Long-term simulation of temporal change of soil organic carbon in Denmark: comparison of three model performances under climate change

The temporal change in soil organic carbon (SOC) was analysed over an 80-year period based on climate change predictions of four regional circulation models under the International Panel on Climate Change (IPCC) A1B emission scenario in the 21st century. A 20-year (1991–2010) set of observed climate data was used to form the baseline, and generate synthetic data for future scenario analyses. With increasing carbon dioxide (CO 2 ) levels, and under continuous winter wheat production with conventional tillage at different nitrogen (N) input rates, three crop-soil models were used to study the temporal changes of SOC. Results indicated that soil carbon (C) generally decreased over the simulation period. In addition, increased N losses through leaching and denitrification were estimated. Decline in soil C under continuous mono-cropping systems indicated increased focus on N fertilization strategies. The results also suggested significant interactive effect of N input rate and climate variables on soil C and denitrification in response to climate change. The uncertainty was addressed by including the crop-soil models in a mixed-effect analysis so that the contribution of the models to the total variance of random variation was quantified. Statistical analysis showed that the crop-soil models are the main source for uncertainty in analysing soil C and N responses to climate change.