The effects of climate as one of the terroir attributes on wine style and quality are generally well understood, while there is much uncertainty concerning those of soil type.
Researchers have reported that soil affects grape composition and subsequently the style and quality of wines through its moisture levels, water holding capacity and drainage properties, nutrient availability, penetrability, heat retaining and light reflecting capacities.1,2 Researchers started studies on the effect of soil type on wine quality under South African conditions during 1973.3 Such knowledge is required in order to make better viticultural and oenological decisions or predictions, as well as to ensure that the potential of a specific vineyard or wine region is fully utilised.
The study investigated the effects of soil types on the performance, wine style and quality of Sauvignon blanc through:
- characterisation of selected climate and soil parameters,
- assessing grapevine responses and
- conducting sensorial evaluation of wines.
Study layout and measurements
A Sauvignon blanc/Ruggeri 140 vineyard in a commercial wine estate at Elgin was studied for four years (2006 – 2010). The vineyard was established during 2000, at an altitude of 508 m, on a 28% slope, with SSW aspect, 1.3 x 3.0 m planting density and under drip irrigation (supplemental). The vineyard had two experimental plots representing soil properties that varied to a certain degree. An automatic weather station was installed at the site and some of the climatic data were obtained from three weather stations positioned in Elgin. Soil and root profile studies were performed at the commencement of the study. Soil water contents and grapevine water stress levels were monitored, while selected grapevine parameters were quantified. Grapevine nutritional status was assessed on leaf petioles and blades, as well as juice. The target sugar levels for harvesting were between 22.5°B to 23.0°B. The must were analysed chemically, thereafter experimental wines were prepared in duplicate by using a standard production technique. An experienced panel of judges performed sensorial evaluations on the wines. Data from the four seasons were used as replicates and analysed statistically in order to determine differences due to soil types (p ≤ 0.05).
Findings and discussion
Seasonal weather conditions
Mean February temperatures ranged between 19.6°C and 20.6°C in Elgin during the study period, qualifying Elgin to be climatically classified as a cool region (Table 1).
Weather conditions during the four seasons could be described with reference to Table 1 as follows:
- 2006/07: Normal annual rainfall, summer rainfall above long term mean (LTM), summer maximum temperature slightly higher than LTM, but Mean February temperature lower than LTM, cool ripening period (wet and slightly warm summer).
- 2007/08: Annual and summer rainfall above long term means (LTMs) (wet season), summer maximum and minimum temperatures slightly above LTMs (slightly wet and warm summer).
- 2008/09: Annual and summer rainfall below the LTMs, but spring rainfall much higher than the LTM, spring mean and maximum temperatures lower, while summer maximum and Mean February temperatures were higher than LTMs, warm ripening period (wet and cool spring, slightly dry and warm summer).
- 2009/10: Annual and summer rainfall below LTMs (dry season), spring rainfall above LTM, spring maximum and minimum temperatures below LTM and summer maximum temperature higher than LTMs (wet and cool spring, dry and warm summer).
Site climatic conditions
The site could be placed in Region I (Winkler index = 1 308 degree days), indicating its potential to produce high quality white table wines characterised by high acids, low pH and excellent cultivar character. While some parts of the traditional South African wine growing areas fall in Region II, large parts of areas, such as Robertson, Stellenbosch/Durbanville and Bottelary-Simonsberg-Helderberg, fall under Region III and even Region IV. Summer maximum temperatures and Mean February temperature were respectively about 1.6°C and 1.4°C lower than the LTM values for Elgin (Table 1). This suggested that the site experienced lower temperatures than the “average” locality in Elgin. In addition, high wind speeds, e g daily maximum wind speed of 6.2 to 7.3 m/s, were measured at the site during some seasons. Generally, cool and windy conditions prevailed at this high altitude site.
Both soils were classified as a Tukulu form (orthic A and neocutanic B horizons underlain by unspecified materials with signs of wetness). One soil contained appreciably higher amounts of gravel than the other, resulting in the soils being labelled as high-gravel and low-gravel (Table 2). Carbon levels in the upper horizons were high for both soils, pointing towards relatively high N supplying capacities. As a result of fertilisation, the high-gravel soil ended up with higher P (topsoil) and K (subsoils) levels than the low-gravel soil (Table 2). Furthermore, the high-gravel soil tended to have lower soil water content and water holding capacity (WHC), especially in the deeper subsoil than the low-gravel soil, which might have caused it to have a better heat retaining capacity.
The fine:medium root density ratios (5.9 and 9.8), indicated healthy root systems for both soils. Fine root density in the low-gravel soil was approximately double (420 roots/m2) than that in the high-gravel soil, suggesting a better root system in the low-gravel soil than in the high-gravel soil.
