Effects of Vineyard Management Practices on Winegrape Composition. A Review Using Meta-analysis

Cluster thinning

Cluster thinning is a way to reduce yield and may be required in overcropping or cool-climate situations to ensure ripeness (Jackson and Lombard 1993, Gil-Munoz et al. 2009, Zhuang et al. 2014) or meet yield targets and restrictions (Keller et al. 2005). As the developing fruit is a strong “sink” (Hale and Weaver 1962), cluster thinning dramatically changes the “source” (leaves) to “sink” relationship (Kliewer and Dokoozlian 2005, Pastore et al. 2011). It is debatable whether the effect of cluster thinning on fruit composition is advantageous. Cluster thinning may lead to excess canopy growth (Smart et al. 1990, Jackson and Lombard 1993), and the remaining clusters are larger and may become more compact due to increased berry number and berry weight (size) (Cameron et al. 2024). This may increase disease risk, and the larger berries may also have less color due to a lower skin-to-pulp ratio (Poni et al. 2018). However, it must be noted that Walker et al. (2005) found little difference in the ratio of skin fresh weight to berry fresh weight across a range of berry masses, with only the smallest berries giving a statistically significant difference and resulting in a higher color due to this difference in ratio. To offset these arguably negative effects, cluster thinning opens the cluster zone and reduces cluster shading by other clusters and may thus improve anthocyanin development and disease resistance. By altering the leaf area-to-fruit ratio, cluster thinning may improve fruit composition (Kliewer and Dokoozlian 2005, VanderWeide et al. 2024).

Grape sugar content increased significantly due to cluster thinning (Table 1). This increase was similar for each timing and severity (Table 2). The remaining fruit has access to a greater proportion of carbohydrates from a largely unchanged canopy, increasing the source-to-sink ratio (VanderWeide et al. 2024). In a study using different categories of cluster thinning severity than used here, when ~25% of the fruit was removed, there was little or no impact on TSS, presumably because the impact on source-to-sink ratio was insufficient to have an effect (VanderWeide et al. 2024).

Is this increase in TSS of practical significance? The response ratio for cluster thinning was 0.043. We can calculate from Equation 1 that, for example, if the control TSS result was 23 Brix, the treatment TSS would be 24 Brix. This difference is both of practical significance (Schmidtke et al. 2020) and a statistically significant increase. TSS per hectare was reduced with cluster thinning (Table 2); we conclude therefore, unsurprisingly, that the increased TSS does not compensate for the yield loss resulting from cluster thinning (Cameron et al. 2024), and overall vineyard productivity of grape sugar is reduced significantly by cluster thinning.

Results for changes in TA and pH due to cluster thinning are varied (Palliotti and Cartechini 2000). We found a significant decrease in TA and an increase in pH, changes that were similar for both cluster thinning timings and severity (Tables 1 and 2). A similar investigation of the practical significance of these results found that, with a response ratio of 0.02, if the control pH was 3.3, the treatment pH would be pH 3.37. For TA, with a response ratio of -0.038, a control TA of 8.0 g/L would correspond to a treatment TA of 7.7 g/L. These changes in pH and TA may be desirable in very cool climate grapes, where pH may be too low and TA, too high. Conversely, the magnitude of these changes in pH and TA may be undesirable in warmer climate-grown grapes, where TA may already be lower, or pH higher, than optimal (Coombe et al. 1980).

The TSS:TA and TSS:pH ratios were used to isolate the impact of cluster thinning from changes in grape maturity indicated by TSS (Table 1), and because TSS:TA ratios can be used as the basis of quality standards (Hunter et al. 2016). With increased TSS and decreased TA due to cluster thinning, it was not surprising that TSS:TA increased, regardless of the timing or severity of cluster thinning (Table 2). The increased TSS:TA ratio suggests a lower TA for a given TSS. This is likely to be undesirable depending on wine style requirements, and particularly in warm climates, where preserving acidity may be difficult (Gatti et al. 2012, Petrie et al. 2017). The TSS:pH ratio increased significantly at both phenological timings of cluster thinning and when ≥50% clusters were removed (Table 2), indicating that despite the TSS and pH response ratios both increasing due to cluster thinning, TSS increased more, relative to the pH. At ≥50% cluster removal, the increased leaf area-to-fruit ratio had more effect on TSS than the reduced cluster shading had on malic acid degradation and pH. The lower pH at a given TSS in cluster-thinned vines when ≥50% clusters are removed could be an advantage, especially in warm climate fruit (Petrie et al. 2017). These increases in TSS:TA and TSS:pH ratios may appear inconsistent, but there is no direct relationship between pH and TA, and the same TA value can have a different pH value (Boulton 1980). This is often due to the buffering effect of potassium in the grapes. We conclude that cluster thinning opened the fruit zone to some degree, enabling warmer temperatures around the remaining clusters, which increased degradation of malic acid and lead to the decrease in TA (Lakso and Kliewer 1975).

