A lysimeter experiment to investigate temporal changes in the availability of pesticide residues for leaching.

Leaching of three pesticides (isoproturon, chlorotoluron and triasulfuron) and a tracer (bromide) were determined in four contrasting soils ranging in texture from sandy loam to clay. The compounds were applied to undisturbed columns of soil and four columns for each soil were randomly selected and leached with 24-mm equivalent of water at prescribed time intervals (3, 9, 24, 37 and 57 d after application). A rapid decline in leached loads of isoproturon and chlorotoluron as time from application to irrigation increased was observed in all soils. In contrast, triasulfuron and bromide loads only decreased rapidly in the clay soil. Bromide losses decreased with decreasing clay contents of the soil and therefore with a decrease in structural development. Magnitudes of pesticide losses varied from soil to soil, depending on structural development and the organic carbon content. Pesticide degradation experiments on disturbed and undisturbed soil samples showed that the rapid decline of leached loads with time was faster than could be explained by degradation alone. Five physico-chemical processes are put forward to explain the different patterns of pesticide leached loads observed in the soils: (1) relative extent of preferential flow, (2) sorption capacity of the compounds to the different soils, (3) extent of degradation of the compounds in the soil, (4) variation in sorption kinetics between compounds associated with pesticide diffusion into soil aggregates, and (5) protection of the compounds by a combination of intra-aggregate diffusion and the presence of preferential flow pathways.