| To study the impact of contamination and possible remedial options on the Lake Ontario ecosystem, a comprehensive toxics mass balance study was conducted to determine the fate and distribution of several compounds of interest, namely PCBs, mirex/photomirex, DDT & metabolites, dieldrin, and 2,3,7,8-TCDD. In the study, a database of the chemicals of interest and the associated parameters for Lake Ontario was compiled first. Then the modeling work that had already been accomplished for toxic contaminants in the lake was thoroughly reviewed and evaluated. After that, a state-of-the-art chemical transport and transformation model, LOTOX2, was developed. Improved spatial, temporal and kinetic resolution was considered in LOTOX2 model conceptualization, which led to a better mathematical description of the system under study. An object-oriented ecosystem modeling library was developed to aid the model implementation. A long term 137Cs/solids mass balance calibration was conducted to insure accurate simulation of solids sedimentation and deposition-resuspension dynamics. Using the well calibrated sorbent dynamics, a carefully designed PCB hindcast was carried out for LOTOX2. The long-term PCB hindcast allowed refinement of PCB dynamics parameterization and reduction of uncertainty in making long-term predictions of the response of Lake Ontario to load control. Using similar approach, the hindcast calibrations were successfully applied to other chemicals of interest. A range of sensitivity studies was conducted on all calibration parameters to reveal the parameters/processes that has the most influence on the model prediction. The model prediction uncertainty was tentatively quantified through a Monte Carlo analysis. Finally, the LOTOX2 model was applied in diagnostic mode to enhance our understanding of the Lake Ontario ecosystem and to quantify the effect of the management options taken before and in the future. The study indicated that all the chemicals of interest are still responding to historical loading with sediment feedback being the most important internal loading source. It was also concluded that further PCB loading reductions will indeed produce in-lake benefits, but those benefits will not be evident for about 10 years. |
