Fast and non-destructive method for estimating grapevine water status

Y. Barnard, A. Strever, G. Bosman, C. Poblete-Echeverría
Appropriate water management assessment methods in vineyards are critical for modern precision viticulture (PV) practices. In this context, traditional methods used for measuring water status in grapevines are costly and time-consuming. The main objective of this study was to estimate grapevine water status using a fast and non-destructive method. The study focused on the potential of using a handheld spectroscopy sensor for fast and accurate estimation of leaf water potential (LWP). The experiment was conducted at Welgevallen experiment farm of the Department of the Stellenbosch University in a vineyard block 'Pinotage'. Sampling points (n=33) were selected considering the different water conditions presented in the experimental block. LWP and spectral data were measured at midday simultaneously in each sampling point using a digital Scholander pressure chamber and a handheld fluorescence detector fitted with a 405 nm laser excitation source and a USB spectrometer (OceanOptics) with a wavelength response from 200 to 900 nm. The data analysis was carried out using a customised Matlab (R) code and the PLS-Toolbox. A variable selection method with the VIP (variable importance in projection) criteria was used to develop the partial least square (PLS) regression model. Results show a high correlation between the measured LWP and the LWP estimated by the PLS model with a coefficient of determination (R2) in the cross validation of 0.79. The error in the model was low with a root mean square error of cross validation (RMSECV) of only 0.89 bars. These results show the potential of this method for a fast and non-destructive determination of grapevine water status under field conditions as a tool for decisions making during irrigation scheduling.
Barnard, Y., Strever, A., Bosman, G. and Poblete-Echeverría, C. (2019). Fast and non-destructive method for estimating grapevine water status. Acta Hortic. 1253, 413-420
DOI: 10.17660/ActaHortic.2019.1253.54
https://doi.org/10.17660/ActaHortic.2019.1253.54
PLS regression, precision viticulture, spectroscopy, VIP variable importance in projection
English

Acta Horticulturae