Depth-Dependent Crustal Deformation from a Physics-Based Joint Inversion of GNSS and InSAR in Southern California

Mradula Vashishtha, William E. Holt, & Jeonghyeop Kim

Submitted August 30, 2026, SCEC Contribution #15580, 2026 SCEC Annual Meeting Poster #065

Crustal strain transitions vertically from locked faulting in the brittle upper crust to distributed shear in the lower crust. Standard geodetic inversions simplify this 3D deformation field by assuming a uniform elastic thickness (H) to convert surface strain rates into seismic moment accumulation (Kostrov, 1974). However, using a single bulk H—even when spatially varied to match seismic catalogs (e.g., Guns et al., 2024)—collapses depth-dependent strain partitioning into a single scalar thickness.

We address this limitation by replacing the single-H assumption with an explicit, depth-dependent strain field inversion. Velocity gradients are initially modeled with linear depth dependence constrained by surface boundary conditions and Vertical Derivatives of Horizontal Shear (VDoHS) stress rates. Where systematic misfits against earthquake moment tensors persist, we apply a higher-order quadratic expansion—yielding eight unknowns per node constrained by surface VDoHS rates, surface velocity gradients, and Saint-Venant strain compatibility equations to ensure a single-valued displacement field at depth. To constrain the 3D strain geometry at depth, we use normalized Kostrov moment tensor sums within discrete crustal volumes (0–5, 5–10, and 10–15 km) to define the unit strain tensor—fixing the relative ratios of all six strain components ε ̇_ij, while geodetic surface data scale the absolute magnitude. Grounded in the self-similarity of seismicity, this approach directly resolves vertical strain partitioning between the locked upper crust and deforming lower crust, providing physically consistent inputs for fault characterization and seismic hazard assessment in Southern California.

Citation
Vashishtha, M., Holt, W. E., & Kim, J. (2026, 08). Depth-Dependent Crustal Deformation from a Physics-Based Joint Inversion of GNSS and InSAR in Southern California. Poster Presentation at 2026 SCEC Annual Meeting.

Related Projects & Working Groups
Tectonic Geodesy