RESUMEN
Monovarietal wines produced in different wine-growing areas may have completely different sensory profiles. As a result, they may be suitable for sale in different regions, depending on local preferences. Better insight into the sensory and chemical profiles of these wines can be helpful in further optimizing commercial strategies and matching supply and demand, which is the main challenge for global wine traders. The training of dedicated sensory panels, together with the correlation of the evaluated attributes with chemical parameters, followed by validation of the obtained models, may yield an improved picture of the overall features associated with products from a specific region. Eighteen samples of international Chardonnay and eighteen samples of international Sauvignon Blanc wines were collected from nine world origins (Northern Italy, Southern Italy, Chile, Argentina, New Zealand, Australia, and South Africa). The overall quality judgement (OQJ) and the sensory attributes were evaluated by a panel trained with a MRATA (Modified Rate-All-That-Apply) method. Moreover, volatile compounds were analysed by HS-SPME-GC × GC-ToF/MS and the phenolic composition, including proanthocyanidins, was determined using HPLC-QqQ/MS. The processing of the data using different multivariate analysis methods, such as multiple factor analysis (MFA), was essential to gain insight into the quality of the samples. The profile of cyclic and non-cyclic oligomeric proanthocyanidins was found to be substantially dependent on the grape variety used in the wines (varietal markers), despite the country of origin of the wine influencing it to a limited extent. The results from the same samples analysed by a sensory panel from Germany and ours were qualitatively compared, highlighting the presence of potential factors inherent to the panels themselves that could influence the different judgments and quality classification of the wines. Consequently, the combination of sensory and chemical analysis, by means of the application of multivariate statistical methods presented in this study proves to be a powerful tool for a deeper and more comprehensive understanding of the quality of the wines under investigation. Overall quality was described as a combination of the sensory attributes, according to the perception process. The attributes were in turn described based on the chemical profiles, which were determined independently by analytical techniques. Eventually, this approach can be very useful not only for basic research on wine quality but also as a tool to aid business-related decision-making activities of wineries and wine traders and to create models that can aid the refinement of marketing strategies.