Comparison of the common spatial interpolation methods used to analyze potentially toxic elements surrounding mining regions

• Published in: Journal of environmental management Vol. 212; pp. 23 - 31
• Main Authors: Ding, Qian, Wang, Yong, Zhuang, Dafang
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 15.04.2018
• Subjects: anthropogenic activities; arsenic; China; copper; correlation; Inverse distance weighting interpolation; kriging; lead; mining; Mining area; Ordinary kriging interpolation; pollution; Potentially Toxic Element; prediction; Radial basis function interpolation; soil; toxic substances; Uncertainty; zinc; China; Mining area; Radial basis function interpolation; Ordinary kriging interpolation; Uncertainty; Potentially Toxic Element; Inverse distance weighting interpolation
• ISSN: 0301-4797; 1095-8630; 1095-8630
• DOI: 10.1016/j.jenvman.2018.01.074

The appropriate spatial interpolation methods must be selected to analyze the spatial distributions of Potentially Toxic Elements (PTEs), which is a precondition for evaluating PTE pollution. The accuracy and effect of different spatial interpolation methods, which include inverse distance weighting interpolation (IDW) (power = 1, 2, 3), radial basis function interpolation (RBF) (basis function: thin-plate spline (TPS), spline with tension (ST), completely regularized spline (CRS), multiquadric (MQ) and inverse multiquadric (IMQ)) and ordinary kriging interpolation (OK) (semivariogram model: spherical, exponential, gaussian and linear), were compared using 166 unevenly distributed soil PTE samples (As, Pb, Cu and Zn) in the Suxian District, Chenzhou City, Hunan Province as the study subject. The reasons for the accuracy differences of the interpolation methods and the uncertainties of the interpolation results are discussed, then several suggestions for improving the interpolation accuracy are proposed, and the direction of pollution control is determined. The results of this study are as follows: (i) RBF-ST and OK (exponential) are the optimal interpolation methods for As and Cu, and the optimal interpolation method for Pb and Zn is RBF-IMQ. (ii) The interpolation uncertainty is positively correlated with the PTE concentration, and higher uncertainties are primarily distributed around mines, which is related to the strong spatial variability of PTE concentrations caused by human interference. (iii) The interpolation accuracy can be improved by increasing the sample size around the mines, introducing auxiliary variables in the case of incomplete sampling and adopting the partition prediction method. (iv) It is necessary to strengthen the prevention and control of As and Pb pollution, particularly in the central and northern areas. The results of this study can provide an effective reference for the optimization of interpolation methods and parameters for unevenly distributed soil PTE data in mining areas. •IDW, RBF, OK were compared for Potentially Poxic Plements (PTEs) in mining areas.•The optimal interpolation methods for As, Pb and Zn were RBF, whereas Cu was OK.•Interpolation uncertainty is positively correlated with PTEs concentration.•Prevention of As, Pb pollution in central and northern areas should be enhanced.•The results can provide references to optimize spatial interpolation methods.