Balancing Stakeholder Priorities and Environmental Dynamics in the Great Lakes: An Integrated Network Flow and Linear Regression Framework for Adaptive Water Management

The Great Lakes, a vital transboundary freshwater system, face escalating management challenges due to climate-driven hydrological variability and competing stakeholder demands. This study develops an integrated framework combining a network flow model with linear regression controls to optimize water levels while addressing ecological, economic, and societal priorities. The network flow model represents lakes as nodes and connecting channels as edges, incorporating stakeholder preferences through constrained optimization. Key innovations include seasonal decomposition of water flows and priority-weighted constraints for competing interests (e.g., shipping needs vs. ecological fluctuations). A linear regression control algorithm subsequently maintains optimal levels by modeling precipitation, evaporation, and river discharge relationships. Validation using 2017 hydrological data demonstrated improved performance over existing methods, with Lake Ontario scenarios showing enhanced equity across stakeholders. Sensitivity analyses revealed the model’s responsiveness to both managed controls (± 5–20% flow adjustments) and environmental variables. The framework reduced water level variance in Lake Ontario by 55% while preserving natural seasonal patterns critical for wetland ecosystems. This work advances water resource management by providing policymakers with a dynamic tool to reconcile human needs with environmental sustainability. Future integration of real-time climate projections and participatory stakeholder modeling could further strengthen decision-making in complex freshwater systems.