Prediction Patterns of Cr and Ni in soils

Correlation and regression analyses were conducted on the data generated from the analysis of soils samples from Bauchi state, in order to comparehow two metals (Cr and Ni) in these soils can be predicted by some soil physico-chemical properties (pH, electrical conductivity, EC; and soil moisture contents, MC). Resulting coefficient of correlation established a linear relationship between the study metals and one or more physico – chemical parameters in all the study sites. Test results of the suitability of the models to the data obtained indicated that linear, logarithmic and quadratic models were found suitable for both metals in the study sites, although quadratic models gave better predictions. The Regression equations obtained showstriking similarities in the prediction pattern of both metals with respect to their predictor variables. The models obtained could be used to predict approximately between 41.0 – 84.0% Cr and 50.0 – 82.0% Ni.