Abstract
Four pruning methods were applied to Cynthiana (Vitis aestivalis Michx.) for four years (2002 to 2005): hand (balanced to 50+10), machine (box cut to 80 nodes), machine + hand (box cut to 110 nodes with hand prune to 80 nodes), and minimal pruning (no pruning). There were only minor differences in vine nutrition and in fruit and wine composition among the pruning methods. Minimal-pruned vines had high yields and less mature fruit in the first year, followed by low yields the second year, with yield stabilization by the third year. Wines produced were similar within year among pruning methods with the exception of wine from minimal-pruned vines in the first year. No sensory differences were found between wines from hand-pruned vines versus other methods in any year (wine from minimal-pruned 2002 excluded). After the first year, all pruning methods produced similar fruit and wine. In the final year, all pruning methods had comparable yields. However, the minimal-pruned vines averaged 38% more clusters over the last two years with a 5- to 10-day delayed harvest as compared with handpruned vines. The use of machine pruning either alone or in conjunction with hand pruning is a viable option for Cynthiana production in regions with a sufficient growing season length. Minimal pruning may also be an acceptable method but additional research is needed.
Dormant pruning by hand is a time-consuming, labor-intensive process used to achieve the desired balance of fruit yield and composition. Hand pruning is the second largest cost factor in vineyard operations behind hand harvesting. Machine farming of grapes reduces cost from 44 to 61% as compared with hand farming for Vitis vinifera grapes depending on training system (Thomsen and Morris 2007). Mechanical winter pruning alone has reduced pruning cost by at least 50% as compared with hand pruning in other studies (Poni et al. 2004). Shortages of labor and increased labor costs are increasing interest in complete vineyard mechanization systems in the United States (Morris 2007). In Australia, widespread adoption of mechanical pruning by hedging and the use of minimal-pruning systems followed the introduction of mechanical harvesters in the early 1970s (Clingeleffer 1993). It was estimated that by 2000 about 80% of Australian winegrape vineyards received some form of mechanical pruning, leading to substantial savings in production costs (Morris 2001).
Cultivars respond differently to pruning (Clingeleffer 1993, Jackson et al. 1984, Poni et al. 2004, Tassie and Freeman 1992) and the response of minimal-pruned vines can differ with growing region (Possingham 1996). Spur length and climate can affect bud fruitfulness in V. vinifera cultivars (Clingeleffer 1993, Jackson et al. 1984). Low-fruitful cultivars like Croatina respond very successfully to mechanical pruning followed by hand finishing (Poni et al. 2004), and conversion of these traditionally long-cane pruned vines to short-cane pruned provides balanced growth and ripening. Some French-American hybrid cultivars have high basal bud fruitfulness and 85% of the crop may originate from noncount shoots (Morris et al. 2004); machine or minimal pruning of this type of grape without hand follow up may require crop thinning (Reynolds and Wardle 2001).
A system involving no pruning, except mechanical skirting (summer trimming of shoots at ~60 cm above vineyard floor) to aid mechanical harvest has been termed minimal pruning. This system increases yield and produces numerous small clusters with delayed fruit maturity (Clingeleffer 1993, Jackson and Lombard 1993, Possingham 1996, Tassie and Freeman 1992). Clingeleffer (1996) concluded from 20 seasons of Australian minimal-pruning trials on V. vinifera that vines have the capacity, through balanced growth and self-regulation, to maintain their shape, productivity, and fruit quality.
Little information is available on factors that influence bud fruitfulness in Cynthiana (Vitis aestivalis Michx.). In a five-year experiment, it was found that pruning Cynthiana to either three or six node spurs did not affect yield, indicating no difference in bud fruitfulness for these spur lengths (J. Morris, unpublished data, 1993). This experiment also examined pruning severities of 30+10 and 60+10 (30 or 60 nodes retained for first 454 g of dormant prunings with an additional 10 nodes retained for each additional 454 g of dormant prunings). Grapes pruned to 60+10 had ~1 kg/vine greater yield with similar fruit composition as compared with the 30+10 treatment.
