Sunday 15 December 2013

Angleterroir

John Atkinson MW has a particular interest in terroir and soil composition, as you can read in his Terroir and the Côte de Nuits. Now he has planted his own vineyard (pictured) in the village of Tixover, Rutland, in the middle of England. Here he explains what happened when he had its soil analysed.
 
 
Superficially, our soil looks like some of the better bits of the Côte de Nuits. The soil is composed of clay (25%), limestone gravel (25%), silt (30%) and sand (20%). In the shallower parts of grand cru Romanée-St-Vivant you hit Bajocian bedrock at a depth of 30 cm, and it's the same for Tixover. Shuffling around on your haunches you might imagine yourself to be in Vosne. The first time I tested our soil, I also submitted samples gathered from across the Côte de Nuits. The results for Romanée-Conti confirmed it as 'a good agricultural soil', with macronutrients falling within the 2-3 range of standard farming indices; endorsing its potential as a wheat field should the wine business ever go to pot. The ratio of potassium to magnesium in Le Chambertin and Clos de la Roche was skewed towards potassium, and the pH for all three sites was between 7.8 and 8.3. The results for the Tixover sample closely resembled those of the grands crus. All four had excessive calcium levels.
The second soil test was analysed in France by AGRO-Systèmes. It introduced me to new concepts such as cation exchange capacity (CEC) and base-saturation. Our soil was high in nitrogen and calcium, but weak in iron and magnesium. The ratio between calcium and magnesium influences soil structure, and soils that are base-saturated with calcium tend to be more porous than soils dominated by magnesium. A high cacium to magnesium ratio generally facilitates water movements within the soil - up, down and sideways.
The most recent test, conducted by Albrecht Agriculture, was also the most thorough, and the results showed worrying discrepancies from the previous two analyses. Iron was recorded at a critically low level, and the CEC was measured at 30 meq, which is high for a soil that is only 25% clay. The 'clay and colloidal matter' bound calcium and magnesium ions in the ratio of 40:1. Our soil, Albrecht warned us, was like a sieve. We needed to add magnesium, lots of magnesium. But at least we were in good company: the three grands crus soils were similarly biased in favour of calcium base-saturation.
Confusingly, iron levels were only measured at 1 ppm by the Albrecht people, but at 15 ppm by the French analysts. We haven't seen signs of chlorosis in the vines, so experience suggests the French analysis is the more accurate of the two. Moreover, despite identifying the chronically deficient iron levels, the Albrecht analysis didn't propose any amendments other than the magnesium. Critics of scientific methodology warn about theory-laden data, and the Albrecht obsession with the calcium:magnesium ratio may explain why in this instance they missed the low iron count.
CEC is a measure of the nutrient holding capacity of the soil, and relates principally to the number of internal and external bonding sites within the sample's clay and colloidal fraction. The meq measurement was elevated in our sample by the mix of clays. Smectite, a volcanic mineral, was incorporated into our clay as it sedimented out, and its inclusion has significantly changed the soil's properties. For agronomist Claude Bourguignon, smectite/montmorillonite clays, with their large internal surface areas, are particularly advantageous for red-wine production. Just why these clays are so interesting to M Bourguignon is worth consideration.
According to Bourguignon, the soils of Pétrus together with most of Vosne's grands crus (Romanée-Conti may be the exception) are luxuriously endowed with smectite. By contrast, the illite-kalolinite clays of the white-wine grands crus have much smaller surface areas, and therefore their capacity to hold and exchange nutrients is considerably lower. Montmorillonite clays are also distinguishable from other clays by the extent to which they expand when wetted. Modest water deficits are seen as advantageous in wine production as they accelerate ripening and generally improve the quality and quantity of extractable solids from the skins, so a clay that absorbs and holds onto copious amounts of water would seem anomalous to the requirements of premium grape production. Montmorillonite clays do have a trick, however. Strong root growth and function require a good level of root oxygenation. The expansion of montmorillonite clays can be so dramatic that root growth and function become impaired. Moreover, the permeability of the clays can decrease due to sealing at their surfaces after wetting. Thus soils may look wet, puddled even, but this moisture is not necessarily available to the roots; a case of water, water everywhere and not a drop to drink. And just as they expand when wetted, so montmorillonite clays shrink when dried. A sustained period of drought will open up capillaries within the soil which then become exploited by the vines' constantly regenerating mesh of short-lived rootlets. Taking into account the close connection between vacillations in clay particle size and Bordeaux's changeable seasons, it is possible to model a shallow, smectite-rich soil in which the availability of moisture to the vine roots is almost continually held at deficit levels - a terroir very much like that of Petrus, in fact.
Beneath the thin clay loams of Vosne's grands crus is limestone. The ability of vine roots to populate hard rock is limited, but this fact hasn't stopped commentators from positing this hidden union as the very foundation of terroir. However, a more probable explanation of limestone's advantage to viticulture is revealed by a consideration of its physical properties. Limestone can hold large volumes of water, but, as was discussed in the context of clays, we must be careful not to confuse capacity with availability. Movement of water through soils is multidirectional, and limestone can irrigate the clays resting upon it through capillary action. The extent to which the capillaries are opened will in turn depend upon the relative expansion of the clays and the degree to which calcium base-saturation has flocculated these, the most minuscule of soil particles, into larger agglomerations. In other words, the limestone's ability to function as an aquifer is itself dependent upon the structure of the overlying soil. As was the case with Pétrus above, we can imagine a possible balance between local effects whose equilibrium point sustains the vines at a slight and ultimately advantageous level of water deficit.
On the Côte de Nuits the mercurial temperament of Pinot Noir brings colour and texture to the bland realms of geology and pedology. In her excellent articles in the Australian & New Zealand Wine Industry Journal on 'Calcium in viticulture', Valerie Saxton talks of 'terroir-ridden France', and Burgundy can seem like yet another scion of the republic's obsession with taxonomy and hierarchy. So completely pixelated has the map of Côte d'Or's vineyards become, that it's now virtually impossible for outsiders to see the whole picture. Twenty years ago, many New World growers censured the Côte d'Or. It just didn't make sense: grands crus a stone's throw from modest village vineyards; and Clos Vougeot, the oversized Circus Maximus at the centre of this antiquated world, just seemed to lump together everything that the rest of the classification had painstakingly tried to keep apart.
There are historical precedents for Burgundy's sub-divisions, but the acute sensitivity of Pinot Noir coupled to the equally acute sensibilities of those that tend to it is the more remarkable story of this segregation. One way of classifying grape varieties is through their differing responses to water deficits. Anisohydric vines are said to be drought tolerant, whilst isohydric varieties are drought avoiding. The divide is not clear-cut, and relates to the strength of response different varieties show towards hormonal signalling (abscisic acid) from the drying roots. The stomata of isohydric varieties progressively close, rationing water uptake and loss, while anisohydric varieties maintain stomatal turgor, such that gas exchange, water uptake and carbon gain are unimpeded. Isohydric vines are 'pessimistic' and anisohydric vines 'optimistic', inasmuch as the latter carry on as if they expect it to rain again tomorrow.
Some varieties seem capable of both responses, but in extremis, Pinot Noir is anisohydric, and Grenache Noir isohydric. Lavish water use, as exemplified by Pinot Noir, brings with it a vulnerability to sustained drought. Pinot Noir, by rapidly depleting soil moisture, can accelerate itself towards conditions under which its own metabolic processes become compromised, just like the man who saws furiously at the branch he's resting on. Without the self-buffering responses of isohydric varieties, Pinot Noir's own water status is so immediately bound to the vicissitudes of soil moisture that it takes, in unirrigated Burgundy, a very special mix of extraneous pedological factors to consistently produce high quality grapes, ie those that are the result of sustained, but not impairing, water deficits. Vosne's grands crus are, indeed, exceptional.
Claude Bourguignon caused me to depart from thoughts of my own soil analyses and on-going struggle to get consistent data. There have been reasons for optimism in most of the results, and I don't doubt the fact that if Tixover's three blocks of brashy soil were panelled into Morey or Chambolle you wouldn't see the joins, but it's ridiculous to start talking about terroir, particularly when we seem to have spent the last few years acquainting ourselves with a whole load of Nature's disadvantages. I remember with incredulity, a vineyard owner in Long Island exploiting the argument that Bordeaux's crus were the finest in the world, that Bordeaux's vineyards were flat and, as his vineyards were flat, too, it followed that his wines were rivals to those of the Médoc.
As said, there was a time when the majority of New World growers rejected terroir, portraying it as either a perspicacious piece of marketing, or homey make-believe. But somewhere along the track these protagonists either gave up on this line of attack or lost the argument, because producers in Chile, New Zealand and Mendoza nowadays reel-off neat invocations about the contingency of their own vineyard work that wouldn't sound out of place in Vosne. Wine now seems so exclusively 'made in the vineyard', it's hard to know if there is anything left for winemakers to do. In the New World, terroir has gone from being nowhere to being everywhere.
The ubiquity of terroir is problematic. I take from the late Peter A Sichel's remark that only a fraction of Bordeaux's vineyards genuinely exhibit terroir characteristics, that terroir is a title bestowed like an honour or a peerage, rather than a democratic entitlement to which anyone with a vine growing up a wall has an equal claim. Accordingly, Richebourg and Pétrus are the exemplars of terroir, the foundations and acmes of a system in which exceptionality and scarcity are paradigmatic. In this context, the work of Claude Bourguignon, Gerard Seguin and Cornelius van Leeuwen is so pertinent because they are trying, in different ways, to cleave some scientific traction into our understanding of a much abused term. For those who crave more metaphysical accounts of grape quality, who will never be moved by terms like 'vine water status' or 'point quadrant analysis', there is always Nicolas Joly of Coulée de Serrant. But this is to miss the point. When physicists revealed diamonds' atomic structure, they didn't stop them from being a girl's best friend.
I do share Peter Sichel's instinct for parsimony, not least because the over-extension of any term, terroir included, eventually runs the risk of draining all significance from it. The meaning of words changes over time, and despite van Leeuwen's attempts to give terroir intellectual rigour, I feel sure that the term's appropriators will win out. 'Sense of place' will read like a postcode.
More optimistically, the weakening of terroir is not going to stop some very clever people from giving us ever more concise formulations of past viticultural accomplishments, those for whom 'sense of place' and 'terroir' prompt the question: 'And how did we get here?' As for the appropriators, I suspect every nascent English sparkling wine venture will claim its unique terroir, as will every newly planted desert of irrigated Sauvignon. Bertrand Russell once observed, 'The method of "postulating" what we want has many advantages; they are the same as the advantages of theft over honest toil.'

