Identificação de áreas de floresta estacional semidecidual e de cerradão por meio de índices de vegetação a partir de imagens de satélite

• Autor(a) principal: Silva, Silvana Barros da
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/18091
• topic: Sensoriamento remoto; NDVI; S2REP; Savana; Sentinel 2; Remote sensing; CIENCIAS AGRARIAS::RECURSOS FLORESTAIS E ENGENHARIA FLORESTAL; CIENCIAS EXATAS E DA TERRA::GEOCIENCIAS; CIENCIAS EXATAS E DA TERRA::GEOCIENCIAS::GEOFISICA::SENSORIAMENTO REMOTO

Brazilian forests have suffered vertiginous reductions over the years, leading phytophysiognomies such as the Atlantic Forest and the Cerrado to be considered priority areas for conservation, however, there is a need to monitor and identify these areas for their effective protection, which is not always easy. Although remote sensing studies are widely used to monitor the state of forest remnants, there are still difficulties in identifying the different phytophysiognomies through satellite monitoring, mainly in phytophysiognomies with floristic similarity Thus, the present work sought to analyze phytophysiognomies associated with the Cerrado and the Atlantic Forest, based on Sentinel-2 satellite images, verifying whether the NDVI (Normalized Difference Vegetation Index) and S2REP (Sentinel 2 Red Edge Position) are viable indices to distinguish the phytophysiognomies of Cerradão and Semideciduous Seasonal Forest. This work has been carried in six locations, twelve of which are forest remnants, in the State of São Paulo. The formations are adjacent areas of Cerradão and the Semideciduous Seasonal Forest. Data were analyzed in two distinct seasons, summer and winter. Descriptive analyses and the Analysis of Variance (ANOVA) test were performed. The results indicated that the NDVI and S2REP can identify significant differences between the Cerradão and FES formations, making it possible to determine if nearby areas are of the same phytophysiognomy. Also, the effect of seasonality on the values of the indexes analyzed was identified, with a greater influence of the seasons than the phytophysiognomy. It was concluded that remote sensing is a complementary tool to the floristic and phytosociological survey in the classification of phytophysiognomies.