Risco integrado de incêndio florestal em áreas de cerrado : contribuições metodológicas

• Ano de defesa: 2016
• Autor(a) principal: Prudente, Tatiana Diniz
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; BR; Programa de Pós-graduação em Geografia; Ciências Humanas; UFU
• Palavras-chave em Português: Cerrados - Aspectos ambientais; Mapeamento florestal; Incêndios florestais - Prevenção e controle; Cerrado; Risco de incêndio florestal; Sensoriamento remoto; Forest fire risk; Remote sensing; CNPQ::CIENCIAS HUMANAS::GEOGRAFIA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/16019; https://doi.org/10.14393/ufu.te.2016.45

The Cerrado has climatic conditions that offer extremely favorable conditions to successive occurrences of fires, which cause serious environmental, social and economic losses. Therefore, this research had as objective to develop a methodology for modeling of forest fire risk in Cerrado areas, from a structural index and other dynamic thereby foster the development of a monthly mapping of integrated risk of fire. For the development and application of the methodology was chosen as the study area the municipality of São Valério, located in the Mesoregion of Eastern Tocantins. This choice was made due to the presence of several phytophysiognomies of Cerrado, large areas of pasture and also recurrent occurrences of fires. It's necessary to highlight that all data used in the proposed methodology are from remote sensing. The structural fire hazard is based on the combination of factors that don't vary in a short period of time, being considered, in this study, the following factors: distance from roads and urban areas, land use and native vegetation cover, hypsometry, slope and orientation from the slopes. The dynamic risk has as an objective detecting changes in the variables capable of causing changes in the flammability of forest fuels, being raised, in this research, the factors: precipitation, water deficit and NDVI. Was generated, for each factor a fire susceptibility map, with classes ranging from high, moderate and low. It was possible to consider as high susceptibility to fire those areas which showed the following conditions: near roads and urban areas; occupied by the categories of landuse and vegetation cover Park Savanna, Grassy-Woody Savanna and Cultivated Pasture; altitudes up to 600 m; slopes over 40%; orientation of slopes facing north, northeast and northwest; monthly precipitation less than 25 mm; monthly water deficit over 80 mm; and NDVI values between 0.00 to 0.25. The fire susceptibility maps were associated from a multi-criteria analysis in a GIS environment, using a weighted linear combination technique, thus generating the structural risk maps and dynamic forest fire risk. Then, the structural and dynamic risks were added, generating monthly maps of integrated fire risk. The results showed that the proposed methodology enables a spatial and temporal analysis, which in turn facilitates the development and formulation of strategic prevention actions in the critical areas, undoubtedly providing greater effectiveness of the measures taken and optimizing material resources to such intervention.