IPSL, Paris, December 19-20th, 2001
Joel
Ferziger Simulation of cloud topped atmospheric boundary layer We shall present simulations of the atmospheric boundary layer under various conditions. By making comparisons with DNS, we are able to show that the dynamic model performs very well without need of the adjustment required by the Smagorinsky model. We then do simulations of the boundary layer with a very simple cloud model-any moisture that cannot be retained as vapor is condensed and considered as `cloud.' Cases are done in which the cloud cover remains solid and others in which the clouds break up. The breakup is very sensitive to the model used. We also look at the so-called clous top entrainment instability mechanism and find that its predictions do not accord with the simulations. New entrainment correlations that perform much better are proposed. |