Abstract: The geographical traceability indicators were selected by analysis of the differences in mineral element contents of Tan sheep meat of different geographical origins and multivariate statistical analysis to construct a discriminant model for the identification of the geographic origin of mutton. Inductively coupled plasma-mass spectrometry was used to determine the contents of 25 mineral elements in the Longissimus dorsi muscle of Tan sheep from Yanchi county of Ningxia, Etuokeqian banner of Inner Mongolia autonomous region, Dingbian county of Shanxi province and Huanxian county of Gansu province. Analysis of variance, principal component analysis and linear discriminant analysis were carried out to develop and validate the discriminant model. The results showed that the contents of 17 of the 25 mineral elements in mutton were significantly different among growing regions. The principal component 1 was mainly composed of six elements: Cr, Mn, Ni, Cu, Se and Rb, with the highest discriminant contribution rate of 26.403%; the principal component 2 was mainly composed of four elements of K, P, Fe and Sn, with contribution rate of 22.285%. Through the stepwise discriminant analysis, Ca, P, Cr, Mn, Ni, Cu, Se, Rb, Mo and Sn were selected for modeling. The overall correct discrimination rate for the geographical origins of mutton was 100%, suggesting the model could accurately distinguish Tan sheep meat from different producing areas. Therefore, discrimination based on mineral element fingerprints can effectively identify Tan sheep meat from different producing areas.

Key words: Tan sheep meat, mineral element, fingerprint, geographical origin, traceability