|
A Technical Guide for Wine Producers
|
|
RECENT ARTICLES | WYNBOER HOME
Soil type may influence wine style: Cabernet Sauvignon from Durbanville and Robertson
Introduction
All over the world the identification of viticultural terroirs (Carey et al., 2002) is currently receiving a lot of attention. A natural terroir unit (NTU) is basically characterised by a relatively homogeneous pattern with regard to topography, climate, geology and soil. It can therefore be defined as a complex of natural environmental factors that cannot easily be changed by the producer. In conjunction with various managerial decisions taken by the producer, this complex may reflect in the end product. The result should therefore be wines with a characteristic, identifiable origin. For the Loire Valley in France, it has been found that basic terroir units may have a significant effect on the character of Cabernet franc wines (Morlat, 1989).
In South Africa research about the effect of various soil types on wine quality was already conducted in the 1970s (Saayman, 1977). Subsequently, for a long time, there was little research about the concept of terroir. However, since 1994 intensive research (partially funded by Winetech) has been resumed by ARC Infruitec-Nietvoorbij, in collaboration with the University of Stellenbosch. These terroir studies basically require two steps (Carey et al., 2002). Firstly all the relevant natural factors (aspect, height above sea level, geology, soil depth, water housekeeping, etc.) have to be identified, so that relatively homogeneous NTUs can be indicated. Secondly wine quality must be determined for vineyards on various terrains, so that those producing comparable products might be grouped together as viticultural terroirs. With regard to the first priority (identification of relatively homogeneous NTUs), much progress has been made. For example, a map has been compiled that identifies natural terroir units for an area of 25 000 ha, west and south west of Stellenbosch, (Carey, 2001). Inasmuch as the second priority is concerned (determination of wine quality for vineyards on various terrains), considerable lacunae remain. This situation is partially to be blamed on the fact that vineyards (terroirs) which constantly produce wines with a specific character, have not yet been clearly identified in South Africa (with its relatively young winemaking tradition), compared to Europe (where the winemaking tradition stretches over many hundres of years).
It is fairly generally accepted (Rankine et al., 1971; Winkler et al., 1974) that climate in the warmer wine producing countries has a dominant effect on wine character. For Sauvignon blanc from various localities in Stellenbosch/Klein Drakenstein it has been found that seasonal variations in temperature and rainfall can have a significant effect on wine style (Bonnardot et al., 2000). The location of specific sites, each with its unique topoclimate, caused seasonal variations in wine style to be either bigger or smaller. For Sauvignon blanc from Stellenbosch/Durbanville, different wine styles could also be identified for different terrains (Conradie, 1998). In some instances climate/soil interactions could be observed, however, consequently wine styles for specific terrains differed from season to season. Similar trends were also observed by Saayman (1977). In other instances wine style remained fairly constant from year to year, regardless of seasonal variations in climatic conditions. In a study on Cabernet Sauvignon in the Stellenbosch/Drakenstein vicinity, seasonal differences were also observed, but the effect of terroir on wine style was still noticeable over the years (Carey, 2002).
The above indicates that climate does indeed have a significant effect on wine character, but that it cannot be seen in isolation. One of the most important aspects of the terroir concept that should be taken into account, is the effect of soil on wine character/wine style. Various pedological factors should therefore be borne in mind. Chemical composition and soil depth definitely have an effect on the growth pattern of the vine, and consequently also on wine quality. However, it is fairly generally accepted that the water housekeeping of soil is the most important pedological factor with regard to wine quality (Seguin, 1986). In Europe (with a high rainfall in summer) there is a marked preference for well-drained soils (Saayman, 1992). The situation in South Africa (low rainfall in summer) is not necessarily comparable. Under dry land conditions soils that drain moderately (e.g. duplex soils), can have a beneficial effect in that they can prevent vineyards from being exposed to excessive water stress during the last part of the growing season. One of the prerequisites for a good terroir is that wine style should remain fairly constant from year to year. In the Western Cape, where soil acidity is a common occurrence, effective soil preparation and liming should result in well-developed root systems, which may serve as a buffer against seasonal variations in climate. Although soil water status was not measured in the study on Cabernet Sauvignon (Carey, 2002), it seemed as though a high soil pH correlated with wine characteristics that might have a positive effect on balance and quality.
