Effect of Pruning, Training, Thinning, and Shoot Positioning on the Yield and Quality of 'Concord' Grapevines ©2001
Principal Investigator: Terence R. Bates, Research Associate, Cornell University, NYSAES, Department of Horticultural Sciences, Fredonia Vineyard Lab, Fredonia, NY 14063
NY Project Cooperators: Richard Dunst, NYSAES; Tim Weigle, NYSAES-IPM Multi-state Cooperators: Marcus Keller, Lynn Mills (WSU); Stan Howell, Bill Nail (MSU), Tom Davenport (National Grape Co-op), John Jugovich (Welch's)

The unique practice in this experiment is that each viticulture treatment is harvested when the fruit reaches 16±0.5obrix. Therefore, harvest date is examined for its effect on juice quality and vine production.

Question 1: Are there differences between the states?
Summary information has been discussed between the researchers in NY, WA, and MI on a yearly basis since the start of the experiment. Further specific analysis of the data is needed to accurately answer question one; however, some general insights are evident. Flavor analysis indicates that tasters are unable to differentiate between Eastern and Western juice if the fruit is all harvested at a target soluble solids (16 obrix). Juice from NY and MI tend to have higher titratable acidity than juice from WA; however, the high soluble solids of the juice masks the slight difference in TA. From a production standpoint, at a given vine size and pruning level, WA vines will have higher yield because of higher bud fruitfulness and greater berries/cluster. The difference in yield components is most likely because of differences in solar radiation, heat accumulation, and vineyard water management between the states. There also appears to be an effect of vine size between the different states' sites. In balanced pruned treatments, WA, NY, and MI tend to have large (4 pound), intermediate (3 pound), and small (2 pound) vines, respectively. Vine size, in turn, will have an effect on maximum sustainable yield potential.

Inter-state data has not yet gone through formal statistical analysis; therefore, general insights are from discussion only.

Question 2: Are there differences between treatments within a state?

Figure 1: Yield and relative harvest date of the 11 viticulture treatments in NY from 1997 to 2001. (The numbers above the bars refer to the average relative harvest date. i.e. on average, Minimal pruned vines reached 16 obrix 11.6 days after balanced pruned vines).

Figure 2. Effect of retained nodes (bud number) on yield, harvest date relative to balanced pruned vines, ripe nodes of periderm, brix, and titratable acidity on single-wire (HRU - A and B) and GDC (C and D) ownrooted, Concord grapevines. Each point is the five year mean for that measurement. As the bud number increases, yield increases to the yield compensation point and relative harvest date increases through the yield compensation point (as seen in figure 1). However, figure 2 shows that periderm (or vine size) begins to decrease at the yield compensation point. This is an indication that the higher bud number treatments are inefficient with a later harvest date and decreased vine size. Since all treatments are picked at 16 obrix, the brix line is flat between all node numbers. However, the rate of sugar accumulation is different for all of the treatments; therefore each treatment is picked on a different date. Relative harvest date and titratable acidity are inversely related, therefore, the higher node number later ripening treatments are the ones with lower titratable acidity. With GDC, the curves are shifted but the relationships remain the same.

Question 3: Are there differences between vines within a treatment?

Figure 3: The effect of retained nodes on yield (A, similar to figure 1). The effect of pruning weight on yield (B) and brix (C) of 130 node and balanced pruned vines. As seen in figures 1 and 2, as the node number increases the yield increases. Balanced pruned, 90 node, and 130 node vines also have the advantage of maintaining vine size over the other HRU treatments. These treatments have maintained canopy and water use efficiency and the difference in sugar accumulation (harvest date) between the three treatments is primarily a yield effect. However, there is variation in yield from year to year and from vine to vine.

Focusing in on two treatments (130 nodes and balanced pruning), vine size can account for much of the vine to vine variation. In balance pruning, node number is adjusted based on vine size. Therefore, small vines have fewer nodes and clusters per vine, which lowers the yield per vine (Figure 3 B). Vine to vine variation in yield is controlled by node number. In balanced pruning, vine leaf area and yield increase and decrease together. Therefore, the crop load (leaf area to fruit ratio) remains fairly constant and fruit maturity remains constant (Figure 3 C).

