Evaluation of EPA SWMM, ASSA and Sewer GEMS Models in Analysis of Urban Flood Collected by Surface Drainage Network (Case Study: Lar New City)

Abstract Introduction: Rainfall-runoff modeling is considered one of the important methods in the study of hydrology and environmental management, especially in urban areas, in which sudden floods lead to significant financial and human losses. Thereupon, various models have been developed for urban flood simulation, among which selecting the best, is of significant consideration in the literature. Methods: Runoff peak flow and hydroghs measured in three nodes of an urban drainage network were applied for both spatial and temporal evaluation of the SWMM hydrological model in EPA SWMM, ASSA and Bently SewerGEMS V8i softwares. Other rainfall-runoff models also evaluated were SCS TR-55, SCS TR-20, Rational, Dekalb Rational, Santabarbara UH models (in ASSA software) and SCS UH model (in SewerGEMS V8i software). The models were calibrated considering the measurements during three precipitation events, and then validated by the measured data of three other events. The Nash-Sutcliffe coefficient along with the BIAS coefficient, the Coefficient of Determination and the Root Mean Square Error were used as the efficiency indicators. Findings: The measured runoff peak flow and hydrograph were most compatible with those simulated by SWMM models of EPA SWMM, SewerGEMS V8i and ASSA softwares, respectively. Regarding the results of the partial parameter and the Spearman correlation coefficient methods, model outputs were most sensitive to the percentage of impervious areas, equivalent width, roughness coefficient of impervious areas, the depth of depression of impervious and pervious areas, the percentage of impervious areas without surface storage and the curve number, respectively. The model was not sensitive to the roughness of the permeable areas. The results suggest EPA SWMM as the software with more reliable simulation results for runoff management projects in the study area and urban basins alike.