水环境综合管理被普遍认为是促进流域可持续发展的基本手段。生态系统服务是生态系统对人类福祉和生存的直接和间接贡献, 是维持与影响流域水生态环境的重要因素。随着流域空间管控日益受到重视, 基于生态特征的"分区分类"管控已成为流域精细化管理的重要内容。论文基于生态系统服务功能与水质关联性, 采用典型对应分析(CCA)选取了土壤类型、土地利用、归一化植被指数和人口密度等4种生态特征作为水生态功能分区指标。结合流域生态特征、汇水区域和行政管理单元的空间分布, 通过聚类分析及空间叠置法, 将浙江省划分了10个水生态功能区和510个水环境控制单元。采用当量因子法对各控制单元的生态系统服务价值进行系统评估。结果表明2015年浙江省生态系统服务总价值为5123.7亿元, 其中控制单元的生态系统服务价值范围为59.8-136.5亿元。利用K-means聚类方法得到了4类生态服务簇, 并探讨了各类生态服务簇的服务类型、空间分布特征以及与各单元水质的关系。研究结果表明, Ⅰ类簇主要分布在西部及西南部的山地丘陵区, 其面积占了全省面积的52.68%;Ⅱ类簇的占地面积最小但水域面积相对较高, 主要覆盖千岛湖和钱塘江河口; Ⅲ类簇分布相对分散, 主要分布在丘陵平原地区; 而Ⅳ类簇主要分布在城镇较集中的平原地区。不同生态服务簇的主要生态服务类型存在明显差异, 如Ⅰ类簇以原料生产、水土保持、生物多样性保护、气体调节和气候调节为主; Ⅱ类簇的主要生态服务类型为调节功能, 包括气体调节、水文调节和废物处理; Ⅲ类簇和Ⅳ类簇以食物生产为主要的生态服务类型。水质特征与生态服务功能密切相关, 生态服务价值越高, 水体水质越好。因此根据各个生态服务簇的水质特征, 将各类生态服务簇划分为保护、维持、改善、整治4种类型, 并提出了不同生态服务簇内水环境控制单元的差异化管控措施, 以推进浙江省水生态环境的差异化管理。;Integrated water environmental management has been world widely treated as a fundamental means to support sustainable watershed development. Ecosystem Services (ESs), which refer to the direct and indirect contributions of ecosystems to human well-being and subsistence, are important factors for water quality maintenance and improvement. With increasing attention paid to the spatial management and control, zoning classification based on the ecosystem characteristics has become an important content of integrated watershed management. Based on the relationships between ecosystem characteristics and water quality, four ecological features including soil type, land usage, normalized vegetation index and population density were selected for aquatic ecoregion zoning using canonical correspondence analysis (CCA). Combining the spatial distribution of ecological features, catchments and administrative management units, Zhejiang Province was divided into ten aquatic ecoregions and 510 water environmental control units through cluster analysis and spatial overlap. The ES value of each unit was evaluated with the method of equivalent factor. The results showed that in 2015, the total ES value in Zhejiang Province reached 512.37 billion RMB, and those of individual units ranged from 5.98 to 13.65 billion RMB. Four ES bundles were obtained by K-means clustering method, and the service types and spatial distribution characteristics of the bundles were analyzed. Subsequently, the relationships between water quality and ecosystem services were analyzed for each unit. The results showed that Bundle I covered most of the mountain and hilly areas mainly located in the west and southwest, accounting for 52.68% of the province's total area, while Bundle Ⅱ had the smallest area but a relatively high water occupation ratio, mainly covering the Qiandao Lake and the estuary of the Qiantang River. Bundle Ⅲ was distributed relatively scattered, mostly in hilly and plain areas, while Bundle Ⅳ was primarily distributed in plain areas where cities and towns were located. The dominant ESs varied significantly in different ES bundles. For example, ESs such as raw material production, soil conservation, biodiversity protection, gas regulation, and climate regulation dominated in the Bundle I, while the regulatory services were the dominant ESs in Bundle Ⅱ, including gas regulation, water regulation and waste treatment. Food production contributed significantly to both Bundles Ⅲ and Ⅳ. The results also indicated that the characteristics of water quality are closely related to the ecological service, i.e., the higher the ES value, the better the water quality. As a result, the units were classified into four types, including protection, maintenance, improvement and remediation according to the characteristics of the water quality of each ES bundle. The differentiated water quality control and improvement approaches were proposed for the units in the four types of ES bundle, to support the water eco-environment management of Zhejiang Province.