Proceedings of the 12th International INQUA meeting on paleoseismology, active tectonic and archaeoseismology

Fig. 2: a) Frictional strength of the AFS. Steady-state shear stress vs normal stress showing the Mohr-Coulomb envelopes. Effective friction coefficient ranged between 0.15 and 0.41 (colored lines). b) Fault friction stability. (a-b) parameter vs velocity steps. All values indicate velocity- strengthening to velocity-neutral behavior. by gradually reducing the normal stress from 10 MPa at a rate of 0.1 MPa per minute while maintaining the shear stress at 36-40% of the shear strength of each gouge. We observed a slow-to-fast transition in sliding velocity, sequentially identifying decelerating creep, steady-state creep, accelerating, and conditionally stable creep. Finally, with the onset of dynamic conditions (< 1 mm/s), we observed two different behaviors. The first behavior was characterized by a sudden increase in fault gouge dilation and a quasi- instantaneous acceleration in slip at velocities exceeding 19 mm/s, representing a seismic event. This phenomenon was observed in gouges dominated by chlorite and granular minerals. The second behavior involves a gradual transition from conditionally stable creep to millimetric oscillatory motion upon reaching slip velocities of 1 - 2 mm/s. This pattern was observed in the illite+muscovite-rich gouges. Finally, we investigated the dynamic fault weakening during seismic reactivation, revealing a significant decrease in the friction coefficient from 0.27 to 0.38 over a short slip weakening distance ranging from 0.2 to 0.6 m. We observed that as the quantity of phyllosilicates increases within the fault bands, the drops in friction diminish. This observation suggests that higher levels of phyllosilicates may attenuate earthquake propagation.