Systematic analysis of freshwater metal toxicity with WHAM-FTOX

Abstract We used the WHAM chemical speciation model and the WHAM- F TOX toxicity model to analyse the published results of laboratory toxicity experiments covering 52 different freshwater biological test species and 24 different metals, a total of 2037 determinations of EC 50 with accompanying data on solution composition. The key extracted parameter was α M , the parameter in WHAM- F TOX that characterises the toxic potency of a metal on the basis of its estimated metabolically active body burden. For 16 data sets applying to metal-test species pairs with appreciable variations in solution composition, values of EC 50 back-calculated from averaged values of α M showed significantly (p  50 values than did the simple average EC 50 , confirming that the modelling calculations could account for some of the dependence of toxicity on chemical speciation. Data for different exposure times permitted a simple parameterisation of temporal effects, enabling values of α M,max (values at infinite exposure time) to be obtained, and the effects of different exposure times to be factored out for further analysis. Comparison of averaged values of α M,max for different metals showed little difference among major taxa (invertebrates, plants, and vertebrates). For Cd, Cu, Ni and Zn (the four metals with most data) there were significant differences among α M,max values for different species, but within-species variabilities were greater. Reasonably similar species sensitivity distributions of standardised α M,max applied to Cd, Cu, Ni and Zn. The average values, over all species, of α M,max increased in the order Al  2 α M,max values, there was a strong dependence (r 2  = 0.56, p  2  = 0.59) if ionic radius was included in the statistical model, indicating that softer, larger cations are the most effective toxicants.