Lakes have faced the challenge of alleviating water deterioration due to rapid socioeconomic development. This fact highlights the urgency to understand the impact of socioeconomic development on water quality. Although existing studies have devoted efforts on the research of the relationship between socioeconomic development and water quality, the theoretical basis and process of factors selection (socioeconomic factors and water quality factors) have often been ignored. A conceptual model was employed in this work to select the candidate indicators. Meanwhile, a new developed method that introduces a monthly correction coefficient was proposed to evaluate socioeconomic indicators and seasonal-trend decomposition (STL) using the locally weighted scatterplot smoothing (LOESS) method. At the same time, the steady-state transformation index (RSI) were adopted to further select water quality indicators. Base on that, canonical correlation analysis (CCA) and the measured and ideal values of water quality indicators were used to quantitatively study the degree of influence of socioeconomic systems and their key factors on water quality. The proposed methodology were applied to the Dongting Lake basin and its four sub-watersheds, the Xiang, Zi, Yuan and Li River watersheds, as these water bodies have been seriously affected by human activities in recent years. The results showed that there was a high correlation between the socioeconomic system and water quality, with significant temporal and spatial characteristics. The average annual impact degree of the socioeconomic system on water quality was at a moderate level for Dongting Lake basin and its four sub-watersheds, while the impact degree for Xiang River watershed was higher than that of the other watersheds. The influence degree decreased in the Dongting Lake basin, Xiang River and Li River watersheds, whereas the trend was the opposite in the Yuan River and Zi River watersheds from 2004 to 2017. Water consumption (WC), percentage of cultivated land area (CA), total nitrogen (TN) input from agricultural non-point sources (A_TN), and human response factors, such as industrial wastewater compliance discharge rate (WR) and sewage treatment (ST), were the main socioeconomic factors affecting the water quality of Dongting Lake. Implementation of an efficient water use plan, controlling the agricultural non-point source pollution and improving the sewage treatment capacity should be given priority in order to effectively improve the water quality of Dongting Lake. The present study showed the relationship between socioeconomic drivers and water quality, and provided a reliable tool for decision makers in watershed management.

由于社会经济的快速发展，湖泊面临着缓解水质恶化的挑战。这一事实凸显了了解社会经济发展对水质的影响的紧迫性。现有研究虽然致力于研究社会经济发展与水质的关系，但往往忽视了因子选择（社会经济因子和水质因子）的理论基础和过程。在这项工作中采用了一个概念模型来选择候选指标。同时，提出了一种引入月校正系数的新方法，以使用局部加权散点图平滑（LOESS）方法来评估社会经济指标和季节性趋势分解（STL）。同时，采用稳态转化指数（RSI）进一步选取水质指标。在此基础上，利用典型相关分析（CCA）和水质指标实测值和理想值，定量研究社会经济系统及其关键因素对水质的影响程度。由于近年来受到人类活动严重影响的洞庭湖流域及其四个子流域湘江、紫江、元江和漓江流域，所提出的方法学已被应用。结果表明，社会经济系统与水质之间存在高度相关性，具有显着的时空特征。社会经济系统对洞庭湖流域及其四个子流域水质年均影响程度处于中等水平，而湘江流域的影响程度高于其他流域。2004-2017年，洞庭湖流域、湘江、漓江流域的影响程度有所下降，而元江、紫江流域的影响程度则相反。 耗水量（WC），耕地面积百分比（CA） 、农业面源总氮（TN）输入（A_TN）以及工业废水达标排放率（WR）和污水处理（ST）等人类响应因素是影响洞庭水质的主要社会经济因素湖。实施有效的用水计划，以控制农业面源污染和提高污水处理能力为重点，有效改善洞庭湖水质。本研究显示了社会经济驱动因素与水质之间的关系，并为流域管理决策者提供了可靠的工具。