Abstract
Adequate yeast assimilable nitrogen (YAN) levels in grape juice are necessary for yeast cells to complete fermentation to dryness. Nitrogen (N) uptake by grapevine roots varies seasonally; therefore, environmental conditions and cultural practices can affect grapevine N status. In addition, genetic differences between rootstock cultivars can influence root dynamics and, subsequently, N uptake, canopy biomass, and fruit composition. Two rootstock cultivars, 1103P and 101-14 Mgt, were fertilized with nitrogen during spring or fall or received no treatment. Vine biomass, leaf N concentration, fruit composition, juice amino-N levels, and fermentation kinetics were measured. The rootstock 1103 Paulsen (Vitis berlandieri × V. rupestris cv. 1103P) has a root system that tends to produce large canopies and high shoot growth. The rootstock 101-14 Millardet et de Grasset (V. riparia × V. rupestris cv. 101-14 Mgt) has a root system associated with smaller canopies and moderate shoot growth. The scion Merlot (V. vinifera L. cv. Merlot clone 1) was grafted onto the two rootstocks in an experimental block in Oakville, California. Merlot on 1103P had higher YAN levels and completed fermentation faster compared to Merlot on 101-14 Mgt. Differences in fermentation kinetics were observed within rootstock N treatments that were not explained by YAN levels, indicating that other factors related to N metabolism may play important roles in fermentation dynamics. Results indicated that Merlot grown on 1103P in the Napa Valley may require little to no N supplementation while Merlot on 101-14 Mgt may require N supplementation to avoid slow fermentations.
Nitrogen (N) is the most important macronutrient for vegetative growth and reproductive development in plants, including grapevines. In winemaking, N is a critical nutrient for yeast cell proliferation (Bisson 1991). Grape juice N concentration and composition can influence wine composition and character (reviewed by Bell and Henschke 2005). YAN is the most widely accepted index of nutritional N available to yeast cells, as it accounts for concentration and form of amino-N and an adequate level is critical for successful fermentations. Estimates for the minimum amount of YAN needed in grape must to complete fermentation to dryness range from 70 to 267 mg N/L, with 140 mg N/L the commonly used threshold (Bell and Henschke 2005). YAN consists of ammonium (NH4+) and all free amino-N compounds but does not include proline. To compensate for low YAN, winemakers typically add diammonium phosphate (DAP) to the juice. Yeasts preferentially assimilate free ammonium ions over amino acids, which can reduce the complexity and desirable flavor aromas of wine (Bell and Henschke 2005, Bisson and Butzke 2000). Excessive N in juice can cause urea production, which might result in the formation of the carcinogen ethyl carbamate (Ough et al. 1989). It is complicated to manage a vineyard so that grape N levels fall between these low and high extremes because many factors influence vine N status, including the environment, cultural practices, and genetic variability (Bell and Henschke 2005).
Researchers have described four seasonal phenological phases of grapevine N supply and demand (Conradie 2005, Kliewer 1967). From budbreak to the end of bloom (phase I), the vine can utilize N reserves stored during the postharvest period of the previous season to supply new growing shoots. From the end of bloom to veraison (phase II), shoot elongation and grape growth is accelerated, the rate of nitrogen uptake from roots is greatest, and the bulk of woody vine biomass is established (Conradie 2005). From veraison to harvest (phase III, fruit ripening), vine growth may slow depending on water supply, while grape clusters continue to be an N sink, but to a lesser extent as ripening progresses (Conradie 2005). Harvest to leaf fall (phase IV) is the period of leaf senescence when N is translocated to and stored in the vine’s permanent structures (wood scaffold and framework roots). Stored N is needed for new growth and fruit set in the following spring (Conradie 2005, Keller et al. 2001, Wermelinger 1991, Williams 1991). These phases represent a general timeline for a woody perennial such as the grapevine; however, they do not account for the diversity of a rootstock root system size and root flush timing (Eissenstat et al. 2006). At least two investigations have indicated that the application of postharvest N during phase IV (fall) can be critical to the N status of the vine in phase I (spring) of the following season (Bettiga and West 1991, Wermelinger 1991), but another has indicated that the effect may be dependent on rootstock (Holzapfel and Treeby 2007).
