AI-based Geo-Spatial Data Modeling of Agriculture for Crop Yield Estimation

Integrating remote sensing satellite data with IoT, advanced sensors, and AI has proven essential in optimizing greenhouse parameters, geospatial technology, and big data analysis enhancing productivity and sustainability. Precise crop yield calculations are crucial for biomass estimation and carbon sequestration, supporting sustainable agricultural practices. Artificial Intelligence tools using high spatial and temporal resolution satellite data must be developed to map sowing dates and yields suitable for Indian conditions. This research investigates six machine learning models applied to predict wheat crop yield using NDVI and NDWI time series data and village-level yield estimates from Karnal district, Haryana. Random Forest Regression emerges with the highest R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.84 indicating strong explanatory power in capturing yield variance, closely followed by Gradient Boosting Regression with R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.81.