Modelling the effect of aggregates on N₂O emission from denitrification in an agricultural peat soil

Abstract

Abstract. Nitrous oxide (N2O) emissions are highly variable in time, with high peak emissions lasting a few days to several weeks and low background emissions. This temporal variability is poorly understood which hampers the simulation of daily N2O emissions. In structured soils, like clay and peat, aggregates hamper the diffusion of oxygen, which leads to anaerobic microsites in the soil, favourable for denitrification. Diffusion of N2O out of the aggregates is also hampered, which leads to delayed emissions and increased reduction of N2O to N2. In this model simulation study we investigate the effect of aggregates in soils on the N2O emissions. We present a parameterization to simulate the effects of aggregates on N2O production by denitrification and on N2O reduction. The parameterization is based on the mobile-immobile model concept. It was implemented in a field-scale hydrological-biogeochemical model combination. We compared the simulated fluxes with observed fluxes from a fertilized and drained peat soil under grass. The results of this study show that aggregates strongly affect the simulated N2O emissions: peak emissions are lower, whereas the background emissions are slightly higher. Including the effect of aggregates caused a 40% decrease in the simulated annual emissions relative to the simulations without accounting for the effects of aggregates. The new parameterization significantly improved the model performance regarding simulation of observed daily N2O fluxes; r2 and RMSE improved from 0.11 and 198 g N2O-N ha−1 d−1 to 0.41 and 40 g N2O-N ha−1 d−1, respectively. Our analyses of the model results show that aggregates have a larger impact on the reduction than on the production of N2O. Reduction of N2O is more sensitive to changes in the drivers than production of N2O and is in that sense the key to understanding N2O emissions from denitrification. The effects of changing environmental conditions on reduction of N2O relative to N2O production strongly depend on the NO3 content of the soil. More anaerobic conditions have hardly any effect on the ratio of production to reduction if NO3 is abundant, but will decrease this ratio if NO3 is limiting. In the first case the emissions will increase, whereas in the second case the emissions will decrease. This study suggests that the current knowledge of the hydrological, biogeochemical and physical processes may be sufficient to understand the observed N2O fluxes from a fertilized clayey peatland. Further research is needed to test how aggregates affect the N2O fluxes from other soils or soils with different fertilization regimes.