Defining regional variability and uniqueness of premium Australian Shiraz
This project is part of a collaborative study with Charles Sturt University and the National Wine and Grape Industry Centre. It will define sensory attributes of wines sourced from multiple regions that contribute to distinctive regional character and will provide objective quality markers for both grapes and wine for future use in vineyard and winery assessment. This project will investigate carefully selected commercially produced wines from premium regions and study the variation in the wines’ sensory properties. Wines of specific criteria will be selected using rapid sensory methods with expert winemaker panels, to characterise the wines from each region and reduce the number of samples to a manageable number for further detailed sensory analysis.
The sensory attributes of the wines from each region will be quantified using sensory descriptive analysis. The associations of the sensory attributes with the concentrations of wine flavour compounds will be determined. The study will allow definition of chemical and sensory attributes which separate wines from the regions, and provide the basis for further investigations into the causes of these differences.
Chemical measures that relate to regional sensory differences
As previously reported, detailed sensory examination of a large number of commercially produced wines from six regions, namely Barossa Valley, McLaren Vale, Heathcote, Yarra Valley, Canberra and the Hunter Valley, was completed. This work involved characterising wines from each region, selecting representative examples, and then quantifying their sensory attributes as a set. There were numerous sensory attributes that separated the wines across the regions. The sensory data were related to almost 70 targeted compositional measures using multivariate regression techniques. The models generated were able to predict most attributes well, with, unusually in this type of exercise, no wines flagged as outliers or as strongly influential samples in the regression models. This reflects the fact that multiple wines representing the range of wine styles from the regions were included, with a good spread across the multivariate space.
Key compounds were identified as major contributors to sensory attributes, some of which had been tentatively identified in previous studies, with others indicated for the first time. Grape-derived compounds such as beta-ionone and beta-damascenone, as well as several fermentation-derived acetate esters, were highlighted as particularly critical to ‘red fruit’ and ‘dark fruit’ characters, while some monoterpenes also featured in the models. Known impact compounds such as isobutylmethoxypyrazine and rotundone were related to ‘green capsicum’ and ‘pepper’ attributes respectively, while mouth-feel and textural characteristics were also modelled well.
Results will be confirmed in a follow-up sensory reconstitution study, and once targets are confirmed, industry practitioners will be able to better control distinctive attributes related to regional differences through viticultural and winemaking adjustments.