A FULLY-AUTOMATED ORCHARD IRRIGATION SYSTEM BASED ON CONTINUOUS MONITORING OF TURGOR POTENTIAL WITH A LEAF SENSOR

Abstract

A sensor, which continuously monitors leaf thickness in the field with an accuracy of ± 1 micron, was developed and field-tested over a period of six growing seasons. The suitability of three different electrical components was investigated as transducers of linear changes in leaf thickness to a measurable electrical signal. A strain gauge based on a miniature printed circuit of a whetstone bridge fastened integrally to the face of a spring steel blade was found to be sufficiently accurate, able to withstand all ambient weather phenomena and agrotechnical practices, without interrupting normal leaf functions. The simplicity of this configuration makes it easy to install, trouble free with a functional life of at least four years. Notwithstanding the low output intensity of the device, a Campbell CR-10 data logger was able to trigger and read with sufficient accuracy, signals transmitted through standard, unshielded cables from a distance of up to 200 meters. The data logger was programmed to monitor and interpret the data received, and to open and close irrigation valves as a real-time response to leaf thickness variations. It created a high frequency, low volume irrigation regime that reacted to and reduced incipient stress with a resultant reduction of total water application of up to 30%. Concurrent research demonstrate that there is a linear and significant correlation between leaf thickness leaf turgor potential (R 2 >0.9), which in turn has been shown to be an accurate and sensitive measure of plant water status as it affects plant metabolism. The system was tested in commercial plots (citrus: 6 years; avocado: 3 years; cotton: one season) and compared with several variations of standard irrigation regimes. Cropping weights and quality of the sensor-triggered treatments were equal to or better than controlled treatments, but water use efficiency (WUE) was consistently and significantly higher in the sensor treatments.