The Atmospheric Infrared Sounder (AIRS) aboard NASA's AQUA satellite provides vertical atmospheric soundings at near radiosonde quality at a resolution of 12 km. AIRS observations are assimilated using an Ensemble Kalman Filter. Our ensemble consisted of 72 9-km resolution WRF model forecasts. Ensemble perturbations consisted of different initial conditions, cumulus parameterization, boundary layer scheme, and moist physics. The objective was to improve forecasts of Mesoscale Convective Systems in the central United States. The period of study was May 2007.

For each of 15 cases, we produced 6 different forecasts: (1) no AIRS data, (2) all AIRS data, (3) only temperature data, (4) only moisture data, (5) all data at 500 hPa and below, and (6) all data at 700 hPa and below. Preliminary results indicate that the use of all AIRS data produced superior forecasts to using no AIRS data. Results also indicate that using data at 500 hPa and below produces nearly as good results as using all data at all levels. Using only temperature data, or only moisture data, resulted in less accurate forecasts than if all data was used.