Abstract
‘Marquette’ is a cold-hardy hybrid grape cultivar that has received increased attention for its use in wine production in the upper midwestern and northeastern United States since it was released in 2006. However, ‘Marquette’ is an early budburst cultivar susceptible to spring freeze damage. We examined the influence of high wire bilateral flat cane (HWC) and four-arm Kniffin (4AK) training systems on young ‘Marquette’ performance during a year with spring freeze damage (2017) and the subsequent season without frost events (2018). In 2017, there were two consecutive spring frost events at the experimental site approximately 2 weeks after the vines reached 50% budburst, which damaged more than 70% of the shoots. The percentage of freeze-damaged shoots and the severity of freeze damage to green tissues did not differ between training systems, but 4AK vines had higher yield at harvest (5.16 kg/vine or 3.12 tons/acre) than HWC vines (3.45 kg/vine or 2.10 tons/acre) because of the greater number of buds retained at winter pruning. There was no freeze damage close to budburst in 2018, and the yield of 4AK vines was still higher (11.74 kg/vine or 7.08 tons/acre) than that of HWC vines (8.20 kg/vine or 4.98 tons/acre). In 2018, the Ravaz index (yield-to-pruning weight) values were lower for HWC vines (3.41) than for 4AK vines (5.39), but the training system did not consistently affect fruit composition in either vintage. Within the 4AK system, shoots that emerged from the lower cane had more freeze damage than those of the upper cane and produced lower crop yield and fruit with lower soluble sugars in both vintages. Our results suggest that ‘Marquette’ vines can be grown on a training system with high cropping potential, such as a divided canopy system or a single canopy, with a higher number of buds and shoots than that of our study. Among divided canopy systems, 4AK might not be the best option for vigorous ‘Marquette’ vines because, in addition to greater susceptibility to freeze damage, the lower cane of 4AK was highly shaded by the upper highly vegetative canopy, which might have caused its lower productivity and soluble sugars at harvest compared with those of the upper cane.
The release of cold-hardy wine grape cultivars has contributed to the expansion of wine production in the northeastern and upper midwestern United States, where more traditional, cold-tender grape cultivars (e.g., Vitis vinifera) could not survive low winter temperatures (Martinson et al. 2016; Schrader et al. 2019). Interspecific hybrid cultivars released from the University of Minnesota and other private breeders in the Midwest can withstand extremely low winter temperatures (e.g., −30 °C) because of the presence of Vitis riparia in their parentage (Hemstad and Luby 2000; Londo and Kovaleski 2019). Breeding efforts for cold-hardy non-V. vinifera cultivars also focus on achieving desirable enological and sensory properties because they have a distinct chemical composition (Hemstad and Luby 2005; Manns et al. 2013; Watrelot and Bouska 2022). Among cold-hardy hybrid cultivars, Marquette (Ravat 262 × MN1094) (Hemstad and Luby 2008) shows promise for use in high-quality wine production (Atucha et al. 2018; Norton et al. 2023; Pedneault et al. 2013; Slegers et al. 2015).
Marquette is a red-fruited and early ripening cultivar that was released in 2006 by the University of Minnesota; it is a complex hybrid with V. riparia and V. vinifera in its parentage, among other Vitis species (Hemstad and Luby 2008). ‘Marquette’ is currently grown across northeastern and midwestern United States and Ontario, Canada. In Canada, in 2019, it was added to the list of cultivars that can be used to produce Vintners Quality Alliance (VQA) wines (https://www.vqaontario.ca/). Despite positive accolades, this cultivar has several traits that might limit adoption by grape growers. ‘Marquette’ vines are prone to high vegetative growth (Schrader et al. 2020), which can require extra labor input for canopy management and lead to undercropping conditions with potential negative effects for fruit ripeness. At harvest, ‘Marquette’ fruit tend to have concurrently high concentrations of soluble solids (TSS) and titratable acidity (TA) (Atucha et al. 2018; Cheng et al. 2023). A high TSS concentration is a positive fruit trait, especially in regions with low heat accumulation and a short growing season, but high TA levels at harvest (above 10 g·L−1) (Bradshaw et al. 2018; Scharfetter et al. 2019) might require the adjustment of traditional winemaking practices (Watrelot and Bouska 2022). Previous work investigated best viticultural practices, from training systems to fruiting zone leaf removal, for managing highly vigorous ‘Marquette’ vines while improving grape and wine quality (Aipperspach et al. 2020; Cheng et al. 2023; Scharfetter et al. 2019; Wimmer et al. 2018).
Another trait of ‘Marquette’ is its low chilling requirement that promotes early de-acclimation and budburst and increases exposure of green tissues to spring freeze damage (Frioni et al. 2017; Schrader et al. 2019). The propensity of ‘Marquette’ for crop losses because of spring freeze damage should be considered when evaluating training systems, especially in frost-prone areas (Frioni et al. 2017; Martinson et al. 2016). Training systems that position green, tender shoots higher from the ground can reduce their exposure to sub-freezing temperature during a radiation frost event because cold air typically settles closer to the ground (Evans 2000). Training systems that bear more buds could be favored to compensate for potential freeze-induced loss of vegetative and reproductive tissues. Increasing the number of buds could also produce more balanced vines with a lower shoot growth rate and increased fruit-to-vegetative tissue ratio (Aipperspach et al. 2020).