Cynthiana grapes require a long growing season, 165+ frost-free days, and generally do not adequately ripen at latitudes above 42°N. Unlike many French-American hybrid and some V. vinifera grapes, Cynthiana basal buds are not very fruitful. Cynthiana vines are less susceptible to spring freeze injury in the Arkansas region because of later budbreak than many cultivars. The practice of allowing extra nodes as a contingency against spring freeze with midseason fruit thinning (Morris 2001, 2007) is not desirable in Cynthiana because it is difficult to mechanically remove green fruit without vine injury (J. Morris, unpublished data, 2001).
The effects of machine and minimal pruning on fruit, juice, and wine composition have not been established for Cynthiana. The aims of this research were to compare and evaluate the effects of hand, machine, machine+hand, and minimal pruning on Cynthiana grapes and wines.
Materials and Methods
Vineyard.
Treatments were established in a 14-yr-old Cynthiana vineyard located at the Arkansas Agricultural Research and Extension Center, Fayetteville (lat. 36°10′N; long. 94°17′W) in a north-south row orientation. Vines were trained on a 1.8-m bilateral cordon without catch wires. Vine spacing was 1.8 m x 2.75 m with two drip emitters per plant for irrigation. Rainfall during the growing season was 47.7, 44.2, 65.6, and 41.9 cm for 2002 through 2005, respectively. Supplemental irrigation was provided so that the vines were not under water stress during low rainfall periods. In the year prior to the study, all vines were hand pruned for uniformity leaving ~60 nodes on primarily two to three bud spurs with some one-bud spurs for wood renewal. During the study, all vines except minimal-pruned vines were shoot positioned by hand in a downward direction two times during the growing season to facilitate pruning. Cynthiana vines have a moderately drooping growth habit with strong tendrils. Orienting the shoots downward facilitates pruning by reorienting shoots that run down the row and allows for better cutter-bar efficiency.
Bird netting, black polypropylene 19.0 x 22.2 mm mesh and 4.27 m wide (J/S Bird Netting, Jim’s Supply Co., Bakersfield, CA) was draped over the row and fixed under the irrigation line at veraison each year. The 0.20-mm diameter of the netting material likely did not cause shading but did restrain leaf movement. Although not measured, restraining the leaves likely reduced light penetration to the interior canopy on all treatments. Shoots were skirted on nonminimal treatment vines at 30 cm above the vineyard floor to facilitate application of bird netting. The hand- and minimal-pruned Cynthiana grapes required ~165 and 170 days, respectively, to reach maturity after budbreak.
Vineyard soil was a Captina silt loam (fine-silty, siliceous, active, mesic Typic Fragiudults) with pH 7.0. The rows had sod middles, and pre- and postemergent herbicides were used on a 1-m strip under the vines. The vineyard soil had average mineral levels of potassium (K), phosphorous (P), calcium (Ca), and magnesium (Mg) of 672, 1035, 5128, and 226 kg/ha, respectively. All soil minerals except Mg were two or more times greater than the levels deemed adequate for grapes in the midwest United States (Dami et al. 2005). Vine nutritional status was measured by petiole sampling at bloom and veraison. Nitrogen (N) and foliar Mg were the only fertilizers applied. Foliar Mg (3.6 kg/378 L MgSO4) was applied to all treatments in two or three applications per year based on observations of leaf color. Nitrogen (25 kg/ha) was split into two applications and applied under the emitters. The minimal-pruned vines received an additional 15 kg/ha N each year applied under the drip emitters; this additional nitrogen was provided based primarily on bloom time N levels and observed differences in leaf color and shoot growth.
Pruning methods.
Node numbers were selected based on previous experience with Cynthiana grapes. A 60+10 pruning level produced pruning weights and fruit composition similar to the 30+10 but with greater yield (J. Morris, unpublished data, 1993). In another Cynthiana study, 60 nodes were retained per vine (Main and Morris 2004). In the current study, a 50+10 pruning severity was used with a maximum bud number of 80. This moderate hand-pruning equates to 28 to 44 buds per meter of row and was used in an attempt to maintain a balance between vegetative growth and fruit production.
Mechanical pruning was in a box-cut style, in which vertical cuts are made on the sides of the cordon and horizontal cuts are made on the top and the bottom of the cordon. The horizontal cut below the cordon was made close to the cordon to leave short spurs with only basal buds. The top and side cuts were made to retain a specified bud number. The cutter bar distance from the cordon varied with year to achieve the desired bud number and to accommodate any movement of fruiting wood away from the cordon that may have occurred in later years.
