Aplicação de Índice de Vegetação para Análise de um Remanescente Florestal da Mata Atlântica

• Ano de defesa: 2022
• Autor(a) principal: Castelo Branco, Antonia Francivan Vieira
• Orientador(a): Zanella, Maria Elisa
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Mata Atlântica; NDVI; Regressão linear múltipla; Tamanho do efeito
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/67937

The Atlantic Forest was the first biome to be devastated in Brazil; however, despite being reduced to fragments, it is relevant in terms of biodiversity. The municipality of Viçosa do Ceará was chosen as the study area, as it is located in the extreme north of the distribution of the Atlantic Forest biome. This study aimed to identify whether there was a “forest transition” of the Atlantic Forest in a remnant located in the extreme north of this biome, from 1987 to 2019. The method consists of the following steps: (1) verification of the vegetation index most used in scientific publications, through bibliometric analysis; (2) application of the vegetation index to measure the size of the Atlantic Forest, using images from Landsats 5/TM, 7/ETM+ and 8/OLI; (3) multiple linear regression analysis to verify the factors that influence the extent of the forest area; (4) effect size statistical analysis to define the degree of impact of “rainfall” on the restoration of vegetation cover. The results showed that the Normalized Difference Vegetation Index (NDVI) has been the vegetation index most used in scientific publications; in general, over the 33 years of analysis, there was a 20% reduction in the forest area; however, due to the tendency towards regeneration, indicated by the data from the last five years, it was evidenced the occurrence of “forest transition” in the study area. The multiple linear regression analysis showed that 79% of the variation dynamics of the vegetation area covered by Atlantic Forest can be explained by the variables “rainfall volume (mm)” and “size of the estimated resident population”, and lower values of rainfall volume combined with the increase in the population tends to reduce the forest area; the positive impact of “rainfall” on forest regeneration ranged from “insignificant” to “very high”.