In order to evaluate the impacts of assimilation of GPS (global position system)-PWV (precipitable water vapor) on the initial field and simulation of different precipitation types in Sichuan Basin, several control tests and assimilation experiments were performed using GSI (gridpoint statistical interpolation) and WRF (weather research and forecasting) model. Two cases, namely CR_2013 and SR_2015, representing the processes of convective precipitation and stable precipitation, respectively were analyzed. The results showed that a great amount of water vapor had accumulated in Sichuan Basin before the occurrence of rain, with the amount in the east being more than that in the west. The rainstorm centers of the two cases were located at the southern and northern gradient zones of GPS water vapor, respectively. The assimilation of GPS-PWV not only improved the humidity field but also ameliorated temperature and wind field to some extent. Although the locations and intensities of simulated precipitation in the control tests were generally similar to the observed precipitation, there was stronger simulated precipitation in small areas. The assimilation of GPS-PWV significantly improved the simulation of precipitation. The range of simulated rainstorm centers of CR_2013 was closer to the observation, while the simulation of SR_2015 significantly weakened the intensity and range of the fake rainstorm centers. The most significant adjustment was located at the gradient zones of GPS water vapor, which was consistent with the observed rainstorm center. The assimilation of GPS-PWV continuously influenced the simulated accumulated precipitation, and the adjustment range of convective precipitation was higher than that of stable precipitation.

GSI  /  GPS-PWV  /  WRF  /  3DVAR  /  moisture gradient