Metabolism and Persistence of Atrazine in Several Field Soils with Different Atrazine Application Histories

To assess the potential occurrence of accelerated herbicide degradation in soils, the mineralization and persistence of (14)C-labeled and nonlabeled atrazine was evaluated over 3 months in two soils from Belgium (BS, atrazine-treated 1973-2008; BC, nontreated) and two soils from Germany (CK, atrazine-treated 1986-1989; CM, nontreated). Prior to the experiment, accelerated solvent extraction of bulk field soils revealed atrazine (8.3 and 15.2 μg kg(-1)) in BS and CK soils and a number of metabolites directly after field sampling, even in BC and CM soils without previous atrazine treatment, by means of LC-MS/MS analyses. For atrazine degradation studies, all soils were incubated under different moisture conditions (50% maximum soil water-holding capacity (WHC(max))/slurried conditions). At the end of the incubation, the (14)C-atrazine mineralization was high in BS soil (81 and 83%) and also unexpectedly high in BC soil (40 and 81%), at 50% WHC(max) and slurried conditions, respectively. In CK soil, the (14)C-atrazine mineralization was higher (10 and 6%) than in CM soil (4.7 and 2.7%), but was not stimulated by slurried conditions. The results revealed that atrazine application history dramatically influences its degradation and mineralization. For the incubation period, the amount of extractable atrazine, composed of residues from freshly applied atrazine and residues from former field applications, remained significantly greater (statistical significance = 99.5 and 99.95%) for BS and CK soils, respectively, than the amount of extractable atrazine in the bulk field soils. This suggests that (i) mostly freshly applied atrazine is accessible for a complex microbial community, (ii) the applied atrazine is not completely mineralized and remains extractable even in adapted soils, and (iii) the microbial atrazine-mineralizing capacity strongly depends on atrazine application history and appears to be conserved on long time scales after the last application.