Botrytis and Oxygen: An Interview With Sandrine Garbay of Château d’Yquem

Dessert wines, particularly those affected by noble rot (botrytis cinerea), are in a category of their own. To understand more about them in terms of oxygen management, we talked to Sandrine Garbay, the cellar master at Château d’Yquem, the most famous of all producers of Sauternes—and possibly the most famous botrytized wine in the world. Garbay, who holds a Ph.D. in oenology and has been working at Yquem for over 20 years, tells us about the aromatic character of these wines and the specific requirements for vinifying and aging them.


© Gerard Uferas Ch‰teau d'Yquem 16_10_12Can you tell us about the influence of botrytis and of relatively late harvest on the way aromatic thiols express themselves in wines like Yquem?

Volatile thiols are a dominant aromatic family in Sauvignon Blanc wines, as studies by DARRIET and TOMINAGA have clearly demonstrated. When botrytis cinerea starts growing on Sauvignon Blanc grapes, it concentrates the volatile thiol precursors in the must. This is likely a reaction to the stress caused to the berries by botrytis. These thiols are released by yeast during alcoholic fermentation, and there four to five times more volatile thiols in Sauvignon Blanc dessert wines than in dry wines. Even better, in 2007, Philippe Darriet’s team identified three new thiols specific to botrytized wines.

What about Sémillon? Are its aromatic components similar to Sauvignon? And do they react the same way?

They are the same, but to a lesser extent. The nature of the thiols is the same, the intensifying effect of botrytis on the thiols is the same, but the amount of thiols is a little lower.

Does the presence of botrytis, in and of itself, change the relationship of wines like Yquem with oxygen in terms of reactivity, aromas, etc.? Are they more resistant because of their concentration?

The oxygen-related processes at work in botrytized wines are actually fairly easy to understand. In the case of a dry Sauvignon Blanc white, for instance, the few phenolic compounds present in the wine react fairly quickly with oxygen and oxidate into quinones, which in turn react with the volatile thiols in the wine and break them down into much less aromatic compounds. So the aroma of dry Sauvignon Blancs is very sensitive to oxidation. In the case of a botrytized wine, however, phenolic compounds are completely oxidized in the must by the laccase produced by botrytis cinerea. Quinones that have developed polymerize and precipitate, so they can’t react with the wine’s volatile thiols anymore. So, not only do botrytized wines hold more thiols, but the thiols are also more restistant to oxidation, whether in Sémillon or Sauvignon Blanc.

In short, everything that could be oxidized in botrytized grapes is effectively oxidized by botrytis enzymes. Wines made from these hyperoxidized musts are more stable to oxidation than any other wines, which explains their incredible aging potential—although it is true that the high level of residual sugar also helps on that front.

Drawing the precious juices from botrytized grapes is a very slow process, which must inevitably cause a lot of oxygen exposure at the press. Can there be too much oxygen at this stage, or is oxygen rather beneficial to the aromatic expression?

The very slow pressing of botrytized grapes (which is necessary to get a good extraction of juice and sugar) and the high level of oxygen that comes in contact with them as a result have no effect, whether positive or negative, on the aromatic quality of must or wine. This is shown by the fact that at Yquem, we have not been adding any SO2 to the musts for over 20 years. There’s no advantage in terms of protection from oxidation, and at the microbiological level (for selecting the fermenting yeasts), there is almost no interest either, since these yeasts are already highly present on botrytized grapes.

During fermentation, do the wines require either protection from or exposure to oxygen?

During alcoholic fermentation, as for any other vinification, there is a need to give the yeasts oxygen so that they develop properly, and that they can be enriched with sterols that will help them resist the alcohol they are producing. That ensures that the alcoholic fermentation will end properly. This oxygen input should take place during the first third of the alcoholic fermentation, on a single occasion, and that’s usually enough.

Thiols_4What about barrel aging? Is it best to protect or expose the wine to oxygen? How is aromatic development affected?

