Exotic prediction is an important content for predicting the occurrence of migratory pests. The simulations and prediction of the migration trajectory of pests is considered an ideal exotic prediction method that can reflect the spatio-temporal dynamics of the pest's migration. The brown planthopper (BPH), Nilaparvata lugens (Stål), is an important migratory pest that affects rice production in China and the accurate forecasting of its migration trajectory can provide a scientific basis for the early warnings and effective prevention of pest-related catastrophes. In order to select appropriate pest migration trajectory models with good accuracy, high resolution, and easy popularization, we screened a great northward migration event of BPH that occurred in Hongjiang City of Hunan Province at the beginning of July in 2006, as a typical case of BPH migration in China. The Weather Research and Forecast (WRF) Model, a popular mesoscale weather research and forecast model used both at home and abroad, was used to simulate and output the atmospheric background fields in high resolution combined with re-analyzed meteorological data from the National Center of Environment Predicting of the USA (NCEP). The forecasting variable fields output by the WRF model as meteorological inputs to drive the two trajectory models were coupled with the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) Model and the Flexible Particle (FLEXPART) Dispersion Model to predict the migration trajectories and landfalling areas of BPH populations. According to the backward calculation results, the differences between some trajectory parameters of the two models, such as the pest sources, migration paths, migration heights, migrating speeds and migration distance, were compared. Finally, the observed data for the habitat, feeding conditions, and three-dimensional atmospheric airflow fields of BPH's migration were used to verify the accuracy and precision of the simulations and calculations. The results suggest several important points. First, the two models show good consistent performance on the calculations of pest sources, migration paths (migration azimuth and flight direction), migration heights, migrating speeds and migration distance. However, the variations in the migration parameters in the FLEXPART model are larger than those in the WRF-HYSPLIT model. Second, as compared to HYSPLIT, the FLEXPART model can better represent the dynamical role of meso-scale weather especially convective processes in the migration processes such as takeoff, flying duration, and landing location, and more realistically simulate land surface processes, atmospheric turbulence structure, and undulating terrain and their impact on the migration of BPH populations. This is because the FLEXPART model has included more realistic parameterization schemes for convection, surface stress, and complex terrain in the trajectory calculations, whereas the HYSPLIT model does not. Third, the two models present reasonable simulations on pest sources, migration paths, and landfalling areas in terms of the selection of BPH populations for habitats and feeding conditions. However, the trajectories simulated by the FLEXPART model showed better agreement with the prevailing winds than the HYSPLIT model's simulations. The evidence can be seen from the three-dimensional airflow fields in which the emigration, flight, and landfalling of BPH populations occurred. Fourth, both HYSPLIT and FLEXPART models demonstrated strong operational forecasting capabilities of migratory pest occurrence.