Imaging upper mantle anisotropy with traveltime and splitting intensity observations from teleseismic shear waves: insights from tomographic reconstructions of subduction simulations

Teleseismic traveltime tomography remains one of the most popular methods for obtaining images of Earth's upper mantle. However, despite extensive evidence for an elastically anisotropic mantle, the isotropic assumption remains commonplace in such imaging studies. This can result in significant model artefacts which in turn may yield misguided inferences regarding mantle dynamics. The nature of anisotropy-induced apparent velocity anomalies has been well-documented in P-wave imaging and various strategies have been proposed to constrain both isotropic and anisotropic heterogeneity from these data. In contrast, few studies have explored the consequences for shear wave tomography and no practical framework for the anisotropic inversion of S-wave delays exists. Here, we propose a new method for constraining arbitrarily oriented hexagonal anisotropy using both traveltime and splitting intensity observations from direct S phases. Our approach accounts for polarization and finite-frequency effects and allows for isotropic starting models. The imaging method is validated through the tomographic analysis of a realistic synthetic dataset produced from waveform simulations through a geodynamic model of subduction. Results illustrate that neglecting anisotropy produces distortions in slab geometry and the appearance of sub- and supraslab low-velocity zones. Anisotropic inversions remove these artefacts while also constraining geodynamically relevant fabric properties including dip.