Lógica Fuzzy na identificação de lotes associados ao risco de inundação no município de Nova Lima/MG

Abstract

Remote Sensing (SR) is one of the usual practices in Earth sciences with applications in urban areas. The data obtained with such process, integrated with a Geographic Information System (GIS) allows identifying, reducing and restructuring different problems in city planning. Flooding, as a natural phenomenon of water overflow, resulting from the concentration of excessive rainfall, damages material goods, interferes in human lives and in the structuring bases of urbanization. The present work proposes to develop a multicriteria analysis model to support the fuzzy type decision with evidence weights allied to the Analytic Hierarchy Process (AHP) and Remote Sensing tools with support of geoprocessing to determine the risks to the flood in the city of Nova Lima / MG, which is located in the metropolitan area of Belo Horizonte. In order to achieve the research, there was a preparation of input maps, like the digital model of elevation, slope, land use and occupation, permanent hydrographic preservation area, climatic isoietas, geological and soil types. These input maps were submitted to the fuzzy logic by weights of evidence, making it possible to ponder suitable and unsuitable regions for flooding, considering each variable and producing individual maps weighted from 0 to 1. Allied to this technique, the method of Analytic Hierarchy Process was used for the multiplication of the maps that were submitted to fuzzy logic by the use of experts notes and literature research. From this application, resulted a critical map with the levels of flood susceptibility in the city. A validation map was previously constructed in which eighty-six percent (86%) of correspondence was reached in the flood spots. The validation data was generated by crossing the information processed about flooding of the Secretariat of State for Environment and Sustainable Development (SEMAD), the Company Mineral Resource Research (CPRM) and local news. In order to point out the urban vulnerability of the lots susceptible to flooding, a new crossing was performed and a map categorizing flood risk spots was obtained for all of the lots in the studied city. Flood-spot analysis based on fuzzy methodology and on AHP proved to be effective, easy to execute, with simple implementation and low financial costs. It is important to point out that this way of identifying areas which need public attention, as well as investments in protection, prevention and mitigation actions, has proved useful for managing risk and disasters in of flood cases.