Crop residues management impact on N2O emissions

Crop residues and their management are important and cross-cutting aspects of cropping systems. In European arable farming systems, crop residues can be exported from the plot to be used in various forms (litter, fuel, raw material, etc.). When they are not exported, they can contribute to carbon storage, physical protection of soils and to the crop nutrition, particularly nitrogen. Management choices between competing options can take into account the environmental impact of practices. However, their effects on N2O fluxes, a powerful greenhouse gas that degrades the tropospheric ozone layer, are still difficult to predict. This thesis aims to assess the influence of crop residues and their management on N2O emissions through two research questions: (i) What is the importance of the quantity and biochemical composition of crop residues and their management in predicting the amount of N2O emitted following their return? (ii) To what extent is it possible to predict N2O emissions and the effect of crop residues on them? Ten years of measurements from the long-term ACBB - field crops system were scrutinized using three approaches. The first one is the ranking of the importance of factors useful for predicting total emissions for a period after crop residues return. The second one compares average emissions related to the export or the return of cereals and pea residues at the scale of the crop succession. The third one uses of the STICS model to verify whether the effects of residues and their management on the dynamics of emissions were properly taken into account by the model. The results confirm the limited influence of crop residues and their management on N2O emissions in field crop systems where the majority of residues come from crops harvested at maturity. The influence of residues on emissions is a function of the biomass C/N ratio, and the use of soluble element inputs has the potential to improve the simulation of emissions occurring after residue return. Despite improvements, there remains a large amount of variance unexplained by the models. These results reinforce those obtained by meta-analyses and controlled experiments while strengthening the idea that predictions based solely on the amount of nitrogen supplied by residues are not satisfactory.