Response of Shiraz grapevines to five different training systems in the Barossa Valley, Australia

Own‐rooted Shiraz, spaced 1.5 m apart in 2.75 m wide rows, were compared under five different training systems in the Barossa Valley, South Australia over five seasons starting 1995–96. Our objective was to determine how training affected fruit composition and crop yield in a context of efficient vineyard management. Training systems were: (i) low single wire (LSW), in which vines were trained to bi‐lateral cordons, 1.0 m above the ground, non‐shoot‐positioned; (ii) high single wire (HSW), a higher cordon (1.8 m) version of LSW; (iii) vertically shoot positioned (VSP); (iv) Scott Henry, where canopies of alternate vines were cordon‐trained and shoot‐positioned upwards (SHU) or downwards (SHD); and (v) minimally pruned (MIN), the same as LSW, except vines were not annually spur‐pruned. Crop yields (kg/m of row), over four seasons, averaged 4.9 (MIN), 2.9 (combined Scott Henry), 3.2 (VSP), 2.6 (HSW) and 2.9 (LSW). The greatest year‐to‐year variation in yield occurred with MIN and Scott Henry training. Crop weight of SHD vines was inferior to that of SHU vines in two seasons due to fewer shoots and bunches per vine, and to fewer bunches per shoot. Individual berry weights (g) were consistently least with MIN (0.89), and greatest with VSP (1.17). Fruit from all training systems exhibited similar rates of sugar accumulation during two seasons in which repeated measures of fruit maturation were made. Excepting the relatively light yields observed in 1999–2000, sugar accumulation was delayed in MIN vines, relative to other training systems, even when MIN harvest was delayed up to two weeks. The delay is most readily explained by the greater crops of MIN vines. Berry total anthocyanins and total phenolics concentrations (mg/g berry fresh wt.) at harvest were not greatly affected by training system. Berry anthocyanins and total phenolics exhibited a negative relationship with crop/m of canopy and a slight positive relationship with bunch exposure when evaluated across all training systems. Experimental plot soil depth and water availability affected cane pruning weights, yield per vine, berry weight, and canopy characteristics. LSW, HSW and MIN training systems all provided good yields of high quality fruit, although MIN did have a tendency to produce excessive crops in some years. VSP and Scott Henry training were less attractive due to their inherently greater canopy management requirements.