Fault reactivation by fluid injection: insights from laboratory friction experiments with multiple reactivation sequences

Faults can be reactivated by fluid injection and pore pressure increase in the rock volumes surrounding the fault zone. Induced earthquakes represent only one of the possible responses of active faults to pore pressure perturbations, since other strain transients characterize the spectrum of fault-slip behavior. A series of fluid injection experiments, designed and undertaken in the framework of the ERC Fault Activation and Earthquake Ruptures (FEAR) project, will be conducted in the Bedretto Underground Laboratory for Geosciences and Geoenergies (BULGG, Switzerland) to understand fault reactivation processes on a target well-identified fault zone. Small and accessible faults are to be instrumented to monitor deformation and seismicity during both fluid injection and fault reactivation.

The mineralogical, microstructural, and hydraulic properties of the target fault zone are investigated to characterize the fault-slip behavior. Characterization of the frictional response is achieved through a suite of laboratory rock-deformation experiments using both double-direct and rotary experimental apparatuses. Fault stimulation by fluid pressurization was also simulated in laboratory by using an injection protocol reliable for the in-situ hydraulic stimulation and consisting of stepwise pore fluid pressure increase. Experiments undertaken at low velocity with the double-direct apparatus (BRAVA) suggest that the selected fault, composed of mixed phyllosilicate-granular materials, is frictionally stable but yet can be dynamically reactivated by hydraulic stimulation.

Experiments were also performed on the fault gouge from the target fault with the rotary experimental apparatus (SHIVA). First, we apply half of the stresses measured at depth in the underground laboratory to accomplish the operating capability of the apparatus: 12 MPa normal stress, 7.5 MPa confining pressure and 1.5 MPa pore fluid pressure. Second, we imposed a slip rate of 10^-5 m/s for 0.01 m to have an equally compacted and textured layer. Third, we applied a shear stress so that an equivalent slip tendency of 0.35 is achieved (ca. 2.7 MPa), and kept it constant. We then increased stepwise the pore fluid pressure by 0.1 MPa every 150 s. This allows the spontaneous nucleation of slip events. After fault reactivation, the maximum slip velocity was set to 0.1 m/s. The fluid injection sequence results in a first reactivation (R1). Thanks to the nominally infinite slip available in SHIVA we run a second injection sequence up to a reactivation (R2).

Our experiments show two different styles of reactivation between R1 and R2. R1 reactivation is abrupt, with slip rate accelerating up to 0.1 m/s. Instead, R2 has a stage in which slip rate oscillates (0.5-3 mm/s) just before the last step of pore pressure increase leads to acceleration to 0.1 m/s. This would suggest a role for the shear fabric developed during the first reactivation, in which extensive grain size reduction might have led to stiffening of the fault, responsible of the oscillatory slip. This frictional behavior suggests the importance of considering the effect of texture development during multiple cycles of seismic slip. The generalization of our data and observations will contribute to shed light on the mechanics of faults and induced earthquakes by fluid pressure increase.