ML20247F396

Forwards RAI Re 980306 Submittal Which Transmitted Paducah Gaseous Diffusion Plant Seismic Hazard Analysis.Addl Info Should Be Provided within 21 Days of This Ltr

ML20247F396
Person / Time
Site:	07007001
Issue date:	05/13/1998
From:	Pierson R NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
To:	Toelle S UNITED STATES ENRICHMENT CORP. (USEC)
References
TAC-L32054, NUDOCS 9805190320
Download: ML20247F396 (4)

Text

- __ _-____ - _ _ ___ - __ _ _ _ ______ _

Mayu13, 1998 i

Mr. Steve A. Toelle, Manager Nuclear Regulatory Assurance and Policy U.S. Enrichment Corporation 2 Democracy Center 6903 Rockledge Drive Bethesda, MD 20817-1818

SUBJECT:

PADUCAH GASEOUS DIFFUSION PLANT SEISMIC HAZARD ANALYSIS (TAC NO. L32054)

Dear Mr. Toelle:

This refers to your submittal dated March 6,1998, which transmitted the Paducah Gaseous Diffusion Plant Seismic Hazard Analysis for staff review.

Our review of your submittal has identified additional information that is needed for the staff to complete its review. The additional information, specified in the enclosure, should be provided within 21 days of this letter. Please reference the above TAC No. in future correspondence related to this review.

If you have any questions conceming this request, please contact me or Andrew Persinko, of my staff, at (301) 415-6522.

Sincerely, M @ fd S19fl8d %

Robert C. Pierson, Chief Special Projects Branch Division of Fuel Cycle Safety and Safeguards Office of Nuclear Material Safety and Safeguards

Enclosure:

As stated (

cc: Mr. Steve Polston, PGDP )

Docket 70-7001 sTRIBuTioN:hM Docket 70-7001 INRC File Center & PUBLIC NMss r/f FCs3 r/f sPB r/f Rlil PHiland. Riti - Ko"Brien, Riti oFC NB h sPB k {s h sPB b SPJA NAME Persinko:lj diey n h erson DATE h8 Y eg(98 96 [M8 i M48

( C = COVER E = cover & ENCLOSURE N = No COPY g:\reiraL51 OFFICIAL RECORD COPY p* - -. w w

~

980519o320 900s S Ijjjg gusg PDR ADOCK 0700 1 L

l l* ,

l..

!' Request for AdditionalInformation l Submittal Dated March 6,1998 United States Enrichment Corporation Paducah Gaseous Diffusion Plant Docket No. 70 7001 l Please provide the following information:

l

1. Assumption of temporal clustering Regarding the discussion of single and clustered events (pages 2-3 to 2-6), the report asserts that an assumption of clustering is more conservative for the large events.

a) Are you assuming that the recurrence of an event can be used in place of the probability of exceeding that event (page 2-3)?

b) In the example, it appears that you use the same recurrence for the three j independent events as for one clustered event to arrive at the probability of exceeding the values of 0.03 and 0.01, respectively. Is that correct or does this analysis assume that the recurrence of a clustered event is exactly 1/3 the l- recurrence of the independent events? Is there any technicaljustification to support this assumption, i.e., could clustered events occur more frequently than the combination of single events?

l c) Because the return period of interest is 250 years, do the large, but infrequent l earthquakes, really control the peak ground motion? Otherwise the assumption of independent events would, according to the calculations, increase the hazard.

d) The effect of the assumption of temporal clustering for large earthquakes occurring in the NMSZ on the recurrence rate of small to moderate earthquakes should be further discussed. It would be helpful to have a figure comparing l magnitude-recurrence curves with and without the assumption of temporal clustering for the NMSZ source. Especially, how would this assumption affect the return period (or the recurrence rate) of a 0.15g earthquake?

2. Logic tree for rates of large event on East Prairie fault and East Prairie extension What is the sensitivity of the weighting given to the weighting of the logic tree on pages 3-14 and 3-15; especially rate of large earthquake, recurrence, and length of East Prairie Extension (because the curve, figure c-1 for example, goes all the way to 100,000 yrs, the effect on the 250 year earthquake is hard to discern)? Van Arsdale and Johnston conclude that assuming greater weight is not valid because of strain accumulation rates.

Would Van Arsdale and Johnston change their weighting given the most recent GPS f data (Weber et al.,1998)?

