Estimating Groundwater Quality Changes Using Remotely Sensed Groundwater Storage and Multivariate Regression

Abstract:

Global groundwater dependence is likely to increase with continued population growth and climate-driven freshwater redistribution. Recent groundwater quantity studies have estimated large-scale aquifer depletion rates using monthly water storage variations from NASA’s Gravity Recovery and Climate Experiment (GRACE) mission. These innovative approaches currently fail to evaluate groundwater quality, integral to assess the availability of potable groundwater resources. We present multivariate relationships to predict total dissolved solid (TDS) concentrations as a function of GRACE-derived variations in water table depth, dominant land use, and other physical parameters in two important aquifer systems in the United States: the High Plains aquifer and the Central Valley aquifer. Model evaluations were performed using goodness of fit procedures and cross validation to identify general model forms. Results of this work demonstrate the potential to characterize global groundwater potability using remote sensing.