Abstract
Concord and Niagara grapevines in Fredonia, NY, were studied from 2001 to 2005 to evaluate the impact of cane length on yield, vine size, crop load, and juice soluble solids (JSS). Concord grapevines were manually pruned to 100 buds using three configurations: two-node spurs or five- or 10-node canes. Niagara grapes were manually pruned to 80 buds using three configurations: two-node spurs or five- or 10-node canes. In addition to standard viticulture measurements, yield was measured separately at each node position along the spur or cane in 2004 to 2005. There was a pattern in bud fruitfulness along the length of the cane, with the greatest yield originating from node positions three to six in Concord and two to six in Niagara; this did not change with pruning treatment. Since all vines were pruned to a constant total node number, five- and 10-node cane treatments had a greater proportion of more-fruitful buds and higher final yield than the two-node spurs in two out of five years. The higher yield on longer canes also led to greater crop load values indicative of overcropped vines, which decreased vine size. In contrast, two-node spur-pruning maintained balanced crop load values and adequate vine size. In the final two years of the study, there was no difference in final yield or JSS among pruning treatments because the longer cane treatments were starting the season with lower vine capacity.
Grape production in New York (NY) state is dominated by Concord (Vitis labruscana Bailey). In 2016, 47% of the United States Concord grape production east of the Continental Divide was in NY, and 71% of the NY grape production was Concord (NASS 2017). For Niagara, 35% percent of total production east of the Continental Divide was in NY, and 9% of the NY grape production was Niagara (NASS 2017).
Traditionally, Concord crop management is done primarily through dormant pruning (Jordan et al. 1980). The need to reduce production costs led to development of hedge and minimal pruning techniques (Pool et al. 1993). Though production costs may decrease with these practices, retaining more nodes per vine with a denser canopy structure is associated with increased crop levels and delayed juice soluble solids (JSS) accumulation (Morris and Cawthon 1981, Miller et al. 1993, Lakso et al. 1996, Miller and Howell 1998). This effect is more pronounced in regions such as the Northeastern US that have attenuated ripening windows due to climatic conditions (Zabadal et al. 2002, Bates and Walter-Peterson 2008).
Vine balance (crop load balance) is maintained by the quality (bud fruitfulness) and quantity (balanced pruning) of bud selection at pruning. Retained node quantity and quality is best obtained through manual balanced pruning. However, the production cost associated with manual balanced pruning makes adoption of new technologies such as mechanical pruning an inevitable choice, even though mechanized pruning decreases the retained node quality and quantity (Cawthon and Morris 1977, Pollack et al. 1977).
A study of the impact of bud density along a foot of row through pruning level and vine spacing showed that adjusting vine spacing to achieve 0.5 kg/m canopy dormant cane pruning weight (vine size) and retaining 44 fruiting buds/kg pruning weight (0.34 lbs pruning weight/ft canopy and 20 buds/lb pruning weight) achieved high yields and balanced crop load (Ravaz index between eight and 11) in NY-grown Concord (Shaulis 1980). Increasing retained node density by decreasing vine spacing and/or increasing retained nodes per unit pruning weight undesirably increased canopy density, internal canopy shading, and crop load (overcropping). In these studies, the vines were high cordon-trained (1.8 m) and long cane-pruned (cordon is still maintained with multiple canes); therefore, buds were also vertically distributed over approximately two-thirds of the cordon height.
Traditionally, long cane-pruning on a high cordon system is called Hudson River Umbrella. Current mechanized pruning systems in Concord tend to leave shorter canes while trying to maintain the same node number as long cane pruning; therefore, horizontal bud density is changed, and the effect requires investigating. Therefore, one major goal of this study was to determine the impact of machine pruning on yield, vine size, and JSS. To achieve this goal, two-node spur pruning was used to mimic mechanical pruning, and its effects were compared to five- and 10-node cane pruning systems, the common practice for the industry.
Concord and Niagara are traditionally cane-pruned because of low fruitfulness in base nodes (Pool et al. 1978). Buds produced at the shoot base usually have fewer inflorescences, making the yield potential for cane-pruned vines greater than that of spur-pruned vines with similar bud numbers (Vasconcelos et al. 2009). Partial yield compensation can occur via more flowers per inflorescence on spur-pruned vines (Huglin and Schneider 1998).
On longer canes, shoots that grow at the end of the cane are more vigorous due to apical dominance (Wolf 2008). Cane length is a key issue with the mechanization of Niagara, since some growers are reverting to umbrella training from high-wire cordon training. Hence, the effect of cane length on yield, vine size, and JSS for Niagara was also tested in this study.
