ALBERT

Description: We investigate charge generation as a function of stress in fine-grained gabbro for both nominally dry samples and samples fully saturated with electrically conductive brine fluids similar to those observed in active earthquake fault zones. These experiments address a number of proposed and reported electrical precursory and coseismic phenomena associated with earthquakes. Compressive load was applied to one end of the sample in repetitive cycles using a pair of precision steel platens driven by a large hydraulic press. The samples were tested by cycling between constant low stress and constant high stress values with a 200-s periodicity. Net charge transport between the stressed and unstressed sample ends was monitored with a picoammeter. For the nominally dry samples, stress-stimulated current (SSC) transients on the order of 50–400 pA peak-to-peak were observed with a decay time constant ~10 s during stress loading and unloading. Under constant compressive loads of ~22 MPa, small negative polarity SSC of ~15 pA magnitude was observed as an offset from the baseline current at low load (5 MPa) conditions. For the fluid-saturated samples, neither transients nor SSCs were observed as a function of stress when the load was cycled, an observation that is consistent with more rapid internal self-discharge due to higher electrical conductivity of the sample. Because the Earth’s crust is fluid saturated, observation of significant electrical charge buildup is not expected during the observed slow stress accumulation prior to earthquakes or during any slow precursory stress release that may occur in the region of earthquake nucleation. However, observation of coseismic charge generation due to electrokinetic, triboelectric, and other processes may occur during earthquake stress drops, surface rupture, and seismic-wave arrivals from dynamic rupture.

Description: We investigate charge generation as a function of stress in fine-grained gabbro for both nominally dry samples and samples fully saturated with electrically conductive brine fluids similar to those observed in active earthquake fault zones. These experiments address a number of proposed and reported electrical precursory and coseismic phenomena associated with earthquakes. Compressive load was applied to one end of the sample in repetitive cycles using a pair of precision steel platens driven by a large hydraulic press. The samples were tested by cycling between constant low stress and constant high stress values with a 200-s periodicity. Net charge transport between the stressed and unstressed sample ends was monitored with a picoammeter. For the nominally dry samples, stress-stimulated current (SSC) transients on the order of 50-400 pA peak-to-peak were observed with a decay time constant approximately 10 s during stress loading and unloading. Under constant compressive loads of approximately 22 MPa, small negative polarity SSC of approximately 15 pA magnitude was observed as an offset from the baseline current at low load (5 MPa) conditions. For the fluid-saturated samples, neither transients nor SSCs were observed as a function of stress when the load was cycled, an observation that is consistent with more rapid internal self-discharge due to higher electrical conductivity of the sample. Because the Earth"s crust is fluid saturated, observation of significant electrical charge buildup is not expected during the observed slow stress accumulation prior to earthquakes or during any slow precursory stress release that may occur in the region of earthquake nucleation. However, observation of coseismic charge generation due to electrokinetic, triboelectric, and other processes may occur during earthquake stress drops, surface rupture, and seismic-wave arrivals from dynamic rupture.

Description: The 3 November 2002 M (sub w) 7.9 Denali fault earthquake triggered deformational offsets and microseismicity under Mammoth Mountain (MM) on the rim of Long Valley caldera, California, some 3460 km from the earthquake. Such strain offsets and microseismicity were not recorded at other borehole strain sites along the San Andreas fault system in California. The Long Valley offsets were recorded on borehole strainmeters at three sites around the western part of the caldera that includes Mammoth Mountain--a young volcano on the southwestern rim of the caldera. The largest recorded strain offsets were -0.1 microstrain at PO on the west side of MM, 0.05 microstrain at MX to the southeast of MM, and -0.025 microstrain at BS to the northeast of MM with negative strain extensional. High sample rate strain data show initial triggering of the offsets began at 22:30 UTC during the arrival of the first Rayleigh waves from the Alaskan earthquake with peak-to-peak dynamic strain amplitudes of about 2 microstrain corresponding to a stress amplitude of about 0.06 MPa. The strain offsets grew to their final values in the next 10 min. The associated triggered seismicity occurred beneath the south flank of MM and also began at 22:30 UTC and died away over the next 15 min. This relatively weak seismicity burst included some 60 small events with magnitude all less than M = 1. While poorly constrained, these strain observations are consistent with triggered slip and intrusive opening on a north-striking normal fault centered at a depth of 8 km with a moment of 10 (super 16) N m, or the equivalent of a M 4.3 earthquake. The cumulative seismic moment for the associated seismicity burst was more than three orders of magnitude smaller. These observations and this model resemble those for the triggered deformation and slip that occurred beneath the north side of MM following the 16 October 1999 M 7.1 Hector Mine, California, earthquake. However, in this case, we see little post-event slip decay reflected in the strain data after the Rayleigh-wave arrivals from the Denali fault earthquake and onset of triggered seismicity did not lag the triggered deformation by 20 min. These observations are also distinctly different from the more widespread and energetic seismicity and deformation triggered by the 1992 M 7.3 Landers earthquake in the Long Valley caldera. Thus, each of the three instances of remotely triggered unrest in Long Valley caldera recorded to date differ. In each case, however, the deformation moment inferred from the strain meter data was more than an order of magnitude larger than the cumulative moment for the associated triggered seismicity.