VS30 at two seismic stations in the Central Valley of California using S-wave refraction tomography and MASW

We estimate VS30 from acquired active-source seismic data near two seismic stations in the California Central Valley towns of Fresno (BK.KARE) and Visalia (BK.LIND). VS30 represents the time-averaged shear-wave velocity in the upper 30 m of the subsurface, and prior site response studies indicate areas with low VS30 values experience greater ground shaking and structural damage during earthquakes; thus, VS30measurements help us understand the potential for amplified ground shaking. We recorded both body and surface waves along linear profiles (up to 118-m long) within 10 m of the stations and used 2D refraction tomography and multichannel analysis of surface waves (MASW) methods to evaluate Rayleigh- and Love-wave-based shear-wave velocities. We calculate VS30 at every meter along the seismic profiles and compare VS30 results among the different methods. Preliminary results from surface wave methods show average VS30 for Fresno seismic station is 573 m/s (Rayleigh-wave) and 390 m/s (Love-wave), and the average VS30 for the Visalia seismic station is 808 m/s (Rayleigh-wave) and 683 m/s (Love-wave). Our Love-wave models show lower shear-wave velocities, which is generally consistent with our prior VS30 studies in California. We find that VS30 for the Fresno Rayleigh-wave model varies by 5.4% across the profile whereas the Love-wave model varies by up to 9.2%. Our Visalia Rayleigh-wave model shows VS30 varies by up to 13.7% across the profile whereas VS30 varies by up to 12.7% for the Love-wave model. The average VS30from our refraction tomography model is 404 m/s for the Fresno seismic station and 1063 m/s for the Visalia seismic station. Finally, comparison among refraction tomography and surface wave models shows average VS30 differs by up to 34.6% (Rayleigh-wave) and 3.52% (Love-wave) for the Fresno profile, whereas average VS30 differs by up to 27.3% (Rayleigh-wave) and 43.5% (Love-wave) relative to refraction-tomography-based VS30 for the Visalia profile. Our results suggest lateral changes in subsurface geology and bedrock structures account for the differences in VS30 across the profiles and among different methods; thus, conventional 1D VS30 would not accurately characterize local sites.