Your browser doesn't support javascript.
loading
Application of the partial least square regression method in determining the natural background of soil heavy metals: A case study in the Songhua River basin, China.
Sun, Yaoyao; Zhao, Yuyan; Hao, Libo; Zhao, Xinyun; Lu, Jilong; Shi, Yanxiang; Ma, Chengyou; Li, Qingquan.
Affiliation
  • Sun Y; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China.
  • Zhao Y; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China.
  • Hao L; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China.
  • Zhao X; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China. Electronic address: zhaoxy15@jlu.edu.cn.
  • Lu J; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China.
  • Shi Y; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China.
  • Ma C; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China.
  • Li Q; College of Geo-Exploration Science and Technology, Jilin University, Changchun 130026, China.
Sci Total Environ ; 918: 170695, 2024 Mar 25.
Article in En | MEDLINE | ID: mdl-38331274
ABSTRACT
The "background" is an essential index for identifying anthropogenic inputs and potential ecological risks of soil heavy metals. However, the lithology of bedrock can cause significant spatial variation in the natural background of soil elements, posing considerable difficulties in estimating background values. In this study, an attempt was made to calculate the natural background through regression analysis of soil chemical composition, and reasonably evaluate the impact of lithology. A total of 1771 surface soil samples were collected from the Songhua River Basin, China, for chemical composition analysis, and the partial least square regression (PLSR) method was employed to establish the relationship between heavy metals (As, Hg, Cr, Cd, Pb, Cu, Zn, and Ni) and soil chemical composition/environmental parameters (SiO2, Al2O3, TFe2O3, MgO, CaO, K2O, Na2O, La, Y, Zr, V, Sc, Sr, Li and pH). The result shows that As, Cr, Pb, Cu, Zn, and Ni have significant linear relationships with soil chemical composition. Each of these six heavy metals obtained 1771 regression background values; some were higher than the uniform background value obtained from the boxplot, while others were lower. The regression background values recognized not only subtle anthropogenic inputs and potential ecological risks in low-background regions but also spurious contamination in high-background areas. All these indicate that the PLSR method can effectively improve the determination accuracy of the natural background of soil heavy metals. More attention should be paid to the serious anthropogenic inputs appearing in some places of the study area.
Key words

Full text: 1 Collection: 01-internacional Database: MEDLINE Language: En Journal: Sci Total Environ Year: 2024 Document type: Article Affiliation country: Country of publication:

Full text: 1 Collection: 01-internacional Database: MEDLINE Language: En Journal: Sci Total Environ Year: 2024 Document type: Article Affiliation country: Country of publication: