Technical opinion piece
Introduction
Brettanomyces has been around Central Otago for a long time. It has appeared in and out of wineries, in individual barrels, particular wines and, occasionally, as a persistent cellar problem. What has changed recently is not necessarily Brett itself, but our ability to see it.
The arrival of accessible, rapid quantitative PCR (qPCR) testing has changed the conversation. Where Brettanomyces was once inferred from a sensory character, a slowly developing plate, or eventually from elevated 4-ethylphenol (4-EP) and 4-ethylguaiacol (4-EG), we can now detect very small populations before the wine smells remotely “Bretty.” Modern molecular methods can detect populations that conventional culture methods may miss, including cells in viable-but-non-culturable states. ETS Laboratories, for example, reports a detection limit of approximately 10 cells/mL for its standard molecular assay, compared with the much longer turnaround associated with culture-based methods (ETS Laboratories, 2025a).
From what we are seeing locally, Brett appears increasingly prevalent in Central Otago wine. But there is an important distinction here: increased detection is not necessarily increased incidence. At present, I am not aware of sufficiently robust regional surveillance data to demonstrate that Brettanomyces has genuinely become more common in Central Otago. What we can say is that we are looking for it more often, looking earlier, and using substantially better tools.
This raises several questions. Is Brett becoming more prevalent, or have we simply discovered what was already there? Does it originate predominantly in the cellar, or are we bringing small populations in from the vineyard? What conditions allow an insignificant population to become a problem? And perhaps most controversially: if microbial diversity contributes to complexity in wine, is every detectable Brettanomyces cell necessarily undesirable?
My argument is that Brett needs to be taken seriously, but perhaps understood more carefully. Overt Brettanomyces spoilage is clearly undesirable and presents a genuine commercial risk. However, Brett is also part of a much larger microbial ecology. Our goal should therefore be less about fearing its mere presence and more about understanding population, activity, environment and risk.
That distinction may be particularly relevant to Central Otago, a region whose best wines are frequently valued not simply for fruit purity, but for complexity, savoury development, bottle bouquet and the ability to evolve.
Brett Hasn't Suddenly Arrived
Brettanomyces bruxellensis is remarkably well adapted to wine. It tolerates ethanol, low nutrient concentrations and relatively low pH, and can remain present under conditions where many other microorganisms have disappeared. My previous review of wine microbial ecology highlighted this ability and the particular suitability of barrel maturation for its persistence (Mattern, 2019b). Similar observations were made in my broader review of yeast and wine quality, where Brett was considered both as a spoilage organism and as a potentially more complicated contributor to wine aroma than its reputation suggests (Mattern, 2019c).
This ability to survive is one reason the historical prevalence of Brett is so difficult to establish.
Culture-dependent microbiology gives us a snapshot of organisms capable of growing under the conditions we provide them. Brett is slow growing, can exist at very low populations and can enter a viable-but-non-culturable state. Molecular methods have fundamentally changed that detection problem. Modern qPCR methods can detect Brett populations at very low concentrations and populations that conventional culture can underestimate or miss (ETS Laboratories, 2025a).
That alone gives us a plausible explanation for what appears to be increasing Brett prevalence in Central Otago.
We may not have more Brett. We may simply have better eyesight.
There is another piece worth considering: the vineyard.
Historically, Brett has often been framed principally as a winery organism associated with old barrels, compromised hygiene and contaminated equipment. There is good reason for that association. Once established in a winery, Brett finds numerous niches, including barrels, valves, transfer equipment and difficult-to-clean surfaces.
But evidence increasingly complicates the idea that the winery is necessarily where Brett begins.
Oro et al. (2019) investigated grapes, winery environments and fermentations and initially failed to detect B. bruxellensis directly on grapes. After enrichment, however, Brett was recovered from eight grape samples. Molecular characterisation showed close relationships between vineyard and winery isolates, supporting the possibility of movement from grapes into the winery. The important point is how difficult the vineyard population was to find: it was there, but conventional detection made it appear absent (Oro et al., 2019).
This fits a much broader understanding of the grape microbiome. Yeast and bacterial populations change throughout berry development and are influenced by climate, moisture, agricultural practice, berry health and damage. My earlier work similarly concluded that substantial microbial diversity arrives at the winery with the fruit, even though many of those organisms subsequently disappear as fermentation changes their environment (Mattern, 2019b).
