Abstract
Background and goals Pinot noir is one of the most popular winegrape varieties worldwide, but because of its origins in more cool-climate wine production regions, there is little information regarding its responses to regulated deficit irrigation practices in warm and semiarid climates. Thus, the objective of this study was to observe the responses of Pinot noir production and fruit quality characteristics to preveraison (anthesis to veraison) and postveraison (veraison to harvest) deficit irrigation over two growing seasons.
Methods and key findings Treatments consisted of pre- and postveraison irrigation levels at fractions (25, 50, 75, and 100%) of estimated crop evapotranspiration (ETc). Vine water status (stem water potential [Ψstem]) was strongly correlated with applied irrigation during both periods, but particularly during preveraison. Accordingly, berry weight was significantly and linearly reduced with preveraison deficit irrigation, but there was no significant berry weight response to postveraison deficit irrigation. All skin phenolic compound classes—anthocyanins, tannins, and iron-reactive phenolics—showed an inverse linear relationship with preveraison deficit irrigation rate, while these responses to postveraison deficit irrigation were quadratic, both on a concentration and a content basis. With preveraison deficit irrigation, a negative linear relationship was observed between berry phenolics such that the greatest preveraison deficit had the highest total phenolics. In contrast, berry phenolic composition improved with moderate postveraison deficit irrigation and no negative yield effects were observed. The preveraison irrigation rate of 25% of ETc and postveraison rates between 50 to 75% ETc maximized berry phenolics without a significant reduction in vine yield.
Conclusions and significance In summary, the results indicate that severe deficit irrigation preveraison, and mild to moderate deficit irrigation postveraison, improve berry phenolics in warm-climate Pinot noir.
Introduction
Deficit irrigation is a widespread practice of restricting supplied irrigation in vineyards to limit vegetative growth and enhance water use efficiency without compromising net photosynthesis (Williams and Matthews 1990, Kriedemann and Goodwin 2003). This practice is aimed at minimizing the competition for photosynthates between ripening fruit and vigorously growing shoots, thus promoting accumulation of sugars and metabolites in the ripening berries (Jackson and Lombard 1993, McCarthy 1997, Williams 2000). Deficit irrigation levels vary greatly, but an average sustained deficit irrigation rate in arid or semiarid region vineyards is often ~60 to 70% of vine evapotranspiration (ETc) (Santesteban et al. 2011). Regulated deficit irrigation (RDI) practices applied at key phenological stages can significantly influence grapevine physiology, vegetative and reproductive development, and fruit quality (Hardie and Considine 1976, Williams et al. 2010a, 2010b, Levin et al. 2020a, Deloire and Pellegrino 2021).
Winegrape fruit quality is often improved by deficit irrigation due to the upregulation of biosynthetic pathways and subsequent increased concentration of secondary metabolites responsible for color, flavor, and wine aroma compounds (Matthews and Anderson 1989, Roby et al. 2004, Castellarin et al. 2007, Ou et al. 2010, Ollé et al. 2011, Torres et al. 2022). This effect is typically more pronounced with preveraison (between fruit set and veraison) rather than postveraison water deficits, as preveraison deficits induce the dual response of increased secondary metabolite production together with reduced berry size (Castellarin et al. 2007). However, insufficient water availability very early in the growing season, i.e., prebloom, can potentially impede the reproductive development of grapevines, as preveraison water deficit can diminish yield by decreasing berry weight and clusters per vine, yet it may not affect skin anthocyanin content (Ojeda et al. 2001, 2002, Williams et al. 2010b, Ollé et al. 2011, Levin et al. 2020a, 2020b). Indeed, if water deficits are severe and occur early enough in the growing season (i.e., before bloom), inflorescence abscission can occur (Williams et al. 2010b). In contrast, yield and yield components (e.g., berry fresh weight) are relatively insensitive to postveraison water deficit, while quality attributes like total soluble solids (TSS) tend to vary with the severity of postveraison deficit (Hardie and Considine 1976, Matthews and Anderson 1989, Levin et al. 2020b).
Among red wine grape cultivars, Pinot noir is the sixth most widely grown in the world and the second most widely planted across the United States (Alston and Sambucci 2019). Generally cultivated within the limits of cool climatic red wine production regions (Shaw 2012), production regions of Pinot noir are becoming increasingly warm and dry (Webb et al. 2012). Seasonal drought, a common occurrence under these climatic trends, may affect fruit production and quality of grapes (van Leeuwen et al. 2024). In the Rogue Valley region of southwestern Oregon where Pinot noir vineyards are irrigated, changing climate is continuing to reduce critical surface water supply during periods of drought and peak water demand for grapes and other crops (Luce et al. 2022). RDI therefore will play a key role in mitigating the effects of drought, while optimizing the use of increasingly limited freshwater resources.
