Classification of Flood-Prone Areas in Padang City Using Landsat-9 Satellite Imagery

Flood-susceptibility mapping requires spatial information that captures the combined influence of hydrological conditions, topography, soils, and land cover. This study classified and mapped flood-prone areas in Padang City by integrating six parameters: rainfall, elevation, slope, distance from rivers, soil type, and land use/land cover. A quantitative spatial approach using Geographic Information Systems (GIS) was employed. Land cover was classified from Landsat-9 Collection 2 Level-2 imagery using the Classification and Regression Trees (CART) algorithm in Google Earth Engine. The six parameters were subsequently scored, weighted, and integrated through weighted-overlay analysis. The land-cover classification achieved an overall accuracy of 85.17%, while the overlay analysis delineated five flood-susceptibility classes. Moderate susceptibility occupied the largest area, covering 30,675.414 ha (45.79%), followed by high susceptibility at 17,631.140 ha (26.32%), low susceptibility at 17,226.524 ha (25.72%), very high susceptibility at 1,420.275 ha (2.12%), and very low susceptibility at 31.048 ha (0.05%). These findings reveal substantial spatial variation in flood susceptibility across Padang City, with moderate susceptibility predominating. The study provides an integrated spatial assessment to support disaster mitigation and environmentally responsive land-use planning.