The brown planthopper, Nilaparvata lugens (Stål), is one of the most damaging pests affecting rice production in China. An outbreak of N. lugens can increase the incidence of disease and insect pest damage, both of which seriously affect the growth and the yield of rice. In order to analyze the influence of atmospheric dynamics, temperature, humidity, and other meteorological factors on the migration and density distribution of N. lugens over complex terrain, numerical simulations were performed. These simulations coupled the Flexpart model with the Weather Research and Forecast Model and the spatial analysis functions of GIS. These models and programs were used to simulate the N. lugens immigration and landing event that occurred at 6 plant protection stations, including Qujiang, Zhaoqing, Meixian in the Guangdong Province and Yichang in the Hubei Province, Ji'an in the Jiangxi Province, and Dongzhi in the Anhui Province, from September 30^th to October 3^rd in 2008 (BST). The simulation was in turn used to investigate the spatio-temporal distribution of the N. lugens' landing and density patterns as they passed through the complex terrain of China. (1) The simulation of the reverse trajectory of the insects showed that the N. lugens at all three stations all originated from sites in the northwest. As the site of origin was in an area of complex terrain, the migration distances of the N. lugens were much shorter, and the direction was changeable and unpredictable. (2) According to the results of simulations of forward trajectories, if the mountains were low and were with a valley as channel for N. lugens' migration, N. lugens could migrate along two mountainsides, lengthening their travel distance. If the mountains were high and there were no obvious valleys, migrating N. lugens could would directions and stay in their location of origin, because their route was blocked by the mountains. (3) When the slope of the mountains was steep, the density of the N. lugens was distributed along the mountains, because N. lugens could migrate along the mountain range. If the mountains consisted of multiple hills and peaks, N. lugens could pass through the areas of lower elevation, leading the density of N. lugens to be sporadically distributed. (4) The northeast wind allowed N. lugens to migrate further. Downdrafts coupled with high temperatures in this area, were conducive to N. lugens making landfall in higher densities. (5) To a certain extent, the result of the analysis of temperature distribution showed that N. lugens migrated towards warm areas. Ultimately, this led to the result that a higher density of N. lugens made landfall in warmer areas and along river valleys. In autumn, N. lugens landed in regions where the relative humidity was over 50%. There are some deviations from N. lugens' immigration trajectories, azimuth angles, and differences between simulated results and actual observations. These results indicate that this simulation can considerably enhance the ability of various operations to forecast of the movements of migratory pests in China. In future, we plan to improve the spatial and temporal resolution of the observational data of N. lugens, which will allow us to achieve more robust simulated results.