Impact of Smoke from Wheat, Oat, and Clover Stubble Burning on Cabernet Sauvignon Grapes and Wine

Shi, Tingting; Ristic, Renata; Wilkinson, Kerry L.

doi:https://doi.org/10.1155/2023/6693220

Australian Journal of Grape and Wine Research

On this page

Abstract Introduction Materials and Methods Results and Discussion Conclusions Data Availability Conflicts of Interest Acknowledgments Supplementary Materials References Copyright Related Articles

Research Article | Open Access

Volume 2023 | Article ID 6693220 | https://doi.org/10.1155/2023/6693220

Impact of Smoke from Wheat, Oat, and Clover Stubble Burning on Cabernet Sauvignon Grapes and Wine

Tingting Shi,¹Renata Ristic,¹and Kerry L. Wilkinson¹

Academic Editor: Bin Tian

Received02 Jun 2023

Revised17 Aug 2023

Accepted19 Sept 2023

Published26 Sept 2023

Abstract

Background and Aims. Stubble burning is an agricultural practice employed by some grain growers to prepare farmland for sowing and/or to control weeds and pests. Grapegrowers and winemakers have questioned whether the resulting smoke can contaminate grapes in nearby vineyards. This study therefore sought to determine the potential for smoke from three different stubble burns to taint grapes and wine. Methods and Results. Excised bunches of mature Cabernet Sauvignon grapes were exposed to smoke from prescribed burning of wheat, oat, and Balansa clover stubble windrows. Environmental sensors monitored the concentration of particulate matter as a measure of smoke density, while chemical and sensory analysis established the extent to which grapes and wine were tainted by smoke. Only grapes that were positioned among the burning wheat windrows or downwind, but in close proximity (∼200 m) to the oat stubble burn, were exposed to sufficient quantities of smoke to result in a detectable concentration of volatile phenols (up to 12 µg/kg), as chemical markers of smoke taint. These grapes yielded wines with two to threefold higher volatile phenol concentrations (up to 18 µg/L) than other wines, including the control wine, and low intensity, but perceptible smoke-related sensory attributes, indicative of low-level smoke taint. Conclusions. Chemical and sensory analyses suggest the risk of smoke taint from stubble burning is low, except where vineyards are immediately downwind and/or prolonged or repeated smoke exposure occurs. However, stubble moisture content and prevailing weather conditions affect smoke density and dispersion, and will therefore affect the potential for smoke contamination by grapes. Significance of the Study. This study will assist colocated grain growers and grape and wine producers to undertake commercial operations, without negatively impacting one another.

1. Introduction

Vineyard exposure to bushfire smoke can lead to “smoke taint” in wine [1–3], i.e., the perception of unpleasant “smoky,” “burnt rubber,” “cold ash” and “medicinal” aromas and flavours, and a “drying” sensation and/or persistent “ashy” aftertaste [4–6]. Economic losses attributed to smoke taint as a consequence of bushfires, including fires that occurred in South Australia, Victoria, Tasmania, and New South Wales during the 2019/2020 wine grape-growing season, have been estimated at several hundred million dollars [1, 7, 8]. While bushfires are a common source of smoke, and therefore cause of smoke taint, vineyard smoke exposure can also arise from prescribed burns, including stubble burning, a practice employed by grain producers to prepare farmland for sowing and/or to control weeds and pests [9, 10]. Consequently, while broadacre farming occurs near wine regions, grapegrowers and winemakers are questioning the potential for smoke from stubble burning to contaminate unharvested grapes.

Stubble refers to the residual stalks or straw that remains in fields after cereal crops have been harvested [10–12]. Agricultural biomass typically comprises varying amounts of lignin, cellulose, and hemicellulose; e.g., wheat straw contains ∼10–25%, ∼30–45%, and ∼20–45% of these macromolecules, respectively [13]. Thermal degradation of lignin, cellulose, and hemicellulose during combustion of plant material results in the emission of volatile organic compounds [14–16], including volatile phenols (guaiacols, cresols, and syringols) which are currently used as compositional markers of smoke taint in grapes and wine in free and glycosylated (bound) forms [17–24]. Smoke arising from the combustion of stubble would similarly be expected to comprise volatile phenols; hence, while stubble burning occurs prior to the completion of harvest, grapes in nearby vineyards could be vulnerable to smoke exposure.

A recent study investigated the uptake of volatile phenols in excised bunches of Cabernet Sauvignon grapes following exposure to smoke during a pea stubble burn [25]. Chemical and sensory data provided evidence of smoke taint when grapes were positioned among the burning pea stubble windrows, but not when grapes were (∼500 m) downwind of the pea paddock. The study concluded that the risk of smoke taint arising from that particular stubble burn was low, due to the rapid dispersion of smoke plumes which resulted in minimal (if any) smoke exposure occurring (at ground level) downwind [25]. This was attributed to the pea stubble burn being undertaken in accordance with guidelines for managing smoke emissions [12, 26]; i.e., dry stubble, raked into windrows, was combusted under favourable prevailing weather conditions, which improved burn efficiency and therefore facilitated smoke dispersion [25].

