Wines produced from smoke-exposed grapes can exhibit unpleasant smoky, burnt or ashtray characteristics. This project investigated the effects that yeast selection, malolactic fermentation and wine style had on the volatile phenols associated with wines produced from smoke-exposed grapes.
The occurrence of wildfires is increasing and affecting many wine-producing countries (United States of America, Spain, Chile, Australia and South Africa). Some of these fires are so devastating that entire vineyards and regions are destroyed. Besides the damage caused by these fires, the smoke that is produced in the vineyards or the surrounding areas could contribute to grapes and wines having an unpleasant, burnt and smoky character or palate, known as smoke taint.1 The compounds usually associated with this taint are known as volatile phenols (VPs), with characteristics ranging from sweet, smoky and toasty to ashtray, burnt and tar.1,3 Guaiacol, 4-ethyl guaiacol, 4-methylguaiacol and 4-ethylphenol are the main VPs used as chemical ‘markers’ for the presence of smoke-tainted wine.1,2 These VPs and their precursors are absorbed by grapes and leaves, and released into the wines during winemaking and storage.
Yeast and lactic acid bacteria (LAB) possess enzymes that assist with improving wine flavour and complexity, but these enzymes also increase the levels of VPs. Taking the above-mentioned into consideration, this project investigated the role that yeast selection, malolactic fermentation (MLF) and wine style played on the levels of VPs found in wines.
Chenin blanc and Merlot wines were produced from smoke-exposed grapes and juice using 40 different commercial yeast strains from various suppliers. Yeast strains produced wines with variable levels of VPs, which ranged from high, intermediate or low. In the first Chenin blanc trial, strains UCLM S325 and Merit produced wines with the lowest guaiacol levels, while wines produced with QA23 and BM4X4 contained the highest levels. Wines produced with QA23 also contained the highest 4-methylguaiacol, 4-ethylguaiacol, phenol and eugenol levels.
In a separate Chenin blanc trial, wines produced with SIHA A3 and AWRI R2 contained the lowest guaiacol levels and those produced with NT 116, ICV GRE and QA23 had the highest guaiacol levels. Wines that underwent spontaneous alcoholic fermentation also contained high guaiacol levels.
In the Merlot trial, wines that underwent spontaneous alcoholic fermentation contained the highest guaiacol levels and those produced with NT 112 had the lowest levels. Merlot wines produced with RX60, WE 372, VIN 13, NT 202 and FX10 contained relatively low VP levels. However, wines produced with D254, Exence, D21, NT 50 and QA23 produced high guaiacol levels.
Smoke-affected Merlot and Shiraz wines were produced with different LAB strains and MLF strategies. Lactic acid bacteria strains and MLF affected the concentrations of VPs found in wines. In a Shiraz trial, wines that underwent sequential MLF with Viniflora® oenos™, Viniflora® CH16™, Enodoc™ ML-Fast and Enoferm V22™ contained lower guaiacol levels than those that underwent co-inoculated MLF. Whereas wines that underwent co-inoculated MLF with Lactoenos® SB3 contained lower levels than those that underwent sequential MLF. Wines without MLF contained lower levels of VPs than wines that underwent MLF. The levels of VPs in wines varied depending on the LAB strain and the MLF strategy (co-inoculation or sequential inoculation) that were used.
Rosé and red wines were produced from smoke-exposed Merlot and Shiraz grapes and juice. Rosé wines contained lower levels of VPs than red wines produced with the same grapes or juice. However, our results show that yeast selection is critical to producing acceptable wines. For example, if high guaiacol producing yeast strains are used, the negative effects associated with VPs will be more prominent. In general, Shiraz wines contained higher levels of VPs than found in other grape cultivars.
The levels of VPs in wine can be reduced by the production of blanc de noir or rosé from red grapes. The advantages of producing blanc de noir or rosé wines are that the wines do not have to undergo MLF and can be produced for early release and consumption.
Fire or smoke-exposed grapes should be sampled and analysed before harvest to determine the levels of VPs in the grape juice. If possible, winemakers should also harvest grapes a few weeks before the normal harvest date to perform laboratory-scale pre-harvest fermentations to determine the potential levels of VPs in their wines. They should avoid high guaiacol producing yeast strains, as well as spontaneous alcoholic fermentations. They should consult yeast suppliers to assist them with selecting low volatile guaiacol-producing yeast strains or should consider the following yeasts: UCLM S325, Merit, SIHA A3, AWRI R2 or VIN 13 for white and rosé wines and NT 112, RX60, WE 372, VIN 13, NT 202 or FX10 for red wine production. Production of blanc de noir or rosé wines should be considered, but if red wines are produced, please note that induction of MLF will increase VP levels. All wines (white, rosé or red) should be prepared for consumption within 12 months after bottling. Wines that were acceptable for consumers at bottling could become unacceptable after six months. Smoke-exposed grapes should not be used for the production of premier class wines and smoke-tainted wines should not be blended with non-smoke-tainted wines.
Commercial yeast strains produced variable levels of volatile phenols, which ranged from high, intermediate or low. In this study, wines produced with some yeast strains, as well as spontaneous alcoholic fermentations, contained high guaiacol levels. While lower guaiacol levels were found in wines produced with yeast strains UCLM S325, Merit, SIHA A3, AWRI R2, NT 112, RX60, WE 372, VIN 13, NT 202 and FX10. Induction of malolactic fermentation resulted in wines with higher levels of volatile phenols. Production of blanc de noir or rosé wines from red grapes that were exposed to smoke contained lower levels of volatile phenols than red wines produced with the same grapes or juice. Winemakers should consult yeast suppliers and use yeast strains that produce low guaiacol levels and, if possible, MLF should be avoided in wines produced from smoke-exposed grapes. Wines should be produced for consumption within 12 months after bottling.
The authors thank the ARC and Winetech for funding. All the students and interns that assisted with this project are thanked for their technical assistance. Collaborators from Stellenbosch University are thanked for their contributions.
- Kennison, K.R., Wilkinson, K.L., Williams, H.G., Smith, J.H. & Gibberd, M.R., 2007. Smoke-derived taint in wine: Effect of postharvest smoke exposure of grapes on the chemical composition and sensory characteristics of wine. Journal of Agricultural and Food Chemistry 55, 10897 – 10901.
- Ristic, R., Osidacz, P., Pinchbeck, K.A., Hayasaka, Y., Fudge, A.L. & Wilkinson, K.L., 2011. The effect of winemaking techniques on the intensity of smoke taint in wine. Australian Journal of Grape and Wine Research 17, S29 – S40.
- McKay, M., Bauer, F.F., Panzeri, V., Mokwena, L. & Buica, A., 2019. Profiling potentially smoke tainted red wines: Volatile phenols and aroma attributes. South African Journal of Enology and Viticulture 40, 231 – 236.
– For more information, please contact Heinrich du Plessis at firstname.lastname@example.org.