Fault core structure and the fault slip behavior during fluid injection: insights from laboratory friction experiments

Fluid pressure plays a crucial role in controlling fault reactivation in both natural and induced seismicity. The effective normal stress is linearly reduced by an increase in fluid pressure (P_f) with σ_eff = σ_n  - S · P_f, which lowers the frictional strength of the fault, τ=µ·σ_eff, increasing the potential for fault reactivation and seismic slip. Upon reactivation, slip can occur quasi-statically or dynamically depending on the interplay between σ_n and P_f and mediated by the S parameter, (the hydromechanical coupling) and by the rate-and-state properties of the fault materials. In this context, the fault zone structure dictates the frictional properties as well as the fault hydro-mechanical coupling, particularly when S ≠ 1 and reactivation might occur at effective stresses different than predicted.

In Bedretto Underground Laboratory for Geosciences and Geoenergies (BULGG, Switzerland), a target fault zone chosen for fluid-induced fault stimulation is characterized by a fractured host rock surrounding a sub-centimetric fault core with fault gouge and bare-rock asperities.  Therefore, to define slip mode of the target fault it is important to characterized the frictional properties of both fault gouge and bare-rock asperities taking advantage of a laboratory controlled experimental environment.

Fault stimulation by fluid injection was simulated in laboratory by increasing the P_f following an injection protocol suitable for the BULGG fluid stimulation. Experiments were performed on both the fault gouge sampled from the target fault and on bare rock surfaces sampled in the surrounding host rock. We employed a rotary shear apparatus (SHIVA) to perform fluid injection experiments. First, we imposed the stress conditions measured at depth in the underground laboratory, halved due to apparatus limitations: 7.5 MPa σ_eff, 7.5 MPa confining pressure and 2.5 MPa P_f. Second, we imposed a slip rate of 10^-5 m/s for 0.01 m to develop a stable fault core structure. Third, we applied a constant shear stress of 2.7 MPa, considering the slip tendency measured on the target fault (0.35). We then increased stepwise the P_f by 0.1 MPa every 150 s. After fault slip initiation, the maximum allowed slip velocity was 0.1-1 m/s. Between each of the experimental stages, permeability and transmissivity were measured with the gradient (Darcy) or P_f oscillations methods.

We show that fault reactivation and slip behavior are different between gouge and bare rocks: in gouge creep and dilatancy precede reactivation, whereas in bare rock surfaces reactivation is sudden and not preceded by neither tertiary creep nor dilatancy indicating that dynamic reactivation is promoted in smooth bare-rock surfaces. Moreover, gouge displays a higher friction (0.58 vs 0.49) and a lower hydraulic transmissivity (i.e., 10^-19 vs 10^-17 m³) than bare rocks.

Here we proceed to test a suite of constitutive models against our data: rate and state friction, (Rudnicki, 2023; Cappa et al., 2022), and a fully coupled poromechanical model (Dal Zilio et al., in prep.), to understand what are the physical processes controlling the onset and style of fault activation in the two fault core structures.