Comparison of measured and predicted concentrations of selected pharmaceuticals in wastewater and surface water: a case study of a catchment area in the Po Valley (Italy).

The aim of this study is to compare measured and predicted concentrations of 11 antibiotics and the antiepileptic carbamazepine in wastewater and surface water and to discuss the accuracy and usefulness of applied predictive models in assessing their concentrations. The occurrence (exposure assessment) of the target compounds was monitored in the influent and effluent of a large municipal wastewater treatment plant, and in its receiving water body, situated in the Po Valley, Italy. Information on the consumption and sales of pharmaceuticals in Italy, along with data related to their excretion and removal during wastewater treatment, were used to predict the concentrations of the selected pharmaceuticals for the studied site. The measured and predicted concentrations were compared for all three sampling points, and according to a criterion available in literature and already adopted in similar studies, the prediction was considered "acceptable" when the ratio of predicted and measured concentrations was between 0.5 and 2. The results show that the concentrations were accurately predicted for ciprofloxacin in wastewaters, and for azithromycin, trimethoprim and carbamazepine in surface water. For all the other compounds and sampling points, the difference between the measured and the predicted concentrations was very high. Possible reasons for these discrepancies are discussed and the main potential factors that influence both measured and predicted concentrations are identified by means of an assessment of the uncertainties in the measured values and sensitivity of all the parameters required for predicting concentration ones. The potential factors are sampling mode for measured values and dilution factor, removal and excretion for predicted ones.