Sensitivity analysis aided calibration of urban drainage modeling

This study presents a new approach for event-based calibration of urban drainage models by coupling conceptual simulation of external sub-catchments with physical representation of underground drainage channels. The proposed approach is applied in two steps: first, the sensitivity analysis for the selection of the most influential parameters; second, two model calibrations one with the full set of parameters and the other one based on the reduced set of parameters. The optimization is performed by the Genetic Algorithm optimizer which enhances efficiency and accuracy in the calibration process. This approach considers a database of 14 events in the Cascina Scala urban catchment, in Pavia, Italy. The study developed urban stormwater drainage within the simulated environment of the SWMM implemented with MATLAB software to explore the potential benefits of the proposed method. Sensitivity Analysis results show Manning coefficient roofs, Manning coefficient streets, Depression storage impervious area, Manning coefficient conduit, Width coefficient, and Percent routed to be vital for their significant influence in describing peak discharge rate and runoff volume. The reduced-parameter calibration approach brings significant computational efficiencies while providing a high degree of accuracy regarding the simulation of hydrographs at the outlet. The results confirm that the proposed methodology can obtain a robust fit to experimental data with a reduction in computational costs compared to conventional calibration approaches. This study emphasizes the potential of sensitivity analysis-aided calibration to enhance the performance and reliability of urban hydrological models and to provide practical solutions for flood risk management and sustainable urban drainage design. The work is part of a larger project in which supervisors are involved. Previous investigations have been already arranged in a Journal paper to be submitted.