Damaged and rot-affected fruit is particularly important because microbial populations can change dramatically when berry integrity is compromised. Brett has also been reported more frequently within the richer microbial populations associated with compromised grapes (Barata et al., 2012).
This does not demonstrate that Central Otago vineyards are the source of the Brett we are currently detecting. That would require regional sampling and strain-level work. But it gives us a reasonable hypothesis worth testing.
Instead of assuming that every Brett-positive wine represents a failure in winery sanitation, we should perhaps ask whether small populations are entering wineries regularly and whether cellar conditions determine which of those populations disappear, persist quietly, or proliferate.
From Organism to Organoleptic Impact
Brettanomyces itself is not an aroma.
Its metabolism is.
The compounds most commonly associated with Brett are 4-EP and 4-EG, formed through metabolism of hydroxycinnamic acid-derived precursors. At sufficient concentrations these contribute the familiar descriptors of medicinal, Band-Aid, horse, barnyard, leather, smoky and phenolic characters.
ETS notes approximate sensory thresholds around 300–600 µg/L for 4-EP and approximately 50 µg/L for 4-EG, although thresholds are highly dependent on the wine matrix and individual taster. Importantly, the ratio between the compounds also varies considerably; ETS reports 4-EP:4-EG ratios ranging from approximately 3:1 to 22:1 in a survey of 300 red wines. Two wines with the same 4-EP concentration can therefore present very differently (ETS Laboratories, 2025b).
This is where the simple equation—
Brett = fault
—starts becoming less satisfactory.
Joseph et al. (2013, 2015, 2017) characterised a considerably broader range of compounds associated with B. bruxellensis. Their work produced the Brettanomyces aroma wheel and associated Brett-containing wines not only with animal and chemical descriptors but also with savoury, spicy, earthy, woody and floral characteristics.
This was something I explored in both of my previous microbial reviews. Brett-associated compounds included descriptors as diverse as smoky, clove, spice, floral, rose, citrus, mint, fruit and woody characters alongside unmistakably negative attributes (Mattern, 2019c). The Brett/LAB comparison reproduced in my earlier review is particularly interesting: wines containing Brett, LAB, or both were frequently described using savoury, spicy, earthy, woody and floral categories, while animal and chemical categories occurred less frequently (Mattern, 2019b).
None of this means that Brett spoilage should be romanticised. It means that microbial aroma chemistry is concentration and matrix dependent.
That is hardly unique to Brett.
Acetic acid, acetaldehyde, higher alcohols, volatile sulphur compounds, esters and numerous other fermentation-derived compounds can contribute positively below certain concentrations and negatively when allowed to dominate. My previous work on wine quality argued that complexity and balance are particularly important to highly involved wine consumers and wine professionals (Mattern, 2019a). Wine aroma is not simply the presence or absence of individual compounds; perception arises from their concentration, interaction and the matrix in which they occur.
Wedral et al. (2010) make essentially this point in relation to Brett: Brett-associated metabolites may add aroma complexity at lower concentrations but become unpleasant as their concentration increases (Wedral et al., 2010).
The question, then, is not merely:
Is Brett present?
It is:
How much is present, is it viable, is it growing, what is it producing, and is the wine moving toward a sensory outcome we don't want?
What Makes Brett Take Off?
Brett's reputation has been earned partly because it is an extraordinary survivor.
It can persist after alcoholic fermentation when sugar is scarce, alcohol is high and most competing yeast populations have collapsed. Renouf et al. (2006) found Brett particularly well adapted compared with other non-Saccharomyces yeasts and demonstrated its persistence as fermentation conditions became increasingly hostile (Renouf et al., 2006).
Temperature matters substantially.
Growth and volatile phenol production have been demonstrated at 15, 20 and 25°C. Other work suggests that keeping wine at ≤15°C, together with adequate molecular SO₂, can substantially restrict Brett activity, although not necessarily eradicate it (Barata et al., 2008; Zuehlke & Edwards, 2013). Zuehlke and Edwards demonstrated an important distinction here: cool conditions and SO₂ substantially reduced Brett activity, but metabolically active cells could remain even when the organism could no longer be cultured normally. Culturability could subsequently return (Zuehlke & Edwards, 2013). (PubMed)
This is worth remembering in Central Otago. A cold cellar is not simply convenient storage. Temperature is part of microbial management.
