Our project had 4 main objectives outlined in the original proposal:
- To conduct a field trial to assess and compare the effects of cluster thinning (CT) and early leaf removal (ELR) on vegetative growth, yield components, Botrytis bunch rot infections, cluster compactness, bud fruitfulness, and fruit composition.
- To assess the impact of CT and ELR on parameters associated with vine susceptibility to winter injury, including single-leaf net carbon assimilation, bud and cane cold hardiness, and carbohydrate reserve storage.
- To perform an economic analysis that estimates the additional price per tonne of fruit and cost per 750 mL bottle of wine needed to maintain grower economic welfare, if CT and ELR practices were adopted compared to if they were not.
- To provide recommendations to stakeholders for managing high-yielding wine grape varieties for optimal fruit composition and vine health.
Towards our objectives, we properly implemented each canopy management tactic at the appropriate time during the 2016 growing season, and collected much of the necessary data to fulfill our first three objectives. We are continuing to collect data on winter tissue cold hardiness, having begun our monthly sample collection in November 2016.
For our fourth objective, I am currently scheduled to present at the Mid-Atlantic Fruit and Vegetable Convention in February 2017 on crop load management practices, as well as present results from 2015-2016 on this project at the Pennsylvania Wine Marketing and Research Board Symposium in March 2017.
Vineyard site and experimental design: A field experiment was conducted in 2015 and 2016 seasons on a high-yielding white grape cultivar Grüner Veltliner (Vitis vinifera L.) grafted on 101-14 Mgt rootstock and planted in 2010 at a commercial vineyard in Lewisburg, PA. Vines were spaced 1.5 m between vines and 2.4 m between rows for a density of 2778 vines per hectare with rows oriented north-south. Vines were pruned to two-bud spurs and trained to a vertically shoot positioned bilateral cordon trellis. Shoot thinning was performed when shoots were at growth stage 14 according to the modified E-L system (1; 25 May 2015 and 26 May 2016). The number of shoots was standardized to approximately 15 shoots per linear meter of row. Canopy disease and insect management were applied according to standard grower practices.
Four adjacent rows were selected for the study. The experimental design was a complete randomized block design with four blocks. Each block consisted of three panels (sections between three post spaces) comprised of four vines each (12 vines per experimental unit). Each of the following treatments was randomized within each block: i) an unthinned, non-defoliated control (C), ii) early leaf removal at trace-bloom (TBLR; E-L 19), iii) early leaf removal at fruit-set (FSLR; E-L 27), and iv) cluster thinning (CT). The first and last four-vine panel of each row were excluded from the experimental area as a buffer. Early leaf removal treatments were implemented by hand-removal of five primary leaves and laterals that developed from nodes 2 to 6 on each shoot. Cluster thinning was performed by removing the distal clusters at bunch-closure (E-L 32) leaving one cluster per shoot. Dates for TBLR performance were 3 June 2015 and 9 June 2016, 18 June 2015 and 21 June 2016 for FSLR, and 1 July 2015 and 13 July 2016 for CT. Treatments were performed on the same vines for both years.
Gas exchange, EPQA, leaf area: Twelve shoots from the four central vines of each experimental unit (48 shoots per treatment) were flagged prior to application of TBLR for data collection. Single leaf gas exchange (assimilation rate (A), transpiration rate (E), and stomatal conductance (gs) was recorded on eight fully expanded primary and eight lateral leaves located at nodes 7 and 12 for each experimental unit at one week post-treatment and repeated 2 to 3 times until harvest. Measurements were taken between 10:00 and 14:00 hours under light-saturated (> 1200 mmol m-2 s-1) conditions using an open-system portable infrared gas analyzer (CIRAS-3, PP Systems, Amesbury, MA, USA). The system utilized a broad leaf cuvette with a 4.5 cm2 window with 400 parts per million (ppm) fixed-rate of CO2 and an adjusted air flow of 400 mL/min.
Enhanced point quadrat analysis (EPQA, 2, 3) was assessed twice, mid-season and pre-harvest, in 2015 (16 July and 25 August) and 2016 (28 July and 7 September) for treatment impacts on canopy density and light availability in the cluster zone. EPQA was conducted as described by Meyers and Vanden Heuvel (2).
All flagged shoots were collected one day prior to the whole plot harvest to assess whole shoot leaf area at the end of the season. Primary and lateral leaves were analyzed separately using a scanning leaf area meter (LI-3100c, LI-COR).
