in the Licensing Board's November 12,1980 Order.
(
18 Q.
Have you read the Board's questions?
19 A.
Yes, I have.
20 Q.
Would you answer the Board's first question? That question is:
l 21 "What is the currently estimated return period 22 for an earthquake producing ground acceleration f
23 of 0.12g at the La Crosse site? Can confidence 24 limits be placed on this estimate?"
l l
l k
soneoo640
. i i
1 A.
Currently estimated return periods for an earthquake producing a L
2 peak ground acceleration of 0.129 at the Ls Crosse site are on the order 3
of 1,000 or 10,000 years. More likely than not the actual value is 4
closer to the longer end of this range. We do not believe that more 5
rigorous " confidence limits" than those indicated above can be specified.
6 Q.
Please answer the Board's second question. That questi n is:
I 7
"What is the basis for detennining that 8
return period (including the specific earth-g quakes considered--e.g., Anna, Ohio; New 10 Madrid, Missouri; etc.)? Do the probabi-11 listic studies take into account the July, 12 1980 earthquake near Maysville, Kentucky?
13 What effect would this earthquake have on 14 the probabilistic computations?"
15 A.
These return periods are based upon documents submitted thus far 16 in the LLL-TERA Site Specific Spectra Program (SSSP) as evaluated by the 17 staff in the June 23, 1980 memo from R.E. Jackson to D. Crutchfield.
18 In the SSSP, return periods were calculated for peak ground accelerations, 19 peak ground velocities and response spectra at the eastern and central l
20 U.S. sites that are part of the Systematic Evaluation Program. Experts 21 on eastern seismicity were polled with respect to seismic zonation, 22 frequency of earthquake occurrences, upper magnitude cutoff, and 23 characterization and attenuation of ground motion.
24
_d
,1
~
I i
1 Peak ground motions and spectra were calculated at selected return 2
periods (200,1000 and 4000 years) for each expert, specifically incor-3 porating uncertainty in the above parameters. The individual results
. }
4 were then synthesized based upon a weighting procedure that utilized 5
expert specified weights in each area.
In the above mentioned review 6
the staff recorrnended use of a specific "1000 year" spectra. We believe
]
that these spectra are conservative and more appropriately characterized 7
8 as having return periods on the order of 1,000 or 10,000 years. TERA Corporation in assessment of the conservatisms believes that the true g
return periods for these "1000 year" spectra are close to 5,000 or 10 jj 10,000 years. The peak ground acceleration associated with the "1000 year" spectra at La Crosse was 0.09g. Based upon our evaluation of the 12 SSSP, we estimate the return period for 0.12g to be about a factor of two 13 14 longer than that for 0.099 Specific earthquakes such as those at Anna, Ohio, and New Madrid, l
15 Missouri, are taken into account by each expert in determining the con-I 16 figuration and frequency of earthquake occurrence and upper magnitude 77 cutoff in each seismic source zone. The attenuation of intensities from 18 3
specific earthquakes is also considered in deriving ground motion models 19 for the central and eastern United States. The above estimates were 20 l
based upon calculations carried out prior to the July 1980 Maysville, l
2)
Kentucky earthquake and the experts could not have taken this event into 22 23 account.
It is difficult to estimate the exact effect of an additional earthquake upon the resulting calculated return period since these 24 ll
I calculations were carried out through a complex integration of a wide 2
range of opinions and uncertainties. However, based upon an examination 3
of individual experts' assumed zone configurations and the size and 4
frequency of earthquake distributions, it appears that inclusion of 5
this earthquake into considerations would have a minimal, if any effect.
6 upon the results. Certainly any small change introduced by this earth-7 quake would be encompassed within the range of return periods indicated 8
in the answer to the Board's first question.
9 Q.
Please answer the Board's third question. This question is:
l 10 "What attenuation model is being used for p
11 the return period determination? How does 12 this model compare with others being used 13 for this purpose? How does use of different 14 atteruation models affect the detemination of 15 the return period of an earthquake producing 16 0.12g ground acceleration?"
17 A.
The attenuation model used in return period estimation for La Crosse
?
18 is that proposed by Gupta and Nuttli (1976).
It is based upon intensity 19 attenuation of several central United States earthquakes and has been used f
20 extensively for sites in the central United States. This intensity 21 attenuation relationship is among the more conservative models that have
?
22 been proposed for the United States east of the Rocky Mountains.
In the SSSP ground motion estimates were calculated assuming several other atten-23 h
24 uation models. These were based upon intensity attenuation from the 1968 l
lA
p.
i J' f' l
1 southern Illinois earthquake and another based upon intensity attenua-2 tion from the 1940 Ossippee, New Hampshire, earthquakes.
3 At La Crosse, the peak acceleration associated with given return 4
periods were similar.
Initial calculations using a theoretical ground 5
motion model proposed by Dr. Otto Nuttli also yielded generally similar 6
results. This is not to say that the above models would yield similar 7
results at all frequencies at all sites in the eastern U.S.
