The Influence of Grape Maturity on Smoke Taint in Wine

Grapes that are exposed to the smoke from wildfires often possess what’s referred to as “smoke taint”: highly undesirable smoky, ashy, or medicinal characteristics that drive quality as well as price downward. This is a phenomenon known all too well in Australia and California, with other grape growing regions such as South Africa, Croatia, Washington state, and others occasionally bearing witness to such catastrophic events.

Research has found links to several compounds in wine to the presence of smoke taint, including guaiacol, 4-methylguaiacol, syringol, 4-methylsyringol, p-cresol, m-cresol, and o-cersol. More recent research has further discovered that it is the glycoconjugate forms of these compounds that when metabolized during fermentation release volatile compounds that give off the trademark smoke taint flavor and aromatic characteristics in contaminated wines.

In terms of the degree of smoke taint present in wines, the timing of the smoke exposure plays a major role, while at the same time so does winemaking techniques, the type of yeasts used, and the use of oak barrels also play a significant role in the amount of smoke taint present in the contaminated wine.

While there is a lot known about smoke taint in wines, it is still very much an active area of research and

Photo Nick carson at en.wikipedia [CC BY 3.0 (http://creativecommons.org/licenses/by/3.0)], from Wikimedia Commons

Photo Nick carson at en.wikipedia [CC BY 3.0 (http://creativecommons.org/licenses/by/3.0)], from Wikimedia Commons

there is still much that is not understood. For example, while it is very well understood how grape maturity affects the aroma, flavor, and overall quality of a wine, it is not known how grape maturity affects smoke taint aromas and flavors in contaminated wines (if it even does).

Therefore, a new study published in the journal Molecules aimed to add to the growing knowledge of smoke taint in wines by examining the effect of grape maturity on smoke taint in wines.

Brief methods (ask for more details if you’d like them)

This was a controlled study in an experimental vineyard at the University of Adelaide, Waite Campus, in Adelaide, South Australia.

Chardonnay, Sauvignon Blanc, Merlot, and Shiraz grapevines were placed in either one of two treatments: smoke or control (no smoke). Vines undergoing the smoke treatment were surrounded by “smoke tents” built for the study, and were exposed to one hour of smoke 7 days after veraison (the start of ripening).

Grapes were harvested at two different time points: the first at a ripeness level appropriate for making sparkling wines or light-bodied wines (TSS 16-20oBrix), and the second at a ripeness level appropriate for full-bodied wines (TSS 22-25oBrix).

Wines were made from these grapes without undergoing malolactic fermentation. For each variety, two wines were made: 1) smoked and 2) control.

The following chemical analyses were made on all wines: pH, titratable acidity, volatile acidity, ethanol content, wine color density, hue, total phenolics, volatile phenols, guaiacol, 4-methylguaiacol, syringol, 4-methylsyringol, and total cresols.

A descriptive sensory analysis was performed by a panel made up of staff and students (10 female, 3 male) from the University of Adelaide and the Australian Wine Research Institute who all (except one) had experience with descriptive analyses of smoke tainted wines.

Training sessions were performed to make sure everyone was on the “same page” with the analysis, with the actual experimental sessions occurring in isolated booths. Panelists sampled and evaluated three replicates for each wine/smoke treatment.

Selected Results

  • There were no differences in TSS (total soluble solids – i.e. sugar accumulation) in wines made from control and smoke-exposed grapes at either harvest date from grapes of the same
  • Grape berry weight was not affected by smoke exposure or fruit maturity in Chardonnay, Shiraz, or Merlot.
  • Grape berry weight was lower for Sauvignon Blanc at the earlier harvest date compared with the later harvest date.
  • Alcohol levels were different by up to 1.1% abv and total phenolics were higher in smoke-exposed Chardonnay and Sauvignon Blanc wine.
  • pH levels were not shown to be significantly different between treatments of any grape varieties and the wines made from those grapes.
  • There was no effect of smoke exposure on total phenolics in the red wines studied.
  • Wine hue and color density were not affected by smoke exposure.
  • Volatile acidity levels were affected by maturity and smoke exposure, though all were below levels accepted in Australian wines, thus assumed to not play a role in the sensory profile of the finished wines.
  • Proline levels were not affected by smoke exposure.
  • No smoke taint associated volatile compounds were found in the control Chardonnay or Sauvignon Blanc wines.
  • Guaiacol, syringol, and cresols were found in control Merlot (up to 3μg/L) and Shiraz (up to 11μg/L) wines, though it is important to note that these compounds are known to be present naturally in some red wines.
  • Significantly higher smoke taint associated volatile compounds were found in smoke-exposed wines.
    • Smoke-exposed Chardonnay wines had up to 3ug/L guaiacol and 5ug/L syringol at the earlier harvest date.
    • Smoke-exposed Sauvignon Blanc wines had up to 4ug/L guaiacol and 6ug/L syringol at the earlier harvest date.
    • 4-methylguaiacol and cresols were found in smoke-exposed Sauvignon Blanc wines from the earlier harvest date.
    • Smoke-exposed Merlot wines had up to 8ug/L guaiacol and 15ug/L syringol.
    • Smoke-exposed Shiraz wines had up to 18ug/L guaiacol and 28ug/L syringol.
  • For Chardonnay and Sauvignon Blanc wines, smoke taint associated volatile compounds were higher at the earlier harvest date compared with the later harvest date.
  • For Merlot and Shiraz wines, smoke taint associated volatile compounds were higher at the later harvest date compared with the earlier harvest date.

