Abstract
Nitrogen (N) fertilization experiments were maintained for three growing seasons in commercial Cabernet Sauvignon (CS), on Vitis riparia rootstock, and Merlot (M), on SO4 rootstock, vineyards in British Columbia. Six treatments were applied annually in a randomized block design with eight replicates and 10 vines per experimental plot. Treatments included the standard commercial vineyard N application rate of 40 (M) or 45 (CS) kg N/ha or double these rates, applied either at budbreak or bloom. A split N treatment involved application of the low N rate at bloom, followed by the same amount applied immediately postharvest. An organic N treatment involved surface application at budbreak of compost estimated to supply the standard N application rate. For Merlot, petiole N concentration before and at bloom was highest each year after application of 80 kg N/ha at budbreak. In two years, delaying N application until bloom was associated with decreased canopy density and increased yield. Berry yeast-assimilable N concentration (YANC) exceeded deficiency only for the high N rate applied at bloom. Bloom-time N application for Merlot has promise as a strategy for targeting berry N status without causing excess vegetative growth. There were few differences in vine performance between those receiving N as compost rather than urea at budbreak. Postharvest N applications did not affect fruit composition or YANC. The Cabernet Sauvignon site had adequate YANC regardless of N treatments. Altering N rate and timing had few multiyear effects on canopy density, yield, and fruit composition. Applying budbreak N as compost rather than urea resulted in similar vine performance. The postharvest N treatment increased cumulative yield without adverse effects.
Nitrogen fertilization can have important effects on grape composition and hence wine quality (Jackson and Lombard 1993, Bell and Henschke 2005). Reports of excessive vine vigor, resulting from high N application rates (Spayd et al. 1993, Bell and Robson 1999) or vigorous rootstocks (Wolf and Pool 1988), have caused many growers in the Okanagan Valley in the southern interior of British Columbia to apply N at modest rates of ~40 to 60 kg N/ha/yr. This strategy persists despite reports of vineyards with slow or stuck fermentations, necessitating the application of high rates of diammonium phosphate to the must (Bell and Henschke 2005). A survey of Vitis vinifera vineyards in the Western coastal states of the United States indicated yeast-assimilable N deficiencies in ~14% of all vineyards (Butzke 1998).
Altering the timing of N application can affect vine vigor and yield (Peacock et al. 1991, Christensen et al. 1994) and inadequate N availability at bloom can reduce yield (Keller et al. 1998). Recent European research has indicated that higher must N concentrations result from N applications at fruit set than at budbreak (Linsenmeier et al. 2008), but there has been little consideration of this research under North American growing conditions. In the Okanagan Valley, N is generally applied to the soil at budbreak despite partitioning studies on grapes (Conradie 1990, 1991) and apples (Neilsen et al. 2001), indicating that early season vegetative growth is associated with previously stored N reserves rather than N applied to the soil in spring. The same research also indicates that fruit N is higher when N is applied postbloom. Application of N later in the growing season, including postharvest, has received little attention locally since grape production occurs in a short season where harvest and leaf senescence occur closely together and growers are concerned about wood hardening to improve winter hardiness. Historically, winter freezes have been a major influence on grape yield when the industry was based primarily on hybrids (Caprio and Quamme 2002).
Recently, N form (organic versus inorganic) has received increasing attention for vineyard application (Conradie 2001, Morlat 2008) because of concern for the low organic matter content of many viticultural soils (Morlat and Chaussod 2008), which can result in deterioration of soil physical properties and create nutrient imbalances in coarse-textured soils such as those in the Okanagan Valley (Neilsen et al. 2003). For these reasons, field experiments were established in representative vineyards of two major commercial cultivars, Merlot and Cabernet Sauvignon, to investigate the effects of rate, timing, and form of N application on vineyard performance including fruit quality. Emphasis was placed upon vine nutrition, yield, and berry yeast-assimilable N.
