As a large country, the impact of China's
agricultural practice on atmospheric N 2O would be highly significant. Field measurements
of N 2O emission from
agricultural land have been carried out. However, it is still difficult to estimate the total soil emission of N 2O in China from these field measurement because of the diversity of climate and cultivation practices that control soil N 2O emission. In this study, in order to estimate N 2O emission more accurately, a process oriented model DNDC (Denitrification and decomposition) was employed. DNDC model has been developed to predict soil carbon and nitrogen biogeochemistry, including N 2O and other trace gas emissions. DNDC was constructed with two components that reflect the two levels of driving force which control geochemical and/or soil biochemical processes related to trace gas emissions. The first component, consisting of soil conditions, crop growth and decomposition sub models, predicts soil temperature, moisture, pH, redox potential (Eh) and substrate concentration profiles based on ecological drivers (e.g., climate, soil, vegetation and anthropogenic activity). The second component, consisting of the nitrification, denitrification and fermentation sub models, predicts NO, N 2O, CH 4 and NH 3 fluxes based on soil environmental variables. Classical laws of physics, chemistry or biology, or empirical equations generated by previous studies were used in the model to parameterize each specific reaction. The entire model forms a bridge between basic ecological drivers and trace gas emissions. DNDC model has been tested against a number of field measurements worldwide. Most of the tests showed that DNDC was able to capture general patterns and magnitudes of N 2O emissions observed in field. To run DNDC using data files of climate, soil and farm management compiled in University of New Hampshire, we validated DNDC against 6 site studies in Nuancheng, Hebei Province, Shenyang, Liaoning Province, Agricutural experiment station and Changqiao Village of Wuxian County, Jiangsu Province, Suburban of Beijing and Guangzhou, Guangdong Province under different climate/soil/land cover/management conditions. DNDC could repeat the range of N 2O emissions observed in 4 sites. These results imply that DNDC may have modeled the fundamental processes influencing directions and rates of the key biochemical and geochemical reactions controlling N 2O emissions from agricultural soils, and thus may be suitable for application to estimate the N 2O emission in China. The result showed that under the climatic, soil and farming management conditions in 1990 0.31 (0 18～0.44) Tg N 2O N was emitted from 96 million hm 2 of cropland in China. The emissions in north and northwestern parts of China where the climate is dry, or moderately dry, or moderately wet were lower than 5.5 kg.hm -2 .a -1 . Higher emissions occurred in western Sichuan Province, northeastern China and Southeastern China because in the former two parts there is high soil organic matter and in the last part there is higher temperature enhancing soil nitrification and denitrification. To study the impact of anthropogenic activities on N 2O emissions at a regional scale, we designed a baseline scenario and several alternative management scenarios to model national N 2O emissions of China. The baseline scenario was set based on current farming practice, such as crop types and rotation, tillage, fertilization, manure amendment, irrigation and flooding. Alternative scenarios were generated by varying some of the components of management (e.g., zero, half or double of current fertilizer) or climate (e.g., increased air temperature by 2℃ or 4℃). The spatial pattern of soil N 2O emission higher in southeastern China than northwestern China was never changed by all alternative scenario. But the difference between alternative and baseline scenarios are always bigger in southeastern China than northwest