Previous research has shown that both smoke density and the duration of smoke exposure influence the extent to which grapes are tainted by smoke [25, 27–29]. As such, it is reasonable to expect that where stubble burns are undertaken under suboptimal conditions, e.g., where damp stubble is burned and/or weather conditions are less favourable, more smoke will be produced (due to lower burn temperatures) and/or smoke will linger (due to inadequate dispersion), thereby increasing the risk of smoke exposure (and contamination) of unharvested grapes in nearby (downwind) vineyards. Anecdotal evidence from grain growers also suggests prescribed burning of stubble from different crops might also affect the density of smoke emissions. The current study initially sought to evaluate the potential for grapes to be tainted by smoke arising from prescribed burning of three different stubbles (wheat, oat, and Balansa clover stubble). However, a combination of rainfall the day before stubble burns were carried out and light wind conditions during stubble burning exacerbated smoke emissions and therefore allowed evaluation of the risk of smoke taint arising under suboptimal conditions.

2. Materials and Methods

2.1. Stubble Burn Trials

Field trials involving exposure of excised bunches of Cabernet Sauvignon grapes to smoke from prescribed burning of wheat, oat, and Balansa clover stubble were conducted on a property (Figure 1) located in Hynam (36°56′S, 140°50′E) in the Wrattonbully wine region in South Australia. Prior to burning, the property manager raked the stubble in each of six paddocks into windrows. The paddocks were then burned (in accordance with a fire permit issued by the local council) in succession, starting with the wheat paddocks (burns 1 to 3), followed by the oat paddock (burn 4) and then the clover paddocks (burns 5 and 6), as shown in Figure 1; i.e., all stubble burns occurred on the same day (burn 1 commenced around midday; burn 6 finished after 6 pm). During burns 1, 4, and 6, grape bunches (∼15 kg) were suspended on wire frames at two positions: (i) among the stubble windrows (Figure S1) and ∼100 m upwind in the oat paddock during burn 1, hereafter “wheat A” and “wheat B,” respectively; (ii) ∼200 and ∼600 m downwind in the wheat paddocks during burn 4, hereafter “oats A” and “oats B,” respectively; and (i) in adjacent vineyards during burn 6, hereafter “clover A” and “clover B,” respectively. Grapes (6 parcels of fruit, each ∼15 kg) were hand-harvested (the afternoon before the stubble burns) from a vineyard located on the same property and stored overnight at ambient temperature, with an additional parcel (∼15 kg) harvested during the oat stubble burn as the control. Each fruit parcel comprised bunches remaining after the vineyard had been machine-harvested, and total soluble solids (TSSs) were ∼28°Brix (Table S1). Portable environmental sensors (R9 series, Attentis Pty. Ltd., Cheltenham, Vic., Australia) were positioned alongside the excised grape bunches during each burn to record particulate matter concentrations (i.e., PM_1.0, PM_2.5, and PM₁₀, being particles that are ≤1, 2.5, and 10 µm in size, respectively) as a measure of smoke emissions. PM concentrations were recorded continuously (at ∼1.6 min intervals) and uploaded via Wi-Fi to the manufacturer’s network. Environmental data (rainfall, wind speed and direction, temperature, and humidity) recorded on the day of the stubble burn by a weather station located among vineyards on an adjacent property were provided by the vineyard manager. Rainfall data recorded prior to the stubble burn (by a weather station at the Naracoorte Aerodrome, 12.6 km from Hynam) were sourced from the Australian Bureau of Meteorology (https://www.bom.gov.au) [30].

On completion of each stubble burn trial (typically after ∼1 hour of smoke exposure), grapes (50 berries, in triplicate, chosen randomly) were sampled and stored on ice (and subsequently frozen) for chemical analysis. The remaining fruit was collected (with bunches from each stubble burn and each location kept separate) and stored overnight (at ambient temperature), together with control fruit, prior to transportation to the University of Adelaide’s Waite Campus for small-scale winemaking.

2.2. Winemaking

Parcels of control and smoke-affected grapes were randomly divided into three winemaking replicates (∼5 kg per replicate, per treatment), destemmed, and crushed. Prior to inoculation with 300 mg/L of Saccharomyces cerevisiae EC1118 (Lalvin, Lallemand Australia, Edwardstown, SA, Australia), 50 mg/L of sulfur dioxide (as a 10% solution of potassium metabisulfite) and 200 mg/L of diammonium phosphate were added, and the pH of must adjusted to 3.5 (with a 20% solution of tartaric acid). Musts were fermented on skins at 25°C, with the caps plunged by hand twice daily. Wines were pressed after 7 days of fermentation and then kept at 25°C until residual sugar was <2 g/L (determined enzymatically). Malolactic fermentation was not performed; instead, wines were cold stabilised at 0°C for 2 weeks. Wines were then racked from total lees, and free SO₂ and pH were adjusted to 30 mg/L and 3.5, respectively. Wine samples were taken for chemical analysis, prior to bottling into 375 mL glass bottles under screw cap.