Anthocyanin accumulation begins at veraison and although anthocyanin composition is genetically determined (Liang et al. 2008), differing among varieties (Rienth et al. 2021), their synthesis can be influenced by a number of factors including light, temperature, and water supply (Guidoni et al. 2008). Anthocyanin synthesis is closely associated with sugar accumulation. Sugar upregulates the expression of flavanone 3-hydroxylase, an important enzyme in the anthocyanin biosynthetic pathway (Zheng et al. 2009). Tannin synthesis begins prior to veraison (Bucchetti et al. 2011). Anthocyanin accumulation is inhibited by temperatures greater than 35°C, which both inhibit anthocyanin synthesis (Mori et al. 2005) and increase anthocyanin breakdown (Mori et al. 2007). Anthocyanin synthesis is favored by daytime temperatures of 15 to 25°C and nighttime temperatures of 10 to 20°C (Kliewer and Torres 1972). Warmer climate (increased hours >22°C between budbreak and bloom) increased phenolic concentrations in Pinot noir (Nicholas et al. 2011), with the anthocyanins, tannins, and thus total phenolics, responding differently depending on the temperatures at different phenological stages. For example, anthocyanins increased at temperatures between 16 and 22°C between veraison and harvest, where tannins increased with warm nights before budbreak and during the budbreak to bloom period. Another study also found that high light levels resulted in lower concentrations of anthocyanins, tannins, and total phenolics (Nicholas et al. 2011). The relationship between light and anthocyanin production has not always been consistent (Downey et al. 2004). It is unclear whether the influence of various management practices on phenolics are due to light, temperature, or other factors (Kennedy et al. 2007). In practice, it is difficult to separate the impacts of temperature and light on these compounds (Spayd et al. 2002).

Anthocyanins and total phenolics both increased due to cluster thinning (Table 1). This is likely due to the increased source-to-sink ratio and the associated TSS increase. The response ratio for anthocyanins was 0.183 (Table 1). Using Equation 1, this corresponds to an 18.3% increase in anthocyanins due to cluster thinning, proportionally more than the 4% increase in TSS. Bobeica et al. (2015) found that carbon limitation reduced anthocyanins more than TSS, leading to speculation that sugar accumulation was maintained at the expense of secondary metabolites such as anthocyanins, meaning that anthocyanins are more sensitive to changes in source-to-sink ratios. Our results agree that anthocyanins respond more to changes in the source-to-sink ratio than TSS. A two-phase response was suggested by Guidoni et al. (2008). In the first phase, up to 6 wk after veraison, sugar and anthocyanin accumulation was almost parallel and correlated significantly, driven by climatic and cultural factors such as photosynthetic activity, leaf area-to-crop ratio, and crop microclimate. In the second phase, lasting ~3 wk, sugar and anthocyanin accumulation was no longer parallel and may be quite different, depending on climatic and management factors. This may explain the large difference in percentage increases of anthocyanins compared to TSS in cluster-thinned vines. It is also arguable that cluster thinning improves light penetration and increases temperature in the remaining clusters, which further improves anthocyanin and total phenolic accumulation (Kliewer and Torres 1972, Morrison and Noble 1990, Mori et al. 2005, 2007), while TSS does not respond to canopy microclimate influences such as light (González-SanJosé and Diez 1992).

The timing of cluster thinning did not affect the increased accumulation of anthocyanins or total phenolics differently (Table 2), but removal of ≥50% clusters was needed to significantly increase anthocyanins and total phenolics (Table 2). This further points to the importance of the source-to-sink ratio in determining anthocyanin and total phenolic content. Tannins were unchanged due to cluster thinning, regardless of timing or severity. This was surprising, as tannins share a common synthetic pathway with anthocyanins and phenolics (Casassa et al. 2015), but it should be noted that the data set for tannins was smaller (Table 1).

The decrease in TSS:anthocyanins ratio when ≥50% clusters were removed indicates a higher anthocyanin content at a given TSS value. This could be a quality benefit, particularly where increased temperatures due to climate change may lead to reduced berry color (Nicholas et al. 2011). The loss of yield due to cluster thinning (Cameron et al. 2024) was not compensated for by the increased TSS and anthocyanins. TSS per hectare were reduced at both timings and proportions of cluster thinning, and anthocyanins per hectare were reduced either by cluster removal E-L 34 to 39 or when ≥50% clusters were removed (Tables 1 and 2).

