Uncertainty assessment of deterministic water quality model for a combined sewer system with the GLUE method

Water quality modeling of the urban drainage system is critical to the assessment of discharged pollutant loads and development of drainage system management. Here, application of water quality modeling to a combined sewer system on a complex network scale in Shanghai, China is presented. The uncertainty of the water quality parameters used in pollutant buildup, wash-off, and sewer sediment erosion/deposition models was assessed based on a well-calibrated water quantity model using the generalized likelihood uncertainty estimation (GLUE) method. Moreover, the influences of a single objective function and a multi-objective function on identification of water quality parameters within the GLUE method were discussed. The identification of water quality parameters was improved and the prediction uncertainty band was reduced when the multi-objective function approach was used. The multi-objective function approach is an effective alternative method to parameter identification and uncertainty analysis that will be useful to similar studies and applications into water quality modeling of combined sewer systems.