Assimilating humidity pseudo-observations derived from the cloud profiling radar aboard CloudSat in ALADIN 3D-Var

This paper describes an experimental procedure for assimilating CloudSat Cloud Profiling Radar (CPR) observations in ALADIN 3D-Var through the use of humidity pseudo-observations derived from a one-dimensional Bayesian analysis. Cloud data are considered as binary occurrences ('cloud' vs 'no-cloud'), which makes the approach feasible to be extended to other cloudiness observations, and to any other binary observation in general. A simple large-scale condensation scheme is used for projecting the prior information from a Numerical Weather Prediction model into Cloud fraction space. Verification over a I month assimilation test period indicates a clear benefit of the pseudo-observation assimilation scheme for the limited CloudSat CPR data set, especially in terms of improved skill scores for dynamical parameters Such as geopotential and wind. Copyright (C) 2009 Royal Meteorological Society