Potential Impacts of Assimilating All-sky Satellite Infrared Radiances on Convection-Permitting Analysis and Prediction of Tropical Convection

Geostationary infrared satellite observations are spatially dense and temporally frequent. This density and frequency suggest the possibility of using these observations to constrain sub-synoptic features over data sparse regions, such as tropical oceans. In this study, the potential impacts of assimilating water vapor channel brightness temperature (WV-BT) observations on tropical convection analysis and prediction were systematically examined through a series of ensemble data assimilation experiments.

The WV-BT observations from the Meteosat Visible Infra-Red Imager (MVIRI) were assimilated hourly into gray-zone resolution ensembles using the ensemble square root filter (EnSRF) within the Pennsylvania State University ensemble Kalman filter system. Comparisons against the independently observed MVIRI window channel brightness temperature (Window-BT) show that the assimilation of WV-BT generally improved the large-scale cloud patterns at wavelengths longer than 100 km. However, comparisons against soundings show that the EnSRF analysis produced a much stronger dry bias than seen in the no data assimilation experiment. Subsequent diagnostics suggest that the strongly enhanced dry bias is due to the use of the simulated WV-BT of the ensemble mean as the mean simulated WV-BT during the update step.

A new stochastic variant of the ensemble Kalman filter (NoMeanSF) is proposed here. The NoMeanSF algorithm was able to assimilate the WV-BT, without causing as strong a dry bias. Furthermore, NoMeanSF’s horizontal cloud pattern performance was comparable to that of the EnSRF. Deterministic forecasts initiated from the posterior ensemble means of both algorithms show that the forecasts initiated from the NoMeanSF analysis have better horizontal cloud patterns than that of EnSRF.