SCEC Award Number 18050 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title Thermomechanical Earthquake Cycles
Name Organization
Eric Dunham Stanford University
Other Participants PhD student (Kali Allison)
SCEC Priorities 3b, 1c, 1e SCEC Groups FARM, SDOT, CXM
Report Due Date 03/15/2019 Date Report Submitted 05/02/2019
Project Abstract
Continental strike-slip faults like the San Andreas extend down into the viscoelastic lower crust and upper mantle. Increasing temperature with depth drives a transition from velocity-weakening to velocity-strengthening friction as well as the onset of viscous flow in the bulk solid surrounding the fault. We have developed 2D antiplane shear simulations of earthquake sequences on rate-and-state friction faults in power-law viscoelastic solids, featuring spontaneously nucleating ruptures that interact with transient and steady viscous flow at depth. Temperature changes from viscous and frictional shear heating are determined by simultaneous solution of the heat equation, with feedback into the mechanics problem through temperature-dependent viscous flow laws. We explore controls on seismogenic depth, recurrence interval, slip per event, lithospheric stress, and partitioning of tectonic displacement into fault slip and viscous flow. We find that seismogenic depth can be controlled by either the frictional velocity-weakening to velocity-strengthening transition or the onset of viscous flow, with the latter occurring primarily in simulations with hot geotherms and significant shear heating. Shear heating is largest when effective stress, which has a prescribed linear increase with depth in this study, is highest. Viscosity reductions from shear heating range from negligible (for low effective stress) to more than one or even two orders of magnitude (for high effect stress). This simulation methodology, available through our open-source code SCycle (, can be used to test candidate rheological models and to construct lithospheric stress profiles.
Intellectual Merit Effects of viscous flow are typically neglected in earthquake sequence modeling, under the assumptions that the depth of earthquakes is limited by frictional rate-dependence and that deeper viscous flow or aseismic fault slip both load the seismogenic zone in an effectively identical manner. However, these assumptions are not necessarily satisfied. This project extends earthquake sequence modeling capabilities from linear elastic solids to viscoelastic solids, allowing for naturally occurring transition from frictional sliding to distributed viscous flow. The modeling framework allows one to incorporate independently derived constraints on thermal and rheological structure to make testable predictions on earthquake depth, lithospheric stress, post- and interseismic deformation, and other observables.
Broader Impacts This project trained one PhD student, allowing her to complete her PhD thesis and transition to an NSF Earth Sciences Postdoctoral Fellow. The modeling software has been released to the community using an open-source license and can be obtained from
Exemplary Figure Figure 2. Cumulative slip and viscous strain from earthquake sequence simulations in viscoelastic solids, without shear heating (top row) and with shear heating (bottom row), with all parameters identical. Slip is contoured every 1 s during the coseismic period (red) and every 10 yr during the interseismic period (blue). This example shows that rupture depth can be limited by the frictional velocity-weakening (VW) to velocity-strengthening (VS) transition (no shear heating) or by the onset of viscous flow (with shear heating) that raises the brittle-ductile transition (BDT) above the VW-VS transition. Note differences in slip per event and recurrence interval. Elevation of the BDT by shear heating is most pronounced in simulations with high effective stress (e.g., hydrostatic pressure gradient, as in this case) and a shallow lithosphere-asthenosphere boundary (i.e., hot ambient geotherm). (From Allison, PhD thesis, 2018.)