Center of Water Resources and Environment, School of Civil Engineering, Sun Yat-Sen University, Guangzhou, 510275, China.
Center of Water Resources and Environment, School of Civil Engineering, Sun Yat-Sen University, Guangzhou, 510275, China.
J Environ Manage. 2024 May;359:121044. doi: 10.1016/j.jenvman.2024.121044. Epub 2024 May 6.
Dams and reservoirs have significantly altered river flow dynamics worldwide. Accurately representing reservoir operations in hydrological models is crucial yet challenging. Detailed reservoir operation data is often inaccessible, leading to relying on simplified reservoir operation modules in most hydrological models. To improve the capability of hydrological models to capture flow variability influenced by reservoirs, this study proposes a hybrid hydrological modeling framework, which combines a process-based hydrological model with a machine-learning-based reservoir operation module designed to simulate runoff under reservoir operations. The reservoir operation module employs an ensemble of three machine learning models: random forest, support vector machine, and AutoGluon. These models predict reservoir outflows using precipitation and temperature data as inputs. The Soil and Water Assessment Tool (SWAT) then integrates these outflow predictions to simulate runoff. To evaluate the performance of this hybrid approach, the Xijiang Basin within the Pearl River Basin, China, is used as a case study. The results highlight the superiority of the SWAT model coupled with machine learning-based reservoir operation models compared to alternative modeling approaches. This hybrid model effectively captures peak flows and dry period runoff. The Nash-Sutcliffe Efficiency (NSE) in daily runoff simulations shows substantial improvement, ranging from 0.141 to 0.780, with corresponding enhancements in the coefficient of determination (R) by 0.098-0.397 when compared to the original reservoir operation modules in SWAT. In comparison to parameterization techniques lacking a dedicated reservoir module, NSE enhancements range from 0.068 to 0.537, and R improvements range from 0.027 to 0.139. The proposed hybrid modeling approach effectively characterizes the impact of reservoir operations on river flow dynamics, leading to enhanced accuracy in runoff simulation. These findings offer valuable insights for hydrological forecasting and water resources management in regions influenced by reservoir operations.
水坝和水库极大地改变了全球河流的流动动态。在水文模型中准确地表示水库运行情况至关重要，但也极具挑战性。详细的水库运行数据通常难以获取，这导致大多数水文模型依赖于简化的水库运行模块。为了提高水文模型捕捉受水库影响的水流变化的能力，本研究提出了一种混合水文建模框架，该框架结合了基于过程的水文模型和基于机器学习的水库运行模块，旨在模拟水库运行下的径流量。水库运行模块采用了三个机器学习模型的集合：随机森林、支持向量机和 AutoGluon。这些模型使用降水和温度数据作为输入来预测水库的出流量。然后，Soil and Water Assessment Tool（SWAT）将这些流出量预测集成在一起以模拟径流量。为了评估这种混合方法的性能，本研究以中国珠江流域的西江流域作为案例研究。结果突出了 SWAT 模型与基于机器学习的水库运行模型相结合的方法优于替代建模方法的优越性。这种混合模型有效地捕捉了峰值流量和枯水期径流量。每日径流量模拟的纳什-苏特克里夫效率（NSE）有了显著提高，从 0.141 到 0.780，相应的决定系数（R）提高了 0.098-0.397，与 SWAT 中的原始水库运行模块相比。与缺乏专用水库模块的参数化技术相比，NSE 的提高范围为 0.068 到 0.537，R 的提高范围为 0.027 到 0.139。提出的混合建模方法有效地描述了水库运行对河流流动动态的影响，从而提高了径流量模拟的准确性。这些发现为受水库运行影响的地区的水文预报和水资源管理提供了有价值的见解。
