Differentiation of ‘two Andalusian DO ‘fino’ wines according to their metal content from ICP-OES by using supervised pattern recognition methods

Abstract The metal content (Ca, Mg, Sr, Ba, K, Na, P, Fe, Al, Mn, Cu and Zn) of several ‘fino’ wines belonging to two Andalusian Denomination of Origin (DO) was determined by ICP-OES. Metal concentrations were selected as chemical descriptors for discrimination, because they play a primary role in the discrimination due to its correlation with soil nature, geographical origin and grape variety. The two studied Andalusian DO were the Jerez-Xeres-Sherry & Manzanilla-Sanlucar de Barrameda (class D) and the Condado de Huelva (class C). Linear Discriminant Analysis (LDA) and procedures based on Artificial Neural Networks (ANN) leads to a perfect separation of classes, especially when applying Multi Layer Perceptrons ANN trained by back-propagation.