Multielement analysis and chemometric methods were proposed to discriminate the Sorrento lemon (PGI) juices according to geographical origin. In 2018 and 2019, 169 fruits from three farms in PGI area and two in not-PGI area were collected and analysed for essential and not-essential elements by ICP-MS. The PCA of multielement fingerprinting grouped lemon juices from PGI farms revealing a strong differentiation at small geographical scale. The S-LDA discriminated lemon juices for Mo, Ba, Rb, Mg, Co, Ca, Fe, Sr on the two production years, giving 97.7% correct classification, 98.5% accuracy and 93.8% external validation. The good correlation lemon juice vs cultivation soil and the soil discrimination by not-essential elements suggested the use of these elements as reliable indicators of lemon juice provenances. Despite lowering the number of variables, constituted by not-essential elements Ba, Rb, Ti, Co, the use of S-QDA discriminated the lemons juices with 87.5% accuracy and 83.9% validation.
Ruggiero, L., Fontanella, M. C., Amalfitano, C., Beone, G. M., Adamo, P., Provenance discrimination of Sorrento lemon with Protected Geographical indication (PGI) by multi-elemental fingerprinting, <<FOOD CHEMISTRY>>, 2021; 362 (N/A): N/A-N/A. [doi:10.1016/j.foodchem.2021.130168] [http://hdl.handle.net/10807/187844]
Provenance discrimination of Sorrento lemon with Protected Geographical indication (PGI) by multi-elemental fingerprinting
Fontanella, Maria ChiaraSecondo
;Beone, Gian MariaPenultimo
;
2021
Abstract
Multielement analysis and chemometric methods were proposed to discriminate the Sorrento lemon (PGI) juices according to geographical origin. In 2018 and 2019, 169 fruits from three farms in PGI area and two in not-PGI area were collected and analysed for essential and not-essential elements by ICP-MS. The PCA of multielement fingerprinting grouped lemon juices from PGI farms revealing a strong differentiation at small geographical scale. The S-LDA discriminated lemon juices for Mo, Ba, Rb, Mg, Co, Ca, Fe, Sr on the two production years, giving 97.7% correct classification, 98.5% accuracy and 93.8% external validation. The good correlation lemon juice vs cultivation soil and the soil discrimination by not-essential elements suggested the use of these elements as reliable indicators of lemon juice provenances. Despite lowering the number of variables, constituted by not-essential elements Ba, Rb, Ti, Co, the use of S-QDA discriminated the lemons juices with 87.5% accuracy and 83.9% validation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.