In conjunction with the above-mentioned results for Cabernet Sauvignon (Carey, 2002), the effect that the water housekeeping of soils may have on wine style will be illustrated in this article. Moreover, the effect of various soil types, as they regularly occur in practice within the same vineyard block, will be pointed out.
Material and methods
The investigation was conducted in two Cabernet Sauvignon vineyards. The first, cultivated under dry land conditions, was situated in Durbanville. Two soil types (Sterkspruit and Oakleaf) were identified in this vineyard. The most important difference between the two soils was an improved rooting system and a slightly higher water retention capacity in the case of the Oakleaf. The second trial vineyard, situated in Robertson, was cultivated under irrigation. In this case too two different soil types (Tukulu and Fernwood) were identified. As indicated in Figure 1, these two soils differed drastically. In the Tukulu soil type a high percentage of stone was present, while the clay content was high (> 40%). Hardly any stones occurred in the Fernwood soil type, while the clay content was low (< 5%). At both localities trial sites were marked out on the various soil types. At neither of the localities these sites were further than 50m from each other. Leaf water potentials and soil water were measured weekly during the 2000/2001 growing season, while grapes were harvested at 24oB. Experimental wines were vinified separately from grapes from the different soil types. Wines did not undergo any wood maturation. After six months various aroma components were evaluated by an experienced panel in the light of the method described by Noble et al. (1987).
Results and discussion
In Durbanville soil water content, especially during the last part of the growing season, was slightly higher in the Oakleaf soil than in the Sterkspruit (not indicated). Leaf water potentials showed that water stress was lower in vines growing on the Oakleaf (Fig. 2). Aroma intensity and fullness were comparable for wines from the two soils (Fig. 3). Where water stress was highest (Sterkspruit), vegetative character (grass/sweet pepper) was dominant, while it was lower in vines experiencing less water stress (Oakleaf). In the latter instance berry character (raspberry/strawberry) was very prominent, while a spicy character (aniseed/black pepper) could also be observed much more clearly. Consequently wine styles differed largely, although not one of the wines could be considered better/worse. It is a known fact that different Cabernet Sauvignon clones result in different wine styles. Producers are therefore advised to plant various clones, in order to obtain a broader spectrum of aroma profiles. However, results in Fig. 3 show that different soil types, even in the same clone, can also result in widely divergent wine styles. In practice grapes from the same block are usually pressed together, which should result in a complex wine with a large variety of aroma components for the Durbanville soils. Producers should also be able to keep wines from the same block separate in order to be able to market a wider spectrum of wines.
The fact that wines from the drier soil, where the vineyard grew slightly less vigorously, tended to a vegetative character is in contrast with the situation for Sauvignon blanc, where vegetative character tends to fall away on drier soils (Conradie, 1998). However, a similar reaction (lower vegetative character in soils with a higher water retention ability) was also obtained for Cabernet Sauvignon in other localities (not indicated). The reason for this 'contradictory' reaction of these two cultivars is currently not clear and is being investigated further. Apparently white and red cultivars do not necessarily react in an identical fashion with regard to the development of aroma components.
Wine from the heavier soil (Tukulu) in Robertson was fuller on taste, with a higher aroma intensity (Fig. 4), compared to that from the more sandy soil (Fernwood). Vegetative characters differed little, but in the former wine berry and spice character was also more prominent. Wine from the sandy soil was therefore in a light style, which is not typical for Cabernet Sauvignon. Where vineyards are cultivated under dry land conditions, or with additional irrigation, water stress is probably the most important factor to reduce wine quality on sandy soils. Under intensive irrigation excessive water stress may be eliminated to a large extent. Consequently, for the vineyard in Robertson leaf water potentials, as well as vigour, did not differ much from vine to vine on the two different soils. Since sandy soils are badly buffered, the availability of water and nutrient elements might differ considerably from day to day, depending on the stage at which irrigation/fertilisation takes place. This situation may result, inter alia, to must with a higher pH, which is negative for wine quality. It is therefore difficult to manage these soils and in practice Cabernet Sauvignon plantings are not usually recommended for sandy soils. Otherwise it is difficult to manage vineyards correctly where two widely divergent soil types (Fig. 1) occur in the same vineyard block. For the vineyard in Robertson blending of the two wines will not result in improved complexity. On the contrary, the typical character of the wine from the Tukulu soil will probably be reduced as a result of the Fernwood soil's lack of character.
