An Initial Evaluation of SSMIS Radiances for Radiance Assimilation Applications

The first Special Sensor Microwave Imager/Sounder (SSMIS) was launched on 18th October 2003 and combines the surface viewing channels of its predecessor (SSMI) with a range of temperature and moisture sounding channels. The accuracy requirements set by numerical weather prediction data assimilation applications for temperature sounding channels are demanding (~0.2 K). The post-launch Cal/Val program highlighted two small, but significant, sources of bias in the SSMIS radiance data associated with solar intrusions into the warm calibration load and with thermal emission from the main reflector. These effects have been studied and initial software mitigation measures are now in place. Data assimilation experiments at the Met Office have shown that the inclusion of SSMIS data reduces short range (day 1-3) forecast errors in southern hemisphere PMSL by 1-3%. Assimilation experiments at ECMWF show that the inclusion of SSMIS radiances from the lower atmospheric temperature sounding channels produces more than half of the impact of NOAA-15 AMSU. Further improvements are expected as the correction algorithms are refined in future