Effects of Berry Size and Crop Yield on Wine Composition and Sensory Quality

The effects of vine irrigation and crop yield on sensory quality of Cabernet Sauvignon wine from the 2000 harvest were investigated using descriptive analysis with a trained panel, quality ratings by members of the wine industry and tannin assays.Wines produced from vines with low irrigation regimes were rated highest in dried fruit/raisin, jam, and red/black berry aromas, fruity by mouth, and acidity and lowest in brown color. The high irrigation wines were rated lowest in bitterness, ethanol, body, and darkness. The medium irrigation wines were ranked highest in veggie aroma, astringency, brown, dark, body, ethanol, and bitterness and were ranked lowest in cherry aroma. Low irrigation wines received slightly higher quality ratings from members of the wine industry than high irrigation wines, but the difference was not significant, possibly because of differences among wine replicates.Lower crop yields tended to produce wines with high bell pepper and black pepper aromas, high astringency and bitterness, and high ethanol and veggie by mouth flavors, whereas the higher yields tended to result in wines with higher red/black berry, jam, and cherry aromas, red color, fruity by mouth, and acidic characters. Tannin concentration was significantly higher in the wines produced from low crop yields, consistent with the higher bitterness and astringency ratings given to those wines by descriptive analysis.We continued our investigation of the effects of crop yield on wine sensory quality using pruning and thinning trials on Cabernet Sauvignon grapes harvested in 2001.In the pruning trial, six treatments that left 12 to 48 buds/vine at pruning were imposed for the second consecutive season on the same vines. Vines were pruned to variations of 1 to 4 bud spurs and shoot thinned accordingly. These treatments resulted in 24 to 60 clusters/vine and produced yields that varied almost 2.5 times from lowest (3 tons/acre) to highest (8 tons/acre). In addition to yield components, a number of vegetative growth parameters were measured including budbreak, shoot diameter, and shoot length. There was only a slight decrease in shoot diameter, but shoot length was a strong inverse function of bud number, varying about 100{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} from highest to lowest bud number per vine. The cluster weight did not differ significantly among treatments. Although berry size was not constant among treatments, the differences were not greater than about 10{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} and did not show any clear relationship to the crop loads.In the thinning trial, vines were pruned to 2 or 4 bud spurs that were then cluster thinned at veraison to create 8 different crop load treatments. Crop load in this cluster thinning experiment was well distributed among the 8 thinning treatments, varying about 4-fold from 4 to over 16 lbs. per vine. These crop loads correspond to yields that vary from about 2 tons/acre to about 8 tons/acre. Thus, we were more successful in accomplishing low yield by cluster thinning than by severe pruning. There were 14 replicated yield treatments, six for the second year and eight for the first time in 2001. Wines were made from each treatment, and they are ready for bottling and subsequent sensory and chemical analysis this spring.

Effects of Berry Size and Crop Yield on Wine Composition and Sensory Quality

We draw these conclusions about the dependence of composition on size and irrigation:

  1. Imposing significant water deficits can decrease berry size and increase skin: juice ratio by approximately 25{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} compared to conventional irrigation regimes. From two separate yield experiments, it appears that smaller berries are not produced when yields are increased from 3.5 to over 10 tons/acre by leaving longer spurs at pruning.
  2. Berry size is highly correlated with seed mass/berry. However, berry growth did not keep pace with seed growth such that the amount of seed per unit berry mass increased with berry size. Low irrigation had less effect on seed growth than on mesocarp growth for any size of berry. Hence, Low irrigated fruit had greater relative seed mass, implying a greater concentration of seed tannin in the resultant wines.
  3. The mass of skin tissue bore a constant relation to berry size, i.e. the mass of skin per unit berry mass was relatively constant over the 3-fold range of berry sizes encountered. Low irrigated fruit had greater relative skin mass per berry. Low irrigation inhibited mesocarp growth more than skin growth such that the relative amount of skin per unit berry mass was always greater than for Control and High irrigated fruit.
  4. The amount of skin tannin per berry was a fairly constant fraction of berry size. Thus, the concentration of skin tannin in must was only slightly lower for large fruit than for small fruit (approximately 10{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23}). However, the concentration of skin tannin was significantly greater in Low irrigation fruit of any size (approximately 30{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} greater in Low fruit).
  5. The amount of anthocyanin per berry increased with berry size, but was a slightly decreasing fraction of berry size. The resulting anthocyanin concentration was approximately 20{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} less from the largest fruit compared to small fruit. Again, the concentration was significantly greater (approximately 30{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23}) in Low irrigated fruit for all fruit sizes.

