BACKGROUND A half of apple production worldwide comes from China. However, the Chinese apple geographic authorization has not been well studied. We highlighted the multi-element-based geographical discrimination of apples from… Click to show full abstract
BACKGROUND A half of apple production worldwide comes from China. However, the Chinese apple geographic authorization has not been well studied. We highlighted the multi-element-based geographical discrimination of apples from the Southwest Cold Highland (SCH) of China. Fifteen elements of 565 samples from the SCH (138) and others (427) were applied for constructing discriminate models. RESULTS The SCH apples from 2017 to 2019 had higher concentration of Mn, Zn, Cr, Cd, Se, Pb and Fe, but lower concentration of Na, B, Ni and P. The representative linear discriminant analysis (LDA) discriminated the SCH with sufficient training and testing accuracy averaged 92.8% and 92.2%. The nonlinear discriminate models were more suitable than the linear models. The optimized random forest analysis was a most fit model, and obtained with averaged training and testing accuracy of 98.2% and 98.8%. CONCLUSION The multielement based discrimination of the SCH apple could aid further geographical origin study. This article is protected by copyright. All rights reserved.
               
Click one of the above tabs to view related content.