Inferring the Place of Origin of Sweet Cherry (Prunus Avium (L.) Moench) from Different Provinces of China Using Stable Isotope and Multi-Element Analysis with Multivariate Modeling

A strategy combining stable isotope and multi-element analysis with chemometrics was developed for origin traceability of cherry products from five provinces of China (Liaoning, Shandong, Hebei, Sichuan, Shannxi). The relationships of stable isotope and multi-element signatures of cherry with irrigation water and cultivated soil in its origin are discussed. One-way analysis of variance (ANOVA) indicates that the regional differences of δ13C, δ2H, δ18O, 87Sr/86Sr, Na, Mg, Ca, Fe, Zn and Se are significant (P<0.05) and can be as indicators for cherry origin traceability. Multivariate modeling of stable isotope and multi-element data of cherry using hierarchical cluster analysis (HCA), partial least squares discriminant analysis (PLS-DA), linear discriminant analysis (LDA) was conducted and compared. The discriminant accuracy of PLS-DA was unsatisfactory for all origins of China, but it was significantly improved by LDA, in which 153 cherry samples from five provinces can be accurately identified with accuracy higher than 90%.