The effects of deficit irrigation on field-grown Pinot noir in warm, semiarid climates remain poorly documented in scientific literature. Some existing reports originating from cool climates (Schreiner and Lee 2014, Zufferey et al. 2017, de Souza et al. 2022), which were conducted on potted vines (Poni et al. 1993, Reynolds and Naylor 1994, Ennahli et al. 2015), used a non-randomized experimental design (Berdeja et al. 2014), or observed vine water stress responses without experimentally manipulating irrigation supply (Rogiers et al. 2009, Wilson et al. 2020). Moreover, none of these studies compared pre- and postveraison RDI treatments applied in the field. Thus, more information regarding the effects of seasonal deficit irrigation strategies on Pinot noir grapevine performance is necessary for the development of precise irrigation management strategies. These strategies may reduce water use and improve Pinot noir fruit and wine quality.
The research question investigated in this current study was how deficit irrigation, applied either pre- or postveraison, influences Pinot noir fruit yield and quality. To achieve this goal, four levels of deficit irrigation treatments (25, 50, 75, and 100% ETc) were independently applied either pre- or postveraison to field-grown Pinot noir vines over two growing seasons. Vine water status was measured weekly. Fruit was harvested at maturity and yield, and primary and secondary berry chemistry was assessed.
Materials and Methods
Experimental site and plant material
The study was conducted over two growing seasons (2017 and 2018) in an eight-row section of a 12-ha commercial vineyard located in the Rogue Valley American Viticultural Area near Wilderville, Oregon (42°23′N; 123°27′W, 278 m asl). Soil at the experimental site was a Central Point sandy loam with an average available water supply of 9.91 cm in the top 100 cm of soil (USDA-NRCS Web Soil Survey, https://websoilsurvey.sc.egov.usda.gov).
The vineyard was planted in 2009 to Vitis vinifera L. cv. Pinot noir (clone 777) grafted onto 101-14 Mgt (Vitis riparia Michx × Vitis rupestris Scheele) rootstock. Rows were oriented north-south with a row spacing of 2.4 m and vine spacing of 1.2 m, for a vine density of ~3363 vines/ha. The vines were trained to bilateral cordons on a vertically shoot-positioned (VSP) trellis and spur pruned during dormancy to retain ~24 buds/vine. All cultural practices besides irrigation management were conducted according to industry standards by the cooperator (i.e., the commercial grower whose vineyard was used to conduct the trial), across all experimental units.
Experimental design
The eight irrigation treatments were arranged in a randomized complete block design with five replications (see Supplemental Figure 1). Each replicated block was 24 vines long and extended across eight rows for a total of 192 vines per block. Treatment plots were randomized within each block, were six vines long, and extended across four rows for a total of 24 vines per plot. Data were collected from the four center vines of the two center rows within each plot, for a total of eight data vines per experimental unit. Thus, each plot was encircled by a ring of border vines. A total of 320 vines were used for data collection. These same vines were observed during both growing seasons. Prior to data analyses, blocks 4 and 5 were excluded because of high powdery mildew disease incidence and fruit infection.
Irrigation treatments and irrigation scheduling
Irrigation treatments were characterized by varying irrigation rates either pre- or postveraison according to fractions of estimated ETc and are described in Table 1. All treatments were imposed once midday stem water potential (Ψstem) reached −0.8 MPa (averaged across all plots) within the range of weak water deficit (van Leeuwen et al. 2009). The preveraison period extended from treatment imposition until veraison (early deficit, ED), while the postveraison period extended from veraison until harvest (late deficit, LD). Two control treatments (well-watered control [WC], dry control [DC]) received a constant rate of irrigation from anthesis until harvest. No irrigation was applied prior to anthesis, and postharvest irrigation was applied evenly to all treatments. The same treatments were applied to each plot during both years of the study.
Description of irrigation treatments based on irrigation rate. Preveraison treatments were imposed from when the stem water potential (Ψstem) threshold of −0.8 MPa was reached (averaged across all plots) until veraison, and postveraison treatments were imposed from veraison until harvest. All plots were irrigated at 100% vine evapotranspiration (ETc) postharvest until the end of October. WC, well-watered control; ED, early deficit; LD, late deficit; DC, dry control.
