生态系统服务功能的可持续对区域生态安全和社会经济发展至关重要。黄河流域是我国重要的生态安全屏障，也是人口活动和经济发展的重要区域，陕西省地处黄河中游，是黄河流域生态保护和高质量发展的核心区域之一，受到气候变化和人类活动的影响，陕西省黄河流域生态环境问题突出。因此，对陕西省黄河流域生态系统服务功能进行科学评估，在此基础上探索环境保护和经济发展的协同路径，是推动该流域甚至整个黄河流域高质量发展的重要保障。基于此，本文以陕西省黄河流域为研究区，基于2000-2020年土地利用、自然地理和社会经济等数据，分析研究区在过去二十年快速城镇化背景下的土地利用格局演变规律，并结合InVEST模型评估产水量、土壤保持、生境质量、碳储量和粮食生产5种生态系统服务功能的时空变化特征，参考熵权-TOPSIS模型构建了单一和综合生态系统服务视角的生态风险评价模型，探讨生态系统服务功能对压力的响应，最后运用地理探测器分析不同生态风险空间分异的影响因素，并针对性地提出风险防控对策。主要结论如下：

（1）在2000-2020年，陕西省黄河流域的土地利用类型以耕地、草地和林地为主，土地利用转移也以这三者间的转移为主，研究期间土地利用变化明显，其中一直正向增长且增速最快的地类是建设用地。

（2）生态系统服务功能存在明显的时空分异。2000-2020年陕西省黄河流域产水量、土壤保持和粮食生产功能有所增强，而生境质量和碳储量功能有所减弱。从空间分布来看，产水量、土壤保持、生境质量和碳储量高值集中于植被覆盖高和自然条件优越的地区，包括秦岭山区和延安市的富县、黄陵县、黄龙县等地，粮食生产高值集中在农业生产条件便利的关中平原。

（3）2000-2020年陕西省黄河流域生态风险总体较高，但呈波动下降的趋势。产水量主要面临高生态风险，主要分布在陕北地区；土壤保持以高生态风险为主，高生态风险主要分布在榆林市和关中平原；生境质量主要面临中等生态风险，主要分布在榆林市东部、南部和延安市北部等地；碳储量主要面临中等生态风险，主要分布在榆林市东部、西南部和延安市的西部、南部以及关中平原等地；粮食生产主要面临高生态风险，高生态风险主要分布在流域北部榆林市、延安市及秦岭山区等地；综合生态风险主要面临较高生态风险，主要分布在榆林市东部及南部、延安市北部和关中平原等地。

（4）自然因子对生态风险的解释力强于社会经济因子，且不同因子交互作用的解释力强于单因子。影响产水量、土壤保持、生境质量、碳储量和综合生态风险空间分异的主要驱动因子是年降水量，影响粮食生产生态风险空间分异的主要驱动因子是年均气温。 基于生态风险的评价结果及实际情况，提出不同风险区的防控对策。建议低和较低生态风险区要保护为先、适度开发，全面深入推进对林地的保护；中等生态风险区设为重点预防区，强化农林复合生态系统的立体保护；较高生态风险区中对于黄土高原中部丘陵沟壑区要加强植被恢复、水土流失治理和水资源保护，关中平原要合理规划耕地和城市的空间布局以及巩固提升耕地生产功能；高生态风险区要持续进行荒漠化的综合防治，提升区域的防风固沙能力。

The sustainability of ecosystem service functions is crucial for regional ecological security and socio-economic development. As a vital ecological security barrier and a key area for human activities and economic development in China, the Yellow River Basin holds significant importance. Shaanxi Province, located in the middle reaches of the Yellow River Basin, is one of the core regions for ecological conservation and high-quality development in the basin. However, influenced by climate change and human activities, the ecological environment in Shaanxi's Yellow River Basin faces prominent challenges. Therefore, conducting a scientific assessment of its ecosystem service functions and exploring synergistic pathways for environmental protection and economic development are essential to promoting high-quality development in this region and the entire basin.This study focuses on the Yellow River Basin in Shaanxi Province, utilizing land use, physiographic, and socio-economic data from 2000 to 2020 to analyze the evolution of land use patterns under rapid urbanization over the past two decades. The InVEST model was applied to assess the spatiotemporal changes in five ecosystem services: water yield, soil conservation, habitat quality, carbon storage, and food production. An ecological risk evaluation model was constructed based on the entropy weight-TOPSIS method to explore ecological risks from both individual and integrated ecosystem service perspectives. Additionally, the study investigated the responses of ecosystem services to external pressures and employed a geographical detector to identify the primary driving forces behind different ecological risks. Targeted risk prevention strategies were proposed. The main conclusions are as follows:

(1) From 2000 to 2020, the dominant land use types in the study area were cropland, grassland, and forestland, with significant transitions occurring among these categories. Construction land exhibited the fastest and most consistent growth.

(2) Ecosystem services showed distinct spatiotemporal variations. Water yield, soil conservation, and food production increased, while habitat quality and carbon storage declined. Spatially, high values of water yield, soil conservation, habitat quality, and carbon storage were concentrated in areas with dense vegetation and favorable natural conditions, such as the Qinling Mountains and counties including Fu, Huangling, and Huanglong in Yan’an City. High food production values were clustered in the Guanzhong Plain, benefiting from superior agricultural conditions.

(3) The overall ecological risk in the basin remained high but exhibited a fluctuating downward trend. High risks for water yield were mainly distributed in northern Shaanxi; soil conservation risks were predominantly high in Yulin City and the Guanzhong Plain; habitat quality faced moderate risks in eastern and southern Yulin and northern Yan’an; carbon storage showed moderate risks in eastern and southwestern Yulin, western and southern Yan’an, and the Guanzhong Plain; food production risks were high in Yulin, Yan’an, and Qinling Mountains. Integrated ecological risks were elevated in eastern and southern Yulin, northern Yan’an, and the Guanzhong Plain. 

(4) Natural factors exhibited stronger explanatory power for ecological risks than socio-economic factors, and interactions between factors outperformed individual drivers. Annual precipitation was the primary driver for spatial variations in water yield, soil conservation, habitat quality, carbon storage, and integrated risks, while mean annual temperature dominated food production variations. Based on risk assessments and regional conditions, differentiated mitigation strategies were proposed: low-risk areas should prioritize conservation with moderate development; medium-risk zones require preventive measures and enhanced protection of agroforestry ecosystems; high-risk areas in the Loess Plateau’s hilly-gully regions need vegetation restoration, soil-water conservation, and rational land-use planning in the Guanzhong Plain; extremely high-risk regions demand comprehensive desertification control and sand stabilization efforts.