Maintaining vine balance through quality and quantity of bud selection is expensive, creating a need for novel techniques to ensure vine balance requirements are met. Current mechanization protocols retain high node number and balance the crop with shoot or fruit thinning (Bates 2008, 2017). The main goal of this study was to investigate the effect of cane length on the yield, canopy development, and JSS of Concord and Niagara grapevines. This study also examined the dynamic impact of spur pruning on vine size and, eventually, vine capacity. The study tested the hypothesis that spur pruning will work on Concord and Niagara. The trial goal was to discover whether the short-term yield loss could be compensated with increased vine size and JSS accumulation in spur pruning.
Materials and Methods
Vineyard description
The experiment was conducted from 2001 to 2005 in Concord and Niagara blocks at the Cornell Vineyard Laboratory in Fredonia, NY (l42°21′N; 79°18′W). The experimental block had an elevation of 231 m asl with a slope of 1 to 2%, and a southeast aspect. The soil was well-drained Chenango gravel-loam with a surface soil pH of 5.5 and 2% organic matter.
The experimental blocks were in the Lake Erie American Viticultural Area, which is characterized by cool and humid conditions. The 30-year (1971 to 2000) weather data were recorded at the Fredonia Vineyard Laboratory site. The average 30-year growing degree days (GDD) accumulation (base of 10°C) from 1 April to 31 Oct was 1460 GDD. Average precipitation accumulation of 69 cm was recorded for the same time, and rainfall was evenly distributed throughout the growing season. Supplemental irrigation is not a common practice for commercial NY Concord production, so this experiment did not use supplemental irrigation. Long-term Julian phenology dates for Concord grapevines at the experimental site were recorded and average 30-year Julian phenology dates are reported (Table 1).
The seasonal GDD accumulations (1 April to 31 Oct, base 10°C) over the course of the experiment were 1597, 1662, 1418, 1561, and 1786 in 2001, 2002, 2003, 2004, and 2005, respectively (Figure 1). All five data-collection years except 2003 were warmer than the long-term average, with 2005 the warmest and 2003 the coolest and closest to the long-term average.
The seasonal precipitation accumulation (1 April to 31 Oct) over the course of the experiment was 54.5, 69.3, 59.4, 83.6, and 64.7 cm in 2001, 2002, 2003, 2004, and 2005, respectively (Figure 2). Seasonal precipitation was below-average in 2001, 2003, and 2005; close to average in 2002; and above-average in 2004.
The own-rooted Concord vines were 30-years-old and the own-rooted Niagara vines were 15-years-old at the beginning of the experiment. Both were planted at a row-by-vine spacing of 2.6 × 2.4 m with rows running east-west. Vines were cordon-trained to a trellis wire at 1.8 m. All vines were trained to a bilateral cordon and fruiting buds were retained on canes from horizontal cordons. In this study, a two-node cane was called a two-node spur; this should not be confused with the terminologies used for head-trained vines that are cane-pruned (Figure 3).
Commercial standards for floor, nutrient, pest, and disease management were adopted (Jordan et al. 1980). Pre- and post-emergence herbicides were used for no-till weed management to maintain a 1.2 m weed-free zone under the vines, and the row centers were treated with one glyphosate application at bloom. Around budbreak, a single application of ammonium nitrate fertilizer was surface-broadcast across the block at a rate of 56 kg/ha of actual N. The NY and PA pest management guidelines for grapes (Weigle 2006) were adopted for choosing and applying proper fungicides and insecticides.
Viticulture treatments
Concord vines were manually pruned to a fixed number of 100 nodes using three different configurations: 1) 50 two-node spurs; 2) 20 five-node canes; and 3) 10 ten-node canes (Figure 3). Niagara vines were manually pruned to a fixed number of 80 nodes using three different configurations: 1) 40 two-node spurs; 2) 16 five-node canes; and 3) eight ten-node canes. Niagara needs fewer buds (60 to 80) to get to the same crop potential due to its larger clusters and berries. These specific fixed bud numbers were based on previous pruning level trials in Concord and Niagara to give moderate to high commercial yields (Bates 2008). Cane length treatments for Concord and Niagara were imposed in a randomized complete block design. The Concord block had three pruning treatments across five replicate plots with three data-collection vines per plot. The Niagara block had three pruning treatments across four replicate plots with six data-collection vines per plot. The pruned spurs and canes were not tied back to the wire. The two experiments were analyzed separately.
