9725406 Peacock We propose to construct thermal and petrologic models of specific convergent plate margins and test these models against geochemical and seismological observations in order to better understand subduction-zone processes at 50-200 km depth. Specifically, we will simulate the thermal evolution of the Andean, Central American, and Cascadia subduction zones in order to test whether gaps in magmatic arcs and variations in the geochemistry of arc lavas can be related to differences in the thermal structure of the subduction zones. For each subduction zone we will construct several 2-dimensional models at different locations in order to generate a pseudo-3-dimensional model. The finite-difference models will incorporate changes in the geometry of the subducting slab, radiogenic heating, and slab ages that vary with time. We will also develop more accurate dynamical flow models for the mantle wedge that lies above the subducting slab in order to test the hypothesis that relatively weak rocks (e.g., sedimentary rocks, serpentinite) which may be present within the subduction shear zone will act to mechanically decouple the slab from the mantle wedge. The results of this study will increase our understanding of convergent plate margins (subduction zones) where most of the Earth's earthquakes and explosive volcanism occurs.