Separate pre- and postveraison driplines were installed in each row and used to irrigate vines during the respective treatment periods. Applied water amounts for each treatment were varied by changing the number of drip emitters per vine, and scheduling pump run times based on the 100% ETc application rate (i.e., in WC treatment). All drip emitters were pressure compensating and had the same flow rate (2 L/hr Woodpecker Jr., Netafim USA). Thus, for example, WC vines were irrigated with four emitters per vine pre- and postveraison, while ED75, ED50, and ED25 vines were irrigated with three, two, and one emitter per vine preveraison, respectively, and four emitters per vine postveraison. In contrast, LD75, LD50, and LD25 vines were irrigated with four emitters per vine preveraison, and three, two, and one emitter per vine postveraison, respectively. Inline water meters (‘M’ series, Netafim USA) were used to quantify applied water amounts. Budbreak, anthesis, and veraison were determined by visual ratings on two tagged spurs per vine based on 50% green tips, capfall, and coloration of inflorescences, and clusters, respectively, averaged across all plots (two per plot).
Vineyard ETc was estimated using the following equation:
where ETo is reference ET and Kc is the crop coefficient. ETo was obtained from a weather station (Vantage Pro 2, Davis Instruments) installed at the research site. Kc was calculated according to Williams et al. (2022) for VSP-specific trellis as follows:
where GDD is accumulated growing-degree days beginning at budbreak (base 10°C) and e is Euler’s number.
Quantification of vine water status
To quantify vine response to irrigation treatments, Ψstem was measured using a pressure chamber (Model 615, PMS Instruments) according to the method described in Levin (2019). Briefly, measurements were made at weekly intervals beginning prior to treatment imposition through harvest at solar noon (i.e., 1300 hr ± 30 min Pacific Daylight Time). Fully expanded leaves located on the shaded side of the canopy were covered with an aluminum foil-lined airtight bag, quickly sealed on the vine, and left to equilibrate for at least 10 min. Upon equilibration, the petiole of the bagged leaf was excised from the stem and quickly brought to the pressure chamber and pressurized within 30 sec of excision. Values were recorded when the wetting front appeared on the excised petiole surface, observed through a handheld magnifying glass. A single leaf from each replicate plot was used for data analysis.
Yield components and fruit composition
To quantify responses of berry fresh weight and fruit composition to irrigation treatments, 50-berry samples were collected prior to harvest in each experimental unit. Yield and yield components (i.e., cluster weight) were then determined by harvesting whole vines. Harvest occurred on the same date for all treatments within a year and was determined by the commercial vineyard operator. All clusters on primary shoots for each vine were counted and weighed in the field. Other yield components such as individual cluster weight and berries per cluster were calculated from berry weight data determined from berry samples.
Upon returning to the lab, 20-berry subsamples were taken from each 50-berry sample and stored at −80°C for subsequent phenolics analysis. The remaining 30 berries were weighed and pressed and the resulting juice was collected and centrifuged. Juice TSS (Brix) were measured using a digital handheld refractometer (AR200, Reichert, Inc.), juice pH was determined using a benchtop pH meter (Orion 3-Star Benchtop pH Meter, Thermo Fisher Scientific), titratable acidity (TA) was determined using an autotitrator (T50 Titrator, Mettler Toledo), and malate was determined using enzymatic assay (L-Malic Acid Assay Kit, Megazyme). Skin phenolics were analyzed using a microplate reader (SpectraMax 190, Molecular Devices) according to the Harbertson-Adams assay described in Casassa et al. (2015).
Data analyses
All data were analyzed and plotted using R software for statistical computing (ver. 4.4.0; R Core Team 2024). Linear mixed models were fit to data using the lmer() function (Bates et al. 2015) from the ‘lmerTest’ package (Kuznetsova et al. 2017), with statistical significance considered at α = 0.05. For analyses of pre- and postveraison Ψstem, two-way analyses of variance (ANOVA) for a split-plot factorial treatment structure were conducted with irrigation treatment as the main plot factor and phenological period as the subplot factor. Separate analyses were conducted each year. For analyses of yield and yield components, berry chemistry, and berry phenolic composition, similar ANOVAs were conducted, with irrigation treatment as the main plot factor and year as the subplot factor. For the latter analyses, if the interaction term p value was >0.05, then a new, more parsimonious linear model was fit to the data excluding the fixed interaction term and compared to the original model using the Chi-square criterion. If the Chi-square value was >0.05, the new model was used for means and contrast computation. Estimated marginal means, Dunnett’s contrasts (trt. vs.ctrl.emmc()), and orthogonal polynomial trend contrasts (poly.emmc()) were computed using the ‘emmeans’ package (Lenth 2025). Regression analyses were conducted using computed emmeans and the lm() function from the R ‘base’ package. Figures were produced using the ‘ggplot2’ package (Wickham 2016).
