A many-objective simulation-optimization framework for integrated water resources management in canal-well combined irrigation area based on FloPy and NSGA-III

The demand for agricultural products has been increasing in many developing countries all over the world due to improved living standards. It has generated significant rise in agricultural water consumption and ever-expanding irrigation area which are very sensitive to both climate change and farmers' irrigation and adaption strategies. In canal and well combined irrigation areas, because of ever increasing need for higher yield and intense variations in surface water supply and extreme weather conditions due to climate change, both groundwater over-exploitation in dry seasons and farmland inundation in wet seasons could happen. The ratios of canal irrigation (CI) to well irrigation (WI) in different regions with various cropping structures have to be carefully determined to avoid negative environmental impacts. To address these challenging and complex issues, this work developed a tightly coupled simulation-optimization framework based on FloPy and the many-objective genetic algorithm (NSGA-III) within the Pymoo framework. This approach enabled the evaluation of five key indicators: water productivity, total income, total water loss, equity (quantified by the Gini coefficient) and groundwater level fluctuation. The framework could provide actionable insights to support decision-making in ecological, environmental and economic contexts. The model was then tested and validated in Jinghuiqu irrigation area in Shaanxi Province (China) to obtain the Pareto Front for wet, normal and dry years. The key findings are as follows. (1) A reasonable canal-well combined irrigation water volume and proportion could be determined by the proposed simulation-optimization framework; (2) the average groundwater level fluctuations of recommend schemes are +0.82 m, +0.063 m, -0.26 m in wet, normal and dry years, respectively; (3) the ratios of canal-well of recommend schemes are 1:2.25, 1:1.19, and 1:1.74 in wet, normal and dry years; (4) different ratios of canal irrigation to well irrigation is crucial for sustainable groundwater development and managed aquifer recharge; (5) the results reveal distinct spatial-temporal variations across different administrative districts, thereby highlighting the considerable potential of the tightly coupled simulation-optimization model for integrated management of surface water and groundwater in climate sensitive canal-well combined irrigation area when dealing with many objectives. • Surface water and groundwater were conjunctively managed in spatial-temporal aspect. • A tightly coupled simulation-optimization model was developed based on FloPy-Pymoo. • Average groundwater level fluctuations are +0.82, +0.063, and − 0.26 m. • Recommended canal-well irrigation ratios of Jinghuiqu are 1:2.25, 1:1.19, and 1:1.74.