The risk also changes with pH and SO₂. It is molecular SO₂ rather than simply the free SO₂ number that has traditionally been considered most important microbiologically, and the proportion of molecular SO₂ falls rapidly as pH rises. ETS cites 0.5–0.8 mg/L molecular SO₂ as commonly suggested inhibitory concentrations while emphasising that SO₂ does not necessarily kill spoilage organisms. At pH 3.5, approximately 40 mg/L free SO₂ is required to achieve 0.8 mg/L molecular SO₂ under their stated assumptions; by pH 3.8 that figure is approximately 79 mg/L (ETS Laboratories, 2025c).
That difference is enormous in practical winemaking.
A Complication to the Molecular SO₂ Story in Red Wine
There is, however, an important complication to the molecular SO₂ story in red wine.
Clark Smith has questioned the usefulness of treating calculated molecular SO₂ in a red wine as though it necessarily represents the same degree of microbiological protection that we expect from the calculation in a simpler wine matrix (Smith, 2013; Duane, 2023). There is good scientific evidence supporting at least part of that argument.
In red wine, bisulfite reacts readily with anthocyanins to form anthocyanin-bisulfite complexes. Once associated in this form, that SO₂ provides little antimicrobial activity. More importantly, some conventional methods used to measure free SO₂ disturb that equilibrium during analysis. Weakly bound SO₂ can consequently be released and measured as though it were freely available in the wine (Howe et al., 2018).
Howe et al. (2018) demonstrated that conventional SO₂ measurements can therefore overestimate the antimicrobial activity of SO₂ in red wine, particularly in wines containing substantial concentrations of anthocyanins. Their work showed that weak anthocyanin-bisulfite complexes can dissociate during conventional analysis and that these complexes themselves have negligible antimicrobial activity (Howe et al., 2018). (ResearchGate)
This distinction is particularly relevant to young red wines.
It would, however, be an overstatement to say that molecular SO₂ itself is irrelevant in red wine. True molecular SO₂ remains microbiologically active. What may be misleading is assuming that conventionally measured free SO₂, followed by a simple pH calculation, necessarily tells us how much microbiologically active molecular SO₂ is actually present in a red wine (Coelho et al., 2015; Howe et al., 2018).
Alternative headspace measurements that do not significantly disturb the existing equilibrium have returned SO₂ values in red wines approximately 30–60% below those obtained through commonly used conventional methods, further illustrating why the analytical number and the microbiological reality may not always be the same thing (Coelho et al., 2015; Howe et al., 2018). (Iowa State University Extension)
This may help explain something winemakers have observed for a long time: two red wines with apparently similar pH and free SO₂ can behave very differently microbiologically.
There is another complication. Brett strains themselves differ considerably in their response to SO₂. Research involving the Institut des Sciences de la Vigne et du Vin and the University of Bordeaux has demonstrated substantial genotype-dependent differences in B. bruxellensis tolerance to SO₂, meaning that the idea of a single molecular SO₂ concentration capable of guaranteeing Brett control is biologically too simple (Avramova et al., 2018). (PubMed Central (PMC))
This does not make SO₂ unimportant. Quite the opposite.
It suggests that SO₂ should be treated as one component of microbial control rather than as an absolute microbiological safety number.
For a red wine, particularly Pinot Noir, I would therefore be reluctant to look at pH, free SO₂ and a calculated molecular SO₂ concentration in isolation and conclude that the wine is protected. Population, strain, temperature, residual substrate, wine composition, anthocyanin concentration, oxygen exposure and actual microbiological monitoring all matter.
Residual sugar, nutrients, oxygen, barrel environment, lees, cellar temperature and microbial competition all influence what happens next.
And this brings us to Clark Smith.
Clark Smith, Oxygen and Redox
In Postmodern Winemaking, Clark Smith (2013) dedicates chapters both to wine redox chemistry and to what he calls Integrated Brettanomyces Management. His view is deliberately contrary to conventional spoilage-control thinking (Smith, 2013).
Smith's model rests on three concepts: creating a nutrient desert, maintaining microbial balance and achieving aromatic integration through appropriate wine structure (Smith, 2011).
His discussion of oxygen is particularly interesting.
