Flood monitoring with high-resolution TWS data from satellite gravimetry

The wetness conditions of a river basin, besides rainfall characteristics, are important factors for the amount of runoff that is generated, eventually leading to a flood event. Satellite gravimetry with GRACE and its successor mission GRACE Follow-On (GRACE-FO) allows for retrieving terrestrial water storage (TWS) anomalies by measuring temporal variations of the Earth’s gravity field.

In this study, we use TWS anomalies derived from daily GRACE/GRACE-FO data downscaled to a 50 km global resolution. This opens the possibility of estimating the wetness conditions before and during flood events. We collect a set of historical flood events at the global scale from multiple sources, including Dartmouth Flood Observatory (DFO). To complement our analysis, we use additional datasets, such as the GRDC global river discharge database. Within these datasets, flood events are identified as periods where hydrometeorological time series (e.g., river discharge, cumulative precipitation, or soil moisture) exceed certain thresholds. During flood events, we check whether exceptionally high storage or discharge anomalies can be observed in the GRACE-based high-resolution TWS. The evaluation is based on correlation analysis of the temporal event dynamics to assess the consistency in the timing and magnitude of peaks. Furthermore, we assess which regional scales (down to <100,000 km²), flood types, and hydro-climatological zones yield the most prominent signals. We expect stronger TWS anomalies for rainfall-driven floods and flood events of larger magnitude. The results can contribute to improving global flood monitoring and flood early warning systems.