Abstract
Canopy architecture, yield components, fruit composition, and vigor of Syrah were measured in response to four canopy management and three regulated deficit irrigation (RDI) treatments. The control consisted of dormant hand pruning to 22, two-node spurs (HP) with no further manipulation. Other treatments consisted of mechanically box pruning vines to 10 cm hedges and mechanically thinning the canopy to a density of 5 (CLL) or 7 (CLM) count shoots per 30 cm of row or mechanically box pruning to a 10 cm hedge with no shoot thinning (CLH). Control vines were irrigated to 70% of evapotranspiration (ETo) from fruit set until harvest (RDIC). Other vines either received 70% of full vine ETo until veraison and 50% of ETo (RDIL) thereafter or received 50% of ETo between fruit set and veraison (RDIE) and 70% thereafter. Mechanical shoot thinning (CLM) removed 25% of the total shoots, exposing 70,600 shoots·ha−1 with a distance of 4.6 cm between count shoots on the cordon, which translated to four leaf layers and 12.6 m2 leaf area. The combination of the CLM and RDIE decreased berry weight at harvest by 12% without decreasing yield compared to HP, resulting in 21.5 tons·ha−1 yield. A combination of CLM and RDIE was needed to achieve vine balance with a crop load of 9.9 kg yield/kg pruning weight and a leaf area to fruit ratio of 0.75 m2·kg−1. The study identified a canopy management method that can be used in combination with regulated deficit irrigation that reduces input costs through mechanization and enhances berry composition with a vine balance that provides sustainable production.
California accounts for 90% of U.S. winegrape production, of which 70% is from the San Joaquin Valley (SJV). However, SJV winegrapes account for only 25% of the total winegrape farm receipts in California. A majority of SJV grapegrowers are unable to apply the principles of ideal canopy management because profit margins are already low. In order to remain profitable, growers tend to retain too many nodes during dormant pruning, resulting in out-of-balance vines with too much fruit for the amount of leaf area. Unbalanced vines tend to develop large leaf canopies with a high water demand and an undesirable microclimate, leading to a lower proportion of fruitful buds in subsequent seasons and fruit of inferior quality and low price. The detrimental effects of out-of-balance vines that were overirrigated have been oft-quantified. Mechanized canopy management options that include the timing and severity of regulated deficit irrigation have not been addressed.
A key component of the efficient use of mechanical canopy management is to achieve balanced cropping together with a favorable canopy microclimate (Morris 2007, Smart and Robinson 1991). Balanced cropping aims to achieve equilibrium between vegetative and reproductive growth of the grapevine, and thus ensures sustainable vineyard production. Canopy management, whether applied by conventional or mechanical methods, includes the following practices: dormant pruning, shoot thinning, shoot positioning, cluster thinning, leaf removal, and hedging/skirting. The effects of various canopy management practices are varied depending on macroclimate, cultivar, and irrigation requirements of a vineyard. They may include a slowing down of vegetative growth (Smart 1985, 1988), a favorable light regime in the defined fruit zone (Dokoozlian and Kliewer 1995, Gladstone and Dokoozlian 2003), enhanced fruit composition (Kurtural et al. 2006, Petrie and Clingeleffer 2006, Smart 1988), and balanced vines for sustained commercial production (Morris 2007, Reynolds and Wardle 1993).
There seems to be agreement that regulated deficit irrigation (RDI) reduces vegetative growth (Shellie 2006), enhances ripening by increasing cluster light exposure and berry temperature (Arozarena 2002, Dokoozlian and Kliewer 1996, Spayd et al. 2002), and improves fruit and wine quality through an increase of skin/pulp ratio and concentrations of phenolic and aroma precursor compounds (Ferreyra et al. 2002, Kennedy et al. 2002, Matthews and Anderson 1988, Romero et al. 2010). There is increasing evidence that water stress uniquely initiates not only whole vine water responses via hormonal and stomatal control of transpiration and photosynthesis (Bota et. al. 2004, Ferreyra et al. 2002, Lawlor and Cornic 2002) but also specific biochemical events that influence synthesis of anthocyanins and flavor precursors in maturing fruit (Kennedy 2008, Kennedy et al. 2002). Common goals of implementing RDI are to maintain moderately severe water stress after fruit set (less than −1.2 MPa midday leaf water potential) and to arrest shoot growth to meet fruit quality objectives. Little is known concerning the long-term impact of water deficits on the overall growth and yield and capacity of vines in subsequent seasons.
While there have been numerous canopy management and deficit irrigation studies conducted on red winegrapes in California and the western United States, the majority have looked at these two factors separately and few have included mechanization of canopy management practices. The objective of this study was to identify mechanized canopy management techniques together with suitable RDI practices that have the potential to optimize fruit composition of Syrah vineyards in San Joaquin Valley.
