摘要：The amount of water stored on continents is an important constraint forwater mass and energy exchanges in the Earth system and exhibits largeinter-annual variability at both local and continental scales. From 2002 to2017, the satellites of the Gravity Recovery and Climate Experiment(GRACE) mission have observed changes in terrestrial water storage (TWS) with anunprecedented level of accuracy. In this paper, we use a statistical modeltrained with GRACE observations to reconstruct past climate-driven changesin TWS from historical and near-real-time meteorological datasets at dailyand monthly scales. Unlike most hydrological models which represent waterreservoirs individually (e.g., snow, soil moisture) and usually providea single model run, the presented approach directly reconstructs total TWSchanges and includes hundreds of ensemble members which can be used toquantify predictive uncertainty. We compare these data-driven TWS estimateswith other independent evaluation datasets such as the sea level budget,large-scale water balance from atmospheric reanalysis, and in situ streamflowmeasurements. We find that the presented approach performs overall as wellor better than a set of state-of-the-art global hydrological models (WaterResources Reanalysis version 2). We provide reconstructed TWS anomalies at aspatial resolution of 0.5∘, at both daily and monthly scales overthe period 1901 to present, based on two different GRACE products and threedifferent meteorological forcing datasets, resulting in six reconstructed TWSdatasets of 100 ensemble members each. Possible user groups and applicationsinclude hydrological modeling and model benchmarking, sea level budgetstudies, assessments of long-term changes in the frequency of droughts, theanalysis of climate signals in geodetic time series, and the interpretationof the data gap between the GRACE and GRACE Follow-On missions. Thepresented dataset is published at https://doi.org/10.6084/m9.figshare.7670849 (Humphrey and Gudmundsson, 2019) and updates will bepublished regularly.
