Water Content Continuous Monitoring of Grapevine Xylem Tissue Using a Portable Low-Power Cost-Effective FMCW Radar

Abstract

This paper presents the real-time monitoring of a grapevine's water content that flows up through the xylem tissue by means of a frequency-modulated continuous-wave radar. The application of an optimization process, based on the superresolution multiple signal classification algorithm, has enabled the reduction of the bandwidth required to discern the xylem water content and, thus, the operating frequency, achieving a depth resolution of at least 3 mm. This design advantage has resulted in a significant step forward toward a real life application, allowing the use of fully-integrated off-the-shelf components in order to implement a completely portable low-power cost-effective radar at 23.1 GHz with a 3.4-GHz bandwidth. The sensor performance has been evaluated by means of three different experiments: irrigation cycles, day/night cycles, and comparison between irrigation cycles at different temperatures. From the experimental results, it is possible to assert that the contactless sensor presented in this paper is very sensitive to changes in the plant's water content, differentiating between daytime and nighttime. In addition, it has been proven that temperature has a noticeable influence over the evapotranspiration, observing negative drying slopes of 5.62 and 6.28 mV/cycle at 23°C and 26°C, respectively.

Remote Sensing for Plant Water Content Monitoring: A Review

“…This type of sensor divides the transmitted power into two signals, the reference signal used as local oscillator (LO) and the RF signal, which is radiated toward the target, reflected, and sent back to the receiver to be mixed with the reference signal. At the mixer input, the time delay between both signals can be estimated from their frequency difference and used to determine the distance D to the reflecting object by applying the following equation [84]:…”

“…For a medium different from vacuum, c must be replaced by the group velocity V g at which the electromagnetic wave propagates in the medium. The depth resolution ∆D, defined as the minimum separation in depth between two targets of equal cross section that makes it possible to identify them as separate targets, can be expressed as follows [84]:…”

“…The application of this technology on the monitoring of plant water content consists in measuring the power reflected on the plant for different degrees of hydration. The part of the plant more commonly subjected to measurement is its trunk [84].…”

“…Remote sensing methods can be classified under multiple criteria and, in this work, they have been categorized according to the part of the plant subjected to measurement. In this way, the different sensors used in the available literature can be applied to the soil [9][10][11][12][13][14][15][16][17], directly to the canopy or to different organs of individual plants, such as leaves or trunks and stems [71][72][73][74][75][76][77][78][79][80][81][82][83][84][85].…”

“…Other studies describe methods that need contact with the trunk or stems to develop the monitoring function [75,[78][79][80]. Among the main sensors that estimate the tree water status include coaxial probes called portable dielectric probes (PDP) [71][72][73][74][75], volumetric soil water content (VSWC) sensors inserted into tree trunks [76], sap-flow measurements based on sensors that measure sap-flow rate (SFR) (g•h -1 ) [78][79][80] or sap-flux density (SFD) (m 3 •m -2 •h -1 ) [77,81], CW vector network analyzer (VNA) spectroscopy [82,83], portable low-power cost-effective FMCW radars [84] and portable nuclear magnetic resonance (NMR) sensors [85]. In relation to the parameters measured by these sensors, PDPs, by using the complex reflection coefficient, estimate the complex xylem dielectric constant (XDC) to be correlated with the xylem SFD, the xylem sap chemical composition and xylem water potential (XWP).…”