Due to increasing water demands globally, freshwater ecosystems are under constant pressure. Groundwater resources, as the main source of accessible freshwater, are crucially important for irrigation worldwide. Over-abstraction of groundwater leads to declines in groundwater levels; consequently, the groundwater inflow to streams decreases. The reduction in baseflow and alteration of the streamflow regime can potentially have an adverse effect on groundwater-dependent ecosystems. A spatially distributed, coupled groundwater–surface water model can simulate the impacts of groundwater abstraction on aquatic ecosystems. A constrained optimization algorithm and a simulation model in combination can provide an objective tool for the water practitioner to evaluate the interplay between economic benefits of groundwater abstractions and requirements to environmental flow. In this study, a holistic catchment-scale groundwater abstraction optimization framework has been developed that allows for a spatially explicit optimization of groundwater abstraction, while fulfilling a predefined maximum allowed reduction of streamflow (baseflow [Q95] or median flow [Q50]) as constraint criteria for 1484 stream locations across the catchment. A balanced K-Means clustering method was implemented to reduce the computational burden of the optimization. The model parameters and observation uncertainties calculated based on Bayesian linear theory allow for a risk assessment on the optimized groundwater abstraction values. The results from different optimization scenarios indicated that using the linear programming optimization algorithm in conjunction with integrated models provides valuable information for guiding the water practitioners in designing an effective groundwater abstraction plan with the consideration of environmental flow criteria important for the ecological status of the entire system.

中文翻译：

---

基于线性规划的地下水抽取多约束流域规模优化

由于全球对水的需求不断增加，淡水生态系统承受着持续的压力。地下水资源作为可获取淡水的主要来源，对全世界的灌溉至关重要。过度抽取地下水导致地下水位下降；因此，流入河流的地下水减少。基流的减少和河流流态的改变可能会对依赖地下水的生态系统产生不利影响。空间分布的、耦合的地下水-地表水模型可以模拟地下水抽取对水生生态系统的影响。约束优化算法和模拟模型的结合可以为水从业者提供客观的工具来评估地下水抽取的经济效益和对环境流量的要求之间的相互作用。在这项研究中，开发了一个整体的流域尺度地下水抽取优化框架，允许对地下水抽取进行空间明确的优化，同时满足预定义的最大允许减少的流量（基流 [Q95] 或中值流量 [Q50]）作为约束流域内 1484 个河流位置的标准。实现了平衡 K-Means 聚类方法以减少优化的计算负担。基于贝叶斯线性理论计算的模型参数和观测不确定性允许对优化的地下水抽取值进行风险评估。不同优化场景的结果表明，将线性规划优化算法与集成模型结合使用，为指导水资源从业者设计有效的地下水抽取计划提供了有价值的信息，同时考虑了对整个系统生态状况很重要的环境流量标准。 .