Grafting a scion onto a rootstock is a standard viticultural practice in North America in order to create a vine resistant to pests (mainly phylloxera) and diseases and able to withstand adverse soil conditions such as high lime content. For winegrapes, the scion is usually a V. vinifera cultivar and the rootstock consists of a number of non-V. vinifera species and crosses thereof. Rootstocks that confer low biomass production onto the scion canopy are desirable in valley floors or other deep, fertile soils. Lowered biomass can reduce labor for canopy management, improve effectiveness of fungicide applications, and increase light levels in the fruiting zone, thus reducing bunch rot and improving flavonoids in red and black grape varieties (Downey et al. 2006). In the Napa Valley region of California, 101-14 Millardet et de Grasset (V. riparia × V. rupestris cv. 101-14 Mgt) and 1103 Paulsen (V. berlandieri × V. rupestris cv. 1103P) are popular rootstocks that produce the upper and moderate extremes of canopy growth when Merlot is the scion. 101-14 Mgt attenuates shoot biomass while 1103P is associated with production of a larger overall woody scaffold and leaf canopy (Wolpert et al. 2002). The primary genetic factor(s) that causes differences in biomass production between rootstocks is not clearly understood. However, factors that are known to influence vine biomass production are water availability, nutrient availability, and root hormonal signaling processes (Dry and Loveys 1998). The observed differences in root dynamics particularly between these two rootstocks contribute to the complexity of managing vine water and N status. The roots of 101-14 Mgt are smaller, shallower, and conduct less water than the roots of 1103P (Alsina et al. 2011). The roots of 1103P produce more fine roots in the summer in response to irrigation and less in the winter months than the roots of 101-14 Mgt (Bauerle et al. 2008). The 101-14 Mgt root system has been reported to produce approximately three-fold more roots in the winter months (December to February) than the 1103P root system, averaged over three years from 2003–2005 (Bauerle et al. 2008), suggesting that 101-14 Mgt rootstocks may benefit more from postharvest N fertilizer application than 1103P rootstocks.
Ough et al. (1968) reported that “the choice of rootstock for a wine variety can alter the juice composition significantly and thus affect the fermentation rate of the must to an economically important degree.” They noted that a rootstock conferring high growth rates on a scion Chardonnay tended to acquire more N, resulting in higher amounts of must amino-N than the intermediate and low growth promoting rootstocks. Some studies have reported on a positive interaction between N application × rootstock-scion combination for grape juice amino-N (Holzapfel and Treeby 2007, Treeby et al. 1998), but none of these studies report subsequent fermentation dynamics. More field-based rootstock and scion interaction research covering the diversity of environmental, cultural, and rootstock-scion combinations will improve our knowledge base for vineyard establishment and management decisions.
The objective of this study was to investigate the response of Merlot juice amino-N levels and fermentation kinetics when N application was undertaken either in spring (phase I) or fall (phase IV), as contrasted with no N application, for two rootstock cultivars (1103P and 101-14 Mgt) that differ in biomass production. The same clone of a Merlot cultivar (scion) was grafted onto each rootstock, and grapevine biomass and N status, grape fruit yields and composition, juice amino-N composition, and fermentation kinetics were measured.
Materials and Methods
Plant material.
Vitis vinifera L. cv. Merlot (clone 1) grafted onto two rootstock cultivars, 1103 Paulsen (V. berlandieri × V. rupestris cv. 1103P) and 101-14 Millardet et de Grasset (Vitis riparia × V. rupestris cv. 101-14 Mgt) were used in this study. These two rootstocks differ in shoot biomass production as well as root proliferation and drought tolerance (Alsina et al. 2011, Bauerle et al. 2008, Carbonneau 1985). Rootstock 1103P is commonly considered by viticulturists to accelerate vine biomass development, while the rootstock 101-14 Mgt is associated with attenuated vine biomass development, depending on site (Wolpert et al. 2002).
