Effects of mixtures of herbicides on nutrient cycling and plant support considering current agriculture practices

García Carriquiry, I.; Silva, V.; Raevel, F.; Harkes, P.; Osman, R.; Bentancur, O.; Fernandez, G.; Geissen, V.


The use of mixtures of pesticides and consecutive pesticide applications challenge current regulations aimed at protecting ecosystem health due to unpredictable effects of complex and dynamic mixtures. In this study, we tested the ecotoxicological effects of mixtures of herbicides, applied following a real application scheme of soybean production on soil health in a mesocosm experiment. The experiment included two sequential applications; first, glyphosate + dicamba + clethodim, and 30 days later, flumioxazin + metolachlor. Commercial products were used at the recommended doses and at two other concentrations: half and double the recommended dose. Soybean plants were exposed to the herbicide-contaminated soil from the time of sowing to the beginning of pod formation. Half of the plants were harvested at the vegetative stage and the remaining plants at the reproductive stage to evaluate endpoints related to plant support and nutrient cycling. Plant biomass was significantly affected during the vegetative stage at the recommended and double the recommended dose, with the effects being mixture-dose dependent. Lower total and arbuscular colonization of mycorrhizas were also observed in double the recommended dose, and intermediate results were observed for the recommended dose. Nodule mass and phosphorous concentration in plants decreased with increasing herbicide doses. By the end of the experiment, nodule mass and total mycorrhizal colonization were low in the plants treated with double the recommended dose of herbicides. However, both endpoints reached similar values to the control at lower herbicide doses. Plant height and phenology were only lower at double the recommended dose during the experiment. The use of non-standard endpoints evidenced that important soil functions were transiently or permanently affected, while the realistic application scheme accounted for the impact of the management practice currently used. Pesticide risk assessment should therefore, incorporate both issues to effectively protect the ecosystems.