The aim of this study was to report physical characterization of the hydrographical basins of Igarapé Canela Fina and Igarapé Preto, both located in the municipality of Cruzeiro do Sul, Acre, Brazil. The ISODATA algorithm was used for the non-supervised classification, drawing on data provided by the TM Landsat 5 sensor from the National Institute for Space Research. The covered period of analysis was from 2005 to 2010. The analysis shows an average increase of 48.11% of deforested areas and a mean reduction of 50.58% in the forested areas for both basins, due to anthropic interventions over those five years. The analysis of physical characteristics shows similarities between the hydrographical basins, with both responding slowly to hydrological events, and thus providing conditions not favourable to flooding. Modifications conducted on the surface of the hydrographical basins along the last few years have, however, possibly altered the hydro dynamics of the rivers and streams. The preliminary results indicate that the Remote Sensing and Geographic Information Systems are efficient tools for the analysis of the space-time dynamics of vegetation and the morphometric characterization of hydrographical basins.
digital image classification; morphometric; map vectorisation
