Total and Available Copper in Some Soil Profile Samples from the State of Sao Paulo

Chemical information obtained through soil analysis requires overall and simultaneous statistical treatment of variables and samples for better data interpretation. Multivariate analysis is the simultaneous analysis of several samples and variables; it provides complementary information not evidenced by the ordinary univariate statistical analysis. Therefore, the objective of the present work was to evaluate the copper (Cu) concentration in several soil profiles and also to evaluate the influence of some soil properties on Cu concentrations, using different extraction methods. Multivariate main component analysis was performed on the dataset. Available and total Cu concentrations were determined in the A and B horizon samples of 28 soil profiles from representative soil unities of the State of Sao Paulo, Brazil. The extracting solutions used to determine the available Cu forms were diethylenetriaminepentaacetic acid (DTPA), pH 7.3 (CuDTPA); Mehlich 1 (CuM1); 0.1 mol L-1 hydrochloric acid (HCl) (CuHCl); ethylenediaminetetraacetic acid (EDTA) (CuEDTA); and 1.0 mol L-1 ammonium acetate (NH4OAc) (CuAc). Most soil samples presented medium to high available Cu concentrations, indicating that such soils are sufficient to supply Cu to plants. All correlations between different Cu forms were positive and significant, indicating that a correspondence exists between the extractive capacities of the methods tested. The most significant correlations were CuDTPA x CuEDTA, CuDTPA x CuHCl, and CuEDTA x CuHCl, at r = 0.98. The soil properties that most influenced Cu concentrations were iron oxides (r = 0.81), magnesium (Mg) (r = 0.69), silt (r = 0.64), and pH (r = 0.63). The main component analysis revealed an association among the following soil properties: silt, pH, Mg, Cation Exchange Capacity (CEC), silicon (Si), aluminum (Al) and iron (Fe) oxides, total copper (CuT), and CuM1, which presented values greater than 0.64 in F1. In F2, the association among clay, fine sand, and aluminum oxide contents was evidenced. It was concluded that most soil samples presented medium to high Cu concentrations; similar forms of Cu were extracted by DTPA, EDTA, and HCl; and the main component analysis was efficient for grouping the soils from the same parent material.