Rethinking the effect of groundwater storage resolution in river discharge estimates: how baseflow contribution governs the benefits of high-resolution data

Abstract. Groundwater storage (GWS) sustains river discharge through its contribution to baseflow, particularly during low-flow periods that are becoming increasingly critical under climate change, yet it remains one of the most difficult components of the hydrological cycle to monitor. The Gravity Recovery and Climate Experiment (GRACE) and its Follow-On mission (GRACE-FO) provide large-scale observations of terrestrial water storage anomalies (TWSA), which can be used to constrain GWS estimates and improve hydrological modelling of river discharge. However, their coarse effective spatial resolution (~300 km) limits their applicability to small and medium-sized river basins. Future gravimetry missions are expected to improve the spatial resolution of gravity-derived water-storage observations, but whether this improvement translates into more accurate river discharge estimates, and under which hydrological conditions, remains unclear. Here, we investigate this question using the semi-distributed hydrological model STREAM forced with GWS anomalies (GWSA) datasets at 0.1°, 0.5°, and 3° spatial resolution. The 0.1° dataset is obtained through constrained Bayesian integration of GRACE/GRACE-FO observations into the World-Wide Water Resources Assessment (W3RA) model, while the 0.5° and 3° datasets are generated by spatially degrading the 0.1° product. The experiments encompass three European pilot basins (Ebro, Danube, and Elbe) and 24 gauging stations representing a broad spectrum of drainage areas and hydrological conditions. Our results show that drainage area alone is not a reliable indicator of the added value of higher-resolution GWSA estimates: improvements in river discharge estimates are generally modest and inconsistent across basin sizes. In contrast, a more pronounced sensitivity is observed across flow regimes, particularly under low-flow (LF) and very-low-flow (VLF) conditions, for which the median reduction in RRMSE across all gauging stations is approximately 10% and 14%, respectively, when moving from the 3° to the 0.1° GWSA product. Moreover, the benefits of higher-resolution GWSA extend to higher-flow conditions as the contribution of baseflow to total river discharge increases. These results identify baseflow contribution as a key predictor of the hydrological value of higher-resolution GWSA estimates and provide a basis for assessing where improved spatial resolution of future gravimetry missions may translate into meaningful gains in river discharge estimation.