Main content area

Measurement and modelling of glyphosate fate compared with that of herbicides replaced as a result of the introduction of glyphosate-resistant oilseed rape

Mamy, Laure, Gabrielle, Benoit, Barriuso, Enrique
Pest management science 2008 v.64 no.3 pp. 262-275
Brassica napus var. napus, glyphosate, herbicide resistance, herbicide residues, soil pollution, trifluralin, metazachlor, field experimentation, laboratory techniques, prediction, rhizosphere, Pesticide Root Zone Model, pesticide persistence, weed control, spatial distribution, France
BACKGROUND: Crops resistant to glyphosate may mitigate the increasing contamination of the environment by herbicides, since their weeding requires smaller amounts of herbicides and fewer active ingredients. However, there are few published data comparing the fate of glyphosate with that of substitute herbicides under similar soil and climatic conditions. The objectives of the work reported here were (i) to evaluate and compare the fate in soil in field conditions of glyphosate, as used on glyphosate-resistant oilseed rape, with that of two herbicides frequently used for weed control on the same crop, albeit non-resistant: trifluralin and metazachlor, and (ii) to compare field results with predictions of the pesticide root zone model (PRZM), parameterized with laboratory data. Dissipation and vertical distribution in the soil profile of glyphosate, trifluralin and metazachlor were monitored in an experimental site located in Eastern France for 1 year.RESULTS: Herbicide persistence in the field increased as follows: metazachlor < glyphosate < trifluralin, contrary to laboratory results showing glyphosate to be least persistent. The main metabolite of glyphosate--aminomethylphosphonic acid (AMPA)--was more persistent than glyphosate. AMPA and trifluralin had the largest vertical mobility, followed by metazachlor and glyphosate. PRZM underestimated the dissipation rate of glyphosate in the field and the formation of AMPA, but its predictions for trifluralin and metazachlor were correct. The simulation of herbicides and AMPA distribution in the soil profile was satisfactory, but the mobility of trifluralin and metazachlor was slightly underestimated, probably because PRZM ignores preferential flow. In general, data from the laboratory allowed an acceptable parameterization of the model, as indicated by goodness-of-fit indices.CONCLUSION: Because of the detection of AMPA in the deep soil layer, the replacement of both trifluralin and metazachlor with glyphosate might not contribute to decreasing environmental contamination by herbicides. PRZM may be used to evaluate and to compare other weed control strategies for herbicide-resistant as well as non-resistant crops.