Shear heating-induced thermal pressurization during earthquake nucleation

S. V. Schmitt, Department of Geophysics, Stanford University
P. Segall, Department of Geophysics, Stanford University
T. Matsuzawa, NIED, Tsukuba, Japan

Abstract

Shear heating-induced thermal pressurization has long been posited as a weakening mechanism during earthquakes. It is often assumed that thermal pressurization does not become important until earthquakes become moderate to large in magnitude. Segall & Rice [JGR, 2006], however, suggested that thermal effects may become dominant during the quasi-static nucleation phase, well before the onset of seismic radiation. Using the slip evolution given by rate- and state-dependent friction—along with reasonable estimates of heat and pore pressure transport parameters—they estimated that thermal pressurization dominates weakening at slip rates in excess of 10-5 to 10-3 m/s.

We further explore this problem numerically, assuming a fault in a 2D elastic medium and accounting for full thermomechanical coupling. The fault is governed by rate and state friction with the radiation damping approximation to simulate inertial effects. Thermal diffusion is computed via finite differences on a grid that adaptively remeshes to minimize computational expense while maintaining accuracy. To start, we neglect fault zone thickness and model the fault as a plane. This approximation is valid for times much greater than the diffusion time across the fault zone. With uniform transport properties, it leads to a direct relationship between pore pressure on the fault and temperature [Rice, 2006, JGR], thus requiring only one finite difference grid.

Our results thus far indicate that thermal pressurization does in fact dominate at modest slip speeds that are slightly lower than those estimated by Segall & Rice [2006]. Interestingly, the thermal pressurization process leads to a contraction of the nucleation zone, rather than the growing crack (aging law) or unidirectional slip pulse (slip law) associated with drained rate- and state-dependent frictional nucleation.

If allowed to proceed to higher—yet still quasi-static—slip speeds, our modeled nucleation zone continues to shrink nearly to zero width. We believe this is a consequence of treating the fault as a planar surface rather than a finite-width shear zone. Such an approximation overestimates the fault temperature at higher slip speeds, when time steps are no longer much greater than the diffusion time across the width of the shear zone. Our current work is to include the finite fault thickness, so that we may conduct simulations up to speeds at which seismic radiation becomes significant.



PDF of poster (886 KB).