Testing DNDC model for simulating soil respiration and assessing the effects of climate change on the CO2 gas flux from Irish agriculture

Simulation models can be valuable to investigate potential effects of climate change on greenhouse gas emissions from terrestrial ecosystems. DNDC (the DeNitrification-DeComposition model) was tested against observed soil respiration data from adjacent pasture and arable fields in the Irish midlands. The arable field was converted from grassland approximately 50 years ago and managed since 2003 under two different tillage systems; conventional and reduced tillage. Both fields were located on the same soil type, classified as a free draining sandy loam soil derived from fluvial glacial gravels with low soil moisture holding capacity. Soil respiration measurements were made from January 2003 to August 2005. Three climate scenarios were investigated, a baseline of measured climatic data from a weather station at the field site, and high and low temperature sensitivity scenarios predicted by the Community Climate Change Consortium for Ireland (C4I) based on the Hadley Centre Global Climate Model (HadCM3) and the Intergovernment Panel on Climate Change (IPCC) A1B emission scenario. The aims of this study were to use measured soil respiration rates to validate the DNDC model for estimating CO2 efflux from these key Irish soils, investigate the effects of future climate change on CO2 efflux and estimate the efflux uncertainties due to using different future climate projections. The results indicate that the DNDC model can reliably estimate soil respiration from the two fields examined. The model underestimated annual measured CO2 efflux from the pasture by only13% (model efficiency: ME = 0.6; root mean square error: RMSE = 1.9 and mean absolute error: MAE = 6.3) and that from the arable conventional and reduced tillage by 9% (ME = 0.6; RMSE = 1.6 and MAE = 2.4) and 8% (ME = 0.23; RMSE = 1.8 and MAE = 2.9), respectively. Short-term land use change had no significant effects on CO2 effluxes from soil. Using the high temperature sensitivity scenario, future C effluxes would increase by 15% for the pasture and 14 and 16% for the arable conventional and reduced tillage systems, respectively. However, under the low temperature sensitivity scenario, lower increases in the C efflux of 6% for the pasture and 5% for the arable field were predicted. The calculated annual CO2 efflux uncertainties for using the high and low temperature sensitivity scenarios were 9% for the pasture and 8% for the arable field.