Mapping of corn phenological stages using NDVI from OLI and MODIS sensors

• Autor(a) principal: Venancio, Luan Peroni
• Data de Publicação: 2020
• Outros Autores: Filgueiras, Roberto, Cunha, Fernando França da, Silva, Francisco Charles dos Santos, Santos, Robson Argolo dos, Mantovani, Everardo Chartuni
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Semina. Ciências Agrárias (Online)
• Texto Completo: https://ojs.uel.br/revistas/uel/index.php/semagrarias/article/view/36641
• topic: Vegetation index; Satellite images; Remote sensing; Zea mays L.; Índice de vegetação; Imagens de satélite; Sensoriamento remoto; Zea mays L.

Crop phenology knowledge is relevant to a series of actions related to its management and can be accessed through vegetation indexes. Thus, this study aimed to evaluate the use of the Normalized Difference Vegetation Index (NDVI), from images of OLI and MODIS sensors, to obtain phenological information from corn crops. To this end, we evaluated two corn cropping areas, irrigated by a central pivot, and located western Bahia state, Brazil. These areas were managed with high technology and had no record of biotic and abiotic stresses. NDVI showed a well-defined temporal pattern throughout the corn cycle, with a rapid increase at the beginning, stabilization at intermediate stages, and decreases at the end of the cycle. Excellent fits for polynomial equations were obtained to estimate NDVI as a function of days after sowing (DAS), with R² values of 0.96 and 0.95 for images of OLI and MODIS sensors, respectively. This demonstrates that both sensors could characterize corn canopy changes over time. NDVI ranges were correlated with the main phenological stages (PE), using the direct relationship between both variables (NDVI and PE) with days after sowing (DAS). For the beginning and end of each phenological stage, NDVI ranges were validated through model identity testing. NDVI proved to be a suitable parameter to assess corn phenology accurately and remotely. Finally, NDVI was also an important tool for detecting biotic and abiotic stresses throughout the crop cycle, and hence for decision making based on corn phenology.