Laserfiche WebLink
SEONFIVE Seismic NMrd INURS <br /> Figure 39 shows a comparison of UHS from this study and the 2013 results. While the PGA and <br /> 1.0 sec SA ground motions generally increased from 2013 to 2017, the ground motions for <br /> spectral periods of 0.2 to 0.8 sec are quite similar. The apparent shift in the peak of the spectra <br /> to shorter periods is due to the increase in the Vs beneath the site from 940 to 1,500 m/sec. <br /> 5.3 COMPARISON WITH USGS NATIONAL HAZARD MAPS <br /> In 1996, the USGS released a "landmark" set of National Hazard Maps for earthquake ground <br /> shaking,which was a significant improvement from previous maps they had developed (Frankel <br /> et al., 1996). These maps have been revised and updated, and the most current version was <br /> released in 2014 (Petersen et al., 2014). These maps were the result of the most comprehensive <br /> analyses of seismic sources and ground motion prediction ever undertaken on a national scale <br /> and they make use of the five NGA-West2 relationships. The maps are the basis for the <br /> International Building Code. The maps are for NEHRP site class B/C(firm rock)or Vs30 of 760 <br /> m/sec. <br /> For a 2,475-year return period (2% exceedance probability in 50 years),the 2014 USGS National <br /> Hazard Maps indicate a firm rock PGA of 0.24 g for the site compared to the site-specific value <br /> of 0.26 g using a Vs30 760 m/sec, which was run as a sensitivity analysis using the site-specific <br /> seismic source model and ground motion models. <br /> The site-specific value is 8% larger due to significant differences in the seismic sources and the <br /> ground motion models. In the 2014 maps, the USGS used (1) a lower Mmax of M 6.5 for the <br /> WFMF compared to our value of M 6.8 to 7.0; (2)a slip rate for the Frontal fault(0.068 mm/yr) <br /> nearly half of the rate used in this study (best estimate of 0.15 mm/yr); and (3) the USGS used <br /> central and eastern ground motion prediction models for the SRM in contrast to our use of the <br /> NGA-West2 models(Section 4.3). <br /> 5.4 COMPARISON WITH DETERMINISTIC HAZARD RESULTS <br /> Given the proximity of the WFMF to the site,situated in the hanging wall at a shortest(rupture) <br /> distance of only 1.2 km, it is the dominant deterministic earthquake at the site. Both median <br /> (50th)and 84th percentile 5%-damped acceleration response spectra were calculated for a M 7.1 <br /> earthquake on the WFMF using the same four NGA models used in the PSHA. The resulting <br /> deterministic median PGA is 0.37 g, and the median at 1.0 sec spectral acceleration is 0.22 g <br /> (Table 8). In comparison,the 84`h percentile PGA is 0.68 g,whereas the 84d'percentile at 1.0 sec <br /> spectral acceleration is 0.44 g(Table 8). Figure 40 shows the 84th percentile spectra for each of <br /> the NGA models along with the lognormal average. An 84th percentile spectrum was calculated <br /> because it is commonly the required level of ground motions used in design or safety evaluations <br /> for high-hazard dams (e.g., California Division of Safety of Dams, and Colorado Office of the <br /> State Engineer). On Figure 41, the median and 84th percentile deterministic spectra are <br /> compared with UHS with return periods ranging from 500 to 30,000 years. The median <br /> spectrum is approximately equal to the 10,000-year return period UHS for spectral periods <br /> greater than 1.0 sec and between the 5,000 and 10,000-year return period UHS for shorter <br /> spectral periods.The 84th percentile spectrum is significantly higher than the 30,000-year return <br /> period UHS(Figure 41). <br /> 22 <br />