Quartet trade-off for regional water resources allocation optimization with multiple water sources: A decentralized bi-level multi-objective model under hybrid uncertainty
Yaru He,Yan Tu,Jun Liu,Hongwei Shi,Benjamin Lev
DOI: https://doi.org/10.1016/j.jhydrol.2023.129341
IF: 6.4
2023-03-04
Journal of Hydrology
Abstract:Multiple uncertainties from water resource allocation have posed new challenges to balanced regional development. On the one hand, in the context of global water shortage crisis, groundwater, as an important water supply module, is facing severe risk of shortage. On the other hand, the regional authorities need to coordinate the contradictory water demand brought by the heterogeneous water use objectives from multiple water use levels, and these demands are also subject to complex uncertainties. How to mitigate the risks posed by uncertainties, ensure regional water security and balance participants, is attracting the keen attention of decision-makers around the world. Therefore, this study establishes a decentralized bi-level multi-objective regional water resources allocation (RWRA) optimization model under hybrid uncertainty, aiming at an optimal quartet trade-off of equity, sustainability, equilibrium, and efficiency. To deal with the hybrid uncertainty in the model, water demand is set as a random fuzzy number, and the groundwater constraint is combined with fuzzy chance-constrained programming (FCCP). Then, the decentralized bi-level multi-objective model is solved by the bi-level interactive method based on the satisfactory solution incorporated with the chaotic-particle swarm optimization (CPSO) algorithm. The proposed model and solution approach are then applied to the Dongjiang River Basin in Guangdong Province of China to verify the validity and utility of the model. The results show that (1) the model can effectively meet the different objectives requirements of regional authority and sub-areas; (2) groundwater constraint parameters has a significant impact on the model results, especially the Gini coefficient decreases by 13.6% when groundwater is used as a water source supplement; (3) from single-level model to decentralized bi-level model, the satisfactory balance of the two levels increases from 0.1457 to 0.793, which indicates that the decentralized bi-level model makes the two levels more balanced than the single-level model. Based on the results, some management recommendations for participants are finally provided.
geosciences, multidisciplinary,water resources,engineering, civil