On constraining 3D seismic anisotropy in subduction, mid-ocean-ridge, and plume environments with teleseismic body wave data

Conventional seismic tomography studies consider the Earth's interior as mechanically isotropic, despite seismic anisotropy being widely observed. This current standard approach to seismic imaging is likely to lead to significant artefacts in tomographic images with first-order effects on interpretations and hinders the quantitative integration of seismology with geodynamic flow models. Although a few methodologies have been proposed for carrying out anisotropic tomography, their ability in simultaneously recovering isotropic and anisotropic structures has not been rigorously tested. In this contribution we use geodynamic and seismological modeling to predict the elastic properties and synthetic teleseismic P- and S-wave travel-time datasets for three different tectonic settings: a plume rising in an intraplate setting, a divergent margin, and a subduction zone. Subsequently, we perform seismic anisotropy tomography testing a recently developed methodology that allows for the inversion of an arbitrarily oriented weakly anisotropic hexagonally symmetric medium using multiple bodywave datasets. The tomography experiments indicate that anisotropic inversions of separate and joint P- and Swave travel-times are capable of recovering the first order isotropic velocity anomalies and anisotropic patterns. In particular, joint P- and S-wave anisotropic inversions show that by leveraging both phases it is possible to greatly mitigate issues related to imperfect data coverage common in seismology and reduce parameter tradeoffs. In contrast, by neglecting seismic anisotropy, isotropic tomographic models provide no information on the mantle fabrics and in all cases are contaminated by strong velocity artifacts. In the inversions the magnitude of anisotropy (as well as that of seismic anomalies) is always underestimated owing to regularization procedures and smearing effects. It follows that the true seismic anisotropy of mantle rocks is likely higher than estimated from anisotropic tomographies, and more consistent with predictions from laboratory and numerical micromechanical experiments. Altogether, these results suggest that anisotropic body-wave tomography could provide unprecedented information about the Earth's deep geological structure, and that the latter could be better recovered by complementing teleseismic body-wave travel-times with other geophysical datasets.