The 30-reservoir controlled system of the upper Yangtze River (source to Yichang, including the lower Jinsha River cascade and the Three Gorges Reservoir) was optimized by Chen et al. (2019, Water 11(12):2542) with a monthly-step, four-objective model (power generation, water supply, ecology, shipping) in which flood control entered only as a hard constraint fixing reservoirs at their flood-limited water levels during the flood season; objective competition was classified only qualitatively as low, moderate, or high. The paper's closing statement asks for three linked refinements: more refined quantitative analysis of the relationships among the benefit objectives, flood control treated as a research objective with a smaller flood-season dispatching step, and coupling of hydraulic and hydrological models to the optimization. No study is known to perform the integrated task: a joint dispatching model of the whole giant system with flood control as an explicit fifth objective at daily-or-finer flood-season resolution and coupled routing, used to quantify pairwise trade-offs among all five objectives across the full Pareto front and to test whether the 2019 qualitative competition structure survives. That integrated characterization is the surviving open core.

Background

A large reservoir system serves functions that compete for the same water: hydroelectric power generation (which wants high, stable water levels and timely release through turbines), water supply to downstream users (minimum flows at prescribed control sections, i.e., fixed river locations where a flow demand must be met), ecology (keeping discharge near an ecologically suitable flow range for river biota), navigation or shipping (minimum channel depths and flows for cargo vessels), and flood control (keeping enough empty storage during the flood season to absorb design floods). Because these objectives conflict, there is no single optimum; instead, multi-objective optimization produces a Pareto front — the set of non-inferior (mutually non-dominated) operating plans, in which no objective can be improved without degrading another. Multi-objective evolutionary algorithms (MOEAs) approximate this front; NSGA-III, the variant used here, is a genetic algorithm that maintains a well-spread set of non-inferior solutions using a grid of reference points in objective space. In Chinese practice, flood control during the flood season is normally not optimized at all: reservoirs are held at or below a flood-limited water level (FLWL), a mandated upper bound on storage that reserves capacity for flood interception. The upper Yangtze system studied here — roughly 30 controlled reservoirs from the river's source to Yichang, including the giant lower Jinsha River stations (Wudongde, Baihetan, Xiluodu, Xiangjiaba), tributary reservoirs such as Pubugou, and the Three Gorges Reservoir — is the largest hydropower cascade on Earth and is jointly dispatched in practice, with substantial documented joint-operation benefits. Chen et al. (2019) built the first four-objective joint optimization of this whole system: a monthly time step over a typical year, water-balance simulation (storage updated from inflows, releases, and evaporation/infiltration losses, without channel flow routing), NSGA-III solutions for wet, normal, and dry design years, and flood control imposed as the FLWL constraint rather than as an objective. Their competition analysis classified each objective pair's interaction strength qualitatively as low, moderate, or high: power generation is the dominant competitor (most strongly with ecology), the water-supply–ecology relationship is competitive or synergistic depending on the magnitudes of the two flow demands at the control section, and shipping is only weakly coupled to the rest. Related work had already shown the shape of the missing pieces in smaller settings: balancing social, economic, and ecological benefits of Three Gorges flood-season operation (Huang et al. 2019, Journal of Hydrology), trading off flood control against power generation through dynamic control of the FLWL (Liu et al. 2019, Water), and flood-season multi-objective operation of the downstream Jinsha cascade by the same research group (Ni et al. 2019; Yao et al. 2019, Water). What none of these or later studies provides is the quantitative competition mechanism of the full system with flood control inside the optimization rather than outside it. Later literature developed the necessary ingredients — copula-based quantitative competition analysis on the full non-inferior set (Zhao et al. 2019, Water), competitiveness and competition-efficiency indices that quantify pairwise objective competition on a Pareto front (Wang et al. 2022, Water Resources Research), joint flood-control optimization of the six key middle/upper-Yangtze reservoirs at fine flood scale (Zha et al. 2021, Water), trade-off-based decision-making among non-inferior solutions by the source paper's own group (Ni et al. 2022, Journal of Hydrology), and coupling of a one-dimensional unsteady-flow hydraulic routing model with flood-control optimal operation for Xiangjiaba and Three Gorges (Yao et al. 2025, Water Resources Management) — but each with fewer objectives, a subsystem, or a single purpose. The unresolved question is whether, and how, the qualitative competition picture changes when the whole giant system is optimized with flood control as a competing objective at the fine flood-season time scale where flood control is actually decided.

