Quasi‐static fault growth and cracking in homogeneous brittle rock under triaxial compression using acoustic emission monitoring

Abstract

Abstract. This paper describes the localization of deformation acceleration in the period prior to dynamic failure in hornblende schist rock under triaxial compression using acoustic emission (AE) monitoring. Rather than stabilize the failure process by controlling axial stress to maintain a constant rate of AE (for monitoring AE hypocenters) as in previous works [e.g., Lockner et al., 1991 ], we have instead developed a rapid multichannel data collection system. This enables us to elucidate the dynamics of fault nucleation under condition of constant stress (creep) loading, which is a better approximation to low strain rate condition in the Earth and allows both quasi-static and dynamic crack growth to occur. The waveforms of more than 8000 AE events which occurred mainly during a 15 s period were recorded on 32 channels, with a sampling rate of 50 ns and mask time of 200 gs. Hypocentral locations of AE sources revealed that the fault initiated at one end of the core and then propagated into the unfaulted rock with a process zone (fault front) of intense cracking. We found that there were two different processes operating during the quasi-static nucleation of a shear fault, namely, a process zone in front of the fault tip and a "wake" of damage zone following the process zone. The process zone had the following features: (1) major tensile cracking, (2) low b value and fewer larger events, and (3) strong self-excitation. The mechanism of crack interaction and fault growth was, therefore, a mutual enhancement on dilatation due to tensile cracking. On the other hand, the damage zone was characterized by (!) major shear cracking, (2) low b value and more larger events, and (3) weak self-excitation, indicating that in the damage zone, following the development of a shear fault, linkage between cracks became the major mechanism of crack interaction and fault development. The mutual changes of b value and self-exciting strength observed in our experiments seem to occur as a result of the hierarchy of fault growth, which was not observed under slowed down loading conditions. Therefore our experimental results, under a realistic approximation of the dynamic condition of the Earth, are meaningful for the problems of earthquakes as well as rock bursts.

Numerical simulation of failed zone propagation process and anomalies related to the released energy during a compressive jog intersection

“…For a low cut-off factor P (omitting many big data), the fluctuation in b 0 is apparent. Experimental results show that the related variable b decreases to a single minimum or double minimum before fault nucleation [Lockner et al 1991;Lei et al 2000] and before rapid stress drop [Main et al 1992;]. The present numerical results support these findings.…”

“…Thus, the relation between the logarithm of the number and the average released energy can be determined. Linearly fitting the relation and multiplying the slope by −1 yields b 0 , whose units are J −1 , different from the common b-value [Lockner et al 1991;Lei et al 2000].…”

Characterization of the Frequency of Extreme Earthquake Events by the Generalized Pareto Distribution

“…First, we note that the exponent 1/ξ is approximately 1 for MORZ (b-value = 1.5) compared to 2/3 for subduction zones (b-value = 1). Such a value 1/ξ ≅ 1 of the exponent is found in many contexts in the distribution of energies associated with acoustic emission records of heterogeneous materials brought to rupture (Pollock, 1989;Omeltchenko et al, 1997;Fineberg and Marder, 1999;Lei et al, 2000) as well as in a model of self-organizing fault networks with faults interacting through long range elasticity, which heal slowly or not at all after an earthquake (Sornette and Vanneste, 1996). The value 1/ξ = 1 thus seems characteristic of events controlled by faults remaining weak or "open'', for a long time compared to the recurrence time between large events.…”

Fracture process zone in granite

“…Second, in the shallow crust a combination of tensile and shear mode cracks may lead to fault rupturing. Third, the controlled variation of the propagation velocity of the fault allows us to bridge the gap between previously published laboratory acoustic emission results on dynamic fracture [Lei et al, 2000] and quasi-static rate-controlled tests [Lockner et al, 1991 ]. A second study with a series of confined asymmetric tests is under way which focuses on the process zone in different materials.…”