Fuzzy logic applied to forest fire risk modeling in the Cajamarca region, Peru

• Ano de defesa: 2021
• Autor(a) principal: Anticona, Alex Joel Vergara
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: spa
• Instituição de defesa: Universidade Federal de Viçosa
• Palavras-chave em Português: Engenharia geotécnica; Análise espacial (Estatística); Processo decisório por critério múltiplo; Fotointerpretação Florestal
• Link de acesso: https://locus.ufv.br//handle/123456789/28256

Forest fires have become more frequent due to anthropogenic activities and climate change. Besides being also a natural phenomenon in all plant ecosystems of the world, it contributes to reduce forest areas when occurs in tropical, boreal and temperate vegetation. The objective of this study was apply Fuzzy logic as an alternative to multicriteria analysis, in order to model forest fire risk in Cajamarca region, Peru. Eight variables have been incorporated to represented biological, topographic, socioeconomic and meteorological factors. Necessary methodological steps for this study were as follows: a) Database acquisition, editing and rasterization, b) Application of Fuzzy membership functions and images fuzzification, c) Fuzzy overlay and d) Spatial reclassification of forest fire risk. According to results, 71.68% of the area is under very low to medium forest fire risk. However, 28.32% of the study area has high and very high fire risk, which makes fire occurrence susceptible to lack of rain and water in the soil. It was found that biological, topographic and socioeconomic factors with their respective variables were directly influenced by the meteorological factor variables, which were temperature, rainfall and water availability. The proposed methodology integrates geotechnology and artificial intelligence to model the complex interactions between vegetation, topography and climate, as well as social, economic and anthropic activities, in order to map vulnerable areas to forest fires. Fuzzy logic provided flexibility to model forest fire risk in Cajamarca region, Peru. By elucidating and mapping fire risk, this approach can provide great environment, economic and social benefits through initiatives that mitigate fire environmental impacts, which improves income, life quality and local population GDP. This methodology can be applied to other areas around the world to provide information about the risk of forest fires. Keywords: Geotechnology. Spatial analysis. Multi-criteria analysis. Membership functions.