Materials and Methods
Two commercial vineyards in the irrigated, grapegrowing region of southern interior British Columbia, with a history of modest N application rates ranging from 40 kg N/ha (Merlot) to 45 kg N/ha (Cabernet Sauvignon) at budbreak, were selected for a N fertilizer trial. The Merlot site, owned and operated by Vincor Canada, was located on a bench ~50 m above Lake Osoyoos, which straddles the Canada/U.S. border (lat. 49°N). The Cabernet Sauvignon vineyard, independently owned, was located at the same elevation on the same bench, 7.5 km to the north. Soil at both sites was an Osoyoos loamy sand, extensively planted to winegrapes (Vitis vinifera L.). The soil properties in the vineyards were typical of this soil series, with low organic matter, coarse texture (70–80% sand), and limited nutrient- and water-holding capacity (Wittneben 1986). Commercial analysis (Waters Agricultural Labs, Camilla, GA) of four composite soil samples collected at each experimental site (0–15 cm depth) before initiation of the experiment indicated similar soil properties including, at the Merlot site, pH (soil:water) averaging 7.7, organic matter 1.7%, Mehlich-I-extractable P and K averaging 66 mg/kg and 158 mg/kg, respectively, and cation exchange capacity 5.1 meq/100 g. The same properties averaged 7.9, 1.9%, 73 mg/kg, 163 mg/kg, and 5.6 meq/100 g at the Cabernet Sauvignon location. Soil NO3-N values (extracted in 2 M KCl) were very low (≤5 mg/kg) at both sites, consistent with extensive research in the area, indicating that soil solution N concentrations are very low in coarse-textured soils of low organic matter content after the irrigation (growing) season when low rates of N have been applied (Neilsen et al. 1998).
Merlot, on moderately vigorous SO4 rootstock (Vitis berlandieri x V. riparia), was planted in 1999 at a 1.25 m (within-row) x 2.5 m (between-row) spacing. Cabernet Sauvignon was on low-vigor V. riparia rootstock planted at 1.17 m (within-row) x 2.5 m (between-row) spacing in 1998. At both sites, the same experimental design was maintained annually from 2005 to 2007. Six N treatments were applied in a randomized complete block design to 10-vine plots with eight replicates. The first and last plant in each plot served as a guard plant separating different treatments and all sampling and measurements were of the central eight plants. Treatments included: (1) standard commercial (S) N application at budbreak of 40 kg N/ha/yr (Merlot) and 45 kg N/ha/yr (Cabernet Sauvignon); (2) twice the standard rate (2S), also applied at budbreak; (3) S applied at bloom; (4) 2S applied at bloom; (5) the same as treatment 3, but with supplemental N applied immediately postharvest at 40 kg N/ha/yr (Merlot) or 45 kg N/ha/yr (Cabernet Sauvignon) and, (6) the S rate, as estimated to be available from annual applications of a poultry-based compost, applied at budbreak. All N was applied as urea (46N-0P-0K). All per area N treatments were calculated and uniformly applied, based on the 10-vine plot length within a 2-m-wide weedfree herbicide strip, centered on the vine row. For the compost treatment, annual N availability was estimated as 30% of total N applied in 2005 and 20% of total N applied in 2006 and 2007 (Table 1⇓), based on typical mineralization values measured after application of organic N to British Columbia soils (Zebarth et al. 2000).
All Merlot plots received a standard commercial PK fertilizer application within the herbicide strips at 40 kg P/ha/yr as triple superphosphate (0N-45P-0K) and 60 kg K/ha/yr as potassium chloride (0N-0P-60K). At both sites, insect, disease, weed control, and sprinkler irrigation from May to September followed commercial recommendations (BCMAL 2006). Irrigation applications were measured in detail at the Merlot site (Figure 1⇓), with amounts adjusted for changes in precipitation and evapotranspiration. A similar schedule was followed at the Cabernet Sauvignon location. Here, irrigation sets were 8-hr duration (0.39 cm/hr) every 12 days from bloom to veraison and weekly from veraison until harvest, resulting in less water applied from bloom until veraison and more water applied between veraison and harvest for Cabernet Sauvignon. Vines were trained to bilateral cordons at 1 m and were spur-pruned to retain 18 to 20 buds per vine each year. Shoots were vertically positioned using three pairs of upper catch wires. Shoot and cluster thinning followed commercial recommendations (BCMAL 2006).