Brett is capable of exploiting oxygen and alternative nutrient sources under appropriate conditions. Smith consequently argues that oxygen availability, residual nutrient status and the wine's ability to consume oxygen—what he terms its reductive strength—are interconnected.
His position is not simply that oxygen is good or oxygen is bad.
Poorly managed oxygen availability can create opportunities for Brett. At the same time, Smith argues that carefully timed oxygen exposure early in red-wine development can build phenolic structure and ultimately increase the wine's reductive capacity. In his model, the important distinction is between purposeful oxygenation during wine development and uncontrolled oxygen ingress during élevage (Smith, 2011, 2013).
This distinction deserves attention.
Cold Cellars, Oxygen Consumption and Temperature Stability
There is also an interesting temperature component to this argument.
Cold storage is conventionally—and correctly—used as a microbiological control tool because lower temperatures substantially restrict Brett growth. Zuehlke and Edwards (2013), for example, demonstrated that maintaining wines at ≤15°C together with appropriate molecular SO₂ substantially restricted Brett activity, although it did not guarantee eradication (Zuehlke & Edwards, 2013).
But the wine itself is also temperature dependent.
Red wine continuously participates in oxidation-reduction reactions involving phenolics, oxygen, metals, SO₂ and other wine components. Wine does not simply have oxygen in it; it consumes oxygen chemically, and different wines do so at markedly different rates depending upon their composition (Carrascón et al., 2015, 2018). The rates of oxygen consumption are influenced by phenolic composition and metal catalysts, among other factors (Carrascón et al., 2015, 2018).
Temperature matters to these reactions as well.
Previous work reported that an air-saturated red wine consumed its dissolved oxygen in approximately 25 hours at 13°C but in only around three hours at 30°C, illustrating just how dramatically temperature can alter oxygen-consumption kinetics (Moutounet & Mazauric, 2001, as cited in del Álamo-Sanza et al., 2017).
This is an important nuance to the idea that colder is always better.
Smith argues that excessively cold cellar conditions can suppress not only microbial activity but also the wine's own oxygen-consuming chemistry. In his discussion with Jim Duane on Inside Winemaking, the wider conversation around Smith's postmodern approach includes making wine with little or no SO₂, micro-oxygenation and the redox concepts central to his winemaking philosophy (Duane, 2023; Smith, 2013).
The basic chemistry underlying the temperature argument is plausible. At lower temperatures oxygen-consuming reactions slow considerably. This means oxygen introduced into a cold wine can potentially remain dissolved for longer than it would in a warmer and more chemically reactive wine. When the wine subsequently warms, both biological activity and chemical reaction rates increase.
What is less established is the next step in the argument: that a very cold cellar followed by warming necessarily creates an oxygen-enriched environment that specifically causes Brett proliferation. I have not found sufficient peer-reviewed evidence to state that as established fact.
It is better considered a credible redox hypothesis.
Imagine two wines receiving the same small amount of oxygen through barrel ingress. In the warmer, more chemically reactive wine, phenolics and other wine constituents may consume that oxygen relatively quickly. In the colder wine, the chemical reactions responsible for consuming oxygen proceed more slowly. If that wine subsequently warms while appreciable oxygen remains available, increased microbial metabolic activity and greater oxygen availability may temporarily coincide.
That combination is potentially important.
The practical implication may therefore be more nuanced than simply keep the cellar as cold as possible.
A consistently cool cellar is clearly preferable to a warm cellar for suppressing Brett. But large temperature swings may deserve greater attention than they receive. A cellar that spends winter extremely cold and then warms substantially in spring or summer is not necessarily microbiologically or chemically equivalent to a cellar maintained at a relatively stable temperature throughout élevage.
For Central Otago this is particularly interesting.
Our seasonal temperature variation can be considerable, and wineries differ enormously in their ability to regulate barrel-hall temperature. A wine sitting at very low temperature through winter may have low microbial activity but also relatively slow oxygen-consumption chemistry. When the cellar warms, microbial activity, chemical reaction rates and oxygen consumption all begin increasing together.
This does not mean warming the cellar to control Brett. Warmer temperatures demonstrably favour Brett growth, and that remains the stronger evidence (Barata et al., 2008; Zuehlke & Edwards, 2013).
It instead raises the possibility that temperature stability may matter alongside absolute temperature.
Rather than viewing cellar temperature solely through the question of “How cold can we keep the wine?”, perhaps the more useful question is: At what stable temperature can we suppress undesirable microbial growth while still allowing the wine to maintain appropriate chemical and redox development?