Four pruning methods were applied in six replications of four-vine plots for four consecutive years (2002 to 2005). (1) Hand-pruned vines (hand) were balance pruned, on primarily three to four bud spurs with some one-bud spurs for wood renewal, to 50 + 10 with an 80-node upper limit. (2) Machine-pruned vines (machine) were box cut with a gasoline-powered hedge trimmer to retain 70 to 80 nodes. The box cut retained two to three node spurs on top and sides of the cordon; hand follow-up was used around post and trunks to cut canes that were missed by the machine. (3) Machine plus hand-pruned vines (machine + hand) were machine pruned to a box cut to retain 100 to 110 nodes with follow-up hand-pruning to 70 to 80 nodes. The box generally had four bud spurs on the top and sides. Hand follow-up consisted of spur thinning to achieve a better distribution of spurs along the cordon. (4) Minimal-pruned vines (minimal) were not touched during dormancy and skirted 60 cm above the ground at veraison. A separation was also made between adjacent vines at veraison to enable data collection on individual vines. The separation was accomplished by shoot positioning and did not involve cutting. These vines were otherwise untouched.
Yield and yield components.
Cluster weight, clusters per vine, and yield per vine were determined by counting clusters and weighing grapes from each vine at harvest. Berries per cluster were calculated as (cluster weight/berry weight). Dormant count nodes were recorded in 2002 and 2003. In 2004 and 2005 when primary shoots were about 10-cm long, count shoots originating from nodes retained at pruning and noncount shoots originating from basal, latent, and secondary buds were counted. The bud counts from the later two years provide more information since buds that developed were counted. No record was made of buds that did not push. Yield per count bud was calculated as the (yield per vine/count buds or count shoots per vine). Dormant pruning weights were recorded for hand-pruned vines. After the fourth year, the 1-yr-old wood was removed from all vines. Ravaz index (kg fruit/kg dormant pruning) for each vine was calculated (Ravaz 1903) to evaluate vine balance.
Fruit composition and analysis.
Berry samples were taken from each pruning method beginning when soluble solids were ~19% to determine harvest time. At harvest, a random 200-berry sample was collected from the four-vine plots for each pruning method and replication and held at −20°C. Frozen berries were recounted and weighed to determine berry weight. A hot-press method was used to determine fruit composition. This method expresses more of the compounds that are available during wine fermentation by extracting more color, acids, and minerals from the skins (Threlfall et al. 2006). For fruit composition, berries were thawed 18 hr at 3°C. The 200-berry sample was placed in 350-mL blender cup to which 50 μL of Scottzyme Pec 5L (Scott Laboratories, Petaluma, CA), a pectolytic enzyme, was added. The berries were then blended (Osterizer model 848-31N, Oster, Jarden, Rye, NY) for 5 sec at the lowest speed. The blended fruit was placed in a 250-mL beaker, covered with a watch glass, and heated in a water bath (90°C) until sample temperature reached 70°C. Samples were placed in a 3°C room and cooled until samples reached 45°C, then the sample was squeezed through cheesecloth. A 175-mL aliquot of juice was collected. The beaker of hot-pressed juice was covered with a watch glass and placed on the counter to cool to room temperature (21°C) prior to centrifugation for 10 min at 13,250 rcf. Laboratory analysis methods and equipment were as described elsewhere (Threlfall et al. 2006).
Wine production.