Cluster thinning timing and proportion were analyzed separately. In a practical situation, different levels of cluster removal may occur at different times. When the data were further segmented to address this question, we found that when cluster thinning occurred at E-L ≤ 33, anthocyanins were only significantly increased when ≥50% of clusters were removed (response ratio = 0.127 ± 0.158 (26) when <50% clusters removed; response ratio = 0.183 ± 0.111 (38) when ≥50% clusters removed). With later cluster thinning at E-L 34 to 39, anthocyanins again only increased when ≥50% clusters were removed (response ratio = 0.116 ± 0.155 (15) when <50% clusters removed; response ratio = 0.171 ± 0.098 (53) when ≥50% clusters removed). Thus, in practice, if more anthocyanins are desired, cluster thinning must remove 50% or more clusters, and this can be done before or during grape ripening. This importance of cluster thinning severity is supported by other studies (VanderWeide et al. 2024).

Despite these potential quality benefits for the winery, cluster thinnning adds additional costs and lowers yield (Cameron et al. 2024). This will likely lower payments to grapegrowers, who are commonly paid on a per tonne basis (Preszler et al. 2010). If growers were compensated to some extent by basing payment on grape anthocyanin or TSS content, this would provide incentive to the grower (optimized income) and benefit the winemaker (improved grape quality).

Irrigation (reduced)

The effect of irrigation on wine quality has attracted considerable debate (Chaves et al. 2010). Grape yield and composition are influenced by water availability (Esteban et al. 1999, Trigo-Córdoba et al. 2014). Too much water can be detrimental to grape yield and composition due to excess vigor and undesirable changes to the source-to-sink balance, unfavorable microclimate conditions within the canopy, or disease risk (Chaves et al. 2007). Too little water can cause photosynthetic decline (Maroco et al. 2002), reduced yield, and loss of canopy leaf area (dos Santos et al. 2003, Mendez-Costabel et al. 2014). Irrigation is required in many warmer grapegrowing regions, such as in the Mediterranean areas, where without irrigation, water stress would result in vine stress and low yields, which could be economically unviable (Buesa et al. 2019). In recent decades, irrigation is implemented more widely as climate change increases global temperatures and rainfall patterns change (Graveline and Grémont 2021). As an example, to ensure viable yields, some areas in France have changed their regulations to allow irrigation in regions where it was previously prohibited (Fandl 2018). Irrigation practices have also changed. Irrigation in winter has become more common in warm areas of Australia to counteract changed rainfall patterns and quantity (Bonada et al. 2021), and spring and summer irrigation has become the primary strategy to mitigate heatwaves (Webb et al. 2010, Garcia-Tejera et al. 2023). With increasing water scarcity and the associated increase in water prices, the efficient use of irrigation is paramount (Medrano et al. 2015).

The papers analyzed here implemented irrigation trials from two perspectives. One involved the application of irrigation to a traditionally rain-fed vineyard. The second investigated different strategies to reduce irrigation in traditionally irrigated vineyards. These strategies included sustained deficit irrigation, where a deficit is applied throughout the season, regulated deficit irrigation, where the deficit is targeted to a specific growth stage, and partial rootzone drying, where roots on one side of the vine are watered and the other side is allowed to dry. The aim of reducing irrigation is to reduce water use and associated costs, while minimizing yield loss and improving grape quality (Chaves et al. 2010, Buesa et al. 2017). Irrigation timing affects fruit composition outcomes (Edwards and Clingeleffer 2013) and the papers analyzed here covered a range of timings, but we segmented our data based only on the percentage reduction in irrigation.

When irrigation is withheld, the vine responds with reduced vigor (Chaves et al. 2007) caused by stomatal closure, which restricts photosynthetic activity and reduces carbon assimilation (Chaves et al. 2010). This reduced vigor comes in the form of reduced lateral shoot growth, which decreases the number of shoot tips competing for carbohydrates and directs them toward fruit and secondary metabolites (Chaves et al. 2007, 2010). Thus, when <75% irrigation was applied, TSS, anthocyanins, tannins, and total phenolics increased. When ≥75% irrigation was applied, there were insufficient changes in vigor and canopy to affect these parameters; for instance, yield was only reduced when <75% irrigation was applied (Cameron et al. 2024). Reduced irrigation reduced both leaf area and yield, so the effect of reduced irrigation on leaf area-to-fruit ratio is uncertain and could vary. The resulting sparser canopy also changes canopy microclimate, leading to more sunlight and fruit exposure, and therefore, higher temperatures (Smart and Sinclair 1976), further favoring the accumulation of anthocyanins and total phenolics (Spayd et al. 2002).

