SCEC Award Number 18047 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title Earthquake Sequences with Fluid Diffusion and Fault Zone Pore Pressure Evolution
Name Organization
Eric Dunham Stanford University
Other Participants 1 PhD student
SCEC Priorities 3f, 3d, 1c SCEC Groups FARM, SDOT, Geology
Report Due Date 03/15/2019 Date Report Submitted 05/03/2019
Project Abstract
Dynamic rupture and earthquake sequence simulations require, as input, the distribution of effective normal stress (total normal stress minus pore pressure) on the fault. While total normal stress is typically close to lithostatic pressure, with deviations that depend on faulting regime, pore pressure is poorly known. Many modelers follow Rice (1992) in choosing an effective normal stress distribution that becomes constant below a few kilometers, which is results in depth-independent stress drops. Such effective stress distributions arise when pore pressure begins to track the lithostatic gradient due to upward flow (and viscous pressure drop) along permeable fault zones. In this project, we have explicitly introduced fault zone fluid flow and pore pressure changes into earthquake sequence simulations of continental strike-slip faults. We account for permeability reduction due to healing and sealing processes in the interseismic period, and permeability enhancement due to mechanical cracking during fault slip. We also account for direct dependence of permeability on effective normal stress. Fluid produced by metamorphic reactions at depth rises along the fault, where it accumulates and pressurizes during the interseismic period. This weakens the fault and triggers an earthquake, which is followed by a fluid surge and partial depressurization of the seismogenic zone. Our simulations thus permit quantitative exploration of Sibson’s fault valve model. We also discovered a new phenomenon, in which the boundary of the locked-creeping transition at the base of the seismogenic zone is driven upward by coupled permeability enhancement and aseismic slip. This process should be observable in geodetic data.
Intellectual Merit Fluids are central to earthquakes and faulting but have generally been neglected or oversimplified in previous earthquake modeling studies. By accounting for fluid pressure and fault zone permeability evolution, fully coupled to frictional slip, one can self-consistently predict how faults respond to natural or anthropogenic fluid sources. This simulation framework is also required to quantitatively connect with geologic observations of mineralized veins and other features that provide evidence of extremely high fluid pressures.
Broader Impacts This project trained three PhD students and one postdoc. One PhD student was a senior PhD student who completed her PhD thesis while conducting this project and then transitioned 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 1. Strike-slip earthquake sequence simulation with coupled fluid migration, pore pressure diffusion, and rate-and-state friction. (a) Fluid overpressure and (b) permeability evolution over two earthquake cycles. Permeability reduction in the seismogenic zone during the interseismic period causes overpressure development as fluids sourced from depth stall. Rupture increases permeability, opening the fault valve and permitting depressurization. Pressure cycles are most pronounced in the aseismically slipping region below ~12 km. (c)-(e) Development and upward migration of an aseismic slip front occurs in the final ~50 yr of the interseismic period due to coupling between flow, pressurization, and sliding. Such behavior ought to be expressed in crustal deformation data.