That is a much more interesting research question, and I am not convinced we currently have a definitive answer.
This also returns us to the question of molecular SO₂. If conventional analysis can overestimate microbiologically available SO₂ in anthocyanin-rich red wine, and if the wine's capacity to consume oxygen is itself affected by temperature and wine composition, then neither free SO₂ nor cellar temperature should be viewed as independent control switches. They form part of a much larger system involving microbial population, pH, phenolics, oxygen, temperature and redox chemistry (Howe et al., 2018; Smith, 2013).
This is, in many ways, the strongest part of Smith's argument even where some of his specific conclusions remain debatable:
Wine microbiological stability is a property of the system, not of a single analytical number.
However, some of Smith's claims—particularly his theories around colloidal aromatic integration and his proposed ability of structured wines to carry comparatively high concentrations of Brett metabolites without sensory intrusion—should be treated as his working winemaking model rather than settled scientific consensus. His ideas are valuable because they force us to think about Brett ecologically and chemically rather than simply through sterilisation, but not every aspect has equivalent experimental support.
The conventional evidence is clearer on one point: warm conditions favour Brett growth. Growth and volatile phenol production occur readily around 15–25°C, while colder storage substantially restricts activity (Barata et al., 2008; Zuehlke & Edwards, 2013).
For practical purposes, therefore, temperature, oxygen management, wine redox condition and microbial monitoring should be considered together rather than independently.
Prevention: Know What Is There Before It Becomes a Problem
The best Brett treatment remains preventing a small population from becoming a large one.
That starts with fruit.
Healthy, intact fruit generally carries a different microbial load from damaged or rot-affected fruit. Where fruit condition is compromised, separating lots, improving fruit selection and considering earlier microbiological surveillance make sense. Brett should also be considered a possible vineyard passenger rather than automatically attributed to poor cellar hygiene (Oro et al., 2019).
During fermentation, the objective should be a strong, complete fermentation leaving as little accessible substrate as practical. Long periods between alcoholic fermentation and MLF deserve particular attention because microbial succession can create an opportunity for Brett. Renouf et al. identified the beginning of MLF as a particularly important stage in Brett contamination and population development (Renouf et al., 2006).
Once wines enter élevage, good cellar fundamentals remain remarkably effective: maintain barrels properly topped, control oxygen ingress, understand pH and molecular SO₂ rather than relying blindly on free SO₂, while recognising that conventionally calculated molecular SO₂ may itself overestimate antimicrobial protection in anthocyanin-rich red wines (Howe et al., 2018), keep susceptible wines cool and preferably at a relatively consistent temperature, minimise residual fermentable substrate, clean transfer equipment thoroughly and pay particular attention to difficult niches such as valves, hoses, fittings and barrels.
Used barrels deserve respect rather than fear. Research supports them as an important ecological niche for Brett, but contamination is not an inevitable consequence of using oak. Maintenance and monitoring matter (Oro et al., 2019; Wedral et al., 2010).
Most importantly, test before sensory evidence appears.
Once you can smell Brett, microbiological testing is no longer early detection.
Sampling Is Half the Test
There is little value in extremely sensitive qPCR if the sample does not represent the wine.
Brett settles.
ETS demonstrated this rather dramatically in a barrel trial. A sample taken from the top of an unmixed barrel returned approximately 18,000 cells/mL. After mixing the same barrel, the result was approximately 1,800,000 cells/mL. Roughly 99% of the Brett population had settled toward the bottom over two weeks (ETS Laboratories, 2025d).
That is a hundred-fold analytical difference created simply by sampling.
Best practice is therefore:
Tanks: mix thoroughly before sampling. If mixing is impossible, collect top, middle and bottom fractions and create a vertical composite.
Barrels: ideally sample immediately after racking when the wine is homogeneous. Otherwise stir before sampling or take a composite from surface, middle and bottom.
Valves: sanitise the sample port and flush several volumes before collecting the sample.
Barrel programmes: recognise barrel-to-barrel variation. One clean barrel does not establish that the rest of the lot is clean.
Trend monitoring: sample consistently. Comparing poorly standardised samples can create apparent population changes that are really sampling artefacts.
