Detalhamento de áreas de savana arborizada no bioma Cerrado a partir da análise de séries temporais MODIS EVI para o período de 2004 a 2008

• Autor(a) principal: PONTES, Marlon Nemayer Celestino de
• Data de Publicação: 2010
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFG
• Texto Completo: http://repositorio.bc.ufg.br/tede/handle/tde/1855
• topic: Bioma Cerrado; PROBIO; Savana Arborizada; detalhamento fisionômico; MODIS EVI; contraste sazonal; Cerrado biome; PROBIO; Arboreous Savanna; Phytophisiognomic detailing; MODIS EVI; seasonal contrast; CNPQ::CIENCIAS HUMANAS::GEOGRAFIA

Land cover and land use maps are essentials for the effective territorial governance, environmental monitoring, and proper understanding of the structure and functioning of the ecosystems. In relation to the Cerrado biome, an important step in this direction was obtained with the PROBIO mapping (Projeto de Conservação e Utilização Sustentável da Diversidade Biológica Brasileira do Ministério do Meio Ambiente), which, based on the interpretation of high spatial resolution imagery (Landsat TM), acquired in 2001 and 2002, mapped the entire biome at the scale of 1:250.000, accorging to the 30 natural and 5 anthropic classes. Although this mapping allowed to know, at high accuracy and precision, the extension and distribution of the major land cover types, its updating and further detailing are necessary. A particular example of such need is the Arboreous Savanna class, which, according to the PROBIO map, occupies an area of about 415.642,58 km² (33,72% of all Cerrado remnant vegetation) and presents an marked variability, 20 to 70% in its arborescent layer. Assuming that the phytophisiognomic variations within this class yield distinct seasonal patterns, in this study we evaluated the potential of the MODIS EVI (enhanced vegetation index) imagery, enhanced in the temporal domain, to futher discriminate among this class sub-types. Based on seasonal contrast images of May and September, it was possible to identify three sub-classes, whose spatial distribution patterns corresponded to the major seasonal domains. On the other hand, and based on the primary productivity concept, it was possible to distinguish five domains, in which (large productivity values were associated to the occurrence of denser typologies, close to ecotones). Our results suggest the use of MODIS or similar images (as the ones to be provided soon to be launched VIIRS sensor onboard the NPP and NPOESS series) for improved differentiation of the Cerrado physiognomies. However, field validation is necessary in order to better understand the biophysical meaning of the intraclass physiognomies identified, as well as a better understanding of the inter-annual patterns are necessary.