Runoff simulation of two typical urban green land types with the Stormwater Management Model (SWMM): sensitivity analysis and calibration of runoff parameters

The characteristics of surface runoff and the infiltration properties of urban green land are important to determine the effects of runoff reduction by low-impact development (LID) facilities. In this paper, two typical types of urban green land (lawn and shrub) in Shanghai were selected to study the runoff characteristics under eight rainfall events. The sensitivity of the runoff parameters was analyzed, and then, the optimal parameters were determined using the Stormwater Management Model (SWMM). The results showed that the interception and infiltration capacities of shrub were greater than those of lawn. The rainfall intensity and rainfall pattern were the major factors that influenced the interception and infiltration of rainwater. The threshold value that generates runoff varied across the eight rainfall events ranged from 1.6 to 28.5mm for lawn and 4.5 to 32.0mm for shrub. The maximum reduction ratios of runoff and peak flow for shrub were 52 and 57% higher than them for lawn, respectively. The parameters for shrub were more sensitive to runoff and peak flow compared with those for lawn. Under light rainfalls with a short duration, the maximum infiltration rate and depression storage were more sensitive than those under heavy rainfalls with a long duration. Antecedent dry weather period was not found to be a sensitive parameter except for the shrub under light rainfalls. The relative errors of runoff and dynamic mean runoff (60min) for lawn and shrub were within +/- 9.5%. The errors of peak flow ranged between -21 and 16.6%. The dynamic runoff characteristics and the parameters for lawn and shrub determined in this study can provide references for simulating urban runoff and planning LID areas.