“New management systems for steeply sloping terraced vineyards. Trials on Merlot in Ticino”.
Summary
A trial of new management systems has been set up with the Merlot grape variety at the Gudo experimental vineyard (TI), cultivated on terraces. The natural slope of the land is 65% and the distance between benches or rows is 3m. Six management systems were tested, with the aim of making better use of the available space. The traditional control, with double Guyot and single vertical trellising, was compared to variants with double trellising, ascending and descending over the slopes, either Guyot or cordon trained. The bud and shoot load of the double trellis systems was increased by 20-60% compared with the control.
On average over the years (1995-2000), this increase in load has made it possible to increase yields by 30-50% without any detrimental effect on grape and wine quality. This quality preservation is essentially due to the ratio of exposed leaf area (ELA) per kg of grapes, which was kept close for all variants. Cordon pruning increased the weight of pruning wood and berries, and reduced bud fertility, as well as the percentage of unbroken buds and weak shoots. The new management systems tested led to an increase in workload of between 30 and 150 h/ha, compared with the control and depending on the system. These extra hours were largely offset by the increase in yield.
Key words
Management systems, canopy management, bench vines, yield, quality, exposed leaf area.
INTRODUCTION
More than half of Switzerland’s vineyards are located on steeply sloping hillsides (>30%).
The creation of benches (flat areas supported by grassy embankments) makes it possible to rationalize the cultivation of these vineyards.
The rows are oriented across the slope, following the contour lines.
This technique protects the soil from erosion and significantly increases mechanization possibilities.
On the other hand, it reduces planting density and space occupation, as the distance between rows often exceeds 3m on very steep slopes (>50%).
In addition to the flat land required for machine passage, the space occupied by the embankments supporting the terraces must also be taken into account (Murisier, 1981; Murisier et al., 1984).
Vineyards are often grown on terraces in Ticino to avoid the risk of erosion caused by high rainfall.
The traditional training system used is double Guyot, with a single vertical trellising system. As a result, leaf area per hectare is low. Depending on yield, this may not be sufficient to obtain optimum quality. Planting density can be increased by tightening the rows of vines. This technique does not alter the occupation of space, nor the exposed leaf surface, nor the level of yield and quality (Pelossi, 1993; Murisicr and Ferretti, 1996). To make better use of the available space and enhance the value of vigile cultivation on steep slopes, new management systems have been tested, dividing the vegetation planes into two (ascending and descending). The overall aim is to improve production while preserving quality.
MATERIALS AND METHODS
Experimental site
The trial of new management methods was instituted in 1989 at the Gudo experimental vineyard (Ticino) with the Merlot grape variety grafted onto 3309C.
The average slope of the land is 65%, with a south/south-east exposure.
The benches were created perpendicular to the slope.
The width of the slope is 1.50m and the distance between benches or rows is 3m.
At Gudo, the average annual temperature (1995-1998) is 12.1°C and rainfall is high (1,666 mm).
Training systems
Six training systems were weighted.
In addition to the traditional control (double Guyot: 300 x120 cm), variants with double trellising, ascending and descending over the slopes, were compared.
In these latter variants, different types of pruning (Guyot and cordon) and various row spacings were studied.
In the ascending and falling trellis systems, the shoot load was increased by 20 to 60% compared with the control, depending on the variety.
The length of ascending and descending shoots was kept constant by trimming, i.e. 120 cm for all systems.
Each variant comprises 4 replicates.
Checks carried out
Agronomic checks covered yield, berry weight, sugar content (g/l), total acidity (g/l), tartaric acid and malic acid, with separate analysis of ascending and descending shoots.
Bud fertility, percentage of uncapped eyes, pruning weight and trimming weight were also monitored.
Exposed leaf area (ELA) was approximated by measuring the area of the outer canopy receiving direct light (Murisier, 1996).
Any large holes in the vegetation are subtracted from this area (Carbonneau, 1976 and 1980).
For the ascending foliage plane, the surface of the two side walls and that of the top of the vegetation were taken into account.
For falling vegetation, only the surface area of the south face was taken into account, as the northern part (on the slope side) is constantly in the shade.
The structure of the leaf surface, as well as the illumination of the foliage and clusters, was analyzed using the methods proposed by Carbonneau (1976 and 1980).
The different variants were vinified separately. The grapes from the descending and ascending shoots were vinified separately until the end of alcoholic fermentation, so that comparative polyphenol analyses could be carried out. The two batches were then blended, to determine the qualitative value of the wines from the entire management system tested.
RESULTS
The increase in bud and shoot load (20-60%) in the double trellised variants resulted in an increase in yield per unit area (30-50%) compared with the control.
The proportion of grapes harvested from ascending and descending shoots was virtually identical, with the exception of variant D, where the proportion of grapes harvested from descending shoots was lower.
This is due to the fact that fewer shoots were kept on the fruiting branch pulled perpendicular to the row.
