Source mapping and determining of soil contamination by heavy metals using statistical analysis, artificial neural network, and adaptive genetic algorithm

In this paper, a novel integrated approach for tracing the heavy metal contamination sources in urban surface soils was proposed by combining statistical analysis, artificial neural network (ANN), and adaptive genetic algorithm (AGA). Total 319 surface soil samples from an area of about 400 km2 including five functional areas have tested here. Firstly, the pollution level of a single heavy metal and the overall status of the urban surface soils contaminated by As, Cd, Cr, Cu, Hg, Ni, Pb and Zn were assessed by the single factor contaminant index and the Nemerow comprehensive index, respectively. Statistical analysis showed that this city has been seriously polluted by heavy metals. Then, the possible sources of heavy metals in the urban surface soils were identified through correlation matrix based on principle component analysis (PCA). At last, the concentrations of heavy metals were estimated by using ANN based intelligent method. And based on this, the AGA was used to accurately search the points with extremely high concentration of heavy metal and their corresponding spatial position. These points especially the one with maximum concentration were regarded as the locations of the contamination source.