Reconstructing terrestrial water storage anomalies based on climate data and pre-GRACE satellite observations

The Gravity Recovery And Climate Experiment (GRACE) and its follow-on mission, GRACE-FO, have observed global mass changes and transports, expressed as total water storage anomalies (TWSA), for over two decades. However, for climate change attribution and other applications, multi-decadal TWSA time series are required. This need has prompted several studies on reconstructing TWSA using regression or machine learning techniques, aided by predictor variables such as rainfall and sea surface temperature. However, the training period is limited to a couple of years, making it hard to capture interannual signals accurately. Furthermore, learned relationships between climate variables and water storage cannot be transferred straightforwardly to the past. To overcome the limitation and provide a more long-term, consistent dataset, we derive a preliminary reconstruction and combine it with large-scale time-variable pre-GRACE gravity information from geodetic satellite laser ranging (SLR) and Doppler Orbitography by Radiopositioning Integrated on Satellite (DORIS) tracking from Löcher et al. (2025). We reconstruct GRACE-like TWSA for the global land, excluding Greenland and Antarctica, from 1984 onward. We find that the seasonal cycle of our new reconstruction is consistent with that of previously published purely climate-data-based reconstructions. Moreover, in many regions, TWSA trends were markedly different in the pre-GRACE timeframe, and we thus suggest caution when interpolating GRACE-derived trends.