2.3. Chemical Analysis of Grapes and Wine

Juice total soluble solids (TSS, expressed as ^oBrix) and pH were measured (after crushing) using a hand-held digital refractometer (PAL-1, Atago, Tokyo, Japan) and a pH meter (Starter 3100, OHAUS, Port Melbourne, Vic., Australia), respectively. Grape anthocyanins and total phenolics were measured, using published methodology [31]. Basic wine chemistry parameters were determined by the Australian Wine Research Institute’s (AWRI) Commercial Services Laboratory (Adelaide, SA, Australia) and comprised: alcohol, residual sugar, pH, titratable acidity (TA, as g/L of tartaric acid), and volatile acidity (VA, as g/L of acetic acid), which were measured using a FOSS WineScan analyser (Mulgrave, Vic., Australia); and red wine colour and total phenolics, measured using the modified Somers method and methyl cellulose precipitable tannin assay [32], respectively. The AWRI’s Commercial Services Laboratory also measured: volatile phenol concentrations in grapes and wine using an Agilent 6890 gas chromatograph coupled to a 5973 mass selective detector (Agilent Technologies, Forest Hill, VIC, Australia); and volatile phenol glycoconjugates in grapes and wine using an Agilent 1200 high-performance liquid chromatograph equipped with a 1290 binary pump coupled to an AB SCIEX Triple Quad^TM 4500 tandem mass spectrometer, with a Turbo VTM ion source (Framingham, MA, USA); in each case using published stable isotope dilution assay methods [18, 20, 33]. These publications report the preparation of internal standards (d₃-guaiacol, d₃-4-methylguaiacol, d₇-o-cresol, and d₃-syringol for GC-MS analysis and d₃-syringol gentiobioside for HPLC-MS/MS analysis), method validation, and instrumental operating conditions.

2.4. Sensory Analysis of Wine

Sensory evaluations were performed in a purpose-built sensory laboratory under controlled conditions. Control and smoke-affected wine replicates were initially assessed by a panel of three experienced wine researchers from the University of Adelaide to confirm there were no wine faults or obvious differences between replicates. Replicates were then blended, and wine sensory profiles were determined using the rate-all-that-apply (RATA) method [34, 35] and a panel comprising staff and students from the University of Adelaide and the Australian Wine Research Institute, and regular wine consumers (n = 48, 12 male and 36 female, aged between 20 and 68 years). Panellists were given a brief induction about the RATA procedure, including the importance of thoroughly rinsing the palate and resting (for at least 1 min) between samples. Wines (30 mL) were then presented monadically in covered, 4-digit coded 215 mL stemmed glasses, using a randomised order across judges. Panellists rated the intensity of 19 aroma, flavour, taste, and mouthfeel attributes adapted from previous smoke taint studies [6, 29] using 7 point scales (where 0 = “not perceived,” 1 = “extremely low,” and 7 = “extremely high”). Between samples, panellists rested for at least 1 min, with water and plain crackers provided as palate cleansers. Data were acquired with Red Jade software (Redwood Shores, CA, USA). Sensory panellists gave informed consent before participating in the study, which was approved by the Human Research Ethics Committee of the University of Adelaide (Ethics Approval No. H-2019-095).

2.5. Statistical Analysis

Chemical data were analysed by the one-way analysis of variance (ANOVA) using GenStat (19^th Edition, VSN International Limited, Herts, UK). Sensory data were analysed by two-way ANOVA using participants as a random factor and wines as a fixed factor, with Fischer’s LSD post hoc test (P ≤ 0.05), to determine significant differences between wines, using XLSTAT (version 2021.3.1, Addinsoft, New York, NY, USA). Principal component analysis (PCA) of sensory data was performed using XLSTAT, with volatile phenol data overlaid.

3. Results and Discussion

3.1. Prevailing Weather Conditions

During the stubble burn trials, the prevailing weather conditions were mild (Figure 2). Between 12 and 6 pm, ambient air temperature ranged from 11.8 to 14.3°C, wind speed was <1 km/h, and wind gusts fluctuated between 12.3 and 21.9 km/h (averaging 17.2 km/h), from the north to northwest (Figure 1). Humidity was from 54.4 to 67.9%. No rainfall was recorded on the day of the stubble burn, but 8.6 mm was recorded on the preceding day (by a weather station at the Naracoorte Aerodrome [30]). As such, stubble was not dry.

The weather conditions (especially the low wind speed) and combustion of damp stubble likely contributed to the observed density of smoke and PM emissions (Figure 3). Prior to commencement of stubble burning, background PM concentrations were <50 µg/m³. The PM levels recorded by the sensor positioned among the wheat windrows remained low during the first ∼25 min of the wheat stubble burn, because downwind windrows were ignited first. However, elevated PM concentrations were recorded soon after ignition of windrows that were upwind of the sensor (Figure 3(a)). PM₁₀ and PM_2.5 concentrations averaged 472 and 319 µg/m³, respectively, while fruit was deployed, with concentrations up to 882 and 645 µg/m³ being recorded; PM_1.0 concentrations were <150 µg/m³. As a consequence, fruit positioned among the wheat windrows was exposed to smoke for ∼90 min. In contrast, with the exception of one data point, PM concentrations recorded by the sensor upwind of the wheat stubble burn (i.e., in the oat paddock) remained <20 µg/m³ (Figure 3(b)), indicating grapes deployed at this position were not exposed to smoke.

