Climate-Resilient Smart Farming: AI and IoT Driven Solutions for Adaptive Agriculture under Extreme Weather Conditions

Agriculture is increasingly vulnerable to the impacts of climate change, with marginal and smallholder farmers facing disproportionate risks from droughts, floods, erratic rainfall, and pest outbreaks. Traditional advisory systems are often reactive, generalized, and inaccessible, leaving farmers without timely, crop-specific guidance. This results in delayed detection of stress conditions and significant yield losses, further exacerbating poverty and food insecurity. To address these challenges, this research proposes an integrated framework that leverages Artificial Intelligence (AI) and the Internet of Things (IoT) to enable climate-resilient smart farming under extreme weather conditions. The methodology is structured into five phases. First, IoT Deployment & Data Acquisition involves installing low-cost soil, weather, and crop sensors, complemented by drone and satellite imagery, to build a unified dataset. Second, Vegetation Index Computation calculates NDVI, SAVI, EVI, and GNDVI to detect early signs of crop stress. Third, AI Model Development applies advanced architectures—LSTM, GRU, Transformers, and CNNs—to forecast yields and classify stress conditions, with hybrid ensembles integrating sensor and imagery data for robust predictions. Fourth, a Decision Support System (DSS) translates complex analytics into actionable advisories via mobile and voice platforms, ensuring accessibility through multilingual support and offline SMS/voice features. Finally, Field Trials & Validation on rice, maize, and cotton farms assess yield improvements, resilience gains, and socio-economic impacts. Expected outcomes include ≥90% accuracy in stress detection and yield prediction, a 20–30% reduction in climate-induced crop losses, and a cost-effective DSS accessible to marginal farmers.