Petiole samples were collected three times annually, 2005 to 2007, at each site for each plot. The first annual sample (end May/early June) was a composite of 16 petioles (2 petioles x 8 vines) from newly mature leaves at shoot tips. Composite 32-petiole samples (4 petioles x 8 vines) were collected from leaves near flowering clusters at full bloom and from mature leaves in the upper third of the canopy at veraison. The sample dates varied by cultivar and year (Table 2⇓). The petioles were oven-dried at 65°C and ground in a stainless-steel Wiley mill (A.H. Thomas, Philadelphia, PA). Petiole N was determined on a 0.5 g sample with a LECO FP-528 (Leco Corporation, St. Joseph, MI) combustion analyzer (Sweeney and Rexroad 1987). Another 0.5 g sample was dry-ashed at 525°C and dissolved in 1.2 M HCl prior to determination of P, K, Ca, Mg, Zn, Fe, Mn, and B by inductively coupled argon plasma (ICP) spectrophotometry (Spectro Analytical Instrument, SCP Science, Vancouver, BC).
Canopy density was estimated each year at veraison. Two visual assessments of percent open space were averaged, one on each side of the row, at the same canopy location for each treatment. Open space was estimated in the area defined by the two central vines and the bottom fruiting and topmost catch-wire. Total harvest weight (kg/vine) was measured each year at commercial harvest for each cultivar and treatment plot from the four central plot vines. A random sample consisting of eight clusters per plot (one from each measurement vine) was used to calculate mean cluster weight. Juice was expressed from 144 berries, randomly selected from the upper, middle, and lower third of each sample cluster for analyses of pH, soluble solids concentration (SSC) by refractometry, and titratable acidity (TA) by titration with 0.1 M NaOH to a pH 8.1 endpoint and expressed as g/L tartaric acid. Yeast-assimilable nitrogen concentration (YANC) was determined from 15 mL juice using the formol titration method (Gump et al. 2000).
Analysis of variance was performed separately for the Merlot and Cabernet Sauvignon sites. The first four treatments were analyzed as a factorial combination of two N times (budbreak, bloom) and two N application rates (S, 2S) from which the main effects of timing and rate of N and their interaction could be determined. Individual degree of freedom contrasts were used to compare the addition of postharvest N applications to bloomtime N only and budbreak application of urea to compost N using SAS software, ver. 6 (SAS Institute, Cary, NC).
Results and Discussion
Harvest was several weeks earlier for Merlot than Cabernet Sauvignon in October (Table 2⇑). Despite the contrast in harvest timing, there was little difference in bloom and veraison date between the cultivars at the two sites.
Petiole nutrient concentration.
Petiole nutrients other than N were generally adequate for Merlot according to BCMAL recommendations (Table 2⇑). Exceptions included low petiole Mg in 2005 and 2006 and low petiole Fe throughout the three years. There were no Mg or Fe deficiency symptoms observed in any plot during the study. In contrast, petiole P concentrations were above the recommended range for Merlot throughout the study. Although the urea N treatments affected petiole nutrient concentrations in some years, effects were inconsistent across years and unrelated to vine yield (data not shown). Application of N as compost consistently decreased petiole Mg concentration all three years, further reducing already low values (data not shown).
Petiole total N concentration was affected by a rate x time interaction across sample times for Merlot in all years (Table 3⇓). Before and at bloom, except for the May 25 sample in 2007, petiole N concentration was highest after application of 80 kg N/ha at budbreak. At veraison, petiole N was increased only by application of 80 kg N/ha at bloom and not by N applied at budbreak. On four of the nine sample times during the three-year study, petiole N concentrations after application of N as compost were lower than in vines where N was applied at the same rate and time as urea. Application of an additional 40 kg N/ha immediately after harvest did not affect petiole N concentration relative to applying 40 kg N/ha at bloom, except in 2006 when it enhanced the N concentration in petioles sampled on 1 June.
