Differential irrigation scheduling by an automated algorithm of water balance tuned by capacitance-type soil moisture sensors

Abstract

Automated software tools are required to undertake the routine tasks and decision-making involved in scheduling irrigation. A key issue in this topic is how to integrate sensors in the scheduling approach. The objectives of this research were to test, in the context of drip-irrigated orchards: (a) the suitability of FAO’s water balance method, locally adjusted by sensors, as the basis for the scheduling algorithm, (b) the suitability of capacitance-type soil moisture sensors, and an approach for their automated interpretation, for providing feedback to the scheduling algorithm, and (c) the performance of these combined approaches in the autonomous scheduling of irrigation in an apple orchard with heterogeneous vigour. The trial consisted of applying for two years the proposed approaches using an experimental web application, IRRIX, which scheduled irrigation of two irrigation sectors, which differed in tree size. The automated system was compared with manual scheduling by a classical water balance and with the actual evapotranspiration determined by a weighing lysimeter located in the same orchard. Results show that the irrigation applied by the automated approach in the sector of larger trees agreed with the ET determined by the lysimeter and, overall, with the scheduling by an experienced irrigator using a classical water balance. Meanwhile, as a result of a different feedback from soil moisture sensors, the same system reduced irrigation in the sector of smaller trees by a similar amount to that expected from the differences between the two sectors in the fraction of photosynthetically active radiation. This study illustrates that the method of water balance complemented with capacitance-type soil moisture sensors provides a sound basis for automated irrigation scheduling in orchards.