Vegetative measurements
During the dormant season, all vines were rough-pruned to a greater number of buds (i.e., 120 nodes for Concord and 100 nodes for Niagara), and the one-year wood was collected and weighed. Vines were then adjusted to a final count of 100 fruiting buds for Concord and 80 fruiting buds for Niagara. Base buds, defined as buds in the axil of a bud scale or bract at the base of a green shoot (Pool et al. 1978), were not included in the count as fruiting buds. These base buds were counted at bud position 0.
Yield measurements
Fruit was harvested manually and weighed from individual count vines during commercial Concord harvest in NY. Just prior to manual harvesting, a berry sample was taken for JSS measurements by collecting the two apical berries off each medial cluster on a count vine. There were the same number of medial clusters on each vine. The berry sample was crushed manually using a hand crusher. JSS was measured with a hand-held, temperature-compensating Leica refractometer (model 10423; Leica, Inc.). The harvest dates were 21 Sept 2001, 28 Sept 2002, 3 Oct 2003, 16 Sept 2004, and 21 Sept 2005 (Table 1). In 2004 and 2005, fruit was harvested and weighed by node position on each count vine in the trial. By-node weight measurements included base buds for all treatments.
Data analysis
Statistical analysis was carried out using the JMP pro 13.1 software (SAS Institute, Inc.). The Concord and Niagara vines were analyzed separately. Data were subjected to one-way (cane length) analysis of variance (ANOVA), using the general linear model, and post-hoc mean comparisons were done using Tukey’s honest significant difference (HSD) test.
Results and Discussion
Cane length and yield of Concord grapevines
In two out of five years of the study, the 10-node, cane-pruned treatment retained buds with overall higher bud fruitfulness and yield. This demonstrates why cane pruning has been used in Concord production. However, the higher yield with the 10-node cane-pruned treatment also led to overcropping (high Ravaz index), which consequently reduced vine size and yield potential in the last two years of the study (Table 2). Cluster number per vine was used to measure yield potential. In contrast, the relatively lower crop on spur-pruned vines maintained balanced crop load values with no reduction in vine size and yield potential. In the last two years of the study, all vines had the same yield regardless of pruning style; however, the vine size had become lower with long-cane pruning.
By-node harvest data for Concord in 2004 and 2005 showed that in five- and 10-node cane-pruned treatments, the majority of fruit was produced on shoots in the middle of a cane, mainly due to differences in the distribution of buds on the cane and fruitfulness of the retained buds (total 100 buds). There was a clear pattern in bud fruitfulness along the cane length (Figure 4C and 4D). Bud position 2 yielded the same amount of fruit whether it was spur- or cane-pruned. Despite the same number of nodes in each treatment (100), the treatments produced a different distribution of buds at each node position (Figure 4A and 4B). The bud number per vine and average crop weight results reflect the field measurement records; therefore, it is still possible to have crop weight reports on bud position 3 in the two-node spur-pruned treatment due to human error in pruning. The greater yields in five- and 10-node cane-pruned treatments occurred because they contained a higher percentage of the more fruitful buds.
The three cane-length configurations influenced yield in three of five years: vines with longer canes had higher yields. Vines with two-node spurs and five- and 10-node canes produced an average of 13.5, 14.2, and 16.1 kg/vine, respectively. Cluster number/vine also increased with cane length (Table 3 and Figure 5). Vines pruned to two-node spurs and five- and 10-node canes produced, on average, 182.6, 188.8, and 211.8 clusters/vine, respectively. Vines with two-node spurs showed slight yield compensation through increased berry weight (3.4 g) over vines with five- and 10-node canes (3.1 g). However, two-node spurs also produced slightly fewer berries/cluster, and the difference in overall cluster weight between treatments was insignificant in four of five years (Table 3).
Cane length and yield of Niagara grapevines
By-node harvest data from Niagara vines showed a pattern similar to Concord. Despite having the same number of buds (80 total buds), the distribution of buds at a given node position was changed by the treatments (Figure 6A and 6B). Bud fruitfulness pattern did not change with pruning treatment in the Niagara vines (Figure 6C and 6D). The greater yield with the five- and 10-node cane-pruned treatments led to overcropping in the Niagara vines (Table 4 and Figure 7), which consequently reduced vine size and yield potential. In contrast, two-node spur-pruned Niagara vines had a relatively lower crop (Table 4 and Figure 7), leading to a balanced crop load and maintaining vine size and yield potential. In the last two years of the study, all Niagara vines had the same yield regardless of pruning style, but vine size was roughly reduced by half in the 10-node cane-pruning treatment in 2004 (Table 4).
