Sensitivity Analysis of Developed Hydrologic, Hydraulic and Stormwater Quality Model

Abstract

Sensitivity analysis of model parameters is the most effective way of evaluating the performances of complex hydrologic, hydraulic and stormwater quality models. Identification of the most influential parameters to model outputs are the important requirements for any modeling task. Sensitivity analysis is the suitable and effective platform to identify the key parameters that has significant effects on various phases of modeling such as optimization, calibration, validation, and uncertainty assessments. In this study sensitivity analysis was performed on seven calibration parameters including initial loss (IL), time of concentration (TC), reduction factor (RF), time-lag (TL), build-up rate (a), time exponent (b) and wash-off coefficient (k) of developed hydrologic, hydraulic and stormwater quality model in R software. The analysis results noted that for peak flow, total runoff volume and mean flow estimations, reduction factor is the most sensitive parameter whilst for total suspended solid (TSS) load estimation; build-up rate is the most sensitive parameters. The approach adopted in this study will help to assessing sensitive parameters enabling effective calibration of hydrologic, hydraulic and stormwater quality models and increase the accuracy in estimation of stormwater quantity and quality for urban stormwater management and hydraulic structure design.