ÍNDICES DE VEGETAÇÃO OBTIDOS DE IMAGENS MULTIESPECTRAIS APLICADOS NA DISCRIMINAÇÃO DA COBERTURA VEGETAL EM RELEVO MONTANHOSO / VEGETATION RATES OBTAINED FROM MULTISPECTRIC IMAGES APPLIED IN THE DISCRIMINATION OF VEGETABLE COVERAGE IN MOUNTAIN RELIEF

• Autor(a) principal: Francisco, Cristiane Nunes
• Data de Publicação: 2018
• Outros Autores: Gruber, Nina Cardoso
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Geo UERJ
• Texto Completo: https://www.e-publicacoes.uerj.br/geouerj/article/view/16931
• topic: normalized difference vegetation index (NDVI); simple ratio (SR); vegetation index; remote sensing; LANDSAT8/OLI; índice de vegetação da diferença normalizada (IVDN); razão simples (RS); índices de vegetação; sensoriamento remoto

The vegetation indices are used to investigate the properties of spectral vegetation and can be related to the biophysics parameters. In addition, they can also be used to minimize illumination effects caused by the topographic on spectral response that difficult to distinguish vegetation physiognomy. To evaluate the efficiency of vegetation indices to distinguish the vegetation cover in mountainous area, it was calculated two indices from LANDSAT8/OLI multispectral bands that cover the Tijuca Massif, situated on central zone of Rio de Janeiro city. The results revealed that the both vegetation indices (NDVI and SR) had efficiency to separate water bodies and to discriminate vegetation and non-vegetation classes. However, the urban area, shadows and rocks were mixed, while it was not possible to classify the vegetation physiognomy because of the effect solar illumination on the vegetation indices. The forest located at southern slopes, which received less solar illumination at the time the orbital sensor collected the data, had indices similar to herbaceous vegetation, and in this way it was not possible to separate them. While the forest located at northern and western, which were received more solar illumination, had the most elevated indices and it was not confused with other types of land cover. In conclusion, the indices vegetation (IVDN and SR) were not efficient to minimize the topographic effects to discriminate the vegetation physiognomy. On other hand, the indices showed efficient to demonstrate that watersheds located at protected area had more elevated indices than the basin located near de urban area.