对流域生态补偿现状进行分析研究，从生态效果、经济水平、社会公平及政策投入 4 个方面构建流域生态 补偿效果评估指标体系；在此基础上，针对评估指标较多和专家评判时容易出现一致性差等问题，提出一种改进 SobolSSA-ANP?(sobol?sparrow?search?algorithm-analytic?network?process) 方法，并基于该方法对太湖流域的生态补偿 效果进行评估。结果表明：太湖流域生态补偿实施效果介于一般和较好之间，且倾向于较好；从 2010 年到 2020 年， 通过实施生态补偿政策，太湖流域生态补偿效果呈逐渐上升趋势；为进一步提升生态补偿效果，太湖流域需加大 资金投入和提升政策满意度。为此，流域可推动多主体参与的横向生态补偿政策以拓宽资金渠道，实现生态资源 到生态资本的转化。研究结果可以为流域生态补偿效果的评估和生态补偿政策的制定提供借鉴与参考，对提升 流域治理能力现代化水平具有重要的现实意义。

To improve the watershed ecological environment, and to realize the sustainable development of water resources, government and scholars have taken a series of investigations. Combined with the flexibility of payment for watershed ecosystem services and the efficient management of government, China proposed watershed ecological compensation and has piloted numerous ecological compensation programs in Xin'an Jiang, Taihu, and other watersheds. The implementation of ecological compensation has played an important role in the improvement of watershed ecology, and it has also impacted the economy, society, and other aspects of watersheds. Therefore, it is crucial to assess the impacts of ecological compensation on the ecology, economy, and society to further promote the implementation and improvement of ecological compensation policy.The evaluation index system, which includes the ecology, economy, social equity, and government investment is established by taking Taihu basin as a case study area. The ecology subsystem includes the per capita water resource, eutrophication index, and cyanobacteria density. The economy subsystem includes the total water supply, per capita disposable income, pollutant emission per unit gross domestic product, and the proportion of tertiary industry. The social equity subsystem includes the disposable income ratio between urban and rural, the health security number of beds per 10 thousand people. The government investment includes the investment in ecological protection. Based on the Dithering Sobol Sequence improved Sparrow Search Algorithm (SobolSSA) and Analytic Network Process (ANP), the evaluation model SobolSSA-ANP is constructed. In the proposed model, the ANP is adopted to analyze the influence relationship among indexes and to determine the evaluation matrix. The SobolSSA model is adopted to optimize the coefficients in the matrix dynamically. Subsequently, the optimized coefficients are input into the AHP model to reconstruct the evaluation matrix and to assess the eco-compensation effect.The results show that: (1) The SobolSSA-ANP improves the consistency and stability of the AHP model, and the evaluation results are more accurate. The CR values decrease by 1~10 orders of magnitude in the multiple ranges given by the experts, and the convergence results are more stable than the AHP model. (2) The implementation of Eco-compensation in the Taihu basin was effective and has improved the ecological environment. Especially, the effect from 2017 is significantly better than that before 2012. (3) The eco-compensation fund is the most important factor that influences the eco-compensation effect. The weight of the eco-compensation fund is 0.191 421, which is higher than the other indexes. Thus, it is necessary to explore market-based and multi-agent eco-compensation mechanisms to broaden the source of eco-compensation funds. (4) The eutrophication index and cyanobacteria density are the second and third important factors, respectively. That is consistent with the actual problems in the Taihu basin. The improvement of the above two factors will improve the ecological environment in the Taihu basin greatly.The implementation of eco-compensation in the Taihu basin has promoted the improvement of ecology, economy, and social equity. However, the eco-compensation funds, the design, and implementation of the eco-compensation mechanism required to be enhanced. It is suggested to promote innovation of eco-compensation mechanisms, explore the multi-agent and market-based eco-compensation, and establish a dynamic evaluation mechanism for the watershed. Meanwhile, establishing the ecological information-sharing mechanism, and transferring the ecological resource to ecological capital promotes the sustainable development of the watershed. The proposed model and results can provide a reference for the evaluation of the watershed eco-compensation effect, and are effective in improving the modernization of the governance capacity. In future research, the blockchain and marker-based eco-compensation mechanism can be coupled together to improve the trading mechanism of ecosystem services.