Melanie Guittet Title: Modelling the galactic thick disc. Abstract: Since the discovery of the galactic thick disc in 1983, the determination of its characteristics such as the age or the mean metallicity has been improved. But there are still some undeterminacies: what is the relation between the thick and the thin discs? Whether or not these populations are completely independant? The galactic disc used to be modeled with two different density profiles which are representative of the thin disc and the thick disc (see the model of Juric et al. 2008 or the Besancon model: Bienayme et al. 1987, Haywood et al. 1997 and Robin et al. 2003). Recently Bovy et al. (2011) suggested that the disc would be composed of a series of continuous sub-populations characterized by their own metallicity, alpha elements, scale height and scale length. In Guittet et al. (2011, 2012 submitted) the study of CFHTLS (Canada-France-Hawaii Telescope Legacy Survey) and SDSS (Sloan Digital Sky Survey) data compared to the Besancon model showed that the observed disc stars seem to be continuous in the Z-[Fe/H] plane (Z is the vertical distance from the galactic plane) which is not the case in the model. This could be in agreement with what Bovy et al. found. I will present star count models having different parameters which I compared with the SDSS data (~ 14000 deg2). Two different models have been realised: one using parameters from Bovy et al. (2011) such as several sub-populations with different scale height and scale length. The second model contains the three main populations of the Besancon model (Robin et al. 2003), the thin disc, the thick disc and the halo, with unique scale height and scale length. The aim is to generate the most realistic model having a disc with properties as close as possible to the observed ones. This will improve our knowldege on the transition between the thin and the thick discs.