Harvest yield, yield components, and grape composition: All clusters of flagged shoots were collected two days prior to the whole plot harvest to assess impacts of treatments on cluster compactness, cluster weight, number of berries per cluster, fruit set, and basic juice chemistries (total soluble solids, pH, and titratable acidity).
The vines were harvested by hand on 24 September 2015 and 28 September 2016. Immediately prior to harvest, 30 randomly selected clusters per experimental unit were visually assessed for Botrytis rot incidence and severity (4). Ten vines per experimental unit, excluding the first and last vine, were used to measure yield per vine and the number of clusters per vine. A hanging scale accurate to 0.1 kg (Pelouze 7710, Rubbermaid, Inc., Huntersville, NC) was used to determine yield, and the average cluster weight calculated by the yield divided by the number of clusters.
The individual clusters harvested from flagged shoots were frozen at -20 °C until deconstruction for individual berry weight and number of berries measurements. Berries were divided in the following categories: normally developed, healthy berries, f berries infected with rot, undeveloped (shot) berries, green ovaries. Berries of each category were counted separately.
Berries were then thawed in a 60 °C water bath, crushed, and strained to remove skin and seeds for juice chemistry. Total soluble solids (TSS) was measured using a hand-held refractometer (Master, Atago USA, Inc., Bellevue, WA), pH was conducted using a pH-meter (Orion Star A111, Thermo Fisher Scientific, Waltham, MA), and titratable acidity (TA, expressed as g/L tartaric acid equivalents) was measured using an autotriator (G20, Mettler Toledo, Columbus, OH). Titrations were made using a 10 mL juice sample size titrated with 0.1 N NaOH to an endpoint pH of 8.2.
Pruning weight were taken during the dormant season using a 0.1 kg accuracy hanging scale (Pelouze 7710). The Ravaz index was calculated as the ratio of yield to pruning weight and used to estimate the crop load of each treatment. Bud fruitfulness was evaluated prior to shoot thinning during the growing season following treatments implementation when inflorescences were clearly visible. The number of shoots and number of inflorescences were recorded to calculate bud fruitfulness (number of inflorescences per shoot).
Dormant tissue cold hardiness: Cold hardiness of dormant cane and primary bud tissue was estimated monthly between November and March, using differential thermal analysis (DTA) according to Mills et al. (6). Eight canes were collected across the ten central vines of each experimental unit, wrapped in moist paper and stored in plastic bags for transport. Canes were cut at internode two in order to leave two basal buds on the vines for the following year vegetative growth. DTA was performed across two days, using four canes per experimental unit each day. Canes were stored at 4 °C until analysis. Five buds were excised from nodes two through six with approximately 2 mm of tissue intact surrounding the bud and placed with bud contact on a thermoelectric module (Melcor Corporation, Trenton, NJ). Two internode sections of approximately 35 mm length were randomly selected from between the first five nodes for estimating cold hardiness of xylem and phloem tissues. Trays were placed in a programmable freezer (Tenney, Thermal Products Solutions, New Columbia, PA). The freezer was programmed to begin at 4 °C and decrease at a rate of 4 °C/hr until -40 °C before recovering to 4 °C. A total of 20 buds and 4 cane segments for each experimental unit were used to estimate the average median low temperature exotherm (LT50) for buds or the temperature at which primary buds experience 50% mortality (7) and LT10 for vascular tissue, the temperature at which 10% injury occurs (6).
Non-structural carbohydrate concentration: Non-structural soluble carbohydrates (CHO) and starch were analyzed in Chancellor (Vitis hybrid) in a parallel study focused on comparing traditional CT and intensity of hand early defoliation (3-leaves vs. 5-leaves) in a high-yielding interspecific hybrid variety. This study was conducted in 2014 and 2015 at The Pennsylvania State University Lake Erie Grape Research and Extension Center in Northeast, PA. The Chancellor experimental design was similar to that of the Grüner Veltliner, in which 16 experimental units of 12 vines was established in a complete randomized block design.
Three tissue-types (root, trunk, and cane) at three time points during the dormant season (November, January, and March) of 2014 and 2015 were collected on four central data vines for each experimental unit. Root sampling was only performed in November and March due to frozen soil in mid-winter. Root samples encompassed multiple branches of roots collected near the trunk to ensure roots emerged from the correct treatment vines. Trunk tissue was sampled using a 1.8 mm diameter drill bit to core half-way through three sections of the trunk height (base, center, and head). Concurrently, 4 cm long cane tissues were collected from nodes three to five. Tissue samples were immediately frozen on dry ice for transport back to storage and stored at -80 °C until processing. Tissues were then dehydrated with a freeze drier (Labconco Corporation, Kansas City, MO) for seven to ten days and ground finely using a Cyclone Mill (UDY Corporation, Fort Collins, CO) with 1.0 mm mesh size.
