Simulated Abiotic Injury Alters Yields of Southern Interspecific Hybrid Grape Cultivars

Abstract

Climate change, including more volatile weather and longer growing seasons, is causing stress on grapevines (Vitis spp.). A change in harvest timing of wine grapes can have significant consequences. Thus, two methods (crop forcing and complete removal of green tissue) were employed to simulate abiotic vine injury. The harvest of bunch grapes in Mississippi occurs during July, a very hot month. ‘Miss Blanc’ and ‘Villard Blanc’ had four different crop forcing treatments imposed to determine yield amount and harvest timing. All treatments reduced yield. Harvest was delayed by 50 days, a potentially positive shift that was not enough to escape high temperatures. ‘Villard Blanc’ had no flower or fruit development after crop forcing treatments in May and June. ‘Miss Blanc’ yields were also significantly reduced by these treatments. Removal of green tissue to simulate injury from weather events such as frost, freeze, wind, or hail in both Mississippi and Oklahoma revealed that lost growth could reduce yields from 19% to 81%, which could influence grape grower management decisions.

1. Introduction

Climate change is a serious issue for growing grapes in many regions of the world [1]. While longer growing seasons may appear beneficial, volatility in early growing season temperatures can lead to abiotic stresses such as frost, freeze, wind, or hail injury because vines will break bud earlier [2,3,4,5]. These types of injuries are of concern to growers who need to know how to manage vines after such events. The expense of pesticide applications and other vine management techniques may not be economically warranted if yields are not at acceptable levels. Although severely damaging events like these are uncommon in the deep South, the unpredictability of climate change may increase the likelihood of them becoming more frequent.

A related issue to longer growing seasons is harvest timing. A change in harvest timing of wine grapes, a high value crop dependent on specific quality parameters, can have significant consequences. Crop forcing is a technique used to manipulate timing of grape harvest in hot production regions due to fruit quality concerns [1,6,7]. It is desirable to move harvest to cooler parts of the growing season. Previous work in California [8] and other countries [6,7,9,10,11] suggests that this could be a viable practice in hot regions where changing the harvest time to cooler conditions would improve fruit quality without being detrimental to future vine productivity. Harvest of bunch grapes in Mississippi and Oklahoma occurs during July and August, very hot months. Delaying harvest weeks or even months until cooler weather may improve fruit quality as well as worker conditions. A recent study by Martinez de Toda et al. [10] found that crop forcing significantly delayed harvest of ‘Tempranillo’ and ‘Maturana Tinta’ in Spain, but also found reductions in cluster number, cluster weight, and vine yield. Zheng et al. [7] reported that late pruning could have multiple year effects on the vine, but other studies showed no changes in subsequent years [6,8]. Thus, any effects of late pruning or crop forcing could be location and cultivar dependent. The technique of crop forcing could also be used to simulate abiotic injury brought on by climate change variables. While it has been primarily used to shift harvest timing, effects of the vine manipulation may also give insight into how vines might respond to destructive weather events.

The objective of this study was to assess compensatory cropping potential after exposure to simulated abiotic injury by use of crop forcing and complete green tissue removal in southern Mississippi and Oklahoma.

2. Materials and Methods

The Mississippi portion of the study was performed in 2016 at the USDA-ARS Thad Cochran Southern Horticultural Laboratory in Poplarville, MS. Own-rooted ‘Miss Blanc’ and ‘Villard Blanc’ grapevines (interspecific hybrids, Vitis spp.) planted in 2012 were used. They were trained to a high curtain bilateral cordon (1.8 m) spaced at 3 m between rows and 2.1 m between vines on an east to west orientation, and spur pruned. Vines were not irrigated, as rainfall was sufficient (

Table 1. Growing season (budburst to harvest) 2016 weather and 30 year historical climate data for average high temperature, average low temperature, and average monthly rainfall for Poplarville, MS.

