ABSTRACT
‘Chardonnay Musqué’ vines in Beamsville, Ontario, were subjected in one trial to three treatments: (1) control (hedged only); (2) basal leaf removal (BLR) and cluster thinning and; (3) cluster thinning during five stages of berry development. Berries and musts from each treatment were analyzed for soluble solids (Brix), pH, titratable acidity (TA), as well as free and potential monoterpenes (FVT and PVT). Wines produced from each treatment were evaluated by descriptive analysis for aroma and flavor intensities. FVT and PVT of berries were higher in thinned vines compared to non-thinned vines. Cluster thinning at veraison yielded fruit with the highest FVT and PVT concentrations. Leaf-pulled vines produced fruit with increased pH, reduced TA and highest FVT and PVT. Wines from BLR and thinning treatments generally had higher muscat and floral/perfumy aromas, and could be separated based on overall quality. The chemical and sensory data were incorporated into a multiple regression model used to construct a grape quality model for aromatic white Vitis vinifera grape cultivars in the Niagara Peninsula. The model developed was able to predict overall quality based on Brix, pH, and berry FVT and PVT concentration. The model was partially validated by correlations between berry FVT and PVT vs. floral and muscat aromas in wines from three previous vintages. This model has potential for use to create a more equitable payment schedule for growers contracted to wineries for the purchase of high-quality grapes.
Key Words: