Direct Computation and Systematic Interpretation of Plate Driving Forces Worldwide

University Of California-Los Angeles, Los Angeles CA

Investigators

Abstract

The dynamic balance of forces that drive the Earth's plates is poorly understood for several reasons: undefined terms, oversimplified models, and the intrinsic difficulty of determining the forces exerted on plates by basal shear tractions and attached subducting slabs. This project uses simple explicit definitions that divide the balanced torques into 4 components: (1) lithostatic pressure torque; (2) side-strength torque; (3) basal drag torque; and (4) net slab pull torque . It applies sophisticated thin-shell modeling tools that incorporate real topography, variable heat-flow, variable crust and lithosphere thickness, realistic nonlinear rheologies, and faults. The balance of torques on each plate are being computed in steps: (1) construct a high-resolution model of the global lithosphere with 52 plates and 13 orogens; (2) impose the correct velocity on each plate, either at edges connected to subducting slabs (if any), or at points in the strong plate interior; (3) compute the reaction forces, and sum them, to find the total torque on each plate that must be supplied by basal shear tractions and net slab pull; (4) redistribute these torques as basal shear tractions on the bases of plates with no slabs. The global dynamic balance is recomputed to obtain more realistic stress and strain rate distributions within the plates and orogens without significant changes in plate velocities. To account for dependence on the effective friction of plate boundary faults (which many previous studies have shown to be low) computations are being carried out for a one-parameter suite of models with different effective fault friction coefficients. Predictions are being scored against (a) azimuths of seismic anisotropy under certain slab-free plates; (b) geodetic velocities, (c) seafloor spreading rates; (d) geologic slip rates on faults; and (e) intraplate stress directions. The sum of basal shear traction and net slab pull torques is being decomposed into its two components for each of the plates with attached slab(s). Basal tractions on plates with no attached slabs indicate the general level of basal tractions. Seismic anisotropy of the upper mantle, where known, gives two possible directions for basal shear traction. These constraints permit a global (52 plates x 3 torque components) inversion study to determine how net slab pull depends on known factors of relative plate velocity, absolute trench velocity, slab age, and/or trench depth. Determining the direction and sense of basal shear traction on (many) plates will constrain global convection models, allowing others to determine the viscosity and density-anomaly structures in the rest of Earth's mantle. Understanding how net slab pull depends on the factors listed above will help understand why plate velocities have changed in geologic history. The final best model of dynamic stress balance in the plates will give global predictions of anelastic strain rates in plate interiors, which will be used in seismic hazard estimation.