Assimilation of thermodynamic information in cloudy regions from advanced IR sounder for tropical cyclone forecasts

Assimilating cloudy radiances is very important in order to take full advantage of infrared (IR) sounder thermodynamic information in the tropical cyclone (TC) inner core region. Usually only clear IR channels (not affected by clouds) are used in most of data assimilation systems, and cloud contaminated channels have not been used effectively due to difficulties in modeling clouds in both forecast and radiative transfer models. An optimal imager/sounder cloud-clearing technique has been developed at the Cooperative Institute for Meteorological Satellite Studies (CIMSS) of University of Wisconsin-Madison, and can be used to retrieve clear column radiances through combining collocated multi-band imager IR clear radiances and the advanced IR sounder cloudy radiances, and no background information is needed in this cloud-clearing. The imager/sounder cloud-clearing is similar to microwave/IR cloud-clearing for deriving the clear equivalent radiances, but retains the original IR sounder spatial resolution, which is critical for regional numerical weather prediction (NWP) applications. Taking AIRS/MODIS as example, the assimilation of IR sounder radiances in cloudy regions with WRF/GSI has been conducted for hurricanes Sandy (2012) and Irene (2011) forecast experiments. Results on including the cloud-cleared IR radiances in partial cloud cover are encouraging, the track forecasts have substantial improvement and the intensity forecasts also have slight improvement. This technique can be applied to process CrIS/VIIRS and IASI/AVHRR for radiance assimilation in NWP.