Generalized 8
attenuation models such as those used above are related in a complex 9
way to resulting ground motion estimates. One model may yield stronger 10 ground motion at shorter distances than another model, while at the 11 same time it could also indicate weaker ground motion than the other 12 model at longer distances. Similarity of results could very well stem 13 from a complex interaction of compensating effects, including the way 14 a particular intensity attenuation model is assumed to be related to 15 ground motion.
16 Q. Would you answer the Board's fifth question? The question is:
17 "Does the probability analysis include any 18 specific conditions which may result in am-f 19 plification at the La Crosse site? (See 20
' Initial Review and Reconsnendations for Site 21 Specific Spectra at SEP Sites', attached to 22 June 23, 1980 Jackson /Crutchfield Memorandum, 23 p.3) Identify any such conditions which may 24 l
2
,,__._,___________.,____.___,____.___-._______m_m
l
_6 t
i I
"be pertinent. What would be the effect 2
of including such conditions in the proba-L 3
bility analysis?"
4 A.
Physical properties of the geologic materials underlying a site 5
can significantly modify the amplitude level and spectral composition 6
of ground motion. Whether or not amplification or deamplification r
t 7
will result at a given frequency is a complex function of the thickness 8
and configuration of the underlying materials, their physical properties 9
and the level of ground motion being considered. The probability analysis 10 discussed above assumes a soil condition typical of the western United 11 States. This stems from the fact that there are virtually no strong
[
12 motion recordings in the eastern United States while there are many in 13 the west, particularly in California. As a result, the amplification 14 conditions inherent in the study are those which might result at " average" 15 western sites. The question posed then is how different are the ampli-f 16 fication conditions at La Crosse from those at the " average" western soil l'
17 site. Preliminary results from a Lawrence Livermore Laboratory study 18 (personal communication, Don Bernreuter) indicate that ground motion i,
ii 19 would not necessarily translate into significant changes in the proba-l 20 bility analysis. The probability analysis is dependent upon ground 21 motion from different sources at different amplitude levels, and ampli-I 22 fication at one level from a particular source could be compensated for 23 by deamplification at another level from anothe-source. The staff has 24 not yet reached any final conclusions with respect to this problem.
1
LEON LITER l'..MR, SEISMOLOGY SECTION GI W1ENCES BR JICH DIV15'0:; 0F ENGliiEERING U. S. NULLEAR REGULATORY COMMISSION "y name is Leon Reiter.
I presently reside at 1960 Dundee Road, 1
%ckville, Maryland 20S50 and am employed as a Seisnologist, Geosciences Branch, Division of Site Saftty and Environmental Analysis, Office of r!uclear Reactor Regulation, 'lashington, D.C. P0555.
i.4FSS10NAL QUALIFICATIONS I received a Bachelor of Arts degree in Geology from Crooklyn College f
in 1958, a Master of Science degree in Geology (Geophysics from the
}
University of Mithigan in 196S, a Master of Arts degree in iW hematics from the University of Michigan in 1970 and a Ph.D. in Geolog (Geophysics) k from the University of Michigan in 1971.
In the year following receipt i
of my Ph,0. I was a National Science Foundation Post-Doctoral Fellow at i
the Institute of Geophysics
'd Planetary Geophysics in La Jolla,' California.
~
From 1972 to 1976 I was an Assistant Professor of Geophysics at the University of Oklahoma. During the sur.uer of 1975 I was a visiting scientist of the U. S. Geological Survey National Center for Earthc. eke Research in Menlo Park, Calif 0rnia.
I joined the NRC in August., i9/6 as a Seismologist and since August 1979 I have been Leader of the Seismolog.e Section in the Geoscicnces Branch of the Division of Engineering.
My researth during my academic career has included the areas of crustal exploration, seismic wave attenuation, midcontinent seismicity and
}
tectonics, earthquake prediction and the application of seismic techniques to engineering problems. At NRC I have been actively involved in review
]'
of sites for nuclear facilities in all parts of the United States and in several foreign countries.
I have also taken a lead responsibility for studies in the ficids of strong motion seismology, near-field groundmotion, i
and probabilistic risk assessments.
I am a member of the American Geophysical Union, the Seismological Society of America, the Society of Exploration Geophysicists and the Earthquake l
Engineering Research Institute.
I have served as a member of the Plate Interiors Working Group of the U. S. Geodynamics Committee, the Interagency I
Committee on Scismic Safety in Construction and the Panel on National Regional and Local Seismograph Networks of the National Research Council-National Academy of Sciences.
I have authored or co-authored papers published in the b
Bulletin of the Seismological Society of America, the Journal of the Acoustical Society of America, the Proceedings of the American Society of Mechanical Engineers and National Science Foundation Conference Proceedings, b
r i
{p e-
.w -
.-..,. v w _ _.- u,
w --.,,.,.,_..,,
m
.