Parameter sensitivity analysis of SWMM: a case study of airport airfield area

Storm Water Management Model and Geographic Information System can provide prediction and management for airport flood problems. Efficient and accurate acquisition of sensitive parameters is the key to real-time model calibration. Due to the influence of special land types, functional zoning, and use requirements of airports, there are many problems in parameter sensitivity analysis, such as large sampling parameters, large amount of calculation, and nonlinear correlation between input and output variables. In this paper, the SWMM of airport airfield area is built, combining GIS and Python programming technology and using Latin hypercube sampling, and a correlation analysis method is proposed to study whether the input parameters have nonlinear correlation with the output results and its strength and compared with the improved Morris screening method. The results show that, the sensitivity of parameters is more balanced for the total inflow, there is no very sensitive parameter, and the nonlinear correlation between the parameters and the total inflow is weak. Manning-N is sensitive to average depth, hour of maximum flooding, and time to peak, which indicates that there is a strong nonlinear correlation between them and Manning-N. From the improved Morris screening analysis, it can be seen that there are no highly sensitive parameters for peak flow, and the sensitive parameters are Zero-Imperv and Manning-N. Highly sensitive parameters for time to peak are Manning-N, N-perv, S-Imperv, and N-Imperv. This paper quantitatively analyzes the influence of input parameters of the storm water management model on the output results, effectively identify the important parameters that affecting the output results, and analyze the nonlinear correlation between the input parameters and the output results. The results can greatly improve the accuracy of airport flood model and provide theoretical guidance for the application and parameter calibration of SWMM in airport.