3. B Values Explain why the b value for the rift is "more stable" than for individual faults. How does application of the system b value compare to the analyses by Hoffman (1996). Have you  ;

i i

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ ._ . . _ _ J

2 tested the sensitivity of this assumption? Have you considered reconstruction of magnitude-recurrence relationships from the historical record, in which the data were divided into several sub-populations instead cf a single population (see e.g., Speidel, 1998)?

4. Fault Geometry in NMSZ The common assumption is that faults in the NMSZ are reactivated features of a Phanerozoic rift system. Characteristics of faults (e.g., orientation) rely on this assumption. However, Schweig and Ellis (1994) point out that the faults in the NMSZ are young and slightly oblique to the assumed fault traces. Have you done any analyses to test for the sensitivit'/ of small changes in fault orientation? Have you, in any preliminary work treated the zone as a background source, and if so how does that change the 250 year PGA?

5. Ground motion attenuation for rock Provide the rationales for varying C1 and C2 (instead of other parameters) to obtain attemative sets of ground motion equations. Using a couple of plots to compare peak and spectral ground motion predictions from the four attemative models used in the Paducah PSHA with those from other attenuation equations in the CEUS., including Atkinson and Boore (1997). According to Abrahamson and Shedlock (1997), Toro et al (1997) predicted lower PGA for smaller earthquakes at all distances and for larger earthquakes at distances greater than about 20 km. Contributions to the PGA at the Paducah site for the 250 year return period are mainly from small and moderate earthquakes (50%), we may not be able to say that the current model is more conservative.

Please provide aerivation of equations 4-3 through 4-5 and rationales for selecting these equations in considering saturation effect. Are parameters a and b the same as those in exponential recurrence model?

6. Ground motion attenuation for soil and vertical component a) Provide the actual equations used to calculate the site-specific soil amplification factors as functions of frequency, how these equations were derived, and the selection of data range for the associated parameters. These equations are not transparent in the cited literature. Dynamic material properties, besides shear wave velocity, that were involved in the calculation and selection of data range for these parameters are not discussed or presented in the subject report.

1 b) Also, provide the equations and processes in deriving such equations for calculating the ratio of vertical to horizo. ital ground motion. Again, the equations are not transparent in the cited literature. Please clarify that the vertical motion is calculated from horizontal motion on rock, not on soil.

)

__ __- . _ _ _ _ __ - _ ._ ---_____________-__--____-_-_________A

e 3

, 4

7. Hazard Reeults a) Provide the uniform hazard spectra for the 0.15g (i.e., return period of 220 year) event. This would provide a more complete picture of the hazard associated with the 0.15g acceleration used in the analyses, b) Provide a version of hazard curves (figs 5-1 through 5-8,5-11,5-12, and other similar figures in appendix C) that emphasize the range of interests (i.e., near return period of 250 years. For example, cut off the plot at annual P of 1E-03 or SE-03).

8. Spectra for Other Damping Ratios Provide the derivation of equations 5-1 and 5-2.

9. Other Verify that the Quality Assurance Plan used during the performance of the seismic hazard analysis was the USEC Ouality Assurance Plan. Discuss how the Plan was implemented.

J References Abrahamson, N.A. and W.J. Shedlock.1997. Overview. Seismological / Research Letters I

68(1):9-23.

Atkinson, G.M. and D.M. Boore.1997. Some comparisons between recent ground motion relations. Seismological Research Letters 68(1): 24-40.

l Hoffman, R.B.1996. Individual faults can't produce a Guttenberg-Richter earthquake recurrence. Engineering Geology 43: 5-9. l l

Schweig and Ellis.1994. Reconciling short recurrence intervals with minor deformation in the New Madrid seismic zone. Science 264: 1308-1311.

Speidel.1998. Seismicity relationships in the Central Mississippi Valley. Final Report to U.S.

Army Corp of Engineers /CRGL.

Toro, G.R., N.A. Abrahamson, and J.F. Schneider.1997. Model of strong ground motions from earthquakes in central and eastern north America: best estimates and uncertainties.

Seismological Research Letters 68(1).

Weber et al.,1998. Estimation ofinterplate strain accumulation in the New Madrid seismic zone from repeat GPS surveys. Tectonics 17:250-266.

l l

t _ _ _ - _ _ -