In urban areas, a large part of the land is covered by impermeable surfaces, which leads to the formation of surface runoff and floods and waterlogging in cities. To conduct this research, in District 12 of Tehran Municipality, using the rainfall parameter and physical characteristics of the basin through the hydrological-hydraulic model, SWMM, the amount of surface runoff in different return periods for three canals of Khayyam, shahrdary and 17 Shahrivar was estimated. Then, through 4 rainfall events on 12/11/2015, 1/12/2015 and 18/4/2016 the model was validated using 3 statistics of Nash-Sutcliffe, square root error and BIAS. The results show the efficiency of the model in estimating surface runoff. The results of sensitivity analysis showed that the percentage of impermeable areas has the most impact and the percentage of impermeable areas without surface storage have the least impact on peak discharges. Since the 17 Shahrivar tunnel in the eastern part of the study area does not have the ability to cross the runoff with a return period of 20 years, so the dimensions of the canal should be reviewed, which was estimated by trial and error method. In general, it can be concluded that the SWMM model has the accuracy required to simulate urban runoff and urban runoff management plans and design a surface water collection network in urban areas.