(Editor’s note: This article is reprinted with permission from the Wine & Viticulture Journal of Australia and New Zealand, where it appeared in the July/August 2012 issue.)

As an investigative story, the hunt for what causes eucalyptus character—and the origin of its aroma compound 1,8-cineole—in wine has the makings of a classic whodunit. The search for the culprit or ally, depending on your preference for or against eucalyptus characters, has thrown up false leads and an unexpected ending. Studying the origin of 1,8-cineole, the Australian Wine Research Institute (AWRI) found that the location and leaves of eucalyptus trees play a direct role in the concentration of 1,8-cineole and occurrence of the “eucalypt,” “fresh” or “minty” characters in wine.

Native to Australia, eucalyptus trees have been planted throughout the world, with large populations of the species now growing in California as well as China, India and Brazil—they live on every continent apart from Antarctica. Hardy and resilient, they grow in a range of different climates and environments, providing raw timber and wood pulp as well as large supplies of eucalyptus essential oil.

It is the oil that matters most to winemakers. Most species of eucalyptus tree contain essential oils in their leaves and, depending on the species, the main component of the oil is a volatile compound called 1,8-cineole, commonly known as eucalyptol. Used as a flavoring agent in a wide range of foods and beverages—as well as being present in a range of therapeutic products—1,8-cineole can also be found in red wine, where it is responsible for characters described as “eucalypt,” “camphor,” “fresh” and “minty.”

For some winemakers these characters are a selling point. Some red wines are well known for their “eucalypt” sensory properties, and the compound responsible is considered a help, not a hindrance to the winemaker’s craft. For other wine producers, however, “eucalypt” characters are something they prefer to avoid—or at the very least limit through effective management strategies. Discovering the source of 1,8-cineole and understanding how it gets into wine has become a detective story: a case that wine scientists have been determined to solve.

Early evidence
For some time, the origin of 1,8-cineole in wine remained a mystery. Scientists had theories, but none were verified: Some researchers believed that “eucalypt” characters were associated with the proximity of vineyards to eucalyptus trees; others proposed that there were compounds in grape berries that acted as precursors for 1,8-cineole.

Further investigations revealed, however, that the precursor proposal did not account for most of the 1,8-cineole found in wine. Research at the AWRI showed that the precursor compounds were unable to generate high enough levels of 1,8-cineole to reach sensory threshold concentrations. Once this potential source was discounted, the AWRI researchers continued to focus on the proximity of eucalyptus trees to vineyards (historically planted as windbreaks) and whether the location of those trees near vines provided a more likely explanation.

The AWRI also compared red and white wines to see whether there was a clear difference between varieties. A survey of 190 commercially available Australian wines found eucalyptol (or 1,8-cineole) in significant amounts in red wine varieties only. The survey led to the daily monitoring of two commercial Shiraz ferments from two different winegrowing regions in South Australia throughout fermentation, revealing a continuous increase in the concentration of 1,8-cineole during fermentation that stopped once the wine was drained from the skins. This indicated that the compound was extracted from the grape skins and/or material other than grapes, commonly known as MOG. How the aroma compound was transferred to grape skins and what is the role of MOG were questions requiring further investigation.

In parallel, consumer studies were carried out by the AWRI sensory team, and they found that overall, participants (104 people) had a slight preference for a wine spiked with 4 micrograms per liter (μg/L) and 30μg/L of 1,8-cineole compared with an unspiked one, with a sizable cluster of consumers (38%) strongly preferring the wine spiked with 30μg/L of 1,8-cineole. Getting the right balance for consumers requires careful management, and to make that happen, winemakers needed to know where the compound 1,8-cineole was coming from. They also needed to know how to control its concentration in wine.

To find out more, the AWRI carried out a detailed study over three vintages to investigate the relationship between grape composition and the proximity of vines to eucalyptus trees. The impact of grape leaves, grape stems and leaves from nearby eucalyptus trees also were included in the investigation. The results of this work provided important information that has the potential to change the way winemakers understand and manage “eucalypt” characters in red wines.

Investigative tools
Key ingredients for the AWRI study were samples of wine, grapes, grape stems and leaves—as well as samples of eucalyptus leaves. Wine samples from Great Southern in Western Australia, Yarra Valley in Victoria and Coonawarra in South Australia were supplied by producers.

Healthy Shiraz grapes were hand-harvested from the Padthaway region of South Australia one day prior to commercial harvest. Samples were taken over three vintages (2008, 2009 and 2011), in the same locations each year. To assess the impact of proximity to eucalyptus trees, three samples of grapes were taken from three separate locations within four different rows of the vineyard (providing 36 samples in all for each vintage). The rows were located at different distances from eucalyptus trees: the first row within about five meters, the furthest around 125 meters from the trees.

Grape leaves also were collected from the same spots in 2009 and 2011, and eucalyptus leaves also were taken from the grapevine canopy of the first row in 2011 for analysis and addition to ferment treatments. Flavor compound traps (consisting of polyethylene sheets) were also installed in a vineyard in 2008 and 2009 to measure airborne 1,8-cineole levels. The samples were supplied, collected and stored according to best practices and then subjected to analysis of 1,8-cineole levels using gas chromatography/mass spectrometry (GC/MS).

Solving the case
The study consisted of a number of stages. In early investigations, wines were made from batches of grapes harvested at set distances from eucalyptus trees in single vineyards in Western Australia and Victoria. The results in the graph 1,8-Cineole Concentration clearly show that the greatest amount of 1,8-cineole was found in wines made from grapes taken from rows closest to the eucalyptus trees. In Victoria, grapes harvested within 50 meters of eucalyptus trees produced wine with a 1,8-cineole concentration of 15.5μg/L, and grapes harvested from rows further away produced a wine with an extremely low 1,8-cineole level of just 0.1μg/L.

