Modelling nitrous oxide emissions from grazed grasslands in New Zealand

In sitit farm-scale measurements are a prerequisite for improving the accuracy of greenhouse gas inventories, and for developing and modifying the modelling approaches to quantify emissions. We describe here the changes made to the process-based model NZ-DNDC to allow us to simulate emissions of nitrous oxide (N2O) from typical New Zealand grazed grassland soils, and subsequent improvements to account for periodic changes in pasture growth, N input from animals, and water balance/soil moisture status. In grazed pastures, NO fluxes varied widely both spatially due to uneven excretal N inputs, and seasonally due to climatic conditions. The model was validated against field measurements from two dairy pastures with contrasting soils and from a sheep pasture. The model simulated effectively most of the soil water-filled pore-space (WFPS) and the general pulses and trends in NO emission from both the sheep- and dairy-grazed pastures, and also captured the observed effects of excretal and fertiliser N inputs. It also fairly reproduced the real variability in underlying processes regulating N2O emissions. A series of sensitivity tests conducted on NZ-DNDC showed the model predicted changes in pasture production and N2O emissions with changes in climate, soil properties, fertiliser management and grazing regimes. Pasture production was more sensitive to rainfall, temperature and initial soil carbon levels than to fertiliser additions, and stocking rate while NO emissions were strongly sensitive to rainfall, quantity and frequency of N inputs through both fertiliser and increased stocking rate, and initial soil carbon levels. (c) 2006 Elsevier B.V. All rights reserved.