1 comment:

  1. John,

    Seven years back, when you wrote this, did you obtain calcaire actif assays from the various labs you mention? How similar were they? I've had a dozen samples analysed from six sites across our three fields, based on their geological heterogeneity (see my posting & discussion on 'Soil2Cellar"), and I think the lab ( Lancrop) have mistakenly reported the data, as a %, but using a soil weight denominator, rather than Total Calcium. I've ostensibly pH values of 7.6 - 8.3, with active calcium carbonate values in range of just <1 - 4%, with lower values for both pH and active calcium as expected for humus organic carbon-rich topsoil.

    Reworking the Lancrop data, using Total Calcium (% in soil, wt/wt) in the denominator, and their original Active Calcium Carbonate % data in the numerator, I get values closer to expectation for a limestone (White Lias-derived) soil,, ie up to 20%, but not as much as a the 30-50% one might anticipate for marl or chalk.

    Stephen Skelton has kindly shared some Lancrop analyses form a couple of the vineyards he planted 8-10 years back, and (on chalk) they return active calcium carbonate data <50%.

    I can only conclude that given the infrequency of this assay, Lancrop have simply forgotten how to frame the data.

    Incidentally they report great CEC values and checking those UK clays data you usefully posted on Twitter in 2018, some of the 'Rhaetic Shales' and 'Lower Lias' map references are on similar geology, just down the road... and Smectitic.

    Comments most welcome,

    Regards,
    Chris Poole

    ReplyDelete