Multi-Temporal Sentinel-1 SAR and Sentinel-2 MSI Data for Flood Mapping and Damage Assessment in Mozambique

Floods are one of the most frequent natural disasters worldwide. Although vulnerability varies from region to region, all countries susceptible flooding. Mozambique was hit by several cyclones in last few decades, and 2019, after Idai Kenneth, country became first southern Africa be two same raining season. Aiming provide local authorities with tools yield better responses before any disaster event, mitigate impact support decision making for sustainable development, it is fundamental continue investigating reliable methods management. In this paper, we propose a fully automated method flood mapping near real-time utilizing multi-temporal Sentinel-1 Synthetic Aperture Radar (SAR) data acquired Beira municipality Macomia district. The procedure exploits processing capability Google Earth Engine (GEE) platform. We map flooded areas finding differences images flooding then use Otsu’s thresholding automatically extract area difference image. To validate compute accuracy proposed technique, compare our results Copernicus Emergency Management Service (Copernicus EMS) available study areas. Furthermore, investigated Sentinel-2 multi-spectral instrument (MSI) produce land cover (LC) estimate percentage each LC class. show that combination SAR MSI damage assessment. mapped an overall about 87–88% kappa 0.73–0.75 directly comparing prediction EMS maps. classification validated randomly collecting over 600 points LC, 90–95% 0.80–0.94.