Dynamic and harmonious allocation of irrigation water resources under climate change: A SWAT-based multi-objective nonlinear framework.

Efficient allocation of water resources in irrigation districts can alleviate regional water shortages and promote sustainable irrigated agriculture development. However, existing research on water resource allocation in irrigation districts does not address the lack of coordination within the "diversion-delivery-irrigation" chain for multiple water sources and users in a changing environment. Hence, poor water supply and demand matching, low efficiency and poor climate change responses pose challenges for efficient water resource allocation in irrigation districts. Therefore, this study couples the SWAT runoff simulation model with a multiobjective nonlinear programming model and proposes a weather-driven dynamic and optimal allocation model for multiple water sources. This model accounts for fluctuations in water supply and fine-tunes the allocation of water resources to different water sources, different channels and different crop fertility periods in the irrigation area. The model is designed to achieve synergistic improvements in water supply and demand, economic efficiency, equity in water distribution and efficiency in water use. The model was applied to the Qindeli Irrigation District in Heilongjiang Province. The results show that an increase in water supply at the head of the channel promotes a synergistic increase in economic efficiency and water supply and demand matching. This model can improve water use efficiency under water scarcity by reasonably optimizing the water use structure of the irrigation district. Compared with the traditional irrigation method, the optimized model saves 4 % of water and increases yield by 399 kg/ha, economic efficiency by 0.2 yuan per cubic meter of water, water use efficiency by 9 %, and water supply and demand matching by >80 % at all stages of fertility. The model ensures that water resources are allocated in an equitable manner at all levels.