Uncertainty Analysis in Wastewater Engineering Modeling: Understanding the Limitations

Design engineers commonly rely on physical and mathematical approximations or models of real complex phenomena. During the development of these physical or mathematical models, simplifications and assumptions need to be made to develop such tools. It is unrealistic to expect that the modeling process will give an answer that is a perfect match to the real case, and it is, therefore, very important to understand the modeling limitations and uncertainties. In 1996, Kops and Vanrolleghen (1996), indicated that numerous researches had been undertaken with regards to the uncertainty analysis (UA) of the sewer systems and receiving waters; however, there had been very little UA research involving the wastewater treatment plants (WWTPs). This statement has not changed too much since then. One reason that uncertainty analysis has been under utilized in WWTP modeling (with the exception of sensitivity analysis) is that practical ways of assessing uncertainty have not been readily applicable. Several techniques have been described in the literature for many years; however, these techniques traditionally require long simulation times. With new developments in computers and software it seems feasible and appropriate to start conducting UA in wastewater engineering modeling on a routine basis. This paper focuses on the development of UA related to the application of numerical models in wastewater treatment process with the main objective of illustrating the use and application of this technique to understand the limitations of the modeling process itself. The paper applies a Computational Fluid Dynamics (CFD) model for the design and upgrade of settling tanks and illustrates through examples the use of historical data and Monte Carlo type simulations to assess the parametric uncertainties associated with the settling and flocculation properties, which are the type of uncertainty that most strongly affects the output of the CFD clarifier model.