A curated sentinel collection to study cascading droughts and extreme precipitation events

The intensification of the hydrological cycle as a consequence of climate change is altering the distribution and intensity of hydro-climatic extreme events. Extremes at both ends of the hydrological cycle,  dry events (droughts) and wet events (heavy rainfall), are increasing in frequency and intensity. When such events occur in cascades, their compounding impacts can increase in severity. However, datasets explicitly designed to study such cascades remain scarce.  Within the ESA-funded ARCEME project (Adaptation and Resilience to Climate Extremes and Multi-hazard Events), we introduce a dataset tailored to study cascading droughts and heavy precipitation events. The events are identified using a dual sampling strategy relying on climatological anomaly detection, and on sampling from the footprints of disaster events reported by the Emergency Database (EM-DAT). The resulting dataset provides over 400 georeferenced datacubes with a spatial extent of 10 by 10 km, each covering one year before and one year after the cascading event, and sampling a wide range of climate zones and terrestrial biomes. Each datacube integrates (i) Sentinel-2 L2A optical imagery, (ii) Sentinel-1 radiometric Terrain Correction (RTC) radar data, (iii) Ancillary Landcover and topographic information. Optimized for the analysis of impacts on natural and managed vegetation, the dataset provides a standardized data collection suitable for data-driven studies of compound and cascading hydro-climatic extremes.