Assessing road-watercourse crossing overtopping potential using GIS and remote sensing: a large-scale screening approach

Road-watercourse crossings (RWCs), such as bridges, are critical components of transportnetworks. Flooding poses a major threat to these structures, with overtopping leading tosafety hazards and traffic disruptions. This study presents a large-scale screening methodusing GIS techniques and remotely sensed data to assess the overtopping potential ofRWCs. The approach integrates road and hydrographic networks with high-resolutionLiDAR-derived DEMs of bare terrain (DTM) and surface (DSM). RWCs create a con-striction of the river runoff cross-section, acting as bottlenecks. The method assumes thatRWCs with a height difference between the road level (DSM) and river thalweg (DTM)lower than the corresponding cross-section height are more prone to overtopping. RWCswere identified, and their remotely sensed height was calculated by extracting elevationdifferences (DSM-DTM). Field measurements were conducted to validate the remotelysensed values. A terrain ruggedness index was used to filter noise in the DSM, assumingroads are the smoothest surfaces. Riverbanks were identified using the raster-based Unsu-pervised approach, and their height from the thalweg was assessed. The method is appliedto the Magra River Basin in Italy (970 km²), a flood-prone area. Results showed that forwatercourses with Strahler order < 4, the median error between remotely sensed and fieldmeasurements height was high (1.2 m, 35%), while for those with a higher order the er-ror was significantly lower (0.5 m, 8%). Among 230 identified bridges, ~ 25% exhibited ahigh overtopping potential. This approach enables the prioritization of bridges for furtherhydrologic-hydraulic and traffic disruption modeling, enhancing infrastructure resilienceand flood risk management.