In another investigation, wines from consecutive vintages were analyzed from the Coonawarra region in South Australia. The vineyard concerned was in close vicinity of well-established eucalyptus trees. In this case, the wines produced from this vineyard contained relatively high amounts of 1,8-cineole, at 47μg/L (2006 vintage) and 81.5μg/L (2007 vintage), and were considered by the winemaker to display an obvious “eucalypt” character. These lots were not sold commercially and may have been blended with other wine, which is a common practice to adjust and refine sensory attributes. These investigations supported the theory that the presence of 1,8-cineole was likely to be related to eucalyptus trees. Additional vineyard studies were still needed, however, to work out how the compound was transferred from the trees to the vineyard and, ultimately, into wine.

To find out, the AWRI turned its attention to the relationship between grape composition and proximity to eucalyptus trees; this included the analysis of grape berries, stems and leaves. The vineyard chosen for the study had eucalyptus trees close to vines with a history of producing wines with 1,8-cineole concentrations well above sensory threshold levels.

Analyses showed that grape skins contained much higher concentrations of 1,8-cineole than grape pulp, and that grape stems and grape leaves had even higher levels. To confirm that airborne transmission was responsible for the transfer of 1,8-cineole (from eucalyptus trees to the vines located close by) passive traps to capture the volatile aroma compound through adsorption onto polyethylene sheets were placed in the canopy at different locations at set distances from the eucalyptus trees.

Again, the results confirmed previous findings: The closer the traps (and vines) were to eucalyptus trees, the higher the concentration of 1,8-cineole. Leaves from eucalyptus trees themselves also appeared to play a role. When the researchers collected clusters of grapes for the study, they often found eucalyptus leaves lodged in the canopy and within the grape bunches in vines closer to eucalyptus trees. The next step, therefore, was to quantify the effect on 1,8-cineole concentration if eucalyptus leaves found their way into ferments, in the form of MOG in the harvest bin.

Five-hundred-and-fifty kilograms of Shiraz were picked by hand from the rows close to eucalyptus trees, taking special care to avoid MOG. The fruit was randomized and split into separate lots (50kg) for different treatments: one lot was pressed immediately (rosé style); a second lot contained crushed berries only with all grape stems and leaves thoroughly removed (no MOG); a third included grape leaves (500g) and stems (1.3kg) and the final batch included four eucalyptus leaves and a small piece of bark (total weight 3.5g). 1,8-Cineole concentrations were determined daily throughout fermentation.

Again, the results were striking. While the inclusion of grape leaves and stems increased the concentration of 1,8-cineole, adding less than a handful of eucalyptus leaves had the most dramatic effect of all: It increased concentrations of the compound from less than 2μg/L (for the control, i.e., no MOG) to above 30μg/L.

Given the high number of eucalyptus trees in the Australian landscape and the fact that large amounts of eucalyptus leaves can be found naturally in grape bunches—the researchers found 33 eucalyptus leaves in just one 550kg lot of hand-picked fruit—the impact of eucalyptus leaves on wine character cannot be underestimated.

Eucalyptus by design
The results were clear: The presence of eucalyptus leaves—and to a lesser extent grapevine leaves and stems—were key drivers behind concentrations of 1,8-cineole in wine. While there were apparent differences between vintages, the proximity of eucalyptus trees had an obvious effect. The impact of MOG—and eucalyptus leaves in particular—was also very clear.

While not all eucalyptus species have high levels of 1,8-cineole in their leaves, many of Australia’s winegrowing regions’ common trees such as eucalyptus leucoxylon (Yellow Gum) have great potential to affect vineyards. In hindsight, it should not be too surprising that eucalyptus leaves or bark falling from trees can be blown some distance by the wind to lodge in grapevine canopies, and from there be picked with the harvest to affect the wine.

This source had not been previously considered, however, with popular thinking that airborne transfer of the eucalypt essential oil volatiles was probably the main avenue. Even though the leaves are dried and brown within vine canopies, they clearly can influence the character of a wine and are of greater importance to ultimate 1,8-cineole levels in a wine than simple aerial transfer of the volatiles from the trees to the berry skins.

For winemakers, this presents a range of management options in terms of minimizing or maximizing “eucalypt” characters. Wine producers may choose to ferment grapes from vines growing near eucalyptus trees separately and use this wine as a blending option; they can hand pick those rows closest to trees; or they can ensure that minimal MOG is included in machine harvest bins of grapes.

Sorting tables, whether manual or automated, also would be effective but obviously more costly. Adjusting machine harvester settings so that less non-grape material is picked, especially in rows closest to trees, would be another straightforward strategy. By paying closer attention to the volume of grape leaves, stems and eucalyptus leaves or bark in their ferments, winemakers can exert greater control over the wines they are seeking to create.

Overall, the results described here give winemakers practical ways to control 1,8-cineole concentrations throughout vineyard and winery operations. The closeness of grapevines to eucalyptus trees has a conclusive effect on 1,8-cineole concentrations in wine, and the presence of MOG can significantly influence 1,8-cineole levels.

Both factors have a major impact on sensory characteristics. Enhancing or reducing “eucalypt” characters is no longer a case of pure chance or serendipity, and winemakers are in a much stronger position to take greater control of 1,8-cineole and adjust eucalyptus character to create wines that express their terroir with market appeal.

The authors wish to acknowledge numerous Australian wine companies for the generous donations of grape and wine samples, and the contributions of numerous colleagues in particular Drs. Mark Sefton and David Jeffery for their contribution to this work and assistance in the preparation of this article. This work was supported by Australia’s grapegrowers and winemakers through their investment body, the Grape and Wine Research and Development Corp., with matching funds from the Australian government.