Geography-Aware Energy-Adaptive IoT-WSN Monitoring for Sustainable Smart Irrigation in Precision Agriculture.

IoT-based wireless sensor networks are increasingly used in smart irrigation, allowing continuous monitoring of soil, weather and crop-water conditions. However, the deployment of sensors knowing only the geographical location cannot ensure sustainable monitoring as the operational constraints such as communication distance, residual energy, packet delivery, dynamic irrigation risk, and node lifetime are still critical. This paper presents a sustainable smart irrigation monitoring framework using a geography-aware energy-adaptive IoT-WSN. The framework encompasses optimized sensor deployment, radio-energy modeling, sink placement analysis, event-driven and adaptive communication, solar energy harvesting, and machine learning-based next-round irrigation risk prediction. Simulations were performed with an irrigation monitoring dataset consisting of multiple measurements of soil moisture, soil nutrient status, temperature, humidity, precipitation, solar radiation, evapotranspiration, spatial coordinates, and critical-event labels. Results show that direct transmission provides complete packet delivery but short lifetime, and event-driven communication reduces transmissions over 94% but weakens the monitoring continuity. ML-assisted adaptive monitoring reduces the transmissions between 56.57%–61.94% while adaptive harvesting achieves the strongest composite sustainability ranking of 0.8317 packet delivery ratio and 1000-round lifetime. This work provides an integrated framework for the joint assessment of energy efficiency, reliability, prediction and sustainability under field constraints for precision agriculture.