Field site.
The experiment was carried out during the 2004–2005 season in a 1.05 ha vineyard in Oakville, California (Napa Valley) (38°25’N; 122°24’W). The vines were in their tenth year and trained as bilateral cordons with vertical shoot positioning. The trellis height was 1.6 m and the canopy was hedged to 2.2 m. The vines on 1103P were established and managed with 28 nodes while those on 101-14 Mgt were established and managed with 24 nodes. The rows were oriented southeast to northwest and spaced 2.4 × 2.2 m between and within rows. The vines were planted in 1995 in Bale gravelly loam (fine-loamy, mixed, superactive, thermic Cumulic Ultic Haploxeroll). Mineral soils in the 0 to 20 cm range had total soil C of 2.48 ± 0.03%, N of 0.21 ± 0.01%, KCl-extractable NH4+-N of 3.02 ± 0.25 mg/kg, and NO3-N of 2.75 ± 0.26 mg/kg (Carlisle et al. 2006). The Oakville region averages 830 mm of precipitation annually and has a mean annual temperature of 14.3°C. An onsite weather station reported 1125 mm of precipitation and 1165 mm of reference evapotranspiration (ETo) during the study period between October 2004 and September 2005 (CIMIS 2003–2007; http://www.cimis.water.ca.gov/cimis/). Soil volumetric water content, midday stem water potentials, and fine root production dynamics (<2 mm) were measured and reported by Bauerle et al. (2008).
Experimental design and treatments.
The experimental vineyard was laid out in a randomized complete block design with six blocks. Each block consisted of subplots of six vines per rootstock × treatment combination. The data rows were bordered on each side by guard rows of the same rootstock × treatment combination. The treatments consisted of two rootstocks (1103P and 101-14 Mgt) fertilized in either the spring (spring) or fall (fall) or not fertilized at all (control). The spring treatment was 16.8 kg N/ha applied on 18 May 2005 and the fall treatment was 16.8 kg N/ha applied on 7 Oct 2004 using potassium nitrate (36.94% K, 13.75% N) applied through the drip line. These vines were previously fertilized in spring 2003. Water was withheld to restrict new leaf area production from the period of fruit set to veraison (onset of ripening) beginning in 2002 and continuing in each subsequent year. Irrigation began on 1 Aug 2005 when midday leaf water potentials reached a threshold value of −1.0 to −1.2 MPa. Leaf water potentials were measured using a pressure chamber (Soil Moisture Inc., Santa Barbara, CA). Replicated (n = 6) midday leaf water potentials were measured on 20 July, 29 Aug, and 31 Aug 2005 for the spring N treatments on both rootstocks. The fourth, fifth, or sixth leaf was selected on a randomly chosen cane, placed in a plastic bag, and immediately inserted into the pressure chamber. Estimates of crop evapotranspiration (ETc) were calculated using the reference evapotranspiration measured from the onsite CIMIS station and adjusted using a grape crop coefficient (Williams et al. 2003). The vines were deficit irrigated biweekly with 40% replacement of ETc, which totaled 60 mm over the course of the summer. Irrigation water and fertigation solution were delivered via one 3.8 L/hr drip emitter per vine, located 50 cm from each trunk. Water was delivered to only one side of each vine and on the same side throughout the study to facilitate a parallel and concurrent study that mimicked partial rootzone drying irrigation (Bauerle et al. 2008). Bloom began on 12 May, veraison began on 26 July, and harvest was on 28 Sept 2005.
Grapevine N status and vine biomass.
