A Note on Smoke Taint

By  Jim HarbertsonThomas Henick-KlingMarkus Keller and Michelle Moyer

Washington State University Viticulture & Enology Department

Wide-spread fires in eastern Washington have filled the valleys with a lingering smoky haze. While warnings have been issued for many areas relating to air quality, there is also concern regarding how this smoke may affect grapes.

Smoke residue contains high concentrations of volatile phenols, such as guaiacol and eugenol. “Smoke taint” has been found in juice and wine made from grapes, as the glycosylated forms of these phenols tend to accumulate in the skin and mesocarp (pulp) of the berry. These compounds are released during alcoholic and malolactic fermentation (2,3), causing the wine to become unpleasantly ‘pharmaceutical’, ‘dirty’, ‘ash tray’, ‘medicinal’, ‘camp fire’, or ‘burnt’, and reduces the perception of varietal fruit aroma.

In the Vineyard. The timing and amount of smoke-exposure can influence the appearance of smoke taint in subsequent wine. For example, taint can develop from low levels of exposure early in the season (6 inch shoot growth to bloom); from variable levels of smoke exposure from pea-size berries to véraison; and from high levels of smoke exposure between véraison and harvest (5,6), with a peak sensitivity about one week after the onset of véraison (4,5). We currently do not know how these controlled levels of smoke exposure relate to natural levels and how long smoke from wildfires needs to be present in vineyards before smoke taint becomes a problem.

With the forecast growth and lack of containment of the 2012 WA wildfires, harvesting sooner rather than later is recommend. This will help to reduce fruit exposure to smoke and likelihood of development of smoke taint.

In the Winery. Because grape skins accumulate smoke-taint associated phenols, reducing skin contact time can help reduce the severity of smoke-taint in wines. This is less problematic for white wines because normally skin contact is limited. We recommend whole cluster pressing, using free-run juice and separating press juice, to minimize the potential for extracting smoke taint aromas. Clarifying white wine must rapidly and thoroughly should help reduce absorption of volatile phenolics into the wine. Yeast hulls can be used as clarifying aid and may even absorb some of the volatile phenols.

While skin contact time cannot be avoided, shortening it and avoiding pectinase or glycosidase enzymatic additions is recommended for red wines. Because anthocyanins are quickly extracted (4-5 days), a short maceration may only limit tannin extraction.

There is evidence (8) that some yeast strains may reduce smoke aromas and flavors without diminishing “fruit” aromas (AWRI 1503, ICV GRE, AWRI Fusion and SIHA active 3) while others exacerbate smoke aromas (AWRI 1176, ICV D254, BDX and S6U) and make higher amounts of volatile acidity (AWRI 1176, S6U). Some yeast strains may have higher amounts of enzymes that cleave smoke taint precursors, or have higher amounts of pectinase enzymes that break down the skin cell walls.

The addition oak chips and tannins has also been looked at,  but the resulting wines were not less smoky, but in fact, were both oaky and smoky. This method was seen as adding complexity as opposed to removal of taint (8).

Glycosylated phenolics are hydrolyzed during winemaking and aging, causing a dramatic increase in the volatile phenolics over a 12 month period (1 µg/L to peak 388 µg/L)  (5). Other research has shown that guaiacol and cresol (4-methylguaiacol) continued to increase over 3 years of bottle aging (7). As a result, methods of removal like reverse osmosis coupled with solid phase adsorption (1) or ultra-filtration will have limited success; smoke taint will return over time.Reverse osmosis with solid phase adsorption, and ultra-filtration are not specific treatments and they can also remove desirable aromas, color, and phenolics, resulting in neutral, but usable, wines (until the smoke taint returns).

Blending is an option, but can be difficult as the odor threshold for smoke taint compounds is low. However, it may work if the volume of smoke tainted wine used for blending is also low. In extreme cases, bottling affected wine as a separate label may be an alternative. Send in suspect fruit and wine samples for analysis. It is important to have an evaluation of both the volatile phenols and their corresponding glycosylated precursors.


Odor Threshold (µg/L): guaiacol 95 (white wine), 75 (red wine); creosol 65 (white and red wine); 4-ethylguaicol 70 (white wine), 110-150 (red wine); 4-ethylphenol 1100 (white wine), 1200 (red wine). Adapted from (3).



1) Fudge et al. 2011. Aus. J. Grape Wine Res. 17:S41-48; 2) Hayasaka, et. al. 2010. J. Agric. Food. Chem. 58:10898-10998; 3) Kennison, et. al. 2008. J. Agric. Food. Chem. 58:7379-7383; 4) Kennison, et. al. 2009. Aus. J. Grape Wine Res. 15:228-237;   5) Kennison, et. al. 2011. Aus. J. Grape Wine Res. 17:S5-S12; 6) Sheppard, et. al. 2009. Am. J. Enol. Vitic. 60: 98-103; 7) Singh, et al. 2011. Aus. J. Grape Wine Res. 17:S13-S21; 8) Ristic et al. 2011. Aus. J. Grape Wine Res. S29-S40