Evaluation of GRACE and GRACE-FO terrestrial water storage change data in California using GNSS elastic deformation measurements 

Abstract

GRACE and GRACE-FO measurements of terrestrial water storage (TWS) change can help with the management of water resources. This is particularly important for California because of its large population, extensive irrigated lands, limited freshwater resources, and the impact of climate change, such as more frequent and prolonged droughts. California’s TWS signal has a significant spatiotemporal variability related to, e.g., snow in Sierra Nevada, groundwater in Central Valley, and large differences in TWS change magnitude in its northern and southern parts. Thus, it is important to evaluate the ability of GRACE and GRACE-FO to measure TWS in this region accurately. We begin by inspecting differences between seasonal variations in GRACE/GRACE-FO TWS data from SDS L2 data and the JPL, GSFC, and CSR mascon solutions. Then, we use elastic deformation measurements from ~1000 permanent GNSS stations in California to validate the GRACE/GRACE-FO observations of TWS changes. We invert 16 years (2006-2021) of GNSS measurements to estimate a high-resolution (25 km) map of seasonal TWS change in California. We then convert the seasonal TWS map estimated from GNSS to spherical harmonic coefficients and truncate at degree 40-60, commensurate with GRACE/GRACE-FO effective spatial scale. We then compare all GRACE/GRACE-FO TWS solutions to the low-pass filtered GNSS TWS map. This analysis demonstrates that the seasonal TWS map from GRACE/GRACE-FO is precisely a low-pass filtered version of the one from GNSS, highlighting the usefulness of GNSS for evaluating GRACE data. Through this comparison, we discuss the (1) similarity and differences among GRACE/GRACE-FO TWS change solutions, (2) decay of TWS amplitude from L2 solutions due to filtering, and (3) spatial regions where each TWS change solution might suffer from errors. This study offers a unique tool for validation various GRACE/GRACE-FO TWS solutions.