Integrating AI and Machine Learning into Agricultural Meteorology for Sustainable and Climate-Smart Farming

Abstract

Climate-driven uncertainties in weather patterns pose significant challenges to agriculture, particularly in regions heavily reliant on climatic conditions. This review examines the role of Artificial Intelligence (AI) and Machine Learning (ML) in agricultural meteorology, exploring their potential to enhance resilience, sustainability, and productivity in farming systems. It highlights how AI and ML contribute to improved weather forecasting, with models such as LSTMs achieving up to 76% accuracy in rainfall prediction and crop yield forecasting, where neural networks have reduced yield deviations to as low as 4-10% compared to 16-35% in traditional models. optimized use of agricultural inputs such as water and fertilizers, and more accurate crop yield predictions by integrating complex environmental datasets. The review further examines applications in early detection of extreme weather events, pest and disease outbreaks, where image-based deep learning models have achieved over 95% precision in pest detection and in climate-resilient crop planning supported by adaptive variety recommendations. In addition, the integration of IoT data, satellite imagery, and ground-based observations is shown to enable AI-driven decision-support systems for farmers. Finally, the paper synthesizes existing research while addressing challenges such as data scarcity, infrastructural limitations, ethical concerns, and accessibility issues, offering insights into the long-term potential of AI and ML in ensuring food security under changing climatic conditions. By bridging technological innovations with practical agricultural strategies, this review underscores the transformative impact of AI and ML in supporting sustainable, climate-smart, and precision agriculture initiatives globally.