Another explanation for increased TSS and anthocyanins under reduced irrigation may be due to reduced berry weight (Cameron et al. 2024) and its effect on the skin-to-pulp ratio (Bravdo et al. 1985) and the concentration of berry quality compounds. TSS increases found with moderate water deficits have been attributed to smaller berry size; however, in berries of any size, TSS were reduced with more irrigation (conversely, increased with reduced irrigation) (Roby et al. 2004). Another study found little difference in skin-to-pulp ratio due to berry size (Walker et al. 2005).

Exposed grapes have a lower total acidity and higher pH than shaded fruit, with malic acid being lower in exposed fruit (Kliewer and Lider 1968) due to lower rates of malic acid accumulation and malate respiration in shaded clusters (Morrison and Noble 1990). Malic acid concentration is particularly sensitive to vine water status, decreasing with decreased water application (Matthews and Anderson 1988). With decreased irrigation, there is less malic acid, and thus lower TA and likely a higher pH.

The increases in pH, anthocyanins, tannins, and total phenolics and the decreased TA could have been due to the fruit being riper. The decreased TSS:anthocyanins ratio when <75% irrigation was applied (Table 3) indicates that this is not the case and that for a given TSS content, the anthocyanin content would be higher with reduced irrigation, an advantage for grape quality. Similar results have been reported (Sadras and Moran 2012). Despite increased fruit TSS and anthocyanins when <75% irrigation was applied, TSS per hectare were reduced at these levels of reduced irrigation and anthocyanins per hectare were reduced when <50% irrigation was applied (Table 3). Therefore, any water cost savings and fruit composition improvements may be offset by reduced vineyard productivity. Noting that TSS:TA ratios can be used as the basis of quality standards (Hunter et al. 2016), we found this ratio was increased when <75% irrigation was applied (Table 3). The increase in this ratio implies a lower TA value at a given TSS value. This could be undesirable in warm climates, where maintaining TA is desired and irrigation is likely to be used (Coombe et al. 1980).

Yield decreased significantly only when irrigation was reduced below ≥75% (Cameron et al. 2024). On this basis, if irrigation is reduced by only a small amount, there is negligible yield penalty and very little change in composition, only a reduction in TA. A small reduction in irrigation could therefore provide some savings in water and water-related expenses. More substantial reductions in irrigation produce greater changes in grape composition, which may benefit wine quality but reduce yield. Ideally, the grower would be compensated for this loss. A range of payment options could be considered that could help support or compensate a grapegrower to implement management practices that reduce yield. These include payments based on vineyard area, as well as bonuses for increasing parameters such as TSS, anthocyanin, or phenolic concentrations. The exact terms are obviously a commercial decision between parties.

Leaf removal

Expanding leaves are a sink for carbohydrates and transition to becoming a source of carbohydrates when they are ~30% of their final size (Hale and Weaver 1962), so the carbohydrate source-to-sink balance will vary depending on the timing, position, and severity of leaf removal (Pastore et al. 2013). It will also vary depending on the effect of leaf removal on yield, which decreased more when leaf removal occurred before or during flowering (≤E-L 26) than during berry development (E-L 27 to 33); there was no impact on yield when leaf removal occurred during veraison and berry ripening (E-L 34 to 39) (Cameron et al. 2024). The effect of leaf removal on the leaf area-to-fruit ratio would therefore depend on the balance between the stage of leaf development and photosynthetic activity (Kriedemann et al. 1970), leaf area removed, and any compensatory lateral leaf growth (Poni et al. 2006).

Traditionally, leaf removal has targeted the basal leaves in the cluster zone prior to veraison (Sivilotti et al. 2016, Poni et al. 2018). The removal of these older basal leaves is not expected to significantly alter vine overall photosynthetic activity or affect yield components and composition parameters (Bledsoe et al. 1988, Vasconcelos and Castagnoli 2000). However, although the removed mature leaves are not as photosynthetically active as younger apical leaves, their large size may offset and compensate for lower photosynthetic rates (Poni et al. 2006). Furthermore, when leaves are removed prebloom, leaf area compensates with increased lateral growth and by harvest, the leaf area-to-fruit ratio may have increased (Poni et al. 2006). Shaded leaves slowed the rates of sugar and preveraison malic acid accumulation and postveraison malic acid decline and resulted in higher malic acid, potassium, and pH at harvest (Morrison and Noble 1990). Shaded clusters did not affect sugar, acid, or potassium accumulation, but decreased anthocyanins and phenolics (Morrison and Noble 1990). These different effects of leaf removal on leaf shading and cluster shading may also influence grape composition.

