Empirical model for forecasting sugarcane yield on a local scale in Brazil using Landsat imagery and random forest algorithm

•Models are calibrated over 4 years to predict sugarcane yield before the harvest.•Combining Landsat images, meteorological and agronomic data improve the sugarcane yield prediction.•The NDMI index derived from Landsat images is one of the most important predictors.•Including expected harvest date i...

Ausführliche Beschreibung