The second COMPARE exercise: A model intercomparison using a case of a typical mesoscale orographic flow, the PYREX IOP3

Fifteen models have been evaluated for their ability to simulate the various phenomena of a mesoscale orographic flow sampled during the Pyrenees experiment (PYREX). A pure forecast exercise has been conducted and model performances were assessed using numerous observations. Two additional experiments were also performed in order to discriminate between small-scale errors and large-scale induced errors, and to discuss an optimal specification of model terrain height and roughness for use with envelope orography and effective roughness length parametrizations. The comparison results reveal systematic errors for all the models: the local winds are too weak, the mountain-wave amplitude is too large and the lee vortices are poorly represented. Since forcing by analyses did not correct the errors, they can be therefore mainly attributed to the model representation of orography. The blocking created by the model topography at low level is under-represented and the model topography does not sufficiently slow the flow. A positive consequence of the effective roughness length parametrization is to reduce the mountain-wave amplitude. Negligible benefit occurs from the use of an envelope orography parametrization. Although it favours the appearance of the lee vortices, the latter appear too early, the local winds remain too weak, and the mountain-wave amplitude is enhanced. The comparison of the computed pressure drag with the observed one is reasonably good for most of the models but the pressure drag is found to be more correlated to the lee vorticity than to the mountain wave.