Materials and Methods
This study was conducted from 2009 to 2010 at a commercial vineyard planted with Syrah/1103P (UC Davis clone 05) grapevines at 2.1 m × 3.4 m (vine × row) spacing. The research site was located in Merced County, CA (lat. 37.1°N; long. 120.2°W; 70 m absolute elevation) and planted in 1999 on a Borden Fine Sandy Loam soil, a fine-loamy, mixed active, thermic, Typic Haploxeralf (https://soilseries.sc.egov.usda.gov/OSD_Docs/B/BORDEN.html). The vines were trained to a bilateral cordon at 1.4 m with two foliage support wires at 1.7 m with a 0.2 m T-top trellis. The vineyard was drip-irrigated with pressure-compensating emitters spaced at 1.1 m, two emitters per vine delivering 2.3 L·h−1. The vines were supplied with 32 kg·ha−1 nitrogen over two years. Pests were managed using an integrated pest management program.
The experiment was a four (canopy management) × three (RDI) factorial with a split-plot design with four replicated blocks. Three rows of 380 vines each comprised a block and one guard row separated each block. The canopy management treatments were randomly applied as main plot to every 80 vines within each row, separated by 15 guard vines. The irrigation treatments were applied randomly as subplot to every three rows, separated by the guard row. There were 240 vines per experimental unit, of which 36 were sampled based on a grid pattern of every fourth vine, or 20 m apart.
Canopy management treatments.
Four canopy management treatments were applied. The control treatment consisting of hand pruning (HP) to 22 spurs with two nodes each per vine. The three mechanized canopy management treatments were initially machine box-pruned to a 100 mm hedge, using a 600 mm double, sprawl-pruner head (model 63700; Oxbo International, Kingsburg, CA) mounted onto an Oxbo 1210 single-row, tractor-mounted tool carrier. All mechanized shoot-thinning treatments were applied at Eichhorn–Lorenz scale stage 17 with a rotary-paddle shoot thinner equipped with an Oxbo 62731 rotary brush. The crop load low (CLL) and crop load middle (CLM) treatments were applied by mechanically shoot thinning to a target of five count shoots (borne from count-buds >5 mm distal to the base of the bearing surface) and seven count shoots per 30 cm of row, respectively. The crop load high (CLH) treatment received no shoot thinning. Every 14 days during the growing season the shoots of all treatments were vertically positioned in a downward direction to reduce intravine shading. There was no skirting of the canopy or leaf removal conducted in the fruit zone in either year.
Timing of RDI severity.
Three RDI levels were applied as splits within each main plot by imposing different irrigation practices to the respective drip line. Precipitation in winter and spring was insufficient to fill the soil profile. The root zone was irrigated beginning in the third week of April based on a crop coefficient (Kc) of 0.2 and 80% of reference crop (grass) evapotranspiration (ETo) obtained from the CIMIS station in Merced, CA. Irrigation was interrupted before bloom, and the soil was allowed to dry down until midday leaf water potential (leaf Ψ) was less than −1.0 MPa to control shoot growth. The RDI treatments were imposed after fruit set. The ETo and Kc (starting at 0.3 at fruit set to 0.7 at veraison and 0.5 postharvest) developed for fully irrigated grapevines in the SJV were used to calculate irrigation amounts based on vine water status (Williams 2001). The commercial control RDI treatment (RDIC) consisted of 70% of daily ETo when leaf Ψ was less than −1.2 MPa starting at fruit set. The early RDI stress treatment (RDIE) consisted of replacing 50% of ETo when leaf Ψ was less than −1.4 MPa from fruit set to veraison, but not thereafter. The late RDI stress treatment (RDIL) consisted of replacing 50% of ETo when leaf Ψ was less than −1.4 MPa from veraison to harvest, but not before or after.
Leaf water potential.
The leaf Ψ was assessed at the end of each weekly irrigation cycle between 1100 and 1600 hr for fully exposed, mature leaves of similar physiological stage that displayed no visible signs of damage and had been in full sunlight. There were 17 and 14 sampling dates occurring weekly between June and September in 2009 and 2010, respectively. Four leaves per vine from 36 vines per experimental unit were used to weekly estimate the leaf Ψ for each sampling date during the growing season as described elsewhere (Williams and Trout 2005). Each leaf sample was covered in a zip-top plastic bag and sealed immediately before excision at the petiole to suppress transpiration.
Shoot counts, canopy architecture, and leaf area.
Two weeks after budburst, total shoots per vine were counted by the addition of count shoots and noncount shoots. After the application of canopy management treatments, shoot density per unit of row length was measured as follows. The full length of vine canopy was divided into seven, 30 cm equidistant sections. Each 30 cm section was assigned a random number. A total of four random, 30 cm sections per canopy were measured to calculate count and noncount shoot numbers, and flowers were retained on count and noncount shoots. Shoot length, number of leaves per shoot, leaf size, and leaf area per shoot and per vine were determined at 50% veraison in each year using four shoots per vine, and leaf area of the vine was measured as described elsewhere (Keller et al. 2008). Indicators of canopy architecture measurements such as leafiness index, exposed shoots per hectare, distance between shoots, and leaf layer number were measured as described by Smart (1985).
Yield, fruit composition, and crop load assessment.