Wines were produced in 2002, 2003, and 2004. Fruit was obtained from representative vines from each pruning method. Grapes harvested from each pruning method were mixed and separated into two 40-kg lots. The clusters were destemmed in a Nuova Zambelli Manta destemmer/crusher (Saonara, Italy) with rollers removed and 1.7 g/100 kg of pectinolytic enzyme Lallz y me EXV (Lallemand, Mont real, Canada) was added. Must was sampled for prefermentation pH adjustments as the fruit exited the destemmer. The must was fermented in containers lined with food-grade polyethylene bags. The bags were loosely taped after the yeast Saccharomyces cerevisiae strain BRL97 (Lallemand) was added at the manufacturer’s recommended rate, and wines were fermented at 21°C. The fermentation cap was mixed two times per day. Before and during fermentation, pH was adjusted to pH 3.6 using tartaric acid to maintain wine quality. After 7 days of fermentation, the must was pressed in a 70-L Enrossi bladder press (Enoagricol Rossi, Calzolaro, Italy) and the wine was placed into 22.7-L glass bottles to settle overnight. The wine was then racked into 18.9-L glass bottles and inoculated with the manufacturer’s recommended rate of Oenococcus oeni MBR strain Lalvin 31 (Lallemand). After malolactic fermentation, wine pH was adjusted to 3.55 with tartaric acid, and the wines were cold stabilized and then filtered with a 1-μM filter. One hundred mg/L of sulfur dioxide (SO2) was added at bottling, and wines were transferred to 750-mL bottles closed with SupremeCorq (Kent, WA) 45-mm closures.
Standard methods for wine analysis were used (Iland et al. 2000). Red color was measured on wine adjusted to pH 3.6 with acetaldehyde added to remove the influence of SO2. Spectral measurements were made to determine red color due to copigmentation, anthocyanins, and polymeric pigments (Levengood and Boulton 2004). The measure “total red pigment” estimates the concentration of red pigments in wines by adjusting wines to low pH by dilution with 1 M HCl.
Sensory analysis.
Wines were evaluated by Vinquiry, Inc. (Napa, CA) using triangle tests. Due to cost associated with the sensory testing, only one wine replicate was used in the analysis. Three of the four pruning methods were compared in 2002, and all four methods were compared in 2003 and 2004. Wines had been in the bottle for approximately one year when evaluated.
All wines were served in clear, tulip-shaped wineglasses of 220-mL capacity that were coded with three-digit random number codes. A 25-mL sample was poured into each glass and then covered with a 57-mm diameter plastic petridish cover at least one hour before evaluation. The tests were conducted in a sensory room illuminated with fluorescent lighting. All wines were served at 18 to 21°C on tables with white surfaces. The panelists were isolated from one another by partitions. Panelists expectorated the wines and rinsed their mouths with distilled water after each test. Water and crackers were provided as palate cleansers. Sensory score sheets were prepared using FIZZ software (Biosystèmes, Couternon, France).
In 2002, wine from minimal-pruned vines was not included in the discrimination test because of obviously lighter color and body. Four trained, experienced wine judges evaluated three wines using triangle tests. For each of the three treatment comparison sessions, 24 triangles were evaluated. Each panelist evaluated six triangles per treatment comparison and evaluated six triangles at one sitting. They took two 15-min breaks: one after comparison of hand versus machine+hand, and another after comparison of machine+hand versus machine. All panelists evaluated wine from comparison of hand versus machine+hand first, and comparison of machine+hand versus machine last.
In 2003 and 2004, six trained, experienced wine judges evaluated four wines using triangle tests. Two sessions were conducted on separate days comparing wine from the pruning methods. In session one, comparisons were made between minimal versus hand, minimal versus machine+hand, and minimal versus machine. In session two, comparisons were made between machine versus hand, machine versus machine+hand, and machine+hand versus hand. At each session, judges evaluated four triangles for each of the three treatment comparisons. Panelists took a 5-min break between evaluations. Treatment comparisons were randomized among panelists. For each of the six treatment comparison sessions, 24 triangles were evaluated.
Experimental design and statistical analysis.
Grapes, juice, and wine.
The experiment was a completely randomized block with four pruning methods (hand, machine, machine+hand, and minimal) in four-vine plots with six replications over four seasons (2001 to 2005). Yield data was averaged for each four-vine plot before statistical application. Wines from the different pruning methods were made in duplicate for three years. There were significant differences in years that resulted in significant year x treatment interactions primarily related to the minimal pruning treatment in years one and two. Data were therefore analyzed and reported as pruning treatments by year. JMP software (version 6.0; SAS Institute, Cary, NC) was used for analysis of variance and Tukey’s honestly significant difference test at the p ≤ 0.05 level of significance was used to separate means of pruning treatments.
Sensory.
FIZZ sensory analysis management software was used for statistical analysis. It was assumed that less than 25% of the population can detect a difference in establishing the power of the sensory test.