The main goal of this study was to compare grape production and fruit maturity parameters of young ‘Marquette’ vines grown on two training systems across two years: 2017, when vines sustained spring freeze damage, and 2018, a vintage with no spring frost events near or after budburst. The two training systems selected for the study, high wire flat cane (HWC) and four-arm Kniffin (4AK), vary in the number of retained buds (Reynolds and Vanden Heuvel 2009). We hypothesized that the training system with more buds and shoots (4AK) produces more balanced vines with a higher fruit-to-vegetative tissue ratio without negative effects on fruit chemistry at harvest compared with that of vines trained on HWC. We also predicted that more shoots per vine (4AK) allows growers to maintain higher yield in the event of freeze damage. A second objective of our work was to assess whether freeze damage, fruit yield, and fruit composition differ systematically between shoots that develop from the lower and upper canes of the 4AK vines. Within the 4AK system, we expected that shoots positioned closer to the ground (lower cane) would have more freeze damage but also lower yield and less mature fruit regardless of vintage and freeze damage because of greater shading conditions of the fruiting zone.
Materials and methods
Experimental design
The experiment was conducted in 2017 and 2018 at the Penn State Russell E. Larsen Agricultural Research Center in Pennsylvania Furnace, PA, USA (lat. 40°71′N, long. 77°97′W). The experimental plot consisted of four north–south-oriented rows planted with 1-year old own-rooted ‘Marquette’ grapevines in May 2015. The vine spacing was 9 ft between rows and 8 ft between vines. Each row had 42 vines across 14 panels (a panel describes the distance between two consecutive posts) of three vines each. The first and last panel of each row were not included in the experimental plot. The vineyard soil is Murrill channery silt-loam (US Department of Agriculture–Natural Resources Conservation Service 2019); the interrow area was planted with hard fescue (Festuca brevipila) in 2015, whereas the soil underneath the vines (85-cm-wide strip) was maintained vegetation-free through repeated herbicide applications. Vines were maintained according to standard management practices for young hybrid grapevines in the eastern United States (Wolf 2008).
The experimental design was a randomized complete block design with eight blocks and two training systems: HWC and 4AK. Each experimental unit consisted of nine consecutive vines (three panels), with the inner seven vines used for data collection. Each block consisted of two training systems randomly assigned across two adjacent rows. Vines were assigned either to the HWC or to the 4AK training system during winter pruning in Mar 2017. In 2017, all grapevines were cane-pruned and trained to bilateral flat cane system, with the goal of filling the trellis length (space between vines) of the young vineyard; however, this was sometimes limited by the cane length. To more accurately compare bud phenological development, freeze damage, and other production parameters between the two training systems and across years, we decided to cane prune all vines in 2018 as well, including HWC vines. In the 4AK system, four or more canes are trained horizontally from the trunk, at least in its original conception (Wolf 2008). Therefore, establishing cordon-trained, spur-pruned vines was not an option for the HWC vines until the study ended. Therefore, if the vines were assigned to the HWC system, one cane (1-year-old stem) was tied to the trellis wire on each side of the trunk at a height of approximately 1.80 m from the ground. If the vines were assigned to the 4AK system, then two canes were tied on each side of the trunk: one to the upper trellis wire and the other to the lower trellis wire at approximately 6 ft and 4 ft from the ground, respectively. The canes tied to the upper and lower wires are referred to as upper and lower canes hereafter. We did not count the number of buds retained at pruning, but we counted the number of shoots in 2018. In 2018, the average numbers of shoots for the upper and lower canes of the 4AK trained vines were 28 and 22 shoots/vine, respectively, whereas HWC-trained vines had an average of 29 shoots/vine. During the season, vines were subjected to shoot positioning to promote downward vegetative growth and uniform foliage distribution. Leaves around the clusters were manually removed from the east side of the canopy close to bunch closure in 2018 only. We were not able to remove leaves in 2017 because of a labor shortage. In Mar 2019, when the study ended, HWC vines were cordon-trained and spur-pruned to increase the number of shoots per vine; during that season, the average was 46.6 shoots/vine.
Weather conditions
Rainfall, air temperature, wind speed, and solar radiation values were recorded at 15-min (2017) and 1-min (2018) intervals by an on-site weather station (Rainwise Inc., Trenton, ME, USA). Daily growing degree days (GDDs) (base 10 °C) were calculated from 1 Apr to 30 Sep as GDD = [(maximum daily temperature + minimum daily temperature)/2] − 10. Wireless temperature dataloggers (iButton Fob, Model DS1921G-F5#; accuracy, ±0.5 °C; Embedded Data Systems, Lawrenceburg, KY, USA) were positioned on the trellis wire to measure air temperature at the height of the canes every 2 min. For the HWC system, one temperature sensor was placed on the trellis wire for each experimental unit, whereas two sensors were placed in each 4AK experimental unit, with one on the upper trellis wire and the other on the lower trellis wire. Two consecutive below-freezing events occurred on 8 and 9 May 2017, when most of the buds had reached or passed budburst. The length of the frost events was calculated as duration (minutes) of air temperatures that were below freezing.
