COMPLEX TERRAIN S.Alessandrini1, E.Ferrero2,3, S. Trini Castelli3, D.Anfossi3 1CESI, Milano, Italy 2Dipartimento di Scienze e Tecnologie Avanzate, Universita’ del Piemonte Orientale, Alessandria, Italy 3ISAC-CNR, Torino, Italy Rivista: Special Issue of the International Journal of Environment and Pollution , 9th International Conference on Harmonisation within dispersion modelling for regulatory purposes, Garmish Germany Summary A dispersion simulation test for the RMS modelling system in real complex terrain is presented. RMS is used to simulate the wind and turbulence fields, to provide, starting from these fields, the necessary input parameters for the dispersion model and to perform dispersion simulations and to produce the concentration fields. As a test case the TRACT campaign was chosen, because this data-set allows to compare both the meteorological variables and the tracer concentrations. A particular attention was focused on testing alternative turbulence closures and assessing their influence on modelling both the meteorology and the dispersion. Firstly, the different turbulence closures introduced in the meteorological model are compared. Sencondly, the dispersion simulations are performed using the turbulence provided by the different models. The results obtained by using different RMS configuration for turbulence, compared with the experimental data, are then presented and discusssed. It is demonstrated that using different closure models for turbulence modifies also the mean flow. Concerning the tracer dispersion, the new closures utilising a fully 3D equation for the turbulent kinetic energy seem to give promising results.