Abstract: Hyperspectral infrared sounders such as Atmospheric InfraRed Sounder (AIRS) provide unprecedented global atmospheric temperature and moisture soundings with high vertical resolution and accuracy. The dimension-reduced projection four-dimensional variational data assimilation (DRP-4DVar) approach has been used to assimilate clear-sky AIRS sounding retrievals in a heavy rainfall storm over Changjiang and Huaihe River basin from 0000 UTC 22 July to 0000 UTC 23 July 2009. Atmospheric soundings of temperature and moisture from AIRS improve the precipitation forecast. Three experiments have been performed to simulate the heavy rain process: A control experiment with the initial conditions from NCEP-FNL, a single assimilation experiment, and a cycle assimilation experiment using AIRS sounding retrievals. Results from experiments show that humidity, geopotential height, and divergence of the initial field are enhanced through assimilating the AIRS temperature and moisture profiles. The 24-h increment in precipitation is consistent with the increment in humidity, geopotential height, and divergence. In this storm event, assimilation experiments have been able successfully simulate the synoptic situation leading to heavy rainfall; the location and intensity of the heavy rainfall event are better simulated when the AIRS data are assimilated. Furthermore, a cycle assimilation framework can absorb more observational data and performs better than a single assimilation framework.