Abstract
Most viticultural regions in the world annually sustain spring frost injury-associated yield loss. Researchers have developed sprayable products to prevent spring frost injury by delaying budbreak. The goal of this research was to evaluate the efficacy of novel products, including Amigo oil,® ProTone,® and FrostShield,® to delay budbreak and the effect of these products on vegetative and reproductive growth in grapevines. The early budbreaking cultivars La Crescent and Marquette were used in field experiments in 2018 and 2019 and in a growth chamber experiment in 2019. Treatments included a control (no spray), 10% (v/v) Amigo oil, 500 mg/L S-ABA ProTone, and FrostShield. In field experiments, FrostShield consistently delayed the budbreak of both cultivars for four to six days, while the other two products showed a very limited effect. In addition, FrostShield did not affect cultivars’ vegetative or reproductive growth. In the growth chamber experiment, FrostShield delayed the budbreak of two cultivars for more than two weeks, while the effect of the other two products was inconsistent between cultivars. Our results indicate that FrostShield outperformed Amigo oil and ProTone by consistently delaying budbreak without negatively affecting vegetative or reproductive growth, suggesting that FrostShield might be an effective product to prevent spring frost injury.
Grape production is threatened by various environmental constraints, including drought and extreme temperatures (Zabadal et al. 2007). It was estimated that nearly 5 to 15% of global grape production suffers from cold-related injuries (Evans 2000), and this threat will be exacerbated because of climate change (Molitor et al. 2014). Cold injury occurs in fall (leading to early leaf abscission), winter (leading to bud and/or vascular tissue injury), or spring (leading to injury of young shoots). Spring frost damage is the most widespread geographically regardless of whether the region is traditionally considered cold or not (Poling 2008, Molitor et al. 2014). In fact, devastating spring frost events have been reported in major wine regions worldwide, including Australia (Webb et al. 2018), China (Li and Bardají 2017), France (Sgubin et al. 2018), New Zealand (Trought et al. 1999), Poland (Lisek 2008), and the USA (Poling 2008). Spring frost damage can be associated with high crop loss because injury is mainly sustained by primary buds, which are the most fruitful (Zabadal et al. 2007). The severity and frequency of spring frost damage also depends on grape genotype. Vitis riparia and some V. riparia-associated interspecific cultivars are widely grown in eastern and midwestern USA for their adaptation to extreme low winter temperatures. However, several of these cultivars tend to break buds early in the spring, which increases their susceptibility to frost damage (Londo and Johnson 2014).
To mitigate spring frost damage, active and passive protection methods have been developed (Evans 2000). Most active methods involve the increase of vineyard ambient temperature during a frost event using expensive equipment such as wind machines, heaters, and helicopters. These methods are used in large commercial vineyards (>4 ha). However, these methods are not always economically feasible in small vineyards because of the high capital cost. Furthermore, these methods are not environmentally sound (Poling 2008). Passive methods aim to prevent or decrease the risk of frost damage through indirect approaches such as site selection, cultivar selection, and certain cultural practices (Trought et al. 1999, Poling 2008). Among cultural practices, chemical applications have also been used to delay budbreak, thus minimizing the threat of injury due to late spring frost events. For example, Amigo oil,® composed mainly of soybean oil, delayed budbreak of some Vitis hybrid cultivars without affecting yield or fruit composition (Dami and Beam 2004). Although the underlying mechanism remains unclear, it was proposed that Amigo oil delayed budbreak by lowering grapevine respiratory activity (Dami and Beam 2004). Follow-up studies were conducted in recent years using the same product in different cultivars. It was concluded that the response to Amigo oil was concentration- and cultivar-dependent, and budbreak delay was often inconsistent (Loseke et al. 2015, Centinari et al. 2018). Plant growth regulators, such as abscisic acid (ABA), also have the potential to delay budbreak. Although ABA application before leaf fall does not affect budbreak in the following year (Li and Dami 2016), recent research suggests that ABA, when applied exogenously on dormant buds, can be potentially incorporated in vineyard practices to delay budbreak. It was reported that grapevine dormancy is always associated with the shift of ABA metabolism pathway. ABA biosynthesis is activated when grapevines enter dormancy, while ABA catabolism is activated when grapevines are released from dormancy (Zheng et al. 2015, Noriega and Pérez 2017, Vergara et al. 2017, Kovaleski and Londo 2019). Some researchers applied ABA on single bud cuttings, potted plants, or field-grown grapevines during the dormant season. However, ABA showed an inconsistent effect on budbreak in those studies (Weaver et al. 1974, Hellman et al. 2006, Zheng et al. 2015, Kovaleski and Londo 2019). Thus, more research needs to be done to understand the underlying mechanism that leads to the variation of ABA’s efficacy.
