SUGARCANE YIELD PREDICTION USING MACHINE LEARNING IN SENTINEL-2 AND GOOGLE EARTH ENGINE FOR ANDHRA PRADESH, INDIA

Sugarcane is one of the major commercial crops cultivated in Andhra Pradesh, India, and accurate yield estimation is essential for agricultural planning, crop management, and food security. Conventional yield estimation methods are labor-intensive, time-consuming, and often limited in spatial coverage. Recent advancements in remote sensing, cloud computing, and machine learning have provided efficient alternatives for large-scale agricultural monitoring and crop yield prediction. In this study, a machine learning-based framework was developed for sugarcane yield prediction using Sentinel-2 multispectral imagery and Google Earth Engine (GEE) in Andhra Pradesh, India. Sentinel-2 data were processed in GEE to generate vegetation indices, including the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Normalized Difference Red Edge (NDRE), and Soil Adjusted Vegetation Index (SAVI). Historical district-wise sugarcane production data from 1998 to 2018 were integrated with remotely sensed features for model development. A Random Forest regression model was employed due to its robustness and ability to handle nonlinear agricultural datasets. The model performance was evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and coefficient of determination (R²). The results demonstrated that Sentinel-2 vegetation indices effectively captured crop health and biomass variability, while the Random Forest model achieved reliable prediction accuracy for sugarcane yield estimation. District-wise analysis indicated higher productivity in Krishna and West Godavari districts, whereas lower yields were observed in Visakhapatnam and Vizianagaram districts. The proposed framework highlights the potential of integrating remote sensing, machine learning, and cloud computing technologies for precision agriculture and sustainable crop management. The developed methodology can support agricultural decision-making, yield forecasting, and policy planning for sugarcane cultivation in Andhra Pradesh.