In the Cabernet Sauvignon plots, petiole Zn was below BCMAL recommendations in 2006 and 2007 and petiole Fe was lower than recommended values in all three years (Table 2⇑). No deficiency symptoms were observed and there was no correlation between yield and petiole Fe or Zn concentration (data not shown), indicating no major implications of the low concentrations of these micronutrients. Petiole P concentrations consistently averaged above the recommended range. Compost application consistently affected only petiole Mg concentration, which decreased each year but stayed within the recommended range (data not shown).
N timing affected petiole N concentration at the earliest two sample times in 2005 and 2006, with N applied at budbreak resulting in higher N than that applied at bloom (Table 4⇓). Application rates affected petiole N at veraison in 2006 and 2007. In 2006, there was a rate × timing interaction in which high N applied at budbreak, but not bloom, resulted in higher petiole N concentrations. In 2007, high N rate vines had higher petiole N concentration than the lower rate vines, regardless of time of application. Application of 45 kg N/ha at budbreak as compost rather than urea resulted in lower petiole concentrations in early June and at bloom in 2005 and at veraison in 2006, but otherwise, petiole N concentrations were the same between these treatments.
Local standards recommend a total N petiole concentration at bloom between 6 and 15 g kg−1 for high-vigor cultivars including Cabernet Sauvignon and Merlot (BCMAL 2006). With the exception of Merlot vines receiving 80 kg N/ha (2S) at budbreak in 2006, this N range was observed during the study at the two sites. The addition of N in excess of adequate levels can further increase petiole N concentrations, lead to increased vine growth, and reduce yield due to increased shading in the renewal zone (Bell and Henschke 2005).
For both cultivars, N application at budbreak generally increased total N concentration of vegetative tissues, as seen in the earliest petiole samples collected from recently matured leaves in the upper canopy in late May/early June and in petioles collected lower in the canopy near the fruiting clusters at bloom. N applied at budbreak was rapidly incorporated into the vegetative tissues, which can be a strong sink for N before bloom. Other N timing trials have found that N applications usually result in higher petiole N at the next petiole sampling (Christensen et al. 1994). Delaying N application until bloom had more modest effects on petiole N concentration. Application of 80 kg N/ha at bloom consistently increased petiole N at veraison more than 40 kg N/ha when measured for Merlot, but there were no consistent differences between rates for N applied at budbreak. The N concentrations of Cabernet Sauvignon petioles showed differences between rates applied at bloom and measured at veraison only in 2007. Changes in petiole N concentrations of Merlot vines were thus consistent with partitioning studies indicating different sinks for N applied at different times during the growing season, so that application of N after bloom is more likely to be allocated to reproductive structures (Conradie 1991).
The application rate of compost was based on an estimate of the N that would be mineralized annually to equal the 40 kg N/ha (Merlot) or 45 kg N/ha (Cabernet Sauvignon) supplied by the standard application of urea. The assumed mineralization rates of 20% (years 2 and 3) and 30% (year 1) of total N derived from the new-year compost application were relatively modest, as compost was reapplied each year and continued mineralization of the previous year N was possible. Mineralization of N in biosolids applied to forage in a cooler region of interior British Columbia ranged from 20 to 50% (Zebarth et al. 2000). The lack of consistent differences in petiole N concentrations between urea and compost applied as N sources at budbreak for both cultivars indicates that compost is a suitable source of applied N. However, occasionally (four of nine measurements for Merlot and three of nine measurements for Cabernet Sauvignon) petiole N concentration was lower for vines receiving compost rather than urea. Based on these results, it will be difficult to use surface-applied compost to predictably supply grapevine annual N needs.
Canopy density and yield.