In this study, we have developed and used a novel product, FrostShield® (proprietary product, The Ohio State University) with the aim of delaying budbreak. The overall goal of this study was to evaluate and compare the efficacy of Amigo oil, ProTone® (ABA; the main ingredient of ProTone), and FrostShield to delay budbreak of two interspecific grapevine cultivars, La Crescent and Marquette. The first objective was to determine the efficacy of these products to delay budbreak in the field and in a controlled environment. Evaluation in a controlled environment would reveal if treatments effectively delay budbreak without weather interference. Field evaluation would confirm whether the effects of these treatments on budbreak are significant in an actual vineyard setting. The second objective was to determine the effects of different field treatments on grapevine vegetative and reproductive growth and on fruit composition. This research contributes to our understanding of commercially available spray products to delay budbreak in grapevines and is useful for developing novel strategies to mitigate spring frost damage in vineyards.
Materials and Methods
Plant materials, vineyard site, and weather conditions
Two interspecific cultivars, La Crescent (Swenson’s St. Pe-pin × Muscat Hamburg; white fruit cultivar) and Marquette (MN 1094 × Ravat 262; black fruit cultivar) were used in this study. Both cultivars break buds early, making them susceptible to spring frost. Grapevines were planted in 2004 at The Ohio State University Horticulture Research Unit 2 vineyard (40°44′N; 81°54′W; elevation: 353 m asl; soil series: Riddles silt loam) in Wooster, Ohio (USDA Plant Hardiness Zone 6a [-23.3 to -20.6°C]; planthardiness.ars.usda.gov/PHZMWeb/). Grapevines were spaced 1.8 m × 3.0 m (vine × row) on east-west oriented rows. Grapevines were trained on a high (height = 1.8 m) bilateral cordon system and were maintained using standard vineyard management practices. All weather data were collected from the Ohio Agricultural Research and Development Center weather station (www.oardc.ohio-state.edu/weather1/). Daily maximum and minimum temperatures, daily precipitation, and cumulative growing degree days (GDD, daily mean temperature above 10°C) are reported from 1 Jan through 30 Sept (Figure 1).
Field experiments
The first field experiment was conducted during the 2018 growing season. One week prior to treatment application, a 1-m cordon (observation zone) was flagged on each grapevine to constitute the experimental unit and was prepruned to five-bud spurs. Within each observation zone, nine spurs, with a total of 45 buds, were retained for data collection of budbreak and vegetative and reproductive growth. For each cultivar, the experiment was conducted using a randomized complete block design, consisting of four blocks with four treatments totaling 16 grapevines. Treatments consisted of (1) a control (no spray); (2) Amigo oil at 10% (v/v) (Loveland Industries) composed of 9.3% soybean oil + 0.7% emulsifier; (3) ProTone, at 500 mg/L S-ABA, a registered growth regulator for grapes (Valent BioSciences Corporation) + 0.05% (v/v) Latron B -1956 (Dow AgroScience); and (4) FrostShield, a proprietary product (The Ohio State University). Based on previous research (Dami and Beam 2004), treatments were applied during ecodormancy, specifically after the satisfaction of chilling requirements and before the rapid accumulation of GDD. This timing also corresponds to a shorter period between treatment application and budbreak, thus resulting in minimum weathering of the applied products. In Experiment 1, treatments were applied on 19 March 2018 at 1300 hr when air temperature was 7.5°C, relative humidity was 53%, and wind speed was 2.7 m/sec. Chilling hours (number of hours between 0 and 7°C from 1 Sept 2017 to 18 March 2018) were 1237. GDD (from 1 Jan to 18 March 2018) were 16, and Eichhorn-Lorenz (E-L) stage was 1 (dormant buds) (Eichhorn and Lorenz 1977). Backpack sprayers (Smith sprayer, The Fountainhead Group, Inc.) were used to apply treatments until runoff on both sides of each vine. After data collection of budbreak on the 1-m cordon observation zone, all grapevines were pruned a second time (double pruning) to 48 to 50 buds/vine on 22 May 2018. Marquette and La Crescent were harvested on 18 and 19 Sept 2018, respectively. Yield components and fruit composition data of the two cultivars were collected and are included in Supplemental Tables 1 and 2.
