摘要: Most researchers have concluded that the uncomplexed Cu ion is the primary toxic or bioavailable species while others have suggested that hydroxide, carbonate and organically complexed species may also be toxic. Because the comput... 展开 Most researchers have concluded that the uncomplexed Cu ion is the primary toxic or bioavailable species while others have suggested that hydroxide, carbonate and organically complexed species may also be toxic. Because the computed activities of the aqueous Cu species are both highly dependent on the thermodynamic data used and highly correlated with each other, it is difficult to show definitively that the individual chemical species-toxicity correlations are valid and it is virtually impossible to determine the statistical relationship between toxicity and copper aqueous speciation using traditional statistical methods. The proposed methodology consists of first computing the aqueous speciation of Cu using the geochemical model, MINTEQ, with a thoroughly reviewed and partially validated thermodynamic data base for Cu. Finally, multivariate statistical techniques applicable to collinear data and underdetermined systems are applied to the activities of the Cu species and the toxicity measurements to determine the relationship between the speciation and toxicity. A reevaluation using this methodology indicates that the primary bioavailable species are the hydroxide species and that the carbonate and phosphate species are not toxic. Based on the results of the statistical analyses and the properties of the statistical techniques, it appears that a combination of factor analysis and ridge regression yield the most information about the structure of the collinearities and provide estimates of the coefficients and the bias. 35 references, 1 figure, 1 table. (ERA citation 09:036383) 收起