Abstract:
Objectives In the inversion of regional terrestrial water storage (TWS) using global navigation satellite system (GNSS) data, it is common to expand the boundary of the study area to account for hydrological loading variations. However, the effects of far-field hydrological loading on GNSS stations are often overlooked. GNSS displacement data are applied in the inversion of regional terrestrial water storage, but the far-field loading contributions in GNSS data may lead to a mismatch between global-scale signals and the local characteristics of the regional inversion model.
Methods Based on the observation data from 49 GNSS stations in the Sichuan–Yunnan region, China, hydrological loading displacements are calculated using both global land data assimilation system （GLDAS) and gravity recovery and climate experiment (GRACE) data, and the loading effects are divided into local-field, near-field, far-field, and total-field components. Simulation experiments and inversion analysis by real GNSS observations are carried out. The inversion results before and after far-field correction are compared with GLDAS and GRACE data.
Results With GLDAS data as the global hydrologic loading, the displacement amplitudes of GNSS sites from local, near, and far fields can reach 3.8 mm, 1.6 mm, and 2.3 mm, respectively. Simulations using GLDAS-calculated site displacement show that the inversion results after removing far-field effects correlate most closely with GLDAS hydrologic loading scenarios, achieving a correlation of 0.98. After removing the far-field effect using GLDAS and GRACE, the extended inversion using GNSS data shows that the average amplitude of TWS inversion results is reduced by 21.0% and 25.4%, respectively. The grid equivalent water height series in the two inversions accounting for far-field effects have a higher correlation with those in GRACE and GLDAS than in the extended inversion.
Conclusions In the inversion of TWS in Sichuan-Yunnan region using vertical GNSS observations, the displacements caused by far-field hydrological loading cannot be ignored, and the far-field effect needs to be corrected while expanding the boundary inversion to ensure the accuracy of the inversion results.
下载:
