Serviços Personalizados
Journal
Artigo
Indicadores
Citado por SciELO
Acessos
Links relacionados
Similares em
SciELO
Compartilhar
Madera y bosques
versão On-line ISSN 2448-7597versão impressa ISSN 1405-0471
Resumo
PEREA-ARDILA, Mauricio A. e MURILLO-SANDOVAL, Paulo J.. Comparison of vegetation indices based on satellite images for the identification of mangrove forests in Colombia. Madera bosques [online]. 2024, vol.30, n.spe, e3042600. Epub 11-Nov-2024. ISSN 2448-7597. https://doi.org/10.21829/myb.2024.3042600.
The use of mangrove-specific vegetation indices (VI) expands the possibility of improving practical mangrove detection. The objective of this study is to compare seven mangrove-specific VIs derived from satellite images to determine their agreement with mangrove classification using the WorldCover 2020 dataset at two mangrove sites in the Colombian Pacific. Four Sentinel-2 images were used, and through oceanographic data, their tidal state was validated in order to show if tidal changes affect mangrove detection. The images were processed in BOA reflectance, the VIs were calculated, and their accuracy was evaluated to identify mangrove/non-mangrove. Then, mangrove points were identified on the WorldCover dataset for 2020 to determine the accuracy of each VI. The average mangrove spectral signature for each tidal state (between 0.23 and 0.25 reflectance units) was identified, and descriptive statistics were determined for each VI. It was observed that the NDMI and MMRI obtained presented a considerable agreement in Chocó for the low tide and an acceptable agreement in Nariño for the high tide, and their spatial extent and the difference in their extent were mapped. The results demonstrate the performance of specific VIs for mangroves in Colombia as a practical alternative for rapid mapping of their extent that contributes to the integrated management of the ecosystem.
Palavras-chave : tidal state; practical monitoring; Colombian Pacific; remote sensing; Sentinel-2.