TSS, anthocyanins, tannins, and total phenolics all increased when leaf removal occurred at or before E-L 33 (Table 4). This indicates that for these preveraison timings, the combination of leaf area removal and yield reduction (Cameron et al. 2024) was such that the leaf area-to-fruit ratio increased, and this assisted carbohydrate partitioning to the fruit. In addition, the compensatory leaf area is attributed mostly to laterals so it had a limited impact on the cluster zone. The cluster zone therefore received increased light and maintained a higher temperature due to the leaf removal, enabling the accumulation of anthocyanins, tannins, and total phenolics, as increased light in the cluster zone upregulates the genes required for their synthesis (further discussed in the Cluster Thinning section). TA decreased due to increased malic acid degradation resulting from warmer temperatures in the fruit zone (Dai et al. 2011, Sweetman et al. 2014). The pH was only altered when leaf removal occurred at or before E-L 18 (Table 4).

When leaf removal occurred during veraison and berry ripening (E-L 34 to 39), the only changes were decreased TA and increased TSS:TA ratio. This suggests that with late leaf removal, the removed leaves were not contributing significantly to photosynthetic capacity (Poni et al. 2006) and the vines compensated by stimulating photosynthesis in the remaining canopy and drawing carbohydrates from reserves to maintain the source-to-sink balance. The decreased TA found with late leaf removal is explained by increased light penetration into the fruit zone and higher fruit temperature, and thus, increased degradation of malic acid (Lakso and Kliewer 1975), as discussed previously.

The increase in anthocyanins with leaf removal at or before E-L 33, and the decreased TA with leaf removal at any stage (Table 4), could be due to the fruit being riper. However, we also found that the TSS:TA ratio increased significantly at all leaf removal timings. The TSS:pH ratio increased and the TSS:anthocyanins ratio decreased at all three stages of leaf removal prior to veraison (Table 4). This indicates that the changes to anthocyanins, pH, and TA are not due solely to increased fruit ripeness. Achieving a given TSS with higher anthocyanins (as implied by the decreased TSS:anthocyanins ratio) and lower pH at a given TSS (as implied by the increased TSS:pH ratio) would be considered a fruit quality improvement, particularly in warmer and warming climates, where pH values may be higher (Gatti et al. 2012), and colors less, than desired. The increased TSS:TA ratio, suggesting a lower TA value at a given TSS value, in contrast may be less desirable in warm-climate fruit. However, this may be beneficial in very cool-climate fruit or in cooler years where a reduced TA at a given TSS would make a more palatable wine.

The proportion of leaf area removed is an important part of this balance. There was a significant increase in TSS when >25 to 50% and >50% leaf area was removed, but no change when ≤25% leaf area was removed (Table 5). Yield decreased significantly when >25% leaf area was removed, but there was no change in yield when ≤25% leaf area was removed (Cameron et al. 2024), because higher proportions of leaf removal reduced fruit set (Frioni et al. 2018). With both leaf area (source) and yield (sink) decreasing, the effect on the leaf area-to-fruit ratio is uncertain. One might expect that more severe leaf removal (>25%) provided less assimilates for the ripening berry, due to a decreased source-to-sink ratio. However, the increased TSS found with >25% leaf area removed (Table 5), combined with decreased yield (Cameron et al. 2024), indicates an increased source-to-sink ratio. We conclude that there is compensation from younger, photosynthetically active leaves that redirect carbohydrates to the fruit more efficiently than the older basal leaves near the fruit that were removed. We suggest too that there is compensatory lateral growth, which increases the leaf area when leaf removal occurs prior to veraison (Poni et al. 2006). TA decreased when >25 to 50% and >50% leaf area was removed, (Table 5) due to more light penetration and higher temperatures in the cluster zone (Percival et al. 1994), and decreased malic acid levels (Esteban et al. 1999). pH was unchanged at any leaf removal severity, perhaps due to the competing influences of leaf shading and cluster shading on acid and potassium, as mentioned earlier (Morrison and Noble 1990). The increased anthocyanins when >25 to 50% leaf area was removed, and the increased total phenolics when >25 to 50% and >50% leaf area was removed (Table 5), are consistent with a higher leaf area-to-fruit ratio and increased light and temperature in the cluster zone. The TSS:anthocyanins ratio was unchanged at any level of leaf area removed (Table 5), suggesting that the timing of leaf removal had more impact on color than its severity, because the TSS:anthocyanins ratio decreased at each phenological timing of leaf removal prior to E-L stage 34 to 39 (Table 4), implying increased color at a given TSS value.

