Using DNDC model to simulate and predict changes in black soil organic carbon

The objectives of this paper were to study long-term patterns of change in black soil organic carbon (SOC) and the suitability of the DNDC model to predict SOC. Based on available data from pre-set long-term experiment in Gongzhuling, Jilin Province, 4 fertilization modes were used to analyze changes in SOC. The 4 fertilization modes included no fertilization (CK), only chemical fertilizer application (NPK), combing chemical and organic fertilizers application (NPKM) and chemical fertilizer with increased organic fertilizer application (M2+NPK). The data were also used to validate the DNDC model. The DNDC verification results showed that root mean square error (RMSE) of each treatment was less than 10%. This suggested that the simulated values were consistent with the observed values, and that the DNDC model was suitable for simulating SOC in the study area. Five factors (chemical fertilizer, organic manure, straw, temperature and precipitation) were used to test the sensitivity of the model. Organic manure application had the most significant and continuous effect on SOC content. Future SOC variation in the next 100 years was also predicted. The results showed that SOC content declined slightly (by 12.5%), from initial value of 13.2 g·kg^-1 to the predicted value of 11.55 g·kg^-1 in 2100 for the CK. For the NPK treatment, SOC content was stable and without any declining tendency. SOC contents of NPKM and M2+NPK treatments increased significantly to 24.4 g·kg^-1 and 27.6 g·kg^-1 in 2100, representing increases over the initial SOC content of 84.8% and 109.1%, respectively.