Vine N status is often measured as either percent leaf N (%N) or concentration of petiole NO3-N (ppm) at flowering and/or veraison, but concentrations vary depending on variety and rootstock and neither technique is consistently reliable (Robinson 2005). Leaf %N was chosen in order to estimate total N partitioned to annual canopy biomass. Twelve leaf samples in each plot were collected on 1 June and 1 Aug and analyzed for total N. Only leaves opposite a fruit cluster were selected and combined. The leaves were oven dried (60°C) and ground using a Wiley Mill to pass through a 60 mesh sieve. Four milligrams of leaf tissue were weighed into tin capsules and sent to the University of California (UC) Davis Stable Isotope Facility for combustion analysis using a PDZ Europa ANCA-GSL elemental analyzer (Sercon Ltd., Cheshire, UK). In May, excess shoots (leaves plus green canes) were removed and collected from four vines in each plot, leaving 28 shoots on 1103P and 24 shoots on 101-14 Mgt. In July, leaves were removed and collected from the fruiting zones of four vines in each plot. All remaining leaves were removed from the lignified shoots (canes) from one cordon (1.1 m) in each plot in October following harvest (n = 6). All biomass removed was dried at 60°C in a forced air oven and weighed. The biomass weights were calculated on a per-vine basis and canopy biomass was calculated as the sum of biomass collected in May, July, and October. Leaf canopy total N was calculated using the canopy biomass and the total %N in August. The remaining cane biomass was collected as pruning weights during February of the following year and was not included in calculations of leaf canopy total N.
Fruit composition, amino acids, and yield.
On 26 Sept 2005, ~200 berries were randomly sampled from four data vines in each field replicate, and tested for Brix, pH, and titratable acidity (TA). Brix was measured with a temperature compensated refractometer (American Optical Corp., Southbridge, MA). TA was measured using established protocols (Amerine and Ough 1980) and pH was measured using a pH meter (model 720 plus; Thermo Orion, Beverly, MA) also using established protocols (Amerine and Ough 1980). A 1.5 mL juice subsample from each field replicate was clarified by centrifugation, frozen, and sent to the UC Davis Molecular Structure Facility for amino acid analysis. The samples were acidified with 10% sulfosalicylic acid to precipitate proteins, then diluted 1:5 using a lithium citrate buffer with an AEcys internal standard. After centrifugation, 50 μL was loaded into a Beckman 6300 amino acid analyzer (Beckman Coulter, Indianapolis, IN), using ion-exchange chromatography to separate amino acids followed by a post-column ninhydrin reaction detection system. YAN was calculated by adding the concentrations of all free amino acids (mg N/L) available to yeast plus ammonium (Bell and Henschke 2005). The yeast assimilable N to yeast nonassimilable N (YAN:YNAN) was calculated by dividing YAN by the sum of proline and hydroxyproline (Bell and Henschke 2005). On 28 Sept, the grapes were harvested by hand between the hours of 0700 and 1000. Grape yield (kg) and number of clusters per vine were recorded. No diseases were observed.
Fermentation.
Approximately 90 kg of grapes per treatment were transported in half-ton macrobins to UC Davis for winemaking. The maximum air temperature that day was 34°C. The grapes were stored in a cold room set at 15°C overnight. The grapes were crushed the following day at the UC Davis winery using a 4.5 tonne per hour crusher/stemmer (Valley Foundry and Machine Works, Fresno, CA). A total of 136.2 L of must was kept for each treatment after crush and was split into triplicate replicates of 45.4 L each in 56.8 L HDPE cylindrical tanks (Nalgene) for fermentation in a non-temperature-controlled room. Sulfur dioxide was added to each tank to achieve a total concentration of 50 mg/L using a 5% solution of potassium metabisulfite. Following mixing, Pasteur Red yeast (Lasaffre Yeast Corp., Milwaukee, WI) was rehydrated according to manufacturer’s instructions and was added at a rate of 5 g/L. Brix values were measured daily until dryness (three consecutive negative Brix readings) using a calibrated hydrometer. The must was pressed at dryness on 17 Oct 2005 and transferred into 20 L glass carboys. Viniflora Oenos (Chr. Hansen, Hørsholm, Denmark) was added directly at a rate of 0.16 g/20 L, following manufacturer’s instructions to initiate the malolactic fermentation. The final volume for each treatment was ~17 L. After malolactic fermentation was complete, the wines were racked on 12 Apr 2005 and bottled in 750 mL standard wine bottles. Initial fermentation rate was calculated as Brix/day using the slope of the linear regression for the first five days.
