Description:
<jats:title>Abstract</jats:title><jats:p>Within continental interiors far from plate boundaries, we hypothesize that the architecture of the lithosphere will influence dynamic pressure gradients produced as the lithosphere moves relative to the underlying asthenosphere. We investigate how these dynamic pressure gradients affect melt migration by percolative flow within the mantle beneath continental interiors. Our models show that for a lithospheric keel that protrudes into the “mantle wind,” the behavior of the system can be classified into three categories: (1) small‐scale convective instabilities modify both the geometry of the lithosphere‐asthenophere boundary (LAB) and associated dynamic pressure gradients so that the surface flux of melt is largest above a narrow downwelling at the center of the keel; (2) weak convection does not substantially modify the keel and persistent gradients at the LAB at margins of the keel control dynamic pressure gradients so that surface melt flux is elevated within the keel relative to its surroundings; and (3) when convective instabilities are suppressed due to sufficiently high viscosity in the asthenosphere, asymmetric dynamic pressure gradients due to mantle shear beneath the keel focus melt streamlines toward the upwind flank of the keel and direct melt away from the downwind flank. We interpret surface patterns of Cenozoic magmatism at the Colorado Plateau, particularly the “upwind‐downwind” asymmetry relative to the direction of motion of North America over the underlying asthenosphere, in terms of category (3) above. Our models quantitatively assess the viability of spatially variable dynamic pressure gradients as a physical mechanism to explain disparate observations at the Colorado Plateau, such as surface magmatism, heat flow, xenolith data, and seismic structure.</jats:p>