Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Journal article
Title: Sensory typicity of regional Australian Cabernet Sauvignon wines according to expert evaluations and descriptive analysis
Author: Souza Gonzaga, L.
Capone, D.
Bastian, S.
Danner, L.
Jeffery, D.
Citation: Food Research International, 2020; 138(Pt A):109760-1-109760-9
Publisher: Elsevier
Issue Date: 2020
ISSN: 0963-9969
Statement of
Lira Souza Gonzaga, Dimitra L. Capone, Susan E.P. Bastian, Lukas Danner, David W. Jeffery
Abstract: The concept of wine typicity has been an important tool for the international wine trade, and especially for Old World wine producing countries, where provenance criteria are regulated and act as a quality indicator. Provenance in Australia is governed by Geographical Indications, for which typicity should also be evident in terms of regional sensory profiles of wine from a given grape cultivar. Two approaches were used to identify sensory drivers for regional typicity of commercial Cabernet Sauvignon wines from three Australian regions, namely Coonawarra, Margaret River, and Yarra Valley. Cabernet Sauvignon-dominant wines from Bordeaux were also assessed for benchmarking purposes. A set of 84 wines underwent a sorting task and rate-all-that-apply (RATA) analysis of the sorted groups with an expert panel. Agglomerative hierarchical clustering of the sorting task data did not show a clear regional driver upon separating the samples into four main clusters, although certain sensory traits could be associated with the different clusters. On the other hand, canonical variate analysis (CVA) of the group-RATA results indicated several sensory drivers for the separation between the regions, such as ‘mint’ and ‘dark fruits’ being important for Coonawarra wine profiles, ‘floral’ and ‘green pepper’ for Margaret River, ‘stemmy’ for Yarra Valley, and ‘barnyard’ and ‘savoury’ differentiating Bordeaux wines from the other regions. A subset (n = 52) of wines was selected for further evaluation by descriptive analysis with a trained panel. Statistical evaluation with CVA revealed similar results to the expert evaluation, with Bordeaux wines showing more dissimilarity when compared to Australian regions, and having ‘savoury’ and ‘earthy’ as significant characters. The results also demonstrated that ‘mint’ and ‘Mallee leaf’ were relevant characters for Coonawarra regional sensory profile, ‘violets’ and ‘red fruits’ for Margaret River, and ‘cooked vegetables’ for Yarra Valley. Analysing both data sets (expert RATA and DA) revealed some agreement between the sets of results for attributes such as ‘mint’, ‘cooked vegetables’, ‘floral’, green-related characters, and oak characters like ‘vanilla’ and ‘chocolate’. Overall, experts and trained panellists were able to distinguish regions based on a few characteristic sensory traits.
Keywords: Regionality; typicality; clustering; rate-all-that-apply; expert panel; sensory assessment; sorting task
Rights: © 2020 Elsevier Ltd. All rights reserved.
DOI: 10.1016/j.foodres.2020.109760
Grant ID:
Appears in Collections:Agriculture, Food and Wine publications
ARC Training Centre for Innovative Wine Production publications
Aurora harvest 4

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.