基于稀土元素指纹分析判别宁都不同基岩区脐橙的研究

Application of Rare Earth Element Fingerprints identification of Navel Orange from Different Bedrock Areas in Ningdu

为探讨基于稀土元素指纹对宁都不同基岩区脐橙判别的可行性,确定具有表征判别特征的稀土元素指标,选择在宁都花岗岩、沉积岩基岩脐橙园采集脐橙果实样品17组,采用电感耦合等离子体质谱法测定15种稀土元素含量,对测试数据进行单因素方差分析、主成分分、Fisher线性判别分析、正交偏最小二乘判别分析。结果表明,宁都不同基岩区稀土元素在土壤、脐橙中的分配曲线基本一致,且脐橙中稀土元素La,Pr,Nd,Gd,Tb,Dy,Y,Sm,Ho,Er平均含量在两个基岩区存在显著性差异;主成分分析得到的分散点图结果显示花岗岩、沉积岩基岩区脐橙得分相距较远,绝大部分样品已明显区分,仅极个别样点存在有交叉现象;Fisher线性判别分析共筛选出5个稀土元素有效判别指标Sm,Eu,Gd,Tb,Er,判别模型的回代检验的整体正确判别率为94.1%,留一交叉检验整体正确判别率为82.4%,检验结果显示模型在溯源不同基岩区脐橙的产地来源是可行的,两个基岩区的判刑函数分别为Group one(沉积岩区)=5.359Sm+10.712Eu-7.704Gd-2.143Tb+4.434Er,Group two(花岗岩区)=-0.874Sm+5.849Eu-0.266Gd+7.675Tb+1.268Er;正交偏最小二乘判别分析结果显示模型的两个主成分对沉积岩区、花岗岩区脐橙样品的预测能力为30%,沉积岩区、花岗岩区脐橙可以较好的区分,对判别模型其重要作用的8个稀土元素指标中Y及Ho的贡献最高。利用稀土元素指纹特征用于判别脐橙不同基岩区来源是可行的。

In order to examine the feasibility of using rare earth elements as a fingerprints to distinguish navel oranges from different bedrock areas in Ningdu,and to determine the rare earth element indexes with distinguishing characteristics,17 groups of navel orange fruit samples were collected in navel orange garden of granite and sedimentary bedrock in Ningdu,and the contents of 15 rare earth elements were determined by inductively coupled plasma mass spectrometry.The test data were analyzed by one-way ANOVA,principal component analysis,Fisher linear discriminant analysis and orthogonal partial least squares discriminant analysis.The results showed that the distribution curves of rare earth elements in soil and navel orange in different bedrock areas of Ningdu were basically the same,and the average contents of rare earth elements La,Pr,Nd,Gd,Tb,Dy,Y,Sm,Ho and Er in navel orange were significantly different in the two bedrock areas.The scatter plot of scores obtained by principal component analysis showed that the scores of navel orange in bedrock area of granite and sedimentary rock are far away from each other.Most samples have been clearly distinguished,and only a few samples have cross phenomenon.Fisher linear discriminant analysis of mercury screened out five effective discriminant index of Sm,Eu,Gd,Tb,Er,the discriminant model of back substitution test overall correct discriminant rate was 94.1%,leaving a cross test overall correct discriminant rate was 82.4%.The test results show that the model is feasible in tracing the origin of navel orange in different bedrock areas,the discriminant function of two bedrock areas are Group one(sedimentary rock area)=5.359Sm+10.712Eu-7.704Gd-2.143Tb+4.434Er,Group two(granite area)=-0.874Sm+5.849Eu-0.266Gd+7.675Tb+1.268Er.The results of orthogonal partial least squares discriminant analysis showed that the prediction ability of the two principal components of the model for navel orange samples in sedimentary rock area and granite area was 30%,and navel oranges in sedimentary rock area and granite area could be well distinguished.Among the eight rare earth elements played an important role in the discriminant model,Y and Ho had the highest contribution.It is feasible to use the fingerprint characteristics of rare earth elements to identify the sources of different bedrock areas of navel orange.