Net fluxes of pesticides from the Scheldt Estuary into the North Sea: a model approach

A hydrological model was constructed with the commercially available modelling environment ECoS for the calculation of mixing plots of dissolved pesticide concentrations in the Scheldt Estuary. The model was based on a one-dimensional advection-dispersion equation and dispersion coefficients were calculated from measured salinity profiles. The model could correctly predict the movement of water masses within the estuary. Nominal flushing times of the estuary were calculated with the fraction-of-freshwater method and ranged between 25 and 95 days, depending on the freshwater discharge. Model calculations demonstrated that the application-related time profiles of pesticide input may well induce curvature in the calculated estuarine mixing plot even if the pesticide behaves conservatively. Calculated mixing plots were compared with experimental data and good agreement was obtained between the measurements and the conservative mixing plot of atrazine. For metolachlor an additional direct emission had to be modelled to explain the differences between the experimental data and the conservative mixing plot. For dichlorvos and simazine, on the other hand, an estuarine loss constant had to be included in the model. Using a least-squares procedure the estuarine loss constants for dichlorvos and simazine were estimated at half-lives of five and 26 days, respectively. Because mixing plots were strongly influenced by the time profiles of pesticide input, standard procedures in which net fluxes are calculated from mixing plots, were not applicable. Therefore, net fluxes were calculated with a newly developed procedure in which the estuarine loss constants and the estuarine flushing time were combined. For the non-conservative compounds dichlorvos and simazine, the net fluxes to the North Sea were found to be 96 and 64%, respectively, lower than the gross fluxes transported by the River Scheidt. (C) 2001 Elsevier Science Ltd. All rights reserved.