(a)

(b)

(c)

(d)

(e)

(f)

Elevated PM concentrations were also recorded during the oat stubble burn (Figures 3(c) and 3(d)). The sensor located ∼200 m downwind from the oat paddock recorded much higher PM concentrations, being 56–1207 and 38–885 µg/m³ for PM₁₀ and PM_2.5, respectively (and 625 and 432 µg/m³, on average), than the sensor located ∼600 m downwind, which recorded PM₁₀ and PM_2.5 concentrations of 9–369 and 4–328 µg/m³, respectively (and 102 and 85 µg/m³, on average). PM_1.0 concentrations of <92 and 45 µg/m³ were recorded at these positions. These results indicate smoke emissions from the oat stubble burn were higher (i.e., more dense) than those from the wheat stubble burn, but that smoke dispersed as it moved downwind. As such, the grapes deployed during the oat stubble burn were exposed to smoke of different densities, reflecting their proximity to the burn. This is notable, given smoke density influences the uptake of volatile phenols by grapes, and thus, the extent to which smoke taint can subsequently be perceived in wine [25, 29].

The positioning of sensors (and grapes) during the Balansa clover burn was constrained by vineyard plantings and the property boundary. Sensors and grapes were deployed between vineyard blocks east of the clover paddocks (Figure 1), and the presence of smoke was apparent from PM measurements (Figures 3(e) and 3(f)); however, smoke density may have been higher further to the south (i.e., within the direct trajectory of smoke). Nevertheless, the sensors recorded PM₁₀, PM_2.5, and PM_1.0 concentrations of 9–1101, 3–885, and 1–201 µg/m³, respectively (97, 74 and 29 µg/m³, on average) for clover A and 7–512, 3–444 and 1–173 µg/m³, respectively (103, 82 and 38 µg/m³, on average) for clover B. Differences in PM concentrations recorded at clover A vs. clover B likely reflect the relative proximity of each position: to the clover paddock, hence the highest PM concentrations were measured at clover A; and to the smoke trajectory, hence the highest overall PM concentrations were measured at clover B. Smoke emissions measured at these positions were far lower (i.e., less dense) than those recorded among the wheat windrows (wheat A) and ∼200 m downwind of the oat stubble burn (oat A), but comparable to or lower than those recorded ∼600 m downwind of the oat stubble burn (oat B).

Substantially, higher PM levels were reported during the aforementioned pea stubble burn [25] than those recorded in the current study. Among the burning pea stubble windrows, PM₁₀ concentrations exceeded 2000 µg/m³ [25], but only briefly (i.e., for ∼5 min), because combustion of well-cured stubble established convective heat columns that achieved vertical dispersion of smoke plumes (Figure 4(a)). In the current study, combustion of damp stubble, coupled with mild prevailing conditions (i.e., wind speeds of <1 km/hr, Figure 2), resulted in smoke plumes being carried laterally (Figure 4(b)), which prolonged smoke exposure.

(a)

(b)

3.2. Grape and Wine Composition

At harvest, Cabernet Sauvignon grapes comprised TSS ranging from 27.6 to 28.6°Brix, resulting in wines with alcohol concentrations from 15.0 to 16.0% alcohol by volume (abv) (Table S1). The observed variation in fruit maturity, which also resulted in significant differences in grape pH and total phenolics, and wine acidity, colour, and total phenolics (Table S1), reflects the use of fruit remaining after commercial harvest; i.e., fruit sampled from the ends of rows may have been slightly riper than fruit sampled midrow. The increased maturity of control grapes may also reflect this fruit parcel being harvested a day later than other fruit parcels. The observed variation in grape and wine chemistry may have been avoided had fruit been randomised prior to the stubble burn trials, but logistically, this was not practical.

Volatile phenols were measured in grapes and wine as compositional markers of smoke taint (Table 1). In grapes, detectable levels of volatile phenols (i.e., ≥1 µg/kg) were only observed in fruit positioned among the wheat stubble windrows (at wheat A) or ∼200 m downwind from the oat paddock (at oats A), being the positions at which the environmental sensors recorded the highest PM concentrations (Figure 3). Syringol was the most abundant volatile phenol (at 10.0 and 11.7 µg/kg for grapes at wheat A and oats A, respectively), in agreement with previous studies [25, 29], while guaiacol, 4-methylguaiacol, cresols, and 4-methylsyringol were only detected at up to 3.7 µg/kg (Table 1). Wheat A grapes had slightly higher concentrations of guaiacol, 4-methylguaiacol, syringol, and 4-methylsyringol than oats A grapes, which might suggest increased smoke contamination. This likely reflects the relative durations of smoke exposure, i.e., fruit was deployed for ∼89 min during the wheat stubble burn vs. ∼67 min during the oat stubble burn, rather than smoke density, which, based on averaged PM concentrations, was higher at oats A than at wheat B (Figure 3).