Month	2016 High Temp. (°C)	Historical Avg. High Temp. (°C)	2016 Low Temp. (°C)	Historical Avg. Low Temp. (°C)	2016 Precip. (mm)	Historical Avg. Precip. (mm)
March	22	21	13	9	300	142
April	25	25	14	12	170	127
May	28	29	16	17	88	119
June	32	32	21	21	88	137
July	33	33	23	22	197	163
August	33	33	23	22	191	137
Sept	31	31	21	19	131	99

). Fungal diseases, primarily anthracnose (Elsinoe ampelina) and black rot (Guignardia bidwellii), were controlled with fungicide applications sprayed as needed, but typically every 10 to 14 days. Pest control sprays were applied as suggested by the Southeast Regional Bunch Grape Integrated Management Guide [12]. Yields were collected for each vine when fruit was ripe. Budburst and anthesis were recorded, which for both cultivars was 14 March and 29 April 2016, respectively.

In Perkins, Oklahoma, at the Oklahoma State University Agriculture Experiment Station, ‘Neptune’ vines grafted on 3309 Couderc rootstock were used in the study during 2009 and 2010. All management was similar to that described above, except vines were spaced at 3.7 m between rows and 2.4 m between vines. Budburst was recorded on 6 April 2009 and 8 April 2010.

Two methods were used to simulate abiotic vine injury, complete removal of all green tissue and crop forcing. Both methods are described below. These methods are used to simulate climate change related issues such as hail, frost and freeze, and wind damages.

2.1. Crop Forcing

The crop forcing treatment was performed consistently with that described by Gu et al. [8]. Briefly, this was done by hand pruning emerged shoots to six nodes, removing all laterals, leaves, and clusters at 21, 28, and 35 days post-anthesis (Figure 1). This equated to 20 May, 27 May, and 3 June 2016. Control vines received no forcing treatment, but were allowed to grow as normal. Cultivars used were ‘Miss Blanc’ and ‘Villard Blanc’. The final design was a 2 × 4 factorial in a randomized complete block with three vine replicates per treatment.

2.2. Complete Removal

The cultivars used were ‘Miss Blanc’ and ‘Villard Blanc’ in Mississippi and ‘Neptune’ in Oklahoma. Experimental design in Mississippi was a randomized complete block with 3 vine replicates for each cultivar-treatment combination. Canopy injury treatments were: (1) remove all green growth at the 10% probability freeze date (−2.2 °C; 24 March); (2) remove all green growth at the 10% probability frost date (0 °C; 11 April); (3) no removal of green growth. These average temperature dates were chosen because −2.2 °C is generally considered a damaging temperature to actively growing tissue, but there could be variability depending on duration of cold temperatures and phenological stage of the vine [13]. Therefore, 0 °C was also used as a worst-case scenario treatment. Ten percent probability dates were chosen to represent worst case scenarios and the data were obtained from the National Oceanic and Atmospheric Administration (NOAA).

In Oklahoma, a completely randomized design was employed with 16 total vines, eight of each treatment. The treatments were: (1) complete removal of all green growth on 1 May of each year (the historical extreme date of 0 °C); and (2) no removal. Vines were pruned to 50 buds on each vine both years. The study was performed in 2009 and 2010 and treatment vines were the same both years.

Data were subjected to analysis of variance with JMP (version 12) (SAS Institute, Inc., Cary, NC, USA) using the fit model procedure. Means were separated using Tukey’s honestly significant difference (HSD) test at p ≤ 0.05. For Oklahoma data, years were analyzed separately.

3. Results

3.1. Crop Forcing

A significant interaction of cultivar and forcing treatment was found. All removal treatments reduced yield in both cultivars (

Table 2. Crop forcing yield response of two interspecific hybrid grape cultivars grown in Poplarville, MS.

Cultivar	Yield Per Vine (kg) z	Total Yield Equivalent (kg/ha)
Miss Blanc	4.2 a	6460
Villard Blanc	0.7 b	1077
Forcing
None	6.6 a	10,151
21 days post-anthesis	1.0 b	1538
28 days post-anthesis	0.2 b	308
35 days post-anthesis	1.9 b	2922
Interaction
Miss Blanc, None	10.6 a	16,303
Miss Blanc, 21 days	2.1 bc	3230
Miss Blanc, 28 days	0.4 c	615
Miss Blanc, 35 days	3.8 b	5844
Villard Blanc, None	2.7 bc	4153
Villard Blanc, 21 days	0.0 c	0
Villard Blanc, 28 days	0.0 c	0
Villard Blanc, 35 days	0.0 c	0
Effect Test	F	p
Cultivar	52.22	<0.0001
Forcing	30.91	<0.0001
Cultivar × Forcing	10.06	0.0007

^z Means within a column and main effects or interactions followed by the same letter are not significantly different by Tukey’s HSD at α = 0.05.

