Abstract. This study aimed to identify risk areas for landslides and flooding in the urban area of Viçosa, Minas Gerais. Remote Sensing and Geographic Information Systems (GIS) were used as methodological tool and it had its basement on satellite and radar images from the RapidEye and TOPODATA/SRTM sensors, respectively. Image analysis showed, through the texture and tone/color, different forms of occupation intensities of the urban area, which was defined from the limits of the IBGE census sectors. Declivity maps and Elevation Model were made based on the local relief steepness which confirmed the predominance of more embedded valleys and steep areas. The Land Use and Cover map analyzed the urban occupation form identifying classes and allowing them to compare the definition of the occupation type and the circumscription of the risk areas. It was evidenced, therefore, that risk areas for landslides are those with higher slope and anthropic interference on the slopes while the floods occur near the confluences of rivers and due to the large volume of precipitation in areas already historically consolidated by the occupation process. In this sense, this research, in addition to identifying the areas of greatest weakness risks and confirm the importance of using contributes geotechnology also provides technical aid to performance measures and adequate prevention of governing bodies in these environments. Palavras-chave: geotechnology, geoprocessing, image processing, natural disasters, geotecnologias, geoprocessamento, processamento de imagens, desastres naturais.