Comparison of Soil N-dynamics across Different Models

Soil nitrogen (N) is essential for plant growth, but in excess it can lead to environmental pollution via leaching and greenhouse gas emissions. A good understanding of soil N dynamics is therefore crucial for improving farm management. Field studies typically measure only some parts of the N cycle as temporal snapshots. Large uncertainties are involved when these observations are scaled up to farm system, regional or national scale and/or extrapolated to produce long-term estimates. Process-based models have the potential to simulate the entire N-cycle, thus enabling the integration of knowledge and improved estimates. However, such models contain a large number of processes that need to be parameterised and validated. Many models have been developed to simulate different aspects of the soil-N cycle. These models include different assumptions and simplifications, depending on the original purpose of the model. Comparing the behaviour of different models for the same dataset can be a useful way of determining which processes lead to significant differences in model output, and therefore where greater understanding of the underlying processes may lead to model improvement. Here we report on existing and planned comparisons between the NZ-DNDC, CenW, APSIM, and OVERSEER models. Introduction Soil N is essential for plant growth. However, in excess it can lead to environmental pollution via leaching, ammonia (NH3) volatilisation, and greenhouse gas emissions (particularly nitrous oxide, N2O). Nitrogen can occur in many chemical forms and there are a large number of interacting processes involved in its cycling through the environment. Figure 1 illustrates the major processes involving N in grazed pasture systems: