Hydrological applications of Gravity Recovery and Climate Experiment (GRACE).

摘要

The NASA/DLR Gravity Recovery and Climate Experiment (GRACE) satellite mission was launched March, 2002. GRACE provides monthly Stokes coefficients of spherical harmonics representing the global gravity field and changes over time with unprecedented accuracy. Gravity changes estimated from GRACE include effects from planets, tides, solid Earth deformation, and motion within the fluid envelopes of the atmosphere, hydrosphere, and oceans. After corrections for known or predictable effects, it is possible to isolate changes due to redistribution of water mass, especially related to storage changes in river basins.; This study addresses four issues related to hydrological applications of GRACE. The first is to examine optimum filter designs to obtain the maximum spatial resolution from GRACE Stokes coefficients and understand possible GRACE errors associated with the filters. One strategy for the optimum filter design is to adjust the filter on a monthly basis according to signal changes. A time variable filter can perform better than a fixed coefficient filter if good model predictions are available.; GRACE aliasing errors are estimated from numerical model fields. Aliasing errors from land and oceans sharply increase up to SH degree 15 and slowly increase thereafter. Atmospheric aliasing errors are significant below SH degree 15.; Global water mass loads are recovered from GRACE using various schemes. The largest water mass load changes are observed in low latitude basins such as the Amazon, Congo and Ganges, and show clear seasonal signals. GRACE signals are comparable to GLDAS prediction, providing some validation GRACE result.; Gravity potential differences along GRACE ground track are computed from numerical model fields. Using the different time scales between changes in terrestrial water storage and atmospheric surface pressure, the surface pressure field can be removed. This suggests the possibility of improved GRACE dealiasing.