Home > Winedr Blog

Terroir and Yeasts Revisited

Once I learnt of it, and really began to understand what it was telling me, the concept of terroir always made sense; in essence, two different sites will always give, even though the variety and the vinifications are the same, two different wines. So a Bourgueil from Les Malgagnes will taste different to one from Les Quartiers, even though both come from the same cellar and the same vigneron. That is terroir.

Defining what lies behind terroir, however, is fraught with controversy; understanding why one site gives different wines to another seems an impossibility. We have climate, soil, bedrock, aspect, drainage and more. Should we include the winemaker? Should we include the local microbiology, either that in the soil, on the grapes or in the winery? All have been mooted.

The thought that terroir might be a yeast effect is a tempting one. After all, as far as I am aware, few agricultural products show this ‘regionality’. Please correct me if I am wrong, but a peach from one orchard tastes much like a peach from an orchard down the road, provided the variety and agriculture is the same, but two wines from fruit grown just metres apart can be radically different. The key difference between the two products is that wine has undergone a microbiological transformation, a process not relevant to the world of peaches (unless you’re into home brewing I guess).

While there is no doubt different yeasts imbue wines with different characteristics (cultured yeasts are sold on this very basis – some types are ‘neutral’, while others produce more aromatic results) I have stated before I find the idea that terroir differences might be due solely to yeasts rather an unlikely one.

A recently published paper from Matthew Goddard and team from the University of Auckland School of Biological Sciences would appear to support my thoughts, even if the authors argue it in the other direction. The authors fermented many (over one hundred) sterilised samples of Sauvignon Blanc juice with genetically diverse isolates of Saccharomyces cerevisiae from six different regions of New Zealand. They then looked at an array of volatile compounds to see if they differed from one S. cerevisiae ferment to the next, and therefore from one region to the next.

Here’s the science bit, part one. Goddard et al found that when the juice was fermented with single strains, there was some difference between the aromatic profiles of the wine that resulted. But there are three important points here: (1) this isn’t surprising; we know different yeasts produce wines of different aromatic qualities, (2) the differences between the six regions when tested with single-strain ferments was only 10% down to the yeast, so even with a single-strain ferment the aromatic differences were 90% due to other (mostly unknown) factors, and (3) the aromatics differed from batch to batch – variation between batches accounted for 7% of the differences in levels of the volatile aromatic compounds (nearly as much as the yeast-effect, which was just 10%).

With the yeast effect hardly stronger than batch variation, is it really a plausible candidate for the cause of terroir?

And here’s the science bit, part two. Because part one is not applicable in the ‘real world’ (because wild ferments involve many different yeasts all working at the same time – they are not single-strain ferments) the team also did co-ferments, with not one strain from each region, but six single-region strains mixed together. With these fermentations, there was no statistically significant relationship between the region from which the strains came and the aromatic profile of the wines. Now this might just be a problem with sample size – perhaps running the test again with several hundred more samples would solve this (yes I know that is easier said than done – this paper reflects a lot of hard work).

Even so, for the moment it appears to me that any regional ‘yeast effect’ is identifiable but small with a single strain, but this appears to be lost in the mix once you have a body of yeasts working together, as in the ‘real world’.

Perhaps yeasts do contribute something to terroir. I am open-minded on the matter, and await some convincing research to persuade me one way or the other. But even if they do contribute, this research suggests yeasts play a minor role, which would therefore seem to indicate that the traditional view of terroir as being related to the physical properties of the site still holds true. Or at least more true than it does for yeast.

Minerality: Gneiss-y Nonsense

There has been a lot of words written on minerality over the past few years; what it means, where it comes from, is it real, and so on. I think the terms minerally and minerality are useful descriptors for wine; I accept, however, that it can be difficult knowing exactly what someone (including me) means when we use one of these terms in a tasting note. I wanted to share some of my own thoughts on minerality over the next few days. In particular I will try and explain my own take on the term, although I will warn you now I’m not going to be as precise as you might be hoping.