After antimicrobial treatment: allow an appropriate period before interpreting microbiological efficacy; ETS cites 20–30 days following some chitosan or DMDC treatment protocols (ETS Laboratories, 2025e).
I would also argue strongly for combining tools.
qPCR tells us about organisms. 4-EP/4-EG tells us about metabolic history and activity. Sensory tells us what the wine actually expresses.
These are related questions, but they are not the same question.
A qPCR-positive wine with low and stable ethylphenols presents a different risk from a wine with a rapidly increasing Brett population and increasing 4-EP/4-EG. Likewise, filtration may remove cells without removing the ethylphenols already produced. ETS therefore recommends looking at microbiological results alongside chemical markers rather than treating either as the whole story (ETS Laboratories, 2025b).
The More Difficult Question: Could Some Brett Be Part of Central Otago Wine?
This is where this piece becomes opinion.
We should be clear first: I am not suggesting that overt Brettanomyces infection is desirable.
A wine dominated by Band-Aid, horse, medicinal phenolics or barnyard character has lost varietal and site expression. Sending unstable wines into market with viable Brett populations is also a commercial risk, particularly where those wines may warm during storage and distribution.
That is not the argument.
The question is whether the presence of Brett—or historical low-level activity from Brett and other microorganisms—has sometimes contributed positively to wines that we regard as complex.
Central Otago places considerable value on Pinot Noir. At its best, Pinot is rarely compelling because it smells solely of pristine primary fruit. The wines we celebrate with bottle age often develop savoury, earthy, spicy, forest-floor and other secondary and tertiary characters. Complexity matters. My earlier work found balance and complexity particularly important in definitions of quality among highly involved and expert wine consumers (Mattern, 2019a).
So what portion of what we historically describe as bottle bouquet, savoury complexity or "funk" is entirely grape and maturation chemistry, and what portion has occasionally been microbial?
We probably don't know.
The work of Joseph and colleagues should at least make us uncomfortable with overly simple answers. Brett strains produce a far broader suite of aroma-active compounds than 4-EP alone, and the wines examined in their aroma-wheel work included savoury, spicy, earthy, woody and floral descriptions alongside negative ones (Joseph et al., 2013, 2015, 2017).
My earlier conclusion was therefore deliberately cautious: despite Brettanomyces being known as a spoilage organism, there is potential for positive impact, and research should investigate the parameters determining positive versus negative aroma contributions (Mattern, 2019b).
That question remains interesting.
The Hospital Disease
Clark Smith pushes the argument much further.
He provocatively describes Brett as a “hospital disease”, arguing that aggressive attempts to sterilise the wine ecosystem can remove organisms that would otherwise compete with Brett, leaving an ecological vacuum in which an exceptionally persistent organism eventually wins (Smith, 2011).
The metaphor is provocative and should not be mistaken for established microbiological doctrine. But the ecological question behind it is legitimate.
In vineyards, we increasingly discuss soil biology in terms of communities rather than individual organisms. Diversity, competition, succession and interaction matter. A healthy system is not necessarily a sterile system.
Why should fermentation ecology automatically be viewed differently?
Wine is unquestionably a microbial ecosystem. Yeasts and bacteria interact from the grape surface through alcoholic fermentation, MLF and maturation, changing nutrient availability and producing compounds that stimulate or inhibit other organisms. Modern reviews describe wine microbiota as a diverse, dynamic ecosystem in which microbial interactions influence both composition and sensory outcomes (Liu et al., 2017).
My earlier review arrived at much the same question. Microbial populations are not static. Different organisms dominate at different stages, and interactions can be competitive, antagonistic, commensal or potentially mutualistic. I suggested that if populations remain modulated by healthy conditions, the result might be more of an “interactive dance rather than an overpowering spectacle” (Mattern, 2019b).
That analogy seems particularly appropriate to Brett.
Perhaps the wrong question is:
How do we eliminate every microorganism except the ones we deliberately purchased in a packet?
Perhaps a more useful question is:
How do we create an ecosystem in which no undesirable organism gets the opportunity to dominate?
That is conceptually very different.
It does not mean abandoning sanitation. Clean equipment remains fundamental. It does not mean tolerating an expanding Brett population. And it certainly does not mean bottling microbiologically unstable wine in the hope that biodiversity will sort things out.
It means acknowledging that sanitation and sterility are not synonyms, and neither is automatically equivalent to microbial stability.