Results
Environmental conditions, vineyard phenology, and applied water amounts
Total precipitation varied over both years of the study, while cumulative GDD, ETo, and ETc were nearly identical between years (Table 2). In general, 2017 was a wetter year compared to 2018. The total seasonal, postharvest, and dormant season (October to March) precipitation in 2017 was about twofold greater compared to that in 2018.
Monthly totals for growing degree days (GDD), reference evapotranspiration (ETo), estimated vine evapotranspiration (ETc), and precipitation.
Major phenological events occurred at a similar date in both years except for anthesis. Budbreak was observed on 17 April and 23 April, while anthesis was recorded on 12 June and 31 May in 2017 and 2018, respectively. Irrigation treatments were imposed on 11 July and 16 July in 2017 and 2018, respectively. Veraison and harvest dates were similar between years; in 2017, veraison occurred on 14 Aug and harvest on 26 Sept, and in 2018, veraison occurred on 13 Aug and harvest on 24 Sept.
In general, applied water amounts across all treatments were similar within treatments across the two study years, with slight differences (Figure 1). Total applied water amounts ranged more than twofold from ~100 mm in DC vines, to more than 250 mm in WC vines, across treatments over both years. Treatments were successfully applied in both years, with water applications decreasing from the control (WC) in a stepwise manner preveraison for all ED treatments, and postveraison for all LD treatments. In 2017, WC was irrigated (266 mm) at 100.7% of ETc (264 mm) during the irrigation treatment period. In 2018, WC was irrigated (265 mm) at 101.5% of ETc (261 mm).
Applied water amounts in each irrigation treatment in both years of the study as a function of phenology. WC, well-watered control; ED, early deficit; LD, late deficit; DC, dry control.
Ψstem responses to treatments
There were strong, statistically significant (p < 0.01) and linear Ψstem responses to irrigation treatments during each phenological period in both years (Supplemental Table 1 and Figure 2). In general, Ψstem values trended lower preveraison and were more responsive to applied irrigation (steeper slopes) compared to postveraison. Ψstem ranged from −0.6 to −0.8 MPa for vines irrigated at 100% ETc, with slightly higher values in 2017 and postveraison. Decreasing applied water amounts linearly reduced Ψstem, with values ranging from −0.9 to −1.3 MPa for vines irrigated at 25% ETc, with lower values in 2017 and preveraison. There was greater variation in preveraison response of Ψstem to applied irrigation in 2017 (R2 = 0.94) compared to 2018 (R2 = 0.76), while the response was similar postveraison between the two years. The slopes of the preveraison responses were significantly different between years (p = 0.01), but this difference was absent postveraison.
Response of stem water potential (Ψstem) to irrigation rate (% vine evapotranspiration [ETc]) pre- and postveraison in 2017 (circles/solid lines) and 2018 (triangles/dashed lines). Data points are irrigation treatment means ± 1 standard error (SE; n = 3). Within each year, data were pooled in each phenological period to include five to six measurement dates. Preveraison treatments were imposed from when the Ψstem threshold of −0.8 MPa was reached (averaged across all plots) until veraison, and postveraison treatments were imposed from veraison until harvest. All plots were irrigated at 100% ETc postharvest until the end of October. Breakdown of each pre- and postveraison treatment is detailed in Table 1. WC, well-watered control; ED, early deficit; LD, late deficit; DC, dry control.
Yield and yield components responses to treatments
Across all yield components, only berry weight was significantly affected by irrigation treatments (Table 3). When comparing individual treatment means, only DC vines had significantly smaller berries compared to WC, with no other significant differences among means. Trend analyses revealed that there was a significant linear trend in berry weight for ED treatments such that preveraison reductions in applied water from WC resulted in smaller berries at harvest. No quadratic trend was found for early deficits. In contrast, there were no significant trends—linear or quadratic—for late deficits. There were significant year effects on some yield components (berry weight, berries per cluster, and cluster weight), but no significant interactions between treatment and year for any single component. Across treatments, berries were smaller in 2018 compared to 2017, but there were more berries per cluster, and thus, cluster weight was higher.
Response of yield components to irrigation treatments at harvest. Means ± 1 standard error (SE; n = 3) were calculated for main effects only. Trend analyses for preveraison deficit (ED) and postveraison deficit (LD) were conducted on treatments arranged according to water application and excluding dry control (DC) (i.e., ED: WC-ED75-ED50-ED25 and LD: WC-LD75-LD50-LD25), where WC is well-watered control: vines were irrigated at 100% vine evapotranspiration (ETc) during both pre- and postveraison; the number within each treatment notation (75, 50, and 25) indicates the irrigation rate in % ETc that the vines received either pre- (ED) or post- (LD) veraison. The vines were irrigated at 100% ETc for the rest of the season. Breakdown of each pre- and postveraison treatment is detailed in Table 1. ANOVA, analysis of variance.
