Close Menu

Smoke taint: a new headache for Bordeaux?

As the dust starts to settle – but with the pervasive odour of smoke still hanging in the air in a number of the vineyards of the Graves – it is perhaps time to start to assess the wider significance of the forest fires that have ravaged the Bordeaux region over the last two weeks.

A firefighter extinguishes dry grass. credit: Олег Копьёв (istock)

In the space of a few days, smoke taint has gone from an appreciable but largely hypothetical risk to an all too palpable and urgent fear in the region. It seems now set to become, if it has not already become, Bordeaux’s latest in a catalogue of climate change related anxieties.

But how genuine and how significant is that threat and what does the latest science contribute to our understanding and assessment of the severity of the risk in the years ahead?

To help me address that question I spoke directly, but off-the-record, to a number of producers (including some of those most immediately impacted) as well as to a number of consultant oenologists (including those with experience of the infamous fires that have ravaged Napa, particularly in the late summer of 2020). I also riffled through some of the most recent academic research on the subject. That new science has changed somewhat our assessment of the risks associated with smoke taint in wine, in ways that some of the more popular literature has yet to catch up with.

Out of the frying pan …

But let’s perhaps start briefly with the forest fires in the Bordeaux region this summer. As is well documented and as has been much discussed, through a combination of misfortune and suspected arson, two major fires broke out on July 12 in La Teste-de-Buch and in Landiras. The first is close to the famous Dune de Pilat just south of Arcachon, but a considerable distance from any Bordeaux vineyard. The second is in the Graves region, close to Sauternes, Barsac and Cérons and itself home to a number of vineyards, including Loïc Pasquet’s Liber Pater (reputedly, the most expensive wine in the world). A third, much smaller, fire broke out in the Northern Médoc at Vensac on the 18 of July, which was much more rapidly controlled.

In both La Teste-de-Buch and Landiras, the fire raged largely out of control for a full two weeks, consuming in the process over 20,000 hectares of forest. In Landiras the entire population of the town was evacuated, including the massive storage facility of Bordeaux négociant, Les Grand Chais de France and, of course, Liber Pater itself (as well as a host of other less well-known vineyards).

No direct fire damage was caused to any vineyard in the region in any of these episodes, though the edge of the fire was at one point less than 500 metres from the Liber Pater vineyard.

Bordeaux is not, of course, new to forest fires. The two most infamous are those of 1949 and 1989 (itself the hottest summer since 1949). By far the worst of these was the great fire of the Landes forest of 1949 which consumed the communes of Cestas, Saucats, Marcheprime and Mios. This was, at the time (and, in fact, until relatively recently), the most deadly recorded forest fire in Europe, with 50,000 hectares destroyed and, tragically, 82 deaths (including many fire fighters). They are commemorated in two memorials, unveiled in 2018, on either side of the D1010 between Cestas and Le Barp. Somewhat poignantly, the location is almost exactly half way between La Teste-de-Buch and Landiras.

No less interesting is that neither episode has tainted the reputation of these almost mythic vintages, even in Pessac-Léognan (parts of which were touched by the fire of 1949).

… And into the fire: the risk of smoke taint in the future

There are reasons for that and they are important in any wider assessment of the risk of fire damage and, ultimately smoke taint, in the region today. So what is the level of the risk and what are the potential implications in the years ahead for the Bordeaux region particularly?

There are at least five important factors to consider here (climate change; topography; the wider environmental damage caused by forest fires; the susceptibility of grapes to smoke taint at different stages in their development; and the geography and meteorology of the region).

Climate change

The first of these is climate change itself. 1949 and 1989 were both exceptionally hot summers. But the point is that they were exceptions. In a context of accelerating climate change what was exceptional then has become more normal today. Thus, whilst one might hope that the summer of 2022 will remain what statisticians call an ‘outlier’ (an event of unprecedented heat and dryness), that would be naïve in the extreme.

Sadly, much more likely is that it will become representative of a new kind of normality. That is what climate models now consistently predict. In such a new normality, the risk of forest fires – above all in the latter parts of the summer and early autumn (August-October) – is greatly elevated.

And that reminds us that, even for 2022, we are far from being out of the woods (as it were). Indeed, the risk of further forest fires in the months to come has only grown in the last two weeks. In the sustained absence of significant rainfall it can only rise further.


