Abstract :
Leaching rates of the herbicide dichlorprop [(±-2-(2,4-dichlorophenoxy)propanoic acid] and nitrate were measured together in field lysimeters containing undisturbed clay and peat soils. The purpose of the study was to investigate the leaching pattern of the two solutes in structured soils under different precipitation regimes. Spring barley (Hordeum distichum L.) was sown on each monolith and fertilized with 100 kg N ha−1. Dichlorprop was applied at a rate of 1•6 kg active ingredient (a.i.) ha−1. Each soil type received supplemental irrigation at two levels (‘average’ and ‘worst-case’), giving total water inputs (irrigation and precipitation) of 664 and 749 mm year−1, respectively. The larger water input approximately doubled the nitrate loads, from, on average, 11•6 to 21•8 kg N ha−1 year−1 in the clay soil and from 37•6 to 65•4 kg N ha−1 year−1 in the peat soil. In contrast, dichlorprop leaching was reduced by more than one order of magnitude when the water input was increased, from average amounts of 3•22 to 0•26 g a.i. ha−1 during an S-month period in the clay and from 28•9 to 2•67 g a.i. ha−1 in the peat. This leaching pattern of dichlorprop was explained in terms of preferential flow. The dried-out topsoil of ‘average’ watered monoliths may have allowed water flow in cracks, thus moving some of the herbicide rapidly through the topsoil to the subsoil. Once the compound reached the subsoil, degradation rates would be reduced and the herbicide residues would be stored for later leaching. Nitrate was presumably more evenly distributed in the soil matrix; therefore, water rapidly moving through macropores would not carry significant amounts of nitrate. In contrast, leaching would occur more evenly through the soil matrix, causing larger nitrate loads in the ‘worst-case’ watered monoliths. These results show that wet years may constitute a worst case scenario in terms of nitrate leaching, but not pesticide leaching, if macropore flow exerts a significant influence on leaching.
