Sheath moisture is an important parameter during the growth period of sugarcane, which is of special importance from the perspective of water stress and field irrigation management. Remote sensing data has a high capacity to update crop growth monitoring systems. In this regard, satellite images that provide a variety of information can be used. In the crop year of 2020, with the aim of predicting the moisture content of sugarcane pods, 4 spectral indices and 7 single band sensors of Sentinel-2 satellite were evaluated. Four methods PLSR, RF, GRNN and SVR were used to model and predict pod moisture. Bayes algorithm was used to optimize the parameters in RF, GRNN and SVR models. In addition, improved sensitivity analysis was used improved stepwise was used to find the most effective input parameter in estimating pod moisture. The results showed that the SVR model provided a more acceptable estimate of sheath moisture content than the other models when the parameters NDVI, EVI, SRWI, Clgreen, B2, B3, B5, B4, B11 and B12 were used as input to the four models. According to the sensitivity analysis, SRWI parameter was considered as the most effective index in the modeling process. Therefore, it can be concluded that among the inputs given to the model, a combination of indices and bands of NDVI, EVI, SRWI, Clgreen, B2, B3, B5, B4, B11 and B12 give a better estimate of sugarcane sheath moisture content.