Chemometric study of Andalusian extra virgin olive oils Raman spectra: Qualitative and quantitative information

Authentication of extra virgin olive oil (EVOO) is an important topic for olive oil industry. The fraudulent practices in this sector are a major problem affecting both producers and consumers. This study analyzes the capability of FT-Raman combined with chemometric treatments of prediction of the fatty acid contents (quantitative information), using gas chromatography as the reference technique, and classification of diverse EVOOs as a function of the harvest year, olive variety, geographical origin and Andalusian PDO (qualitative information). The optimal number of PLS components that summarizes the spectral information was introduced progressively. For the estimation of the fatty acid composition, the lowest error (both in fitting and prediction) corresponded to MUFA, followed by SAFA and PUFA though such errors were close to zero in all cases. As regards the qualitative variables, discriminant analysis allowed a correct classification of 94.3%, 84.0%, 89.0% and 86.6% of samples for harvest year, olive variety, geographical origin and PDO, respectively. [Display omitted] •412 samples of Andalusian Extra virgin olive oils were studied by Raman spectroscopy.•Harvest year was included as a qualification variable.•PLS factors were progressively increased to analyze the evolution in the goodness of the statistical models.•New dimensionless calibration and prediction error measures were proposed.