WBM November 2006
From the tasting bench

Our knowledge of Brett has come a long way in a short time, but there’s still plenty we don’t understand.

Dekkera (Brettanomyces) bruxellensis is a fermenting yeast present in the bloom of grapes and on – and in – surfaces in wineries. AWRI research has found six races of Brett in various regions around Australia (Curtin et al, 2005). It’s a weak fermenter, but can nonetheless utilise low concentrations of sugars left after primary fermentation. It can produce a wide range of compounds which affect a wine’s smell and taste, the most familiar being 4-ethyl phenol and 4-ethyl guaiacol which are often present in a ratio of roughly 10 to 1. 4-EP has a smell described as Bandaid®, horse dung or medicinal, while 4-EG is more pleasant, smelling of cloves or wood-smoke. Both impart a very gritty, metallic hardness to tannins in red wines.

This much we know.

We don’t yet know much about the relative importance of Brett on fruit and Brett in the winery. We suspect the latter is more critical, but Brett is sometimes more common in particular vineyards and in some years, 2001 in the Hunter, for instance. Some wines appear badly affected on nose, but show little of the tannin effect; others are the reverse. This is possibly related to the proportions of 4-EP and 4-EG, and possibly other compounds like 4-ethyl catechol, but we don’t yet know.

The main issue with Brett in finished wine is that our sensitivities to it vary greatly. In fact, it often seems to inspire a competition – “I can spot lower levels of Brett than you can!” One person’s complexity is another one’s pong, which makes it difficult to set standards for assessing quality. And remember that most barrel-aged reds contain at least some of 4-EP and 4-EG. I find that, at close to the subliminal level, a hint of earthiness from Brett can be positive and add complexity, but once it begins to smell horsey or the tannin balance gets out of whack, you’ll lose me. But is there an inconsistency here? We accept moderate levels of VA in sweet whites and even encourage funky, sulfidic characters in barrel-fermented Chardonnay and yet, if you’re a member of the Brett police, the merest trace of 4-EP is enough to condemn a wine.

I’m only moderately sensitive to Brett myself, so it may sound contradictory when I say I sympathise with the Brett polices’ view. At this stage, when we have much to learn about Brett, zero tolerance is a pretty good idea, at least during wine production, particularly as we know that Brett can grow in bottle. A hint of Brett may bring welcome complexity, but how can you safely restrict it to a hint?

This brings us to a more market-related issue. Some consumers seem to like obvious Brett characters, or at least are unable to recognise them for what they are. Otherwise, why would many Bordeaux and Rhône reds be so popular? But Brett is erratic – obvious one year, not detectable the next. So, if you are trying to build consistent clientele who will seek out your wines for qualities they recognise, do you want to mess with inconsistency?

The biggest issue with Brett I’ve seen in my work is with winemakers, particularly in cool regions, trying to get wines through MLF in barrel. Reds staying on low SO2 levels and at temperatures too cool for a reliable MLF are an open invitation to Brett, particularly as they warm in spring, unless you are absolutely sure of your microbial cleanliness. You could invite Acetobacter and Lactobacillus, too. A solution? Get MLF through quickly, either in a warm room or by transferring to tank.

Other tips for Brett-avoidance include:

  • Use 50ppm of SO2 at the crusher.
  • Avoid high must sugars and sluggish ferments.
  • Avoid adding DAP late in the ferment. Brett will thank you.
  • Keep pH under 3.55 if possible and TSO2 over 70ppm.
  • Wash and clean all barrels very thoroughly.
  • Be very careful of second-hand barrels.

Parish et al (2003) have more detailed recommendations.

If you suspect you have a Brett problem, an AWRI analysis is relatively inexpensive. Alternatively, do your own SNIFF’ Brett test*.

Quarantine any affected barrels, keep their SO2 levels up, only include them in your blend if a Brett-sensitive colleague gives the OK and make sure you sterile filter to at least yeast sterility.

Lastly where, you may ask, was Brett before it sprang to prominence a few years ago? We almost certainly thought it was mercaptan. The horse-dung manifestation is not dissimilar and the gritty tannin effect is not unlike the bitterness of some mercaptans.
We’re still learning.

Curtin, C.D., Bellon, J.R., Coulter, A.D., Cowey, G.D., Robinson, E.M.C., de Barres Lopes, M.A., Godden, P.W., Henschke, P.A., and Pretorius, I.S. (2005) The six tribes of ‘Brett’ in Australia. Aust. N.Z. Wine Industry J. 20(6):28-36

Parish, M., Kelly, M. and Baldwin, G. (2003) Managing Brett Australia. Aust. N.Z. Wine Industry J. 18(3):12-15

SNIFF’ Brett, from Fleurieu Winery Supplies, [email protected]

This is the full text of the article submitted. The published version may have been edited.