Combining simulation modelling and statistical power analysis to evaluate and optimize contaminant monitoring programs in Lake Ontario

This study combines a Lake Ontario chemical fate food-chain and bioaccumulation model with statistical power analysis to identify effective polychlorinated biphenyl (PCB) monitoring strategies in Lake Ontario. The main steps of the research were to apply the ecosystem model, conduct a statistical power analysis on the ecosystem model output, and conduct an analysis of the sensitivity of various existing monitoring programs to selected parameters.;Three loading scenarios were introduced into the ecosystem model to produce quantitative predictions of total PCB concentrations in water, sediment, lake trout, smelt, and herring gull eggs, over a hypothetical time-line of 100 years.;A statistical power analysis model developed in SAS (Statistical Analysis System), was applied to evaluate the ecosystem model output and to determine the time it would take for contaminant monitoring programs of various environmental compartments (i.e. water, sediment, lake trout, smelt and herring gull eggs), to achieve a minimum statistical power of 0.8 to detect specified decreases in PCB loading. (Abstract shortened by UMI.).