Spatial variation and pollution risk assessment of heavy metals in industrial soil based on geochemical data and GIS-A case of an iron and steel plant in Beijing，China

Abstract

ABSTRACT The precise range of contamination determines the cost of treating heavy metals contamination of soils at industrial sites and was a key aspect of soil heavy metals contamination management. The pollution level and spatial correlation of As, Cd, Cr, Cu, Hg, Ni, Pb and Zn in one hundred and thirty topsoil samples from a typical contaminated site was analysed. Geochemical analysis showed that Hg pollution was the most serious in the site, followed by Cd, Pb, Zn, Cu and Cr. The contents of As and Ni in the soil did not exceed the background values. The interpolation accuracy, interpolation results and spatial distribution of inverse distance weight method (IDW), local polynomial method (LPI), radial basis function method (RBF) and ordinary kriging method (OK) were compared, respectively. The results show that the range, mean and coefficient of variation of the four interpolation methods were generally lower than that of the sampled data, which was mainly caused by smoothing effect. The smoothing effect of LPI and OK were the most serious, while the smoothing effect of IDW and RBF were not obvious. The optimal interpolation method of Cd, Cr, Hg, Pb and Zn in polluted sites was RBF and the optimal interpolation method for As, Cu and Ni was IDW. The evaluation of polluted area by IDW and RBF interpolation method was ranked as IDW > RBF, which was much larger than that calculated by sampling point. Both interpolation methods could increased the area of polluted area. The probability kriging (PK) method was similar to that based on sampling points, which can greatly improve the calculation accuracy of heavy metals polluted area in industrial contaminated sites. Geochemical data and spatial interpolation results show that phytoremediation techniques was the most suitable for remediation of heavy metal contaminated soils at this site.