sensitivity analysis of the swmm model for runoff simulation in an arid urban catchment

Objective: hydrological models' performance is highly dependent on the accurate calibration of multiple parameters. sensitivity analysis can play a key role in optimizing this process by identifying the most influential parameters. the present study investigates the sensitivity of the storm water management model (swmm) in simulating urban runoff in shahroud, a city located in an arid region of iran. methods: the sensitivity of the swmm to changes in seven key calibration parameters, including the width factor of the subcatchment, curve number (cn), manning's roughness coefficients for pervious (n-perv) and impervious (n-imperv) areas, the percentage of impervious areas (imperv factor), and depression storage depth for both pervious (des-perv) and impervious (des-imperv) areas, was quantified using the sensitivity factor (s) proposed by morris. the study area was divided into 18 relatively homogeneous sub-catchments, with runoff from these areas drained through two outlets. during the sensitivity analysis process, peak discharge and runoff volume responses to these parameters were examined across seven rainfall events characterized by different intensities and durations.results: the results showed that peak discharge (qp) and runoff volume (rv) are sensitive to the imperv factor(s= 0.64 and 0.93, respectively). qp is generally sensitive to the width factor (s=0.17) and n-imperv (s=0.13), while showing no sensitivity to other parameters (s<0.05). longer rainfall durations increased the sensitivity of qp to both cn and the imperv factor, while sensitivity to other parameters decreased. cn significantly affected both rv and qp only during long-duration events. conclusions: longer rainfall durations increased the sensitivity of qp to both cn and the imperv factor, while sensitivity to other parameters decreased. cn significantly affected both rv and qp only during long-duration events.