Pesticide retention in an experimental wetland treating non-point source pollution from agriculture runoff

Pesticide losses to the environment are unwanted due to possible environmental and health hazards. An experimental wetland is established to study the efficiency with respect to retention of sediments, nutrients and pesticides. Pesticides were applied on an arable soil plot in the watershed. Statistical analyses were carried out on three selected pesticides: propachlor, metalaxyl and chlorfenvinfoss. All pesticides were found in the experimental wetland, with peak concenttrations shortly after spraying. In 2003 pesticide retention varied from 11% to 42% and in 2004 retention varied from 19% to 56%. Comparing eight different wetland filters, we found that L6 and L8, with flagstones and straw, respectively, had a higher total pesticide retention than a standard Norwegian wetland (L4). When the compounds were treated separately, however, the picture was different. Statistical analyses showed that the treatments were signficantly different from zero in six of the wetlands for remowal of propachlor, for removal of metalaxyl none were significantly different, and for removal of chlorfenvinphos four treatments were significantly different. For the three compounds none of the relative treatments were significantly different from L4. Chemical properties of the pesticides could explain some of the behaviour in the watershed and in the wetland.