The purpose of this research is to evaluate the amount of nitrogen and leaf sheath moisture using remote sensing and artificial intelligence models. For this purpose, three models of Support Vector Machine , Random Forest (RF) and Artificial Neural Network were used, It should be noted that the sensitivity analysis of all input parameters to simulate nitrogen and leaf sheath moisture values was done before running the model using Pearson correlation and only those factors that were more important on nitrogen or leaf sheath moisture were considered. Single bands B2, B3, B4, B5, B6, B7, B11, B12 and NDVI vegetation index and in the simulation of leaf sheath moisture in bands B2, B3, B4, B5, B6, B7, B11, B12 and vegetation indices NDVI, NDMI and LAI are more important and were considered as inputs to the models. Also, among the models, the random forest (RF) model in nitrogen simulation with R2 equal to 0.88 and RMSE equal to 0.06 in the training phase and R2 equal to 0.93 and RMSE equal to 0.05 in the validation phase had the best performance. In addition, leaf sheath moisture with R2 equal to 0.88 and RMSE equal to 0.71 in the training phase and R2 equal to 0.92 and RMSE equal to 0.59 had a better performance in simulation. The results showed that the combined use of vegetation index, sentinel 2 satellite bands and artificial intelligence models gives an acceptable estimate of the amount of nitrogen and leaf sheath moisture in sugarcane fields.