With constant 130 node pruning, the bud number is not adjusted with vine size. Large (3 pound) vines produce large crops of high quality (Figure 3B and C). At this site, 3 pound vines are balanced at 130 nodes. As vine size decreases, the crop level is forced to remain high by the constant 130 node pruning. In small vines, the relative decrease in leaf area is more than the decrease in crop level. This creates an overcrop which leads to lower yield (fewer clusters per node, fewer berries per cluster) and lower brix (Figure 3B and C).

130 Node pruning can be an efficient compromise between the needs of Concord growers (maximum potential yield) and processors (good juice quality at an acceptable harvest date) IF AND ONLY IF ADEQUATE VINE SIZE CAN BE MAINTAINED! The potential high reward of constant high node number pruning is matched by the potential high risk of overcropping lower capacity vines. 130 Node pruning is on the edge of the performance cliff. If the vines are pushed over the edge, yield and quality drops dramatically and vine recovery is costly.

If adequate vine size cannot be maintained, more conservative pruning is required.

Question 4: Can we control Concord Juice Quality?

Figure 4: The effect of crop load on juice soluble solids (A) and titratable acidity (B) from veraison to harvest in a single year. The effect of year on juice soluble solids (C) and titratable acidity (D) from veraison to harvest on a single treatment.

Figure 4A shows that increasing crop load decreases the rate of sugar accumulation in the fruit. A dense and inefficient canopy with a lot of shaded leaf area can further delay brix accumulation and delay harvest. At veraison, Concord fruit is typically between 7 and 7.5 oBrix and veraison marks the point at which there is a rapid build up of sugars in the fruit. Although there can be year to year differences, Figure 4B shows that there is no effect of crop load on the reduction of juice titratable acidity (TA) in any given year. From other research done at the lab, vines with larger crops tend to have lower TA but this is usually not statistically different.

In addition to the effect of crop load on sugar accumulation within any given year, there can be differences in fruit development from year to year within a given crop load. Figure 4C shows sugar accumulation from veraison to harvest in a cool (2000) and a warm (1999) year when the vine crop load is held constant (fixed node pruning). Note: the data is based on "days from veraison" not "day of the year." Veraison will be later in a cool year based on calendar date. At constant crop load there is little difference in sugar accumulation from year to year once veraison starts. On the other hand, the growing season heat accumulation can have a large effect on TA (Figure 4D). A cool year produces juice of higher acid because the berries enter veraison with higher acid concentrations. The literature indicates that temperatures during fruit development stages other than fruit ripening (ie. stages I and II) can affect final fruit acidity. Theoretically, temperatures from veraison to harvest will determine the rate of acid reduction post-veraison. Unusually hot weather in late August and early September could increase the rate of acid reduction. However, the weather information shows how we typically loose heat not gain heat during the same period. Therefore, temperatures in June and July may be more important in determining final TA in Lake Erie Concord fruit.

General Conclusions

Pruning
In both HRU and GDC treatments, less severe pruning leads to greater retained nodes, higher yield, and later relative harvest date. However, it is important to note that the relationship between retained nodes and yield is not linear. There is a point at which greater retained nodes does not translate into higher yields. Furthermore, excessive retained nodes leads to high shoot density and greater shaded leaf area which delays fruit and wood maturity at the same crop level.

Training
GDC vines have greater exposed leaf area/acre than HRU vines. We continue to see the advantage of higher yields and higher juice soluble solids (or earlier harvest date for the same crop level) in Geneva Double Curtain.

Thinning
Thinning minimal pruned vines increases the leaf area to fruit ratio and provides for an earlier harvest date. Thinning also appears to have a strengthening effect on the vines that is measured as greater ripe nodes of periderm and earlier leaf area development than standard minimal pruned vines.

Shoot Positioning
Shoot positioning appears to have a negative effect on the growth of both balanced pruned and minimal pruned vines. In balanced pruned vines, shoot positioned vines have less pruning weight in years of similar crop levels or they do not have more pruning weights in years of lower crop levels. In minimal pruned vines, shoot positioning increases canopy leaf layers (average of 4.5 leaf layers on one side of the canopy), increases canopy shading, and delays both wood and fruit maturity.