Application of hydrological models for water resources management at large continental river basins is often limited by the scarcity of in situ meteorological forcing data. Remote sensing information provides an alternative to in situ data, with observations that are, in some cases, at higher spatial and temporal resolutions than those available from traditional ground sources. In this work, the water balance equation is solved using precipitation retrieved from Tropical Rainfall Measuring Mission, water storage from Gravity Recovery and Climate Experiment satellite data and ground discharge. Evapotranspiration (ET) is then computed as a residual term of the water balance. Satellite data are compared with ground data to understand to what extent remote sensing observations can be used to improve estimates of the terrestrial water balance at regional scale. ET estimates are also compared with the ET computed from a detailed distributed energy water balance model and with the ET product from the Moderate Resolution Imaging Spectroradiometer Global Evapotranspiration Project. These analyses are performed for the Upper Yangtze River basin (China) in the framework of NRSCC-ESA DRAGON-2 Programme.