Parameter, Data Input and Structural Uncertainty Propagating through Coupled Models in a River Water Quality Modelling System

Abstract: This paper investigates three sources of uncertainty in a river water quality modelling system: parameters, input data and model structure. Emphasis is placed on structural uncertainty since this aspect has received less attention in the literature compared to parameter and input data uncertainty. Model structure is understood to be the equations and algorithms used to describe processes of substance transport and transformation in a model. Focus is given to two processes: i) sorption of heavy metals to suspended solids as a function of the fraction of organic carbon constituting the solids and ii) phytoplankton growth limitation by light extinction through the water column as a function of the concentration of suspended solids and chlorophyll-a in the water. Both processes are described by empirical relationships using regression curves derived from field data. The variance in the y-axis interceptions of the curves (also called regressor uncertainty) is taken to be the uncertainty bounds for the Monte Carlo simulations used for the uncertainty analyses. The three models of the WASP5 (Water quality Analysis Simulation Package), DYNHYD (hydrodynamics), EUTRO (dissolved oxygen, nutrient and phytoplankton dynamics) and TOXI (transport and transformation of sediments and micro-pollutants) were used for the study and were coupled in the HLA (High Level Architecture) platform to allow interactions between the models during simulations. The results show that structural uncertainty can be more significant than parameter and input data uncertainty to the overall uncertainty ranges of the model output, especially if the processes are very sensitive to the state variables. In addition, the behaviour of uncertainty propagation through a system of models is investigated