If the gradient decreases along the terraces then the slope gradient will also
have to decrease to keep
the height constant. Therefore, the horizontal ledge of the slope (distance between two consecutive
terraces) will increase. With this new geometry, it may be of interest to continue planting on the terra-
ce or change from the terrace to the slope. Depending on the natural gradient and on the extent of the
lengthways variations of the slope, the optimum will be different. To obtain a precise result, a 3D model
Manual of techniques for sustainable mountain viticultur
e - 67
Chart 4.1 Comparison between slope and terrace planting with constant natural gradient
Theoretic ELA: effective leaf area of all stock planted in 1 ha; calculated for shoots of 0.14 m
2
of ELA with an average space
between shoots of 7 cm.
(1) The limiting factor is the horizontal ledge of the slope (parameter p in Figure 2.2), which must be over 0.6 m so that the
ring of the upper terrace does not interfere with passing along the lower terrace (with a sufficient gap). This prevents the use
of slopes with a higher gradient.
(2) The limiting factor is the height of the slope (maximum 1.5 m), which prevents more gentle gradient slopes from being used.
Where the height of the slope is limited to 2 m, the resulting ELA increases to the values of the lower row.
Natural slope (%)
20
30
40
50
60
Terrace width (m)
1.3
1.3
1.3
1.3
1.3
Slope gradient (º)
32
43
51
58
62
Slope height (m)
0.38
0.57
0.77
0.95
1.15
Slope length (m)
0.7
0.8
1.0
1.1
1.3
No. of terraces per ha
52
52
52
52
52
Theoretic ELA (m
2
/ha)(1)
24,504
24,504
24,504
24,504
24,504
Terrace width (m)
1.3
1.3
1.3
1.3
1.3
Slope gradient (º)
14
22
32
42
52
Slope height (m)
1.31
1.51
1.44
1.46
1.47
Slope length (m)
5.4
4.0
2.7
2.2
1.9
No. of terraces per ha
15
19
27
34
40
Theoretic ELA (m
2
/ha)(2)
18,371
17,796
16,774
16,605
16,920
Theoretic ELA
19,757
19,036
18,629
18,680
18,856
(slope height <2 m)
Circular
plantation
on terrace
Double vine
training
plantation
on slope
must be prepared to introduce the topography of the land and the design conditions of the terraces.
However, certain illustrative criteria can be established based on a 2D model:
•
If the average gradient of the land is only slightly less than the maximum (e.g. up to 33% less), all
the vineyard should be planted on the terrace.
•
If the average gradient of the land is much less than the maximum (e.g. up to 66% less), all the
vineyard should be planted on the slope.
•
In intermediate cases, planting should be started on the terrace and the slope then used at the
point
where the vineyard develops, which will depend on the average gradient: the further away
the average gradient from the maximum, the earlier planting should be changed to the slope.
In all cases, planting on slopes makes viticulture work more difficult:
•
Precise irrigation of intermediate stock (that is not on the terrace but on the slope) is more com-
plicated because it is on a gradient.
•
Access to intermediate stock may require climbing the slope on foot. Where the slope gradient
exceeds 30-35%, steps must be installed.
It also has advantages, however:
•
The height of the slopes is greater and, therefore, fewer terraces are required.
•
Double vine training is cheaper than ring vine training.
In general, it is only wise to plant on slopes when the ELA gain is significant (e.g. over 15%), although
this will depend on the criteria of each vine grower.
One way of making slope growing easier is to plant only the end two stocks on the terraces and leave
the liana plant to develop its production branch along the slope with no intermediate stock. To do so,
Mas Martinet is carrying out experiments in order to answer two questions:
•
Is it possible to accelerate plant growth to form the entire production branch in fewer years through
irrigation?
•
How does the grape quality vary as the shoots grow further away from the stock?
The results of these experiments are not yet available.
Manual of techniques for sustainable mountain viticultur
e - 68
To show the increase in productivity achieved by combining the terracing techniques with those of
vigour control, Chart 4.2 compares the soil area required to produce 10,000 kg of grapes (equivalent
to around 8,600 bottles of wine) for three forms of cultivation:
•
Conventional terraces with the following characteristics:
- Terrace width: 2.3 m (2 rows of stock on each terrace).
- Slope gradient: 45º (100%).
- Formation: cordon royat.
•
Mas Martinet terraces and ring vine training on the terrace.
•
Mas Martinet terraces and double vine training on the slope.
In al cases, the natural gradient of the land remained constant at 40%.
As shown by experience, the real ELA is seen to be 65% its theoretic value (see Section 3.4). This also
occurs in conventional plantations, given that some stock is not feasible and other does not reach the
expected ELA.
In line with Chart 4.1, ring vine training provides the best results. Somewhat more than 1 ha of land is
required to produce 10,000 kg of quality grapes. Conventional planting has a lower productivity and
requires more than 3 ha of land.
Note that, if 6,000 kg/ha are collected in a conventional plantation like the one considered, the resul-
ting production per m
2
of ELA is 1.3 kg, which is very high for preparing a quality wine that can withs-
tanding a good ageing process. The significant parameter is not production per ha, as regulations are
often limited to, but production per m
2
of ELA truly developed on productive stock.
In other words, for each plot, variety and weather, etc. the optimum ratio between ELA and the pro-
duction of a quality grape can be assessed, although there is no optimum ratio per ha. As explained
in the previous sections, it is worth noting that production is not linked to the number of shoots but
to the ELA. The number of shoots depends on the vigour of the stock in order to obtain grapes with
the appropriate morphology for their quality.
Manual of techniques for sustainable mountain viticultur
e - 69