As with cluster thinning, with leaf removal, the increases in anthocyanins and total phenolics were proportionally larger (~15%) than the increases in TSS (~1 to 4%) (Table 4). Another review on leaf removal reported a similar finding but did not comment on possible causes (VanderWeide et al. 2021). We suggested earlier that with cluster thinning, anthocyanins respond to the increased light penetration and temperature of the cluster when leaves are removed, where TSS does not (Morrison and Noble 1990). We propose the same explanation for the relative greater effect of leaf removal on anthocyanins than on TSS. This proposal assumes that any lateral compensatory regrowth is predominantly in the apical parts of the shoot and not in the cluster zone. It is often not clear where the additional lateral growth was located in leaf removal studies (Poni et al. 2006, 2009, 2013). However, the results in these studies were consistent with the cluster zone being more open, suggesting that any lateral growth is mainly apical. This larger increase in anthocyanins than in TSS due to leaf removal is reflected in the decreased TSS:anthocyanins ratio (Table 4).

Viewing these results in light of vineyard productivity, the increased TSS and anthocyanins in the fruit did not fully compensate for the loss in yield (Cameron et al. 2024), as the overall production of TSS per hectare was reduced at all leaf removal timings prior to E-L 34 to 39, and when >25% leaf area was removed (Tables 4 and 5). Anthocyanin productivity per hectare may be unchanged if leaf removal occurs before or after flowering (E-L 19 to 26) (Table 4) or if 50% or less leaf area is removed (Table 5).

Leaf removal can improve fruit composition through increased TSS, anthocyanins, tannins, and total phenolics, depending on the timing and severity of leaf removal, and can decrease disease incidence through lower canopy density (Poni et al. 2023). However, it must be viewed in the context of a warming and unpredictable climate. If the temperature of the exposed fruit is too high, phenolic synthesis is inhibited (Bergqvist et al. 2001), the fruit may sunburn (Gambetta et al. 2021) and shrivel (Bonada et al. 2013), and the increased TSS from preveraison leaf removal may result in an undesirably earlier and compressed vintage (Cameron et al. 2020).

Pruning timing

Different pruning timings have been investigated with the aims of avoiding early spring frosts (Friend and Trought 2007), delaying sugar accumulation, and spreading out the harvest period to mitigate harvest compression (Petrie et al. 2017).

Late pruning removes significant leaf area, in which the vine has already invested significant carbohydrate resources (Frioni et al. 2016). It can also result in a smaller canopy size, which reduces carbohydrate availability (Silvestroni et al. 2018). Yield was unchanged when pruning occurred at E-L 2 to 10, but significantly reduced at E-L 11 or later (Cameron et al. 2024). Therefore, pruning at E-L 2 to 10 could be expected to reduce the leaf area-to-fruit (source-to-sink) ratio, but for even later pruning at E-L 11 and after, the effect on the leaf area-to-fruit ratio is less certain because both leaf area and yield are reduced.

Late pruning is used to delay budbreak and to maintain this delayed phenology throughout the growing season (Petrie et al. 2017). Consistent with this, the TSS was lower when vines were pruned at E-L 2 to 10 (Table 6). The smaller canopy size of these later-pruned vines can also lower the leaf area-to-fruit ratio, meaning there is less carbohydrate available to ripen the fruit, further reducing TSS. When pruning occurred even later, at E-L stage 11 or after, TSS was not altered (Table 6). There was a marked reduction in yield for vines pruned at or after E-L 11 (Cameron et al. 2024). The improved source-to-sink ratio is likely to have advanced grape sugar accumulation and maturation enough to catch up to the control treatment. Anthocyanins and total phenolics increased when pruning occurred later, at or after E-L 11 (Table 6). This suggests that the reduced canopy size from this late pruning opened the cluster zone, allowing more light penetration and increasing temperature, which facilitated synthesis of these components (as discussed earlier). Although overall, pH decreased and TA increased with late pruning (Table 1), consistent with the fruit being less ripe in the treatment than in the control, there was no change to TA or pH when the data were split between the two pruning timing groups (Table 6). The decreased TSS:pH and TSS:TA ratios when pruning occurred at E-L 2 to 10 (Table 6) point to higher pH and TA at a given TSS value. The increase in TA relative to TSS can increase fruit quality, particularly in warmer climates struggling to maintain adequate acid in the fruit (Gatti et al. 2012). Gatti et al. (2012) point to the work of Kliewer and Schultz (1964) and suggest that tartaric acid synthesis may be assisted by increased light due to defoliation, and that this may offset some loss of malic acid and raise TA in berries that are more exposed.

Although TSS per hectare and anthocyanins per hectare were unchanged overall (Table 1), both decreased significantly when pruning occurred at E-L 11 and later (Table 6). This reflects lost vineyard productivity and reinforces the notion that if growers prune at or after E-L 11, they will have significant productivity losses, whether measured as tonnes, TSS, or anthocyanins per hectare. If delayed pruning is imposed by the winery purchasing that grower’s fruit, careful consideration may be needed to ensure the grower is compensated for implementing this management practice.

