Approach to Determine Extent and Depth of Highway Flooding

Flooding and flash flooding pose serious infrastructure hazards to human populations in many parts of the world. During a flood, it is critical to identify road segments that are flooded so that rescue and response routes can be determined and rescue personnel and supplies can be distributed promptly. Presently there is not an information system that, given a specific flood level, can accurately predict flooded road segments and their depth and provide this information for rescue activities. This paper reports on a study to develop a prediction prototype to identify flooded road segments during a flood. It includes two tasks: determining flood extent and depth and identifying road segments that are flooded. Unlike a traditional approach, which determines flood extent and depth by comparing a water surface and a terrain surface, this study determines flood extent and depth using light detection and ranging data in a geographic information system. This flood extent and depth information is integrated with 3D road centerline data created in a related study. It is believed that the proposed prediction model and algorithms in this study can provide a practical and efficient approach to identify road segments that are flooded in a timely manner to help determine rescue routes. These models show promise. Additional research is required to fully validate the models.