Full-Wave Anisotropic Tomography in Southern California

Tuesday, 16 December 2014: 11:50 AM
Yu-Pin Lin, Department of Geoscience, National Taiwan University, Taipei, Taiwan, Li Zhao, Academia Sinica, Taipei, Taiwan and Shu-Huei Hung, NTU National Taiwan University, Taipei, Taiwan
The interpretation of shear wave splitting, one of the principal observables in the investigation of anisotropy in the crust and upper mantle, has been largely based on the ray-theory modeling of a single vertically incident plane wave. This approximation results in the rejection of shear-wave splitting measurements in many cases, thus severely limits our ability to make full use of shear-wave splitting data to resolve the spatial variations in anisotropy. Here we adopt a full-wave approach to the inversion of 3D anisotropy structure using the sensitivity (Fréchet) kernels calculated by an efficient and flexible algorithm based on the normal-mode theory. The full-wave sensitivity kernels accurately account for all the interactions of multiple phases and widen the range of possibilities in the source-receiver geometry amenable for making shear-wave splitting measurements. We apply the full-wave kernels to the splitting measurements using records of the Southern California Seismic Network from ~150 events, and combine them with a wavelet-based model parameterization in a multi-scale inversion for the anisotropic structure. In addition, we quantify the spatial resolution of our tomography inversion by a rigorous analysis based on the statistical resolution matrix approach, which provides an efficient way to estimate the resolution lengths of the inversion without explicitly calculating the resolution matrix. Results show that the finest resolution length in our inversion can be ~25 km in the region with densely-distributed stations. The anisotropic model shows that the azimuths of the fast axis in the lithosphere can be related to surface geologic features such as the Salton Trough, the Transverse Ranges and the San Andreas Fault. A weaker anisotropy in the asthenosphere may imply a vertical flow beneath the North American Plate.