ABSTRACT
Nitrate leaching can lead to numerous environmental hazards; therefore, research on balancing water input, fertilization, N leaching and crop yield is necessary. Previous research has shown the ability of the DNDC model to accurately predict N leaching, which was a result of embedding the details of N transformation processes in the model. In this study, the N-leaching data obtained from previous experiments conducted in No. 3 field (Bush Estate) were used as validation data. This field consisted of 12 plots treated with different management scenarios from 1989 to 1992. After the calibration of crop growth parameters according to field measurements, the DNDC model was applied in this field for simulating nitrate leaching. Both the predicted and measured results indicate the value of 90–150 kg N ha−1 as the appropriate amount for fertilization to avoid N leaching. The comparison of the simulation and validation data shows that the simulated results were totally dependent. The simulated and observed RMSEs of the accumulated N-leaching amount in different plots for two sampling dates were in the range of 1.14–1.97. However, there were some problems that affected the accuracy of the DNDC model, including the underestimation of the N fertilizer effect and the overestimation of the water input effect.
