Sensitivity Analysis for the Identification of Important Parameters in a Water Quality Model

Surface water quality models are always developed as universal frameworks so that they are flexible and can be employed to simulate a large variety of water bodies. As a result, they are often over-parameterized (more parameters than needed are used) and it is necessary to identify sensitive parameters when they are applied in the simulations of certain water bodies. Sensitivity analysis is an effective tool to accomplish this purpose. In this study, a simple approach consisting of the parameter perturbation method and the Morris method was proposed for the sensitivity analysis of the model parameters. It was illustrated with an application to the evaluations of all adjustable parameters in the simulation of the hydrodynamic and kinetic conditions in the Lake Maumelle in central Arkansas, USA. The obtained information associated with the parameters is important for the calibration, validation and application of the model.