The salar de Uyuni is a massive dry salt lake that lies at the lowest point of an internal drainage basin in the Bolivian Altiplano. A kinematic GPS survey of the salar in September 2002 found a topographic range of only 80 cm over a 54 × 45 km area and subtle surface features that appeared to correlate with mapped gravity. In order to confirm the correlation between topography and gravity/geopotential, we use local gravity measurements and the EGM96 global geopotential model to construct a centimeter-level equipotential surface corresponding to the elevation of the salar. Our comparison of GPS survey elevations with the equipotential surface estimate shows that 63% of the variance of the GPS elevations can be explained by equipotential surface undulations (and long-wavelength error) in the EGM96 model alone, with an additional 30% explained by the shorter-wavelength equipotential surface derived from local gravity. In order to establish a physical connection between topography and the geopotential, we also develop and test a simple surface process model that redistributes salt via the dissolution, transport, and redeposition of salt by precipitated water. Forcing within the model pushes the system to evolve toward constant water depth, with the salt surface approximating the shape of the local equipotential surface. Since the model removes almost all topographic relief with respect to the equipotential surface within a matter of decades, it appears that observed (~5 cm amplitude, ~5 km wavelength) residual topography is actively maintained by a process independent of gravity-driven fluid flow.

A key element in the preparation for NISAR's launch (see Section 1.2) in 2023 is developing a methodology for validating NISAR-estimated displacements against GNSS displacements. In collaboration with the NISAR Solid Earth Science team I developed a Jupyter notebook workflow to download and generate timeseries of Sentinel-1 interferograms and GNSS daily solutions, project their displacements into radar line-of-sight (LOS), and derive velocity trends for each GNSS station in the given study site region and its collocated InSAR pixel. These velocity trends are then compared to asses the agreement of Interferometric SAR (InSAR) with GNSS. The effect of varying the following parameters were investigated: the GNSS station velocity approximation model, the reference station selected, various applied InSAR corrections, and the number of InSAR pixels averaged to form the InSAR collocated velocity. Analysis was conducted at two study sites, the Mojave Desert in California and a section of the Central Valley in California. The Mojave study site did not meet the validation requirement set by the NISAR team while the Central Valley did. Another key element of this project is providing future users with a clear and simple workflow to compare GNSS and InSAR data. The workflow discussed in this thesis has been handed off to the Solid Earth Science team to maintain and improve the user accessibility by moving the workflow into a cloud based environment.

The higher variability of surface-water supplies due to a changing climate has motivated an increased reliance on groundwater for urban, industrial, and agricultural purposes. This is particularly evident in California’s San Joaquin Valley (SJV) where overexploitation of the underlying aquifer system threatens the quality and availability of future groundwater resources as well as lead to large magnitude land subsidence threating critical infrastructure. Direct observation of these reservoirs on policy relevant scales, however, has remained elusive. The three principle chapters of this dissertation explore the extent to which geodetic tools and data can be used to better characterize regional subsurface processes such as groundwater recharge, flow, and extraction. Chapter 1 provides the motivation of this research and an introduction to the data used in the subsequent chapters. Chapter 2 presents a new methodology for combining interferometric synthetic aperture radar (InSAR) and Global Positioning System (GPS) displacements. Chapter 3 explores the efficacy of seasonal surface displacement time series for characterizing locations of groundwater recharge and flow in the SJV. Chapter 4 describes a method for analyzing networks of groundwater level records and how they relate to an extension of Chapter 3 results.

We develop and test an algorithm for modeling and removing elevation error in kinematic GPS trajectories in the context of a kinematic GPS survey of the salar de Uyuni, Bolivia. Noise in the kinematic trajectory ranges over 15 cm and is highly autocorrelated, resulting in significant contamination of the topographic signal. We solve for a noise model using crossover differences at trajectory intersections as constraints in a least-squares inversion. Validation of the model using multiple realizations of synthetic/simulated noise shows an average decrease in root-mean-square-error (RMSE) by a factor of four. Applying the model to data from the salar de Uyuni survey, we find that crossover differences drop by a factor of eight (from an RMSE of 5.6 to 0.7 cm), and previously obscured topographic features are revealed in a plan view of the corrected trajectory. We believe that this algorithm can be successfully adapted to other survey methods that employ kinematic GPS for positioning.