Assessing the capability of MODIS to monitor mixed pastures with high-intensity grazing at a fine-scale

MODIS time series carries valuable long-term data essential to support several studies such as biogeochemical modelling. However, there is a lack of validation studies applying MODIS data at a fine-scale to monitor pasture management practices. In this study, we assessed the potential of MODIS sensor in monitoring at a fine-scale four intensively managed mixed-pastures fields located in Sao Paulo State, Brazil. The MODIS spectral response was compared with Sentinel-2, and the ability of the two sensors in predicting aboveground biomass (AGB) and canopy height (CH) was assessed using the Random Forest algorithm. EVI images from MODIS and Sentinel-2 were correlated with field measurements of AGB and CH. The prediction performance of AGB (R-2: Sentinel-2 = 38%; MODIS = 42%) and CH (R-2: Sentinel-2 = 69%; MODIS = 85%) models was superior using EVI data from MODIS than Sentinel-2, highlighting MODIS ability to monitor small and intensively managed pasture fields.