MODIFICATION OF GREAT-LAKES REGULATION PLANS FOR SIMULATION OF MAXIMUM LAKE-ONTARIO OUTFLOWS

Nearly 50,000 years of Lake Ontario outflows were simulated and analyzed as a part of the Hydro-Quebec Beauharnois-Les Cedres spillway rehabilitation study. Historical Lake Ontario outflows could not be used for the study because of anthropogenic effects reflected in the data, the statistically short record length, and autocorrelation of the data. Stochastically generated Great Lakes net basin supplies were used in a Great Lakes hydrologic response model to obtain Lake Ontario levels and outflows. A significant problem was the lack of robustness in the Lake Superior and Lake Ontario regulation plans during simulations with extreme water supplies. The regulation plans were modified consistent with the International Joint Commission's regulation criteria and past operational actions to give reasonable results under these conditions. The simulated Lake Ontario levels and flows had a greater range than those resulting from historical water supplies. The maximum simulated Lake Ontario quarter-monthly level was 76.41 m (IGLD 55), and the maximum simulated quarter-monthly outflow was 14,160 m3s-1. The maximum-flow limitation of Lake Ontario's regulation plan, 8,780 m3s-1 was exceeded 0.27% of the time. The upper lake-level regulation limit, 75.22 m, was exceeded 0.15% of the time. It was found that some of the regulation criteria cannot be met simultaneously under extreme conditions. Historical water supplies, the current standard for the design and evaluation of modifications to the operational regulation plans, should no longer be the sole test of the plans.