Research On The Origin Of Guizhou Tea Based On The Content Characteristics Of Multi-element And Biochemical Components

Tea is a perennial cash crop.The chemical molecular composition of tea varies with the growth environment and processing technology.As a major tea planting province in China,Guizhou’s tea industry has gradually developed into a pillar industry in the agricultural field,making significant contributions to the economic development of Guizhou.Therefore,in order to explore the chemical composition characteristics of tea in different regions of Guizhou,safety supervision of Guizhou green tea is carried out,and data support is provided for the identification of Guizhou tea production areas.The mineral element content,stable isotope ratio and biochemical component content of tea in six regions of Zunyi,Duyun,Anshun,Bijie,Southwest and Southeast Guizhou were measured.In addition,this study also collected soil samples from tea gardens in Zunyi,Duyun,and Anshun,and analyzed the mineral element content and p H distribution characteristics of the soil.The main conclusions obtained are as follows:（1）The average content of harmful elements Pb,Cd,As,and Cr in tea from six regions in Guizhou did not exceed the limit values specified in the Chinese tea standards.The content of Fe,Ni,Zn,Mg,and Pb in the soil of tea gardens in Zunyi,Duyun,and Anshun was lower than the risk control value of agricultural land soil in China.However,Ni element exceeded the limit at certain points in some regions.After analyzing the soil p H value,it was found that the soil in the tea gardens of Duyun and Anshun showed excessive acidification,while the soil in the tea gardens of Anshun（p H=4.206）and Duyun（p H=4.444）showed extreme acidity.The results of stable isotope analysis show that The range ofδ²H variation is relatively large,ranging from-60.777 to-109.864 mg/kg;The variation range ofδ¹³C is relatively small,ranging from-25.08 to-30.632 mg/kg.amongδ²H is an effective variable for identifying the origin of tea.In addition,the biochemical components of tea from various regions,such as free amino acids,caffeine,and tea polyphenols,are abundant and appropriate,and the content of each component meets national standards.（2）The mineral element content,stable isotope value,and biochemical component content of tea in different regions of Guizhou have their own unique characteristics.According to the one-way ANOVA results,there are significant differences（p<0.05）among 13chemical components in tea in different regions,including mineral elements Pb,Zn,Cd,As,Ni,Co,Mn,Cr,Mg,and Ca,as well as stable isotopesδ²H,biochemical components of tea polyphenols and caffeine.Based on the 13 differential chemical components mentioned above,principal component analysis（PCA）,orthogonal partial least squares discriminant analysis（OPLS-DA）,and linear discriminant analysis（LDA）were used to distinguish tea from different regions in Guizhou.Through the PCA two-dimensional scatter plot,it can be found that there is a certain degree of differentiation between tea samples from Zunyi,Duyun,and Anshun,but there is serious confusion among tea samples from Bijie,Qianxinan,and Qiandongnan.Subsequently,seven effective discriminant variables were successfully selected by combining LDA and OPLS-DA,including Ca,Zn,Mn,Cr,Pb,As,and caffeine.The LDA results indicate that the LDA model constructed based on the above seven variables has good generalization ability and discrimination effect,with an accuracy rate of 93.9%for origin discrimination.In summary,it is feasible and effective to use differential chemical components between tea producing areas for identifying tea producing areas.（3）Comparing the discriminative model established based on different variables,it is found that the accuracy of the discriminative model generated by combining mineral elements,stable isotopes and biochemical components is higher than that of the discriminative model established by using mineral elements alone.When using mineral element variables alone to establish a discriminative model,the LDA model with the best effect has a discrimination accuracy of 90.3%,and the accuracy of cross validation by leaving one method is 71.4%.After adding two stable isotopes and three biochemical components,The discrimination rate of the LDA model was increased to 93.5%,and the accuracy of the left one method cross validation was increased to 83.3%.Therefore,it is an effective way to improve the accuracy of the discriminative model to clearly increase the discriminant variables.（4）LDA performs better than PCA and OPLS-DA in discriminating tea producing areas.Although PCA cannot effectively distinguish tea producing areas,it can screen out some effective variables for discriminating tea producing areas.In this study,OPLS-DA was not ideal in discriminating tea producing areas,resulting in overfitting.Therefore,overall,LDA has the best discriminating effect in this study.In addition to the discrimination method,the number of samples also has an impact on the model’s discrimination performance.When the number of samples is small,the sample characteristics of each sampling point are not obvious enough,which can easily misjudge the origin of the sample.Increasing the number of samples can effectively increase the discrimination accuracy.