The HYDRUS model is an efficient technical means to study the process of complex nitrogen (N) transport in farmland. Nonetheless, spatial variability in soil water and N parameters has been observed at the field scale, leading to uncertainties in the HYDRUS model. In this study, the HYDRUS‐1D model was corrected and verified by the measured data. Generalized likelihood uncertainty estimation (GLUE) was used to estimate the uncertainty and sensitivity of the model parameters. The differences in simulated and measured data were expressed in terms of root mean square error, normalized root mean square error and coefficient of determination, and the model calibration and verification indicated that the result was reliable. The soil saturated water content, dispersion and sorption coefficient were well identified, with values ranging from 0.42 to 0.45 cm3/cm3, 0.26 to 0.29 cm and 0.141 to 0.158 L/mg, respectively. The denitrification rate and saturated hydraulic conductivity greatly influenced the uncertainty in the model results. Considering the uncertainty of the parameters, the predicted leaching of N accumulation from 0 to 100 cm depth ranged from 9.5 to 12.7 kg/ha. The HYDRUS‐1D model can simulate water and N transport in paddy fields. The prediction interval can be narrowed by considering the uncertainty of the model parameters.