Juice chemistry responses to treatments
Juice pH and TA were significantly affected by irrigation treatments, but there were no significant differences among treatments in juice TSS or malate (Table 4). In general, these juice acidity variables responded similarly to treatments and were highly correlated among one another. Deficit irrigation during either phenological period reduced malate and TA and increased pH compared to the control, but individual treatment means were not significantly different from WC except for the DC treatment. However, a linear trend was found to be significant (p < 0.05) for pH. There were also strong significant year effects on TSS, pH, and TA.
Response of berry chemistry to irrigation treatments at harvest. Means ± 1 standard error (SE; n = 3) were calculated for main effects only. Trend analyses for preveraison deficit (ED) and postveraison deficit (LD) trends were conducted on treatments arranged according to water application and excluding dry control (DC) (i.e., ED: WC-ED75-ED50-ED25 and LD: WC-LD75-LD50-LD25), where WC is well-watered control: vines were irrigated at 100% vine evapotranspiration (ETc) during both pre- and postveraison; the number within each treatment notation (75, 50, and 25) indicates the irrigation rate in % ETc that the vines received either pre- (ED) or post- (LD) veraison. The vines were irrigated at 100% ETc for the rest of the season. Breakdown of each pre- and postveraison treatment is detailed in Table 1. TSS, total soluble solids; TA, titratable acidity; ANOVA, analysis of variance.
Berry skin phenolics responses to treatments
The irrigation treatments had a significant effect on berry skin phenolics, both on a concentration and a per berry basis (Table 5). There were also no significant interactions between irrigation treatment and year for any variable, although year main effects were significant for anthocyanin and tannin concentrations and for iron-reactive phenolics (IRP) content. Berries from the WC treatment exhibited significantly lower anthocyanin concentrations and total anthocyanin contents compared to the deficit irrigation treatments (DC, LD50, LD75, and ED25). A similar trend emerged for tannin and IRP concentrations, however, the total tannin and IRP content in the DC treatment did not differ significantly from WC.
Response of skin phenolic composition to irrigation treatments at harvest. Each compound was evaluated on a concentration (mg/g) and content (mg/berry) basis. Anthocyanins are in malvidin-3-glucoside equivalents. Tannins and iron-reactive phenolics (IRPs) are in catechin equivalents. Means ± 1 standard error (SE; n = 3) were calculated for main effects only. Trend analyses for preveraison deficit (ED) and postveraison deficit (LD) trends were conducted on treatments arranged according to water application and excluding dry control (DC) (i.e., ED: WC-ED75-ED50-ED25 and LD: WC-LD75-LD50-LD25), where WC is well-watered control: vines were irrigated at 100% vine evapotranspiration (ETc) during both pre- and postveraison; the number within each treatment notation (75, 50, and 25) indicates the irrigation rate in % ETc that the vines received either pre- (ED) or post- (LD) veraison. The vines were irrigated at 100% ETc for the rest of the season. Breakdown of each pre- and postveraison treatment is detailed in Table 1. ANOVA, analysis of variance.
Trend analysis revealed a strong linear decrease in the concentrations of anthocyanins, tannins, and IRPs in berry skins, as well as their respective total contents in berries, with increasing irrigation rates preveraison (Figure 3). However, the relationship between these phenolic compounds followed a quadratic trend postveraison. Both phenolic concentrations and contents increased from 25 to 50% ETc irrigation rates, plateaued between 50 and 75%, then declined in 100% ETc. Notably, the 25% ETc irrigation rate postveraison yielded lower phenolic levels compared to the same irrigation rate preveraison. Conversely, the remaining postveraison irrigation rates yielded equal or greater phenolic concentrations and contents relative to their corresponding preveraison irrigation rates. The relative percentages of phenolic concentrations and contents exhibited distinct responses to irrigation rates and timings (Figure 4). Preveraison, phenolics decreased linearly with increasing irrigation rate. Postveraison, however, this relationship was quadratic in nature. The greatest increase in phenolics in response to postveraison irrigation was observed at 50 and 75% of ETc.
Responses of berry phenolic concentration (mg/g) and content (mg/berry) to irrigation rate (% vine evapotranspiration [ETc]) pre- and postveraison. Data points are irrigation treatment means (excluding dry control [DC]) ± 1 standard error (SE; n = 3) taken from Table 5. IRPs, iron-reactive phenolics.
