PERFORMANCE ANALYSIS OF DIGITAL ELEVATION MODELS - A CASE STUDY OF BASIN LOW MIDDLE SÃO FRANCISCO RIVER

Corresponding author: madson_tavares@hotmail.comThe performance assessment of digital elevation models, from the handling of the mission SRTM elevation data , ASTER imager instrument , and project TOPODATA compared to elevation data obtained in the field from GPS was the aim of this work, in view the adequacy of the best model for use in studies of morphometric analysis of watersheds , as well as applications in hydrological models . To conduct this study we used the ArcGIS 9.3 software. The results demonstrate that the elevation data extracted from SRTM , ASTER and TOPODATA are extremely satisfactory for presenting high correlation with the data collected by GPS (r = 0.97 , 0.96 and 0.97 , respectively) , was also observed that the smallest errors associated were obtained for the SRTM DEM , being observed PBIAS equal to 2.71 % and MAE of 11.45 m . Finally from the index c Camargo & Sentelhas be excellent in the criterion of model performance. In conclusion it was observed that the DEM SRTM showed the best results, possibly due to the lower number of cells to compose your image, thus generating more homogeneous areas than other models.