Modelling approach for identifying management practices that reduce GHG emissions from cropping systems in Canada

Nitrous oxide (N2O) concentration in the atmosphere has risen from 270 ppb in 1750 (IPCC) to approximately 328 ppb in 2015. Higher N2O contributes towards increased global warming potential and stratospheric ozone layer depletion. Observations from research sites are rarely available to describe interacting soil-plant-atmospheric processes thus it becomes important to develop a tool which can simulate these processes and estimate interactions between agronomic practices and trade-offs in environmental outcomes. A Canadian version of the DNDC model has been under development since 2011, originally to improve the simulation of crop cultivars and agricultural management in cool weather climates but more recently to expand the models functionality for simulating additional management interactions and to integrate more detailed mechanisms. Focus has mostly been placed on improving the simulation of reactive N losses and hydrological processes rather than a direct focus on denitrification and N2O emissions. It was reasoned that these components needed to be improved before the more complex microbial and chemical processes effecting denitrification could be addressed. Several Canada DNDC developments were implemented which had notable impacts on soil C&N cycling and trace gas emissions including improved i) crop growth and development and winterkill impacts ii) temperature stress on crops and the effects of CO2 fertilization on C assimilation, water and N use efficiency, iii) we reformulated transpiration algorithms iv) a new sub-model was developed for predicting NH3 volatilization from slurry and urea, v) the ability of the model to characterize effects of management practices, snow cover, and soil texture on soil temperature was improved and vi) the model was restructured to simulated a deeper and heterogeneous soil profile, revised root density functions and quasi-2D tile drainage. Model developments have in general improved the simulation of N2O emissions even though the conceptualization of the denitrification process remains similar to the original model.