School of the Environment, University of Windsor, Ontario, Canada.
Department of Water Engineering, University of Tabriz, Tabriz, Iran.
Environ Res. 2024 Dec 15;263(Pt 1):120015. doi: 10.1016/j.envres.2024.120015. Epub 2024 Sep 14.
Monitoring water quality and river ecosystems is vital for maintaining public health and environmental sustainability. Over the past decade, data-driven methods have been extensively used for river water quality modeling, including dissolved oxygen (DO) concentrations. Despite advancements, challenges persist regarding accuracy, scalability, and adaptability of data-driven models to diverse environmental conditions. Previous studies primarily employed singular models or basic combinations of machine learning techniques, lacking advanced integration of adaptive mechanisms to process complex and evolving datasets. The current study introduces innovative hybrid models that integrate temporal pattern attention (TPA) mechanisms with advanced neural networks, including feed-forward neural networks (FFNNs) and long short-term memory networks (LSTMs). This approach leverages the synergistic strengths of individual models, significantly enhancing the accuracy of DO predictions. The models were rigorously tested against water quality data obtained from two distinct riverine environments, the Illinois River (ILL) and Des Plaines River (DP). Daily measured water quality data, including DO, chlorophyll-a, nitrate plus nitrite, water temperature, specific conductance, and pH, from 2017 to 2024 provided a robust foundation for comprehensive analysis of DO dynamics in these rivers. We conducted 10 scenarios with different model inputs, wherein the hybrid TPACWRNN-LSTM-10 model particularly excelled, achieving coefficient of determination values of 0.993 and 0.965, and root mean squared errors of 0.241 mg/L and 0.450 mg/L for DO predictions at the ILL and DP stations, respectively. The model's reliability was further confirmed by Willmott's index values of 0.998 and 0.992 and Nash-Sutcliffe efficiency values of 0.990 and 0.961 at the ILL and DP stations, respectively. Additionally, Shapley additive explanations (SHAP) values were utilized to interpret each predictor's contribution, revealing key drivers of DO predictions. We believe the novel hybrid modeling approach presented in this study could benefit utilities and water resource management systems for predicting water quality in complex systems.
监测水质和河流生态系统对于维护公共健康和环境可持续性至关重要。在过去的十年中，数据驱动方法已被广泛用于河流水质建模，包括溶解氧（DO）浓度。尽管取得了进展，但数据驱动模型在准确性、可扩展性以及适应不同环境条件方面仍然存在挑战。以前的研究主要采用单一模型或机器学习技术的基本组合，缺乏先进的自适应机制来处理复杂和不断发展的数据集。本研究引入了创新的混合模型，该模型将时间模式注意（TPA）机制与先进的神经网络（包括前馈神经网络（FFNN）和长短期记忆网络（LSTM））相结合。这种方法利用了各个模型的协同优势，显著提高了 DO 预测的准确性。模型经过严格测试，使用了来自伊利诺伊河（ILL）和德斯普兰斯河（DP）两个不同河流环境的水质数据。从 2017 年到 2024 年，每天测量的水质数据，包括 DO、叶绿素-a、硝酸盐加亚硝酸盐、水温度、比电导和 pH 值，为全面分析这些河流中的 DO 动态提供了坚实的基础。我们进行了 10 种不同模型输入的场景，其中混合 TPACWRNN-LSTM-10 模型表现特别出色，在 ILL 和 DP 站的 DO 预测中，分别达到了 0.993 和 0.965 的决定系数值，0.241 和 0.450 的均方根误差值。该模型的可靠性还通过 ILL 和 DP 站的威尔莫特指数值 0.998 和 0.992 以及纳什-苏特克里夫效率值 0.990 和 0.961 得到进一步确认。此外，还利用 Shapley 加法解释（SHAP）值来解释每个预测因子的贡献，揭示了 DO 预测的关键驱动因素。我们相信，本研究提出的新颖的混合建模方法可以为公用事业和水资源管理系统提供帮助，以预测复杂系统中的水质。
