Harnessing Machine Learning Techniques for Forecasting Crop Yields under Changing Climatic Conditions

Sarmistha Borgohain & Chayanika Nath

Abstract:

Global agriculture is facing increasing uncertainty due to climatic variability, which significantly affects crop growth and productivity. Accurate crop yield prediction under changing environmental conditions has become essential for ensuring food security and sustainable agricultural development. Traditional statistical and mechanistic models have been extensively used for yield estimation, yet they often fail to capture complex, nonlinear relationships among climatic, soil, and management factors. Machine learning techniques have emerged as powerful data-driven tools that can process large and heterogeneous datasets to predict crop yields with higher precision. Various algorithms such as regression models, random forests, support vector machines, artificial neural networks, and deep learning architectures have been successfully implemented for yield forecasting under variable climatic conditions. These methods demonstrate superior adaptability and accuracy in modelling the intricate interactions between environmental variables and crop performance. Integration of machine learning with remote sensing data, precision agriculture technologies, and climate simulation models can further improve the reliability and spatial resolution of yield predictions. The adoption of these advanced computational methods holds significant potential for enhancing agricultural resilience and facilitating informed decision-making in the face of climatic challenges.