wa2.gif (4241 bytes)

abut9.gif (3095 bytes)

abut12.gif (3207 bytes)
abut10.gif (3636 bytes)

abut11.gif (4039 bytes)


Grappling with the science of terroir  
Viña Casa Silva’s microterroir project

Soil profile descriptions

Chilean producer Viña Casa Silva is currently involved with a research project on terroir, in conjunction with Professor Yerko Moreno at the University of Talca. Their goal has been to identify small terroir units, which they have dubbed ‘microterroirs’, in their properties in the Colchagua Valley that are best suited to specific grape varieties.

Professor Yerko Moreno of the University of Talca

This project considers several different aspects of terroir, including climatic variables, soil properties, plant characteristics and also viticultural management strategies, with a view to uncovering special sites for Carmenère, Cabernet Sauvignon, Cabernet Sauvignon, Petit Verdot, Syrah and Viognier. In the past, educated guesswork based on weather data and soil profiles would have been the only way to decide which varieties to plant where. The problem with this approach is that each trial planting takes a number of years before any results are known, and replanting or grafting over to a different variety is an expensive process. With good scientific understanding of terroir characteristics, informed decisions can be made, shortening the time frame until and ideal match between site and variety is made. This sort of study also helps inform producers about the right management techniques to use with different varieties and soil types in order to get best results.

A ‘microterroir’ is defined as ‘a small sector (0.1–0.5 hectares) of a base terroir with special features of soil and plant growth, which when combined with viticultural management gives a wine of unique value.’

Casa Silva presented the example of their Los Lingues vineyard, where they have undertaken a range of studies, generating a significant volume of data which they have then tried to make sense of. 

Soil characteristics of selected Carmenere plots

What is the best methodology for collecting data on natural variation within a vineyard? One way might be to sample randomly throughout the vineyard and then look at the pattern of variation that is revealed. A more sophisticated method would be to do a coarse sampling, and then let these initial data points guide where subsequent samples should be taken from. This approach takes account of the fact that some portions of the vineyard will be more variable than others, and thus need more closely spaced data points to build up an accurate picture of the pattern of variation.

A third, and perhaps more effective strategy is to use precision viticulture approaches to split a large vineyard up into homogeneous management blocks which can then be studied in isolation and compared. Aerial imaging during the growing season, coupled with statistical analysis of the data that result, enable growers to see the pattern of natural variation reflected in vine vigour. Then it’s possible to split the vineyard up into management zones on the basis of what are presumably differences in terroir.

Casa Silva began by doing a series of detailed soil studies in which they generated soil profile descriptions that included measures such as sand content per soil horizon and water accumulation capacity.

Then they turned their attention to the plants, looking at factors such as canopy density, bunch (cluster) sun exposure level, evolution of berry temperature over 24 hours and pruning weights.

They have also looked at different responses to water stress, examining spatiotemporal variability of hydric potential across plots.

This was followed by commercial small lot winemaking of reds from different microterroirs in 500 litre French oak barrels. In the resulting wine, variables such as anthocyanin profiles, tartaric/malic acid and sensory wine descriptors were recorded.

This work generated a large volume of data which need to be made sense of through statistical analysis. For example, it is possible to look at how factors such as water deficit affect the wine style of each grape variety. Also, one of the big questions concerns year-to-year variability. ‘Terroir’ features, if they are to be useful in management decisions, need to be relatively stable across vintages.  


Initial conclusions from the study are interesting. The statistical analysis showed that wine composition variability is dependent on the following factors, and to the following degree:

  • Soil characteristics                      32%

  • Climatic factors                           25%

  • Cultivar (including management)     19%

  • Other factors not quantified           24%

The most important individual factors were:

  • January average temperature

  • Average temperature in the last month of maturity

  • Seasonal rain in spring

  • Root exploration depth

  • Soil volumetric moisture content

The best Carmenère is achieved in the blocks where the roots have to go deeper (at least 1 m) to get a water supply.

For Cabernet Sauvignon, vegetative equilibrium is critical. Shoots shorter than 80 cm lead to dry tannins; those longer than 130 cm lead to green characters and hard tannins.

The stage at which canopy growth ceases affects fruit composition significantly. Active canopy is needed until quite late in the season.  

These initial results show the potential value to producers of studies such as this. For Casa Silva, these complex research results have enabled them to produce a blueprint of the optimum elements required for a microterroir plot to produce top quality Carmenère, Cabernet Sauvignon, Syrah, Viognier or Petit Verdot. It has led to them reviewing some of their current management practices such as late irrigation and nitrogen nutrition. The blueprint has enabled them to re-map their vineyards and regraft those vines that were underperforming.  Winemaker Mario Geisse says, ‘The improvement in quality is already clear in the most recent vintages of our top wines and will become increasingly evident in the next few years.’

Viña Casa Silva Microterroir de los Lingues Carmenère 2005 Colchagua, Chile
14.5% alcohol. Deep coloured and dense, this has a classic Carmenère nose of brooding, sweet red fruit pastille and blackberry fruit with a spicy, chalky dimension. The palate is concentrated, smooth and quite lush with an appealing, smooth grainy tannic structure. Like many serious Carmenères it is very ripe and full, but is far from jammy, with grainy, chalky, spicy notes keeping the fruit really well defined. There’s a hint of dark chocolate, too, but the emphasis here is really on the bold fruit. 92/100 (UK retail c. £25)

See also: my series on Chilean wine

Wine tasted as 09/09  
Find these wines with

Back to top