A wine that has been stripped of competitors but still contains a handful of resilient Brett cells may not necessarily be more biologically secure than a wine in which nutrients have been exhausted, populations have competed through their natural succession, pH, temperature and SO₂ are appropriate, and the system has reached stability.
That hypothesis deserves proper testing rather than dogma from either side.
It also gives a slightly different perspective to SO₂ and temperature management. If microbiological stability is ecological rather than simply numerical, then neither an apparently correct molecular SO₂ calculation nor an extremely cold cellar should provide false confidence. Both are valuable tools. Neither tells us everything.
Perhaps this is the common thread running through Brett management: the organism does not respond to our spreadsheet; it responds to the actual environment in the wine.
Where Does That Leave Us?
Brettanomyces is real, persistent and capable of destroying wine quality. Better access to qPCR in Central Otago is allowing us to detect it sooner and at substantially lower populations than was previously practical. That increased detection should not automatically be interpreted as evidence of a regional outbreak.
We are learning what was previously invisible.
The vineyard also deserves greater consideration. Brett can occur on grapes at very low populations and molecular work has demonstrated relationships between vineyard and winery isolates. Whether that is an important source in Central Otago remains an unanswered regional question and, in my view, an excellent one to investigate.
In the winery, the practical response is relatively straightforward: sound fruit, complete fermentations, sensible nutrient management, controlled oxygen exposure, good barrel and equipment hygiene, appropriate SO₂ management, stable and appropriately cool storage, representative sampling and routine monitoring of at-risk wines. Where Brett is detected, following population and 4-EP/4-EG trends gives considerably more information than waiting for someone to smell it.
But there is a larger question that deserves to remain open.
Wine has never been microbiologically simple. Saccharomyces, non-Saccharomyces yeasts, Oenococcus, Lactobacillus, Pediococcus, Brettanomyces and numerous other organisms occupy different stages of a changing ecosystem. Their relationships influence fermentation, aroma and ultimately what we call wine quality. My earlier reviews argued that non-Saccharomyces organisms traditionally categorised as spoilage organisms may have more nuanced roles in wine composition than those categories imply (Mattern, 2019b, 2019c).
I still think that is a question worth pursuing.
For Central Otago, perhaps the opportunity presented by better microbiological testing is not simply to eradicate Brett more efficiently. It is to finally study it properly.
We can begin asking whether populations differ between subregions, vineyards and vintages; whether damaged versus pristine fruit alters Brett incidence; whether particular cellar practices favour or suppress populations; how temperature and oxygen exposure interact with those populations; what relationships exist with LAB and other yeasts; whether conventional molecular SO₂ calculations accurately predict microbiological protection in our Pinot Noir; whether seasonal cellar-temperature fluctuations influence Brett risk; and whether wines showing low-level historical Brett activity are sensorially preferred, rejected, or simply different.
Most importantly, we can distinguish presence from proliferation and proliferation from spoilage.
There is a great deal we now know about Brettanomyces. There is considerably more that we do not.
And somewhere between trying to sterilise it out of existence and allowing it to run wild is probably the much more interesting territory: understanding the ecology well enough that the winemaker, rather than the microorganism, remains in control.
References
Avramova, M., Vallet-Courbin, A., Maupeu, J., Masneuf-Pomarède, I., & Albertin, W. (2018). Molecular diagnosis of Brettanomyces bruxellensis sulfur dioxide sensitivity through genotype specific method. Frontiers in Microbiology, 9, 1260.
Barata, A., Malfeito-Ferreira, M., & Loureiro, V. (2012). Changes in sour rotten grape berry microbiota during ripening and wine fermentation. International Journal of Food Microbiology, 154(3), 152–161.
Barata, A., Pagliara, D., Piccininno, T., Tarantino, F., Ciardulli, W., Malfeito-Ferreira, M., & Loureiro, V. (2008). The effect of sugar concentration and temperature on growth and volatile phenol production by Dekkera bruxellensis in wine. FEMS Yeast Research, 8(7), 1097–1102.
Carrascón, V., Vallverdú-Queralt, A., Meudec, E., Sommerer, N., Fernandez-Zurbano, P., & Ferreira, V. (2018). The kinetics of oxygen and SO₂ consumption by red wines. What do they tell about oxidation mechanisms and about changes in wine composition? Food Chemistry, 241, 206–214.
