The influence of vine variety and vineyard management on the effectiveness of canopy sensors to predict winegrape yield and quality

Abstract

Precision Viticulture (PV) is the application of precision agriculture techniques in vineyards. The ultimate aim of PV is to manage vineyards at sub-field scale delineating management zones. In winegrapes, successful zoning depends on the relevancy of management zones in terms of grape quality to justify selective handling of grapes. The aim of this study was to evaluate the effectiveness of a canopy sensor commonly used in PV, in predicting grape yield and quality variability under different varieties and vineyard conditions. The study was conducted in a commercial vineyard in Northern Greece in the 2013 vintage. The vineyard was planted with Vitis vinifera cvs. Sauvignon Blanc and Syrah in 2.5 and 1.7 ha respectively. Vines were trained to a bilateral cordon and were spaced 1.2 x 2.2 m. Sauvignon blanc received many operations for summer training techniques and extensive irrigation, while Syrah only received a limited number of manual operations and restricted irrigation. The soil characteristics were determined using an elevation map and an electrical conductivity map. Canopy properties in terms of NDVI values (measured using the Crop Circle sensor) were taken at six different stages of the vine growing period . A 0.1 ha regular grid was formed to assess grape yield at harvest and grape composition parameters throughout ripening. NDVI measurements successfully monitored canopy growth variations within the vineyard in both varieties. However, NDVI was not related to vine water status in any variety and was significantly correlated with ECa only in Sauvignon blanc. The results showed a strong influence of grape variety and vineyard management decisions on the effectiveness of canopy sensors in describing the spatial variability in grape yield and quality. NDVI was positively correlated with yield and berry weight and negatively correlated with must total soluble solids and anthocyanins in Syrah. On the other hand, no correlations between NDVI, yield and berry quality parameters were observed in Sauvigon blanc. It is proposed that the lower effectiveness of NDVI in predicting yield and quality in Sauvignon Blanc, as opposed to Syrah, was related to the increased number of management practices, reducing the initial variability induced by soil and topography factors.