The second experiment was conducted during the 2019 growing season. Experimental design, setting of observation zone, vineyard management, and the procedure of treatment application were the same as described in Experiment 1. Before treatment application, one spur per treated vine was randomly flagged to photograph budbreak progression. Treatments were applied on 18 March 2019 at 1300 hr when air temperature was 1.5°C, relative humidity was 83%, and wind speed was 0 m/sec. Prior to treatment application, chilling hours (from 1 Sept 2018 to 17 March 2019) were 1444, GDD (from 1 Jan to 17 March 2019) were 7, and E-L stage was 1. After data collection of budbreak on the 1-m cordon observation zone, all grapevines were pruned again to 48 to 50 buds/vine on 18 May 2019. Marquette and La Crescent were harvested on 4 and 24 Sept 2019, respectively.
Budbreak progression and maximum budbreak
Budbreak was determined when buds reached E-L stage 5, which is defined as “green shoots first clearly visible” (Eichhorn and Lorenz 1977). The budbreak progression of each treatment was monitored by counting open buds every two days from the start of budbreak (the day when the very first budbreak occurred among all the experimental units per cultivar) until no change of budbreak was observed. On each collection date, percent budbreak was computed by dividing the number of open buds by the total number of buds/m of cordon × 100. The maximum budbreak was also determined and consisted of the highest budbreak percentage with no change after successive observations.
Budbreak relative to control
The date when 50% of buds opened in the observation zone of each treated vine was re corded. Budbreak relative to the control for each treatment was calculated by subtracting the mean Julian day when 50% budbreak occurred in the control treatment from the Julian day when 50% budbreak occurred in each of the treated vines. A positive value of budbreak relative to the control indicates that 50% budbreak occurred on a later date than the control, thus a delay of budbreak. On the contrary, a negative value of budbreak relative to the control indicates that 50% budbreak occurred on an earlier date than the control, thus an advance of budbreak.
Grapevine phenology after budbreak
In both years and after budbreak, grapevine development was monitored until early July using the E-L system (Eichhorn and Lorenz 1977), where higher E-L values indicate more advanced phenology. During assessment, each experimental grapevine was assigned with an appropriate E-L stage value based on the average of all shoots. The E-L stage values of each experimental grapevine was subjected to data analysis.
Vegetative and reproductive growth
The effects of different treatments on grapevine vegetative and reproductive growth were determined by recording grapevine shoot number and cluster number, respectively. Shoot and cluster number/m of cordon (observation zone) of each grapevine were counted after fruit set.
Yield components and fruit composition
In two field experiments, yield components and fruit composition data of the two cultivars were collected and are included in Supplemental Tables 1 to 4.
Growth chamber experiment
A third experiment was conducted in a controlled environment to evaluate the effect of Amigo oil, ProTone, and FrostShield using grape cuttings. Solutions of the three products were prepared based on the formulas described in Experiment 1. On 18 Jan 2019 (E-L stage 1), canes with eight buds (node positions 3 to 10) were collected from the dormant Marquette and La Crescent vines used in Experiments 1 and 2. Each of the four treatments were assigned to eight to 10 canes and the tips of all canes were sealed using Parafilm (Sigma Aldrich) to eliminate the direct penetration of solutions into the vascular system. Hand sprayers were used to apply treatments until runoff; then sprayed canes were dried at room temperature for two hours. Canes were then subject to a budbreak assay based on the method described in Zhang et al. (2011). Briefly, treated canes were excised into single bud cuttings and were inserted into water-saturated foam medium (Smithers-Oasis) in plastic trays (T.O. Plastics). Four treatments were randomized in each tray with 18 cuttings per replicate. Each tray was considered as a block for data analysis, and four trays (n = 4) were used for each cultivar. Thus, for each cultivar the experiment was conducted as a randomized complete block design with four treatments and four replications. Trays were subsequently placed in a growth chamber (Conviron) with the following environmental settings: 14 hr photoperiod with 220 μmol/m2/sec photosynthetically active radiation, 23°C, and 80% relative humidity. Photographs were taken daily to record the progression of budbreak.