Fruit yield and cluster numbers for each treatment were measured at 23 Brix by hand-harvesting 36 vines from each experimental unit. Average cluster weight was calculated by dividing fruit yield per vine by the number of clusters harvested. Fruit composition was measured starting at 1000 GDD until harvest at nine and six dates in 2009 and 2010, and at harvest. On each date, a random 100 berry sample was collected from 36 vines from each experimental unit, placed in polyethylene bags, stored on ice, and analyzed within 24 hr. Before analysis, the 100 berry sample was weighed and average berry size was determined. The samples were then crushed by hand and the juice was placed in 100 mL beakers. A 5 mL portion of each sample was used to determine the percent total soluble solids (TSS) measured as Brix using a digital refractometer (Spec Scientific Ltd., Scottsdale, AZ) The juice pH was determined with a glass electrode and a pH meter (model AR15; Fisher Scientific, Pittsburgh, PA). The titratable acidity (TA) of each sample was determined by titrating to pH 8.2 with 0.1 N sodium hydroxide and expressed as grams per liter (Iland et al. 2004). Samples of 20 random berries at 23 Brix for each treatment at harvest were used for determination of anthocyanins, total iron reactive phenolics, and tannins. Berry skins were removed from the pulp by hand, rinsed with tap water followed by distilled water, and blotted dry with paper towels. The skins were then extracted in 30 mL 50% acetone solution in darkness for 24 hr. After removal of acetone by evaporation, anthocyanins, total iron reactive phenolics, and tannins were determined using the University of California assay (http://wineserver.ucdavis.edu/adams/tannin/totalassay.pdf). Crop load was calculated by dividing the yield per vine by the dormant pruning weight per vine and is expressed as kg·kg−1. Leaf area to fruit ratio was calculated by dividing the leaf area of each vine by the yield per vine and is expressed as m2·kg−1.
Statistical analyses.
The results were subjected to three-way (canopy management × regulated deficit irrigation × year) analysis of variance and F test and were considered significantly different (p < 0.0001). However, year and year × RDI failed Levene’s test due to difference in variance among years (Levene 1960). Data were analyzed by year, as a split-plot (canopy management as main plot, RDI as subplot), where replication × canopy management was the random variable, using the Type III tests of fixed effects with the MIXED procedure of SAS (version 9.1; SAS Institute, Inc., Cary, NC). Treatment means were separated by Tukey’s honestly significant difference test at p ≤ 0.05. Simple linear regression analyses using the REG procedure of SAS were used to conduct regression analyses where appropriate.
Results
Climate, irrigation regime, and vine water status.
The GDD accumulation was 2188 and 2021 in 2009 and 2010, respectively. There was a 167 GDD difference between 2009 and 2010, making 2010 cooler than average for the study area. The ETo was 6.7 mm·day −1 and 6.5 mm·day −1 between 300 and 1800 GDD in 2009 and 2010, respectively (Table 1). The vines under the RDIC treatment received 1.5 and 1.2 megaliters (ML) irrigation in 2009 and 2010, respectively. The vines under both the RDIE and RDIL treatments received 1.2 and 1.0 ML irrigation in 2009 and 2010, respectively.
Cumulative growing degree days, weekly evapotranspiration (ETo), megaliters (ML) irrigation applied per week, and midday leaf water potential (Ψ) of Syrah/1103 vines as affected by regulated deficit irrigation (RDI) treatments in 2009 and 2010.
The leaf Ψ of Syrah in both years responded to the RDI treatments during the prescribed intervals for RDIE and RDIL (Table 1). In 2009, the leaf Ψ of RDIE was retained at −1.32 MPa for six weeks, whereas the leaf Ψ of RDIL was retained at −1.4 MPa for seven weeks compared to the RDIC at −1.2 MPa. In 2010, the leaf Ψ of RDIE was retained at −1.4 MPa for six weeks, whereas the leaf Ψ of RDIL was retained at −1.4 MPa for six weeks compared to RDIC at −1.28 MPa for the duration of the experiment.
Effect of canopy management on retained shoot and cluster numbers.
In both years of the study, prior to application of the shoot-thinning treatments, vines that were mechanically pruned had 33% more count shoots compared to the hand pruned (HP) control (Table 2). In 2009, mechanically pruned vines had 33% more noncount and total shoots compared to HP. In 2010, compared to HP, mechanically pruned vines had 20% and 40% more noncount and total shoots respectively.
Effect of mechanical canopy management on average number of count, noncount, and total shoots and cluster numbers on shoots per 0.3 meters of row on Syrah/1103P grapevines in 2009 and 2010 (n = 48).
