2018-04-02_REVISION - M1977342 - Laserfiche WebLink

Home Browse Search

SEO NFOUR SOIsmIC NEW Inputs estimated based on whether the earthquake magnitudes were based on a maximum felt MM intensity or whether they were instrumentally determined. First the recurrence was calculated for the SRM catalog. Then the recurrence was re-calculated 500 times, first fixing the b-value obtained initially, and applying the two magnitude corrections discussed above each time (first the rounding error, then the magnitude error for each magnitude in the catalog) to obtain an average corrected a-value. The average corrected a-value of 0.036 events for M 5 and above for the SRM zone(area of 103,475 km2)is included in the PSHA. Because of the limited duration and incompleteness of the historical catalog and the very small number of events and their narrow magnitude range used in the recurrence calculations (Figure 6),uncertainties in the recurrence parameters for the background seismicity are large. To incorporate the uncertainties into the hazard analysis, we used three b-values for the regional seismic source zones, the best estimate, and plus and minus 0.1 values, weighted 0.6, 0.2, and 0.2, respectively. An inspection of the resulting recurrence intervals for M 5 and 6 events was performed to check the reasonableness of the three b-values. The a-values were held fixed because the recurrence curve was better constrained at the smaller magnitudes. For the SRM zone, the calculated recurrence interval for an "1882-like" M 6.6 earthquake is approximately 1518 years,not an unreasonable value given the seismotectonic setting. The use of seismic source zones assumes that background earthquakes are uniformly (randomly) distributed throughout the seismogenic crust. However, some seismicity may also be stationary through time (at least over the next few decades of interest) and thus the use of Gaussian smoothing. In the Gaussian smoothing approach, we smoothed the historical seismicity on a grid at 0.2 degree intervals to incorporate a degree of stationarity. The version of Gaussian smoothing adopted in this study (Frankel, 1996) is the same as that used in the National Seismic Hazard Maps (Petersen et al., 2008). This scheme addresses both the spatial stationarity of seismicity and its randomness. In the smoothing approach, we smoothed the historical background seismicity out to 500 km from the site to incorporate a degree of stationarity, using a spatial window of 15 km. Thus the hazard from seismicity that clusters in a specific seismic zone is retained spatially rather than being smoothed to a uniform distribution as in a seismic source zone. We weighted the use of seismic source zones and Gaussian smoothing at 0.5 and 0.5, respectively (Figure 4). Recent seismicity may be considered more likely representative of seismicity occurring in the next 100 years. However, given the short 144-year incomplete historical record, the possibility exists that the record is not representative of the next 100 or so years and thus recommend that the two approaches be weighted equally. 4.2 SITE CONDITIONS At the request of the TRB, Vs surveys were performed at 1 and 3 Dams due to the lack of any site-specific information. Geovision Geophysical Services performed both Rayleigh wave spectral-analysis-of-surface-wave (SASW) and multi-channel-analysis-of-surface-wave (MASW) surveys in addition to S-wave seismic refraction. Their report is contained in the Appendix. Contrary to the information on the foundation geology that was used in the previous study, neither 1 or 3 Dams were founded directly on Troublesome Formation or gneiss, respectively. Both dams were located on unsaturated Quaternary glacial drift and outwash deposits underlain by a zone of weathered rock and then more intact rock. The intact rock with a Um %27_K%R_,m 17