Relative increase of berry phenolic compound concentration (mg/g) and content (mg/berry) in response to irrigation rate (% vine evapotranspiration [ETc]) pre- or postveraison. Percent increase data were calculated relative to well-watered control (WC) using means from Table 5 (excluding dry control [DC]). Separate linear and polynomial regressions were fit to pre- and postveraison data sets, respectively. IRPs, iron-reactive phenolics.
Discussion
This study highlights the effect of early- and late-season deficit irrigation on V. vinifera L. Pinot noir grapevine productivity and fruit quality in a warm, semiarid growing region. Ψstem responded strongly and predictably to the applied irrigation rate (i.e., % ETc) and timing (i.e., pre- or postveraison). While deficit irrigation during both early and late seasons strongly influenced berry phenolics, most yield and quality components remained unaffected. Sustained deficit irrigation at 25% of ETc was the only treatment to affect fruit juice chemistry at harvest. Skin phenolics were increased by deficit irrigation, but the response differed between early and late season deficits. This study demonstrates that Pinot noir fruit quality can be improved by early season, late season, and sustained deficit irrigation, with some variation by irrigation rate.
Midday Ψstem showed a consistent, significant, and moderate-to-strong positive relationship with applied irrigation over two consecutive years characterized by different precipitation amounts. This validates the irrigation scheduling method used in this study, which determined ETc from Kc values originally developed for Chardonnay grapevines grown in a different region but trained on a similar VSP trellis as the one used in this study (Williams et al. 2022). These values may differ for VSP-trained Pinot noir grown in southwestern Oregon. However, because of the lack of published region- and cultivar-specific Kc values, the Chardonnay values were used as the closest available approximation and were able to generate a broad range of Ψstem responses.
The preveraison response of midday Ψstem to applied irrigation was weaker in the second year compared to the first year. While the current study did not directly measure canopy growth, it is possible that the water deficits experienced by vines in the first year negatively affected early season canopy size in the second year. Other studies have shown that water requirement decreases in vines in the season following deficit irrigation treatment, and vine canopy size is strongly correlated with vine water requirement (Williams et al. 2003, Williams and Ayars 2005, Picón-Toro et al. 2012). The lower slope value for the linear relationship between preveraison irrigation rate and Ψstem in 2018 may be a consequence of altered canopy size and Kc, which potentially overestimated the irrigation requirement for low rate (i.e., 25% ETc) treatments and underestimated the irrigation requirement for high rate (i.e., 100% ETc) treatments.
The current study did not find an effect of deficit irrigation treatments on vine yield components, except for the DC berry weight compared to the WC berry weight. The second year showed a notable increase in cluster weights due to a higher number of smaller berries compared to the first year, but there were no significant treatment effects. The results presented here do not align with other reports that yield and yield components are sensitive to severe and preveraison water deficits (Matthews and Anderson 1989, Levin et al. 2020a, 2020b). Poni et al. (1993) observed similar results in potted Pinot noir insofar as yield for preveraison deficit irrigated vines was numerically lower (25 to 27%) than postveraison deficit irrigated vines and an untreated control, but in the end there was no significant difference. Ultimately, the level of water stress, expressed here with Ψstem, was not severe enough to significantly reduce yield, even under irrigation rates as low as 25% of ETc. Levin et al. (2020b) observed significant reductions in yield across 15 cultivars from preveraison water deficits, but preveraison leaf water potential (Ψleaf) values were severe (−1.64 to −1.05 MPa) compared to the mild-to-moderate Ψstem values observed here (van Leeuwen et al. 2009).
Berry weights were reduced linearly in response to decreasing preveraison irrigation rates, highlighting the sensitivity of berry growth to water stress early in the season. Water uptake of grapevines is a principal factor that determines berry size (Hardie and Considine 1976, Matthews and Anderson 1989, Matthews and Shackel 2005). Preveraison water deficits can irreversibly restrict pericarp development during the cell expansion phase of berry growth, which limits cellular extensibility and potential maximum berry weight (Ojeda et al. 2001). Levin et al. (2020b) found berry fresh weight to be the most sensitive yield component to water deficits preveraison in 15 winegrape cultivars. Under the conditions of the present study, this trend did not result in a significant treatment effect for Pinot noir berry weight.