Carrascón, V., Vallverdú-Queralt, A., Meudec, E., Sommerer, N., Fernandez-Zurbano, P., & Ferreira, V. (2015). Oxygen consumption by red wines. Part I: Consumption rates, relationship with chemical composition, and role of SO₂. Journal of Agricultural and Food Chemistry, 63(51), 10928–10937.
Coelho, J. M., Howe, P. A., & Sacks, G. L. (2015). A headspace gas detection tube method to measure SO₂ in wine without disrupting SO₂ equilibria. American Journal of Enology and Viticulture, 66(3), 257–265.
Duane, J. (Host). (2023, June 26). Clark Smith—Postmodern winemaking and video course (Episode 153) [Audio podcast episode]. Inside Winemaking.
ETS Laboratories. (2025a). Scorpions genetic detection of spoilage organisms. ETS Laboratories.
ETS Laboratories. (2025b). Brettanomyces monitoring by 4-EP and 4-EG. ETS Laboratories.
ETS Laboratories. (2025c). Controlling microbial growth with molecular SO₂. ETS Laboratories.
ETS Laboratories. (2025d). Collecting samples for microbial analysis. ETS Laboratories.
ETS Laboratories. (2025e). Scorpions genetic detection: Sampling and treatment considerations. ETS Laboratories.
Howe, P. A., Worobo, R. W., & Sacks, G. L. (2018). Conventional measurements of sulfur dioxide (SO₂) in red wine overestimate SO₂ antimicrobial activity. American Journal of Enology and Viticulture, 69(3), 210–220.
Joseph, C. M. L., Albino, E., & Bisson, L. F. (2017). Creation and use of a Brettanomyces aroma wheel. Catalyst: Discovery into Practice, 1(1), 12–20.
Joseph, C. M. L., Albino, E. A., Ebeler, S. E., & Bisson, L. F. (2015). Brettanomyces bruxellensis aroma-active compounds determined by SPME GC-MS olfactory analysis. American Journal of Enology and Viticulture, 66(3), 379–387.
Joseph, C. M. L., Gorton, L. W., Ebeler, S. E., & Bisson, L. F. (2013). Production of volatile compounds by wine strains of Brettanomyces bruxellensis grown in the presence of different precursor substrates. American Journal of Enology and Viticulture, 64(2), 231–240.
Liu, Y., Rousseaux, S., Tourdot-Maréchal, R., Sadoudi, M., Gougeon, R., Schmitt-Kopplin, P., & Alexandre, H. (2017). Wine microbiome: A dynamic world of microbial interactions. Critical Reviews in Food Science and Nutrition, 57(4), 856–873.
Mattern, T. H. (2019a). Balance, complexity and wine quality. Lincoln University.
Mattern, T. H. (2019b). The relevance of Lactobacillus, Pediococcus and Brettanomyces to wine production: Considering microbial interactions and symbiosis. Lincoln University.
Mattern, T. H. (2019c). Yeast and wine quality. Lincoln University.
Oro, L., Ciani, M., & Comitini, F. (2019). Occurrence of Brettanomyces bruxellensis on grape berries and in related winemaking cellar. Frontiers in Microbiology, 10, 415.
Renouf, V., Falcou, M., Miot-Sertier, C., Perello, M. C., De Revel, G., & Lonvaud-Funel, A. (2006). Interactions between Brettanomyces bruxellensis and other yeast species during the initial stages of winemaking. Journal of Applied Microbiology, 100(6), 1208–1219.
Schumaker, M. R., Chandra, M., Malfeito-Ferreira, M., & Ross, C. F. (2017). Influence of Brettanomyces ethylphenols on red wine aroma evaluated by consumers in the United States and Portugal. Food Research International, 100, 161–167.
Smith, C. (2011). Integrated Brett management. Wine Business Monthly.
Smith, C. (2013). Postmodern winemaking: Rethinking the modern science of an ancient craft. University of California Press.
Wedral, D., Shewfelt, R., & Frank, J. (2010). The challenge of Brettanomyces in wine. LWT – Food Science and Technology, 43(10), 1474–1479.
Zuehlke, J. M., & Edwards, C. G. (2013). Impact of sulfur dioxide and temperature on culturability and viability of Brettanomyces bruxellensis in wine. Journal of Food Protection, 76(12), 2024–2030.