Development and Recovery of Stress‐Induced Elastic Anisotropy During Cyclic Loading Experiment on Westerly Granite

Abstract

In the upper crust, where brittle deformation mechanisms dominate, the development of crack networks subject to anisotropic stress fields generates stress‐induced elastic anisotropy. Here a rock specimen of Westerly granite was submitted to differential stress cycles (i.e., loading and unloading) of increasing amplitudes, up to failure and under upper crustal conditions. Combined records of strains, acoustic emissions, and P and S elastic wave anisotropies demonstrate that increasing differential stress promotes crack opening, sliding, and propagation subparallel to the main compressive stress orientation. However, the significant elastic anisotropies observed during loading (≥20%) almost vanish upon stress removal, demonstrating that in the absence of stress, crack‐related elastic anisotropy remains limited (≤10%). As a consequence, (i) crack‐related elastic anisotropies measured in the crust will likely be a strong function of the level of differential stress, and consequently (ii) continuous monitoring of elastic wave velocity anisotropy along faults could shed light on the mechanism of stress accumulation during interseismic loading.

Seismic Attenuation Monitoring of a Critically Stressed San Andreas Fault

“…The dilatancy model (Scholtz, 2019) predicts time-dependent variability of crack density within crustal rocks subject to long term tectonic stressing. Cycles of increased permeability with increasing applied stress, and subsequent healing after failure are shown in laboratory experiments (Faulkner & Armitage, 2013;Mitchell & Faulkner, 2008;Passelègue et al, 2018;Vasseur et al, 2017). Time-dependent Q −1 S anomalies are thus expected to be enhanced by prefailure permeability changes caused by tectonic stress-induced dilatancy of crustal rocks.…”

Off‐Fault Damage Characterization During and After Experimental Quasi‐Static and Dynamic Rupture in Crustal Rock From Laboratory P Wave Tomography and Microstructures

“…These fractures decrease path‐averaged P wave velocities along horizontal ray paths more than those along angled ray paths, causing the observed anisotropy (Figure 3). The closing of the vertical microfractures, caused by the synfailure differential stress drop, results in partial P wave recovery (Passelègue et al., 2018). The lowest V P along ray paths intersecting the failure zone are thus observed when the contribution of fracture opening in the rupture process zone dominates over the contribution of fracture closure due to decreasing differential stress (Figure 3a, asterisks).…”

Brittle Faulting of Ductile Rock Induced by Pore Fluid Pressure Build‐Up

“…Here, such variation could not be measured during the sample macroscopic shear failure due to the large interval between two Vp measurements (i.e., every 2 min, while stress drop is quasi‐instantaneous). Note that the crack densities estimated at the failure of the specimens are smaller than the percolation threshold (i.e., critical crack density) for failure observed in previous studies (i.e., ≈ 0.13 rather than 0.3) (e.g., Passelègue et al., 2018; Wang et al., 2013). This is probably related to an underestimate of the crack density in fluid saturated media, as observed in previous studies (e.g., Sarout & Guéguen, 2008a; 2008b).…”