A bioavailability model predicting the toxicity of nickel to rainbow trout (Oncorhynchus mykiss) and fathead minnow (Pimephales promelas) in synthetic and natural waters.

The effects of Ca, Mg and pH on the toxicity of Ni to juvenile rainbow trout (Oncorhynchus mykiss) were examined during 17-26-day exposures to Ni in 15 synthetic test solutions. Higher chemical activities of Ca2+, Mg2+ and H+ reduced Ni toxicity, as demonstrated by increased 17-day median lethal concentrations expressed as Ni2+ activity (17-d LC50(Ni2+)). A non-linear increase of the 17-d LC50(Ni2+) with increasing H+ suggested that the effect of pH could not be appropriately described by single-site competition between Ni(2+) and H+ for sensitive sites on the fish gill. Instead, a linear increase of pNi2+ (=-log 17-d LC50(Ni2+)) with increasing pH was observed with a slope of 0.32. This slope was used as the basis for modelling the effect of pH. The effects of Ca and Mg were modelled according to single-site competition with logK(CaBL)=logK(MgBL)=3.6, both assumed to be independent of pH. The effect of pH was superimposed on this competition effect and was also assumed to be independent of Ca and Mg concentrations. The model was able to predict 17-d LC50s (expressed as dissolved Ni) in most synthetic test waters within a factor 2 deviation from observed toxicity. The model's predictive capacity was also evaluated using results of similar laboratory toxicity tests with juvenile rainbow trout in Ni-spiked European natural surface waters. For most of these waters, predicted 17-d LC50s did not deviate more than a factor 2 from observed toxicity. The same model, calibrated to account for sensitivity differences between species, life stages and/or exposure durations, was able to accurately predict 96-h LC50s for larval and juvenile fathead minnow (Pimephales promelas) and juvenile rainbow trout, based on data taken from literature. Although the developed model seems very promising, the uncertainty around the role of alkalinity and the exact mechanisms by which Ca, Mg and pH modify Ni toxicity need to be further explored.