Depending on the exact timing, late pruning could result in lower TSS and a delayed maturity, which is useful in regions experiencing high TSS levels and compressed harvest periods (Petrie and Sadras 2016, Petrie et al. 2017, Cameron et al. 2020). However, the yield penalty may be severe depending on the timing of pruning (Cameron et al. 2024). Delayed pruning could provide additional quality improvements due to lower pH, higher TA, increased anthocyanin and phenolic content, and changes in some fruit composition parameter ratios, depending on the timing of pruning.

Pruning severity

Bud load is an important factor affecting the yield of grapes, as it determines canopy size and numbers of shoots and clusters per vine (Tangolar et al. 2015). Bud load can be easily manipulated at pruning to obtain the desired yield (Greven et al. 2014). However, pruning is expensive and requires skilled labor (Poni et al. 2004). Mechanical pruning techniques have been investigated, such as minimal pruning of cordon-trained vines, to reduce vineyard input costs (Clingeleffer 2010) or to delay fruit maturity under changing climate conditions (Zheng et al. 2017). Researchers have investigated the effect of increased bud number due to pruning systems, both hand-pruned or mechanical, to determine the relationship between yield and wine quality (Howell et al. 1987, Clingeleffer 1989, Poni et al. 2004, Tangolar et al. 2015).

The only composition change found due to bud number was reduced TSS when more than twice the bud numbers were retained (Table 7). Increased bud numbers led to increased yield, more so when >5x buds remained than when >2 to 5x or ≤2x buds remained (Cameron et al. 2024). When bud number was more than twice the control, the leaf area-to-fruit ratio decreased, leading to more competition between clusters for the available carbohydrate being synthesized by proportionally fewer leaves, thus lowering TSS. With up to twice the bud numbers in the treatment compared to the control, the leaf area-to-fruit ratio was not altered sufficiently to change TSS in the fruit. The minimal impact of bud number on other composition parameters suggests that the change in leaf area-to-fruit ratio was not sufficient to alter their synthesis or degradation. This supports findings that, despite a five-fold difference in yield due to increased bud number, juice composition was largely unaffected and there was an adequate leaf-to-fruit ratio, regardless of increasing bud number (Clingeleffer 1989).

Despite few compositional changes, some ratios changed due to pruning severity. TSS:TA and TSS:pH decreased when more than twice the number of buds were retained (Table 7). An increase in TA relative to TSS could be a positive quality indicator, particularly in a warmer climate. An increase in pH relative to TSS, on the other hand, may be a negative quality indicator because it shows that for the same TSS, the pH is higher, likely to be considered a disadvantage in warmer regions.

Although the TSS decreased, the increased yield (Cameron et al. 2024) meant that the overall total production of TSS per hectare increased significantly. This increase was not significant when bud numbers were >2 to 5x, but data were limited (Table 7). Although there was no significant change in anthocyanins with increased bud numbers (Table 7), there were more anthocyanins per hectare when bud number increased to >5x, reflecting the increased yield when more buds are retained (Cameron et al. 2024).

Despite increasing yield, increased bud number did not change grape composition other than lowering TSS (>2x bud number) and pH, which may be an advantage in regions where advanced grape ripening and high sugar content has been a problem.

Shoot thinning

Shoot thinning reduces canopy density by removing fruit, leaves, shoots, and shoot tips to optimize yield, meet yield targets, and improve grape quality (Tardaguila et al. 2008, Bernizzoni et al. 2011) by optimizing canopy density (Poni et al. 2018). Light penetration and canopy temperature can increase with shoot thinning, which could lead to decreased TA and increased anthocyanins and phenolics (Reynolds et al. 2005, Poni et al. 2018). As shoot thinning removes both fruit and leaves, the leaf area-to-fruit ratio does not necessarily change, despite yield reduction (Cameron et al. 2024). However, when imposed early in the season, there is often compensation and full recovery of leaf area and photosynthetic activity (Bernizzoni et al. 2011, Poni et al. 2023). Shoot thinning can be biased toward removing unfruitful shoots, which could lower the leaf area-to-fruit ratio. TSS increased with earlier shoot thinning, at or before E-L 18 (Table 8). This is consistent with there being compensatory canopy growth that increased the leaf area-to-fruit ratio and the availability of carbohydrates for the remaining fruit.

