Modeling of nitric oxide emissions from temperate agricultural soils

Arable soils are a significant source of nitric oxide (NO), most of which is derived from nitrogen fertilizers. Accurate estimates of NO emissions from these soils are essential to devise strategies to mitigate the impact of agriculture on tropospheric ozone production and destruction. This paper presents the implementation of a soil NO emissions submodel within the environmentally-orientated soil-crop model, CERES-EGC. The submodel simulates NO production via the nitrification pathway, as modulated by soil environmental drivers. The resulting model was tested with data from 4 field experiments on wheat- and maize-cropped soils representative of two agricultural regions of France, over three years, and encompassing various climatic conditions. Overall, the model provided accurate predictions of NO emissions, but shortcomings arose from an inadequate vertical distribution of N fertilizer in the soil surface. Inclusion of a 2-cm thick topsoil layer in a ‘micro-layer’ version of CERES-EGC gave more realistic simulations of NO emissions and under-lying microbiological process. From a statistical point of view, both versions of the model achieved a similar fit to the experimental data, with respectively a MD and a RMSE ranging from 1.8 to 6.2 g N–NO ha−1d−1, and from 22.8 to 25.2 g N–NO ha−1d−1 across the 4 experiments. The cumulative NO losses represented 1–2% of NH4+ fertilizer applied in the case of maize crops, and about 1% in the case of wheat crops. The ‘micro-layer’ version may be used for spatialized inventories of NO emissions to improve air quality prediction.