SCEC Award Number 15112 View PDF
Proposal Category Collaborative Proposal (Integration and Theory)
Proposal Title Joint inversion of direct P and S waves, head waves and noise dispersion data for the San Jacinto fault region
Investigator(s)
Name Organization
Clifford Thurber University of Wisconsin, Madison Yehuda Ben-Zion University of Southern California
Other Participants Xiangfang Zeng, Pieter-Ewald Share
SCEC Priorities 6a, 4a SCEC Groups Seismology, USR, CS
Report Due Date 03/15/2016 Date Report Submitted 03/13/2016
Project Abstract
 Thurber, Ben-Zion, their research groups, and collaborators are pursuing several lines of work aimed at improving the knowledge of the 3D crustal structure in the SJFZ region: (1) evaluating the efficacy of two automatic P- and S-wave arrival pickers and using them to expand the body-wave first arrival dataset; (2) expanding the database of fault zone head waves (FZHW) and associated direct P-wave arrivals; (3) carrying out joint body and surface wave inversions using several different starting models. The UW team completed a detailed comparison of the performance of the Ross and Ben-Zion (2014; RBZ14) and Rawles and Thurber (2015; RT15) auto-pickers, finding that the RBZ14 method outperforms RT15 for P waves but RT15 outperforms RBZ14 for S waves. The USC team worked to identify FZHWs recorded by linear arrays crossing the SJFZ at several locations, to be used in subsequent tomographic inversions. The bulk of our effort has been devoted to joint body and surface wave inversions, in collaboration with Prof. Haijiang Zhang and Hongjian Fang at USTC, using existing body- and surface-wave datasets. Joint inversions were done using four different starting models: a simple 1D model, the SCEC models CVM-H and CVM-S, and the waveform tomography model of Tape et al. (2010). Surprisingly, the inversion with the 1D starting model yields the optimal results. We used that model result to generate full-wave synthetic seismograms for an example earthquake, finding a good fit to the body-wave parts of the wave-forms up to 2 Hz.
Intellectual Merit This research establishes an approach for improving the SCEC CVM and ultimately its ability to predict observa-bles (seismic wave travel times, waveforms) by utilizing higher frequency data types, specifically P and S arrival times and ambient noise surface wave dispersion, and carrying out a joint inversion. Joint body and surface wave inversions at crustal scales are a relatively recent development with good promise for further improvement.
Broader Impacts This project has furthered the cross-fertilization of ideas and collaborations among the UW-Madison, USC, and USTC seismology research groups. It has led to a 2016 SCEC proposal involving an early career PI from an underrepresented group. The ultimate goal of the research is improved ground motion prediction capability. The obtained velocity model can be useful for various studies by other researchers, including earthquake location.
Exemplary Figure Figure 3. Horizontal slices of Vp (a, b, c, d), and Vs (e, f, g, h) at depths of 3 km, 7 km, 11 km, and 16 km from the joint inversion (Fang et al., 2016).
Linked Publications

Add missing publication or edit citation shown. Enter the SCEC project ID to link publication.