Effects of Crust and 3D Mantle Heterogeneity on Shear Wave Splitting Using 3D Wave Simulations

Shear wave splitting is a powerful technique to evaluate deformation within the Earth. However, finite frequency effects with a 3D heterogeneous Earth have yet to be fully investigated. Complications from 3D mantle and crustal heterogeneity may affect shear wave splitting. Ray theory is commonly used in modeling seismic anisotropy and is appropriate within certain limits, but not all implications have been explored. The goal of this work is to explore the effects of ray theoretical assumptions made in shear wave splitting as well as tease out discrepancies between different background mantle models and observations of anisotropy within the Earth. In this work, we explore 1D and 3D background models: PREM, S40RTS (Ritsema et al., 20110, and GLAD-M15/25, and the 3D mantle and crustal effects on core phases used for shear wave splitting (i.e., SKS, SKKS, and PKS). We conduct our numerical simulations via 3D global wave propagation solver SPECFEM3D_GLOBE (Komatitsch & Tromp, 2002). Ultimately, we aim to improve the understanding of how seismic anisotropy observations, like SKS shear wave splitting, are related to models of deformation in the Earth's mantle and the sources of anisotropy in the crust and mantle, specifically the D" layer. We also calculate sensitivity kernels of various phases for travel time to evaluate how wave sensitivity may change with 3D structure down to 9s. Overall, we do see polarization anomalies and anomalous energy on the transverse component of core phases that could introduce splitting delay time errors up to ~0.5s for SKS and SKKS waves, but PKS splitting delay time error is much smaller and on the order of ~0.1s. Most of these splitting errors can arise from multiple phase arrivals, Earths Coriolis effect, 3D Earth structure, and source effects. Additionally, we see complications of multiple phase arrivals that could introduce additional errors into splitting measurements.