Phenology and freeze damage
Grapevine growth stages were assessed on 20 and 25 Apr and 4 May 2017 using four randomly selected vines for each experimental unit. Measurements of the same 20 buds per vine were conducted using the modified Eichhorn-Lorenz (E-L) system (Coombe 1995). Budburst was defined as E-L stage 5 (“rosette of leaf tips visible”). During the following year, phenology measurements were collected only until vines reached 50% budburst because the 2017 data indicated no training system effect on bud phenological development. Freeze damage was assessed as the percentage of freeze-damaged shoots and severity of green tissue and inflorescence damage. Data were collected on 16 May and again on 1 Jun 2017 using the same vines and buds used for phenology measurements. The severity of freeze damage was visually graded for each shoot using a scale from 1 to 5 modified from Centinari et al. (2016) [1 = no visible damage; 2 = slight damage to green tissue, no visible damage on inflorescence(s); 3 = moderate damage to green tissue and inflorescence(s); 4 = severe damage to green tissue and necrotic inflorescence(s); and 5 = complete kill]. The number of secondary shoots was also counted on the same vines and days.
Yield components, fruit composition, and pruning weight
Experimental vines were hand-harvested on 20 Sep 2017 and 13 Sep 2018. During both years, clusters were counted and weighed separately for each experimental unit using a hanging scale accurate to 0.02 kg (ES-22 electro; Brecknell, Fairmont, MN, USA). Clusters of the 4AK upper and lower canes were weighed and counted separately. Fruit mass with late-season rot (i.e., Botrytis cinerea, non-Botrytis fungal rots, sour rot) was collected and weighed separately. At harvest, five clusters were randomly collected from each HWC experimental unit, whereas two five-cluster samples were collected for the 4AK units, with one from the upper cane and the other from the lower cane. Samples were stored at –20 °C until the juice chemical analysis was performed.
A subsample of 200 berries was used to determine the average berry weight. The rest of the berry sample was thawed in a water bath at 60 °C, crushed, and strained through a cheese cloth for the juice chemistry analysis. Although freezing and heating the fruit might slightly alter the juice pH and TA, the consistency of procedures is considered more important to providing reliable results with low variability (Threlfall et al. 2006). The TSS were measured using a hand-held refractometer (Master, Atago USA, Inc., Bellevue, WA, USA). The pH was measured with a pH meter (Orion Star A111; Thermo Fisher Scientific, Waltham, MA, USA). The TA was measured using an auto-titrator (G20; Mettler, Toledo, OH, USA). A 10-mL juice sample was titrated with 0.1 N NaOH to an endpoint pH of 8.2.
During both years of the study, vines were cane-pruned during the dormant season, in Mar 2018 and 2019, to assess the treatments effect on the vegetative growth of the previous season. For each experimental unit, canes removed were weighed with a hanging scale accurate to 0.02 kg (ES-22 electro, Brecknell). The pruning weight was used as an indicator of vine vegetative growth in the previous season. The Ravaz index, a proxy for crop load, was calculated as the ratio of the yield at harvest and pruning weight collected during the following dormant season.
Data analysis
The data analysis was conducted using SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA). Data were subjected to an analysis of variance using Proc MIXED with year, training system (HWC, 4AK), and cane height (4AK lower cane vs. upper cane) as fixed effects and block as a random effect. Model assumptions were confirmed using the UNIVARIATE procedure. Proportional data and graded scale data were log-transformed, and analyses of the transformed data were performed. The training system and cane height effects on freeze damage, vine growth, production, and fruit composition were analyzed separately. To assess freeze damage severity, data were analyzed within each damage class separately. Differences between air temperatures of HWC, 4AK upper cane, and 4AK lower cane were evaluated with Tukey’s honestly significant difference pairwise comparison test in the case of F test significance. Data from the 2017 and 2018 seasons were not combined over the years because of significant treatment × year interactions for some of the parameters analyzed. In addition, we wanted to distinguish training system effects on vine growth, production, and fruit composition in the year with freeze damage from those in the year without frost events.
Results
Weather conditions
Overall, the 2017 growing season was cooler and drier than that during the following year (Fig. 1). The GDDs accumulated between 1 Apr and 30 Sep were 1436 in 2017 and 1677 in 2018, and the total rainfall measurements for the same period were 600 mm in 2017 and 932 mm in 2018. The lowest daily temperatures occurred in Jan 2018, when the minimum temperatures (Tmin) were –20.9 °C and –21.1 °C on 1 and 7 Jan, respectively (Fig. 1B). In 2017, air temperatures were below freezing on 8 May approximately between 3:00 AM and 5:00 AM, and again on 9 May, from approximately 3:50 AM to 6:30 AM (Fig. 2). During these times, the average wind speed was between 0.76 m·s−1 (8 May) and 0.20 m·s−1 (9 May). There were no sub-freezing events close to or after budburst in 2018 (Fig. 1B). The lowest temperatures during the frost events were measured in the lower part of the vineyard and were –2.5 °C on 8 May and –2 °C on 9 May (data not shown). The length of the frost events and Tmin did not vary between the 4AK upper cordon, 4AK lower cordon, and HWC cordon (8 May: Plength frost = 0.209 and PTmin = 0.801; 9 May: Plength frost = 0.456 and PTmin = 0.727). However, air temperatures tended to be lower at the height of the 4AK lower cordon, specifically between 3:00 AM and 4:30 AM on 8 May and between 4:00 AM and 4:45 AM as well as 5:45 AM and 6:15 AM on 9 May (Fig. 2).
