I think I was possibly the first to publicly raise and discuss the ‘vintage character’ of 2004 reds – anecdotally whites are also affected – Christophe Roumier is one who says this – but I’ve not experienced any obvious examples, so from here-on, red it is…
[Edit: Whites – Now I have]
Early on, I’d empirically estimated that about 30% of cuvées had a strange aromatic profile, elevated levels of which also affected the taste. I assumed a large but limited extent to (let’s call it for now…) the ‘taint’ and one which would hopefully subside, but an October 2008 lunch with fellow enthusiasts that focused on 2004’s left me questioning that perspective.
It was only by chance that I mentioned my experience of grape-baskets often with dozens of ladybirds (ladybugs or coccinella depending on your location) at harvest-time, but quick as a flash, Don Cornwell found a potential link – pyrazines – I’ll come back to those later.
But first, let us take a step back. As the wines matured in their barrels, the 2004 reds were better than many had expected – much rot had to be triaged (I was there), but the result seemed worth it. The wines showed ripe fruit coupled to a nice freshness, and many looked a good buy after the somewhat bizarre 2003’s. At en-primeur tastings all was fine, and for the first two or three months in bottle they repaid my confidence, but less than 6 months after the bottling there was an obvious ‘odour’ problem. Today there are some wine-makers (David Croix for instance) that say they spotted a disturbing ‘greenish’ element early in the elevage, but that it quickly disappeared – maybe it was only Lalou Bize-Leroy who truly spotted it, and THAT was the reason she blended all her wines together…
[Edit: See document #4 at the end of the page – from the conclusions I note: “This result, and the limited performance of deodourized oak here, suggest that the oak chips are masking the taint with oak-derived volatile components, such as vanillin (Perez-Coello et al., 2000) and cis oak lactone (Sauvageot & Feuillat, 1999).” If volatile oak components do have a masking effect on pyrazines, this might explain why the taint only came to the fore a few months after bottling, as those volatile oak components started to fade. Only conjecture of-course. It might also explain why some found a nasty green aroma early in the elevage, before it was coverd (masked) by barrel extracts….]
This ‘character’ of the vintage.
The closest for me was a peculiarly old English thing – the smell of Wright’s Coal Tar Soap – an almost mineral, chemical, carbolic smell; from the ‘natural world’ I would describe it as similar to a cedar/sandalwood mix. Others say ‘mirepoix’ or very simply say ‘green’ which is an unfortunate use of the word/concept – even though I slip into using the term myself – because traditionally when discussing red wines, ‘green’ implies just about any element that might indicate ‘unripe’, and 2004 wines are far from unripe. I have come across truly green (unripe) 2004’s – but in this case my descriptor would be pine-needles and it is not the ‘character’ we focus on here. The latest description comes mainly from french tasters who describe the aroma of gentian flowers – it sounds lovely, but you have been warned!
Most lists of ‘green descriptors‘ don’t describe very well the 2004 character. The only other vintage I consistently noted a similar descriptor was young 1998’s but it was very mild; given their astringent sand-paper tannins a number of people said that ‘green’ came from unripe tannins. I no longer find it in the typical 1998.
Who is affected?
Close to everyone and no-one depending on your sensitivity. Bottles opened by me between September 2006 and June 2007 indicated about a 30% level of ‘infection’. Later tastings, but smaller samples, had a majority showing the effect. The issue was that in the early months after bottling, the producer was not an indicator – find two cuvées whose elevage were side-by-side and in many cases only one wine would show the taint. You could find it on a low level in the wines of Domaine de la Romanée-Conti, but here’s the interesting thing; at low levels, it’s just another small note of complexity and it’s actually rather nice – e.g. Fourrier’s Morey Clos Solon tasted in October 2008 – but increase the concentration and it’s nasty. A much bigger concern to me is that some bottles I opened 2 years ago, bottles that were fine, now show the ‘issue’ so the ‘infection rate’ certainly hasn’t ‘peaked’ yet. Some people, as ever, are significantly more sensitive to it – I seem to be able to pick it up at very low levels. For those of low sensitivity, this remains a relative bargain of a vintage.
What is it?
Well chemically, a number of more technically oriented wine-makers have told me the smell is of pyrazines, most likely methoxy-pyrazines (MP’s). If so, that doesn’t bode well for a slow reduction over time – pyrazine odours can often get worse with time. Pyrazines also brings us back to our cute little beetle friends. The linked report in the third paragraph (above) shows that coccinella use methoxy-pyrazines as sexual attractants, they are also used as a defence alert – pick one up and the yellow colour that they leach onto your fingers also contains pyrazines.
Alkoxypyrazines; ‘alkoxy’ implies any of thousands of pyrazine combinations with methyl (methoxy), ethyl (ethoxy) etc., etc.. This group of naturally ocurring compounds have long been found in grape-juice and wine; some varieties like Savignon Blanc derive a portion of their characteristic aroma profile (e.g. when in New Zealand) from these strongly smelling compounds, whereas with Cabernet Sauvignon, the compounds show ‘bell-pepper’ aromas that people associate with cool climate wines and less ripe fruit.
