Differentiation of port wines by appearance using a sensory panel: comparing free choice and conventional profiling

Citation
E. Cristovam et al., Differentiation of port wines by appearance using a sensory panel: comparing free choice and conventional profiling, EUR FOOD RE, 211(1), 2000, pp. 65-71
Citations number
15
Categorie Soggetti
Food Science/Nutrition
Journal title
EUROPEAN FOOD RESEARCH AND TECHNOLOGY
ISSN journal
14382377 → ACNP
Volume
211
Issue
1
Year of publication
2000
Pages
65 - 71
Database
ISI
SICI code
1438-2377(2000)211:1<65:DOPWBA>2.0.ZU;2-W
Abstract
Appearance in port wines is often a key quality parameter. Although expert tasters can determine age through subjective colour judgements, it is not c lear how meaningful this is to individuals. Twelve commercial ports were as sessed for sensory quality, using free-choice profiling (FCP) employing a p anel of 18 subjects. Individual vocabularies ranged from 2 to 11 appearance attributes. The consensus product space obtained by generalised Procrustes analysis, showed differentiation of ports mainly on two factors, on the ba sis of colour related to port style and length of wood maturation. These un ique sets of descriptors were used to develop a formal consensus vocabulary for conventional profiling of retailed ports. In a second experiment 24 co mmercial ports were assessed for sensory quality using this consensus vocab ulary, employing 12 individuals, from the original panel. The consensus pro duct space, obtained using principal component analysis, showed differentia tion of both the 12 original ports and the extended set of 24 ports again o n the basis of colour related to port style and length of wood maturation. FCP generated three significant factors explaining 69% of the sensory varia nce while conventional profiling generated four, explaining a total of 99%. The trained panel using the consensus vocabulary was therefore more discri minative, It was also possible to derive model appearance profiles for thre e styles of port from this strategy.