Phenology and freeze damage
Grapevine phenological development did not differ by training system (HWC vs. 4AK) or cane height (4AK upper vs. lower cane) for any of the three dates in 2017 (data collected on 4 May are shown in Table 1). We estimated that 50% budburst was reached on 23 Apr 2017 and 2 May 2018. The grapevine growth stage averaged to E-L 11 “four leaves separated” for HWC and 4AK upper and lower canes 4 d before the first frost event 8 May 2017 (Table 1). Therefore, freeze damage intensity and severity were unaffected by the difference in shoot phenological development between training systems or cane position. Freeze damage assessment data reported in Table 1 and Fig. 3 were collected on 1 Jun, almost 4 weeks after the last frost, when the extent of damage could be clearly quantified, but trends were similar on 16 May.
Shoot growth stage [Eichhorn-Lorentz (E-L)], percentage of freeze-damaged shoots, and severity of green tissue damage for ‘Marquette’ vines trained on high wire cane (HWC) and four-arm Kniffin (4AK) and for shoots located on the 4AK upper and lower canes. Phenology and freeze damage data were collected on 4 May and 1 Jun 2017, respectively. The severity of green tissue damage was graded using a scale from 1 to 5 (1 = no visible damage; 2 = slight damage; 3 = moderate damage; 4 = severe damage, necrotic inflorescence; 5 = complete kill).
Overall, the percentage of freeze-damaged shoots and the severity of the green tissue and inflorescence damage did not differ between HWC and 4AK vines (Table 1). However, when comparing the severity of freeze damage within each class, HWC vines had higher percentages of shoots with “slight damage” (class 2) and “moderate damage” (class 3), but lower percentages of shoots with “severe damage” (class 4), including necrotic inflorescences, compared to those of 4AK vines (class 2: HWC = 15.9%, 4AK = 12.4%, P = 0.006; class 3: HWC = 30.5%, 4AK = 23.2%, P = 0.023, class 4: HWC = 29.7%, 4AK = 38.8%, P = 0.002) (Fig. 2A). The percentage of shoots assigned to class 5 “complete kill” was below 0.5% for HWC, and it was zero for 4AK; however, the percentage of shoots with highly or completely necrotic inflorescence was close to (HWC) or more than (4AK) 30% of the total shoots measured (class 4) (Fig. 2A). The percentage of nodes with secondary shoots was only 10% for both training systems.
Within the 4AK system, the lower cane had a greater percentage of freeze-damaged shoots and greater damage severity compared to those of the upper cane (Table 1). The percentage of shoots on the lower cordon with “severe damage” (class 4) on green tissues and inflorescences was almost double that of the upper cane (upper cane = 26.7%, lower cane = 50.9%, P < 0.001) (Fig. 2B). There were also fewer shoots in the lower cane with “no visible damage” (class 1: upper cane = 29.8%, lower cane = 21.3%, P = 0.001) or “slight damage” (class 2: upper cane = 17.2%, lower cane = 7.8%, P < 0.001).
Yield components, fruit composition, and pruning weight
Grapevines trained on 4AK had more clusters and greater yield than those on HWC in both years (Table 2). The yields of 4AK vines were 49.6% (2017) and 43.2% (2018) higher than those of HWC-trained vines. However, when yield was standardized by the number of shoots in 2018, the average of the HWC vines was 284.1 g/shoot and that of the 4AK vines was 237.4 g/shoot (Table 2). The overall lower yield/shoot of 4AK was mainly driven by the lower fruitfulness (yield/shoot) of the lower cane compared with that of the upper cane (188.7 g/shoot vs. 276.2 g/shoot, respectively), whereas the yield/shoot values of the HWC and 4AK upper cane were numerically similar. The training system effect on cluster weight was not consistent across years (Pyear × training system = 0.018); the cluster weight of 4AK vines was lower than that of HWC vines in 2018 only. In 2018, the average cluster weights of the upper 4AK and lower 4AK canes were 23.8% and 39.0% smaller, respectively, than the cluster weights of the HWC vines.
Yield components, pruning weight, and crop load (Ravaz index) of ‘Marquette’ vines trained on high wire cane (HWC) and four-arm Kniffin (4AK) and shoots located on the upper and lower cane of 4AK-trained vines at harvest 2017 and 2018.
In general, within the 4AK system, the yield (cane−1 and shoot−1), cluster weight, and number of clusters were consistently higher for the upper cane than for the lower cane (Table 2). In both years, the upper cane produced more than 60% of the total vine yield; in 2018 the average yield produced by each shoot was 46% higher for the upper can than for the lower cane. The numbers of clusters were 55.7% (2017) and 41.6% (2018) higher for the upper cane than for the lower cane. Clusters produced by the shoots on the upper cane were 19.3% (2017) and 24.9% (2018) heavier than those of the lower cane. Yields for both the upper and lower canes were more than double in 2018 compared with those of the previous vintage (Table 2).
