Sensitivity analysis of SWMM model parameters for urban runoff estimation in semi-arid area

Modelling of urban runoff is important for flood prevention and storm water management. For urban runoff modelling, estimate and collection of input parameters (measured and inferred) is very important, but accuracy of the results depend to the precision of the input data and calibration of the model that need to the highly detailed input data. Sensitivity analysis should indicate the parameters with greater effect on the results. In this study, sensitivity analysis of SWMM model parameters was done for urban runoff estimation in a semi-arid area located in the Northwest of Iran (Zanjan city watershed). According to results, depth of depression storage, percent of impervious area and Manning’s roughness coefficient of impervious area were the most sensitive parameters of SWMM that affect peak and volume of the runoff. The properties related to the previous surfaces such as curve number, Manning’s roughness coefficient and depth of depression storage have not a significantly effect on model outputs. The results of the goodness-of-fit test show the accuracy of the model outputs, consequently, SWMM souled propose for simulating the urban drainage systems in semi-arid area. NOF and NSC criteria indicate that the prediction errors are also well balanced. We can conclude that validated model can use for rainfall-runoff simulation in the area located in the semi-arid area.

Sensitivity analysis, Calibration, Urban runoff peak and volume, SWMM, Zanjan city watershed