Hydraulic Diffusivity Enhancement during Injection-Induced Fault Reactivation

It has been well acknowledged, from in situ observations of fluid induced fault slip reactivation, that the interactions between fluids and pre-existing faults can be complex: while fluid flow can reactivate pre-existing faults generating seismic or aseismic slip, the shear slip can affect the hydraulic properties as well. The relationship between slip front and fluid front remains not fully understood. In this study, we use a series of laboratory injection reactivation tests to investigate hydraulic diffusivity evolution during fracture shearing. Fault reactivation tests are conducted under triaxial conditions, at different confining pressures (30, 60 and 95 MPa) for a saw-cut Andesite rock sample. Fluid pressure is measured by two pressure sensors located at the two opposite ends of the experimental fault, which we use to estimate the history of the effective hydraulic diffusivity and its associated uncertainties. For this, we develop a deterministic and a probabilistic inversion procedure, which is able to reproduce the experimental data for a wide time range of the different experiments. During the injection reactivation test, the hydraulic diffusivity increases by one order of magnitude: hydraulic diffusivity changes are mainly governed by the reduction of the effective normal stress acting on the fault plane, with a second-order effect of the shear slip.