Geraldo Jose Lucatelli Doria de Araujo Jr.
1997Remote sensing and geoprocessing techniques were used in running two soil erosion predictive models (USLE Wischmeier and Smith, 1978 and MUSLE - Williams, 1975). The research area in the surroundings of Barra Bonita, Santa Maria da Serra, SP. Several input parameters were integrated in a Geographic Information Systems (GIS, SIG/INPE) according to each model logic. To provide the 1994 soil use and occupation parameter was digitalized from Castro, 1992. Field research was realized to obtain the soil/erodibility parameter. Paper topographic charts were used to obtain the topographic factor parameter Erosion Potencial (PE), Natural Erosion Potencial (PNE), Sediments apport (Y) and the Criticity Index (IC) were obtained as the models output to both 1988 and 1994 situations. Through the outputs (PE, PNE, Y and IC) comparation, between 1988 and 1994, it was found that there is greater degradation expectative in 1994 than in 1988. The main reason is the event of the more intense antropic activity. The GIS as model support and the use of remote sensing techniques in input parameters providing, proved to be an operationally efficient approach.