Deficit irrigation also had a weak influence on fruit juice chemistry. Water deficit can influence berry sugar content at harvest, and the effect usually scales with the severity of water deficit (Ojeda et al. 2002). Berry sugar concentration was not higher with the pre- and postveraison water deficits imposed in the present study, which were characterized by high Ψstem values that are indicative of moderate-to-weak water deficit (van Leeuwen et al. 2009). The berry sugar content, calculated as the product of TSS and berry weight, also exhibited low variability, with the lowest value of 0.183 g/berry for DC and the highest of 0.224 g/berry for WC. Rapid sugar accumulation may be of little concern in warm regions where TSS targets are easily achieved, and furthermore, the ripening target may be achieved too soon under the conditions of climate change projections (Kurtural and Gambetta 2021). Deficit irrigation has been shown to delay ripening (Matthews and Anderson 1988) and perhaps could be useful to offset this effect. However, in this study we have not observed any compelling evidence of this phenomenon insofar as TSS values recorded at harvest were not significantly different among irrigation treatments. The low TA observed in this study could be attributed to high temperatures during ripening and low potassium content in regional soils (Terrier and Romieu 2001, Sweetman et al. 2014). Still, deficit irrigation has in some cases reduced the rate of sugar translocation into berries (Torres et al. 2021). The switch from a preveraison water deficit—which reduced berry size and, as a result, sink demand—to postveraison irrigation at 100% ETc may have altered carbon partitioning to favor sugar export toward other sinks (e.g., roots and shoots) or reduced the size and sink demand of ripening fruit (Gómezdel-Campo et al. 2005, Torres et al. 2021). Therefore, more severe preveraison deficit irrigation followed by close-to-full irrigation (100% ETc) postveraison could be an effective strategy to counteract the effect of rapid sugar translocation within berries in a hot climate.
This study found that deficit irrigation produced higher berry skin phenolics compared to the WC treatment, and response dynamics depended on phenological period. The concentration and content of all phenolic classes—anthocyanins, tannins, and IRPs—increased with greater pre- and postveraison water deficit, but while responses were linear preveraison, they were quadratic postveraison. Water deficit is known to increase the skin-to-pulp ratio, resulting from reduction in growth of mesocarp tissue, thus enhancing the concentration and extraction of skin tannins and other phenolics (Bravdo et al. 1985, Kennedy 2002, Ojeda et al. 2002, Roby and Matthews 2004, Keller et al. 2008, Chacón-Vozmediano et al. 2021). The present study does not characterize the effect of Ψstem values below −1.4 MPa on berry phenolics, but some reports suggest that even severe water stress can improve the concentration of phenolic compounds in berries. Keller et al. (2008) observed high color density associated with postveraison Ψstem values as low as −1.76 MPa.
It is well-established that biosynthesis of phenolic precursors is promoted by water deficit in grape berries (Matthews et al. 1990, Kennedy et al. 2002, Ojeda et al. 2002, Castellarin et al. 2007, Chaves et al. 2010, Ollé et al. 2011, Zarrouk et al. 2012, Chacón-Vozmediano et al. 2021). Deficit irrigation increases flavonoid concentrations by inducing earlier expression of genes regulating anthocyanin biosynthesis in berries, when compared to well-watered vines (Castellarin et al. 2007). This same study found increases in anthocyanin concentrations and content in berries in response to both pre- and postveraison deficit irrigation and achieved greater water stress than presented here (e.g., minimum Ψleaf ≈ −1.45 MPa). Ojeda et al. (2002) found significantly higher skin anthocyanin concentration and content during postverasion deficit irrigation treatment compared to preverasion treatments, although the maximum water stress was greater than that observed in the present study (i.e., Ψleaf > −1.0 MPa). However, a majority of this early work was conducted on Cabernet Sauvignon and Syrah. Phenolics responses were not directly related to deficit irrigation levels, but were merely shown accumulating over time as a function of deficit irrigation treatment.
The current study provides a greater resolution to different degrees of pre- and postveraison water deficits to phenolics profile in berry skins in field-grown Pinot noir. The linear increase in phenolics concentrations, as well as their content with an increasing degree of preveraison water deficits, suggests an upregulation of pertinent biosynthetic genes in relation to water stress due to the lack of significant influence on berry size. This is further evident from the response to late deficit treatments, which increased skin phenolics concentrations and content without influencing berry size. Although the current study did not find a strong linear response of phenolics to late deficit treatment, the significantly greater phenolics concentrations and contents compared to control, especially at moderate deficit levels, suggest an effect on flavanol biosynthesis even postveraison (Ojeda et al. 2002). Anthocyanins are synthesized postveraison and although tannins are synthesized preveraison, they share several early steps in the pathway (Castellarin et al. 2007). Studies have suggested that the berry is hydraulically isolated postveraison (Coombe 1992, Greenspan et al. 1994, 1996), but newer evidence exists for sustained xylem connection (Bondada et al. 2005, Knipfer et al. 2015) and a signal translocation pathway (Castellarin et al. 2007) during ripening. Therefore, the high concentration of phenolics resulting from late deficits is not solely due to passive desiccation of berries, it is also due to active synthesis of these compounds during the postveraison period. Both the concentration and content of phenolic compounds were highest for high preveraison deficit (ED25), moderate postveraison deficit (LD75, LD50), and sustained deficit (DC). This study indicates that phenolics biosynthesis could be triggered by early (between fruit set and veraison) and severe water deficit, moderate deficit later (postveraison) during the season, and by severe season-long deficit, adding new insights into the effects of water deficit on berry phenolics.
