Integrated Process Chain for Reservoir Inflow Prediction-Multi-objective Joint Optimal Scheduling-Risk-Informed Decision-making Considering Multiple Uncertainties Transmission

Significant climate change and human activities have decreased the stability of water resource systems, leading to multiple uncertainties in streamflow prediction, reservoir operation optimization, decision-making, and adaptive adjustments for water resource scheduling. Understanding the impact of uncertainties on reginal streamflow is necessary and crucial to identifying reservoir operation strategies and decision-making responses. We proposed an integrated systematic “inflow predition”– “reservoir operation”– “optimization”– “decision-making risk analysis” chain considering the transmission of multiple uncertainties.The uncertainty of streamflow prediction is disclosed based on error analysis and reservoir inflow process is simulated by stochastic scenario model. Then, the modified stochastic multi-criteria decision-making model were applied to identify the effects of inflow prediction on reservoir multi-objective operation and decision-making. Moreover, risk quantification indices were used to determine the uncertainty propagation and potential risks accumulated in the chain. We applied this framework to reservoir system in typical basins. The results indicate that the uncertainty of inflow predictions leads to stochastic process of reservoir operation and decision-making. The reservoir decision-making error risk is quantified and enhanced with deep uncertainty. We identified the preferred solutions for reservoir operation under different uncertainty levels with risk information to enhance the robustness of reservoir operation and decision-making.