Cane length had similar effects over the same five seasons in Niagara. On average, vines with two-node spurs and five- and 10-node canes produced 16.3, 16.8, and 18.8 kg/vine, respectively. Vines with longer canes also produced more clusters/vine, with two-node spurs and five- and 10 node canes averaging 134, 137.7, and 157.7 clusters/vine, respectively (Table 5). Unlike Concord, cane length did not affect berry number/cluster for Niagara vines in any year, and differences in berry weight were insignificant in four of five years (Table 5).
The higher yield in longer cane treatments in both Concord and Niagara can be attributed to bud fruitfulness. Although base buds were included in “by-node” weight measurements, since this was the same across all treatments, their impact can be ignored except for Niagara in 2005. Additionally, the leaf distribution and the leaf area to fruit weight ratio may also play a role in the greater yields associated with longer canes. Our results again confirm the role of bud position in determining fruitfulness, although a slight yield compensation occurs in two-node, spur-pruned Concord grapevines, mainly due to increased berry weight. However, the berry number per cluster was always lowest for the two-node, spur-pruned Concord.
JSS and vine size in Concord
The difference in yield among treatments had an effect on JSS and pruning weight (Table 2). This was most noticeable in 2003, when the crop produced by vines with 10-node canes was large (19.6 kg/vine) and JSS dropped to 13.8% and pruning weight to 0.76 kg. Overcropping was better controlled on vines with two-node spurs, which produced 20.3 t/ha and allowed the fruit to ripen to 15.5% JSS. On average, vines with two-node spurs and five- and 10-node canes allowed fruit to ripen to 16.4, 16.0, and 15.9% JSS with pruning weights of 1.3, 1.1, and 0.9 kg, respectively. The crop load results for both Concord and Niagara indicated that the two-node spurs were balanced (crop load range of eight to 11), while the 10-node canes were overcropped (Tables 2 and 4).
Overall, longer cane lengths significantly increased yield in only two of five years of the study, which, in heavy crop seasons, led to reduced vine size and low soluble solids concentration. In these seasons, vines with two-node spurs had smaller, but still acceptable, yields, and allowed JSS concentrations to reach closer to the desired 16%. In lower-yield seasons, all fruit ripened to at least 16% JSS.
JSS and vine size in Niagara
The difference in yield among treatments affected JSS in two of five years, and pruning weight in four of five years (Table 4). These effects were most noticeable in 2003, when the difference in yield between vines with two-node spurs and 10-node canes was 6.5 kg/vine. In this season, the 10-node vines had a pruning weight of 0.6 kg and JSS of 11.8%, while two-node spur-pruned vines had a JSS of 13.3% and pruning weight >1 kg. On average, vines with two-node spurs and five- and 10-node canes produced fruit with 13.7, 13.3, and 13.0% JSS, with pruning weights of 1, 0.8, and 0.6 kg, respectively.
In general, longer cane lengths on Niagara vines produced higher yields, which slightly reduced vine size and JSS. As with Concord, shorter canes better controlled overcropping in high-yield seasons and allowed fruit to reach higher JSS concentrations.
In the long run, overcropping reduces vine size and pruning weight (Bates 2008, Wolf 2008). A reduction in vine size has been attributed to reduced yield potential, inadequate fruit maturation, and reduced wood maturation (Shaulis and Oberle 1948, Pool et al. 1978). The results of this study show the effect of different cane length on pruning weights, with longer cane length linked to reduced pruning weights. The treatments modified the distribution of buds along the cane, leading to larger crop levels in longer cane treatments, thus reducing vine pruning weight.
Conclusion
An experiment using three pruning configurations, two-node spur pruning and five- and 10-node cane pruning, was tested on Concord and Niagara grapevines for five years. In Concord, the two-node spur-pruned vines had a higher JSS in one year out of five and a higher pruning weight in three out of five years, compared to cane-pruned vines. In Niagara, the two-node, spur-pruned vines had a higher JSS in two out of five years, and a greater pruning weight in four out of five years. However, the yield was slightly lower in two out of five years in two-node, spur-pruned vines due to fewer clusters per vine. These results suggest that two-node spur pruning can be used in Concord and Niagara vines without compromising fruit quality.
Acknowledgments
This research was funded by the Lake Erie Regional Grape Processor Fund and the New York Wine and Grape Foundation. The authors thank Mr. Rick Dunst, Ms. Paula Joy, and Ms. Madonna Martin for their help with data collection. They also thank Ms. Jackie Dresser for reviewing this manuscript.
- Received February 2018.
- Revision received April 2018.
- Accepted May 2018.
- Published online October 2018
- ©2018 by the American Society for Enology and Viticulture