Evaluation of a cloud system life‐cycle simulated by the Meso‐NH model during FASTEX using METEOSAT radiances and TOVS‐3I cloud retrievals

The evaluation of a simulation over the life‐cycle of a cloud system during the Fronts and Atlantic Storm‐Track EXperiment is made by comparing synthetic and METEOSAT observed radiances. The usefulness of a ‘model‐to‐satellite’ approach, combining an explicit cloud scheme implemented in a mesoscale model (the Meso‐NH model) with a detailed radiative‐transfer code, is shown. In clear sky, accuracy in the brightness‐temperature simulations are, respectively, about 4.5 K for the infrared channel and 2 K for the water‐vapour channel. In cloudy areas, comparison with the TOVS‐3I cloud parameters shows that these errors become twice as large. The occurrence of key patterns in the simulated infrared images confirms the good quality of the model forecast and the accuracy of the radiative code. On the other hand, the discrepancies identified by the approach show that the model overestimates the upper‐level cloud cover. It is suggested that this overestimation might result either from a too large water detrainment or from an overestimated water residence time in the upper atmosphere. Finally, using an ice‐phase parametrization in the cloud scheme significantly improves the simulated cloud cover, although this more advanced scheme does not correct the overestimation of upper‐level cloud cover.