The absence of detectable volatile phenols in fruit deployed at oats B, clover A, and clover B (Table 1), despite PM monitoring confirming the presence of smoke at each of these positions (Figure 3), suggests smoke density was not sufficient to cause any contamination. This was further supported by wine volatile phenol data (Table 1). Control wine contained 1 µg/L of guaiacol and 6.3 µg/L of syringol, whereas other volatile phenols were not detected. Wines corresponding to oats B, clover A, and clover B had comparable volatile phenol concentrations, with one exception: the oats B wine contained 1.7 µg/L of guaiacol, being a small, but statistically significant increase in concentration. As expected given wheat B grapes were upwind of the stubble burns, the volatile phenol content of wheat B wine was also equivalent to that of the control wine. In contrast, wheat A and oats A wines had significantly elevated concentrations of guaiacol, o- and m-cresols, syringol, and 4-methylsyringol (Table 1), with syringol again being the most abundant volatile phenol, at 12.3 and 18.0 µg/L, respectively. Interestingly, whereas wheat A grapes had higher volatile phenol levels than oats A grapes, the oats A wine had higher guaiacol and syringol concentrations than the wheat A wine. While these results provide compositional evidence of smoke contamination, it is worth noting that the volatile phenol concentrations observed in wheat A and oats A wines were comparable to those reported in previous studies, for wines that were not perceived to be smoke tainted [25, 29]. Nevertheless, the wheat A and oats A wines did contain several-fold higher volatile phenol concentrations than wines made from grapes exposed to smoke during the pea stubble burn [25]. This suggests the risk of smoke contamination does increase when prescribed burning occurs under suboptimal conditions.

Grape and wine volatile phenol glycoconjugates were also measured, but as expected given the limited time for glycosylation (i.e., <48 hours lapsed between smoke exposure and fermentation), few were detected, and none occurred at concentrations above 8 µg/kg for grapes or 10 µg/L for wine (Table S2). In grapes, small (i.e., <1 µg/kg) but significant differences were observed for 4-methylguaiacol rutinoside and syringol gentiobioside; however, no differences were observed for total grape volatile phenol glycoconjugates. In wine, modest concentration increases (i.e., <14 µg/L) were observed for the total glycoconjugates of several volatile phenols, typically for wheat A and oats A wines (relative to other wines), demonstrating some glycosylation of smoke-derived volatile phenols occurred in fruit between smoke exposure and fermentation.

3.3. Wine Sensory Profiles

Sensory analysis results showed good agreement with PM and volatile phenol data. The control wine and wines corresponding to wheat B, oats B, clover A, and clover B were characterised by fruit aromas and flavours, with comparable intensity ratings for acidity, astringency, and a drying aftertaste (Table S3). Wines made from grapes deployed at wheat A and oats A were perceived to exhibit the most intense smoke, cold ash, and burnt rubber aromas, smoky and burnt rubber flavours, and ashy aftertaste, with the oats A wine also having slightly diminished fruit expression, relative to other wines (Table S3). Although statistically significant differences were observed among several basic wine compositional parameters, including alcohol content, pH, TA, and total phenolics (Table S1), the sensory panel did not perceive significant differences in the acidity, bitterness, hotness, drying aftertaste, or astringency of wines (Table S3).

Principal component analysis (PCA) of wine sensory data (overlaid with wine volatile phenol data) showed clear separation of wheat A and oats A wines from the other wines (Figure 5). The first principal component explained 62.9% of variation and separated wines according to the intensity of cold ash and burnt rubber aroma, smoky and burn rubber flavour, and volatile phenol concentrations. The second principal component explained a further 24.3% of variation, primarily reflecting the intensity of fruit aroma and ashy aftertaste. The positioning of wheat A and oats A wines on the right side of the PCA biplot, with the remaining wines on the left side (Figure 5), provides further evidence that wines made from grapes deployed further away from the stubble burns (i.e., at oats B, clover A, and clover B) were not exposed to smoke of sufficient density, for a sufficient duration of time, to impart a perceptible smoke taint. Nevertheless, it is worth noting that although chemical and sensory data confirm grapes positioned at wheat A and oats A (and therefore, their corresponding wines) were tainted by smoke, the extent to which they were tainted was low, especially when compared with chemical and sensory data reported for smoke-tainted wines in other studies [6, 24, 29]. In the current study, grape bunches were deployed for ∼1 hour, whereas cumulatively, the stubble burns occurred over several hours. As such, it is reasonable to expect that had fruit been deployed for the duration of the stubble burns, and within the direct trajectory of smoke in the case of the clover stubble burn, greater smoke exposure, and thus contamination, may have occurred. As such, had vineyards been located immediately downwind of stubble burns, there may have been some risk of a low, but perceptible level of smoke taint occurring in unharvested grapes. The risk and perceived intensity of smoke taint should decrease as smoke disperses, i.e., as a function of distance from the smoke source. Importantly, any risk of smoke taint would also be alleviated by burning dry stubble, in part, because dry stubble would burn rather than smoulder, yielding less dense smoke, but the increased burn temperatures would also assist vertical smoke dispersion via heat convection. Burning under moderately windy conditions would also be expected to improve smoke dispersion.

