在生态补水工程的持续实施下，白洋淀水位逐渐恢复，生态环境有所好转，而2017年雄安新区的成立则进一步对白洋淀恢复和保持其“华北之肾”功能提出了新要求。为量化分析生态补水下白洋淀生态服务时空动态变化情况及其权衡协同关系，促进白洋淀在雄安新区建设中最大化其生态安全保障作用，本文基于土地利用数据，通过InVEST模型评价1991年、1996年、2005年、2015年及2019年白洋淀生境质量、碳储量及产水量服务，并基于SPSS和GeoDa软件进行相关性分析和空间分析，得到的主要结论如下：

（1）在生态补水及人类活动等因素综合作用下，近30年来白洋淀浅水区面积提升约40%，深水区面积有所减少，2019年时水域总面积已恢复至与1991年相当的水平；城镇用地、耕地面积在近30年内大幅提高，但由于退耕还淀政策影响，2019年耕地面积较2015年减少约16.5%。

（2）在生境质量上，五年平均分值为0.73、0.74、0.60、0.54及0.56，2019年有略微好转，但生境退化指数在五年中逐年升高，且栖息地质量较差的区域主要对应耕地、城镇用地及未利用地，人类活动对生境产生的威胁仍不可忽视。在碳储量上，五年碳储量总量分别为1.1225?Tg、1.1926?Tg、1.1608?Tg、1.0273?Tg、1.0275?Tg，其中2005至2015年平均碳密度减少幅度较大，达4.51?t/hm²，而2015至2019年平均减少幅度下降（0.07?t/hm²），表明白洋淀在生态补水作用下固碳功能在逐步恢复和提高。在产水服务上，五年总产水量体积分别为69.57×10⁶、84.06×10⁶、50.72×10⁶、56.45×10⁶、27.77×10^6?m³，产水量变化趋势与降水量变化趋势相同，与蒸散发量相反。总体上看生态服务高值区分布逐渐破碎化，总体生态服务呈现先恶化再逐渐缓解的趋势，2019年较2015年低值区比例有所下降。

（3）在生态服务权衡与协同关系上，五年间生境质量与碳储量均呈显著正相关关系，相关系数分别为0.117、0.158、0.625、0.461及0.429，表现出显著协同关系，产水量则与生境质量、碳储量既非权衡关系也非协同关系。在空间分布上，2019年生境质量及碳储量在空间上均有较为显著的聚集特征，单变量空间自相关莫兰指数（Moran's I统计量）分别为0.925及0.863，二者在空间上也呈现协同关系，双变量空间自相关莫兰指数为0.437。

本文通过对白洋淀生态服务时空动态变化情况进行定量评价，并进一步分析其生态服务间的权衡协同关系，量化了生态补水对白洋淀生态环境的影响，为白洋淀生态管理规划提供支撑。

In recent decades, the water level of Baiyangdian Lake has gradually recovered and the ecological environment has also improved due to the implementation of environmental flow releases, while the establishment of Xiong'an New Area in 2017 has further put forward new requirements for the restoration and maintenance of Baiyangdian Lake's function as "the Kidney of North China". To quantitatively analyze the temporal and spatial changes of the ecosystem services of Baiyangdian Lake and their trade-off and synergies, and to maximize its function of ecological security guarantee in the construction of Xiong'an New Area, this research is conducted based on land use data, mainly focusing on habitat quality, carbon storage and water yield services in 1991, 1996, 2005, 2015 and 2019 respectively. Firstly, the ecosystem services were evaluated by InVEST model, and then correlation analysis and spatial analysis were conducted using SPSS and GeoDa software. The main conclusions are as follows:

(1) With the combined effect of environmental flow releases and human activities, in the past 30 years, the area of shallow water has increased by 40%, while that of deep water has decreased dramatically. And in 2019, the total area of water has recovered to the level of 1991. Urban and cultivated land area had increased enormously by the end of the period, but that of cultivated land in 2019 decreased by approximately 16.5% compared with 2015 after the implementation of returning farmland to lake.

(2) In terms of habitat quality, the average score in five years was 0.73, 0.74, 0.60, 0.54 and 0.56, with a slight improvement in 2019. However, the habitat degradation index increased year by year in the five years. It is also noteworthy that the area with poor habitat quality mainly corresponded to cultivated land, urban land, and unutilized land. Therefore, the threat to habitat caused by human activities should not be ignored. In terms of carbon storage, the total carbon storage in five years was 1.1225?Tg, 1.1926?Tg, 1.1608?Tg, 1.0273?Tg and 1.0275?Tg respectively. The average carbon density decreased significantly from 2005 to 2015, at 4.51?t/hm². However, the average reduction from 2015 to 2019 decreased (0.07?t/hm²), indicating that the carbon sequestration function of Baiyangdian Lake is gradually recovering and improving with the implementation of environmental flow releases. As for water yield, the volume of total water yield in five years was 69.57×10⁶, 84.06×10⁶, 50.72×10⁶, 56.45×10⁶ and 27.77×10^6?m³ respectively. The variation trend of water yield was the same as that of precipitation, but opposite to evapotranspiration. Overall, the distribution of high-value ecosystem services has become more fragmented, and the total ecosystem services showed a trend of first deteriorating and then being gradually alleviated. In 2019, the proportion of low-value ecosystem services decreased compared with 2015. 

(3) There was a significant positive correlation between habitat quality and carbon storage, with correlation coefficients of 0.117, 0.158, 0.625, 0.461 and 0.429 respectively, indicating a significant synergistic relationship between them. However, there was no significant correlation between water yield and these two services. In terms of spatial distribution, there was a significant spatial aggregation characteristic in habitat quality as well as carbon storage, with the univariate spatial autocorrelation Moran's I of 0.925 and 0.863 respectively. The two services also show a synergistic relationship in terms of spatial distribution, with the bivariate spatial autocorrelation Moran's I of 0.437. 

This study quantitatively evaluated and analyzed the spatial-temporal dynamic changes of ecosystem services, analyzed the trade-off and synergy relationship among them, and quantified the impact of environmental flow releases on the ecological environment of Baiyangdian Lake. It can provide support for regional eco-environmental planning and management.