Budbreak progression, maximum budbreak, and budbreak relative to control
Single bud cuttings were monitored daily for budbreak (E-L stage 5) at 1100 hr from 18 Jan to 8 March 2019, corresponding to one day to 50 days in forcing, respectively. The recording of budbreak progression, the determination of maximum budbreak, and budbreak relative to the control were the same as described in Experiments 1 and 2.
Statistical analysis
The data of budbreak progression (Experiments 1, 2, and 3) and grapevine phenology (Experiments 1 and 2) after budbreak were subjected to repeated measures analysis of variance (RMANOVA), with block included in the model, to determine treatments effect. When RMANOVA was significant, Fisher’s least significant difference (LSD) test was used to compare means among different treatments at each time point at 95% confidence. The data of budbreak relative to control in two field experiments (Experiments 1 and 2) were subjected to two-way ANOVA with treatment and year included in the model, and LSD was used as post-hoc test. All the other data, including budbreak relative to control in Experiment 3, vegetative and reproductive growth, yield components, and fruit composition were subjected to one-way ANOVA with LSD used as post-hoc test. Data analysis was conducted using R Studio (version 1.1.456). All tests were considered significant when p ≤ 0.05.
Results
Budbreak progression and maximum budbreak
Experiment 1
In both cultivars, different treatments led to distinguishable progressions of budbreak during our observation from 30 April to 12 May 2018 (Figure 2). RMANOVA revealed a significant effect (p < 0.001) of treatment on budbreak in both cultivars. In La Crescent, the control, ProTone-, and Amigo oil-treated grapevines initiated budbreak on 2 May 2018 (Figure 2A). Two days later (4 May 2018), more than half of control and ProTone-treated buds were open, while Amigo oil- and FrostShield-treated buds had only 17 and 7% budbreak, respectively (Figure 2A). Between 6 and 10 May 2018, budbreak of control, ProTone-treated, and Amigo oil-treated grapevines surpassed 90%. However, FrostShield-treated grapevines maintained the lowest budbreak at each observation during this period (Figure 2A). By 12 May 2018, 90% of buds in all four treatments were open, with no further increase in budbreak (Figure 2A).
In Marquette, the control, ProTone-, and Amigo oil-treated grapevines initiated budbreak on 2 May 2018, but ProTone treatment had a significantly higher budbreak than that in the other treatments (Figure 2B). On 4 May 2018, >50% of buds in the control and ProTone-treated grapevines were open, while the budbreak of Amigo oil- and FrostShield-treated grapevines remained low at 25% and 11%, respectively (Figure 2B). Between 4 and 8 May 2018, lower budbreak was observed in FrostShield-treated grapevines (Figure 2B). By 10 May 2018, budbreak gradually increased to >80% in all Marquette grapevines (Figure 2B). Budbreak remained the same after 12 May 2018, and the maximum budbreak was at the same range (80 to 90%) among all four treatments.
Experiment 2
Similar to Experiment 1, different treatments led to distinguishable progressions of budbreak during our observations from 22 April to 8 May 2019 in both cultivars (Figure 3A and 3B). RMANOVA revealed a significant effect (p < 0.001) of treatment on budbreak in both cultivars. Pictures taken on 2 May 2019 showed that different treatments led to different bud phenology (Figure 3C and 3D). In La Crescent, grapevines initiated budbreak on 24 April 2019 (Figure 3A). Six days later, more than half of the control, ProTone-, and Amigo oil-treated buds were open, while the budbreak of FrostShield-treated grapevines was 23% (Figure 3A). Photos taken on 2 May 2019 showed that the control and Amigo oil-treated buds were at E-L 9, ProTone-treated buds were at E-L 8 to 9, and FrostShield-treated buds were at E-L 3 (Figure 3C). After 2 May 2019, budbreak of control, ProTone-treated, and Amigo oil-treated grapevines surpassed 90%, except FrostShield (Figure 3A). By 8 May 2019, all vines reached maximum budbreak at >90% (Figure 3A).
