Significance of soil chemical heterogeneity for spatial behavior of cadmium in field soils

To investigate the spatial behavior of Cd in an arable field and to analyze its dependency on pH and organic carbon content (OC), soil was sampled on a 0.5-ha field and analyzed for HNO3-extractable Cd and Zn contents (Cd(T) and Zn(T), respectively), soluble Cd and Zn contents (Cd(s) and Zn(s), respectively), pH, and OC. All parameters were spatially variable and anisotropic. Within the sampled area the measurements became independent of each other for OC, for Cd(T) and Zn(T) in the east-west direction, and for Cd(s) in the north--south direction at distances of 5 to 20 m. For the other variables and directions, no range was observed. Similarity between anisotropy of HNO3-extractable heavy metal contents and OC, as well as between soluble contents and pH, suggested a functional (spatial) relationship between these parameters. Because the field was anisotropic, the spatial structure of the parameters was different from the spatial structure obtained from transect analysis in the same field. Correlation between Cd and Zn contents was high, both in soluble and in HNO3-extractable form. No correlation was observed between HNO3-extractable and soluble heavy metal contents. Soluble Cd and Zn were highly correlated with pH. A significant correlation was found between Cd(T) and Zn(T) and pH, and between HNO3-extractable contents and OC. Using a scaled sorption model, variability of pH and OC was explicitly accounted for when calculating Cd(s) from HNO3-extractable Cd contents. Predicted Cd concentrations in the soil solution correlated well (r = 0.87) with measured Cd concentrations, whereas prediction when homogeneity of the sorption parameters is assumed yielded no significant correlation. Apparently, pH and organic-matter content are the most important soil parameters that regulate Cd sorption in this soil.