Statistical analysis.
A linear mixed effect model was used to assess the effects of rootstock and fertilizer treatment on various yield variables. Fixed effects were included for rootstock, fertilizer treatment, and their interaction. Random effects were included for blocks and all block-by-fixed-effect interactions. To assess fertilizer rate effects, orthogonal contrasts were used to separate main effects and interaction effects into a control versus average of spring and fall tests and a spring versus fall test. In cases of interactions, the simple effects of fertilizer were tested within rootstock. Model fit was assessed using graphical analysis of residuals and the Shapiro-Wilk test for normality. A natural logarithmic transformation was used where appropriate. All analyses were performed using SAS for Windows (ver. 9.2; SAS Institute, Cary, NC). Statistical significance was declared at α ≤ 0.05.
Results
Prior to the onset of irrigation, on 20 July 2005, the water status of 1103P was higher than that of 101-14 Mgt (Table 1). After weekly irrigations began on 1 Aug 2005, no difference in water status was observed between the two rootstock treatments (Table 1). The balance of canopy vegetation versus yield of fruit, the Ravaz index, indicated that all treatments and rootstock/scion combinations were in the desirable range of 5 to 10 (data not shown) (Ravaz 1911).
The 1103P vines had more leaf N in both June (bloom) and August (veraison) than 101-14 Mgt vines (Table 2). The 1103P vines had 40 to 45% higher aboveground biomass than 101-14 Mgt (Table 3), and therefore more total N was acquired by 1103P and fixed into canopy biomass than by 101-14 Mgt (Table 2). During veraison, a rootstock × fertilizer interaction was observed in leaf %N. Fertilizer rate (control vs. average of spring and fall), not timing of N (spring vs. fall), drove the interaction effect. In the fertilized vines, leaf %N decreased in 1103P and increased in 101-14 Mgt compared to the control (Table 2).
Rootstock differences were seen in all vineyard response variables except in the pH of the fruit juice (Table 3). With the exception of fermentation rates, timing of N had no effect on any of the variables. The only fertilizer rate effect was observed in pruning weights. Pruning weights increased when N was applied. A rootstock × fertilizer interaction was observed in the fruit composition variables and it was the fertilizer rate effect that drove the interaction. Within the 101-14 Mgt treatments, application of fertilizer generally increased Brix by 0.8 compared to the nonfertilized control, whereas a drop in Brix by 0.50 was observed in 1103P compared to the nonfertilized control. Application of fertilizer increased the pH of 101-14 Mgt juice by 0.08 compared to the control but no change was seen in in pH of 1103P juice. Application of N fertilizer, whether in spring or fall, decreased TA by 0.46 g/L in 101-14 Mgt juice compared to the control, whereas in 1103P there was a trend of increasing TA compared to the control (Table 3).
