Biochemical modeling of the Nhue River (Hanoi, Vietnam): Practical identifiability analysis and parameters estimation

Ecological modeling of a complete river system integrates the knowledge of various disciplines. The essential problems are to estimate how many parameters can fit into a context corresponding to a specific site and how to find identifiable parameter subsets given by the experimental layout, while avoiding over-parameterization of the model. It is the aim of this paper to address this question for the parameters of a derivative version of the recently published case study of the river water quality model no. 1 (RWQM1 [Reichert, P., Borchardt, D., Henze, M., Rauch, W., Shanahan, P., Somlyódy, L., Vanrolleghem, P., 2001. River Water Quality Model no. 1 (RWQM1). II. Biochemical process equations. Water Sci. Technol. 43 (5), 11-30]) of the Nhue River and the periphery of Hanoi city (Vietnam). The selection of practically identifiable parameter subsets is discussed for typical boundary conditions as a function of the experimental layout and of the hydrological regimes (steady and unsteady states). The results show that for steady state conditions, the field determination of principal environmental variables on an annual basis, a maximum nine kinetic parameters (among a total of 51 parameters) appear to be identifiable. In an unsteady state condition, and with only three measured environmental variables, a maximum of three maximum kinetic rates (among a total of 42 parameters) can be identifiable. The identifiable parameter subsets in both hydrological regimes are subject to parameter estimation. The successful performance of the simulation of the post-estimated parameter model proves the usefulness of parameter estimation techniques employed in this study.