Soluble CHOs including sucrose, fructose, glucose, raffinose, and stachyose were extracted with a procedure modified from Reed et al. (8). Fifty milligrams of dry, ground tissue was washed three times at 80 °C for one hour with 3 mL of 80% analytic-grade ethanol adjusted to pH 6.75 using formic acid; 1.5 mg/g dry tissue was added as an internal standard for extraction efficiency. Samples were dried under nitrogen evaporation and residue was re-suspended in analytic-grade water and filtered through a 0.2 mm nylon disk filter into a 1.5 mL HPLC vial in order to be quantified using liquid-chromatography tandem mass-spectrometry (LC-MS/MS). Standard curves for CHOs quantification were established with 1, 0.5, 0.25, 0.13, 0.06, and 0.03 mg/mL of glucose, fructose, sucrose, and lactose and 0.1, 0.05, 0.03, 0.01, 0.006, and 0.003 mg/mL for oligosaccharides raffinose and stachyose.
Following soluble CHO extraction, samples were freeze-dried for 24 hrs. Non-structural starch remaining from soluble CHO extraction was hydrolyzed with 5 units of amylogucosidase and 2.5 units of a-amylase (Millipore Sigma, Merck KGaA, Germany) for glucose quantification using colorimetic analysis (9, 10).
Economic costs of canopy management: An economic analysis was performed to determine the additional price per tonne and price per 750 mL bottle required to maintain the same economic welfare of using TBLR, FSLR, and CT compared with no adoption of the practices (11). The time necessary to perform each treatment was recorded each year. The additional production costs per hectare (i.e., labor) was estimated according to Yeh et al. (12). The additional price per 750 mL bottle was calculated by dividing the additional production cost/ha by the yield (tonne/ha). Expected revenue/ha was then calculated by multiplying yield by the average industry price per tonne for Grüner Veltliner for 2015 and 2016 (https://flgp.cce.cornell.edu). The additional price per tonne required to maintain grower economic welfare (grower preferred price) was calculated by dividing the expected revenue per hectare for the C by the yield and adding the additional cost per tonne of production for each treatment. The price per 750 mL bottle required to maintain grower economic welfare was then derived by subtracting the average price per tonne from the grower preferred price and dividing by 655.2, the number of 750 mL bottles from one tonne of grapes.
Statistical analysis: Data analysis was performed using JMP statistical software (v. 12.1, SAS Institute, Cary, NC). All data were analyzed across years using a mixed-effects analysis of variance (ANOVA) model, with year and treatment as fixed-effects, blocks as random-effects, and a treatment by year interaction effect. In the instance of a treatment by year interaction effect, parameters were analyzed separately within each year.
- Coombe, B.G. 1995. Adoption of a system for identifying grapevine growth stages. J. Grape Wine Res. 1:100-110.
- Meyers, J.M. and J.E. Vanden Heuvel. Enhancing the precision and spatial acuity of point quadrat analyses via calibrated exposure mapping. Am. J. Enol. Vitic. 59(4):425-431.
- Smart, R.E. and M. Robinson. Sunlight into Wine: A Handbook for Winegrape Canopy Management. Winetitles, Adelaide.
- Horsfall, J.G. and R.W. Barratt. An improved grading system for measuring plant disease. Phytopathology. 35:655.
- Ipach, R., B. Huber, H. Hofmann, and O. Baus. 2005. Richtlinie zur Prü- fung von Wachstumsregulatoren zur Auflockerung der Traubenstruktur und zur Vermeidung von Fäulnis an Trauben. Outline for an EPPO-Guideline.
- Mills, L.J., J.C. Ferguson, and M. Keller. Cold-hardiness evaluation of grapevine buds and cane tissues. Am. J. Enol. Vitic. 57(2):194-200.
- Wolf, T.K. and R.M. Pool. Factors affecting exotherm detection in the differential thermal analysis of grapevine dormant buds. J. Amer. Soc. Hort. 112:520-525.
- Reed, A.B., C.J. O’Connor, L.D. Melton, and B.G. Smith. Determination of sugar composition in grapevine rootstock cuttings used for propagation. J. Enol. Vitic. 55(2):181-186.