Summary
If the terroir is good, wine style should remain fairly constant from year to year. Results reported above were gathered over one season only and should be deemed preliminary. It is clear however, that the water housekeeping of soils, especially under dry land conditions, may have a significant effect on the wine style of Cabernet Sauvignon. It appears as though good soil preparation/liming, together with good water housekeeping, may result in wines with a fairly prominent berry character. In the case of "drier" soils, character tends more towards grass/sweet peppers. Even under intensive irrigation it may be difficult to obtain a typical Cabernet Sauvignon wine from sandy soils. Under such conditions a lighter style wine is apparently obtained. Since soils in the Western Cape are usually not very homogeneous, the same vineyard block will in many instances produce wines in wide-ranging styles.
Literature
BONNARDOT, V, CAREY, V & SCHMIDT, A, 2000. The effect of vintage and location on Sauvignon blanc wine aroma in the Stellenbosch-Klein Drakenstein winegrowing area (Vintages 1996-1999). Wineland, October, 114-118.
CAREY, VA, 2001. Spatial characterization of natural terroir units for viticulture in the Bottelaryberg-Simonsberg-Helderberg winegrowing area. M.Sc. Agric Thesis, University of Stellenbosch. 90pp + annexes.
CAREY, VA, 2002. Cultivar x environment interaction: example of Cabernet Sauvignon in the Stellenbosch-Drakenstein wine growing area, South Africa. Paper presented at the OIV Groupe d'Experts: Zonage Vitivinicole, March 2002, Paris.
CAREY, VA, ARCHER, E & SAAYMAN, D, 2002. Natural terroir units: What are they? How can they help the wine farmer? Wineland, February, 86-88.
CONRADIE, WJ, 1998. The effect of soil and climate on the character of Sauvignon blanc wine. In: Proc. SASEV Congress, November 1998, Cape Town, South Africa.
MORLAT, R, 1989. Le terroir viticole: contribution á l'étude de sa caractérisation et de son influence sur les vins. Application aux vignobles rouge de Moyenne Vallée de la Loire. PhD Thesis, University of Bordeaux II. 289pp + annexes.
NOBLE, AC, ARNOLD, J, BUECHSENSTEIN A, LEACH, EJ, SCHMIDT, JO & STERN, PM, 1987. Modification of a standardized system of wine aroma terminology. Am. J. Enol. Vitic. 38, 143-146.
RANKINE, BC, FORNACHON, JCM, BOEHM, EW & CELLIER, KM, 1971. Influence of grape variety, climate and soil on grape composition and quality of table wines. Vitis 10, 33-50.
SAAYMAN, D, 1977. The effect of soil and climate on wine quality. In: Proc. Int. Sym. Quality of the vintage, February 1977, Cape Town, South Africa, 197-208.
SAAYMAN, D, 1992. Natural influences and wine quality. Part 2: The role of soil. Wynboer, August, 49-51.
SEGUIN, G, 1986. 'Terroirs' and pedology of wine growing. Experientia 42, 861-873.
WINKLER, AJ, COOK, JA, KLIEWER, WM & LIDER, LA, 1974. General Viticulture, 2nd ed. University of California Press, California, 710 pp.
|
|
Wynboer is incorporated in WineLand, magazine of the SA wine producers.
Visit our sister sites:
|
||||||
Kobus Conradie, ARC Infruitec-Nietvoorbij, Stellenbosch 7599
| UP |
COPYRIGHT (C) 2000 WineLand |