This class of compounds can be detected by the humble human nose at incredibly low concentrations, not unlike those other unwelcome guests in your wine-glass, TCA and Geosmin. Some of us are sensitive at levels of parts per trillion – that’s much more sensitive than you are to cork taint – TCA. Although the human nose can detect these at extremely low levels, there is considerable variation not only from person to person, but also the same person depending on the time of day.
In his essay ‘Drinking in Minerality‘, writer Jeff Cox thinks that pyrazines may be partly what we percieve as ‘minerality’, that said, the following quote is most relevant to our discussion:
Human beings can detect pyrazines in vanishingly small amounts, such as parts per trillion. “One drop in the Reflecting Pool in Washington, D.C., and you’d smell bell peppers all the way to the Beltway. [So] minerality could be really low levels of pyrazines.” He points out that “grassiness” in Sauvignon Blancs is caused by pyrazines.
So is there any good news? Not yet, it seems that even in the bottle (assuming that MP’s are our problem) MP’s seem stable; “Aging in a dark cellar for 3 years did not change the concentration of MP in the wine.” It seems we know quite a lot about the development of MP’s in the vineyard/grapes, but less about them once they are bottled.
How did it get there?
A link could be the Coccinella. Some people pooh-pooh the idea of ‘ladybugs’ being the source – ‘you expect me to believe that every vineyard was infested with these things?‘ Actually no, though it can happen. The vineyard is in many respects a red herring; 2004 was a year with an over-abundance of coccinella, the following year the cuveries were full of fruit flies, the year after that the grapes were full of earwigs. In 2008 there grapes were almost fauna-free as they were so cold! There are natural cycles and some insects dominate for a year and are seen less the following year. It’s not about how many coccinella were in the vineyard, it’s about how many were in the cap of the fermenters or on the triage table (left) – remember 1 part per trillion is probably 0.01 of a coccinella in a large fermenting tank!
That coccinelle can taint a wine is empirically demonstrated by a number of authorities; A department of the Swiss government put out a press communqué in August 2007 alerting vineyard owners to the arrival of the asian version in some vineyards, indicating that the beetles could affect the taste of the wines. They further did some follow-up vinification work on both Pinot Noir and Chasselas (Chasselas is a popular Swiss white variety) with these Asian Coccinelle in the mix – both wine varieties were equally affected. If you are interested to read the report in German, it is here:
“In 2007, they collected adult beetles in fields of sunflowers. Then they made two tests. On the one hand, they threw ladybugs live in pots containing crushed-stemmed grapes of pinot noir. On the other they have pressed chasselas with the small animals. After a standard microvinification they put the wine bottle in March 2008 and they have tasted in May. The addition of the beetles does not affect the basic composition of the wine: pH, acidity and ethanol content are unchanged. However, they are very clearly the cause of bad tastes.
The tasters noted an imbalance to the Chasselas wine and a strong smell of rancid oil. The finesse and fruitiness were also heavily penalized. The addition of five beetles per kg grapes to Pinot Noir equally penalised the red wine. It affects the quality of tannins, the finesse, fruit and the structure of the wine. Researchers estimate that one can see the bad taste of 0.3 Asian Coccinella per grape cluster for Chasselas and 0.2 for Pinot Noir. This corresponds to 18% (chasselas) and 12% (pinot noir) clusters settled by at least one beetle.”
That the taint was visible so quickly in the Swiss tests as opposed to the months it took to develop in burgundy’s 2004’s could be down to either (or both) of 2 things:
- they added a lot of ladybirds to get the effect i.e. much more than the Côte d’Or experienced. 12% of the clusters also sounds quite high – I’d ‘guesstimate’ that it was more like 3-5% on our sorting table.
- or that the population of 2004 could have been the smaller european version that maybe is less ‘potent’ – I have photos of both, but I’d need an expert to judge
Of-course there’s a third possibility – Coccinelle are not involved in the 2004 ‘vintage character’ at all…
This is conjecture. I’ve ‘hypothesised’ Coccinella as a plausible reason for the malaise that affects so many wines from the 2004 vintage; many winemakers accept that possibility whilst others remain unsure. I fully accept that Coccinelle may not be THE reason – but thus-far no-one can tell you otherwise and there is certainly no alternate and ‘viable’ theory yet proposed. I expect that no wineries will be sponsoring research to find a ‘root-cause’ as they have to concentrate on selling their 2007’s.
Other than to ‘keep my hand in’ I’ve all-but stopped opening 2004’s as I don’t like the ‘taint’ whatever its source, so the vast majority of my bottles will stay in the cellar a good few more years to see whether MP’s really do diminish with time – and that’s a shame as the vintage had much early charm.
I’ll end with a list of 2004’s that I’ve tasted, and I’ll leave you to work out the code. The list has a large measure of chronology about it, making the colour-change an interesting but depressing story…