In 2018, the yield per vine was more than double that of the previous vintage for both training systems (Table 2). Greater yield parameters in 2018 compared with those of the previous year were likely the results of the older age of the vines in addition to the lack of freeze damage. In 2018, the average numbers of clusters per vine were 53.4% and 70.6% higher for HWC and 4AK vines, respectively, compared with those of same training system in 2017. Similarly, the average cluster weight was 75.7% greater for HWC vines and 38.9% greater for 4AK vines compared to that of the same training system for the previous vintage. The 2018 season was wetter and the percentage of fruit mass with rot was greater for HWC than for 4AK vines, and there were no differences between 4AK lower and upper canes (Table 3). Pruning weight did not differ between training systems in either year, whereas the crop load (i.e., the Ravaz index) of 4AK vines was higher than that of HWC vines in 2018 only (Table 2).
Juice composition, percent of fruit with rot and berry weight of ‘Marquette’ vines trained on high wire cane (HWC) and four-arm Kniffin (4AK), and shoots located on the upper and lower canes of 4AK-trained vines at harvest 2017 and 2018.
Overall, there were no consistent differences in fruit composition between the two training systems (Table 3). The TSS was higher for HWC in 2017 only; however, the TA and pH did not vary between HWC-trained and 4AK-trained vines. The only consistent trend between years was the higher TSS of the fruit produced on the 4AK upper cane compared to that of the lower cane (Table 3).
Discussion
There is a growing interest in identifying best cultural practices for cold-hardy grapevine cultivars; however, our understanding of how these practices influence vine productivity under environmental constraints, such as spring freeze damage, is still limited. In this study, we assessed the influence of training systems on young ‘Marquette’ productivity and recovery after vines were exposed to sub-freezing temperatures for 2 consecutive days in early May 2017. In general, growing conditions at our site appeared to be well-suited for ‘Marquette’ production. Seasonal GDDs were well within the range reported by previous studies of ‘Marquette’ conducted across the midwestern and northeastern United States (Atucha et al. 2018; Martinson et al. 2016). Furthermore, grapevines did not show symptoms of winter cold damage on buds or vascular tissues despite temperatures reaching –21 °C in Jan 2018. However, the extreme cold-hardiness of ‘Marquette’ vines must be weighed against its early budburst and potential for spring freeze damage to green tissues, which occurred during one of the two years of the study.
Our data indicated that the percentage of freeze-damaged shoots and the severity of green tissue damage did not differ between HWC and 4AK vines, and that the crop level was low in both training systems. The frost events occurred approximately 15 to 16 d after 50% budburst, when the shoots had an average of “four leaves separated”; therefore, they were very sensitive to freeze damage (Centinari et al. 2016). Less than 1% of the shoots measured were killed by the frost, but more than half sustained severe damage on the inflorescences (ranked as “3” or “4”), which resulted in lower yield potential at harvest. The whole vineyard was exposed to freeze damage; therefore, it was not possible to compare yield components of freeze-damaged vines to those of vines with no frost damage. However, it is reasonable to assume that freeze damage was one main factor that explained the markedly lower yield at harvest 2017 (58% lower for HWC and 56% lower for 4AK) compared with the following vintage. The lower cluster weight and fewer clusters per vine in 2017 were likely, at least to some extent, a result of partially or fully freeze-damaged inflorescences. It is also important to note that the vines were still young when the freeze damage occurred during their third growing season and were not up to their full production potential yet. A study conducted in Wisconsin that included several cold-hardy hybrid grapevine cultivars reported that Marquette was the most consistent cultivar among those tested in terms of yield production during the first 6 years of vineyard establishment (Atucha et al. 2018). In their study, however, the yield of ‘Marquette’ vines during the third season after planting was approximately 23% lower than that of the following year.
The relative difference in yield between vintages was similar for the two training systems, but 4AK vines produced higher yield in both years. The yield of 4AK vines was maintained above 5.00 kg/vine, or 3.12 tons/acre , despite the freeze damage. Leaving a higher number of buds during winter pruning was confirmed to be an effective strategy for reducing crop and monetary losses in case of freeze damage. The 5-year old ‘Marquette’ vines grown on HWC in Michigan produced 3.30 kg of fruit/vine (1.8 tons/acre) despite severe freeze damage killing most primary buds (Frioni et al. 2017). In 2017, the average yield of 4-year-old HWC vines at our site was 3.45 kg/vine, which was very similar to what was reported for ‘Marquette’ in Michigan, where vines were 1 year older and planted with the same in-row spacing as that of our site. In that study, there were multiple frost events that ended approximately 1 month after budburst (Tmin = –4.6 °C). The fruit was mostly produced by secondary shoots that developed after the frost, suggesting that ‘Marquette’ can produce fruitful secondary shoots, which limits the spring freeze losses. In our study, the percentage of secondary shoots that developed after the frost events was much lower, approximately 10%, perhaps because the two frosts occurred when the shoots were already at stage E-L 11. Most primary shoots were not completely killed by subfreezing temperatures and, if the shoot tip was necrotic, often a lateral shoot emerged near the tip to resume shoot growth (Centinari M, personal communication). Nonetheless, our results confirmed that young ‘Marquette’ vines can produce crops of approximately 2.10 tons/acre (HWC) and 3.12 tons/acre (4AK) despite their susceptibility to spring freeze damage. Despite their young age, the yield of 4AK vines was closer to 3.52 tons/acre, which is considered economically viable for cold-hardy hybrid cultivars (Martinson et al. 2016).