). By applying crop forcing 21 days after anthesis, yield was significantly reduced in ‘Miss Blanc’ by 80%. At 28 days, yield was reduced by 97%, but rebounded at 35 days to 64%. Why the yields at 28 days were less than at 35 days is unknown, however, flowering could have coincided with above average high temperatures (35 °C+) near the end of June and early July, thus negatively affecting pollination. The effect was even more dramatic for ‘Villard Blanc’. ‘Villard Blanc’ had no flower or fruit development after any crop forcing treatment (Table 2) and displayed poor overall regrowth. Harvest for all three forcing treatments was delayed by 50 days compared to the control from 21 July to 9 September 2016, yet temperatures were still very high (Table 1) and the desired effect was not achieved. Overall, the temperatures and rainfall were not much different from an average season, thus one would not expect further delays in ripening in other years.

3.2. Complete Removal

When green tissues were completely removed in Mississippi there was no significant interaction effect. Both cultivars responded similarly to the treatments, however, levels of production were not the same (

Table 3. Grapevine response to growth removal at times simulating abiotic (cold) injury by 10% probability date for each event in Poplarville, MS.

Cultivar	Yield Per Vine (kg) z	Total Yield Equivalent (kg/ha)
Miss Blanc	8.3 a	12,765
Villard Blanc	1.7 b	2615
Removal
None	6.6 a	10,151
−2.2 °C (March 24)	5.0 ab	7690
0 °C (April 11)	3.5 b	5383
Interaction
Miss Blanc, None	10.6	16,303
Miss Blanc, −2.2 °C	8.1	12,458
Miss Blanc, 0 °C	6.4	9843
Villard Blanc, None	2.7	4153
Villard Blanc, −2.2 °C	2.0	3076
Villard Blanc, 0 °C	0.6	923
Effect Test	F	p
Cultivar	132.54	<0.0001
Removal	10.50	0.0035
Cultivar × Removal	1.37	0.2975

^z Means within a column and main effects followed by the same letter are not significantly different by Tukey’s HSD at α = 0.05.

). Cultivars differed in yield, with ‘Miss Blanc’ being more productive than ‘Villard Blanc’. Removal of green growing tissue at the last average freeze date (−2.2 °C; 24 March) in Mississippi reduced crop yield by 24% over the control (no removal). Later removal at the last average frost date (0 °C; 11 April) further reduced crop yield to 47% of the control. Production levels for ‘Miss Blanc’ would still be considered viable after both removal treatment timings, though reduced by as much as 40%. The late removal treatment on ‘Villard Blanc’ had substantially less yield than the control and the early removal treatment. While yield was reduced by 26% for the early removal treatment, the crop would still be protected during the season. Whereas the late removal treatment decreased the crop by 78%, essentially rendering it a loss. Harvest was delayed by only five days later than the control for both removal timings from 21 July to 26 July 2016.

In Oklahoma, harvest date for ‘Neptune’ was delayed by 11 days in 2009 and 16 days in 2010 when initial growth was removed. Vines that did not receive the treatment produced an average of 8.5 kg per vine in 2009 and 12.4 kg per vine in 2010 (

Table 4. Yield and cluster weight response of ‘Neptune’ grapevines to simulated abiotic injury in Perkins, Oklahoma, over two years.

Year	Removal Treatment	Yield Per Vine (kg) z	Total Yield Equivalent (kg/ha)	Average Cluster Weight (g) z
2009	No	8.5 a	9472	301.3 a
2009	Yes	3.0 b	3475	228.5 b
2010	No	12.4 a	13,918	271.0 a
2010	Yes	2.3 b	2513	124.5 b

^z Means within a column and year followed by the same letter are not significantly different by Tukey’s HSD at α = 0.05.

). Treated vines averaged 3.0 kg per vine in 2009 and 2.3 kg per vine in 2010. Individual average cluster weights were also greater on untreated vines vs. treated vines, but the treatment did not affect berry size or Brix within one year (data not shown). In this experiment, treated vines produced 35% of untreated vines in 2009 and just 19% in 2010.

