A coastal region is influenced by many meteorological phenomena due to air-sea interactions to intense large-scale wind systems. Mesoscale air flows in coastal regions are mainly determined by land-sea temperature contrast that drives land-sea breezes, and by the orography that drives mountain-valley breezes while the shape of coastline has an effect on mesoscale wind flow. All these phenomena strongly influence various scalar fields including moisture and air pollution concentrations. The Esino Valley (in the middle of Italy) is particularly interesting due to its composite orography and the presence of an inner city that can complicate small-scale circulation patterns due to its urban heat island effect. The Valley is surrounded by hillsides sharply rising to mountain height. Its climate is classified under sub coastal where there is an all year round sea breeze. Topography in the Valley area is fairly complex and the proximity to the coast implies a significant interaction between the sea and valley breezes. In such coastal areas, regional and local climate behaviours are strongly affected by microclimate phenomena that perturb the general atmospheric circulation. The aim of this study is to model the mesoscale circulation in this complex coastal region both to improve our understanding of the phenomena and to forecast them. The first step towards the modelling of convective, valley and/or sea breeze circulation for the Esino Valley was accomplished through the application of RAMS model for several events. RAMS is a non-hydrostatic model based on a terrain following coordinate system built around the full set of primitive dynamical equations governing atmospheric motions, and supplements these equations with optional parameterisations for turbulent diffusion, solar and terrestrial radiation, moist processes including the formation and interaction of clouds, sensible and latent heat exchange between the atmosphere and multiple soil layers. RAMS model proved to be suitable for numerical studies of sea breeze circulations and it is capable to capture diurnal variability of wind and convective cloudiness. For this study RAMS 4.3 was run with three nested grids, the finest of which was 1 km horizontally spaced, and the vertical distribution of atmospheric flow and meteorological parameters have been simulated considering effect of complex terrains. Initial and boundary upper air data have been downloaded from National Center for Environmental Prediction (NCEP) reanalysis products. Two cases of mesoscale circulation during the summer of 2001 are presented. The main requirement for the selection was the presence of calm conditions with weak synoptic forcing, like the high-pressure system that dominated over the selected area for both days. After completion of 72–hour simulations resulting fields of temperature, wind and cloudiness (cloud liquid water content) were examined. The interactions of land-sea convective cells and mountain-valley convective cells have been evaluated as a first step to determine both their respective importance and the importance of their interactions in determining moisture and pollutant concentration fields.