Before I get to that though, I first wanted to consider the notion that minerality in wine reflects minerality in the soil.

The major problem with minerality is that some may have assumed that its presence in a wine reflects a direct relationship between the wine and/or fruit from which it was made, and the soil or rock in which the vines are planted. In other words, minerals from the ground somehow find their way into the grapes, and then to the wine. This has always seemed like nonsense to me. We probably all know that plants absorb water, nutrients and minerals through their roots; this is why the health and growth of plants can be influenced by adding fertiliser to the soil. The fertiliser contains nutrients and minerals valuable to the plant, which can be absorbed across the huge surface area offered by its extensive network of fine, branching roots. It is important to understand exactly what we mean by ‘minerals’ though. It is tempting to think they are rock minerals, a molecule of gneiss here, a granite molecule there. That’s always seemed like nonsense to me, for several reasons. Take gneiss – a commonly encountered metamorphic rock in the Muscadet region – as an example; this is made from a variety of minerals, but is likely to be a mix of the following:

  Feldspar, chemical formula : (Ca,Na)AlSi3O8 or KAlSi3O8

  Quartz, chemical formula : SiO2

  Muscovite, chemical formula : KAl2(AlSi3O10)(F,OH)2 or (KF)2(Al2O3)3(SiO2)6(H2O)

  Biotite, chemical formula : K(Mg,Fe)3(AlSi3O10)(OH)2

Gneiss, as an aside, is not defined by a specific mineral composition (so it can be a mix of the above, and other minerals too) but by a texture; it is a metamorphic rock, often granite-derived in the Muscadet region I believe, where the minerals are banded into dark and light strands a result of extreme heat and pressure.

No, I haven’t surreptitiously pulled out a geology degree you never knew about; I simply pulled these formulae off Wikipedia, which only seems fair as they have been mining Winedoctor for wine-related facts ever since a wiki author decided that Latour and Le Pin warranted entries just as much as Miley Cyrus and all the episodes of South Park and The Simpsons ever made did.

Looking at the above minerally formulae, several things spring to mind:

  1. Gneiss is made up of some huge molecules; it would be very difficult (impossible I think – but I’m not a plant scientist and I’m trying to remain open and balanced) to transport them across a cell membrane.

  2. There looks as though there is a lot of potential toxicity there – do plants really want to be hoovering up Aluminium-containing minerals?

  3. Active transport into the cell requires energy – why would the plant use energy to absorb such huge molecules, when it is the constituents (iron, manganese, etc.) that might be useful. Why not just transport in iron and manganese ions from the surrounding soil?

  4. Transport into the cell requires the nutrient or ion to be water soluble; any gardener knows this. In high-pH (alkaline) soils plants suffer from chlorosis. This disease was a big problem for those planting experimental grafted vines in the early years after phylloxera, as some rootstocks when planted on limestone couldn’t handle the conditions. In such conditions much-needed iron is sequestered as a solid, bound with calcium. Solids cannot be absorbed by roots, the ‘minerals’ need to be dissolved. There are two solutions; alter the pH, for a long term solution, or add chelated iron (iron in soluble compounds) for a quick fix. It is impossible to imagine the molecules of stone, described above, dissolving in water and being absorbed. Gneiss, and other rocks, just aren’t water-soluble. They erode, yes, but they don’t dissolve into water like sugar.

  5. Discount all of the above; assume gneiss is soluble, and magically taken up by the roots. How and why would this gigantic minerally molecule be ascended through the structure of the plant to be deposited in the grapes. Would the normal xylem channels achieve this? What would be the purpose of this energy-expensive process?

It all seems like gneiss-y nonsense to me, and seems to discount both the notions that ‘minerality’ originates with the absorption of minerals, and also – on a related issue – that a particular terroir might be expressed in a wine through the absorption of the very minerals that comprise that terroir.

More minerally thoughts on another day.