Metodologia computacional para ajuste de talvegue usando modelo digital de elevação para prospecção de potenciais hidrelétricos

Abstract

Hydropower generation projects are generally expensive, highly complex projects that require large investments and take years to materialize. The use of Digital Elevation Models (DEMs) in hydroelectric prospecting studies brings several advantages to the entrepreneur. However, for hydrological analyzes to be based on DEMs it is necessary to perform topological corrections on these, in order to eliminate depressions and other interferences, making them hydrologically consistent. In this work the results of two methods of depressions correction in DEMs are contrasted, a filling one (Fill Depressions) and a rupture one (Breach Depressions). The objective was to develop an algorithm (StretchStream) capable of adjusting these results on the real planimetric layout of the basin's watercourse, since the two methods generate different representations of the same thalweg. When applying this correction the thalwegs extensions derived from the DEM treated by the two methods become the same as that of a reference trace, allowing the comparison, in scale, of the corresponding vertical profiles generated. A case study was carried out for the upper section of the Sapucaí River, a tributary of the left bank of the Rio Grande, which has a hydroelectric inventory approved by the National Electric Energy Agency (ANEEL). This peace of river was selected because it was the object of aerial surveying by laser tillering (Lidar) and also because it had a high resolution orthophoto mosaic (0.30m per pixel). The work concludes out that the use of the breach method for pretreatment of DEM of rivers in mountainous regions, presents results significantly closer to the reference values than the other method, which is quite widespread and used in a transparent way for the end user in several commercial GIS programs. It is also concluded that the critical analysis and the use of the breaching depressions method contribute to the greater assertiveness of the initial hydroelectric prospective studies, reducing costs and development times for new potentials for the entrepreneur. The algorithm developed in the work was compiled into a plugin for GIS QGIS, called StretchStream, made available to the entire GIS community for use and future improvements.