Wang,X.J., Liu,R.M., Wang,K.Y., Hu,J.D., Ye,Y.B., Zhang,S.C., Xu,F.L. and Tao,S.(2006): Application of multivariate spatial analysis in scale-based distribution and source study of PAHs in the topsoil: an example from Tianjin, China. Environ.Geol., 49, 1208-1216.


 Polycyclic aromatic hydrocarbon (PAHs) in soil originated from various sources under different spatial scales. Coregionalization analysis is more revealing than univariate geostatistical analysis. Scale-dependent spatial features of variables reflect different sources of spatial variability. In this study, 188 topsoil samples in the Tianjin area were collected. The contents of 16 PAHs and soil background properties were determined for all samples. A multivariate geostatistical approach was used for multi-scale spatial analysis for PAH compounds. Results show that coal combustion was the major source for the spatial distribution patterns of PAHs in the topsoil of the studied area. It worked mainly at the short-range scales (5-10 km). Significant spatial variation patterns were identified. In contrast, no significant spatial distribution trends at the nugget (0-5 km) or long-range scales (10-50 km) were seen. Long-range transport and site contamination of PAHs might not be key contributors in forming the distribution pattern of PAHs in the topsoil of Tianjin area.

Keywords: PAHs; Multi-scale spatial distribution; Sources; Tianjin; China』

Experimental section
 Sample extraction, cleanup and analysis
 Data analysis
Results and discussion
 Coregionalization analysis
 Structural correlation coefficients and regionalized factors
 Cokriging analysis for PAHs, soil properties and regionalized factors
 Sources of PAH contamination