MR33C-2691
Fault rupture as a series of nano-seismic events during high-velocity shear experiments

Wednesday, 16 December 2015
Poster Hall (Moscone South)
Ximeng Zu1, Zeev Reches1, Xiaowei Chen2 and Jefferson C Chang3, (1)University of Oklahoma Norman Campus, Norman, OK, United States, (2)University of Oklahoma Norman Campus, School of Geology and Geophysics, Norman, OK, United States, (3)Oklahoma Geological Survey, Leonard, OK, United States
Abstract:
The rupture process of experimental faults is investigated here by monitoring nano-seismic events (NSE) during slip in a rotary shear apparatus. Our experimental faults are made of two rock blocks with one to four miniature 3D accelerometers that are glued to the stationary block at distance of ~ 2 cm from the fault surface. Accelerations in the frequency range of 1 Hz to 200 kHz are recorded in slip-parallel (x), slip-transverse (y), and slip-vertical (z) directions. We conducted a series of 45 experiments on diorite and dolomite samples in two loading styles: classical velocity controlled loading, and power-density loading, in which the power-density (shear stress times slip velocity) is selected, and stick-slip events develop spontaneously according to the experimental fault response. The 3D accelerometer data were recorded at 106 samples/s, with acceleration resolution of 10 mV/g in recoding range of +/- 5 V.

The experiments were conducted at slip-velocity of 0.001-0.8 m/s and slip distance up to 1.38 m. The accelerometer observations revealed tens to hundreds of NSEs per slip in both loading styles; peak acceleration ranged from 1g to over 500g. A typical stick-slip with tens of NSEs in Fig. 1, shows: (1) An initial NSE at ~59.72 s (green) that coincides with a small stress drop (~10%, red); (2) Simultaneous macroscopic slip initiation (blue); (3) A swarm of NSEs occur as the fault slips, each NSE lasts 1-2 milliseconds; and (4) Details of the initial NSE are shown in Fig. 2. Based on waveform cross-correlation between frequency band from 20 kHz to 70kHz, we identify several groups of NSE clusters, and apply empirical Green’s function method to analyze event source spectra based on Brune-type source model. These NSEs are indicators of rupture propagation during the experimental fault slip.