- Somogyi M. 1952. Notes on sugar determination. Biol. Chem. 195:19-23.
- Nelson, N. 1944. A photometric adaptation of the Somogyi method for the determination of glucose. Biol. Chem. 153:375-380.
- Sun, Q., G.L. Sacks, S.D. Lerch, and J.E. Vanden Heuvel. Impact of shoot and cluster thinning on yield, fruit composition, and wine quality of Corot Noir. AJEV. 63(1):49-56.
- Yeh, A.D., M.I. Gomez, and G.B. White. 2014. Cost of establishment and production of vinifera grapes in the Finger Lakes region of New York-2013. Department of Applied Economics and Management, Cornell University, Ithaca, NY.
Yield components and cluster morphology: Yield, crop load, cluster compactness, and percentage of visible surface area infected with rot (rot severity) were significantly affected by ELR and CT treatments as compared with the C (Table 1). There were no significant differences in yield and yield components between the timing ELR (at TBLR and FSLR). CT was the most effective treatment for reducing crop level. Compared to the C vines, yield was 49%, 14%, 13% lower in CT, TBLR and FSLR vines, respectively (Table 1).
Crop load, determined by the Ravaz index, was only significantly reduced in the CT treatment (Table 1). As expected, CT decreased yield through a reduction in the number of clusters per vine, which had 30.6% fewer cluster per vine than the C (Table 1). Cluster weight was higher in the CT vines as compared to the other treatments (Table 1).
Levels of cluster area infected by rot were significantly lower in the FSLR compared with the C. Correspondingly, FSLR vines had the loosest clusters, with an assessed visual cluster compactness value of 2.88 out of 5 (Table 1).
Berry composition, berry weight, and number of berries per cluster: For the distal cluster, there were not significant differences in the number of berries per cluster, berry weight and berry maturity parameters (TSS, pH, TA) amongst treatments (Table 2). For the basal cluster, berry weight was higher in the CT vines as compared to TBLR, but there were not significant differences in berry number per cluster and TA amongst treatments. Due to a significant treatment by year interaction effect, TSS and pH data of the basal clusters were analyzed by year (Table 3). In 2015, FSLR fruit had a significantly higher TSS compared to the other treatments, however, there was no difference in pH between treatments. In 2016, CT exhibited higher TSS and pH as compared to the C. Data for categorized berries (shot, green ovaries, and rot berries) will be analyzed in early 2018.
Canopy characteristics: Mid-season (16 July 2015 and 28 July 2016) TBLR and FSLR had significantly lower canopy density (leaf layer number; LLN) as compared with C (Table 4). The significant decrease in LLN from the C was maintained through pre-harvest (25 August 2015 and 7 September 2016) in the FSLR treatment. The percentage of gaps and amount of light interception to the leaves (LEFA) and clusters (CEFA) were similar among all four treatments at both time points.
Additional data analysis: Cold hardiness, bud fruitfulness, gas exchange, and economic costs data will be analyzed in during the winter and spring 2018.
Education & Outreach Activities and Participation Summary
The topic of early leaf removal (ELR) as an innovative yield management tool and results from the 2015-2016 growing seasons have been presented in various programs throughout 2017, including presentations and extension publications. Three presentations were given in 2017 that targeted a diverse range of stakeholders, including wine grape producers of the Northeastern and Mid-Atlantic regions and scientists. One newsletter was created in 2017 to address timing of ELR application, as well as potential benefits and drawbacks of using the practice. This newsletter is available to the public through the Penn State Wine and Grapes U. (https://psuwineandgrapes.wordpress.com/) blog. Additional upcoming outreach will be provided through preparation of the data for manuscript publication in early 2018.
In a preliminary assessment of bud tissue cold hardiness, we found no relevant or consistent differences in cold hardiness amongst treatments. Because of these lack of differences, we decided to not perform the time consuming and expensive work of extracting and quantifying carbohydrates in Grüner Veltliner, that would have helped to determine how differences in cold hardiness relate to carbohydrate storage. Instead, we are performing carbohydrate extraction and quantification in a closely related project on intensity (instead of timing) of early leaf removal for crop load management in Vitis hybrid ‘Chancellor’, which will additionally be correlated with cold hardiness DTA data. This quantification will incorporate three different storage tissues (roots, trunk, and cane) across three time periods (November, January, and March) for both soluble carbohydrates and non-structural starches associated with cold hardiness. This process is currently ongoing and will be completed in early 2018.