4. Conclusions

The potential for grapes to be tainted by smoke from stubble burning depends on both the density of smoke and the duration of smoke exposure. Combustion of windrows comprising dry stubble improves burn temperature and efficiency, thereby mitigating the propensity for stubble to smoulder (increasing particulate emissions) rather than burn, as well as establishing convective heat columns that facilitate vertical smoke dispersion. The prevailing weather conditions also affect smoke production and dispersion; stubble is more likely to smoulder under cool, still, and humid conditions, whereas moderately windy conditions will assist smoke dispersion.

In the current study, combustion of damp stubble coupled with light wind conditions generated smoke plumes that were carried laterally (i.e., close to the ground), such that elevated PM levels (∼800–1200 µg/m³ for PM₁₀) were recorded among burning wheat windrows and ∼200 m downwind from the oat stubble burn. Grapes deployed at these locations were exposed to sufficient quantities of smoke that resulting wines contained detectable levels of volatile phenols and perceptible smoke-related sensory characters, which provided evidence of low-level smoke taint. The composition and sensory profiles of wines made from grapes that were either upwind, some distance downwind, or not in the direct trajectory of smoke from stubble burns were comparable to those of the control wine. This suggests that with sufficient dispersion of smoke, the risk of grapes being tainted significantly diminishes.

Where policy guidelines/codes of practice for managing smoke emissions are followed (especially the combustion of dry stubble), and vineyards are not located immediately downwind, the risk of smoke contamination by grapes is considered to be low. Nevertheless, repeated and/or prolonged smoke exposure may arise where prescribed burning is widespread (i.e., where stubble burns occurs on multiple properties in succession), and this could increase the potential for smoke taint to occur. Grain growers undertaking stubble burning also need to be aware of the weather conditions that lead to temperature inversions, a phenomenon during which warm air covers cooler air, which can trap smoke in low-lying areas, preventing dispersion and thus, potentially increasing the risk of prolonged smoke exposure to any nearby vineyards.

Beyond consideration of combustion efficiency (i.e., stubble moisture content), the area to be burned, and the prevailing weather conditions, there is little that can be done to further mitigate smoke emissions from prescribed burns. Landowners could consider alternative practices to stubble burning, which would also alleviate potential health and environmental impacts, but in some cases, burning is still considered best practice. While stubble burning is deemed to be necessary, communication with local councils, neighbouring landowners, wine industry associations, and/or the Country Fire Service is essential to ensure the risk of smoke taint to unharvested grapes is minimised. In this way, it should be possible for the grains and wine industries to work together to negate untended impacts of each industry’s activities on the other.

Data Availability

The data used to support the findings of this study are included within the article and the supplementary information file.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Acknowledgments

The authors thank: Gary and Trace Burgess from Langkyne Estate for providing access to both grapes and stubble burns; the vineyard manager of the neighbouring property who provided weather data; Attentis Pty. Ltd. and Chris Wilkin for the use of environmental sensors and invaluable technical support; the AWRI’s Commercial Services Laboratory for grape and wine compositional analysis; and the panellists who participated in sensory analysis. This research was funded by the Department of Primary Industries and Regions South Australia as part of the South Australian Wine Industry Development scheme. Tingting Shi thanks the University of Adelaide’s School of Agriculture, Food, and Wine for an Honours scholarship.

Supplementary Materials

The following data are provided as supplementary material: Table S1: Basic composition of control and smoke-affected Cabernet Sauvignon grapes and wine. Table S2: Concentration of volatile phenol glycoconjugates in control and smoke-affected Cabernet Sauvignon grapes (µg/kg) and wine (µg/L). Table S3: Mean intensity ratings for the sensory attributes of control and smoke-affected Cabernet Sauvignon wine. Figure S1: Grape bunches (suspended on wire frames) deployed alongside environmental sensors (a) among the wheat stubble windrows (i.e., at wheat A) and (b) ∼100 m upwind in the oat paddock (i.e., at wheat B). (Supplementary Materials)