Our results confirmed that a training system that positions shoots higher from the ground than a typical low cordon with vertically positioned shoots (VSP) might be a better choice for protecting cold-hardy hybrid vines against radiation frosts (Trought et al. 1999), which are more common than advective frosts in the eastern and midwestern United States (Poling 2008). Shoots that emerged from the lower cane were closer to the ground and had higher freeze damage. Air temperature tends to increase up to several meters from the ground during clear, calm nights; therefore, raising the shoots and leaves from the ground might reduce exposure to below-freezing temperatures (de Rességuier et al. 2023). In our study, air tended to be colder at the lower cane height for part of the frost event, although differences with the higher cordon were often below 0.5 °C and the trend was less clear on the second day. During a radiation frost, plant tissues are initially warmer than the sky; therefore, they radiate their own heat to the surrounding air, becoming colder than the sky (e.g., 1 to 2 °C colder) (Evans 2000). Although we did not measure plant tissue temperatures, we can speculate that they follow similar temperature trends of the surrounding air; therefore, shoots positioned closer to the ground might have been colder than those farther from the soil. Beyond higher susceptibility to freeze damage, a low-wire training system is not recommended for ‘Marquette’ because the vines tend to produce a lower crop when trained on VSP (Luby 2012; Martinson et al. 2016; Wimmer et al. 2018).
Shoots produced from the lower cane not only had greater freeze damage but also produced fewer and smaller clusters with lower TSS in both years, regardless of the frost. In our study, although not directly measured, it appeared that the upper part of the canopy was shading shoots and clusters of the lower cordon (Wolf 2008). Increased shading conditions might have decreased the leaf photosynthetic capacity of the lower cordon and consequently lowered sugar accumulation in the berries and reduced bud fruitfulness (i.e., number of clusters/shoot) (Keller 2020). Therefore, a horizontally divided canopy, such as the Geneva Double Curtain, could be a better option for highly vegetative ‘Marquette’ to achieve high yield while raising the shoots from the ground and improving fruit sunlight exposure (Reynolds and Vanden Heuvel 2009). Although limited to 1-year data, work performed in Minnesota suggested that ‘Marquette’ vines trained on the Geneva Double Curtain produced a higher yield than that of HWC and VSP-trained vines without penalizing wine quality (Luby 2012). An additional, but important, consideration when selecting a training system is the labor input required for managing the vines. We did not quantify labor input or estimate whether an increased labor cost would be offset by greater yield and economic return. However, based on our experience over the duration of the trial, 4AK vines needed more labor hours for pruning, shoot positioning, and harvest.
Vine vegetative growth was lowest in 2017, but it did not differ between training systems in either year. In addition to younger vine age (Atucha et al. 2018), freeze damage could explain lower pruning weight in 2017 because there was considerable damage to vegetative tissue and a temporal delay in growth. Regardless, vines exhibited high vegetative growth and low crop load ratios, which were always below the range suggested for other hybrid cultivars (Ravaz index of 8 to 12) (Bordelon et al. 2008). However, the experimental vines were still young, and our Ravaz index values aligned with those reported in other cool climate regions for young vineyards of Marquette (Scharfetter et al. 2019; Wimmer et al. 2018) and other highly vegetative hybrid cultivars (Vanden Heuvel et al. 2013). The tendency toward higher crop load ratios for the 4AK system confirmed that a training system with greater cropping potential should be considered for ‘Marquette’ to produce more balanced vines and potentially increase economic return for grape growers, similarly to what was reported by Wimmer et al. (2018).
Differences in the fruit composition between the two training systems were limited to lower TSS (1.6 °Brix) of 4AK grapes in one of the two years of the study. Differences between training systems were potentially driven by grapes of the 4AK lower cane, which had consistently lower TSS than that of the upper cane, as discussed previously. Fruit TSS and TA varied between vintages, and they were both numerically higher in the year with the frost events. Cooler seasonal conditions and the lack of fruiting zone leaf removal in 2017 might have contributed to lower berry temperatures decreasing the rate of malic acid degradation during fruit ripening (Keller 2020). Frioni et al. (2017) reported that ‘Marquette’ clusters that developed from secondary shoots had higher TA than those of primary shoots, but there were no difference in TSS. We cannot exclude that higher TA values measured in 2017 may have also been the result of a greater proportion of secondary shoots that developed after the frost events, although the number we were able to quantify was limited (10% of the total nodes).
