View article

Flash floods are dangerous and unpredictable. This study aimed to improve flash flood prediction in Algeria’s Hodna Basin using advanced AI models and GIS (GeoAI). Each watershed exhibits unique characteristics that contribute to flooding, primarily driven by hydrological and topographic factors. To capture and incorporate these distinctive attributes, a wide range of data sources were integrated, including topographic features, hydrological parameters, and remote sensing data. These data encompassed slope, rainfall, aspect, elevation, land use/land cover (LULC), topographic wetness index, distance from rivers, stream power index, curvature, hill shade, and geology. These diverse factors served as input variables for the present models. The data sources employed were Landsat 8, Sentinel-2 imagery, climate hazards group infrared precipitation with station data (CHIRPS) data and USGS data, which were …