基于元素及稳定同位素分析的大黄鱼产地区分技术

为实现对4种大黄鱼（Larimichthys crocea）产地（分别来自福建宁德养殖、浙江温州养殖、浙江舟山养殖和舟山渔场自然海域捕捞）的区分,应用电感耦合等离子体质谱仪（inductively coupled plasma mass spectrometry,ICP-MS）及元素分析-同位素质谱仪对大黄鱼肌肉中19种矿物元素及δ¹³C、δ¹⁵N和C、N元素的质量分数进行了测定,基于质量分数差异并结合主成分分析（principal component analysis,PCA）、聚类分析（cluster analysis,CA）和Fisher判别分析对4种不同产地大黄鱼进行区分。单因素方差分析（ANOVA）及事后多重比较（LSD和Tamhane T²检验）结果表明,12种元素（C、N、Mg、Al、K、Ti、Cr、Cu、Zn、Se、Sn、Ba）及δ¹³C、δ¹⁵N在不同产地大黄鱼中具有较明显的地域特征差异。PCA结果表明,前2个主成分（累计方差贡献率为64.1%）的得分散点图分成了4个相对集中的区域,4种不同产地大黄鱼彼此无重叠,能被有效区分。CA结果表明,4种不同产地大黄鱼中除温州养殖大黄鱼样本被分为两类,其他3种大黄鱼样本均各自聚为一类。Fisher判别分析结果显示,原始分类结果的判别准确率为100%,“留一法”交叉验证分类结果显示,有85.7%的舟山养殖大黄鱼样本被正确分类,14.3%的被误判为温州养殖大黄鱼,其余3种大黄鱼样本分类准确率均为100%。研究结果表明,利用元素及稳定同位素比值差异结合化学计量学分析手段,可有效实现大黄鱼的产地区分,为大黄鱼产地鉴别提供参考。

Geographical identification of Larimichthys crocea based on analysis of elements and stable isotope ratios

In order to distinguish the geographic origins of four Larimichthys crocea (L. crocea,three species rearing in Ningde, Wenzhou, and Zhoushan, and one farming in sea area of Zhoushan), the contents of 19 mineral elements, carbon stable isotope ratio (δ¹³C), nitrogen stable isotope ratio (δ¹⁵N), carbon (C) and nitrogen (N) in the muscle tissues were determined by inductively coupled plasma mass spectrometry (ICP-MS) and elemental analysis-isotope mass spectrometry. Based on the difference on contents of such elements, and combined with principal component analysis (PCA), cluster analysis (CA) and Fisher discriminant analysis, the four producing areas of L. crocea were distinguished. Single factor ANOVA test and multiple comparisons (LSD and Tamhane T² test) showed the contents of C, N, Magnesium (Mg), Aluminium (Al), Kalium (K), Titanium (Ti), Chromium (Cr), Cuprum (Cu), Zinc (Zn), Selenium (Se), Stannum (Sn), Barium (Ba), δ¹³C and δ¹⁵N had obvious regional differences. PCA analysis showed that the score scatter plot of the first two principal components (with the cumulative variance contribution rate of 64.1%) was divided into four relatively concentrated and independent regions without overlapping. CA analysis showed that samples of L. crocea rearing in Wenzhou were divided into two groups and samples of the other three species each were clustered into one group respectively. Fisher discriminant analysis showed that the accuracy of the original classification results was 100%, while the 'leave-one-out' cross-validation classification showed that 85.7% samples of L. crocea rearing in Zhoushan were correctly classified (the other 14.3% were misclassified as rearing in Wenzhou), and the samples of other three species of L. crocea were 100% correctly identified. The results show that the application of analysis on elements and stable isotope ratios, combined with chemometric analysis can effectively distinguish the geographical origin of L. crocea, provide reference for origin authenticity identification.