Canopy density was affected by N treatments in 2006 and 2007 for Merlot (Table 5⇓). Canopy density increased as N rate increased from 40 to 80 kg N/ha in 2006. In both 2006 and 2007, canopy density was increased by applying N at budbreak rather than bloom. Application of an additional 40 kg N/ha to Merlot postharvest increased canopy density in both 2006 and 2007 relative to the bloom-time application of 40 kg N/ha. Canopy density was unaffected by application of N as compost rather than urea.
Per vine yield of Merlot was affected by N treatments each year and cumulatively (Table 6⇓). Effects on the yield components of average berry and cluster weight were measured each year, but there was no consistent relationship (data not shown). Thus, this discussion is limited to yield per vine, which integrates the effects of its components. Timing of N application had the most effect on Merlot yield, with application of N at bloom rather than budbreak resulting in higher yields in 2005, 2007, and cumulatively (2005–2007). Increasing the N rate from 40 kg N/ha to 80 kg N/ha reduced yield in 2007, when annual yield was lowest. In 2006 only, vines to which postharvest N was applied in 2005 had higher yield than vines receiving only standard (S) N applications at bloom. Yield was unaffected by application of N as compost rather than urea.
Canopy density of Cabernet Sauvignon was unaffected by N treatments (Table 5⇑). Cabernet Sauvignon yield response to N treatments was minimal (Table 6⇑). Effects on average berry and cluster weight were also minimal (data not shown) and limited over the three-year trial to increased berry weight for the postharvest treatment over the 45 kg N/ha bloom-time application in 2007. The postharvest N treatment involved application of an additional 45 kg N/ha after harvest to vines receiving 45 kg N/ha at bloom and increased yield compared with vines receiving only the bloom-time N in 2007 and cumulatively 2005 to 2007. There was also a significant rate × time interaction in 2005 and cumulatively for yield of Cabernet Sauvignon; higher yields occurred at the standard 45 kg N/ha application rate for N applied at budbreak and at the 90 kg/ha rate applied at bloom. There were no differences in yield between treatments involving N application at budbreak as urea or compost.
Most N fertilization studies place optimal vine vigor and yield at relatively low N application rates, which in irrigated vineyards frequently are less than 100 kg N/ha (Neilsen et al. 1989, Spayd et al. 1993, Bell and Robson 1999). These data are consistent with local industry practices of applying low N rates (40–45 kg N/ha/yr, applied at budbreak) to optimize vine performance (BCMAL 2006). Much less is known concerning the effects of timing of N application on canopy density and yield (Linsenmeier et al. 2008). Results from the Merlot site suggest that canopy open space is increased when N is applied at bloom rather than at budbreak or postharvest. Higher yield of Merlot was associated with bloom rather than budbreak N applications and was consistent with improved yield performance of Riesling fertilized late at fruit set in a long-term N timing trial in Germany (Linsenmeier et al. 2008).
The minimal effects of N applications at budbreak and bloom on annual yield of Cabernet Sauvignon may reflect its higher N status, as indicated by elevated petiole N concentrations throughout the growing season. The cumulative yield effects for Cabernet Sauvignon suggest that superior long-term yield performance occurs with low N applications at budbreak (45 kg N/ha), higher N applications at bloom (90 kg N/ha), or 45 kg N/ha applied both at postharvest and at bloom.
At both the Merlot and Cabernet Sauvignon locations, there were no differences in canopy density and yield between N applied at budbreak as urea or compost, consistent with minimal differences in petiole N concentration between these treatments. The maximum annual compost application rates in our study ranged from 200 kg/ha (Merlot) to 225 kg/ha (Cabernet Sauvignon) and were based upon the ability to supply N at the same rate as urea. European research suggests that organic amendment application rates as high as 20 tons/ha/yr (much higher than the rates applied in this project) do not affect vine yield or N nutrient status (Morlat 2008).
Fruit composition.
