Collaborative Proposal: Understanding Subduction Systems by Linking Anisotropic Seismic Imaging and Geodynamic Modeling

Recordings of distant earthquakes will be used in this project to infer the velocity at which the waves traveled beneath the Earth's surface. In turn, we use this information to understand the Earth's structure and, in particular, how subduction (the descent of an old tectonic plate into the Earth's mantle) produces volcanism and affects surface topography. When conducting this type of study, it is currently common to assume that the speed at which the waves travel is independent of their direction. In reality, seismic waves travel at different speeds depending on what their direction is relative to the orientation of minerals in the mantle. This phenomenon is known as anisotropy, and when not taken into account, it can bias the image of the subsurface in ways that can compromise the understanding of the subduction processes. In this project, seismologists will investigate ways to reduce this problem by incorporating physically-based estimates of the distribution of anisotropy into the imaging procedures. In this way, more accurate images of the subsurface will be determined advancing the understanding of how subduction affects the surface, including the distribution of volcanism.

The project will combine travel-time tomography, SKS splitting observations and geodynamic flow modeling to produce self-consistent models of mantle structure (isotropic and anisotropic) in the vicinity of two subduction zones. The new approach will allow us to fully integrate mantle anisotropy into teleseismic tomography of subduction zones. By incorporating estimates of the anisotropy field into travel-time tomography the scientitists will significantly reduce major artifacts stemming from the isotropic assumption and obtain a physically-based model for the strain in the mantle wedge and beneath the subducting plate. The improved imaging will allow to better recover the mantle physical state (temperature, composition and melt fraction). This methodology will be applied to existing data from the western Mediterranean and Cascadia subduction zones.