Zoning risk to the occurrence of landslides: an application on watershed of Matirumbide Stream, Juiz de Fora, MG

Abstract

In Brazil, there is a great influence of natural processes such as landslides, characteristic of morphologically hilly areas, in shaping the Earth's surface. In the populated urban areas, this process involves the generation of risk to the resident population in hillside areas. Thus, studies aimed at monitoring and forecasting these events become necessary, since that it would avoid further damages, material and human. There are some methodologies for forecasting landslides based on the use of mathematical and operationalized by GIS models. Thus, the present study aimed to analyze the spatial distribution of the risk the occurrence of landslides in the Watershed of Matirumbide Stream, in the city of Juiz de Fora, MG, using statistical mathematical prediction model SINMAP. For the representation of the risk could be taken was necessary to prepare the map of the use, occupation and land cover, as well as the probability map the occurrence of landslides. From the overlap of these two maps could be represented risk areas of the watershed, as well as the representation of the instability of areas from the relationship with the occupation of the land and the processes and dynamics of the Earth surface. In the final map, we find that 24% of the watershed presents high risk of landslides occurrence. The presence of these areas was mainly related to the morphology of the land, through the degree of slope, and the presence of low occupations infrastructure along the areas of default risk. The results show that part of the basin is characterized by the relation of social fragility linked to occupations in high-risk areas. Finally, SINMAP is a relevant model for the representation of auxiliary risk areas and subsequent analysis.