GNSS/InSAR Integration for 3-D Deformation Time Series in Southern California

Zheng-Kang Shen, & Zhen Liu

Submitted August 30, 2026, SCEC Contribution #15574, 2026 SCEC Annual Meeting Poster #085

We have developed a method to integrate GNSS and InSAR observations to produce a three-dimensional (3-D) surface velocity field of the Earth [Shen and Liu, 2020, 2025]. The method consists of the following key components: (a) an optimal interpolation approach that converts discrete GNSS velocity measurements into a continuous velocity field; (b) a pragmatic strategy for estimating uncertainties in InSAR and GNSS measurements, and incorporating these uncertainties as weighting factors in the data integration; (c) a global optimization procedure to determine ramp parameters for multiple InSAR tracks, thereby minimizing systematic biases in the solution; and (d) averaging InSAR line-of-sight (LOS) observations within small grid cells, followed by combining the de-ramped InSAR data with the interpolated GNSS velocities to estimate the 3-D deformation field in each grid cell using least-squares regression. We have successfully applied this method to integrate GNSS and InSAR data for investigating the 3-D crustal deformation field of California and western Nevada [Shen and Liu, 2025].

In the present study, we further extend this framework in a proof-of-concept application to southern California by integrating GNSS and InSAR time series to derive a 3-D surface deformation time series. We use Sentinel-1 SAR observations from two tracks (ascending track 64 and descending track 71), spanning the period from mid-2015 to mid-2019, together with 3-D position time series from approximately 1,000 continuous GNSS stations obtained from the MEaSUREs project. The two datasets are integrated using a Kalman filter that explicitly accounts for time-dependent observational and process uncertainties. The resulting product provides not only a spatially continuous 3-D deformation field, but also its temporal evolution. The results reveal predominantly quasi-linear horizontal deformation across most of the study area, together with pronounced temporal variations in the vertical component, including widespread seasonal signals and nonlinear deformation in regions influenced by groundwater extraction and recharge.

Key Words
GNSS, InSAR, deformation, time series, Southern California

Citation
Shen, Z., & Liu, Z. (2026, 08). GNSS/InSAR Integration for 3-D Deformation Time Series in Southern California. Poster Presentation at 2026 SCEC Annual Meeting.

Related Projects & Working Groups
Community Earth Models (CEM)