Eikonal-Based Adjoint-State Inversion for the Structure of the Los Angeles Basin

Eli Bird, Ettore Biondi, Robert W. Clayton, & Zhongwen Zhan

Submitted August 30, 2026, SCEC Contribution #15435, 2026 SCEC Annual Meeting Poster #TBD

Due in part to computational limitations and variations in data quality, many models of seismic structure in Southern California are limited to a small subset of the available data. Typically, broader scale inversions focus on relatively sparsely-spaced permanent conventional sensors operated by the Southern California Seismic Network, while finer scale temporary arrays are used to create local models. The process of merging these results to create a multi-scale model remains an open research question, but creating such a model is critical from the perspective of hazard assessment.

We propose a lightweight inversion, focused on data products such as body wave travel times from earthquakes and ambient-noise-derived surface wave dispersion curves. This is a relatively computationally manageable approach to large quantities of data, and in moving to simple data products, variations in data quality become less problematic. We jointly invert these products on arrays at both the sparser and finer scale, first focusing our attention on the Los Angeles Basin, one of the least well-resolved regions in existing models of Southern California. We incorporate data from distributed acoustic sensing arrays, conventional seismometers, dense nodal deployments, and the Community Seismic Network. Considerable care must be taken in weighting data and normalizing results when combining datasets across meter and kilometer scales. We present the resulting tomographic models and compare predicted waveforms to real data.

Citation
Bird, E., Biondi, E., Clayton, R. W., & Zhan, Z. (2026, 08). Eikonal-Based Adjoint-State Inversion for the Structure of the Los Angeles Basin. Poster Presentation at 2026 SCEC Annual Meeting.

Related Projects & Working Groups
Seismology