The tendency of ‘Marquette’ to produce grapes with high TA in some vintages is not uncommon (Atucha et al. 2018; Bradshaw et al. 2018), and enological research is addressing this issue to mitigate its effect on wine quality (Martinson et al. 2016). From the perspective of vineyard management, increasing fruiting zone sunlight exposure and temperature can reduce fruit TA and improve other wine quality parameters (Cheng et al. 2023; Scharfetter et al. 2019). Improving the microclimate condition in the fruiting zone can also improve air movement and pesticide penetration for fungal disease control. Under our experimental conditions, ‘Marquette’ appeared to be prone to bunch rot, which affected 20% (4AK 2017) to 48% (HWC 2018) of the total fruit mass. Not surprisingly, bunch rot incidence was higher in the rainy 2018 season, with HWC vines having a greater percentage of fruit with rot in that vintage. Dividing the vegetation between upper and lower canes in the 4AK system might have created a less crowded fruiting zone and improved air flow and disease control (Reynolds and Vanden Heuvel 2009), although these parameters were not directly measured in this study.
In conclusion, our results collected in a young Marquette vineyard provide initial guidance regarding selecting a training system for this newer and promising cultivar that has benefited rural economies in cold climate regions of the United States (Atucha et al. 2018; Bradshaw et al. 2018). Although more long-term studies are needed, our work indicated that ‘Marquette’ can be grown on training systems with higher cropping potential. Leaving a higher number of buds could produce more balanced vines (greater fruit-to-vegetation ratio) and provide better insurance to growers against crop losses because of freeze damage. Based on our observation, 4AK vines should require greater labor input, and shading conditions around the lower cane might have compromised its productivity; therefore, other divided canopy systems might be a better option for highly vigorous hybrid cultivars such as Marquette. However, leaving more buds and switching from a cane to a spur-pruning system on high wire cordon-trained vines could be another option for increasing the number of shoots per vine and producing more balanced ‘Marquette’ vines on a single canopy system with a lower risk of freeze damage.
References cited
Aipperspach A, Hammond J, Hatterman-Valenti H. 2020. Utilizing pruning and leaf removal to optimize ripening of Vitis riparia-based ‘Frontenac Gris’ and ‘Marquette’ wine grapes in the Northern great plains. Horticulturae. 6(1):18. https://doi.org/10.3390/horticulturae6010018.
Atucha A, Hedtcke J, Workmaster BA. 2018. Evaluation of cold-climate interspecific hybrid wine grape cultivars for the upper Midwest. J Am Pom Soc. 72:80–93. http://www.pubhort.org/aps/72/v72_n2_a2.htm.
Bordelon BP, Skinkis PA, Howard PH. 2008. Impact of training system on vine performance and fruit composition of Traminette. Am J Enol Vitic. 59(1):39–46. https://doi.org/10.5344/ajev.2008.59.1.39.
Bradshaw TL, Foster JA, Kingsley-Richards SL, Berkett LP. 2018. Horticultural performance and juice quality of cold-climate grapes in Vermont USA. Europ J Hortic Sci. 83(1):42–48. https://doi.org/10.17660/eJHS.2018/83.1.6.
Centinari M, Smith MS, Londo JP. 2016. Assessment of freeze injury of grapevine green tissues in response to cultivars and a cryoprotectant product. HortScience. 51(7):856–860. https://doi.org/10.21273/HORTSCI.51.7.856.
Cheng Y, Gapinski AD, Buren L, Nonnecke GR, Watrelot AA. 2023. Impact of post-fruit set leaf removal on Marquette phenolic compounds during berry development and ripening. Am J Enol Vitic. 74(2):0740027. https://doi.org/10.5344/ajev.2023.22054.
Coombe BG. 1995. Growth stages of the grapevine: Adoption of a system for identifying grapevine growth stages. Aust J Grape Wine Res. 1(2):104–110. https://doi.org/10.1111/j.1755-0238.1995.tb00086.x.
de Rességuier L, Pieri P, Mary S, Pons R, Petitjean T, van Leeuwen C. 2023. Characterisation of the vertical temperature gradient in the canopy reveals increased trunk height to be a potential adaptation to climate change. OENO One. 57(1):41–53. https://doi.org/10.20870/oeno-one.2023.57.1.5365.
Evans RG. 2000. The art of protecting grapevines from low temperature injury. Proc ASEV 50th Anniversary Mtg. 60–72.
Frioni T, Green A, Emling JE, Zhuang S, Palliotti A, Sivilotti P, Falchi R, Sabbatini P. 2017. Impact of spring freeze on yield, vine performance and fruit quality of Vitis interspecific hybrid ‘Marquette’. Sci Hortic. 219:302–309. https://doi.org/10.1016/j.scienta.2017.03.026.
Keller M. 2020. The science of grapevines: Anatomy and Physiology (3rd ed). Academic Press, Cambridge, MA, USA.
Hemstad PR, Luby JJ. 2000. Utilization of Vitis riparia for the development of new wine varieties with resistance to disease and extreme cold. Acta Hortic. 528:487–496. https://doi.org/10.17660/ActaHortic.2000.528.70.
Hemstad PR, Luby JJ. 2005. ‘‘Marquette’’, a new wine grape, named in Minnesota. Wine East. 33(4):7–8.
Hemstad PR, Luby JJ. 2008. A grape plant named ‘Marquette’. University of Minnesota. (assignee). US Plant Patent 19,579 (Filed 13 Oct 2006, granted 16 Dec 2008).
