黄土高原是我国气候变化敏感区，研究该区相对湿度（RH）时空演变及其与地理因子、气象要素之间的关系，有利于全面了解黄土高原气候变化规律。基于黄土高原及其周边地区90个气象站点逐日RH观测资料，利用趋势分析、敏感性分析和空间插值等方法，分析了1960—2021年及生态工程实施前后RH变化特征。结果表明：（1） 时间上，黄土高原地区年均RH为57.66%，以-0.376%·(10a)^-1速率显著减少（P<0.05），经历了“减弱-增强-减弱”年代际变化特征；除秋季以微弱幅度增大外，其他季节皆呈减少趋势，且减少幅度春季最高[-0.945%·(10a)^-1]，冬季最少[-0.194%·(10a)^-1]。（2） 空间上，黄土高原年及春夏秋季平均RH皆呈自南向北逐渐降低，而冬季表现为南部最高、中北部自东向西逐渐降低；年均及四季平均RH变化趋势空间态势各异。（3） 在黄土高原生态工程实施后，年及春夏冬季平均RH不同程度减少，年及夏冬季的趋势变化皆由增大转为减少；所有时序RH均值及趋势变化的空间特征存在较大差异，最典型的趋势变化组合类型是一致偏低型。（4） 引起黄土高原年均RH长期变化的主要季节因素是春季，空间上则主要是春季主导型、春夏季组合主导型。（5） 黄土高原年均及四季平均RH与纬度呈极显著负相关（P<0.01），与降水量呈极显著正相关（P<0.01），夏季平均RH受各地理因子的协同作用最为明显；年及春夏季平均RH对平均气温最为敏感，秋冬季对风速最为敏感。

The Loess Plateau, a region in China that is highly sensitive to climate change, serves as a focal point for investigating the spatiotemporal evolution of relative humidity (RH). Understanding the interplay between RH and geographical and meteorological factors is essential for comprehending the climate change dynamics within the plateau. This study leverages daily RH observations from 90 meteorological stations in the Loess Plateau and its environs to analyze the temporal and spatial variations in RH from 1960 to 2021, both before and after the implementation of ecological initiatives such as the conversion project of farmland to forest and grass. Using trend analysis, sensitivity analysis, and spatial interpolation, we unveil the following findings: (1) The average annual RH in the Loess Plateau exhibited a notable decrease of -0.376%·(10a)^-1 (P<0.05) from 1955 to 2021, undergoing a decadal variation characterized by a “weakening-strengthening-weakening” process. While autumn experienced a slight increase, all other seasons displayed a declining trend, with the most pronounced decrease observed in spring [-0.945%·(10a)^-1] and the least in winter [-0.194%·(10a)^-1]. (2) Spatially, the winter average RH in the Loess Plateau peaked in the south, gradually diminishing from east to west in the central and northern regions. Conversely, other time series demonstrated a gradual decline from south to north. The spatial patterns of the annual and seasonal average RH variation trends in the Loess Plateau were different. (3) Postimplementation of the ecological project, the average RH throughout the year, as well as in spring, summer, and winter, exhibited varying degrees of decrease. The trends in annual, summer, and winter RH shifted from an increasing to a decreasing trajectory. Noteworthy differences emerged in the spatial distribution characteristics of annual and seasonal average RH, coupled with their respective trend changes. The prevailing trend change combination type for all temporal RH patterns was consistently low. (4) The primary seasonal factor influencing long-term changes in annual RH in the Loess Plateau is spring, with spatial dominance primarily by a single dominant type in spring and a combination of dominant types in spring and summer. (5) The annual and seasonal average RH in the Loess Plateau demonstrated a significant negative correlation with latitude (P<0.01) and a positive correlation with precipitation (P<0.01). The geographical factors exerted the most significant influence on summer average RH. Annual, spring, and summer average RH were most sensitive to average temperature, whereas autumn and winter were most responsive to wind speed.