Use of statistical and GIS techniques to assess and predict concentrations of heavy metals in soils of Lahore City, Pakistan.

Soils from different land use areas in Lahore City, Pakistan, were analyzed for concentrations of heavy metals-cadmium (Cd), chromium (Cr), nickel (Ni), and lead (Pb). One hundred one samples were randomly collected from six land use areas categorized as park, commercial, agricultural, residential, urban, and industrial. Each sample was analyzed in the laboratory with the tri-acid digestion method. Metal concentrations in each sample were obtained with the use of an atomic absorption spectrophotometer. The statistical techniques of analysis of variance, correlation analysis, and cluster analysis were used to analyze all data. In addition, kriging, a geostatistical procedure supported by ArcGIS, was used to model and predict the spatial concentrations of the four heavy metals-Cd, Cr, Ni, and Pb. The results demonstrated significant correlation among the heavy metals in the urban and industrial areas. The dendogram, and the results associated with the cluster analysis, indicated that the agricultural, commercial, and park areas had high concentrations of Cr, Ni, and Pb. High concentrations of Cd and Ni were also observed in the residential and industrial areas, respectively. The maximum concentrations of both Cd and Pb exceeded world toxic limit values. The kriging method demonstrated increasing spatial diffusion of both Cd and Pb concentrations throughout and beyond the Lahore City area.