STURM-Flood (STURM-Flood: a curated dataset for deep learning-based flood extent mapping leveraging Sentinel-1 and Sentinel-2 imagery)

The repository hosts the STURM-Flood dataset, an open-access resource designed for flood extent mapping using Sentinel-1 and Sentinel-2 satellite imagery. The dataset comprises 21,602 Sentinel-1 tiles and 2,675 Sentinel-2 tiles, each of size 128 × 128 pixels at a resolution of 10 meters, along with corresponding water masks covering 60 flood events globally. This curated dataset is optimized for deep learning applications and provides ground-truth data from the Copernicus Emergency Management Service to facilitate robust model development. We invite researchers and developers to utilize this resource for advancing flood mapping techniques in disaster management. For further details on the methodology, results, and implementation, please refer to our study published in Big Earth Data (2096-4471): https://doi.org/10.1080/20964471.2025.2458714 .