In Marquette, the control, ProTone-, and Amigo oil-treated grapevines initiated budbreak on 24 April 2019, while Frost-Shield-treated grapevine initiated budbreak on 26 April 2019 (Figure 3B). On 30 April 2019, >50% of buds in the control, ProTone-, and Amigo oil-treated grapevines were open, while the budbreak of FrostShield-treated grapevines remained low at 20% (Figure 3B). Photos taken on 2 May 2019 showed that the control, ProTone-, and Amigo oil-treated buds were at E-L 9, while FrostShield-treated buds were at E-L 3 (Figure 3D). By 6 May 2019, the budbreak gradually increased to >90% in all Marquette grapevines. During this period, lower budbreak was continuously observed in FrostShield-treated grapevines (Figure 3B). Budbreak of all Marquette grapevines remained the same after 6 May 2019, and maximum budbreak was >90% with no difference among all four treatments.
Experiment 3
In both cultivars, different treatments led to distinct progressions of budbreak (Figure 4A and 4B). RMANOVA revealed a significant effect (p < 0.001) of treatment on budbreak in both cultivars. Pictures taken after 25 days in forcing showed that different treatments led to different bud phenology (Figure 4C and 4D). In La Crescent, the control and ProTone-treated buds began opening after 16 days in forcing, but ProTone treatment had a higher budbreak than the other treatments (Figure 4A). Between 16 and 24 days in forcing, the budbreak of control and ProTone treatments gradually increased to ~95% and remained unchanged thereafter (Figure 4A). Amigo oil-treated buds began opening after 18 days in forcing, and the budbreak increased for 16 days thereafter and reached 94% (Figure 4A). Between 18 and 30 days in forcing, the percentage of open buds in Amigo oil treatment was consistently lower than control and ProTone treatment (Figure 4A). FrostShield-treated buds began opening after 22 days in forcing (six and four days later than control and Amigo oil-treated buds, respectively). Budbreak increased for 22 days thereafter and remained unchanged at 93% (Figure 4A). Between 22 and 42 days in forcing, the percentage of open buds in FrostShield treatment was consistently lower than all the other treatments (Figure 4A). Maximum budbreak was the same in all four treatments. Photos taken after 25 days in forcing showed that control buds were at E-L 8 to 9, ProTone-treated buds were at E-L 9 to 10, Amigo oil-treated buds were at E-L 6 to 7, and FrostShield-treated buds were at E-L 1 to 2 (Figure 4C).
In Marquette, control buds began opening after 16 days in forcing, while ProTone-treated buds began two days earlier (Figure 4B). After 28 days in forcing, the budbreak of the control and ProTone treatment gradually increased to 95% and remained unchanged thereafter (Figure 4B). Amigo oil-treated buds began opening after 18 days in forcing, and the budbreak increased to 92% after 34 days in forcing (Figure 4B). However, from 28 to 30 days in forcing, fewer buds were open in the Amigo oil treatment than in the control (Figure 4B). FrostShield-treated buds began opening after 20 days of forcing; the budbreak increased for 24 days and reached 93% (Figure 4B). From 20 to 42 days in forcing, the percentage of open buds in FrostShield treatment was consistently lower than all the other treatments (Figure 4B). Maximum budbreak was the same in all four treatments. Photos taken after 25 days in forcing showed that the control and ProTone-treated buds were at E-L 8 to 9, Amigo oil-treated buds were at E-L 7 to 8, and FrostShield-treated buds were at E-L 1 to 2 (Figure 4D).
