Multivariate classification of rosé wines from different Spanish protected designations of origin Original Research Article

Wine composition depends on many factors which are especially important for quality wines from specific regions, such as protected designation of origin (PDO) wines. Nineteen analytical parameters were analysed in commercial rosé wines from different Spanish PDOs (Ribera del Duero (RD), Rioja (Rj), Valdepeñas and La Mancha (M-V)), and from two consecutive vintages. Stepwise linear discriminant analysis (SLDA) was used to differentiate and classify these wines by their geographical origin. All the wines were made from “similar” varieties of grapes. The final model selected 12 variables, being ethanol and calcium the most important variables for the differentiation of the three PDOs. The evaluation of the model was done by crossvalidation, obtaining a global percentage of correct classification of 98.8% and of global prediction of 97.3%. These results can be considered satisfactory and acceptable, and the selected variables can be useful to differentiate these wines by their origin.