社会-生态系统恢复力理论为可持续发展研究提供了新视角。从社会、经济、生态三大子系统,脆弱性和适应能力两大要素着手,建立社会-生态系统恢复力评价指标体系,采用集对分析法测度黄土高原2000—2018年各子系统及社会-生态系统恢复力,利用探索性空间数据分析法进行时空演变格局分析,并甄别社会-生态系统恢复力的主要影响因素。结果表明：（1） 2000—2018年社会-生态系统恢复力由0.522增强至0.721。社会恢复力由0.548增强至0.629后减弱至0.525;经济恢复力由0.401持续增强至0.850;生态恢复力由0.725减弱至0.607后增强至0.734,子系统恢复力演化趋势均不协同,经济系统与社会-生态系统恢复力演化趋势协同。经济系统恢复力的增强对于社会-生态系统恢复力增强具有显著促进作用。（2） 社会-生态系统恢复力出现显著空间集聚趋势,较高地区除省会城市和包头等能源富集区外,关中平原地区始终呈现高-高（H-H）集聚格局,其余地区恢复力普遍相对偏低。（3） 2000年以来黄土高原地区社会-生态系统恢复力生态维度障碍度始终高于社会和经济维度,不同地区指标层首要影响因素为人均GDP。

The resilience of social-ecological systems provides a new perspective for sustainable development research. In order to solve the problem of highly unbalanced development in ecologically vulnerable areas and achieve the goal of sustainable development, the resilience of social-ecological systems was studied by establishing an evaluation index system. The set pair analysis method was then used to measure the resilience of each subsystem and the social-ecological system in the Loess Plateau of China from 2000 to 2018. Finally, exploratory spatial data analysis was used to analyze spatio-temporal evolutionary patterns, and barrier degree models were used to identify the main influencing factors. As a result, we found the following: (1) social-ecological system resilience gradually increased from 0.522 to 0.721, with the Guanzhong Plain region always a higher value than others. The social resilience increased from 0.548 to 0.629 and then decreased to 0.525. The regions with high social resilience are mainly in provincial capitals such as Lanzhou and Tongchuan City, and the enhancement is closely related to policy support for western regions. Economic resilience increased from 0.401 to 0.850, and its distribution pattern was consistent with urban agglomeration. Ecological resilience decreased from 0.725 to 0.607 and then increased to 0.734. However, ecological resilience presented a gradual distribution from west to east, and its enhancement was closely related to policy and climate warming and humidification trends. This indicates that the evolutionary trend of each subsystem’s resilience was not coordinated, but the evolutionary trend of economic resilience was coordinated with the social-ecological system. Improving the resilience of economic systems significantly promotes the resilience of social-ecological systems. (2) There was a pronounced spatial agglomeration trend in the resilience of social-ecological systems. Except for the provincial capitals and energy-rich regions such as Baotou City, the Guanzhong Plain region always showed a high-high cluster. In contrast, the resilience of other regions was relatively low. The areas with low resilience were mainly concentrated in western regions such as Yinchuan City. In contrast, the areas with high resilience were concentrated primarily on the Guanzhong Plain such as Baoji and Xi’an City. (3) Since 2000, per capita GDP has been a major influential factor in all respects. The barriers to ecological maintenance of social-ecological resilience in the Loess Plateau were higher than the socio-economic aspects. Except for a slight increase in ecological barriers in individual areas, a gradually decreasing trend can be observed in most areas, indicating that ecological policies are successful and that ecological building is one of the important strategies to improve the resilience of the social-ecological systems.