The mechanical shoot-thinning treatments affected the number of count, noncount, and total shoots and clusters retained on the shoots in both years (Table 2). In 2009, compared to HP, the number of count shoots was reduced by 45% and 36% with CLL and CLM treatments, respectively, but increased by 45% with CLH. A similar response was evident for noncount and total shoots in 2009, where CLL and CLM were most effective in controlling shoot numbers. The CLL and CLM treatments also provided the most control of clusters on count shoots in 2009, and CLL was the only treatment that provided control of number of clusters on noncount shoots and total shoots. In 2010, the numbers of count shoots and total shoots were almost halved for the HP treatment compared to 2009. As in the previous year, CLL provided the most control for number of count shoots when compared to CLH. CLL and CLM treatments provided the same amount of control of noncount and total shoots compared to CLH in 2010. The CLL treatment provided the most control for clusters on count, noncount, and total shoots in 2010.
Effect of canopy management and RDI on canopy architecture.
There was no interaction of canopy management and RDI treatments on canopy architecture in either year of the study (Table 3). Both canopy management and RDI treatments affected leafiness index in 2009. CLM and CLH reduced the leafiness index of Syrah compared to HP and CLL in 2009. Limiting irrigation application with RDIE between fruit set and veraison also reduced the leafiness index in 2009. There was no effect of canopy management or RDI on leafiness index in 2010.
Effect of mechanical canopy management and timing and severity of regulated deficit irrigation (RDI) on canopy architecture of Syrah/1103P grapevines at 50% veraison on 23 Jul 2009 and 5 Aug 2010 (n = 48).
In both years, the CLL and CLM treatments provided the most control on the number of exposed shoots per hectare (Table 3). In 2009, CLL and CLM reduced the number of exposed shoots per hectare by 68% and 57%, respectively, compared to CLH. In 2010, the number of exposed shoots per hectare was reduced by 28% and 20% by CLL and CLM treatments, respectively. The number of exposed shoots per hectare increased by 22% with the RDIL treatment compared to RDIC and RDIE in 2010.
The CLL and CLM treatments were more effective in controlling the distance between the shoots in both years of the study compared to CLH (Table 3). Likewise, CLL and CLM treatments reduced leaf layer numbers by 63% and 36% in 2009 and 2010, respectively, compared to CLH. Limiting irrigation amount between fruit set and veraison with RDIE also reduced leaf layer numbers by 20% compared to RDIC and RDIL in 2009. However, the RDI treatments did not affect leaf layer numbers in 2010. The CLL and CLM treatments had the most effect in controlling vine leaf area in both years of the study. The leaf area was reduced by 60% with CLL and CLM in 2009 compared to CLH or HP. In 2010, CLL and CLM reduced leaf area by 60% and 50%, respectively, when compared to CLH. The RDI treatments did not affect leaf area in either year of the study.
Effects of canopy management and RDI on yield components.
In 2009 there was no interaction of canopy management and RDI on berry weight, clusters per vine, or yield (Table 4). However, in 2009 average cluster weight and pruning weight per meter of row (vigor) were affected by the interaction. In 2010, the interaction of canopy management and RDI affected yield components.
Effects of mechanical canopy management and timing and severity of regulated deficit irrigation (RDI) on average yield components and pruning weight per meter of row of Syrah/1103P grapevines at 23 Brix in 2009 and 2010 (n = 48).
In 2009, CLH and CLM reduced berry weight by 5% and 10%, respectively, compared to HP. The RDIE treatment also reduced berry weight by 16% in 2009 (Table 4). In 2010, CLM under the RDIE regime and HP under the RDIE regime produced the smallest berries at harvest. Conversely in 2010, HP vines under the RDIC and RDIL regimes and CLL under the RDIL regime produced the largest berries at harvest.
In 2009, the CLH treatment produced the highest number of clusters harvested per vine (Table 4). The CLL and CLM treatments produced 47% and 16%, respectively, fewer clusters harvested compared to CLH. There was no effect of RDI on the number of clusters harvested. In 2010, the number of clusters harvested increased for mechanically pruned vines under the RDIC regime, regardless of canopy management treatment. The highest cluster numbers per vine were produced by HP under the RDIC regime. The lowest cluster numbers were seen with HP under the RDIC regime in 2010.
In 2009, the HP vines under the RDIC regime had the heaviest clusters at harvest, whereas the CLH vines under the RDIE regime had the lightest clusters. The average cluster weight at harvest of CLM vines under the RDIE regime was 46% less than HP vines under RDIC, but was 16% greater than CLH under RDIE (Table 4). In 2010, the same interaction response was evident for average cluster weight at harvest.
The CLL and CLM treatments reduced yield compared to CLH in 2009 (Table 4). The CLL treatments reduced yield by 42% and 33% compared to CLH and HP, respectively. The RDIE treatment also reduced yield by 27% compared to RDIC and RDIL in 2009. In 2010, vines that were mechanically pruned had higher yields under RDIC than under RDIE and RDIL. The one exception was CLL vines under the RDIL regime, which had a yield similar to the CLL and RDIC treatment combination because of berry compensation in these treatments.
