SCEC Award Number 24155 View PDF
Proposal Category Individual Research Project (Single Investigator / Institution)
Proposal Title Generation and Validation of a SCEC Multi-scale Community Velocity Model
Investigator(s)
Name Organization
Kim Olsen San Diego State University
SCEC Milestones A1-1, D3-2, D3-3 SCEC Groups CEM, GM, Seismology
Report Due Date 03/15/2025 Date Report Submitted 03/21/2025
Project Abstract
 The goal of this project is to incorporate velocity models of multiple resolutions available for Southern California into the SCEC community velocity model CVM-S4.26.M01 (CVM-SI hereafter) to generate a single CVM with improved resolution and accuracy. We perform 0-0.5 Hz 3D wave propagation simulations for a suite of 21 small-to-moderately sized events throughout Southern California in order to validate the improvement from the refined candidate velocity models. We propose a data-informed weighting strategy to further refine the merging of the candidate models. Our analysis first confirms that CVM-SI is overall the best performing regional model, particularly inside the major Southern California basins, with smaller regions in need of refinement. Using our proposed refinement method, we succeed in lowering the Fourier Amplitude Spectral bias to data near Santa Barbara, in the Central Valley, as well as in parts of the Salton Trough, by incorporating more accurate basin structures provided by the candidate models.
SCEC Community Models Used Community Velocity Model (CVM)
Usage Description We use CVM-SI as the starting model for this study where other candidate models with various resolutions are subsequently merged into CVM-SI. The model update process involves two main parts, namely model merging and refinement. We start out with models covering the largest areas, which are the models by Berg et al. (2021) and Fang et al. (2022), followed by the models with medium coverage, such as Fang et al. (2019), and the models covering the smallest areas.
Intellectual Merit We propose a novel refinement method to merge 3D velocity models, that ensures that only the fraction of a candidate model in which the fit to data is improved will remain in place as the next current model, using a weighting technique.
Broader Impacts The project aims at producing a unified and more accurate 3D velocity model of Southern California, which will lead to improved seismic hazard analysis.
Project Participants The work was carried out by Postdoc Te-Yang Yeh under supervision of PI Olsen. Input from SCEC director BenZion was also considered in the project.
Exemplary Figure Figure 2. (left) Improvement map for the combined bias, , from including the model by Berg et al. (2021). The boundary of Berg et al. (2021) is depicted by the green dashed polygon. (right) Weighting function informed by the improvement shown in the panel on the left, which is naturally incorporated with the boundaries of the model by Berg et al. (2021). Credit: Te-Yang Yeh.
Linked Publications

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