Multi-Elemental Analysis for Geographical Tracing of Chickpeas Produced in Nearby Locations Around a Protected...

The multi-elemental profile has repeatedly been proposed as a reliable indicator of the geographical origin of plant-derived foods, as mineral composition accurately reflects the local soil geochemistry and environmental factors. However, this approach may fail in distinguishing specimens from nearby locations, which are expected to be exposed to similar geoclimatic conditions. Herein, we studied 70 chickpea samples collected in four southwestern Spanish provinces, two located within the Protected Geographical Indication 'Garbanzo de Escacena' (i.e., Huelva and Sevilla), as well as other two boundary areas (i.e., Cádiz and Córdoba). Then, inductively-coupled plasma mass spectrometry was employed to simultaneously determine 31 trace elements and 16 rare-earth elements. Interestingly, we found great similarities in the mineral content of chickpeas cultivated in the regions ascribed to the Protected Geographical Indication, but these could be clearly discriminated from the rest of the samples. Afterward, the application of state-of-the-art machine learning tools provided predictive models with good performance in terms of classification accuracy, sensitivity, and specificity. In conclusion, we have demonstrated that the combination of multi-elemental analysis and advanced chemometrics could be a powerful strategy for food authentication and traceability according to the geographical origin.