Characterising spatio‐temporal variation in fruit composition for improved winegrowing management in California Cabernet Sauvignon

Abstract

Background and Aims Spatial variability in yield and fruit composition in winegrape vineyards has been demonstrated, but few chemical compounds responsible for impacting wine composition have been analysed at a sample density high enough to compare with variability in remotely sensed imagery. The aims of this project were to evaluate spatial variability in grape composition at harvest in three seasons and to compare this with remotely sensed canopy vegetation data to assess its utility in underpinning targeted management. Methods and Results The composition of fruit samples were analysed to compare their spatial variability with aerial imagery products, the normalised difference vegetation index (NDVI) and the difference between canopy temperature from imagery (Tc) and ambient temperature from ground weather stations (Ta), (Tc − Ta). Zonal discrimination of fruit composition using k-means clusters generated from seasonal aerial imagery showed a difference as high as 2.7 kg/m in vine yield, up to 0.3 mg/g anthocyanins and 1.2 pg/g carbon-6 alcohols and aldehydes (C6) with these ‘quality zones’ reflected by the imagery in some vineyards and/or seasons. Conclusions The NDVI and (Tc − Ta) data collected at multiple time points were correlated with several attributes of fruit composition evaluated at harvest, but most correlations peaked at veraison. They were also strongest in vineyards in which the spatial variation showed stronger spatial structure. Significance of the Study Spatial variations in berry chemistry followed similar patterns to those seen in aerial imagery of vineyards with structured vigour zones. Furthermore, as most of the spatial structure in the variation of fruit composition is dominated by flavanols, opportunities for reduced analytical costs in winery laboratories also arise.