Yields per vine are a function of row spacing and bud load.
Variants C and F, with an inter-row spacing of 60cm, have the lowest yields per vine, but yields/m2 comparable to variants B, D and F.
Bud fertility
Bud fertility in permanent cordon pruning (variants B, E and F) was significantly lower than in Guyot pruning.
The differences were significant, ranging from 0.2 to 0.3 bunches per bud.
Differences in fertility between ascending and descending shoots are small.
This probably reflects the effect of comparable foliage exposure at flower initiation.
Bay peas
Berry weight was notably higher in short pruning (cordon) than in long pruning (Guyot).
This phenomenon compensates for the lower fertility observed in cordon pruning.
Bunches of grapes from falling shoots tended to bear slightly smaller berries than those from ascending shoots.
However, the differences were not significant.
Sugar content
Overall, the different management systems produced similar sugar contents, despite the higher yields observed in the double trellised variants.
The new management systems increased yields by 30-50%, without compromising grape quality.
This is due to the increased exposed leaf area (ELA) achieved by doubling the trellising planes.
The ELA/kg of grapes ratios were fairly close for all variants, between 0.8 and 1.0 m2 of illuminated leaves per Kg of grapes.
These values are slightly lower than the optimum values defined by Murisier (1996) and Murisier and Zufferey (1997) for the Chasselas and Gamay grape varieties.
The significantly higher sugar content of variant F is difficult to explain by a single parameter.
It may be due to the combined effect of different factors (double vegetation, cordon pruning or row narrowing).
Musts acidity
No effect was observed on total must acidity or tartaric acid content.
On the other hand, in terms of malic acid, grapes from falling shoots have a systematically lower malic acid content than grapes from ascending shoots.
The lower development of drooping shoots alone cannot explain this phenomenon.
On the other hand, control A (Guyot), which has a vigor comparable to that of the drooping part of variant E (cordon), has a malic acidity close to that of grapes from ascending shoots.
Measurements of the cluster microclimate could help explain the differences observed.
Pruning and trimming weights
As with ELA/m2, double trellising systems produced significantly higher pruning weights than the control (A).
In addition, cordon pruning (H, E and F) produced significantly higher total shoot weights than Guyot pruning (C and D).
The differences are mainly due to the ascending shoots.
The same observations can be made for the weight of prunings.
In terms of weight per shoot (main shoot + internodes), it can be seen that in each variant, falling shoots are weaker than ascending shoots.
Measurement of trimming weight shows that falling shoots are less developed than ascending shoots.
It would be important to check the weights of shoots and trimmings from the main shoots and from the internodes separately.
Merlot frequently presents bud-break difficulties (Madonna, 1988; Murisier el al.,l990). Cordon pruning (B, E and F) markedly reduces bud-break problems. Falling trellising tends to produce more regular bud break.
Wine quality
Differences in the quality of wines from different variants were small both analytically and sensorially.
In terms of total polyphenols and anthocyanins, analyzed after alcoholic fermentation, the double trellising variants tended to have slightly higher values than the control.
Differences between wines from ascending and descending shoots were also small.
The tasting carried out in February 2001 on all variants and years did not reveal any significant differences between the wines of the various management systems tested.
Table 1. ESSAi systems sm Meriot in Gudo (Tl). Analysis of po]ypelols after aleoolic]ue cImentation (~I 19%-1999) and results of wine tasting on 27.02.2001.
Production costs
Double trellising systems require more care to be taken with the vines, in particular to separate upward-facing and downward-facing vegetation.
For all vine care operations (pruning and green harvesting), the additional labor required by the new double trellising systems compared with the control system varies from 30 to ISO h/ha, or from 600 CHF/ha to 3,000 CHF/ha.
The extra work is largely offset by the higher yields.
In fact, 30 to 150 h/ha of extra work correspond to 600 to 3,000 CHF/ha for 3,000 kg/ha more harvest, or around 13,000 CHF/ha.
CONCLUSIONS
– In steeply-sloping vineyards grown on terraces, the increased shoot load combined with the doubling of the trellising planes (ascending and descending) increased yields per unit area by 30-50% compared with the conventional system with a single vertical trellising plane.
– The increase in yields obtained with double trellising systems has not led to any reduction in grape or wine quality.
– The quality level is essentially explained by the SFE ratio per kilo of grapes, which was very close for all variants.
– Cordon pruning increased the weight of berries and pruning wood. It reduced bud fertility and the percentage of buds that remained unfilled.
– Overall, falling branches were less vigorous than ascending branches. It would be important to check whether this phenomenon is also evident in the between-hearts.
– Cluster-specific microclimate control could help explain some of the differences observed in malic acid levels.
– The new management systems required an increase in vineyard work of between 30 and 150 h/ha, depending on the system. These extra hours were largely offset by the increase in yields.