Problem Statement

For the giant reservoir system of the upper Yangtze River — the roughly 30 controlled reservoirs from the river's source to Yichang, including the lower Jinsha River cascade and the Three Gorges Reservoir — construct and solve a joint multi-objective dispatching model that (1) treats flood control as an explicit fifth optimization objective alongside power generation, water supply, ecology, and shipping, replacing the current treatment in which flood control is only a hard constraint holding reservoirs at their flood-limited water levels during prescribed flood-season months; (2) uses a refined flood-season dispatching step of daily or finer scale (the scale at which flood-control fine operation is conventionally decided), instead of the monthly step, so that system benefit is optimized over the whole year; and (3) replaces pure water-balance simulation with a formulation coupled to hydraulic routing (unsteady-flow wave propagation and channel attenuation between reservoirs) and/or distributed hydrological simulation (interval inflows and evaporation losses). On this refined model, quantify the pairwise competition among all five objectives across the full Pareto front — replacing the qualitative low/moderate/high classification with measurable trade-off relations — for representative wet, normal, and dry inflow conditions, and determine how the competition mechanism changes relative to the monthly four-objective findings: whether power generation remains the dominant competitor, whether the water-supply-ecology relationship remains regime-dependent, whether shipping stays weakly coupled, and which new trade-offs or synergies appear once flood control competes as an objective. The scientific objective is the quantitative characterization of the five-objective competition mechanism of this giant system, which comprises the refined model, the quantitative pairwise trade-off characterization across the full front, and the explicit comparison with the qualitative conclusions of the monthly model; the refined model is the instrument.

References

  1. Chen, M.; Dong, Z.; Jia, W.; Ni, X.; Yao, H. (2019). 'Multi-Objective Joint Optimal Operation of Reservoir System and Analysis of Objectives Competition Mechanism: A Case Study in the Upper Reach of the Yangtze River.' Water 11(12):2542. DOI: 10.3390/w11122542. https://doi.org/10.3390/w11122542
  2. Ni, X.; Dong, Z.; Xie, W.; Jia, W.; Duan, C.; Yao, H. (2019). 'Research on the Multi-Objective Cooperative Competition Mechanism of Jinsha River Downstream Cascade Reservoirs during the Flood Season Based on Optimized NSGA-III.' Water 11(4):849. DOI: 10.3390/w11040849. https://doi.org/10.3390/w11040849
  3. Yao, H.; Dong, Z.; Jia, W.; Ni, X.; Chen, M.; Zhu, C.; Li, D. (2019). 'Competitive Relationship between Flood Control and Power Generation with Flood Season Division: A Case Study in Downstream Jinsha River Cascade Reservoirs.' Water 11(11):2401. DOI: 10.3390/w11112401. https://doi.org/10.3390/w11112401
  4. Zhao, M.; Huang, S.; Huang, Q.; Wang, H.; Leng, G.; Liu, S.; Wang, L. (2019). 'Copula-Based Research on the Multi-Objective Competition Mechanism in Cascade Reservoirs Optimal Operation.' Water 11(5):995. DOI: 10.3390/w11050995. https://doi.org/10.3390/w11050995
  5. Huang, L.; Li, X.; Fang, H.; Yin, D.; Si, Y.; Wei, J.; Liu, J.; Hu, X.; Zhang, L. (2019). 'Balancing social, economic and ecological benefits of reservoir operation during the flood season: A case study of the Three Gorges Project, China.' Journal of Hydrology 572:422-434. DOI: 10.1016/j.jhydrol.2019.03.009. https://doi.org/10.1016/j.jhydrol.2019.03.009
  6. Liu, G.; Qin, H.; Shen, Q.; Tian, R.; Liu, Y. (2019). 'Multi-Objective Optimal Scheduling Model of Dynamic Control of Flood Limit Water Level for Cascade Reservoirs.' Water 11(9):1836. DOI: 10.3390/w11091836. https://doi.org/10.3390/w11091836
  7. Zha, G.; Zhou, J.; Yang, X.; Fang, W.; Dai, L.; Wang, Q.; Ding, X. (2021). 'Modeling and Solving of Joint Flood Control Operation of Large-Scale Reservoirs: A Case Study in the Middle and Upper Yangtze River in China.' Water 13(1):41. DOI: 10.3390/w13010041. https://doi.org/10.3390/w13010041
  8. Wang, H.-R.; Li, F.-F.; Wang, G.-Q.; Qiu, J. (2022). 'Shrinking Pareto Fronts to Guide Reservoir Operations by Quantifying Competition Among Multiple Objectives.' Water Resources Research 58: e2021WR029702. DOI: 10.1029/2021WR029702. https://doi.org/10.1029/2021WR029702
  9. Ni, X.; Dong, Z.; Jiang, Y.; Xie, W.; Yao, H.; Chen, M. (2022). 'A subjective-objective integrated multi-objective decision-making method for reservoir operation featuring trade-offs among non-inferior solutions themselves.' Journal of Hydrology 613:128430. DOI: 10.1016/j.jhydrol.2022.128430. https://doi.org/10.1016/j.jhydrol.2022.128430
  10. Yao, L.; Peng, Y.; Yu, X.; Zhang, Z.; Luo, S. (2025). 'Improving Flood Control Optimal Operation of River-Type Cascade Reservoirs through Coupling with 1D Hydrodynamic Model.' Water Resources Management 39(7):3443-3466. DOI: 10.1007/s11269-025-04116-7. https://doi.org/10.1007/s11269-025-04116-7