Budbreak relative to control
Field experiments
In both cultivars, results from ANOVA showed significant differences (p < 0.001) of treatment effect but no significant differences of year effect (p = 0. 684 and 0.723 in La Crescent and Marquette, respectively) or year × treatment effect (p = 0.190 and 0.546 in La Crescent and Marquette, respectively). Thus, LSD was conducted to separate means of budbreak relative to control among treatments. In Experiment 1, FrostShield delayed the budbreak of La Crescent and Marquette by 4.5 and 4.2 days relative to control, respectively (Figure 5A). Amigo oil delayed the budbreak of La Crescent and Marquette by 2 and 1.5 days, respectively (Figure 5A). ProTone showed no effect on delaying budbreak in either cultivar (Figure 5A). In Experiment 2, FrostShield delayed the budbreak of La Crescent and Marquette by 5.5 and 4.5 days relative to control, respectively (Figure 5B). ProTone and Amigo oil showed no effect on delaying budbreak in either cultivar (Figure 5B).
Growth chamber experiment
In both cultivars, results from ANOVA showed significant differences (p < 0.001) of treatment effect. Thus, LSD was conducted to separate means of budbreak relative to control among treatments. FrostShield delayed the budbreak of La Crescent and Marquette by 18.5 and 15 days relative to control, respectively (Figure 5C). Amigo oil delayed the budbreak of La Crescent by seven days but showed no effect in Marquette (Figure 5C). ProTone advanced the budbreak of La Crescent by 1.5 days but showed no effect on Marquette (Figure 5C).
Grapevine phenology after budbreak
Experiment 1
RMANOVA revealed a significant effect (p < 0.001) of treatment on grapevine phenology after budbreak in both cultivars. The delay or advance of budbreak resulting from different treatments affected grapevine phenology during the early growing season, but these differences were no longer observed by the middle of the growing season (Figure 6A and 6B). In La Crescent, Amigo oil only decreased the value of E-L stage on 12 June 2018 (Figure 6A). However, the treatment that consistently affected grapevine phenology was FrostShield, which led to lower values of E-L stage (delayed phenology) from 18 May to 12 June 2018, but the difference was no longer observed on and after 21 June 2018 (Figure 6A). In Marquette, ProTone treatments advanced grapevine phenology (greater E-L stage) on 18 May 2018, while Amigo oil had no effect throughout the growing season (Figure 6B). FrostShield treatment delayed grapevine phenology during the first five observations, and this delay was no longer observed on and after 21 June 2018 (Figure 6B).
Experiment 2
RMANOVA revealed a significant effect (p < 0.001) of treatment on grapevine phenology after budbreak in both cultivars. Except for FrostShield treatment, all the other treatments did not significantly affect grapevine phenology in either cultivar after budbreak (Figure 6C and 6D). In La Crescent, the values of E-L stage in the control, Amigo oil-, and ProTone-treated grapevines increased from 12 to 17, from 13 to 27 May 2019 with no difference between each other (Figure 6C). The values of E-L stage in FrostShield-treated grapevines were lower than the other three treatments from 13 to 24 May 2019, but the difference was no longer observed on 27 May 2019 (Figure 6C). In Marquette, the values of E-L stage increased from 11 to 17 in the control, Amigo oil-, and ProTone-treated grapevines from 13 to 27 May 2019 with no difference between each other (Figure 6D). The values of E-L stage in FrostShield-treated grapevines was lower than that in the other three treatments from 13 to 24 May 2019, but the difference was no longer observed on 27 May 2019 (Figure 6D).
Vegetative and reproductive growth
In both experiments and in both cultivars, treatments did not affect shoot or cluster number (Supplemental Figure 1). In Experiment 1, the shoot number in La Crescent and Marquette was not different among treatments and ranged between 36 and 43 and 28 and 37, respectively (Supplemental Figure 1A). The cluster number in La Crescent and Marquette was not different among treatments and ranged between 43 and 48, and 38 and 52, respectively (Supplemental Figure 1B). In Experiment 2, the shoot number in La Crescent and Marquette was not different among treatments and ranged between 30 and 34, and 24 and 29, respectively (Supplemental Figure 1C). The cluster number in La Crescent and Marquette was not different among treatments and ranged between 40 and 47, and 36 and 43, respectively (Supplemental Figure 1D).
