Automatic Detection of Potential Dam Locations in Digital Terrain Models

Structural measures for retaining and distributing wateri.e., reservoirs, flood retention and power plantsplay a key role to protect and feed a growing world population in a rapidly changing climate. In this work, we introduce an automated method to detect potential reservoir or retention area locations in digital terrain models. In this context, a potential reservoir is a larger terrain form that can be turned into an actual reservoir by constructing a dam. Based on contour lines derived from terrain models, potential reservoirs are found within a predefined range of dam lengths, and the locally optimal ones are then extracted. Our method is to be applied in the very early stages of project planning and for area-wide potential analysis. Tests in a 100 km(2) study area bring promising results, but also show a certain sensitivity regarding terrain model quality and resolution. In total, 250-300 candidate polygons with a total volume of more than 6 million m(3) were found. In order to facilitate further processing, these are stored as a GIS vector dataset.