A Novel Environmental Flow Modeling Calibration Approach based on Multi-Objective Performance Characterization

Abstract:

Human interventions are significantly altering natural streamflow and the connectivity of freshwater ecosystems, which is a major issue all over the world. As a result, water resources managers are increasingly dealing with the allocation of environmental flows to balance water demands and ecosystem health. Hydrological models are widely used by water resource managers to address this complex issue. These models are typically used to simulate changes in streamflow regimes due to human interventions in land-use and environmental conditions. However, since the ability of hydrologic models to represent all aspects of the flow regime is limited, they are commonly used to address some aspects of environmental flows while ignoring others. To advance towards a more comprehensive approach, we propose a framework to evaluate different characteristics of flow regimes during model calibration in order to develop more robust model outputs. Here, we include the use of traditional efficiency metrics, error transformations, and functional data analysis. In the process, we identify the best-suited method in representing each environmental flow component (i.e., magnitude, duration, frequency, rate of change, and timing). Then, by implementing a multi-objective calibration approach and by constraining the error level for critical hydrologic indicators, we obtain sets of optimal solutions to improve the overall streamflow regime representation. The proposed framework here can help decision-makers and stakeholders to develop more robust management strategies that can shape water resources policies at local and regional scales.