Discussion
Amigo oil versus FrostShield
Centinari et al. (2018) reported that the concentration of 10% Amigo oil was phyto-toxic because it caused higher bud mortality, compromised vegetative growth, and decreased yield in some cultivars, including Riesling and Lemberger. However, other researchers reported that this concentration was safe for other cultivars including Chambourcin and Chardonel (Dami and Beam 2004) and Edelweiss (Loseke et al. 2015). In this study, we did not observe any phytotoxicity symptoms with 10% Amigo oil in La Crescent or Marquette. This reaffirms that the optimum concentration of Amigo oil is cultivar- and species-dependent. Similar to previous studies, we observed budbreak delay in grapevines treated with 10% Amigo oil (Dami and Beam 2004, Loseke et al. 2015, Centinari et al. 2018). However, the effect of Amigo oil in delaying budbreak showed significant variation among experiments and cultivars. In the field, Amigo oil effectively delayed the budbreak of both cultivars for two to three days in year 1 (Experiment 1), while this delay was not observed in year 2 (Experiment 2). In the growth chamber, Amigo oil significantly delayed the budbreak of La Crescent cuttings by seven days, while no delay was observed in Marquette. Although the year × treatment interaction was not statistically significant, the difference of Amigo oil effectiveness between the two years might be explained by oil weathering under different meteorological conditions (precipitation, temperature, cumulative chilling units, and GDD). For example, a period of five days of precipitation was observed in Experiment 2 immediately after treatment application, while this did not occur in Experiment 1. Thus, it is likely that the precipitation after treatment in year 2 resulted in product wash off, which in turn caused the lack of response from Amigo oil. Amigo oil’s delaying effect on budbreak might be explained by its inhibition of respiration (Dami and Beam 2004). Respiratory activity gradually increases in grapevine buds during the transition from ecodormancy to budbreak, which is likely due to changes of metabolic activities (Gardea et al. 1994, Díaz-Riquelme et al. 2012). Reduced respiration and slower metabolism in Amigo oil-treated buds may ultimately lead to the delay of budbreak.
We did observe that Amigo oil significantly delayed budbreak in one of our field experiments, but the duration of delay was very short (e.g., two-day delay in La Crescent and 1.5-day delay in Marquette), which might not be economically practical in a commercial setting. Some researchers prolonged the efficacy of Amigo oil to delay budbreak by using multiple applications (Loseke et al. 2015). However, using multiple applications may not be warranted for commercial vineyards because of its increased cost. Based on interviews, most growers with small vineyards (less than 4 ha) in the midwestern USA were willing to invest less than $500/ha for spring frost management (Wang and Dami, unpublished data), whereas a single application of Amigo oil costs $532/ha (Centinari et al. 2018). It is estimated that the cost of FrostShield/ha will be nearly half that of Amigo oil.
Compared with Amigo oil, FrostShield showed a superior effect by delaying the start and the end of budbreak of both cultivars in all three experiments. The budbreak delay by FrostShield could potentially reduce the threat of spring frost injury because the historical median (i.e., 50% probability) of the last spring frost (temperature below 0°C) date is 3 May at the research location site (www.oardc.ohio-state.edu/weather1/). In both years and in both cultivars, FrostShield-treated vines had the lowest budbreak on the median date. Furthermore, the variation of FrostShield’s effectiveness between field and growth chamber experiments can be associated with weather. For example, precipitation and high solar radiation (UV) are eliminated in a growth chamber environment, which results in reduced product weathering and preservation of active ingredients, thus contributing to a longer delay of budbreak (Lee and Rawlings 1982, Potvin and Tardif 1988).
In the field experiments, FrostShield-treated grapevines also showed a delayed phenology (lower values of E-L stage) after budbreak. In Experiment 1, FrostShield delayed grapevine phenology for about one month after budbreak, and this delaying effect was no longer observed after grapevines reached E-L 29 (peppercorn berry size) in late June. These data also indicate that flowering (from E-L 19 to E-L 26) was delayed by FrostShield in Experiment 1. In Experiment 2, FrostShield delayed grapevine phenology for about two weeks after budbreak, and this delaying effect was no longer observed after grapevines reached E-L 17 (12 leaf stage) in late May. These data indicate that flowering was not affected by FrostShield in Experiment 2. However, regardless of the duration of delay and its effect on flowering, the effect of FrostShield disappeared at harvest because yield and fruit maturity were not negatively affected (Supplemental Tables 1 to 4). Nevertheless, the delay on grapevine phenology, perhaps after multiple applications, might create additional benefits for commercial grape production.
