SCEC Award Number 25229 View PDF
Proposal Category Individual Research Project (Single Investigator / Institution)
Proposal Title Incorporating data from H/V observations into joint body wave-surface wave tomography: proof of concept
Investigator(s)
Name Organization
Clifford Thurber University of Wisconsin, Madison Hao Guo University of Wisconsin, Madison
SCEC Milestones A1-1, A2-1 SCEC Groups CEM, Seismology, RC
Report Due Date 03/15/2026 Date Report Submitted 03/09/2026
Project Abstract
 We proposed to test a new idea for incorporating results from Rayleigh wave ellipticity (RWE) inversions, obtained from H/V measurements (Lin et al., 2012), and potentially from receiver function (RF) inversions into joint body wave-surface wave tomography as a proof of concept. The ultimate goal is to greatly improve constraints on S-wave velocity (Vs) structure in tomographic models, especially at shallow depths and in basins. PI Thurber has both a current project and a pending proposal with the USGS external grants program on development and validation of P-wave velocity (Vp) and Vs models for the San Francisco Bay (SFB) region using joint body wave-surface wave tomography, employing an improved version of the joint inversion code of Fang et al. (2016). PI Lin has a current SCEC project and a pending USGS proposal on measuring and inverting RWE and RFs in northern California and the SFB area, respectively. We will use some initial results from these projects for our proof of concept. Our idea is to use the layered 1-D Vs models derived from the RWE observations to generate one-way S-wave travel times from each model interface up to the station on the surface and includes these as "virtual shot" data in our tomographic inversion. We recognize that these travel times are approximate, and the joint inversion will not fit them exactly (just as all data are generally not fit exactly), but we expect that using these travel times will lead to improvements in our Vs model, especially near the surface.
SCEC Community Models Used Community Velocity Model (CVM)
Usage Description The SCEC CVM model for the San Bernardino Basin was used to create the virtual VSP data.
Intellectual Merit We are developing a new paradigm for adding independent constraints to travel-time-based tomographic inversions. The concept is based on using independently derived seismic velocity models from a variety of possible sources to extract 1-D velocity models beneath many points at the surface (either with or without a seismic station present, depending on the type of source) to create "virtual vertical seismic profiles" (VVSP) of one-way travel times from depth to the surface. These VVSP travel times are then added to tomographic inversions just as other active-source data are included. This strategy has proven successful in two demonstration cases.
Broader Impacts Given the broad use of travel-time-based tomographic methods by the scientific and industry communities, our novel paradigm can have very wide applicability. With support from other sources, the VVSP concept will be presented at the April 2026 SSA meeting, freely allowing others to take advantage of this new method. A proposed NSF project extending this work, if funded, will expand the assessment of the applicability of this paradigm to multiple additional scenarios, with an emphasis on contributions intended to improve earthquake hazard estimation by improving seismic velocity models.
Project Participants Fan-Chi Lin, University of Utah, provided preliminary layered velocity models beneath seismic stations in the Sam Francisco Bay region that were used to create virtual VSP data for four stations.
Exemplary Figure Figure 3. (left) Initial Vp/Vs model and (right) Vp/Vs model obtained by inverting the San Bernardino Basin Virtual VSP data. Cross-sections are at 34.4° N.
Linked Publications

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