By measuring the thermal radiation emitted by the earth and its atmosphere in thousands of spectral bands, hyperspectral infrared sounders like AIRS (Atmospheric InfraRed Sounder) or IASI (Infrared Atmospheric Sounding Interferometer) provide information on temperature and chemical composition very useful for NWP (Numerical Weather Prediction). Unfortunately, as the impact of clouds on the measured radiances is important and difficult to model, the use of this data is limited to clear channels or fields of view.

In order to simplify the description of the cloud, which would require in principle the knowledge of vertical water content (liquid and ice) and cloud cover profiles, local effective parameters such as cloud top pressure, cloud over and possibly other microphysical parameters can be introduced to describe the impact of clouds on infrared radiances. Following encouraging results obtained in the context of theoretical one dimension variational assimilation studies, such an approach using effective local parameters was introduced in the 3D/4D variational assimilation code of the Canadian Meteorological Center.

In this poster, the results of some experiments using this approach with AIRS data will be presented and discussed.