Assimilation of cloud affected radiances in NCEP GDAS

One of major challenges of data assimilation is incorporating the satellite measurements over cloudy regions. Radiance measurements over cloudy regions possibly provide us with the atmospheric total liquid/ice water content, humidity, and temperature. This additional information could help to improve the numerical weather forecast skill. NOAA is currently working to include cloudy radiance components in NOAA data assimilation systems. Tangent linear (TL) and adjoint (AD) models consistent with NCEP Global Forecast System (GFS) have been developed as a part of efforts to directly assimilate cloud and precipitation affected microwave radiances in NCEP Global Data Assimilation System (GDAS).

In the first part of this study, these TL and AD models are applied in the strong constraint in the GDAS to improve dynamic balance in the regions where moist processes are important. Differences in analyses before and after applying the moisture physics in strong constraint are compared and impacts in the forecast skills are discussed.

In the second part of this study, specific issues such as observations errors, bias correction, and quality control processes in assimilating cloudy radiances are discussed. Radiance data assimilation in cloudy regions requires rapid and accurate radiative transfer and radiance gradient models. For a vertically stratified scattering and emitting atmosphere, the Community Radiative Transfer Model (CRTM) was developed at the Joint Center for Satellite Data Assimilation (JCSDA). This CRTM is employed in this study to calculate radiances and Jacobians at various wavelengths for radiance assimilation under all weather conditions. Moisture physics model generated jacobians combined with CRTM generated jacobians are applied in GDAS system and preliminary impact studies are implemented.