The initial and boundary conditions are critical to the numerical weather prediction (NWP) model. It is known that satellite observations can overcome the limitations of the terrain, especially over the oceans where conventional observations are difficult to obtain. Therefore, the use of satellite data will expect to improve those regions where lack of traditional observation. The Advanced Microwave Sounding Unit (AMSU) and Atmospheric InfraRed Sounder (AIRS) onboard NASA's EOS Aqua satellite, represent microwave and hyperspectral infrared observations, respectively. Both of them may provide atmospheric temperature and moisture soundings with complementary characteristics. For example, AMSU has the advantage to give cloudy retrievals while AIRS may retain the atmospheric gradient due to its finer high spatial resolution. Both data could estimate atmospheric thermodynamic state with substantial accuracy to improve high impact weather forecast In this study, we adopt the Weather Research and Forecasting (WRF) model and the community Gridpoint Statistical Interpolation (GSI) data assimilation system to evaluate the use of AMSU/AIRS retrievals for severe precipitation at Taiwan. The front, UTC 2016/01/05 22Z, is selected to demonstrate the benefit of using sounding data. The preliminary results shows a positive impact on total precipitable water while the time slope may need further investigation.