The only measured differences in juice amino N were between rootstocks. Rootstock differences were observed for YAN, ammonium, YNAN, and all other amino acid levels; no fertilizer timing, fertilizer rate effects, or rootstock × fertilizer interactions were observed (Table 4, Table 5). YAN was 2.5× higher in 1103P juice than in 101-14 Mgt juice (Table 4). Proline was the dominant amino acid in this Merlot juice on both rootstocks and is the major constituent of yeast nonassimilable amino-N (YNAN). Unlike the more common proline to arginine ratio used to determine grape yeast N status, the YAN to YNAN ratio (YAN:YNAN) provides a broad-based N nutritional index for yeast growth during fermentation because it accounts for ammonium availability and all other primary amino-N sources (Bell and Henschke 2005). The 1103P treatment had almost double the YAN:YNAN ratio of 101-14 Mgt treatment (Table 4). Therefore, not only did 101-14 Mgt have considerably less YAN than 1103P, but more of the total N in the 101-14 Mgt juice was found in proline, which cannot be utilized by yeast under anaerobic conditions. On the other hand, the concentration of arginine was six times higher in 1103P juice than in 101-14 juice, while most other amino acids were approximately two times higher (Table 5).
Differences emerged in the distribution of representative amino acids into amino acid family groups between the two rootstocks; that is, amino acids that share common biosynthetic pathways (Table 5). For both rootstock treatments, the majority of amino acids detected were in the glutamate family. However, in fruit from 101-14 Mgt, a higher percentage of the total amino acids was found in the glutamate family compared to fruit from 1103P (85% vs. 82%). Instead, in 1103P fruit, a higher percentage of amino acids was diverted in the aspartate (5% vs. 3%) and pyruvate (7% vs. 5%) families compared to 101-14 Mgt fruit. Within the glutamate family, 101-14 Mgt fruit synthesized more γ-aminobutyric acid (GABA) relative to the free amino acids than 1103P fruit (21% vs. 14%). Conversely, 1103P fruit synthesized more arginine relative to the free amino acids in 101-14 Mgt fruit (19% vs. 7%). These percentages were averages of all treatments within rootstock; no differences were found between fertilizer treatments within rootstock (data not shown).
Fermentation rates were lower for 101-14 Mgt juice than for 1103P; 101-14 Mgt completed fermentation an average of six days later than 1103P, varying somewhat by N treatment (Table 4, Figure 1). In addition to the fermentation rate differences between rootstocks, a fertilizer rate effect was observed within rootstock N treatments; compared to the control, the application of N generally increased initial fermentation rates (Table 4). In addition, there was a fertilizer timing effect within the 1103P treatments, whereby spring N correlated with faster initial fermentation rates than fall N. For 101-14 Mgt, fermentation from the nonfertilized control was slower throughout fermentation than either of the seasonal N application treatments for that rootstock (Figure 1).
Discussion
This study documented the amino-N and fermentation response to seasonal fertilizer (N) application in Merlot on two rootstocks that differ in potential growth rates and biomass production, root foraging, and drought tolerance (Alsina et al. 2011, Bauerle et al. 2008). Grapevine N status and biomass productivity should be strongly correlated with a rootstock system’s ability to acquire and assimilate nitrogen from the soil N pool. Peak production of fine roots occurs during the late spring/early summer root flush (Eissenstat et al. 2006, Volder et al. 2005), and it has been shown that these fine roots absorb N actively during the first few days of their lifetime (Volder et al. 2005). The concurrent study of root foraging by these vines (Bauerle et al. 2008) showed that the 101-14 Mgt root system produced more new roots during winter compared to the 1103P root system, which produced more new roots during the summer months. This study reports on the vines response to seasonal N fertilizer application in relation to the timing of fine root flushes between these two rootstocks.
According to a thorough review of the effects of N nutrition on grape composition (Bell and Henschke 2005), applying N generally increases YAN. YAN is measured by winemakers to ensure enough yeast assimilable N is available to promote yeast growth and support fermentation to dryness (Bisson 1991). YAN was expected to be the variable to most strongly affect fermentation rates because the same clonal yeast population was inoculated into all fermentations and the Brix levels were within ±1 range. However, juice YAN levels did not significantly increase when 16.8 kg N/ha fertilizer was applied compared to the unfertilized control. It was interesting that fermentation rates increased when N fertilizer was applied compared to the no N fertilizer control despite negligible differences in YAN. Factors other than YAN, perhaps also related to N metabolism, may have influenced fermentation rates.