Conclusion
This study sought to characterize the vine response of V. vinifera L. Pinot noir and its fruit grown in a warm and semiarid climate. It aimed to explore the timing and intensity of deficit irrigation while moving toward more specific water status targets for achieving fruit quality goals. No strong influence of deficit irrigation on yield or fruit juice chemistry was observed. Preveraison deficit irrigation linearly increased phenolics as Ψstem decreased, down to approximately −1.3 MPa, below which was not observed. Postveraison water deficits had positive effects on berry skin phenolics as well. The highest phenolics concentrations from postveraison deficit irrigation were at the 50 and 75% ETc treatments, which produced average Ψstem values between −0.9 and −0.5 MPa. The effects of more severe postveraison water status values were not observed in this study. The results of this study indicate that severe preveraison deficit irrigation (e.g., 25% of ETc) and mild-to-moderate postveraison deficit irrigation (e.g., 50 to 75% of ETc) of Pinot noir in a warm, semiarid climate can maximize phenolic composition without a significant yield penalty.
Supplemental Data
The following supplemental materials are available for this article in the Supplemental tab above:
Supplemental Table 1 Mean stem water potential (Ψstem) response to irrigation treatments during two phenological periods: from treatment imposition to veraison (preveraison), and from veraison to harvest (postveraison). Treatment means (and standard error [SE]) are calculated from pooled data encompassing five to six measurements for each period (n = 3 to 5). P values adjacent to means represent comparisons against WC treatment and are adjusted for multiplicity using Dunnett’s method. Trend analyses for preveraison deficit (ED) and postveraison deficit (LD) trends were conducted on treatments arranged according to water application and excluding dry control (DC) (i.e., ED: WC-ED75-ED50-ED25 and LD: WC-LD75-LD50-LD25), where WC is well-watered control: vines were irrigated at 100% vine evapotranspiration (ETc) during both pre- and postveraison; the number within each treatment notation (75, 50, and 25) indicates the irrigation rate in % ETc that the vines received either pre- (ED) or post- (LD) veraison. The vines were irrigated at 100% ETc for the rest of the season. Breakdown of each pre- and postveraison treatment is detailed in Table 1.
Supplemental Figure 1 Soil properties of the vineyard site. Data was obtained from the USDA-NRCS Web Soil Survey (https://websoilsurvey.sc.egov.usda.gov). Experimental blocks (B1 to B5) are marked with a red border. Block 1 (B1) was expanded to the right to show the assignment of irrigation treatments. Rectangles in the middle of each treatment plot indicate eight data vines. WC, well-watered control; DC, dry control; LD, late deficit; ED, early deficit.
Data Availability
All data underlying this study are included in the article and its supplemental information.
Footnotes
We thank Andy Pearl and Antonio Nolasco of Pearl Family Vineyards for providing the research site and for vineyard management services. We also thank Kathryn E. Lundquist for helping with the editing and presentation of the figures. This work was supported in part by the Oregon Wine Board (award numbers 2018-2223 and 2019-2223), the Oregon Wine Research Institute, and the Rogue Valley Winegrowers Association. In addition, this work is based on research that was supported, in part, by the Oregon Agricultural Experiment Station with funding from the Hatch Act capacity funding program from the USDA, National Institute of Food and Agriculture. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and should not be construed to represent any official USDA or U.S. Government determination or policy.
Kar S, Copp CR, DeShields JB, Clark RW and Levin AD. 2025. Severe preveraison and moderate postveraison deficit irrigation improves berry phenolics in warm-climate Pinot noir. Am J Enol Vitic 76:0760024. DOI: 10.5344/ajev.2025.25016
By downloading and/or receiving this article, you agree to the Disclaimer of Warranties and Liability. If you do not agree to the Disclaimers, do not download and/or accept this article.
- Received April 2025.
- Accepted July 2025.
- Published online September 2025
This is an open access article distributed under the CC BY 4.0 license.
References
Vol 76 Issue 2
Jump to section
Related Articles
Cited By...
More from this TOC section
Similar Articles