References

M. P. Krstic, D. L. Johnson, and M. J. Herderich, “Review of smoke taint in wine: smoke-derived volatile phenols and their glycosidic metabolites in grapes and vines as biomarkers for smoke exposure and their role in the sensory perception of smoke taint,” Australian Journal of Grape and Wine Research, vol. 21, pp. 537–553, 2015.
View at: Publisher Site | Google Scholar
Y. A. Mirabelli-Montan, M. Marangon, A. Graça et al., “Techniques for mitigating the effects of smoke taint while maintaining quality in wine production: a review,” Molecules, vol. 26, no. 6, p. 1672, 2021.
View at: Publisher Site | Google Scholar
V. Summerson, C. Gonzalez Viejo, A. Pang, D. D. Torrico, and S. Fuentes, “Review of the effects of grapevine smoke exposure and technologies to assess smoke contamination and taint in grapes and wine,” Beverages, vol. 7, no. 1, p. 7, 2021.
View at: Publisher Site | Google Scholar
K. R. Kennison, M. R. Gibberd, A. P. Pollnitz, and K. L. Wilkinson, “Smoke-derived taint in wine: the release of smoke-derived volatile phenols during fermentation of Merlot juice following grapevine exposure to smoke,” Journal of Agricultural and Food Chemistry, vol. 56, no. 16, pp. 7379–7383, 2008.
View at: Publisher Site | Google Scholar
M. Parker, P. Osidacz, G. A. Baldock et al., “Contribution of several volatile phenols and their glycoconjugates to smoke-related sensory properties of red wine,” Journal of Agricultural and Food Chemistry, vol. 60, no. 10, pp. 2629–2637, 2012.
View at: Publisher Site | Google Scholar
R. Ristic, A. L. Fudge, K. A. Pinchbeck et al., “Impact of grapevine exposure to smoke on vine physiology and the composition and sensory properties of wine,” Theoretical and Experimental Plant Physiology, vol. 28, no. 1, pp. 67–83, 2016.
View at: Publisher Site | Google Scholar
J. Whiting and M. P. Krstic, in Understanding the Sensitivity to Timing and Management Options to Mitigate the Negative Impacts of Bush Fire Smoke on Grape and Wine Quality– Scoping Study, Department of Primary Industries: Knoxfield, Vic, Australia, 2007.
M. Walpole, “The financial impact of 2020 summer bushfire smoke on the wine regions of north east Victoria,” 2020, https://www.tafco.com.au/farmsmart/r-d/143-bushfire-2020.html.
View at: Google Scholar
M. Walsh and P. Newman, “Burning narrow windrows for weed seed destruction,” Field Crops Research, vol. 104, no. 1-3, pp. 24–30, 2007.
View at: Publisher Site | Google Scholar
M. I. Abdurrahman, S. Chaki, and G. Saini, “Stubble burning: effects on health and environment, regulations and management practices,” Environmental Advances, vol. 2, Article ID 100011, 2020.
View at: Publisher Site | Google Scholar
B. J. Scott, P. L. Eberback, J. Evans, and L. J. Wade, “Graham centre monograph No. 1: “stubble retention in cropping systems in southern Australia: benefits and challenges,” in Industry and Investment: Orange, pp. 1–105, NSW, Australia, 2010.
View at: Google Scholar
New South Wales Environment Protection Auhority, “Local government air quality toolkit. Module 3: guidelines for managing air pollution,” 2021, https://www.epa.nsw.gov.au/your-environment/air/air-nsw-overview/local-government-air-quality-toolkit.
View at: Google Scholar
L. Zhang, A. Larsson, A. Moldin, and U. Edlund, “Comparison of lignin distribution, structure, and morphology in wheat straw and wood,” Industrial Crops and Products, vol. 187, Article ID 115432, 2022.
View at: Publisher Site | Google Scholar
R. J. Evans and T. A. Milne, “Molecular characterization of the pyrolysis of biomass,” Energy and Fuels, vol. 1, no. 2, pp. 123–137, 1987.
View at: Publisher Site | Google Scholar
P. J. Crutzen and M. O. Andreae, “Biomass burning in the tropics: impact on atmospheric chemistry and biogeochemical cycles,” Science, vol. 250, no. 4988, pp. 1669–1678, 1990.
View at: Publisher Site | Google Scholar
R. Wittkowski, J. Ruther, H. Drinda, and F. Rafiei-Taghanaki, “Formation of smoke flavor compounds by thermal lignin degradation,” in Flavour Precursors: Thermal and Enzymatic Conversions, R. Teranishi, G. R. Takeoka, and M. Güntert, Eds., pp. 232–243, American Chemical Society, Washington, DC, USA, 1992.
View at: Google Scholar
Y. Hayasaka, K. A. Dungey, G. A. Baldock, K. R. Kennison, and K. L. Wilkinson, “Identification of a β-D-glucopyranoside precursor to guaiacol in grape juice following grapevine exposure to smoke,” Analytica Chimica Acta, vol. 660, no. 1-2, pp. 143–148, 2010.
View at: Publisher Site | Google Scholar
Y. Hayasaka, G. A. Baldock, M. Parker et al., “Glycosylation of smoke-derived volatile phenols in grapes as a consequence of grapevine exposure to bushfire smoke,” Journal of Agricultural and Food Chemistry, vol. 58, no. 20, pp. 10989–10998, 2010.
View at: Publisher Site | Google Scholar
