Integrating Digital Agriculture and Smart Farming Technologies for Climate-Smart Crop Management: A Field-Based Evaluation of AI, IoT, and Precision Systems

Abstract

We conducted a field-based evaluation of digital agriculture and smart farming technologies across 20 research farms in diverse agroecological regions, employing 360 plots in a randomized complete block design over three cropping seasons. The study integrated IoT-based soil moisture monitoring, AI-driven crop health assessment, drone surveillance, GPS-guided precision farming, remote sensing, automated irrigation, a cloud-based farm management platform, and a climate forecasting decision support system. A hybrid deep learning framework combining Long Short-Term Memory and Convolutional Neural Network models was developed for crop growth prediction, disease identification, and irrigation scheduling, achieving a prediction accuracy of 97.1%, precision of 96.5%, recall of 96.0%, and an F1-score of 96.2% with a root mean square error of 0.16. The implementation of these technologies resulted in a 32.8% increase in crop yield, a 30.6% improvement in water-use efficiency, a 26.9% reduction in fertilizer consumption, and a 23.8% decrease in crop disease incidence compared to conventional farming. Multiple linear regression revealed that AI-based decision support was the strongest predictor of climate-smart crop management (β = 0.48, p < 0.001), followed by IoT soil moisture monitoring (β = 0.43, p < 0.001), with the model explaining 80.1% of variance in crop productivity. Principal component analysis extracted three components accounting for 85.8% of total variance, representing AI-driven crop management, IoT-based water optimization, and climate resilience. The Smart Farming Index showed a strong positive correlation with crop productivity (r = 0.88, p < 0.001). These findings demonstrate that integrating AI, IoT, remote sensing, and precision systems provides a robust pathway toward sustainable agricultural intensification under climate variability, significantly improving resource-use efficiency and environmental outcomes.