GPS Imaging of Crustal Strain and Uplift Rates in the Western United States Abstract


  • We present detailed maps of vertical and horizontal land motion imaged using data from many thousands of GPS stations located in the Pacific/Juan de Fuca/North America plate boundary deformation zone. These images are based on GPS coordinate time series produced as a part of the NASA Plug and Play Project that provides open access to data products for over 14,200 GPS stations on Earth.  "GPS Imaging" incorporates new sophisticated signal processing and image filtering techniques to generate images of land motion with highly improved levels of robustness, spatial and temporal resolution. Our prototype method uses an automatic, robust, and accurate non-parametric estimator of station velocity called MIDAS, which is insensitive to common problems that plague GPS coordinate time series analysis, including unknown step discontinuities, outliers, seasonality, and heteroscedasticity. GPS Imaging filters and interpolates the robust velocities using Delaunay triangulation and median-based spatial filtering to create velocity fields, strain rate maps, and uplift rates.  The method preserves rate field edge-like discontinuities and resolution is limited only by the density of the GPS network rather than an arbitrary predefined grid.   The images of strain rate and uplift reveal deep Earth processes with unprecedented scope and accuracy. The results provide a synoptic view that enhances interpretive value of the signals from postseismic relaxation, interseismic strain accumulation on fault systems, volcanic uplift and subsidence, anthropogenic groundwater withdrawal subsidence, elastic rebound from water mass unloading, and range-wide uplift of the Sierra Nevada.  At regional scales, the images reveal that lithospheric flexure and mantle flow processes can dominate vertical land motion over extended regions.

publication date

  • 2016

presented at event