These results show that the amount of each skin solute is not a constant that is diluted to varying degrees dependent upon fruit growth. Accumulation of skin solute is coordinated with growth. Accordingly, oft observed changes in fruit composition caused by water deficits are not attributable simply to differences in berry growth. There is an independent effect of deficit irrigation on composition. Because deficit irrigation produces lower yield, it is important to evaluate whether there is an independent effect of yield on these quality parameters.

The fruit size and irrigation differences in fruit composition were carried through to the resulting wines.

The differences in wine composition were slightly less dependent upon fruit size than the juice composition, suggesting a possible difference in “extractability”. And, both fruit and wine composition were much less sensitive to fruit size than the theoretical dependence that is predicted from differences in surface: volume of the berries. A preliminary investigation found differences among the low, control and high irrigation wines in the intensities of green beans aroma, red fruit by mouth, bitterness and astringency as shown below. When wines were made from different sized berries that were irrigated similarly, small berry wines were more astringent than large berry wines.

PDF: Effects of Berry Size and Crop Yield on Wine Composition and Sensory Quality

Determination and Consequences of Berry Size in Winegrape Production

A series of field experiments were conducted with Cabernet Sauvignon to investigate means of creating differences in berry size and the consequences thereof for winemaking. The primary focus was the hypothesis that irrigation alters wine composition primarily by changing berry size and the “dilution” of compounds of sensory significance. Additional questions included testing whether pruning to a high initial crop load and subsequent thinning could be used to manipulate berry size; attempting to make wine from berries of different size; and evaluating berry size distribution in clusters and on vines. Putting on a heavy crop load and thinning later was ineffective as a means of realizing smaller berries. Cluster thinning prior to veraison created four-fold differences in crop load, the highest being about 11 tons/acre. Yet the results clearly showed no differences in final berry size or color extracted from skins. This somewhat surprising result implies that even larger crop loads are required before berry growth becomes limited. It is in part for this reason that we propose to impose a wider range of crop loads and to investigate the consequences of crop load differences for wine chemical and sensory attributes in the coming season. When vines were differentially irrigated, mean berry volume was 17{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} greater with High irrigation (2x normal) compared to Low irrigation (minimal water applied). The predicted differences in must and wine composition would be of similar magnitude if there were no effects of vine water status on the amount of skin or seed solutes per berry. However, both anthocyanin and skin tannin per berry were greater in Low irrigated fruit than in the other treatments. Thus, irrigation regimes altered fruit composition by means other than simply berry expansion. Accordingly, the wine of Low irrigated fruit had 28{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} greater tannin concentration than the High irrigated wine (vs. the theoretical 17{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23}). Seed tannin per berry (and seed mass) did not differ among irrigation treatments for any berry size class. In order to test the feasibility of sorting berry size for winemaking, hand harvested clusters were carefully destemmed, fruit sorted according to size, and made to wine. The tannin concentration was 13{aed9a53339cdfc54d53cc0c4af03c96668ab007d9c364a7466e3349a91bf0a23} greater in the small berry wine than in the other treatments; less than predicted on size alone. The results were promising in terms of the feasibility of developing technology to create small berry wines post harvest. However, it is not yet clear how important sorting might be to wine flavor, which is what we propose to address in the next funding cycle. Our informal bench tasting indicated treatment differences among irrigation and berry size treatments, the latter apparently differing in vegetative character. This report is limited to the data that have been analyzed to date with some limited interpretation of the implications of the data. However, the major tests were accomplished. We found that irrigation affected wine composition more than predicted by berry size alone and that sorting berry size had less than the hyothetical effect on wine tannin. Large differences in yield had no effect on berry size. We hope to pursue the sensory consequences of differences in yield, berry size, and irrigation.