Pre-flight Characterization and Fast Radiative Transfer Forward Model Development for the NPOESS Advanced Technology Microwave Sounder (ATMS) for Use in Operational Numerical Weather Prediction and Data Assimilation Systems

The NPOESS Advanced Technology Microwave Sounder (ATMS) is scheduled to fly on the NPOESS Preparatory Project (NPP), as well as NPOESS operational flight units C1 and C3. The ATMS is a lighter and lower power version of the current operational Advanced Microwave Sounding Unit (AMSU) and the Microwave Humidity Sounder (MHS) combined into a single sensor package, with an additional channel in the oxygen absorption region at 51.76 GHz and 3 additional channels in the water vapor absorption band at 165.5 and 183 GHz. Numerical Weather Prediction (NWP) and associated Data Assimilation Systems (DAS) routinely assimilate microwave radiance data directly using the model background state in conjunction with a fast radiative transfer model (RTM). The RTM maps the NWP model state variables into radiances. Prior to developing the fast RTM a Line-by-Line (LBL) RTM is developed using the best available sensor channel characteristics including central frequencies, passband shapes and widths, polarizations, polarization purity, scan geometry, antenna patterns, etc., in digitized form. Fast RTMs use coefficients generated via regression against a set of predictors, such as temperature and water vapor, which are used to produce transmittances that nearly match those of the LBL RTM for a wide sampling of real atmospheres. The accuracy of the LBL RTM is dependent upon the validity of the pre-flight characterization measurements. Historically, the detailed passband shape and polarization purity measurements were not produced digitally, and have been laboriously digitized from spectrum analyzer snapshots captured and included in engineering reports. Antenna pattern contours and their variation with respect to scan position may play a role in the observed scan biases. The status of the RTM development as well as the impact of the pre-flight characterization on the RTM accuracy will be presented.