References of "Wang, Lin 012101825B"
     in
Bookmark and Share    
Full Text
See detailRegional Continental Water Storage Variations Inferred from Three-dimensional GPS Coordinates Time Series
Wang, Lin UL

Doctoral thesis (2016)

Recent advances in space geodetic techniques (including but is not limited to, Global Navigation Satellite System, GNSS, Very-long-baseline interferometry, VLBI) allow us to observe changes in continental ... [more ▼]

Recent advances in space geodetic techniques (including but is not limited to, Global Navigation Satellite System, GNSS, Very-long-baseline interferometry, VLBI) allow us to observe changes in continental water storage (CWS) depending on the extent and the amplitude of the load. Among the geodetic techniques, GPS is the most common observational tool because of its global distribution. GPS observations are used for many fields of studies, including seismology and tectonics. This thesis presents a method to obtain regional changes in continental water storage by inverting the three-dimensional GPS time series. The error sources from a regional study are studied first. In theory, the surface motions from each GPS station are caused by loads acting over the entire surface of the Earth. As we are only interested in the changing water storage in a particular basin, the loading signal from the far field, outside the region of interest, must be accounted for. From our simulation studies, we conclude that the mass changes locate outside of the study region cannot be neglected. We find that the coverage of the area needs to extend to about 20 degrees (about 20 000 km) of the basin center for a regional study. The second concern is the GPS time series. We find discrepancies over the globe between GPS observed displacements and forward modelled displacements using models of water storage. At annual periods, the thermal expansion of the GPS monuments and underlying bedrock, atmospheric loading, and the draconitic signal if not accounted for will introduce an error into the inversion. These errors may contribute to the disagreement between our forward modelled and observed ground motions. For 88% of the stations analyzed, we are able to reduce the WRMS on the GPS vertical time series by removing the modelled displacements using estimates CWS loading obtained from WaterGAP. We conclude that the most likely cause of the discrepancies come from the GPS observations themselves. Due to the observed discrepancy, we find that the uncertainties of the GPS time series should be re-estimated in any inversion study. Finally, we determine monthly CWS variations from GPS three-dimensional coordinate time series for the major river basins in Europe and North America. The results at the basin scale are validated against GRACE and hydrological models, the correlation between inferred CWS and GRACE or models are close to 0.9 and WRMSR are as high as 50% for some basins. We also demonstrate that the relative contributions of the GPS horizontal coordinates are about one third those of the vertical signals. We prove that by including the horizontal coordinates in the inversion that we are able to improve the inversion results. [less ▲]

Detailed reference viewed: 50 (8 UL)