Trajetória e contexto espacial dos desmatamentos no bioma cerrado

• Ano de defesa: 2014
• Autor(a) principal: Rocha, Genival Fernandes
• Orientador(a): Ferreira Junior, Laerte Guimarães
• Banca de defesa: Ferreira Junior, Laerte Guimarães, Cherem, Luis Felipe Soares, Oliveira, Ivanilton José de, Brito, Jorge Luís Silva, Carvalho Júnior, Osmar Abílio de
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Goiás
• Programa de Pós-Graduação: Programa de Pós-graduação em Geografia (IESA)
• Departamento: Instituto de Estudos Socioambientais - IESA (RG)
• País: Brasil
• Palavras-chave em Português: Cerrado; Detecção de desmatamentos; Mudanças na cobertura e uso da terra; SIAD; PMDBBS
• Palavras-chave em Inglês: Cerrado; Deforestation monitoring; Land-cover and land-use changes.
• Área do conhecimento CNPq: CIENCIAS HUMANAS::GEOGRAFIA
• Link de acesso: http://repositorio.bc.ufg.br/tede/handle/tede/4986

The Cerrado biome, occupying approximately 25% of the Brazilian territory, is distinguished for its rich biodiversity and for being a major water producer for the the most important South American basins (e.g. Amazonian, São Francisco, Tocantins-Araguaia and Paraná). With vast flat areas, it is a major industrial food producer and the main agricultural frontier in the country. On the other hand, the large-scale land conversion made the Cerrado one of the 34 biodiversity hotspots in the world, i.e. regions under severe environmental pression due to the reduction and fragmentation of the natural landscapes. The remote sensing monitoring of the Cerrado is very recent, with only two systems currently in operation: the warning deforestation system (SIAD Cerrado), developed by the Image Processing and GIS Lab of the Federal University of Goiás (LAPIG – UFG) and the Satellite Deforestation Monitoring of the Brazilian Biomes Project (PMDBBS), an initiative of the Brazilian Institute of the Environment and Renewable Natural Resources (MMA – IBAMA). Specifically, this doctoral thesis pursued the following goals: 1) comparative evaluation of these two monitoring initiatives; 2) analysis of ten years of SIAD deforestation data, the only Cerrado deforestation time-series available; 3) assessment of possible alternatives aiming at the improvement of SIAD-like semi-automated monitoring systems. Our results demonstrate that both the SIAD and PMDBBS deforestation spatial distribution patterns are very similar and complementary. The analysis of SIAD data for the 2002 – 2012 period indicates that the remnant decreased from 61.2% to 58.5% relatively to the biome area (~ 2 million km2), with about 80% of the total cleared area constrained to terrains with mild slopes (< 3%) and concentrated in only 100 of the 1,384 municipalities. It is worth mentioning that the states with highest deforestation rates are also the ones with the largest native vegetation remnants, suggesting that key agricultural frontiers are still very active. Potential improvements for making systems like SIAD more robust and reliable include the use of time-series and spatial heuristics criteria for filtering the automated generated deforestation warnings, which demand very time-consuming visual inspection approaches. We estimate that about 506,906 km2 of remnant vegetative cover are highly vulnerable and under risk of being converted over the years. Thus, improving the current systems and monitoring models, in order to make them more reliable and capable of operating at multiple scales of observation, is fundamental for the effective territorial governance of the Cerrado biome.