In both years of the study, canopy management and RDI interacted to affect vigor (pruning weight per meter of row) of Syrah. In 2009, HP vines under the RDIC regime had the highest vigor, whereas CLM vines under the RDIC regime had the lowest (Table 4). Vine vigor decreased when the mechanical canopy management treatments were combined with RDIC. However, vine vigor increased, close to the value of HP under RDIC, when mechanical canopy management treatments were combined with RDIE. In 2010, CLM vines under the RDIE regime achieved the same vigor as HP vines under the RDIC regime. Vigor of vines treated with HP was reduced 25% under RDIE regime compared to CLM under RDIE, but not with HP under RDIL. As in the previous year, the RDIE irrigation method provided vigor control for hand-pruned Syrah. Retaining more shoots and clusters with CLM and CLH reduced vigor in 2010 as in 2009, but a similar trend was evident with an increase in vigor when CLM and CLH were irrigated with RDIE in 2010.
Effects of canopy management and RDI on fruit composition.
There was no interaction of canopy management and RDI on Brix, juice pH, or TA in 2009 or 2010. The time to reach harvest target of 23 Brix was affected by the canopy management treatments in 2009 and 2010. HP, CLL, and CLM reached 23 Brix four days earlier than CLH (Figure 1A). In 2010, HP and CLL reached 23 Brix five and seven days earlier than CLM and CLH, respectively (Figure 1B). The RDI treatments also affected the time to reach 23 Brix in 2009. Applying the RDIC treatment delayed harvest by five days, whereas the RDIL treatment delayed it by three days compared to RDIE (Figure 2A). In 2010, the RDIL treatment delayed harvest by six days, whereas the RDIE treatment delayed it by three days compared to RDIC (Figure 2B).
Effect of mechanical canopy management on Brix of Syrah/1103 grapevines in 2009 (A) and 2010 (B). Bars with *, **, and *** and different letters indicate statistical difference at p < 0.05, 0.01, and 0.001, respectively, according to Tukey’s HSD test.
Effect of regulated deficit irrigation (RDI) treatments on Brix of Syrah/1103 in 2009 (A) and 2010 (B). RDI control (RDIC): 70% ETo applied between fruit set and harvest; RDI early (RDIE): 50% ETo applied between fruit set and veraison, 70% ETo applied postveraison; RDI late (RDIL): 70% ETo applied between fruit set and veraison, 50% ETo applied between veraison and harvest, but not thereafter. Bars with *, **, and *** and different letters indicate statistical difference at p < 0.05, 0.01, and 0.001, respectively, according to Tukey’s HSD test.
Juice pH was affected by the canopy management treatments in both years. The HP treatment consistently had higher juice pH throughout the season compared to CLL, CLM, and CLH (Figure 3A, B). Juice pH of HP was at least 10% higher in both years when compared to CLH. In 2009, there were no effects of RDI treatments on juice pH (Figure 4A). Conversely, in 2010, the RDI treatments affected juice pH throughout the season and at harvest (Figure 4B). Juice pH of RDIC and RDIE was 5% higher than RDIC at harvest in 2010.
Effect of mechanical canopy management on juice pH of Syrah in 2009 (A) and 2010 (B). Bars with *, **, *** and different letters indicate statistical difference at p < 0.05, 0.01, and 0.001, respectively, according to Tukey’s HSD test.
Effect of RDI treatments on juice pH of Syrah/1103 in 2009 (A) and 2010 (B). RDI control (RDIC): 70% ETo applied between fruit set and harvest; RDI early (RDIE): 50% ETo applied between fruit set and veraison, 70% ETo applied postveraison; RDI late (RDIL): 70% ETo applied between fruit set and veraison, 50% ETo applied between veraison and harvest, but not thereafter. Bars with *, **, *** and different letters indicate statistical difference at p < 0.05, 0.01, and 0.001 according to Tukey’s HSD test respectively.
Canopy management treatments affected TA in both years. During the early part of the 2009 season, CLH TA was the highest of all canopy management treatments. As the season progressed, CLH TA declined and at harvest it was similar to HP and CLM, while CLL TA was the lowest (Figure 5A). In 2010, a similar trend was evident with CLH TA, which was higher than that of HP and CLL but not of CLM (Figure 5B). However, in 2010 at harvest, compared to 2009, HP TA was lower than that of the other treatments. RDI treatments also affected TA in both years. Early in the season, TA was 12% lower under RDIE compared to RDIC and RDIL. This trend was evident for five more weeks. However, at harvest, RDIE TA was similar if not higher than under RDIC and RDIL (Figure 6A). In 2010, earlier in the season, TA was higher under RDIE, compared to RDIC and RDIL (Figure 6B). However, as the season progressed, RDIE TA decreased and by harvest, although statistically significant, the differences were not great enough to be of viticultural significance.
Effect of mechanical canopy management on TA of Syrah/1103P grapevines in 2009 (A) and 2010 (B). Bars with *, **, *** and different letters indicate statistical difference at p < 0.05, 0.01, and 0.001, respectively, according to Tukey’s HSD test.
