Validating Nevada ShakeZoning Predictions of Las Vegas Basin Response against 1992 Little Skull Mountain Earthquake Records

摘要

Over the last two years, the Nevada Seismological Laboratory has developed and refined Nevada ShakeZoning (NSZ) procedures to characterize earthquake hazards in the Intermountain West. Simulating the M_L 5.6-5.8 Little Skull Mountain (LSM) earthquake validates the results of the NSZ process and the ground shaking it predicts for Las Vegas Valley (LVV). The NSZ process employs a physics-based finitedifference code from Lawrence Livermore Laboratory to compute wave propagation through complex 3D earth models. Computing limitations restrict the results to low frequencies of shaking. For this LSM regional model the limitation is to frequencies of 0.12 Hz, and below. The Clark County Parcel Map, completed in 2011, is a critical and unique geotechnical data set included in NSZ predictions for LVV. Replacing default geotechnical velocities with the Parcel Map velocities in a sensitivity test produced peak ground velocity amplifications of 5%-11% in places, even at low frequencies of 0.1 Hz. A detailed model of LVV basin-floor depth and regional basin-thickness models derived from gravity surveys by the U.S. Geological Survey are also important components of NSZ velocity-model building. In the NSZ-predicted seismograms at 0.1 Hz, Rayleigh-wave minus P-wave (R ? P) differential arrival times and the pulse shapes of Rayleigh waves correlate well with the low-pass filtered LSM recordings. Importantly, peak ground velocities predicted by NSZ matched what was recorded, to be closer than a factor of two. Observed seismograms within LVV show longer durations of shaking than the synthetics, appearing as horizontally reverberating, 0.2 Hz longitudinal waves beyond 60 s after Rayleigh-wave arrival. Within the basins, the current velocity models are laterally homogeneous below 300 m depth, leading the 0.1 Hz NSZ synthetics to show insufficient shaking durations of only 30-40 s.