Sensory Anaylsis

  • Earlier harvest Chardonnay:
    • No smoke taint characteristics noted in control or smoke-exposed wines.
  • Later harvest Chardonnay:
    • Smoke taint characteristics noted: “smoke”, “cold ash”, “woody”, “smoky flavor”, and “ashy aftertaste”.
  • Earlier harvest Sauvignon Blanc:
    • Smoke taint characteristics noted: “smoke”, “cold ash”, “smoky flavor”, and “ashy aftertaste”.
  • Later harvest Sauvignon Blanc:
    • Smoke taint characteristics noted: “smoke” and “ashy aftertaste”.
  • Red wines were significantly more affected by smoke taint associated aromas and flavors than white wines.
  • Earlier harvest Merlot:
    • Smoke taint characteristics noted: “smoke”, “cold ash”, “smoky flavors”, “medicinal”, “ashy aftertaste”, and “diminished fruit aroma and flavor”.
  • Later harvest Merlot:
    • Intensity of all smoke taint characteristics increased for the later harvest wines compared with the earlier harvest wines.
  • Earlier harvest Shiraz:
    • Smoke taint characteristics noted: moderate “smoke”.
  • Later harvest Shiraz:
    • Smoke taint characteristics noted: moderate “smoky flavor” and “ashy aftertaste”.
  • The earlier harvest Sauvignon Blanc was considered the most tainted white wine by sensory analysis.
  • The earlier harvest Merlot was considered the most tainted red wine by sensory analysis.
  • The relative abundance of some volatile compounds (ex. isobutyl alcohol, ethyl butanoate, ethyl 2-methylbutanoate, and 2-phenylethyl acetate) were significantly different between the earlier harvest and the later harvest in wines.
    • Specifically, relative abundance of these compounds were higher in Sauvignon Blanc wines at the later harvest, thus potentially explaining the decreased perception of smoke taint in these later harvest wines.
    • The same principle did not apply to Shiraz, as even though specific esters were increased in the later harvest wines, the perception of smoke taint was still higher in the later harvest wines compared with earlier harvest wines.

Conclusions

The results of this study seem to indicate that fruit maturity does play a role in the presence and perception of smoke taint in wines, though this effect appears to be variety-dependent.

Specifically, for Chardonnay wines, there did not appear to be any smoke influence on the sensory characteristics in the earlier harvest wines, while significantly more smoke taint characteristics were noted in the later harvest wines. This result is interesting considering the fact that smoke taint associated volatile

Photo by Flickr user DVIDSHUB (https://www.flickr.com/photos/dvids/5664249483/)

Photo by Flickr user DVIDSHUB (https://www.flickr.com/photos/dvids/5664249483/)

compounds were higher in Chardonnay at the earlier harvest date compared with the later harvest date. This seems to suggest some sort of “dampening” of smoke taint characteristics by other compounds present in the Chardonnay at an earlier maturity level, though this is just conjecture and would need to be studied further.

For Sauvignon Blanc wines, the influence of smoke on the sensory characteristics in the earlier harvest wines appeared to be greater than in the later harvest wines, which is supported by the chemical analysis showing that like Chardonnay, smoke taint associated volatile compounds were higher at the earlier harvest date compared with the later harvest date in Sauvignon Blanc. Sauvignon Blanc seemed to be the white wine most affected by smoke taint in this analysis overall.

For Merlot and Shiraz wines, the influence of smoke exposure was greater overall compared with the white wines. This makes sense since many of the same compounds associated with smoke taint are naturally present in the skins of grapes, and since red wine typically experience greater contact with the skins during the winemaking process, thereby allowing greater opportunity for increased extraction of these compounds into the wine. For both Merlot and Shiraz, smoke taint characteristics were more noticeable in the later harvest wines compared with the earlier harvest wines, which correspond well with the chemical analysis results.

The current practice for testing grapes for smoke taint in Australia suggests that grapes be sampled as soon as possible after smoke exposure, conducting small lot fermentations, and running chemical and/or sensory analyses on the finished wine. The issue with this standard practice is that it may result in some smoke tainted wines being missed. This can be seen with the Chardonnay example from this study—if the grapes in this study were tested early on, and assuming the results of this study are in fact repeatable, then the Chardonnay would show up as not being affected. However, after leaving the remaining grapes on the vine until a more typical harvest date, the wine would actually be influenced by smoke taint associated characteristics, which was unfortunately not caught earlier on (“false-negative”).

More research needs to be done to first confirm or refute the results of this study, and then more to further evaluate the effects of grape maturity levels on smoke taint in many different wines, particularly those that appear to produce these “false negative”-like results, to potentially improve upon the current standards for evaluating possible smoke taint contamination in wines exposed to the smoke from wildfires.

Source:

Ristic, R., Boss, P.K., and Wilkinson, K.L. 2015. Influence of fruit maturity at harvest on the intensity of smoke taint in wine. Molecules 20: 8913-8927.

1 comment for “The Influence of Grape Maturity on Smoke Taint in Wine

  1. July 23, 2015 at 10:47 am

    Fascinating stuff, as always.

Comments are closed.