SCEC Award Number 17133 View PDF
Proposal Category Individual Proposal (Integration and Theory)
Proposal Title Imaging sharp lateral velocity gradients using scattered waves on dense arrays: faults and basin edges
Name Organization
Zhongwen Zhan California Institute of Technology
Other Participants Chunquan Yu
SCEC Priorities 2d, 3a, 4a SCEC Groups Seismology, CXM, GM
Report Due Date 06/15/2018 Date Report Submitted 06/10/2018
Project Abstract
Sharp lateral velocity contrasts often exist across major faults, at basin edges, or along other geological features. While the SCEC community velocity models continue to improve in resolution, sharp velocity gradients remain difficult to image, due to limitations in data coverage and damping in inversions. Here, we analyzed broadband waveforms recorded by the Southern California Seismic Network and observed strong body-to-surface wave scattering for various teleseismic incident wave types. We focused our study on strongly scattered Love waves following the arrival of teleseismic SH wave. These scattered Love waves travel approximately in the same (azimuthal) direction as the incident SH wave at a dominant period of ~10 s but at an apparent velocity of ~3.6 km/s as compared to the ~11 km/s for the SH wave. Back-projection suggests that this strong scattering is associated with pronounced bathymetric relief in the Southern California Continental Borderland, in particular the Patton Escarpment. Finite-difference simulations using a simplified 2-D bathymetric and crustal model are able to predict the arrival times and amplitudes of major scatterers. The modeling suggests a relatively low shear wave velocity in the Continental Borderland.
Intellectual Merit This project demonstrated how scattered waves can be used to image sharp structural boundaries in Southern California. More specifically, we used the teleseismic body-to-surface wave scattering recorded on the Southern California Seismic Network (SCSN) to image the sharp boundaries and basin edges in the borderland area. This project lays down the foundation for systematic imaging of sharp edges in southern California seismic velocity structure.
Broader Impacts The method developed here demonstrated a new way of using the data collected on the Southern California Seismic Network. With the dense array, we are able to migrate the scattered energy back to its source without much aliasing. Our study shows that scattered waves could be used to calibrate/update existing SCEC community velocity models, e.g. CVM-S4.26 (Lee et al., 2014) and CVM-H15.1.0 (Shaw et al., 2015), in particular across the California Continental Borderland, where stations are sparse. Such practice may enable to more precise offshore earthquake locations as well as improved seismic hazard assessment through, for example, CyberShake seismic hazard model calculations (Graves et al., 2011). The methodology in this study can essentially be applied to other seismic arrays, such as the EarthScope Transportable Array, to image potential scatterers on a broader scale. This project has provided training to a postdoctoral researcher, and has inspired another SCEC project to use even denser nodal seismic arrays in the Los Angeles Basin to under the long shaking durations, which is critical in earthquake hazard assessment.
Exemplary Figure Figure 2. Back-projection of scattered wave envelope for the selected Tonga-Kermadec-Fiji event. Amplitude is normalized by that of the direct SH phase and muted if the phase coherence is not significant. Constant Love wave velocities 3.1 km/s and 3.6 km/s are used for the region to the west and east of the coast line, respectively. Inset is a comparison between topography and back-projected waveforms (black for stacked waveform, red for envelope) along AA’ profile. Dashed lines show perturbations of back-projected waveforms assuming a 0.1 km/s difference in the Love wave velocity of the Continental Borderland. Envelope peaks correlate with pronounced topographic relief at the Patton Escarpment and the Santa Cruz Basin.

Yu et al., GRL, 2017