The motion of tectonic plates at the surface of the earth is caused by forces within the earth's mantle: the push of positive buoyancy at spreading ridges and the pull of sinking plates (slabs) at subduction zones. At the earth's surface these forces result in earthquakes where the two plates slide past one another (plate boundaries). The ability of the force from sinking slabs to effectively pull tectonic plates behind them depends on how the slab deforms within the earth's mantle, which in turn depends on how its material properties change as it deforms. In addition, the total force associated with the sinking slab depends on changes in the crystal structure of the minerals (phase changes), which lead to changes in density within the slab. Most of these phase changes occur between 410 and 660 km beneath the earth's surface, a region known as the transition zone. Ultimately, the deformation of the sinking slab inside the mantle is manifest in seismicity occurring within the slab to depths of 660 km beneath the earth's surface, and observables changes in plate motions at the earth's surface. The purpose of this study is determine, 1) how surface plate motions and the state of stress within surface plates react to, and provide feedbacks for, slab dynamics in the transition, 2) what is the origin of deep slab seismicity, and 3) if the observed shape of slabs is related to intrinsic properties of the subducting plate and plate boundary, or is instead a reflection of the time-dependent evolution of the slab. While the focus of this study is on deformation within the earth?s mantle, this deformation couples to the motions of plates at the earth's surface, which can cause destructive earthquakes and tsunamis.
To address these questions with will develop three-dimensional numerical models (simulations) of subduction dynamics. More specifically we will use the best laboratory and observational constraints on the mineral composition of the plate (the crust, the residual harzburgite and the mantle layers), phase transitions (including all major mineral components) and rheology of the plate and mantle. We will enable dynamically mobile plates and plate boundaries, which are essential for understanding the physical connection between slab deformation and surface plate motions. These models eliminate several simplifying assumptions used in previous studies allowing us to make connections between slab deformation and surface plate motions and slab seismicity. Model results will be compared to global data sets on slab shape, plate characteristics and kinematics, as well as regional seismic observations on the state of stress within slabs and seismic discontinuities (which occur at phase transitions) across subducting lithosphere. The results of this project will provide a much more complete and realistic understanding of the physical processes that control the deformation of slabs and their ability to pull tectonic plates, and insight into how slab seismicity is related to the larger scale deformation slabs. In addition, the research will support the training of a PhD student, contribute software to the modeling community, introduce undergraduate students to scientific research, and contribute to the material used in undergraduate instruction.