ProTone effect on grapevine budbreak
In our study, the application of ProTone at 500 mg/L S-ABA showed a minor and inconsistent effect on budbreak. For example, ProTone treatment slightly advanced budbreak of La Crescent and Marquette in the growth chamber experiment but did not show any effect in two field experiments. In fact, ProTone slightly advanced grapevine phenology (higher E-L stage values) in Experiment 1, but it showed no effect throughout the growing season in Experiment 2. The lack of ProTone influence on budbreak delay remains unclear in this study, and the following are suggestions of possible reasons. First, the efficacy of ProTone on grapevine budbreak or dormancy release might be cultivar- or species-dependent. Previous studies using exogenous ABA were conducted with Vitis vinifera (Weaver et al. 1974, Hellman et al. 2006, Zheng et al. 2015, Kovaleski and Londo 2019), while we used two interspecific hybrids in our study. Different cultivars might respond to exogenous ABA differently. Second, the inconsistency between our findings and others might also be explained by different ABA application protocols. Previous researchers applied ABA by immersing single bud cuttings in ABA solutions (Weaver et al. 1974, Zheng et al. 2015) or at more advanced phenological stage when bud scales were more open (Kovaleski and Londo 2019). The exposure of bud interior tissues to exogenous ABA would facilitate its penetration into the vascular system. In our study, we sprayed ProTone at E-L stage 1 (buds fully closed and protected by scales) in field experiments and on canes with sealed tips in the growth chamber experiment. These conditions may have prevented exogenous ABA penetration into the bud and vascular tissues. Penetration of ABA might be crucial for its function, because ABA degrades rapidly under experimental conditions where light intensity is generally high (Hellman et al. 2006, Gao et al. 2018). Further studies are warranted to test timing and methods of ABA application that facilitate its tissue penetration. Third, ABA’s effect on plants might be determined by the concentration of ABA applied. For example, a high level of exogenous ABA (>250 mg/L) led to oxidative damage to plant cells on leaves of maize seedlings, while a low level of exogenous ABA (<25 mg/L) induced antioxidative defense response (Jiang and Zhang 2001). The rate of 500 mg/L used in this study may be too high or too low for ABA to effectively delay budbreak in grapevines. However, this hypothesis needs further investigation.
Conclusion
Amigo oil showed varied effects and ProTone showed no effect on delaying budbreak of La Crescent and Marquette under field and growth chamber conditions. This inconsistency raises doubt of their suitability to be used commercially. On the other hand, FrostShield delayed budbreak consistently and for a longer duration, although it also delayed grapevine development for about a month postbudbreak. The use of FrostShield to delay budbreak may effectively reduce spring frost injury, a desirable outcome under the ongoing climate change. Considering that all products did not compromise grapevine vegetative or reproductive growth, FrostShield appears to be a better alternative to Amigo oil or ProTone as a vineyard management practice to mitigate spring frost injury. Owing to its greater efficacy in this study, FrostShield is being considered for release to the market for commercial use in vineyards.
Acknowledgments
This research was supported by funds appropriated by The Ohio State University Department of Horticulture and Crop Science and the Ohio Grape Industries Committee. The authors would like to thank Andrea Chapman and Juliet Freed for maintaining experimental vineyards and collecting data; Diane Kinney for assistance with fruit composition analysis; Logan Walter, Joshua Heller, and Becky Colon for assistance with vineyard management.
Footnotes
Supplemental data is freely available with the online version of this article at www.ajevonline.org.
By downloading and/or receiving this article, you agree to the Disclaimer of Warranties and Liability. The full statement of the Disclaimers is available at http://www.ajevonline.org/content/proprietary-rights-notice-ajev-online. If you do not agree to the Disclaimers, do not download and/or accept this article.
- Received October 2019.
- Revision received December 2019.
- Revision received January 2020.
- Accepted January 2020.
- Published online July 2020
- © 2020 by the American Society for Enology and Viticulture