The role of aseismic creep, elastic-stress transfer, and pore-pressure diffusion on injection-induced earthquakes: Insights from meter-scale experiments and numerical models
Join this presentation In-Person: BEG Building 130, VR Room 1.116C Or Via Zoom
Join Zoom Meeting
Meeting ID: 932 0137 7630
Passcode: 013011
Presenter
Chas Bolton, PhD
Research Assistant Professor
UTIG
Jackson School of Geosciences
The University of Texas at Austin
Description
Pore-pressure diffusion and the effective stress principle are routinely used to explain induced seismicity. However, as induced seismicity has become more widespread over the past 10+ years it has become apparent that additional mechanisms are needed to explain the richness in fault behavior associated with subsurface fluid injection. Here, I will present results from meter-scale laboratory experiments and numerical modeling that provide new insights into the physics of injection induced earthquakes. The laboratory fault is heavily instrumented with arrays of strain gauges, slip sensors, and pore-pressure transducers that permit spatiotemporal measurements of fault slip, shear stress, and pore-pressure. Frictional instabilities are induced by injecting fluid at constant rate of 5 ml/min into a fault zone that is pre-stressed to ~ 96% of its failure strength. We observe a spectrum of slip modes, including aseismic creep, slow-confined ruptures, and fast-fault spanning instabilities. With progressive fluid injection, instabilities evolve from fast-fault spanning (peak slip rates ~ 20 mm/s) to slow stick-slip (peak slip rates < 0.005 mm/s). Fluid-induced instabilities are preceded by accelerating aseismic creep that is modulated by both fluid pressurization and elastic stress transfer generated by aseismic creep. Fast-fault spanning ruptures occur when the critical nucleation length scale, h*, propagates beyond the pressurized region and into frictionally unstable regions of the fault (i.e., where w/h* < 1). In contrast, slow slip instabilities occur at the end of the experiment when the pore-pressure front as propagated across the fault and increased h* such that w/h* ~ 1. In contrast to some theoretical arguments, laboratory ruptures often propagate well-beyond the leading edge of the pore-pressure front. Together our work shows that that fluid-induced fault slip is driven by a combination of fluid-induced aseismic creep and elastic stress transmission, and could help explain long distance triggering that is common in many areas that host induced seismicity.