Assimilation of satellite derived surface heating rates in a Numerical Weather Prediction model

In a simplified data assimilation scheme, surface temperature heating rates were used to adjust the soil moisture content in the limited area atmospheric model RACMO. These heating rates were derived from the geostationary MEIEOSAT satellite IR clear sky pixels after sunrise. In this paper we will discuss the applied cloud detection algorithm, atmospheric correction, specifications of the observation error and the formulation of the data assimilation method. Further, a case study applying the scheme here described will be presented, extending over the Southern and Central parts of Europe berween 1 May and 1 October 2000. Comparison of the model results and a large number of observed 2m temperaturcs at synops locations revealed a small but systematic improvement in the central part of Europe, but a slight deterioration over the Iberian peninsula. To some extend, this might be related to the uniform specification of the aerodynamic coupling between the surface and the atmosphere within the model.
Altogether, we may conclude that satellite data contain a useful additional signal to correct model soil moisture content and aerodynamic parameters simultaneously.