Os sistemas de Informações Geográficas são atualmente uma das alternativas mais eficientes para o armazenamento, manipulação, análise e monitoramento de informações ambientais referenciadas geograficamente. Bacias hidrográficas constituem unidades naturais para a análise destas informações ambientais, pois apresentam características próprias as quais são utilizadas para estudar os efeitos antrópicos sobre os ecossistemas. Este trabalho tem por objetivo a caracterização da Bacia Hidrográfica do Rio Boa Esperança, Reserva Biológica do Tinguá - RJ, através da utilização de um Sistema de Informações Geográficas, o OSUMAP. Para tal foram utilizadas: variáveis topográficas, morfométricas, satelitárias e qualidade d'água, que cruzadas resultaram no plano de informações de susceptibilidade à Erosão. No sentido de analisar o comportamento das variáveis traçados variogramas. O sistema de informações utilizado, aliado a outros programas de tratamento e manipulação de dados, foi suficiente para o nível proposto, alcançando os objetivos esperados / The Geographical Informations Systems (GIS's) are nowadays one of the most efficient altematives for the retrieval, manipulation and analysis of georeferenciated environmental informations. Watersheds constitute natural units to the analysis of these environmental informations, because they have own characteristics that are used to study the effects produced by the man over the ecosystems. This dissertation has as its object the characterization of the Boa Esperança watershed, at the Biological Reserve of Tinguá in Rio de Janeiro, throught the use of a Geographical Informations System, the OSU-MAP. To do so, were used: topographic variables (gradient, slope, aspect and convexity) and morphometric ones (drainage density, river density, texture ratio, geological texture and form ratio), that were got from a Digital Elevation Model (DEM); remote sensing variables (reflectance), got from the bands 2, 4 and 5 ofthe Landsat-TM, tha were used to make an unsupervised classificatio by Principal Components Analysis. All these variables were crossed to obtain anew layer, the Erosion Susceptibility. And this layer was crossed with water quality layers (total sediment, conductivity and pH) that were got from analysis of the hidric material collected along the basin principal drainages, in order to averiguate the proposed model. Variograms in two directions, for all variables, were also plotted in order to analyse variables behavior in the basin and to verify the suitability of the geometric resolution (50m x 50m) to the scale (1 :50000). The GIS used was suitable to the proposed manipulation, despite of its robustless, and the chosen variables reach their finallity in composing the Erosion Susceptibility layer, that were verified throught the water quality layers. The band 2 ofthe Landsat-TM was not significantly in the proposed reflectance model, and so, it didn't allowed a better different land uses discrimination at the basin.The variograms form showed that the geometric resolution esed was suitable to the scale, dispite some variables anisotropy. The Geographical Informations System, theOSU -MAP, with other data handling and manipulation softwares, was enough on the proposed leveI and got its expected objectives