Results
Overall, yield components were similar within year for hand, machine+hand, and machine vines (Table 1⇓). There was little difference in berry weight among pruning methods. The greatest differences in berry weight occurred in 2004 between the machine+hand and minimal vines. Minimal vines had lower berry weights than hand vines in 2002 and 2004. Minimal vines had lower cluster weight, more clusters per vine (except 2003), fewer berries per cluster, more count nodes per vine and lower yield per count node in each year than the hand vines. Minimal vines pushed more noncount shoots than vines from other pruning methods in both years, which is not surprising since there were more buds on the minimal vines. However, the average ratio of noncount shoots to count nodes was about 0.9 to 1 for hand, machine+hand, and machine vines while minimal vines had a ratio of 0.5 to 1. Yield per vine was higher on minimal vines as compared to hand vines in 2002 and 2004.
Pruning weights for hand vines were 0.92, 0.39, 0.55, and 0.98 kg/vine for 2002 through 2005, respectively. After the fourth year, the 1-yr-old wood was removed from all vines. The pruning weights were 0.98, 1.04, 0.99, and 0.70 kg/vine and the Ravaz indices were 12.1, 11.7, 12.3, and 17.0 for hand, machine + hand, machine, and minimal vines, respectively.
Vine nutritional status was measured in plant petioles at veraison. There were minor differences in petiole mineral content between pruning treatments within year. However, year-to-year patterns associated with pruning treatments were not apparent. The nutrient values ranges, in percent, across years and treatments were N (1.3 to 1.5), P (0.38 to 0.54), K (2.1 to 3.0), and Ca (1.6 to 2.5). These values were within or slightly higher than the normal range for midwestern grapes (Dami et al. 2005). The values for Mg (0.17 to 0.29%) were in the low to normal range. Minor elements manganese, iron, zinc, copper, and boron were also within the normal range for each element, with no apparent relationship to pruning method.
Juice.
The higher yield on minimal vines in 2002 resulted in lower soluble solids (21.3%) that year (Table 2⇓). Soluble solids were not different among pruning methods for other years. There were slight differences for juice pH, titratable acidity, and potassium among pruning methods. Juice organic acids were similar among pruning methods after the first year. Tartaric acid did not differ among pruning methods in any year and ranged from 6.5 to 7.2 mg/L in 2004 and from 8.0 to 8.5 mg/L in the other years. Total red pigments and total phenolics were similar within year except 2002 when the juice from the minimal treatment had lower values than juice from the other treatments.
Wine.
Wines were analyzed after bottling each year. Wine pH was adjusted with tartaric acid to pH 3.55 before cold stabilization on all treatments and therefore did not differ among treatments. Titratable acidity did not differ in 2003 (Table 3⇓), and differences in 2002 and 2004 were probably related to prebottling pH adjustment with tartaric acid rather than pruning method.
Wine color was measured after adjusting wine samples to pH 3.6 and removing the influence of sulfur dioxide with acetaldehyde (Table 3⇑). In 2002, total anthocyanin and total red pigments in the wine from minimal vines were lower than in the other treatments. A breakdown of anthocyanin composition into percentage-free, polymerized, and copigmented did not reveal any differences attributable to pruning method. The percentage free anthocyanin did not differ among treatments and was remarkably similar, with an average among treatments and years of 29.5% with a standard deviation of 3.4. The polymerized and copigmented anthocyanin was more variable with year. Total phenolics had similar values in all three years, whereas total anthocyanin and total red color were much higher in 2004 than in 2002 or 2003.
There were small differences in wine organic acids among pruning treatments, but the differences were usually not significant or recurrent with year and did not appear to be associated with pruning methods. The values obtained across years and treatments ranged from 0.5 to 0.7 mg/L for citric acid, 2.1 to 2.6 mg/L for tartaric acid, 0.29 to 0.49 mg/L for malic acid, and 3.1 to 4.7 mg/L for lactic acid. Ethanol among treatments and years ranged from 11 to 12.6% with the exception of the minimal treatment in 2002 (10.3%), reflecting the lower sugars obtained in the grapes that year.
The results of sensory discrimination testing of the wine using triangle tests are shown in Table 4⇓. Only wines from three pruning methods were evaluated in 2002 because wine from minimal vines obviously had less color. The panel of expert wine judges could only differentiate wine from machine vs. machine+hand wine in 2003. They could not differentiate any other wine pair in other years at the 5% level of significance.