The uptake of P was significantly better for grapevines on the high-gravel soil than those on the low-gravel soil. This was reflected on the leaf-petioles, -blades and juice (data not shown), and largely attributed to the higher soil P levels found in the high-gravel soil.
Grapevine water stress, vigour, yield, berry parameters and juice analyses were not significantly affected by soil type. Yields were exceptionally low (< 3 t/ha) at both sites, but bunches of grapes on the high-gravel soil weighed significantly more than those of grapes on the low-gravel soil (data not shown). Must of grapes on the low-gravel soil generally had a higher total titratable acidity (TTA), i.e. 9.35 g/ℓ, than that of must of grapes on the high-gravel soil (8.89 g/ℓ).
Some notable differences in must composition that were likely to affect wine quality were observed during specific seasons. Sugar levels of must from grapes on both soils were similar and of the desired levels for winemaking purposes, but those of must on the low-gravel soil were lower by at least 1°B than those of must on the high-gravel soil, during 2007 and 2008. Sugar levels of must of grapes on both soils were above the desired values during 2010, but sugar levels of must of grapes on the high-gravel soil were 2.4°B lower than those of must from grapes on the low-gravel soil.
Wine style and quality
Only acidity was significantly affected by soil type when data from all vintages were combined and it was higher for wines produced using grapes on the low-gravel soil than for those made using grapes on the high-gravel soil (Table 3), which aligns with previous observations in relation to TTA. Noticeable acidity is also typically associated with white wines from cooler climates. Indication is that the low-gravel site was probably cooler than the high-gravel site, due to higher soil moisture.
Furthermore, it is noteworthy that soil type had significant effects on certain wine attributes in some vintages. The 2008 vintage wines produced using grapes on the high-gravel soil were more aromatic and with lower acidity than those produced using grapes on the low-gravel soil (Table 3). Wines produced using grapes on the high-gravel soil were better than those produced using grapes on the low-gravel soil, especially pertaining to overall quality, and vegetative aromas identified during the 2009 and 2010 vintages. The 2009 vintage was perceived to have been the best produced during the entire study (Table 3). The superiority of the 2010 vintage wines made using grapes on the high-gravel soil over those produced using grapes on the low-gravel soil, might have been largely attributed to lower sugar levels.
Although grapevine water stress did not differ considerably between the two soils, grapevines on the high-gravel site might have experienced sufficient water stress, which resulted in better grape-juice composition, especially during seasons with wet conditions. This could be because the high-gravel site had soils with the ability to retain less water and probably more heat, as well as less developed root systems compared to the low-gravel site. However, during dry and/or exceptionally warm seasons, a reverse situation may be expected at such cool localities, and soils with high a WHC and better developed root systems may produce better and more aromatic wines.
The effects of soil types on the grapevine performance, wine style and quality were assessed for a Sauvignon blanc vineyard at a high altitude (508 m) with two Tukulu soils (high-gravel and low-gravel) over four years (2006 – 2010). The site was in Elgin (Overberg district, Western Cape) and cooler than most of South Africa’s traditional wine growing areas. The two Tukulu soils differed regarding gravel and water contents, water holding capacities and mineral contents. The root system in the low-gravel soil was regarded as better developed than that in the high-gravel soil. The P-levels, as well as bunch weight, were significantly higher for grapevines on the high-gravel soil than those on the low-gravel soil. The low-gravel soil produced wines with significantly higher acidity than the high-gravel soil, which was attributed to must total titratable acidity levels. Additionally, the high-gravel soil generally produced wines that were more aromatic and of better quality than the low-gravel soil, when wet conditions and warm summers prevailed. The study could not pinpoint the exact mechanisms grapevines respond to soils, but it creates a grower’s awareness of possible soil-effects on wine style and quality via grape composition and the extent thereof under specific climatic conditions.
Funding for this project was provided by the Agricultural Research Council (ARC) and Winetech. The authors wish to thank the staff of the Soil and Water Science Division of the ARC Infruitec-Nietvoorbij for technical assistance, D. Rowswell and I.O. Joubert from the ARC Institute of Soil, Climate and Water for climate data.
Jackson, D.I. & Lombard, P.B., 1993. Environmental and management practices affecting grape composition and wine quality – A review. American Journal of Enology and Viticulture 44, 409 – 430.
Seguin, G., 1986. ‘Terroirs’ and pedology of wine growing. Experientia 42, 861 – 873.
Saayman, D. & Kleynhans, P.H., 1978. The effect of soil type on wine quality. In: Proc. SASEV, October 1978, Stellenbosch, South Africa. pp. 105 – 119.
– For more information, contact Reckson Mulidzi at firstname.lastname@example.org.