Urban flood management by simulation of surface runoff using SWMM model in Gorgan city, Golestan Province-Iran

Water collection systems and disposal of superficial rain runoff are of the important components of urban planning and development, and any negligence in the system design could be problematic for human societies. In the superficial water collection networks design, design errors increase with the increment and extension the network. To reduce the error, the network design should be accurate. Computer models have greatly contributed to the accurate design and played a very useful role in achieving economical plan.
A study was done to determine the optimum dimensions of Gorgan city drainage network using hydrologic-hydraulic model SWMM. In this study, to calibrate the model, four rainfall events in dates 22/11/1390, 30/11/1390, 12/01/1390, 12/02/1390, were used and the speed of the corresponding runoffs in the chosen sub basin were recorded. In this study, Nash-Sutcliff, square root of error and bias error were used as model performance indices in the estimating peak discharge and flow volume.
The model calibration results showed that the simulated peak discharge and flow volume are in good correspondence with the observed values (BIAS%=11.44, RMSE 0.006 and NS=0.70) and the calibration results were used for the optimum values of parameters. To evaluate and test model validation, two rainfall events in dates 2/02/1391, 16/02/1391, were used. The evaluation results were confirmed the efficiency and accuracy of the model, so that the model performance indices were acceptable in both cases (BIAS%=8.01, RMSE 0.00043 and NS=0.69). After evaluating model performance, design rainfall with return period was calculated and model with design return period was performed in respect to the optimal parameters and dimensions of the drainage network. Peak discharge passing through the channels was identified based on design rainfall intensities. The adequacy of existing network to carry runoff with design return period was assessed with regard to model results based on presence or absence of added load and flood conditions at the nodes and channels. According to the model results, 40.11, 58.30 and 48.12 percent of drainage network channels possess critical and flood conditions in 2, 5 and 10 return periods respectively. Finally, the optimal dimensions of flood channels were determined by the model through trial and error test according to the type of optimal dimensions for each design return period.
The results of the study indicate that the model has the accuracy required for urban runoff simulation and it can be used for urban runoff management plans and designing superficial water collection and disposal networks.