K. L. Wilkinson, R. Ristic, K. A. Pinchbeck et al., “Comparison of methods for the analysis of smoke related phenols and their conjugates in grapes and wine,” Australian Journal of Grape and Wine Research, vol. 17, no. 2, pp. S22–S28, 2011.
View at: Publisher Site | Google Scholar
Y. Hayasaka, M. Parker, G. A. Baldock et al., “Assessing the impact of smoke exposure in grapes: development and validation of a HPLC-MS/MS method for the quantitative analysis of smoke-derived phenolic glycosides in grapes and wine,” Journal of Agricultural and Food Chemistry, vol. 61, no. 1, pp. 25–33, 2013.
View at: Publisher Site | Google Scholar
M. Noestheden, K. Thiessen, E. G. Dennis, and B. Tiet, “Quantitating organoleptic volatile phenols in smoke-exposed Vitis vinifera berries,” Journal of Agricultural and Food Chemistry, vol. 65, no. 38, pp. 8418–8425, 2017.
View at: Publisher Site | Google Scholar
M. Noestheden, E. G. Dennis, and W. Zandberg, “Quantitating volatile phenols in Cabernet Franc berries and wine after on-vine exposure to smoke from a simulated forest fire,” Journal of Agricultural and Food Chemistry, vol. 66, no. 3, pp. 695–703, 2018.
View at: Publisher Site | Google Scholar
P. Crews, P. Dorenbach, G. Amberchan et al., “Natural product phenolic diglycosides created from wildfires, defining their impact on California and Oregon grapes and wines,” Journal of Natural Products, vol. 85, no. 3, pp. 547–561, 2022.
View at: Publisher Site | Google Scholar
J. W. Favell, K. Wilkinson, I. Zigg et al., “Correlating sensory assessment of smoke-tainted wines with inter-laboratory study consensus values for volatile phenols,” Molecules, vol. 27, no. 15, p. 4892, 2022.
View at: Publisher Site | Google Scholar
K. Wilkinson, R. Ristic, I. McNamara, B. Loveys, W. W. Jiang, and M. Krstic, “Evaluating the potential for smoke from stubble burning to taint grapes and wine,” Molecules, vol. 26, no. 24, p. 7540, 2021.
View at: Publisher Site | Google Scholar
South Australian Country Fire Service, “Broad acre burning code of practice,” 2015, https://ablis.business.gov.au/service/sa/broad-acre-burning-code-of-practice/46764.
View at: Google Scholar
K. R. Kennison, K. L. Wilkinson, A. P. Pollnitz, H. G. Williams, and M. R. Gibberd, “Effect of timing and duration of grapevine exposure to smoke on the composition and sensory properties of wine,” Australian Journal of Grape and Wine Research, vol. 15, no. 3, pp. 228–237, 2009.
View at: Publisher Site | Google Scholar
K. R. Kennison, K. L. Wilkinson, A. P. Pollnitz, H. G. Williams, and M. R. Gibberd, “Effect of smoke application to field-grown Merlot grapevines at key phenological growth stages on wine sensory and chemical properties,” Australian Journal of Grape and Wine Research, vol. 17, no. 2, pp. S5–S12, 2011.
View at: Publisher Site | Google Scholar
C. Szeto, R. Ristic, D. Capone et al., “Uptake and glycosylation of smoke-derived volatile phenols by Cabernet Sauvignon grapes and their subsequent fate during winemaking,” Molecules, vol. 25, no. 16, p. 3720, 2020.
View at: Publisher Site | Google Scholar
Australian Bureau of Meteorology, “Daily rainfall at Naracoorte Aerodrome,” 2021, http://www.bom.gov.au/jsp/ncc/cdio/weatherData/av.
View at: Google Scholar
P. Iland, N. Bruer, E. Wilkes, and G. Edward, “Anthocyanins (colour) and total phenolics of grape berries,” in Chemical Analysis of Grapes and Wine: Techniques and Concepts, p. 44, Winetitles, Broadview, Australia, 1 edition, 2004.
View at: Google Scholar
M. D. Mercurio, R. G. Dambergs, M. J. Herderich, and P. A. Smith, “High throughput analysis of red wine and grape phenolics – adaptation and validation of methyl cellulose precipitable tannin assay and modified Somers color assay to a rapid 96 well plate format,” Journal of Agricultural and Food Chemistry, vol. 55, no. 12, pp. 4651–4657, 2007.
View at: Publisher Site | Google Scholar
A. P. Pollnitz, K. H. Pardon, M. Sykes, and M. A. Sefton, “The effects of sample preparation and gas chromatograph injection techniques on the accuracy of measuring guaiacol, 4-methylguaiacol and other volatile oak compounds in oak extracts by stable isotope dilution analyses,” Journal of Agricultural and Food Chemistry, vol. 52, no. 11, pp. 3244–3252, 2004.
View at: Publisher Site | Google Scholar
G. Ares, F. Bruzzone, L. Vidal et al., “Evaluation of a rating-based variant of check-all-that-apply questions: rate-all-that-Apply (RATA),” Food Quality and Preference, vol. 36, pp. 87–95, 2014.
View at: Publisher Site | Google Scholar
L. Danner, A. M. Crump, A. Croker, J. M. Gambetta, T. E. Johnson, and S. E. P. Bastian, “Comparison of rate-all-that-apply and descriptive analysis for the sensory profiling of wine,” American Journal of Enology and Viticulture, vol. 69, no. 1, pp. 12–21, 2018.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2023 Tingting Shi et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies