The fate of the herbicide isoproturon [3‐(4‐isopropylphenyl)‐l,1‐dimethylurea] was investigated in soil from a grassed buffer strip. Compared to a cropped soil originating from the same experimental site (Kd = 1.8 L kg−1), sorption of isoproturon was enhanced in the grassed soil and especially in the surface layer (0‐2 cm) containing high proportion of nondecomposed plant residues (Kd = 5.0 L kg−1). Nonhumified organic fractions isolated from the surface soil layer and corresponding to above and bdow‐ground plant residues derived from the grass exhibited high sorption coefficients Kd and Koc compared to the rest of the soil. Reversibility of sorption was lower in the grassed soil than in the cropped soil and decreased rapidly with time. A rapid degradation of isoproturon was observed at different depths of the grassed soil whereas most of the herbicide remained nondegraded in the cultivated soil: half‐lives were respectively 72 d in the cultivated soil, and only 8 d in the superficial layer (0–2 cm) of the grassed soil. The highest mineralization rate of the isoproturon ring (20% after 35 d) was observed in the top layer (0–2 cm) having the highest mineralization rates of organic matter. In relation with this fast degradation, a large proportion of isoproturon residues became nonavailable to water and methanol extractions (54% of the initial applied isoproturon found as nonextractable (bound) residues). Thus the grassed strip surface soil had a high potential to dissipate isoproturon trapped from run‐off.
: Experiments on grassed bu †er strips have been conducted since 1993 by ITCF (Institut Technique des Ce re ales et des Fourrages) at three research farms (La Jaillie`re, Bignan and Ple lo). Literature data and conclusions drawn from previous work with isoproturon and diÑufenican were conÐrmed in a range of soil and cropping conditions : grassed bu †er strips are e †ective in restricting pollutant transfer in runo † ; those with widths of 6, 12 and 18 m reduced runo † volume by 43 to 99É9%, suspended solids by 87 to 100%, lindane losses by 72 to 100% and loss of atrazine and its metabolites by 44 to 100%. More than 99% of isoproturon and 97% of diÑufenican residues in runo † were removed by bu †er strips. Nitrate and soluble phosphorus in runo † were reduced by 47 to 100% and by 22 to 89%, respectively. At La Jaillie`re, a rainfall simulator was used in 1995 to verify that bu †er strips are still e †ective in conditions of intense runo †. Investigation of the inÑuence of sowing direction during the 1994È95 cropping period at Bignan showed that sowing perpendicular to the slope seemed to be beneÐcial in reducing pesticide content in runo †.
The pollution of ground and surface waters with pesticides is a serious ecological issue that requires adequate treatment. Most of the existing water pollution models are mechanistic mathematical models. While they have made a significant contribution to understanding the transfer processes, they face the problem of validation because of their complexity, the user subjectivity in their parameterization, and the lack of empirical data for validation. In addition, the data describing water pollution with pesticides are, in most cases, very imbalanced. This is due to strict regulations for pesticide applications, which lead to only a few pollution events. In this study, we propose the use of data mining to build models for assessing the risk of water pollution by pesticides in field-drained outflow water. Unlike the mechanistic models, the models generated by data mining are based on easily obtainable empirical data, while the parameterization of the models is not influenced by the subjectivity of ecological modelers. We used empirical data from field trials at the La Jaillière experimental site in France and applied the random forests algorithm to build predictive models that predict "risky" and "not-risky" pesticide application events. To address the problems of the imbalanced classes in the data, cost-sensitive learning and different measures of predictive performance were used. Despite the high imbalance between risky and not-risky application events, we managed to build predictive models that make reliable predictions. The proposed modeling approach can be easily applied to other ecological modeling problems where we encounter empirical data with highly imbalanced classes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.