4. Discussion

4.1. Crop Forcing

Harvest was delayed by 50 days, which could be beneficial in a hot climate if yields were sufficient, but it was still not delayed enough to avoid high temperatures. ‘Villard Blanc’ yields may have been low due to Pierce’s disease (PD) (Xylella fastidiosa), an infection that was detected later in the summer, after harvest. Vines displayed weak regrowth after being forced, which is in contrast to what Gu et al. [8] reported with ‘Cabernet Sauvignon’. This may be attributed to an interaction of the PD infection, and a loss of essential carbohydrates that could not be regained [9,11]. The vines have since died, many as early as the following winter of 2016–2017. Conversely, ‘Miss Blanc’ vines were vigorous and productive, possibly tending to overproduction. All crop forcing treatments for both cultivars significantly reduced yields to the point of being undesirable. Lower yields were reported by others [1,7,10], but no reports of total crop failure were found. Moran et al. [6] stated that late pruning must not have detrimental effects on the vine yield if it is to be a useful management technique. Yet, this was not the case for ‘Miss Blanc’ and ‘Villard Blanc’ in the present study. Even if crop loads were high enough, there would be a need for a longer pesticide spray regimen. Late summer in south Mississippi is still hot and wet (Table 1), and control of fungal diseases is difficult even under the best situations. While other researchers proclaimed that late pruning was a promising canopy management technique [7], as applied in south Mississippi to simulate abiotic injury, the delay would need to be longer, essentially into October to have the desired effect. Martinez de Toda et al. [10] stated that vine fruitfulness was in direct relation to phenological stage when forcing was done. Forcing should be done earlier than 21 May in south Mississippi if vines are to retain sufficient fruitfulness. Therefore, severe abiotic injury occurring later than this time, which would include hail and wind damage, would essentially result in loss of the crop.

4.2. Complete Removal

Earliest removal still allowed for marketable yields for both ‘Miss Blanc’ and ‘Villard Blanc’ in Mississippi, but did not delay harvest timing much (5 days). One practical application of this study was to understand the potential reduction in yield that abiotic events such as frost, freeze, wind, or hail could have on grapevines. Jones et al. [13] stated that vines could be affected in current and subsequent years by direct and indirect effects of frost damage. Hail events can cause loss of shoots, leaves, and fruits [14]. When severe, it can lead to negative vine effects in multiple years [14,15]. In the present study, both ‘Miss Blanc’ and ‘Villard Blanc’ responded similarly, thus one might expect other cultivars to do the same, although further testing is needed to bear this out. Since budburst starts in early March most years in Poplarville, MS, any very late loss of growth such as that simulated by the late removal treatment in this study, would result in substantial crop losses which would not be economically advantageous to maintain at a high level.

The results were similar in Oklahoma for ‘Neptune’. Removal of green growth on 1 May resulted in extreme losses, especially in the second year of the study. There may have been a carryover effect as suggested by the Australian Wine Research Institute [14] and Agosta et al. [15]. Some of the loss can be attributed to reduction in cluster weights, as was also found by Zheng et al. [7] and Martinez de Toda et al. [10]. While severe losses were observed, ‘Neptune’ is a table grape that is eaten fresh and sold in a different market than wine grapes. The higher retail price obtained in a fresh market may lead a grape grower to keep managing the vines at a high level in contrast to wine grapes that are processed and typically bring a lower price in the southern U.S.

5. Conclusions

The crop forcing imposed in the present study to simulate abiotic injury showed that yield losses are substantial and harvest is delayed. Further research could be done on other cultivars. Disease control and depredation obstacles were also significant. Timing of forcing for each cultivar and location would need to be evaluated to determine how vines respond to destructive manipulation, thus allowing researchers to make recommendations to grape growers.

Abiotic injury simulation showed that early loss of growth can still produce yields that are acceptable, however, later injury would impose a difficult decision making process upon a grape grower. While these types of events are not common now, they may become more problematic as the climate continues to warm, budburst occurs earlier, and frost events become more unpredictable. A standardized methodology could establish a baseline of crucial information to grape growers who must make difficult economic management decisions after abiotic injury events. More work must be done to determine effects of timing, location, cultivar, rootstock, and other variables on yield potential after abiotic injury.