A Cloud-Native Python GIS Framework for Flood Susceptibility and Critical Facility Exposure Assessment: Evidence from Miami, Florida

Urban coastal cities increasingly confront compounded flood hazards driven by sea-level rise, intense precipitation, and dense impervious surfaces. This study evaluates a cloud-native Python GIS framework for flood susceptibility mapping and critical facility exposure analysis in the City of Miami, Florida, being one of the most flood-exposed coastal cities in the United States. Implemented entirely within a Google Colab cloud native environment, the workflow integrates three open-source spatial indicators: (i) terrain elevation retrieved via the py3dep interface to the USGS 3D Elevation Programme at 10 m resolution; (ii) Euclidean proximity to water bodies extracted from OpenStreetMap (OSM) using OSMnx; and (iii) building footprint density as a proxy for impervious surface cover, also sourced from OSM. These raster-based indicators were standardised, weighted using a Multi-Criteria Decision Analysis (MCDA) framework (water proximity: 0.40; elevation: 0.35; building density: 0.25), and combined via weighted overlay to produce a continuous flood risk index. The index was classified into low, medium, and high susceptibility zones using quantile thresholds at the 33rd and 66th percentiles. Results show that high-susceptibility areas cover 48.66 km² (34.0%) of the city, concentrated along coastal waterfronts and inland water corridors. Exposure analysis reveals that 9 of 16 hospitals (56.2%), 61 of 244 schools (25.0%), and 5 of 17 fire stations (29.4%) are situated in high-susceptibility zones. The framework is fully reproducible, cost effective, low hardware requirement and transferable decision-support methodology.