For Merlot on 101-14 Mgt, applying N fertilizer to the vines in the fall was just as effective at increasing fermentation rates as applying N fertilizer in the spring. For Merlot on 1103P, applying N fertilizer in the spring was more effective at increasing fermentation rates than applying N fertilizer in the fall. This finding supports that a proportionally greater winter root flush was seen in 101-14 Mgt vines compared to 1103P vines and that 1103P vines had a proportionally greater summer root flush (Bauerle et al. 2008). It is important to note that the initial levels of YAN in juice from 101-14 Mgt grapes were below estimates of the minimum amount of YAN needed in grape must to complete fermentation, yet sugar consumption was completed in these musts over time. These results suggest that N fertilizer application may have more benefits to yeast nutrition than the simple provision of ammonia and amino acids.
The N status of a plant is strongly dependent on N availability, which in turn depends significantly on water availability. The lack of N or water can elicit a stress response (Taiz and Zeiger 2010). The results of this study indicated that the 101-14 Mgt vines were N stressed. Plant stress can affect carbon:nitrogen metabolism, shifting amino acid metabolism toward the synthesis of GABA, for example, which is in the glutamate family (Bouche and Fromm 2004). In this study, there were higher levels of GABA in 101-14 Mgt fruit than in 1103P fruit and higher levels of arginine in 1103P fruit than in 101-14 Mgt fruit. GABA is a four-carbon nonprotein amino acid that can account for 10 to 25% of the free amino acids and is utilized by the yeast Saccharomyces cerevisiae (Bach et al. 2009). GABA can have a positive effect on wine composition, specifically for the metabolite succinate (Bach et al. 2009). Certain amino acids, such as arginine, are favored by S. cerevisiae, and the initial concentrations of these amino acids can affect fermentation rates (Monteiro and Bisson 1991). The concentration differences of GABA and arginine between rootstock treatments may have contributed to the differences in fermentation rates between rootstock treatments. However, the amino acid data do not explain the fermentation rate differences within rootstock N treatments.
Conclusion
This study examined the effects of viticultural practices of N fertilization on fermentation kinetics of two rootstocks with different root dynamics and biomass production. The Merlot scion responded very differently depending on whether it was grafted to 1103P or 101-14 Mgt. The Merlot fruit from 1103P had approximately two times more YAN than the fruit from 101-14 Mgt, resulting in faster fermentation rates. The application of N at a moderate amount of 16.8 kg N/ha increased fermentation rates on both rootstocks. Seasonal differences in application of N affected fermentations differently depending on rootstock. In 101-14 Mgt must, both fall and spring N applications increased fermentations proportionally compared to the control. In 1103P must, spring N fertilizer application increased fermentation rates more than fall N application. Regardless of N fertilizer application or timing, the 101-14 Mgt root system was not able to acquire adequate levels of N for timely completion of fermentation to dryness. Within the rootstock treatments, YAN alone did not explain differences in fermentation rates. Fruit from 101-14 Mgt produced more GABA than fruit from 1103P, an amino acid synthesized in response to stress and measured as a proportion of the total free amino acids assimilated by wine yeast, suggesting that these vines were stressed under the overall conditions of these experiments.
Acknowledgments
Acknowledgments: This project was financially supported by the American Vineyard Foundation, California Competitive Grants Program for Research in Viticulture and Enology, the Maynard A. Amerine Endowment, and Jastro Shields Graduate Research Scholarship. The authors thank Geoffrey Dervishian, Travis Pritchard, Helina Chin, and Alison Breazeale for technical assistance in the field and laboratory and especially Jason Benz, Vineyard Manager at the Oakville Research Station, for all of his assistance and support.
- Received April 2012.
- Revision received December 2012.
- Accepted January 2013.
- Published online June 2013
- ©2013 by the American Society for Enology and Viticulture