| Abstrakt: |
Outflows from Lake Ontario are regulated at the Moses-Saunders Dam. The current control policy, Plan 2014, is the first of this system to use supply forecasts to inform release decisions. In this work, we use multi-objective optimization to identify alternative control policies to Plan 2014 and explore how these policies perform using forecasts at different lead-times, skill levels, and under plausible future climate scenarios. This information is necessary to understand how policy robustness varies with forecast attribute and how system performance across objectives can change in the future. Specifically, we create forecasts at 1, 3, 6, and 12-month timescales and vary skill between a baseline level and a perfect forecast. We then discover optimal policies using this forecast information under historic supply conditions that balance flood risk reduction, hydropower production, commercial navigation, wetland biodiversity, and recreational boasting. A subset of policies is evaluated under a stochastic and climate-change driven dataset of water supplies to assess policy robustness outside of historical supply conditions. Results from this work are directly supporting decision makers in their review of the current policy and in the formulation and evaluation process of a new operating policy for the system. [ABSTRACT FROM AUTHOR] |