There were few consistent effects of N treatment on Merlot fruit composition. Multiple year effects were only apparent for Merlot berry SSC, which was lower at high N application rates, in 2005 and 2006 (Table 7⇓). Juice pH and TA were not consistently affected over the three-year period (data not shown). There were also no consistent multiyear effects of N treatment on Cabernet Sauvignon fruit composition, including SSC (Table 7⇓), TA, or juice pH (data not shown). In a single year (2005), Cabernet Sauvignon SSC increased by 0.4% in vines receiving 45 kg N/ha at budbreak from compost rather than urea.
The negative effect of increased rate of N fertilization on Merlot SSC and the lack of response in Cabernet Sauvignon are the most frequently reported responses of vines to high rates of applied N (Bell and Henschke 2005). None of the reductions in Merlot berry SSC exceeded 0.4%, which has little practical winemaking significance. Annual average juice pH means ranged from 3.8 to 3.9 and TA means ranged from 4.8 to 7.1 g/L tartaric acid equivalents for Merlot and from 3.7 to 3.8 and 8.0 to 8.7 g/L, respectively, for Cabernet Sauvignon, posing no limitations for production of quality wine (Jackson and Lombard 1993).
Yeast-assimilable N concentration.
Grape berry YANC was the only measured harvest fruit quality indicator affected every year by N treatments for Merlot (Table 8⇓). There was a consistent interaction between N rate and timing for YANC. Application of 80 kg N/ha (2S) at bloom consistently increased YANC relative to either rate (S or 2S) applied at budbreak. In two of the three years, Merlot vines receiving N as compost rather than as urea had lower YANC. Postharvest N did not affect YANC.
Cabernet Sauvignon berry YANC was affected more by rate than by timing of N application (Table 8⇑). YANC was higher at the high N rate at both application times in 2007 and at budbreak applications in 2006. There were no differences in YANC between compost and inorganic N. Postharvest N had no effect on YANC other than in 2005 (lower values) as berries were harvested before fertilizer application.
In contrast to other berry quality characteristics, there was a wide variation in YANC among years and treatments. It is widely known that the N concentration of grape juice affects the rate and duration of fermentation. A minimum of 140 mg/L YANC is recommended to avoid stuck fermentation in both white and red wine fermentation (Bisson 1991, Butzke 1998). In our study, the majority of YANC in Merlot but not Cabernet Sauvignon fruit averaged below this critical threshold. A wide variation in YANC was observed in V. vinifera vineyards throughout the western United States, with significant differences among cultivars and over 15% of sampled vineyards for Merlot, Cabernet Sauvignon, and Cabernet franc with deficient YANC values (<140 mg/L) (Butzke 1998). At our Merlot site, only the application of 80 kg N/ha at bloom consistently increased YANC above the 140 mg/L threshold. This implies that time of N application is an important consideration for increasing YANC of Merlot grapes when a low N fertilization regime is practiced. A similar conclusion that N fertilizer applied at fruit set was more effective in increasing must N of Riesling was reached from a long-term N timing study (Linsenmeier et al. 2008). The effectiveness of bloom-time N applications for increasing must N is consistent with the seasonal development of the grape berry as a strong sink for N between bloom and veraison (Conradie 1980). This raises the possibility that strategically timed foliar N sprays may, if applied soon after bloom, be effective at increasing harvest YANC (Schreiber et al. 2002). Research is needed to determine if sufficient N could be applied in this manner. Responses to N timing may be cultivar- and rootstock-specific, as suggested here by results for Cabernet Sauvignon on V. riparia rootstock, which, despite a similar range of N treatments to Merlot on SO4 rootstock, had higher petiole N, YANC adequate for fermentation, and indications that YANC was affected more by rate than timing of N application. Rootstocks have long been known to influence vine N composition (Bell and Henschke 2005), including scion-fruit amino-N concentrations (Sponholz 1991). In our study, it was not possible to determine whether differences in response to N timing at the two sites were caused by differences in cultivars, rootstocks, or site-specific factors.
