Effect of Porosity and Permeability Enhancement in Injection-Induced Aseismic Slip

Abstract:

Fluid injection can trigger seismicity in regions that are otherwise geologically stable. To effectively assess earthquake hazards, a better understanding of the physical mechanisms governing fluid-induced seismicity and pore pressure transmission in reservoirs is essential.

Several processes have been proposed for injection-induced seismicity, mainly pore pressure diffusion, poroelastic stress transfer, and fault loading by aseismic slip. The last process has come to the fore recently in numerical and observational studies, and it is the focus (together with pore pressure diffusion) of our work. Although we now have a more holistic view of the mechanisms responsible for injection-induced seismicity, there have been few quantitative modeling studies of this phenomenon that account for potentially important hydromechanical processes. Most models either prescribe a simple frictional and rupture model, or neglect the evolution of permeability and porosity that accompanies slip and pore pressure diffusion, which could have significant effects on the resulting slip. This motivates the more comprehensive modeling approach that we take.

In this study, we investigate the propagation of aseismic slip that occurs in response to fluid injection in a 2D strike-slip fault in anti-plane shear embedded in a semi-infinite elastic medium, with fluid transport confined to the fault zone. The fault is governed by rate-and-state friction, and injection occurs at a specified rate into the center of the fault. Porosity evolves with slip following Segall & Rice (1995). Permeability changes with porosity via a power law relation. We observe how effects of pore dilation and permeability enhancement compete to hinder/facilitate pore pressure diffusion, which impacts the propagation of aseismic slip. We analyze the conditions under which aseismic slip outpaces the pore pressure diffusion front, and the role of permeability enhancement accompanying pore dilation, which is often ignored in numerical models. We also make a comparison to the case when porosity and permeability evolution are neglected to quantify the importance of these nonlinearities. To more accurately assess injection-related earthquake hazards, we believe more efforts should be directed at incorporating these hydromechanical processes.