Effect of RDI treatments on TA of Syrah/1103 in 2009 (A) and 2010 (B). RDI control (RDIC): 70% ETo applied between fruit set and harvest; RDI early (RDIE): 50% ETo applied between fruit set and veraison, 70% ETo applied postveraison; RDI late (RDIL): 70% ETo applied between fruit set and veraison, 50% ETo applied between veraison and harvest, but not thereafter. Bars with *, **, *** and different letters indicate statistical difference at p < 0.05, 0.01, and 0.001, respectively, according to Tukey’s HSD test.
There was no interaction of canopy management and RDI on berry skin phenolics at harvest during either year (Table 5). However, in 2009 CLM, CLL, and CLH increased the total phenolics by 10% compared to HP. The RDIE regime increased total phenolics by 17% and 21% compared to RDIC and RDIL, respectively. CLM and CLH had the greatest effect on anthocyanin, with an increase of 18% compared to HP. The RDIE regime increased anthocyanin by 22% compared to RDIC and by 42% compared to RDIL. CLM increased tannins by 11% compared to HP. The RDI treatments did not affect tannins in 2009.
Effects of mechanical canopy management and timing and severity of regulated deficit irrigation (RDI) on average berry skin phenolics, anthocyanins, and tannins of Syrah/1103P grapevines during harvest at 23 Brix in 2009 and 2010 (n = 48).
In 2010, the response of total phenolics was similar to that of 2009 (Table 5). CLM increased phenolics by 18% compared to HP. The RDIE also increased total phenolics by 15, a similar amount of increase as in 2009. There was no effect of canopy management treatments on anthocyanin. However, as in 2009, RDIE improved anthocyanin by 8% in 2010. There was no effect of canopy management on tannins in 2010, whereas RDIE and RDIL increased tannins at harvest by 25%.
Effects of canopy management and RDI on crop load and leaf area to fruit ratio.
There was no interaction of canopy management and RDI on the crop load or leaf area to fruit ratio of Syrah during the study (Table 6). In 2009, crop load was affected both by canopy management and RDI treatments. Crop load decreased by 34% with CLL but increased by 10% and 25% with CLM and CLH, respectively, compared to HP, and decreased by 64% and 18% with RDIE and RDIL, respectively. In 2010, CLM and CLH increased the crop load by 33% and 58% compared to HP. Conversely, RDIE reduced the crop load by 34% compared to RDIC or RDIL in 2010.
Effects of mechanical canopy management and timing and severity of regulated deficit irrigation (RDI) on average crop load and leaf area to fruit ratio of Syrah/1103P grapevines in 2009 and 2010 (n = 48).
The leaf area to fruit ratio of Syrah was affected by the canopy management treatments during the study (Table 6). In 2009, the amount of leaf area required to ripen one unit of fruit was similar for HP and CLH treatments. However, the leaf area to fruit ratio of CLL and CLM was 30% and 50% lower, respectively, compared to HP and CLH. In 2010, there was a decrease in the leaf area to fruit ratio of 38% with CLL and CLM compared to HP and CLH. The RDI treatments did not affect the leaf area to fruit ratio.
Discussion
Canopy architecture.
Our results indicate that dormant pruning or RDI alone was not enough to control the shoot numbers and hence to manipulate canopy architecture and yield to achieve vine balance. There were no untoward weather events such as a late spring frost that would have limited the shoot numbers of HP or CLH vines. Shoot thinning therefore is advantageous because it maximizes sunlight penetration into the fruit zone (Gladstone and Dokoozlian 2003) and can be applied rapidly (Morris 2007). Retaining all the shoots with CLH increased the proportion of noncount shoots to 55% in both years. That, in turn, led to the positioning of count shoots at a distance of 2.6 cm apart, increasing the leaf layer number to 7.66 with 120,000 exposed shoots per hectare. These conditions are considered unfavorable for winegrape productions (Smart 1988). Alternatively, shoot thinning with either CLL or CLM reduced the contribution of noncount shoots to ~50% of the total shoots and positioned the count shoots farther along the cordon, ~5.7 cm for CLL and 4.5 cm for CLM, reducing the leaf layers to ~3.5 for both treatments while exposing 60,000 and 70,000 shoots per hectare, respectively. The precise reduction in shoot density by the mechanical application of shoot thinning with CLL or CLM gives the viticulturist the ability to manipulate the canopy architecture. Since more closely spaced shoots per unit row length result in greater leaf area and hence shade within the canopy (Smart 1985), the average main and lateral leaf area of a vigorous cultivar such as Syrah would increase. The wide row spacing (>3 m) and high shoot numbers used in SJV vineyards (>100,000 exposed shoots per hectare) further exacerbates the leafiness of the shoots retained on the vines. Although the leafiness index of Syrah shoots was fairly uniform in this study, during the initial year it was evident that the application of CLM was the most beneficial in reducing the leafiness index of shoots for the amount of shoots that were being exposed per hectare to sustain yield as reported elsewhere (Smart 1985). The reductions in leaf area with the application of CLL and CLM treatments compared to CLH were proportional to the increase in distance between shoots along the cordon and the decrease in the number of shoots per hectare. There was no interaction of canopy management and RDI treatments on the canopy architecture variables presented in this study. Although there was precise control of vine water status in both years of the study in regard to application of RDI treatments, only the RDIE treatment showed a reduction of leaf layers in year one as reported by Shellie (2006) and and the number of shoots exposed per hectare in year two as reported by Keller (2005).