Failure to achieve adequate must N concentrations for rapid and effective yeast fermentation can be overcome by the addition of N supplements, including diammonium phosphate (DAP), by the winemaker (Sponholz 1991). However, it is considered desirable to fine-tune berry N concentrations in the vineyard (Bell and Henschke 2005), so that an adequate N status occurs as fruit arrives from the field. In this context, fruit YANC is a valuable quality indicator. It would be useful if some related variable could be used as an indicator of berry YANC status. In the Merlot planting, petiole N concentration at veraison, but not early in the growing season or at bloom, was correlated annually with YANC (Figure 2⇓). The relationship varied by year, with YANC values particularly low in 2006 when yield was highest. If petiole N at veraison is to be used as an indicator of potential fruit YANC, an adjustment for yield may be required. In 2005 and 2007, years with similar yields, on average the 140 mg/L YANC threshold would have been exceeded if a minimum 0.5% petiole N concentration had been measured in mature leaves in the upper portion of the canopy at veraison. It is also noteworthy that average petiole N concentration at the Cabernet Sauvignon site exceeded 0.5% at veraison, consistent with its relatively higher YANC status.
Conclusion
In general, variation in timing of N application over a three-year period affected the performance of mature grapevines at two sites with a history of modest N application rates and no other known nutritional inadequacies. For both Merlot and Cabernet Sauvignon, traditional spring N applications were rapidly incorporated into vegetative tissue, as indicated by increased petiole N concentrations before and at bloom. Delaying N application until bloom resulted in smaller increases in petiole N concentration by veraison, a possible consequence of reduction in vegetative sink strength after bloom. Merlot was favorably affected by bloom-time N application, which reduced canopy density and increased yield and berry YANC relative to traditional budbreak N applications. Only 80 kg N/ha applied at bloom consistently produced YANC values considered adequate for fermentation of Merlot must. Application at the same rate at budbreak resulted in adequate values in 2005 and 2007. Despite a similar range of treatments applied on the same soil series, Cabernet Sauvignon on V. riparia rootstock had higher petiole N and nonlimiting YANC, a possible result of either historically slightly higher N application rates or more efficient N uptake by this cultivar/rootstock combination. Furthermore, delaying N application until later in the season did not increase fruit YANC of Cabernet Sauvignon, unlike Merlot, suggesting that applying soil N when grapes are a stronger sink may only be effective for some cultivar/rootstock combinations and when cultivar N status is not replete. The efficiency of fruit crop N uptake in irrigated, coarse-textured soils, as at both experimental sites, is affected by frequency and quantity of irrigation, which has a major influence on root-zone N retention (Neilsen and Neilsen 2002). Altering the timing of soil N applications was effective at increasing YANC for Merlot in the season of N application. However, further research is required to determine whether foliar applications of N to Merlot, which could be applied directly to the developing berries between bloom and veraison, would be equally effective at increasing YANC. Postharvest N applications are generally considered impractical under cool-climate growing conditions as in British Columbia, where there is a short time-lapse between harvest and leaf loss. Consistent with this view, few effects of postharvest N applications were observed for Merlot. However, surprisingly, the yield performance of Cabernet Sauvignon receiving postharvest N was good and may warrant further research, although these results were measured during three years when winter cold damage did not occur. The short-term availability of N from compost was sometimes less than the 20 to 30% of total N estimated, reducing tissue N concentrations and implying that higher compost N applications are needed to be consistently effective. Additional research is required to calculate the amount and timing of N mineralized from surface-applied composts, minimally mixed in the soil and subjected to periodic drying between irrigation applications.
Footnotes
Acknowledgments: The authors are grateful for technical support from Bill Rabie and Linda Herbert. Funding for this research was provided by the BC Grape and Wine Council and matched by Agriculture and Agri-Food Canada’s Matching Initiative (MII) program.
- Received October 2009.
- Revision received February 2010.
- Accepted February 2010.
- Published online September 2010
- Copyright © 2010 by the American Society for Enology and Viticulture