SCEC Award Number 20187 View PDF
Proposal Category Collaborative Proposal (Integration and Theory)
Proposal Title Development and Implementation of Full 3-Dimensional Waveform Tomography Applied to the Greater Los Angeles Region
Name Organization
John Vidale University of Southern California Thomas Jordan University of Southern California
Other Participants Brianna Birkel, USC graduate student
Alan Juarez, USC graduate student
Leonardo Ramirez-Guzman, Universidad Nacional Autonoma de Mexico
Armando Espindola, King Abdullah University of Science and Technology
SCEC Priorities 3a, 4a, 4c SCEC Groups CXM, Seismology, GM
Report Due Date 03/15/2021 Date Report Submitted 03/16/2021
Project Abstract
Southern California contains strongly heterogeneous structures that have been studied, mapped, and integrated into the SCEC CVMs. However, further refinement is required to achieve seismic wave propagation simulations that are accurate up to one second period, which would be a critical improvement for seismic hazard estimation.

In the first stage of this ongoing project, we are comparing seismograms observed from five recent moderate magnitude earthquakes within the Los Angeles Basin with calculations incorporating the CVM-S4 and CVM-H models. The synthetic seismograms used for comparison are those simulated via forward modeling with the octree-based full-3D tomography Hercules toolchain (Taborda et al. 2016) and the finite-difference code developed and operated by Rob Graves.

In Los Angeles basin, we see significant differences between the observations and predictions, even at periods longer than 5s. We are still assessing how systematic the differences are, and which aspects of the structure cause them. COVID disrupted our efforts, leaving us well behind schedule
Intellectual Merit Los Angeles mitigation is the primary rationale in the United States for understanding the amplification of shaking by large sedimentary basins. The basins are deep and extensive. The cities in the basin are replete with critical and aging buildings and infrastructure. The earthquakes are frequent. Lessons learned from Los Angeles will have application to similar earthquake-prone cities such as Seattle, Salt Lake City, and places with smaller basins such as the Bay Area and Portland. Structural models are now approaching the fidelity needed to model shaking to estimate seismic hazard down to roughly three seconds period. Engineers need estimates at even shorter period, which requires identification of shortcomings in the current models and paths toward improvement.
Broader Impacts The ultimate goal of this ongoing project is identification of regions within existing seismic velocity models that are in need of further study and improvement, as well as improvement of seismic velocity models that can be incorporated into Cybershake calculations of the seismic hazard for Southern California. Our work thus far identifies significant misfit between empirical and synthetic seismograms, which demonstrates the need for a study of this kind.
Exemplary Figure Figure 3 shows an example of the dramatic shortcomings of the current models.