Discussion
Vine canopy structure differed among pruning methods (Figure 1⇓). The hand, machine+hand, and machine vines were similar in appearance as compared with the minimal vines. The machine vines had a bushy appearance as compared with hand and machine+hand treatments and had some dead spurs. The machine treatment had a tendency for the growing area to expand away from the cordon each year. Shoots also accumulated near the head of the trunk on machine vines, which may necessitate occasional hand pruning to thin this area in order to maintain shoot distribution along the cordon. This clustering of shoots near the head may be variety specific for Cynthiana. A 1.8 m interrow vine spacing was used in this vineyard because it had been noted in a previous study that cordon size diminished at the terminal end of the cordon when vines were spaced at 2.4 m. This smaller cordon had reduced shoot growth and was especially evident on vines trained to a double-curtain system with a 4.8 m cordon (J. Morris, unpublished data, 1993).
Minimal vines had many dead, small diameter canes with short internodes, as is often found with minimal pruning (Clingeleffer 1993). Additional permanent cordon like wood (2- or more year-old wood with canes) also developed off the bilateral cordon on some vines. This wood can be seen above the right cordon in the minimal vine in Figure 1⇑, but was more prominent on some vines. In the initial year, the fruiting area ranged from above the cordon to floor skirting area. In later years, the fruiting area migrated closer to the cordon, perhaps due to shading as seen in other cultivars (Possingham 1996) or to poor shoot lignification. Many of the canes on the lower portion of the minimal canopy did not produce mature wood or did not push shoots from buds (Figure 1⇑).
Mineral nutrition did not appear to be a concern with any of the pruning methods. Mg levels in the petioles were marginal on all pruning methods even though foliar Mg sprays were applied and soil Mg was in the normal range. The additional N applied to the minimal vines each year maintained petiole nitrogen at a level similar to the other pruning methods.
Minimal pruning had the greatest impact on yield and fruit composition among the treatments. This was primarily during the first year when minimal vines were overcropped and fruit was not as mature when harvested. In the second year, the minimal vines had reduced yield in response to the high crop load from the previous year. This yield increase during the first year after implementation of minimal pruning is normal with yield stabilization occurring in subsequent years (Clingeleffer 1996, Jackson and Lombard 1993). Yield stabilized in minimal vines by the third year, and the vines produced fruit of similar composition to the other pruning methods. Many noncount shoots arose from secondary buds and were not usually fruitful, producing zero to one cluster. Minimal vines had three times as many shoots as hand vines. Although the minimal vines produced more noncount shoots than the other pruning methods, they produced proportionally fewer shoots per bud retained, indicating that basal and secondary buds were suppressed more by the minimal treatment than by the other treatments. The minimal vines were self-regulating shoot production. The shoots produced were also shorter and smaller in diameter than the shoots from the other pruning methods, which were consistent with minimal pruning in V. vinifera cultivars (Sommer and Clingeleffer 1993, Clingeleffer 1996, Jackson and Lombard 1993, Possingham 1996).
The pruning weights in the final year equate to about 0.56 kg/m of cordon for the hand and machine treatments and 0.39 kg/m for the minimal treatment. Smart and Robinson (1991) suggest that an ideal canopy has an optimal pruning weight of 0.3 to 0.6 kg/m. Values under or above would indicate that canopy was too sparse or too shaded. Minimal vines had significantly lower pruning weights and higher Ravaz indices than the other methods. However, the Ravaz index is not a good measure for minimal vines because these vines divert more carbohydrate to perennial wood than to canes and because terminal shoot growth fails to mature and lignify (Clingeleffer 1993). As previously noted, a permanent vine framework develops on minimally pruned vines with a reduction in 1-yr-old wood. Clingeleffer and Krake (1992) indicated that yield and 1-yr-old wood are poorly correlated and that old wood must be considered when measuring vine capacity in minimally pruned vines.