Yield components.
Data from both years indicate that berry size at harvest was smaller with the mechanical canopy management treatments, which is mostly an artifact of the mechanical dormant pruning for these treatments (Zabadal et al. 2002). Limiting irrigation application to 50% of ETo between fruit set and veraison also reduced berry weight at harvest compared to RDIC in both years as reported elsewhere (Keller 2005, Romero et al. 2010). As the number of clusters increased with the decrease in amount of shoots removed by mechanical shoot thinning, the average berry size at harvest also decreased. The larger berry size seen with HP and RDIC treatments and their combination may not be as desirable, as larger berries would have a lower skin-to-berry volume ratio (Morris 2007). Conversely, the interaction presented in 2010 with CLH and RDIE would have the smallest berry size with the highest skin-to-berry ratio preferred by winemakers (Ojeda et al. 2002). The number of clusters harvested was affected by the number of count shoots retained by canopy management treatments, as it explained 62% (p < 0.0001) and 71% (p < 0.0001) of total variation in 2009 and 2010, respectively. There was also a strong negative relationship between distance between shoots and the number of clusters harvested, as it explained 54% (p < 0.0001) and 61% (p < 0.0001) of the total variation in 2009 and 2010, respectively. Cluster weight in both years was influenced by the interaction of the factors tested. Generally, the trend seen in cluster weight mimicked that of berry size, where a reduction in shoot thinning levels coupled with limiting of irrigation to 50% of ETo between fruit set and veraison resulted in smaller clusters. However, even with smaller berry and cluster sizes the yields of mechanical CLM and CLH treatments were at or above SJV economic thresholds.
There was a strong positive relationship between yield per hectare and the number of count shoots retained (r2 = 0.68, p < 0.0001), number of shoots exposed per hectare (r2 = 0.84, p < 0.0001), and clusters harvested per vine (r2 = 0.78, p < 0.0001). The precision in achieving similar yields in consecutive years was possible with the mechanical canopy management approach. One common drawback to previous mechanical canopy management was the decline in yield after an initial increase compared to conventionally managed vineyards. In this study, there was also a decline in yield with CLH from 30 t·ha−1 to 26 t·ha−1 and with HP from 26 t·ha−1 to 14.4 t·ha−1. This decline was explained by the strong positive relationship between leaf layer numbers and the yield seen in 2009 (r2 = 0.42, p < 0.0001). However, when yield exceeded 27 t·ha−1 (CLH irrigated with RDIC, HP irrigated with RDIC in 2009), the canopy leaf layers also exceeded 5.0, which in turn reduced return budburst and bloom in the following season and lowered yields. Recommended leaf layers from side-to-side of canopy is three to four in warm climate viticulture (Smart 1985), and this target was achieved with combinations of CLL or CLM with RDIE in both years of the study with stable yields. Limiting irrigation to 50% of ETo between fruit set and veraison in both years reduced yields as reported previously (Keller et al. 2008, Matthews and Anderson 1988, Ojedo et al. 2002, Romero et al. 2010).
The vigor (pruning weight per m cordon) control of Syrah was achieved by combinations of mechanical canopy management and RDI treatments. The concomitant decrease in vigor with decreased shoot thinning when irrigated with the RDIC was balanced by limiting irrigation between fruit set and harvest in both years in similar amounts. Values of 0.3 to 0.6 kg·m−1 pruning weight are generally considered to be in the optimal range (Kurtural et al. 2006, Smart and Robinson 1991) for cool-climate growing regions. However, our data is consistent with recent reports from warm-climate growing regions (Dokoozlian and Kliewer 1995, Kliewer and Dokoozlian 2005) where values up to 1.0 kg·m−1 for red winegrape cultivars are indeed capable of producing high-quality wines without loss of productivity. There was a strong negative relationship between yield and vigor of Syrah in both years (r2 = 0.84, p < 0.0001 in 2009; r2 = 0.78, p < 0.0001 in 2010). Reducing shoot density (hence leaf layers and yield) with the CLM and limiting irrigation application between fruit set and veraison with RDI resulted in vigor values slightly less than the optimum vigor value of 1.0 kg/m (Kliewer and Dokoozlian 2005).
Fruit composition.
Recent studies have confirmed significant effects, especially of radiation microclimate, on fruit composition (Smart 1985, Keller et al. 2008). Our results are consistent in demonstrating the deleterious effects of high leaf layer numbers on fruit composition. The higher leaf layer numbers achieved with CLH in both years delayed ripening based on the 23 Brix target of the wineries. Contrary to this finding is the Brix of the HP treatment in 2009 where a lower crop level was more effective in not depressing it. Contrary to previous reports, increased canopy leaf layers as seen in CLH in 2009 and 2010 did not increase TA (Smart and Robinson 1991).
