Efficient irrigation management is essential to improve water-use efficiency under increasing freshwater scarcity and climate variability. However, accurate automation remains limited by the difficulty of estimating the crop coefficient (Kc), a key parameter in the FAO-56 evapotranspiration model, which typically requires periodic experimental recalibration. This paper addresses this limitation by proposing an IoT-enabled smart lysimeter architecture integrating heterogeneous industrial sensors, wireless communication, cloud-based data management, and artificial intelligence. A Dynamic Bayesian Network models Kc as a virtual sensor inferred from environmental and hydraulic measurements, enabling predictive computation of crop evapotranspiration (ETc) and adaptive irrigation threshold control. Field experiments conducted over two months under Mediterranean conditions demonstrate a 16.6% reduction in irrigation volume and a 64% decrease in runoff compared to conventional farmer-managed irrigation, while maintaining equivalent ETc. These results confirm that IoT-driven virtual sensing enables data-driven and resource-efficient irrigation management.
IoT based Smart Lysimeter for Automated Crop Coefficient Prediction
Kocian A.;Cela F.;Carmassi G.;Chessa S.;Milazzo P.;Incrocci L.
2026-01-01
Abstract
Efficient irrigation management is essential to improve water-use efficiency under increasing freshwater scarcity and climate variability. However, accurate automation remains limited by the difficulty of estimating the crop coefficient (Kc), a key parameter in the FAO-56 evapotranspiration model, which typically requires periodic experimental recalibration. This paper addresses this limitation by proposing an IoT-enabled smart lysimeter architecture integrating heterogeneous industrial sensors, wireless communication, cloud-based data management, and artificial intelligence. A Dynamic Bayesian Network models Kc as a virtual sensor inferred from environmental and hydraulic measurements, enabling predictive computation of crop evapotranspiration (ETc) and adaptive irrigation threshold control. Field experiments conducted over two months under Mediterranean conditions demonstrate a 16.6% reduction in irrigation volume and a 64% decrease in runoff compared to conventional farmer-managed irrigation, while maintaining equivalent ETc. These results confirm that IoT-driven virtual sensing enables data-driven and resource-efficient irrigation management.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
