Monitoring time-varying terrestrial water storage changes using daily GNSS measurements in Yunnan, southwest China

• Published in: Remote sensing of environment Vol. 254; p. 112249
• Main Authors: Jiang, Zhongshan, Hsu, Ya-Ju, Yuan, Linguo, Huang, Dingfa
• Format: Journal Article
• Language: English
• Published: New York Elsevier Inc 01.03.2021; Elsevier BV
• Subjects: Anomalies; Daily precipitation; Data assimilation; Data collection; Drought; Drought characterization; Environmental monitoring; Extreme weather; Global navigation satellite system; GRACE (experiment); Heavy precipitation; Hydrologic drought; Hydrologic loading; Hydrology; Hydrometeorological extremes; Hydrometeorology; Independent component analysis; Inversion; Monitoring; Monitoring instruments; Navigation satellites; Navigation systems; Precipitation; Precipitation anomalies; Rainfall; Rainstorms; Satellite-borne instruments; Satellites; Terrestrial environments; Terrestrial water storage; Time series; Time-varying inversion; Water storage; China; Yunnan China; Hydrologic loading; Time-varying inversion; Drought characterization; Terrestrial water storage; Hydrometeorological extremes
• ISSN: 0034-4257; 1879-0704
• DOI: 10.1016/j.rse.2020.112249

Global Navigation Satellite System (GNSS) instruments provide a powerful tool to investigate spatiotemporal variations in regional-scale terrestrial water storage based on the solid Earth's elastic response to hydrologic loading signals. Here, we implemented an independent component analysis-based inversion method to investigate water storage changes and hydrometeorological extremes (e.g., heavy precipitation and droughts) in Yunnan. Our time-varying inversion allows us to reproduce the spatiotemporal evolution of terrestrial water storage changes. Three independent components (ICs) were chosen in our time-varying inversion model. The first two ICs contribute to 96.9% and 2.1% of data variance, respectively, and reveal annual water storage changes at different temporal scales. The third IC slightly explains the network time series and possibly correlates with the nonlinear water changes. The averaged GNSS, Gravity Recovery and Climate Experiment (GRACE), and Global Land Data Assimilation System (GLDAS) based multi-annual seasonal water changes have good consistency in their spatiotemporal signatures. All datasets suggest a gradually increasing trend in seasonal water storage variations from northeast to southwest Yunnan. The peak annual amplitudes of ~355 mm in the GNSS-inferred water estimates are larger than those of ~167–226 mm derived from the GLDAS and GRACE models. Hydrometeorological extremes are tracked in the various water time series, and GNSS-derived water deficits contribute to hydrological drought characterization. We also find good agreement between daily precipitation anomalies, GNSS-, and GLDAS-based water changes during the 2015 winter rainstorm. Our results demonstrate that a continuously operating GNSS network is a complementary tool to remotely measure terrestrial water storage changes and to provide valuable insights into operational hydrological monitoring. •Daily water changes are inferred from the solid Earth's vertical elastic response.•Water inversion scheme is based on the independent component analysis technique.•Geodetic and hydrological data show coherent spatiotemporal changes of water cycle.•GNSS-inferred water estimates allow for the characterization of drought events.•A GNSS network could be used to remotely monitor large precipitation events.