Exciting news! We're transitioning to the Statewide California Earthquake Center. Our new website is under construction, but we'll continue using this website for SCEC business in the meantime. We're also archiving the Southern Center site to preserve its rich history. A new and improved platform is coming soon!

Group B, Poster #062, Tectonic Geodesy

Investigating the Effects of the GACOS Tropospheric Correction on 6+ years of Integrated GNSS+InSAR Time Series Over the San Andreas Plate Boundary, California

Katherine A. Guns, Xiaohua Xu, David T. Sandwell, & Yehuda Bock
Poster Image: 

Poster Presentation

2023 SCEC Annual Meeting, Poster #062, SCEC Contribution #13171 VIEW PDF
Time series derived from Synthetic Aperture Radar (SAR) data, in combination with Global Navigation Satellite System (GNSS) vector time series have enabled us to observe and explore evolving crustal deformation processes at high spatial resolution, with regular temporal sampling and at plate boundary scales. In this study, we have extended the nine tracks (5 descending and 4 ascending Sentinel-1 satellite tracks) of integrated InSAR+GNSS time series of Xu et al. (2021) past the July 2019 Ridgecrest earthquake sequence in time to span the years 2014.85 – 2022.0, for the length of the San Andreas fault system in California and northernmost Baja California. Integrating with GNSS displacement ti...me series allows us to correct for long-wavelength atmospheric effects and adhere to an underlying precise terrestrial reference frame. Atmospheric delay effects and in particular, delays caused by the troposphere layer of the atmosphere, are one of the largest sources of noise in InSAR data, and its mitigation still presents a challenge for the community.

Here, we assess the strengths and weaknesses of a weather model-based correction in this area of the globe by applying the Generic Atmospheric Correction Online Service (GACOS) tropospheric delay correction grids (see Yu et al., 2018) to our 6+ year time series. We compare our integrated InSAR+GNSS time series with a twin GACOS-corrected integrated InSAR+GNSS time series version through interferogram-level statistics, displacement time series-level statistics and by differencing time series model-derived 2D parameter maps for each. These parameter maps are calculated from a pixel-by-pixel time series model that estimates the parameters of velocity, annual and semi-annual seasonal terms (both amplitude and phase offset from January 1), and for the six tracks that observe deformation from the 2019 Ridgecrest earthquake sequence, coseismic and postseismic terms using a spatially-varying distance-weighted inversion. We display the visual comparisons, and our initial statistical calculations show that the GACOS correction decreases the mean standard deviation of interferograms in 5 out of our 9 tracks of data (max decrease of 1.57 mm compared to our original time series). These results indicate a mixed improvement across our coverage area in California. We also plan in the future to investigate a tropospheric correction based on an available high-rate time series of troposphere zenith delays over the area of interest.