BIPLOT ANALYSIS APPLIED TO ENOLOGICAL PARAMETERS IN THE GEOGRAPHICAL CLASSIFICATION OF YOUNG RED WINES

Fifteen variables including conventional enological parameters, phenol ics, and color-related parameters were analyzed in 45 young red wines belonging to the Spanish zones Ribera de Duero and Toro with Certified Brand of Origin (CBO). The data obtained were subjected to HJ biplot analysis to determine whether, with the variables employed, it would b e possible to differentiate the wines according to their CBO. Alcohol, fixed acidity, pH, degree of ionization of anthocyanins, total phenol ics, and procyanidins were the variables with the best relative contri butions to discrimination between the two CBO's. A logistic model for estimating the probability of a given sample belonging to a certain CB O is proposed. According this prediction model, 100% of the observatio ns for Ribera de Duero and 81.81% for those of Toro are well classifie d.