Minimal pruning reduced berry weight in two of the four years as compared to hand pruning. Minimal pruning also reduced berry weight in other cultivars (Sommer and Clingeleffer 1993, Clingeleffer 1996, Reynolds and Wardle 2001). Cluster weight was lower, and clusters per vine were higher in minimal vines than hand, consistent with other reports (Sommer and Clingeleffer 1993, Clingeleffer 1993, Jackson and Lombard 1993, Possingham 1996). Minimal vines had an average of 38% more clusters than hand vines in the last two years. There was a trend toward lower cluster weights in vines from machine versus hand pruning, with the machine vines having lower cluster weights in all years and significantly lower weights in two years.
The low yields in 2004 were possibly due to the poor wood lignification in 2003 that resulted in fewer nodes retained. Poor wood lignification affecting pruning weights has been previously noted in Cynthiana (Main and Morris 2004). The low pruning weight (0.39 kg/vine) for hand vines in 2003 was partially due to immature (dried, hollow) wood at pruning. The low pruning weight was reflected in the low bud numbers for the hand vines in 2004 as well as in yield. Based on number of buds retained, it is evident that the machine vines were too severely pruned in spring 2004.
Minimal vines had a yield of 11.6 kg/vine in year four with a 4-yr average of 11.2 kg/vine (19.9 t/ha equivalent). The 4-yr average yield for the hand vines was 10.1 kg/vine (17.95 t/ha equivalent). That was higher than the 7.7 kg/vine average obtained when vines were pruned to 60 buds (Main and Morris 2004). The 4-yr average yield increase for minimal vines versus hand vines was ~10%, much lower than the 25 to 50% increase reported for some V. vinifera cultivars (Clingeleffer 1993, Possingham 1996). Typical commercial yields of Cynthiana are in the 3.5 to 13 t/ha range, which is much lower than obtained in our vineyard. Therefore, it would appear that in grapegrowing areas with a sufficient growing season, 165 to 175 days, commercial yields could be increased while retaining good fruit composition by applying any of the pruning and management conditions used in this study. However, the vineyard location in this experiment provides a 30+ day foliated postharvest period, which may help in vine recovery by increasing carbohydrate reserves, thus perhaps allowing for ripening of a larger crop (Howell 2001).
Harvest was delayed on minimal vines 5 to 10 days each year as compared with the other pruning methods to reach similar fruit maturation, consistent with minimal pruning of other cultivars (Clingeleffer 1993, Jackson and Lombard 1993, Possingham 1996). Juice composition differences among pruning treatments were usually minor, were not recurrent with year, and therefore did not appear to be associated with pruning method. Fruit from minimal vines had lower soluble solids in 2002 despite a delay in harvest. That was not ref lected in increased acidity, but red color was reduced almost 2-fold as compared with other treatments. Red color in juice and wine was lower in 2002 and 2005 than in the other years; both these years had temperatures above 35°C during veraison or a temperature spike that occurred during or slightly before veraison. By contrast, the exceptional color year of 2004 had only a few days where the maximum temperature reached 33°C. Red color accumulation was decreased by warm temperatures during the growing season in some cultivars (Jackson and Lombard 1993) and decreased red color has been previously associated with warm veraison periods in Cynthiana (Main and Morris 2004).
Further testing on minimal pruning is needed in areas north of latitude 36° because of the shorter growing season. Possingham (1996) reported that minimal pruning in region I and II vineyards improved the exposure and microclimate of bunches but that winter skirting and side trimming may be required to adjust crop levels so the fruit matures before the end of the cooler growing season.
Conclusions
Cynthiana vines that were machine pruned with or without hand follow-up produced fruit yield, fruit composition, and wines that were similar to hand-pruned vines. Therefore, the use of machine pruning either alone or in conjunction with hand pruning is a viable option for Cynthiana production in regions with a long, 165+ days, growing season. Minimal pruning also produced fruit and wine composition similar to hand-pruned fruit after the vines stabilized in fruit production. However, additional research is required before minimal pruning can be recommended for Cynthiana grapes in all growing regions.
Footnotes
Acknowledgments: The authors thank Renee Threlfall and Janice Blevins for their assistance in collecting field samples and for laboratory analysis.
This project was partially funded by Viticultural Consortium-East through a subcontract with NYSAES, Cornell University, and the University of Arkansas Board of Trustees.
- Received July 2007.
- Revision received November 2007.
- Copyright © 2008 by the American Society for Enology and Viticulture