The effect of elapsed time between spraying and first leaching event on the leaching behavior of five herbicides (terbuthylazine, S-metolachlor, mesotrione, flufenacet, and isoxaflutole) and two metabolites (desethyl-terbuthylazine and diketonitrile) was evaluated in a 2011-2012 study in northwest Italy. A battery of 12 lysimeters (8.4 m(2) long with a depth of 1.8 m) were used in the study, each filled with silty-loam soil and treated during pre-emergence with the selected herbicides by applying a mixture of commercial products Lumax (4 L ha(-1)) and Merlin Gold (1 L ha(-1)). During treatment periods, no gravity water was present in lysimeters. Irrigation events capable of producing leaching (40 mm) were conducted on independent groups of three lysimeters on 1 day after treatment (1 DAT), 7 DAT, 14 DAT, and 28 DAT. The series was then repeated 14 days later. Leachate samples were collected a few days after irrigation; compounds were extracted by solid phase extraction and analyzed by high-performance liquid chromatography and gas chromatography-mass spectrometry. Under study conditions, terbuthylazine and S-metolachlor showed the highest leaching potentials. Specifically, S-metolachlor concentrations were always found above 0.25 µg L(-1). Desethyl-terbuthylazine was often detected in leached waters, in most cases at concentrations above 0.1 µg L(-1). Flufenacet leached only when irrigation occurred close to the time of herbicide spraying. Isoxaflutole and mesotrione were not measured (<0.1 µg L(-1)), while diketonitrile was detected in concentrations above 0.1 µg L(-1) on 1 DAT in 2011 only.
Keywords: Herbicides; contamination; leaching; lysimeters; metabolite; time; water.