Results and Discussion

We computed three FDPs for three different 3D Santa Clara Valley velocity models. The first model is the San Francisco Bay Area model from USGS (USGS Bay Area Velocity Model 05.1.0). The second model is the UC Berkeley (UCB) 3D model (Stidham et al., 1999). The third model is a modified USGS model. In this modified velocity model, in order to enhance the effect due to the basin structure, we kept the velocity structure inside the basin, but assumed only 1D reference velocity structure for the regions outside of the basin. The comparison between synthetic and observed FDPs shows that FDPs for USGS and modified USGS models follow a similar depth dependent trend as observed FDPs, and their absolute values are in agreement as well (Figure 2.17). The synthetic FDPs for the UCB model show that FDP correlates with basin depth up to 2 km, but the correlation coefficient is not high and individual values are more scattered than in the case of the other models, especially for shallow depths. We also cannot match the observed shift of the trend in the observed FDP curve at around 2 to 3 km depth because the maximum thickness of the basin obtained in the UCB model is only 2 km. The observation and synthetic results show that the variations in FDP are sensitive to the thickness of the basin up to a certain depth and then becomes stable. The simulation of the P-wave time delays from teleseismic events for the USGS 3D velocity model at SCVSE stations shows a similar saturating trend as the FDPs for the USGS model (Dolenc et al., 2005). Synthetic time delay increases with increasing basin thickness up to a certain depth and then stabilizes. Since the P-wave time delay depends entirely on the mean velocity and thickness of the basin, this indicates that velocity contrast between the basin and the background medium is negligible at deeper depth, which is true for USGS 3D velocity model. In the case of the Santa Clara Valley model, FDP is only sensitive to the shallow structure, and the apparent thickness of the deeper part of the basin is much shallower than the model. It is likely due to the negligible velocity contrast in Santa Clara Valley model. But we still need more experiments to confirm which factor really controls the depth of the change in the depth-dependent trend of the FDP.

Figure 2.17: A comparison among observed (open circles)) and three synthetic FDPs for USGS (Open squares), modified USGS (Solid circles) and UCB (Gray triangles), as a function of depth.