Abstract: Continuous and dense global navigation satellite system (GNSS) observations of surface deformation can be used to recover regional terrestrial water storage (TWS) changes with high spatiotemporal resolution, which is of great significance for the study of regional water cycle and climate changes and has become a hot topic in the field of hydro-geodesy. This paper introduces the basic theory and inversion algorithms of the loading Green's function method and Slepian basis function method, and the stability and performance of these two methods in retrieving regional TWS changes by using GNSS surface deformations are evaluated. Meanwhile, the latest progress in using GNSS-derived TWS changes to study regional water cycles and extreme climate changes is reviewed. The existing problems and future development trends of the inversion of GNSS observations (e.g. refined GNSS data processing and joint inversion of multisource data) are summarized and analyzed. It provides important insights into the use of GNSS observations to investigate regional TWS changes and related applications in hydro-geodesy.