【摘要】 评估地表水水质变化和趋势,准确把握从上游流域到下游接收水体的水质成分的质量或负荷信息对于水资源的有效管理至关重要.近年来,伴随水质数据集长度的增加、统计方法的提升、计算机软件和硬件的改进,一系列统计方法用于探索和分析水质趋势和通量变化.时间、流量和季节加权回归模型(WRTDS)因其相对复杂和灵活性的特点,不断发展成为了长时间序列水质趋势分析的主要工具.本文围绕WRTDS模型在水质分析中的研究进展展开综述,总结了WRTDS模型基本原理(浓度/通量和归一化流量浓度/通量的估计)和发展状况,梳理了当前水质趋势和通量估计的方法和应用,汇总了WRTDS模型在长时间序列水质变化和趋势中应用情况和与其他模型对比情况.分析了WRTDS模型目前存在的不足,并结合我国实际水环境问题,展望了未来WRTDS模型可结合流域模型、遥感反演、人工智能等技术手段进行拓展和延伸,以期更好地指导未来流域水环境管理工作.更多还原

【Abstract】 Evaluating changes and trends in surface water quality and obtaining accurate information on water quality or nutrient loading from upstream to downstream in a watershed is crucial to the effective management of water resources. Recently, along with the increase in the length of water quality datasets, the upgrading of statistical methods, and the improvement of computer software and hardware, a series of statistical methods have been used to explore and analyze water quality trends and flux changes. Weighted Regressions on Time, Discharge, and Season(WRTDS) have evolved into the primary model for long-term continuous water quality trend analysis due to the relative complexity and flexibility. This article reviewed the research progress of WRTDS model in water quality analysis, summarized the basic principles(concentration/flux and normalized flux concentration/flux estimation) and development of the model, sorted out the current methods and applications on water quality trends and flux estimation, and outlined the application of the model to long-term water quality changes and trends and comparison with other models. Lastly, we analyzed some shortcomings of the WRTDS model, and combined with the actual water environment problems in China, we foresee that the WRTDS model can be extended and combined with watershed models, remote sensing inversion, artificial intelligence and other technical means to better guide watershed water environment management in the future.更多还原