Atmospheric temperature and water vapor information are the key parameters needed in the regional numerical weather predition (NWP) model for accurate convective storm forecasting. Radar measurements have been used widely in the regional NWP model and have demonstrated very useful resource, however, radar provides useful information mostly when the storm is initiated. In the earlier stage of the storm, the clear sky atmosphere is dynamically unstable, the spatial, temporal and vertical distributions of the atmospheric temperature and moisture information in pre-convection environment are important for NWP. The AIRS, IASI and CrIS provide temperature and moisture profiles with high vertical resolution and good accuracy; both AIRS single field-of-view (SFOV) soundings and science team sounding product are used in the regional WRF (Weather Research and Forecasting) model with 3DVAR, Ensemble Kalman Filter (EnKF), and 4DVAR assimilation techniques. The AIRS soundings are applied to severe storms such as convective precipitation and hurricane for forecast improvement. Results show that AIRS soundings with both WRF (Weather Research and Forecasting)/3DVAR and WRF/DART (data assimilation research test bed) assimilation and forecast systems improve the hurricane track and intensity forecasts, AIRS soundings also improve the precipitation forecast with 4DVAR assimilation technique.