Mapping methodological analysis of wildland-urban interface for wildfires in south of Brazil

Abstract

Wildfires cause several damages to flora, fauna, soil, atmospheric air and human health. The places where there is a concentration of forest fuels in contact or very close to the homes represent a great risk, because the human presence can contribute to the start of fires and can also be drastically affected in the event of a fire. These environments are called wildland-urban interface, represented by places where human beings and anthropic structures meet or mix with vegetation. There are different methodologies for the classification of the wildland-urban interface, aiming to serve as a subsidy for preventive and vegetation fire-fighting policies. In Brazil, discussions about the theme are incipient. Campina Grande do Sul and Quatro Barras are two Brazilian municipalities with an extensive vegetative area with social interaction (areas of wildland-urban interface) and, in addition, suffer from the presence of wildfires. From 2011 to 2020, a total of 797 occurrences were registrated in this area. The objective of the present study was to delimit the wildland-urban interface for Campina Grande do Sul and Quatro Barras municipalities, located in the state of Paraná, Brazil, relating to the occurrence of wildfires. To achieve the proposed objective, it was necessary to use a methodology for classifying wildland-urban interface areas, already established globally, based on the density of households and forest cover, dividing into intermix and interface zones. Two methodologies from the Brazilian Institute of Geography and Statistics were used to obtain the density of households (statistical grid and census sectors). The statistical grid presents fixed cells of 0.04 km² for urban areas and 1 km² for rural areas, while the census sectors have variable areas. The results showed that when using the statistical grid, Campina Grande do Sul and Quatro Barras had coverage of 12.5 and 15.2% of the municipal areas classified as wildland-urban interface, respectively. In the use of census sectors, Campina Grande do Sul presented 25.7% and Quatro Barras 13.6% of the areas as a wildland-urban interface. This discrepancy observed when using the census sector may be related to an overestimation of household densities due to the variable size of the census sectors. When distributing the 314 geolocation of fires that occurred in the municipalities from 2011 to 2016, it was observed that when using the statistical grid, 15.3% of these were present in areas of wildland-urban interface and using the census sectors, 17.2 % of occurrences located in these areas. Based on the results obtained, it can be concluded that the delimitation of the wildland-urban interface using the method of household information contained in the statistical grid presented greater detail than use of the census sectors. However, it is necessary to evaluate the methodology used to quantify coverage forestry. The presence of fires in areas with a wildland-urban interface demonstrates the need for preventive policies to be applied to the habitants of these places. Finally, it is recommended to continue studies, as well as to create or adapt methodologies for classifying the wildland-urban interface based on local characteristics.