Field assessment of sequential herbicide mixtures on nitrogen cycle-related functions in soybean production

Summary

Herbicide applications worldwide generally intend to protect crop yields from weed interference while avoiding mechanical disturbance in the topsoil layer. Their intensive use often leads to cocktails of chemicals in soils, which may interfere with natural soil functions, and productivity. The effects of a sequence of herbicides typically used in soybean crops were tested on different endpoints related to the nitrogen cycle. The field experiment was repeated in 2020–2021 and 2021–2022, from fallow to harvest time of the soybean crop. The treatments were: a control without herbicides, a sequence of three herbicide applications at the label recommended rate, all the six possible combinations of the three herbicide applications, and a treatment with the typical, complete scheme but at twice the recommended rate. The first application was a mixture of glyphosate, dicamba, and clethodim; the second was a mixture of glyphosate, S-metolachlor, and flumioxazin, and the third herbicide application moment included fomesafen only. All the treatments remained weed-free, either by herbicides or manual removal. Significant negative effects were detected for nodulation in some treatments at the vegetative stage compared to the control, but no dose-dependent response was observed. Plant biomass and nodulation were not significantly related to herbicide mixtures at the reproductive stage, nor was the soybean yield at harvest time. These results engender a complex scenario for farmers to fully grasp the potential risks associated with the use of herbicides. However, potential nitrification was affected after the third herbicide application moment in the first year of the experiment, in all the treatments exposed to at least one herbicide application, while the abundance of ammonia oxidizers showed no effects. This comprehensive field assessment is relevant to evaluate herbicide environmental risks, accounting for plant-microbiome interactions under real pedo-climatic conditions and stress factors.