Analyzing GNSS Time-Series Parameterization of Interseismic and Early Postseismic Deformation

Aubrey Bennett, Yehuda Bock, & Lavoisiane Ferreira

Submitted August 30, 2026, SCEC Contribution #15411, 2026 SCEC Annual Meeting Poster #068

Both the 2003 Mw 6.5 San Simeon and the 2004 Mw 6.0 Parkfield earthquakes were recorded by a near-fault GNSS network that captured coseismic and postseismic deformation. However, the short pre-earthquake record produced strong tradeoffs among interseismic velocity, coseismic offsets, and two episodes of postseismic deformation. Conventional decay models have difficulty representing both the rapidly evolving first postseismic days and the slower return toward steady interseismic motion.

We investigate whether the postseismic transients can be represented by a small number of network-derived temporal modes. First, model parameters for each station’s component time series are estimated independently using least squares. An independent interseismic model supplies the long-wavelength velocity field that is poorly constrained by the Parkfield observations. Principal component analysis is applied to both the transient model components and to residuals from the generalized least squares solutions to determine the dimensionality of the unmodeled transients. A single dominant component explains approximately 98% and 97% of the horizontal postseismic variance for Parkfield and San Simeon respectively, indicating that the transients are dominated by a coherent network-wide postseismic signal.

This single principal component is compared with the station models with independently estimated amplitudes, especially in the early postseismic period. The single principal component is differenced with each station’s horizontal timeseries at individual epochs to assess the spatial cohesion of the postseismic signal through time. This information is paired with the individually determined postseismic decay and amplitude from the least-squares solution to assess the spatial variability of the postseismic signal through time. This framework is used to observe the spatial coherence of postseismic transients and residuals, while evaluating the fit of conventional timeseries parameter models.

Citation
Bennett, A., Bock, Y., & Ferreira, L. (2026, 08). Analyzing GNSS Time-Series Parameterization of Interseismic and Early Postseismic Deformation. Poster Presentation at 2026 SCEC Annual Meeting.

Related Projects & Working Groups
Tectonic Geodesy