Flood Inundation Mapping Using Earth Observation Data in the Po River (North of Italy)

Floods have caused severe consequences on the environment in different areas of the world and also in Italy. Therefore, this study explores the opportunity of combining the Earth Observation data (both radar and optical images) for flood mapping in the Po River, located in the north part of Italy. Based on a multi-temporal approach, synthetic aperture radar Sentinel 1 images between August and November 2016 were used in the period. The methodology adopted is semi-automatic method using GEE to detect a flooded area. We selected images before and after flood event using VH band flood detection. It was concluded that the use of synthetic aperture radar (SAR) data produces results with low accuracy especially in building areas due to the spectral signatures of urban features. To deal with this, we used an optical image (Landsat 8 OLI) based on automatic threshold using Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI) for change detection to accentuate changes likely caused by flooding. The results were calibrated by hydraulic model using HEC-RAS software. Thus, the approach developed in this study could be a crucial step toward adequate flood risk mitigation and management strategies.