The most recent episode of unrest in the Long Valley Caldera (LVC) manifested itself in 1997 as an increase in the rate of inflation of the resurgent dome and an increase in seismic activity in the south moat. Ground deformation in the LVC reached a peak in November 1997 and was accompanied by three M4 earthquakes with significant volumetric components, indicating that fluids were involved in the source process of these earthquakes (Dreger et al., 2000). This project seeks to ascertain the extent of fluid influenced faulting triggered by the hydrothermal or magmatic system under LVC.
We inverted regional broadband data to compute deviatoric and full moment tensor (FMT) solutions for events greater than M4.0 since 1993. We solve for the FMT to determine the coseismic volume changes in the source region and compare it to the deviatoric solution which assumes no volume change. We performed a time domain inversion of complete three-component displacement seismograms. Seven stations were chosen within the Berkeley Digital Seismic Network (BDSN) that provided the best azimuthal coverage and data quality. Green's functions were computed using the SoCal velocity model which is appropriate for the eastern California and Sierra Nevada regions. Both the data and the Green's functions were bandpass filtered between 0.02 to 0.05 Hz using a causal Butterworth filter.
The F-test was used to determine if the improvement in fit between the deviatoric and FMT solutions is significant. The volumetric components of FMT solutions which exceed the 90% confidence level are taken to be statistically significant. To determine the stability of our solution, we use a Jackknife test to iterate over all possible subsets of data. Solutions from different station combinations can be compared to determine if solutions are stable or are influenced by path effects. The results from this test indicate that our solutions are stable for sub-station combinations of two or more.
In the original study of the 1997 swarm, four earthquakes were found to have significant volumetric components including one which had waveforms similar to nearby tectonic events. In this more detailed study, the volumetric component of that earthquake was determined not to be statistically significant. In addition to the three 1997 earthquakes that were previously discovered, two other events were found south of the caldera wall with statistically significant volumetric components (Figure 27.1). These new events were not located near the 1997 events. Instead, they formed part of a 1998 earthquake swarm within the Sierra Nevada block. Interestingly, this sequence had hints of magmatic involvement that suggested some degree of magma or fluid movement (Hough et al., 2000).
The majority of the events in the LVC area do not have significant isotropic components. Out of the 32 events studied with M4.0 or greater, only five earthquakes had a significant volumetric component in their source region. All of the five earthquakes occurred in areas which had evidence suggesting active magmatic activity.
We appreciate support for this project by NSF through contract EAR-0087147.
Dreger, D. S., H. Tkalcic, and M. Johnston, Dilational Processes Accompanying Earthquakes in the Long Valley Caldera, Science, 288, 122-125, 2000.
Hough, S. E., R. S. Dollar, and P. Johnson, The 1998 Earthquake Sequence South of Long Valley Caldera, California: Hints of Magmatic Involvement, Bull. Seism. Soc. Am., 90, 752-763, 2000.