How Planning and Management Can Benefit from Uncertainty and Risk Analyses: A Case Study on the Niagara River

Abstract

Management decisions and policy development are oftentimes limited by the incomplete knowledge of the phenomena under study and by the inability to quantify the effective reliability of the system to be managed. It is possible to circumvent these limitations by including in the management decision process, an analysis of the variability and uncertainty of the model results. One regulatory process that can benefit from coupling the traditional deterministic modeling with the uncertainty analysis is the computation of the total maximum daily load (TMDL). By definition, in the TMDL, the margin of safety (MOS) is added to account for the inherent variability of the natural phenomena and for the limitations of their mathematical representation. In practice selection of the MOS, rarely reflects the data variability or the model uncertainty. This paper suggests a scientific method to incorporate variability and uncertainty in the computation of the TMDL by estimating a MOS based on the risk of the system of exceeding a critical concentration. The work presented here reports the results of the uncertainty and risk analyses conducted using the modified Rosenblueth Point Estimate Method to estimate the variability of lead concentrations in the Niagara River. A TMDL for lead is also computed for the first time by choosing a MOS based on the uncertainty analysis results and a selected risk of exceeding the most stringent water quality criteria