SCEC Award Number 14105 View PDF
Proposal Category Collaborative Proposal (Integration and Theory)
Proposal Title Estimating Directivity and Rupture Velocity for Moderate Earthquakes in the San Jacinto Fault Zone
Investigator(s)
Name Organization
Jeffrey McGuire Woods Hole Oceanographic Institution Yehuda Ben-Zion University of Southern California
Other Participants
SCEC Priorities 3c, 4b SCEC Groups Seismology, FARM, DRCV
Report Due Date 03/15/2015 Date Report Submitted N/A
Project Abstract
 Several studies of the San Jacinto Fault Zone (SJFZ) over the last few years have found evidence for contrasts of seismic velocities along with asymmetric fault-zone damage patterns and additional observables that may result from a preferred rupture propagation direction (time domain signatures of directivity, along-strike asymmetry of aftershocks, reversed polarity secondary deformation structures near step overs). We proposed to perform a systematic seismological analysis of rupture velocity and directivity for M > 3 SJFZ earthquakes that can be compared to and augment the existing results. We have performed empirical Green’s function deconvolutions and second moment inversions for a number of M > 4 earthquakes and are in the process of developing a catalog of these results. A USC graduate student is being trained in the second moment inversion approach to carry this work forward. One of the most interesting result to date is an apparently fairly clear correlation between the second moment estimates and measures of peak ground motions. We find that the Peak Spectral Accelerations (PSAs) in the frequency band around the corner-frequency are highly correlated with the expected directivity signal from our second moment estimates for the March, 2013, M5.1 earthquake on the SJFZ. We are currently extending this study to smaller earthquakes.
Intellectual Merit Moderate earthquakes are not well suited for finite fault inversions because of a lack of geodetic data to help improve the uniqueness of the solution. However, their finite source properties are still well resolved from seismic data in a gross sense. The second moments inversion procedure is a relatively novel approach that quantifies the length, width, duration, and propagation velocity of a rupture in a well constrained (6 parameter) inverse problem. Our formulation is general, e.g. applicable to any earthquake without bias. The SJFZ dataset is unique in Southern California in terms of the density of on-scale recordings of magnitude 4-5 earthquakes. We will develop the first real catalog of rupture velocity estimates for earthquakes in this magnitude range.
Broader Impacts A USC graduate student is being trained in the second moment inversion approach to carry this work forward. Perhaps the most interesting result to date is an apparently fairly clear correlation between the second moment estimates and measures of peak ground motions. We find that the Peak Spectral Accelerations (PSAs) in the frequency band around the corner-frequency are highly correlated with the expected directivity signal from our second moment estimates for the March, 2013, M5.1 earthquake on the SJFZ. We are currently extending this study to smaller earthquakes.
Exemplary Figure Figure 3. Peak Spectral Acceleration values from accelerometer data in the SJFZ data for the M5.1 March 2013 earthquake at periods from 0.1 to 3.0 seconds. The black curves denote the ‘Phase 1’ GMPEs of Kurzon et al. [2014]. Each data point is the average of the two horizontal components. The points have been color-coded based on the expected source time function duration, τc(s), predicted at this station from on our estimate of the 2nd moments. The color scale is in seconds, similar to Figure 2. In the period bands around the corner frequency (~0.3s to 0.8s) there is a significant separation between the forward direction stations (blue colors) and the backward direction stations (red) resulting from the unilateral propagation to the NW. The forward direction stations having higher peak ground motions as expected. No corrections have been made to the data for site effects, which are undoubtedly present.