Londo JP, Kovaleski AP. 2019. Deconstructing cold hardiness: Variation in supercooling ability and chilling requirements in the wild grapevine Vitis riparia. Aust J Grape Wine Res. 25(3):276–285. https://doi.org/10.1111/ajgw.12389.
Luby C. 2012. The effect of training system and yield on fruit quality of ‘‘Marquette’’ and ‘La Crescent’ wine grapes (Vitis spp.) in a Vermont vineyard. J Am Pom Soc. 66:34–38. https://www.pubhort.org/aps/66/v66_n1_a6.htm.
Manns DC, CoquardLenerz CTM, Mansfield AK. 2013. Impact of processing parameters on the phenolic profile of wines produced from hybrid red grapes Maréchal Foch, Corot Noir, and ‘Marquette’. J Food Sci. 78:C696–C702. https://doi.org/10.1111/1750-3841.12108.
Martinson TE, Mansfield AK, Luby JJ, Gartner WC, Dharmadhikari M, Domoto P, Fennell A. 2016. The Northern Grapes Project: Integrating viticulture, enology, and marketing of new cold-hardy wine grape cultivars in the Midwest and Northeast United States. Acta Hortic. 1115:3–12. https://doi.org/10.17660/ActaHortic.2016.1115.2.
Norton EL, Talbert JN, Sacks GL. 2023. Consumer hedonic testing and chemical analysis of Iowa wines made from five cold-hardy interspecific grape varieties (Vitis spp.) Am J Enol Vitic. 2023. 74(1):0740010. https://doi.org/10.5344/ajev.2022.22002c.
Pedneault K, Dorais M, Angers P. 2013. Flavor of cold-hardy grapes: Impact of berry maturity and environmental conditions. J Agric Food Chem. 61(44):10418–10438. https://doi.org/10.1021/jf402473u.
Poling EB. 2008. Spring cold injury to winegrapes and protection strategies and methods. HortScience. 43(6):1652–1662. https://doi.org/10.21273/HORTSCI.43.6.1652.
Reynolds AG, Vanden Heuvel JE. 2009. Influence of grapevine training systems on vine growth and fruit composition: A review. Am J Enol Vitic. 60(3):251–268. https://doi.org/10.5344/ajev.2009.60.3.251.
Scharfetter J, Workmaster BA, Atucha A. 2019. Preveraison leaf removal changes fruit zone microclimate and phenolics in cold climate interspecific hybrid grapes grown under cool climate conditions. Am J Enol Vitic. 70(3):297–307. https://doi.org/10.5344/ajev.2019.18052.
Schrader JA, Cochran D, Domoto PA, Nonnecke GR. 2019. Phenology and winter hardiness of cold-climate grape cultivars and advanced selections in Iowa climate. HortTechnology. 29(6):906–922. https://doi.org/10.21273/HORTTECH04475-19.
Schrader JA, Cochran D, Domoto PA, Nonnecke GR. 2020. Yield and berry composition of cold-climate grape cultivars and advanced selections in Iowa climate. HortTechnology. 30(2):193–203. https://doi.org/10.21273/HORTTECH04557-19.
Slegers A, Angers P, Ouellet É, Truchon T, Pedneault K. 2015. Volatile compounds from grape skin, juice and wine from five interspecific hybrid grape cultivars grown in Québec (Canada) for wine production. Molecules. 20(6):10980–11016. https://doi.org/10.3390/molecules200610980.
Threlfall R, Main G, Morris J. 2006. Effect of freezing grape berries and heating must samples on extraction of components and composition parameters of red wine grape varieties. Aust J Grape Wine Res. 12(2):161–169. https://doi.org/10.1111/j.1755-0238.2006.tb00056.x.
Trought MCT, Howell GS, Cherry N. 1999. Practical considerations for reducing frost damage in vineyards. Rpt NZ Winegrowers. 10 Oct 2020. https://researcharchive.lincoln.ac.nz/handle/10182/4236. [accessed 4 Jul 2024].
US Department of Agriculture–Natural Resources Conservation Service. 2019. Web soil survey. https://websoilsurvey.nrcs.usda.gov/app/. [accessed 4 Jul 2024].
Vanden Heuvel JE, Lerch SD, Lenerz CC, Meyers JM, Mansfield AK. 2013. Training system and vine spacing impact vine growth, yield, and fruit composition in a vigorous young ‘Noiret’ vineyard. HortTechnology. 23(4):505–510. https://doi.org/10.21273/HORTTECH.23.4.505.
Watrelot AA, Bouska L. 2022. Optimization of the ultrasound-assisted extraction of polyphenols from Aronia and grapes. Food Chem. 386:132703. https://doi.org/10.1016/j.foodchem.2022.132703.
Wimmer M, Workmaster BA, Atucha A. 2018. Training systems for cold climate interspecific hybrid grape cultivars in northern climate regions. HortTechnology. 28(2):202–211. https://doi.org/10.21273/HORTTECH03946-17.
Wolf T. 2008. Wine grape production guide for Eastern North America. Natural Resource, Agriculture, and Engineering Service, Ithaca, NY, USA.