Abstract: Using the Weather Research Forecast (WRF) model and WRF data assimilation system (WRFDA), sensitivity experiments are conducted to simulate a heavy rainfall caused by the coupling of the Tibetan Plateau vortex (TPV) and an abnormal-path southwest China vortex (SWCV) in Sichuan Basin with assimilation of operational sounding data and the intensive observed data obtained through the SWCV Scientific Experiment. The model simulations of precipitation and the vortex moving path using different initial conditions are compared, and the impacts on the structure evolution of the SWCV and associated precipitation by assimilating intensive sounding data are analyzed. The results show that the model capability for the forecast of precipitation and the moving paths of TPV and SWCV is improved by assimilating both intensive sounding data and conventional sounding data, while the forecast is less satisfactory with assimilation of only the conventional sounding data. By introducing intensive sounding data, the intensities of initial TPV and SWCV are enhanced due to the increased cyclonic perturbation on the initial wind field. In addition, the more unstable stratification adjusted by data assimilation makes it easy for the model to produce large precipitation with a magnitude that is consistent to observation at the initial time. The experiment reveals that the generation and maintenance of the SWCV at 700 hPa in this case not only rely on high potential vorticity in the upper levels, but also rely on increases in positive vorticity caused by horizontal convergence in the lower levels and its upward transport, which is favorable for the SWCV to develop and move. Latent heat release associated with precipitation at the beginning of the simulation also plays an important role in the development of SWCV. Results of the present study are helpful for better understanding the mechanisms for the development of SWCV and associated precipitation process.