The only other change in composition was increased pH, which occurred regardless of timing of shoot removal (Table 8), but was not altered based on the severity of shoot removal (Table 9). A recent review indicated that in most cases, increased TSS and anthocyanins were found with decreased shoot density (Poni et al. 2023). Our analysis found no significant changes in TSS or anthocyanins based on the severity of shoot removal (shoot density), but there were no available data on anthocyanins when >50% shoots were removed (Table 9).

The small composition changes led to one composition ratio change: an increased TSS:TA ratio when shoot thinning occurred at E-L 19 to 39 (Table 8). As discussed previously, this change in TSS:TA ratio could be advantageous or disadvantageous, depending on the expectations for the region and/or desired wine style. With respect to overall vineyard productivity, TSS per hectare and anthocyanins per hectare were reduced, regardless of timing, when >25% shoots were removed (Tables 8 and 9). This reflects the yield reductions found due to shoot thinning regardless of timing, and when >25% of shoots were removed (Cameron et al. 2024).

Shoot thinning, commonly used to obtain a desired shoot density, is a time consuming and expensive operation that seeks to improve canopy microclimate (De Bei et al. 2020). Given the minimal improvements in fruit composition and significant vineyard productivity reductions found in yield, TSS, and anthocyanins per hectare, implementation of shoot thinning should be considered carefully (De Bei et al. 2020), taking into account the particular challenges, such as disease or vigor, that might be present in each vineyard.

Shoot trimming

Shoot trimming is employed widely in viticulture. At a practical level, it is used to reduce foliage to facilitate vineyard operations such as tractor and human access, or to make bird netting installation easier, potentially without much thought to its effect on yield or grape composition (Poni et al. 2023). A light shoot trimming around flowering was examined with the aim to increase fruit set and yield (Collins and Dry 2009). Shoot trimming is also used to reduce and open up the canopy. This could reduce disease pressure in sprawling vines (Bettiga et al. 1989, Poni et al. 2014) and improve anthocyanin content by increasing light penetration, temperature, and air circulation in the fruit zone (Cañón et al. 2014). This depends on the timing of shoot trimming, as later shoot trimming is more likely to permanently change the leaf area-to-fruit ratio because there is minimal regrowth. The severity of the trimming determines whether the carbohydrate limitation significantly reduces TSS (Poni et al. 2023). In this way, shoot trimming has been used to reduce berry sugar accumulation (Filippetti et al. 2015, Herrera et al. 2015) by removing a proportion of the carbohydrate source (Valentini et al. 2019). If not too severe, shoot trimming may reduce TSS without decreasing anthocyanin content (Poni et al. 2023). Shoot trimming can also promote lateral shoot growth, which redirects carbohydrates from the fruit to this lateral growth (Collins and Dry 2009). When shoot trimming is applied to vertical shoot-positioned canopies, the fruit zone microclimate is largely unchanged, as lateral shoots develop mainly on apical nodes (Poni et al. 2009, Molitor et al. 2015).

The main change in fruit composition was decreased TSS, the magnitude of which was similar regardless of timing of shoot trimming (Table 10). A decreased pH was also found when shoot trimming occurred during berry development, i.e., E-L 27 to 33 (Table 10). These changes are consistent with a loss of leaf area reducing the source-to-sink ratio, meaning there are less available carbohydrates for the fruit. Yield was not impacted by shoot trimming (Cameron et al. 2024). The TSS decrease only occurred when >25% leaf area was removed and was most reduced when >50% leaf area was removed (Table 11), due to the greater leaf area reduction and restricted carbohydrate availability for the fruit. The severity of shoot trimming was more important than timing: when shoot trimming removed >50% leaf area, both anthocyanins and pH were reduced, in addition to the TSS reduction (Table 11).

Given the challenges of high sugars and advancing ripeness due to climate change, the restriction in TSS from shoot trimming could be advantageous in regions where increasing grape sugar content is an issue. TSS could be reduced without affecting anthocyanin content, by shoot trimming to remove >25 to 50% leaf area across any phenological stage (Tables 10 and 11). This is consistent with the proposal that “vine metabolism tends to delay sugar accumulation more than color accumulation” (Poni et al. 2023).

Despite no reduction in yield found due to shoot trimming (Cameron et al. 2024), the decreased TSS meant that overall productivity for the vineyard was reduced with a decreased TSS per hectare, when shoot trimming occurred during berry ripening at E-L 34 to 39 (Table 10). Anthocyanins per hectare also decreased when trimming occurred during berry development, at E-L 27 to 33 (Table 10), further pointing to potential productivity loss if shoot trimming occurs during berry development.

Shoot trimming has the potential to reduce TSS in the fruit. The interest in shoot trimming is its potential for mechanization, to provide a relatively inexpensive way to address the challenges and changes found with increased sugar content of fruit in a changing climate (Poni et al. 2023).