Improvements of the atmospheric general circulation model LMDZ in the perspective of the development of the new IPSL coupled model

In the course of the development of the IPSL coupled model, a significant work has been devoted during the last three years to the improvement of the atmospheric physics. Switching from Tiedtke (Mon. Wea. Rev., 117:1179-1800, 1989) to Emanuel (J. of Atmopsh. Sci., 48:2313-2335, 1991) convection scheme resulted in a better distribution of convective precipitations in the tropics, with a significant enhancement on Amazonia and Africa during the rain season. However, when using the original cloud scheme of LMD (Le Treut and Li, Clim. Dyn., 5:175-187, 1991), the model was strongly underestimating short wave radiative forcing of convective clouds, even in those regions were the precipitation itself was greatly improved. This problem was solved by introducing a more sophisticated parameterization (Bony and Emanuel, J. Atm. Sci., 58:3158-3183, 2001) based on a generalized log normal PDF of the subgrid scale distribution of total water (in place of the old top-hat distribution). More importantly, in this new parameterization, the width of the distribution is not imposed but computed interactively with the convection scheme so as to simulate the proper incloud water content. Finally, a particular care was given to a rewritting of an interface between a four-column boundary layer scheme and four sub-surfaces (ocean, continents, sea-ice and glaciers) for each GCM mesch, insuring a numerical conservation of energy and water. This interface is used for coupling with both oceans (ORCA) and continental surfaces ORCHIDEE). We present the impact of these recent updates focussing on those aspects which were crucial for the improvement of the climatology of the coupled model (joint poster) such as energy balance at the top of the atmosphere and simulation of surface fluxes, in particular as influenced by convection and clouds.