Nowadays, the quality of food products is strongly linked, among other factors, to their geographical origin, leading to the implementation of quality schemes such as the Protected Designation of Origin (PDO), enhancing the value of the products from certain regions. Given the differences in quality and prices of several food products, cases of fraud related to geographical origin mislabeling are a growing concern. In the particular scenario of honey, due to its high value, fraud cases arise from price differences between products of different origins and qualities. For this reason, in order to ensure its high quality and prevent fraud cases, it is important to recognize the differences in the chemical attributes of honey from different origins.

In this work, the characterization and classification of 68 honey samples from eight countries and diverse botanical origins (including acacia, chestnut, and eucalyptus, among others) relied on physicochemical parameters (pH, conductivity, sugar and water contents), spectrophotometric indexes (total phenolic content, ferric reducing antioxidant power, and 3,5-dinitrosalicylic acid assay), and LC-MS/MS polyphenolic profiling. The data generated in the studies were analyzed statistically (ANOVA and t-test) and with exploratory and classification methods (PCA and PLS-DA).

The statistical and chemometric analyses helped to identify different trends among the sample groups. Good classification error values were obtained with the parameters studied, especially with LC-MS/MS data, obtaining cross-validated values of classification error lower than 34.4%. Therefore, the variables studied in the present contribution were effective in distinguishing honey from diverse geographical origins, which will be helpful when addressing authentication issues.