Amplification and Attenuation in Southern California Basins Empirically Calculated from the Ambient Seismic Field

Modeling wave propagation through the complex geology of the crust is one of the foremost challenges to a physics-based approach to seismic hazard analysis, largely because we lack the complete description of elastic and anelastic structure needed for accurate ground motion simulations. We use the ambient seismic field to improve these simulations in several ways. First, we use ambient field Green's functions to study ground motion amplification from sedimentary basins. Our results are validated as accurate using ground motion observed in moderate earthquakes at long periods. To date we have used first order coherency for these comparisons, but this suffers from potential biases due to the non-uniform distribution of ambient noise sources. We use higher order coherencies to reduce directional excitation effects. The resulting Green's functions show better agreement with observations. Further improvements can be realized by accounting for more realistic earthquake source parameters as well. Second, we use the ambient field to recover spatially variable anelastic structure in South California. This information is critical to strong ground motion prediction for cases where propagation paths are long. This includes scenarios of particular concern, such as waveguide effects into the Los Angeles basin from a great earthquake on the San Andreas Fault.