The influence of a mega-city on the atmospheric boundary layer and surface conditions was examined in the complex-terrain, semi-arid Tehran region using the Pennsylvania State University/National Center for Atmospheric Research fifth-generation Mesoscale Model (MM5) during a high pollution period. In addition, model sensitivity studies were conducted to evaluate the performance of the urban canopy and urban soil model "SM2-U (3D)" parameterization on the meteorological fields and ground level air pollutant concentrations in this area. The topographic flows and urban effects were found to play important roles in modulating the wind and temperature fields, and the urbanized areas exerted important local effects on the boundary layer meteorology. An emission inventory of air pollutants and an inventory of heat generation were developed and updated for 2005 in this work. Emissions from on-road motor vehicles constitute a major portion of the emission inventory and play the most important role in terms of contributions of air pollutants to the atmosphere in Tehran. By using a detailed methodology, we calculated spatial and temporal distributions of the anthropogenic heat flux (Qf) for Tehran during 2005. Wintertime Qf is larger than summertime Qf, which reflects the importance of heating emissions from buildings and traffic during cold and warm period respectively. Different urban parameterizations were used as a tool to investigate the modifications induced by the presence of an urban area in the area of interest. It was found that, for local meteorological simulations, the drag-force approach coupled with an urban soil model (DA-SM2-U) is preferable to the roughness approach (RA-SLAB). The comparisons indicated that the most important features of the wind, temperature and turbulent fields in urban areas are well reproduced by the DA-SM2-U configuration with the anthropogenic heat flux being taken into account (i.e., "DA-SM2-U Qf: On" option). This modeling option showed that the suburban part of the city is dominated by topographic flows whereas the center and south of Tehran are more affected by urban heat island (UHI) forcing especially during the night. The chemical transport modeling, including a model sensitivity study, was used to investigate the impact of the different urban parameterization on the dispersion and formation of pollutants over the Tehran region. Results show that applying DA approaches leads to significant improvements in the simulated spatial and temporal distribution of air pollutant concentrations in the city area and affects significantly the size of the urban plumes.