Assessing WHAM/Model VII against field measurements of free metal ion concentrations: model performance and the role of uncertainty in parameters and inputs

A key question in the evaluation of chemical speciation models is: how well do model predictions compare against speciation measurements? To address this issue, the performance of WHAM/Model VII in predicting free metal ion concentrations in field samples has been evaluated. A statistical sampling method considering uncertainties in input measurements, model parameters and the binding activity of dissolved organic matter was used to generate distributions of predicted free ion concentrations. Model performance varied with the metal considered and the analytical technique used to measure the free ion. Generally, the best agreement between observation and prediction was seen for aluminium, cobalt, nickel, zinc and cadmium. Important differences in agreement between model and observations were seen, depending upon the analytical technique. In particular, concentrations of free ion determined with voltammetric techniques were largely over-predicted by the model. Uncertainties in model predictions varied among metals. Only for aluminium could discrepancies between observation and model could be explained by uncertainties in input measurements and model parameters. For the other metals, the ranges of model predictions were mostly too small to explain the discrepancies between model and observation. Incorporating the effects of uncertainty into speciation model predictions allows for more rigorous assessment of model performance.