Fluid-induced aseismic slip can outpace pore-fluid migration – evidence from in situ data
Pathikrit Bhattacharya, & Robert C. ViescaPublished August 15, 2018, SCEC Contribution #8814, 2018 SCEC Annual Meeting Poster #185
Earthquake swarms attributed to subsurface fluid-injection are usually assumed to occur on faults destabilized by elevated pore-fluid pressures. But theory suggests that fluid-injection could also activate aseismic slip which might ultimately outpace pore-fluid migration and transmit earthquake triggering stresses beyond the fluid-pressurized region. We test this prediction by comparing numerical models of fluid-induced aseismic slip against in-situ data derived from fluid-injection experiments that activated and measured slow, aseismic slip on pre-existing, shallow faults (Guglielmi et al., 2015).
In our model, we couple axi-symmetric pore-pressure diffusion away from the borehole (within a permeable and compliant fault damage zone) to the slip response of a quasi-circular, planar, shear rupture governed by a frictional strength criterion. We use this model within a Bayesian inference framework to compare the numerical predictions to the in-situ pore-pressure and slip observations of Guglielimi et al., (2015). The inferred model provides direct in-situ constraints on the joint evolution of the hydrological and mechanical properties of the fault and its surrounding medium with active slip and the ambient state of stress.
We find evidence for permeability enhancement of about 60% with accumulating pore-pressure and slip -- the resultant model captures the pore-pressure evolution well with hydrological parameters that are consistent with laboratory-derived values. We further find that the modeled aseismic slip, induced by the pore-pressure diffusion model that best explains the hydrological data, fails to fit the observed acceleration in aseismic slip when prescribed a constant frictional strength. This accelerating slip is, instead, captured well by the model when friction is allowed to weaken linearly with slip. Remarkably, the joint observational constraints on both the modeled pore-fluid migration and the rupture front require that aseismic slip indeed outpace pore-pressure diffusion during late stages of injection. This suggests that aseismic slip mediated stress transfer could represent a plausible mechanism for trigerring induced-seismicity even on faults that are located outside the subsurface volume pressurized by pore-fluid migration. We also discuss preliminary results on the response of such large, spatially-extended, regions of aseismic slip to injection shut-off at the borehole and how such models might explain observations of continued induced-seismicity after the termination of injection.
Key Words
fluid-induced fault slip, rupture nucleation, linear slip-weakening, Bayesian inversion, fault hydrology
Citation
Bhattacharya, P., & Viesca, R. C. (2018, 08). Fluid-induced aseismic slip can outpace pore-fluid migration – evidence from in situ data. Poster Presentation at 2018 SCEC Annual Meeting.
Related Projects & Working Groups
Fault and Rupture Mechanics (FARM)