Experimental testing of a model-based decision support system integrated with smart sensors to optimize irrigation strategies for processing tomato: A case study in southern Italy

Advanced digital technologies, model-based Decision Support Systems (DSS) and soil/plant sensors, are promising tools to support the ‘real-time’ decision-making process, in the contest of precision farming. The purpose of this research was to evaluate the reliability of a ‘cloud-based’ DSS system (Bluleaf™) to support the management of irrigation strategies for processing tomato, in relation to the prediction of the soil water balance and the level of crop stress. An experimental research was conducted in Southern Italy (Foggia, 41°46’N, 15°54’E; altitude 74 m a.s.l.) over two growing seasons (2015 and 2016). A processing tomato hybrid was grown under two irrigation regimes (‘full irrigation’ – IR100 and ‘deficit irrigation’ – IR75). Irrigation scheduling was computed by means of Bluleaf™ DSS (based on the FAO-56 approach) integrated with weather/soil sensors for the continuous monitoring of the system. Stomatal conductance (gs) and crop water stress index (CWSI) were respectively measured/calculated several times during the cropping cycle, as reliable indicators of plant water stress. At harvest, the yield was measured and the irrigation water use (I-WUE) was computed. A good agreement was found between model-based simulations and field measurements for both years and irrigation regimes, in terms of prediction of CWSI and soil water balance components. As expected, IR75 showed values of gs and CWSI significantly different from IR100, and a linear relationship between the two indicators was found. Accordingly, soil moisture content (as measured by sensors and simulated by the model) in IR75 was progressively lower than for in IR100 along the growing seasons.