The investigator will undertake a study of non-volcanic tremor in southern California using continuous seismic data made available by the Southern California Seismic Network (SCSN). Initially, they will focus our search for tremor near Anza, California in a densely instrumented region of southern California that has both surface and borehole continuous seismic data. These high-quality data will be used to determine the frequency content and duration of tremor that occurs near Anza allowing for tuning of future searches to these features. An automatic tremor detection scheme will then be used to search for tremor across all SCSN stations over a 5-year time period. These efforts provide the foundation to address fundamental questions about physical conditions necessary for non-volcanic tremor in California.
They will also determine whether tremor is correlated with local earthquakes, teleseismic wave arrivals, Earth tides, or observations of slow slip or aseismic creep. And, they will determine whether tremor occurrence correlates with known fault geometries or material properties. The major questions to be addressed in this proposal are: 1) Does tremor occur primarily along well-developed, through-going faults such as the San Andreas or San Jacinto fault zones?; 2) Does non-subduction zone tremor occur primarily in regions of a fault that are likely to be in conditionally stable such as: a) at depths less than 5 km in shallow sediments and/or b) deep (>15 -20 km) on crustal faults beneath the seismogenic zone?, and 3) Is timing of tremor in California controlled by stressing events such as the passage of teleseismic waves, changes in local seismicity rates, or tidal forces? If so, what is the stress amplitude and period required to trigger tremor?
Tremor are low amplitude, long duration seismic signals that have been only recently discovered in seismic data recorded continuously. The first observations of tremor were made in Japan and found to occur deep on the subducting plate interface. Tremor has since been observed at multiple plate boundaries around the globe, most recently along several faults in California. Tremor typically occurs deep on faults, below the portion of the crust that ruptures during an earthquake. It turns out that tremor is part of a suite of seismic and geodetic phenomena likely associated with deep slip on faults. By studying the details of tremor occurrence, including when and where tremor is typically observed, under what conditions tremor can be triggered, and the relationship between tremor and regular earthquakes, we can investigate the frictional properties of faults to greater depths than was previously possible using seismic data.
We were awarded funds for the project: ‘Exploring deep fault mechanics by identifying non-volcanic tremor on Southern California faults’. We undertook a study of non-volcanic (aka tectonic) tremor in southern California using continuous seismic data made available by the Southern California Seismic Network (SCSN). Tremor is a seismic signal associated with slip occurring on deep faults, below the region where earthquakes typically occur. Tremor can give us a window in to the slip mechanics of the deep fault and also may help us understand variation in fault properties by comparing regions with and without tremor. We focused our search for tremor near Anza, California along the San Jacinto fault in a densely instrumented region of southern California that has both surface and borehole continuous seismic data. Tremor triggered by waves from a teleseismic (distant) earthquake had been observed in this region once previously, but the rate of tremor occurrence had not been systematically studied. Using the available high-quality data from the Southern California Seismic Network, we examined when tremor was triggered by passing waves from 44 M>7.4 teleseismic earthquakes, and 1 regional earthquake, that occurred between 2001-2011. The goal of the study was to determine the stress and perhaps period of the waves that resulted in triggering of tremor. We observe only one instance of triggered tremor during the 2002 M7.8 Denali earthquake (Figure 1). The Denali earthquake imposed the highest shear stress on the fault during the study period, with the exception of the nearby regional earthquake which resulted in even higher shear stresses. The regional earthquake did not trigger tremor, which suggests either the triggering waves have to be dominated by longer periods or that the fault was not close enough to a critical stress to fail. The observation of only one episode of triggered tremor was somewhat unusual as most regions that had previously been examined (e.g. Cascadia, San Andreas fault) had several instances of triggered tremor over relatively short time periods. These previous studies suggested that tremor was easily triggered because of very low effective friction (likely due to high pore pressures in the fault). Here, we found that a much higher stress was required to trigger tremor along the San Jacinto fault than had previously been reported for the San Andreas fault. Figure 1: Example waveform from station RDM showing the surface waves of the 2002 M7.8 Denali earthquake in the upper trace and an enlarged section below. The upper trace is original (raw) data and the lower trace shows the unfiltered data (grey) and the data filtered between 2-6 Hz (black). Tremor is visible in the lower trace and the tremor amplitude modulated by the amplitude of the passing surface waves. From Wang et al., Bulletin of the Seismological Society of America, 2013.