My answer to this question is going to be a little more empirical. During the first year of aging, the wines always seem to react very little to oxidation. We age them on whole lees for one month, which probably adds to their glutathione content, a component that protects thiols. Then, over the course of that first year, we perform two classic, barrel-to-barrel rackings, which means the wines get oxygenated. It seems to us that, at that stage, a bit of oxygen helps the wine’s typicity and aromatic expression. After blending, which is done after 10 months in barrel, we protect the wines from oxygen during the rackings, to preserve the aromatic potential that has been developed during the first year of aging.

Recent work by Stamatopoulos (2013) has shown that botrytized wines express their “confit” character (that orange marmalade note) thanks to barrel agine. A certain type of lactone (2-nonen-4-olide) that is synthesized through the development of botrytis on the berries, combined with the wood lactones, increases the presence of that “candied orange” note, through the way various elements come to interact—which in turn influences the way they are perceived.

Is there a risk of reduction with dessert or botrytized wines?

There is practically no risk of reduction with dessert wines. It only happened with a single barrel at Yquem over 21 vintages, and after a few rackings, the reductive character was gone.

How do you manage sulfur levels in dessert wines?

Sulfites have to be rationally managed. They’re mainly needed to provide microbiological stability in the wines. Taking into account the wine’s pH, you have to aim for an active sulfur level of about 0.5 mg/L. Since botrytized wines combine huge amounts of SO2, this often means relatively high total SO2 levels. That being said, over 20 years, we’ve cut the amount of SO2 by almost 40 percent by better managing hygiene in the barrels and material and by better oxygen management during aging and at bottling.

Is volatile acidity a greater risk in dessert wines than in others? And if so, how should oxygen ingress be managed—or other things like sulfite levels—in that context?

It’s true that dessert wines made with noble rot have much higher volatile acidity levels than their dry counterparts. These high levels are tied to their sugar content: when yeasts are in a very high osmotic pressure, at the beginning of alcoholic fermentation, they react by accumulating ethanal, which is then turned into acetic acid. This phenomenon is not related to oxygen, since its origin is not bacterial. On the other hand, it is indeed important to limit oxygen ingress during aging and to control sulfite levels to avoid any formation of acetic bacteria, which could then add even more volatile acidity to the wines.

© Gerard UferasChateau D'Yquem  juin 2012You’ve worked a lot on “mapping” the whole range of aromatics in Yquem wines. Can you tell us more about this, especially in terms of reactivity to oxygen?

We have worked a lot on creating an aromatic cartography of our wines, but not as specifically on the varieties—or I should say that we’ve found the same type of molecules in both varieties at play in our wines, Sémillon and Sauvignon Blanc.

As mentioned, volatile thiols (citrus peel and passionfruit notes) but also lactones (candied fruit notes) are dominant elements in the specific aromatic expressions of dessert wines. Furanones (caramel notes) and aldehydes (honeyed notes) also have a part to play.

It’s been shown that botrytized wines have an uncommon aromatic intensity, thanks to a much higher than average amount of aromatic molecules in the wines, but also to very specific aromatic characters.

About the Author

Rémy Charest is a Quebec City-based journalist, writer, and translator. He has been writing about wine and food since 1997 in various Canadian and American print and online publications, includijng Le Devoir, Le Soleil, Coup de Pouce, EnRoute, Palate Press, Punch Drink, WineAlign and Châtelaine, and has been a regular radio columnist for Montreal's CJAD and CBC/Radio-Canada. He is additionally a wine judge for national and international wine competitions, notably the National Wine Awards of Canada and the World Wine Awards of Canada organized by WineAlign.

Rémy Charest Photo Credit: Jason Dziver


    1. Those would be a very different category of wines, indeed. The grapes aren’t botrytized, normally, and the process (through alcohol addition, the “mutage”) is a whole different ball game.

  1. Wow. Good job Remy!
    Like most winemakers, I have never made a botrytized wine (except having to deal with it on Pinot noir), but have been curious about how it is made. Having had the opportunity to drink well-aged classic Sauternes peaked the curiosity.
    This is the quality of info I value highly, the nature of which was the basis for a book I wrote titled, “Creating World Class Red Wine”

Leave a Reply