If that sounds bleak, and it does, then the second factor brings with it just a little glimmer of hope – at least for the Bordeaux region. It is topography, a factor stressed by practically everyone I spoke to – and most of all, by the consultant oenologists.

Bordeaux is, basically, flat (parts of it very flat indeed); and where it is not flat, it is not close to sizeable areas of forest. That is very good news when it comes to the threat of smoke taint in the wine itself, in 2022 and more generally. For two key factors elevating the risk of smoke taint (see the table below for a full summary) are the level and duration of the plant’s exposure to it.

In a flat landscape (such as the Médoc), smoke dissipates more quickly and without the localised high concentrations that are produced in a more densely contoured terrain such as the Napa Valley.

The wider environmental damage caused by forest fires

That in turn brings us to a third factor. In Bordeaux especially (and in other topographically similar regions with vineyards surrounded by either natural or commercial forests), the principle risk associated with forest fire may well not be smoke taint itself but eco-systemic restructuring and the associated loss of biodiversity. As Loic Pasquet makes clear, “the lake bordering Liber Pater was emptied by firefighters to try to save the forest which burned anyway. It took only 10 days of fire to consume the entire forest on the southern slope of the anticline and for us to lose all of its biodiversity”. Only some of that biodiversity is likely to return. The consequences of this cannot be underestimated.

That, too, is likely to sound very depressing. But in one sense at least it is probably overly optimistic. For it implies that smoke taint is not much of a threat in Bordeaux and that the 2022 vintages is unlikely to be remembered as the one lost to the forest fires. That may well be true; and of course everyone reading this will join me in hoping that this turns out to be accurate.

The susceptibility of grapes to smoke taint

But there are two crucial caveats. The first is that, as already suggested, the risk of further significant forest fires has not been reduced by the bringing under control of those that started earlier this month. The second is that the most recent research suggests that smoke taint can influence the quality of the wine rather earlier than was previously thought. The emerging consensus is that smoke taint is a tangible and significant risk at any point after floraison (flowering) and not veraison (colour change). To make matters worse, it is also extremely likely (given the mechanism through which taint comes to be present in the wine) that thick-skinned and small-berried varietals, with a greater surface area to volume ratio (such as Cabernet Sauvignon and Cabernet Franc) will prove more susceptible still to smoke taint than the typically larger berried and thinner skinned varietals on which most of the existing academic research has been conducted.

Whilst this remains a hypothesis to be tested, it is a very credible one and one likely to give Bordeaux vignerons (and those who empathise with their plight) additional sleepless nights.

Geography and meteorology of the region

Finally, even the most cursory reflection on the geography and meteorology of the Bordeaux region leads rapidly to the conclusion that it is the Médoc that is likely to experience the greatest relative risk of smoke taint, as it is to the associated risk of biodiversity loss and eco-systemic restructuring caused by rampant and contagious forest fires. The combination of a massive strip of forest running north-south down the western half of the Medoc and a predominant flow of air from west to east is not auspicious.

Bordeaux and, above all, the Médoc may well be temporarily out of the proverbial frying pan, but it is certainly not out of the fire.

Smoke taint: the scientific basics

Where does it come from?

  • free volatile phenols are released with the combustion of lignin in wood

How does it enter the plant?

  • volatile phenols are absorbed through the grape’s waxy cuticle
  • there is no significant transfer between the leaves and the grapes

What factors increase its prevalence?

  • proximity to the fire
  • stage of grape development (risk elevated by floraison and then veraison)
  • smoke density and concentration
  • length of exposure
  • smoke composition
  • the grape varietal (with the risk increasing with the surface area/volume ratio of the grape and the ratio of skin/juice)
  • length of maceration (increasing the transfer of phenols to the juice)
  • amount of press wine used (since press wine will be higher in phenolic concentration)

Relevant contextual factors

  • the topography of the area
  • wind speed and direction
  • the density of forest plantation

Myths associated with smoke taint

  • that it leaves a trace in subsequent vintages
  • that there is no significant risk before veraison (the grape’s change of colour)


It looks like you're in Asia, would you like to be redirected to the Drinks Business Asia edition?

Yes, take me to the Asia edition No