Simulation of the persistence of atrazine, linuron and metolachlor in soil at different sites in the USA

The effects of soil temperature and soil moisture content on the rates of degradation of atrazine, linuron and metolachlor were measured in the laboratory in soil from different sites in the USA. Persistence of the herbicides was measured in the same soils in the field during the summers of 1978 and 1979. Weather records from the different sites for the periods of the field experiments were used in conjunction with appropriate constants derived from the laboratory data in a computer program to simulate persistence in the field. There was a general tendency for the model to overestimate the observed soil residues. For example, with atrazine, 40 of the 48 measured residues were lower than those predicted by the model; seven were more than 30% below and two were more than 50% below. With metolachlor, 16 of the 48 measured residues were more than 30% below those predicted and six were more than 50% below; almost identical results were obtained with linuron. When the model overestimated late-season residues by a large amount, the discrepancies between predicted and observed data were usually apparent from early in the experiment. Possible reasons for the discrepancies are discussed.