Three-dimensional Subduction Models: Implications for Plate-Mantle Coupling and Length-scales of Seismic Anisotropy

Brown University, Providence RI

Investigators

Abstract

The theory of plate tectonics predicts that the outer layer of the Earth is comprised of lithospheric plates (0 - 250 km thick) that are in motion with respect to one another (at rates on the order of 1 - 20 cm/yr), with the majority of deformation concentrated at the plate boundaries. At the Earth's surface, this deformation is commonly manifested in the form of increased and localized seismicity, volcanism, and mountain building. The expression of plate boundary zone deformation underneath the plates, in the Earth's mantle, however, is not well understood. Geodynamic modeling is a valuable tool that can be used to predict the Earth's viscous mantle response to and interaction with the tectonic plates, as the mantle cannot be accessed directly. The proposed work will use high-resolution, three-dimensional geodynamic modeling as a tool to simulate how the Earth's viscous mantle responds to and interacts with the tectonic plates at subduction zones, regions where one tectonic plate slides beneath another and induces motion within the underlying viscous mantle. Recent seismological observations of shear wave splitting indicate the portion of the mantle wedged between the overriding plate and subducting plate, the mantle wedge, is commonly characterized by seismic fast axes oriented oblique to plate motion, indicative of a complex mantle flow field in many subduction zones. However, far from the plate boundary, the seismic fast axes are commonly oriented sub-parallel to plate motion, indicative of coupling between the plate interior and underlying mantle flow field. A two-dimensional subduction regime cannot explain the along strike and across strike variations in mantle flow implied by the shear wave splitting, thus requiring a three-dimensional framework. Previous three-dimensional numerical simulations of subduction zones predict that trench-parallel flow can be driven by pressure gradients in the mantle wedge and that toroidal flow can be generated around lateral slab edges due to steepening of the subducting plate or rollback in the trench. In addition models of small-scale convection within the mantle wedge predict complex mantle flow and seismic anisotropy in the mantle wedge. However, what controls the transition from complex mantle-overriding plate motion near the subduction zone to aligned mantle-overriding plate motion in the plate interiors has not been investigated. Furthermore, although recent work indicates the rheological flow law governing deformation in the mantle may be instrumental in controlling the magnitude and length-scales of the subduction induced viscosity reduction and complex mantle flow field, this has not been quantified in three-dimensional models of subduction. For this research, three-dimensional numerical models will be constructed, run, and analyzed, to systematically test the controls on the lateral extent of the slab driven viscosity reduction in the mantle wedge due a strain-rate dependent viscosity. The results have implications for the length-scales of complex seismic anisotropy observed in subduction zones and may place constraints on the magnitude of coupling between the mantle and overriding plate. Moreover, understanding how the rheology modulates the viscous flow of the mantle has important implications for understanding the rates of tectonic plate motion, the length-scales of plate boundary zone deformation, as well as the three-dimensional transport of geochemical signatures within the volcanic front.