Dr. James Case, Geologist
Geologic Hazards Section
Wyoming State Geological Survey
Box 3008, University Station
Laramie, Wyoming 82071-3008

SUBJECT: SEISMIC DESIGN FOR URANIUM MILL TAILINGS SITES IN WYOMING

Dear Dr. Case:

Thank you for meeting on May 21, 1996, at the Kennecott Sweetwater (KSW) Uranium Mill with members of the U.S. Nuclear Regulatory Commission staff. The staff has informed me that the discussions and field visit to faults in the KSW site area provided important input to the staff's review of the KSW Revised Environmental Report. Based on information and alternatives you presented, the staff has recommended that Kennecott submit an addendum to its regional seismicity report.

During the KSW site visit you suggested four possible alternatives for the estimation of the peak ground acceleration (PGA) at the site. Those alternatives were based on the consideration of the Chicken Springs fault system, located approximately 15 km from the KSW site, as capable. The selected alternative considered all segments of the Chicken Springs fault separately, and resulted in a magnitude 6.5 event. The staff considers this approach to be acceptable and the resultant magnitude to be technically justified.

The staff also reviewed your report, "Recommendations Regarding Seismic Design Standards for Uranium Mill Tailings Sites in Wyoming," transmitted to the NRC on March 5, 1996, by a letter from Governor Geringer of Wyoming. In that report you recommended that Wyoming uranium mill sites be designed to withstand a peak horizontal acceleration (PHA) with a recurrence time of 2000 years, which would equate to a 200-year design life. Your report was in response to the report, "Seismic Hazard analysis of Title II Reclamation Plans," generated for the NRC by Lawrence Livermore National Laboratory (LLNL). The LLNL report proposed a 0.33 g PHA with a 10-4 annual probability of exceedance, appropriate for a 1000 year design for the geologic province containing KSW site. For a 200-year design life, the LLNL report suggests a 0.18 g PHA for KSW. Although a 200-year design may be sufficient for a remote site, NRC regulations in 10 CFR Part 40, Appendix A, require that a design for the disposal of tailings from a mill provide "reasonable assurance of control of radiological hazards to ... be effective for 1,000 years, to the extent reasonably achievable, and, in any case, for at least 200 years...." The staff considers that a 1000-year design life is "reasonably achievable." Appendix A further requires that "Tailings should be disposed of in a manner that no active maintenance is required...." Therefore, the staff concludes that the alternative proposed in your report of recommendations would not

provide the level of assurance necessary to comply with the regulations. The staff does agree that the LLNL analysis may impart an overly conservative design value for the KSW site, and the alternative approach you proposed for the KSW site on May 21 will provide sufficient conservatism.

Although the LLNL report provides an acceptable value for the seismic design for each uranium mill site in Wyoming, NRC licensees have the option to propose a more robust seismic design for reclaimed tailings piles or alternatives for the estimation of the seismic design value for their site. The staff is willing to review, on a case-by-case basis, alternative approaches proposed for estimating ground motion, assuming the alternative approach is well-based technically and results in conservative design assumptions. As discussed with you on May 21, a probabilistic seismic hazard assessment, such as those documented in the LLNL report, provides one alternative for calculating seismic design values, but may not be necessary. Another acceptable alternative to using the LLNL value is a deterministic analysis which includes consideration of floating earthquakes. Such an analysis includes a deterministic evaluation of nearby capable faults, as you have done for the Chicken Springs fault system, and calculation of the PGA from the maximum magnitude floating earthquake characteristic of the site's tectonic province, and placement of this event 15 km from the site. The PGA from all capable faults and the floating earthquake is the seismic design value.

Again, thank you for sharing the results of your work on Wyoming fault systems with the NRC staff and for providing useful and thoughtful suggestions. If you have any further questions regarding the KSW or other uranium mill sites in Wyoming, please contact Ms. Charlotte Abrams or Mr. Daniel Rom, of my staff, at (301) 415-5808 and 415-6704, respectively.

Sincerely,
SIGNED
Joseph J. Holonich,
Chief Uranium Recovery Branch
Division of Waste Management
Office of Nuclear Material Safety and Safeguards
cc: G. Glass, Wy State Geologist
O. Paulson KSW