Limiting irrigation to 50% of ETo between fruit set and veraison only influenced canopy architecture in the initial year to affect fruit composition, which elsewhere has been indicated to be due to the reduced canopy area (Romero et al. 2010). Contrary to that report, our findings indicate that this effect was due to the reduction in canopy leaf layers when measured at 50% veraison, not necessarily to reduced canopy leaf area. Decreasing the leaf Ψ with RDIL after veraison was not effective in controlling canopy leaf layer number or vine leaf area and therefore did not have much effect on fruit composition during stage III of grape berry growth. The lower Brix accumulation with RDIL in 2009 and 2010 is explained by lower carbon dioxide assimilation as a consequence of water stress (Romero et al. 2010).
In warm, semiarid growing regions such as the San Joaquin Valley, increased direct solar exposure due to reduced canopy leaf layers (r2 = 0.38, p < 0.0001 in 2009, r2 = 0.41, p < 0.0001 in 2010), as a result of shoot positioning farther along the cordon (r2 = 0.44, p < 0.0001 in 2009, r2 = 0.63, p < 0.0001) by mechanical shoot thinning, explained the increase in berry skin anthocyanins (Smart 1985, Smart 1988). The reduction in berry size had an indirect effect on berry skin phenolics (Morris 2007, Romero et al. 2010). The results presented here confirm previous reports where a reduction in berry size by mechanical dormant pruning and canopy management led to higher berry skin phenolics, anthocyanins, and tannins (Keller et al. 2008, Zabadal et al. 2002). Berry size decreased as the intensity of shoot removal decreased. However, any further decrease in berry size beyond the size achieved by CLM did not improve berry skin phenolics, anthocyanins, or tannins. Any further reduction in berry size achieved by CLH was coupled with increased shoot density (>7 count shoots per 30 cm of row), canopy leaf layers (>5.0), and yield (>27 t·ha−1) along with decreased berry skin phenolics, thereby negating the indirect effects of higher skin-to-pulp ratio desired for winemaking purposes.
Crop load and leaf area to fruit ratio.
There was a strong, positive relationship between crop load and leaf area to fruit ratio of Syrah in both years (r2 = 0.77, p < 0.0001 in 2009; r2 = 0.69, p < 0.0001 in 2010). Generally, vines with crop load values between 5.0 and 10.0 kg·kg−1 and leaf area to fruit ratio values between 0.8 and 1.2 m2·kg−1 are required for maximum levels of Brix, berry weight, and berry coloration indicating vine balance (Kliewer and Dokoozlian 2005). Concomitant reductions in berry weight due to increased shoot density and consequently crop load were reported (Naor et al. 2002, Reynolds and Wardle 1993) where similar crop load values were achieved by canopy management. The linear increase in yield with crop load values up to 9.9 kg·kg−1 and leaf area to fruit ratio values up to 0.75 m2·kg−1 indicated a lack of source limitation in the study presented here. However, decreases in the rate of yield above the aforementioned values were probably associated with limited availability of assimilates to retained clusters post-canopy management. Therefore, to achieve vine balance with a crop load value of 9.9 kg·kg−1 and a leaf area to fruit ratio of 0.75 m2·kg−1 for Syrah in the San Joaquin Valley, a combination of CLM (7 count shoots per m of row) and RDIE (limiting irrigation application to 50% of ETo between fruit set and veraison) was needed.
Conclusion
The methods presented here demonstrate the potential effects of mechanical canopy management and limited irrigation application at key phenological stages on Syrah grapevines grown in a warm region. Two types of responses to mechanical canopy management were shown: a direct response to dormant pruning and shoot density manipulation that was positive depending on the concomitant decrease of canopy leaf layers related to canopy shading affecting berry skin phenolics and an indirect but positive response to the effect of increased berry skin phenolics (a concentration effect) due to berry size reduction. Two types of responses to water deficit were also shown: a direct response to limiting irrigation early during berry growth where yield and berry size were reduced and a direct but negative response to restricting irrigation later during berry growth where Brix accumulation was reduced due to severe water stress. An indirect response was also shown to restricting water early during berry growth where a reduction in berry size led to an increase in berry skin phenolics (a concentration effect). The study identified a canopy management method that can be used in combination with RDI that reduces input costs through mechanization and enhances berry composition with a vine balance that provides sustainable production. The precise role of mechanical canopy manipulation and water restriction will be better understood once their collective influence on wine composition is examined.
Acknowledgments
Acknowledgments: This project was funded in part by the American Vineyard Foundation and the Bronco Wine Company Viticulture Chair Trust Funds.
The authors gratefully acknowledge the cooperation of Bronco Wine Company, Greg T. Berg of Oxbo International Corporation, and West Coast Grape Farming during the execution of this project.
- Received February 1, 2011.
- Revision received June 1, 2011.
- Accepted July 1, 2011.
- Published online December 1969
- © 2011 by the American Society for Enology and Viticulture
Literature Cited
Vol 62 Issue 4
Jump to section
Related Articles
Cited By...
More from this TOC section
Similar Articles
