Abstract: Objective　The purpose of this study is to comprehensively analyze the long-term spatiotemporal evolution of aerosol optical depth (AOD) in the Fenwei Plain, a region recognized as one of the key areas for atmospheric pollution control in China, from 2000 to 2022. Specifically, the study aims to (1) quantify interannual and seasonal variations of AOD, (2) characterize the spatial clustering and persistence of high- and low-value AOD regions, and (3) assess the relative contributions of socioeconomic activities, meteorological conditions, precipitation characteristics, air pollution events, and atmospheric diffusion capacity to the spatial heterogeneity of AOD. The ultimate objective is to provide scientific evidence and insights to support long-term air quality improvement strategies, environmental governance, and sustainable development of the Fenwei Plain. Methods　AOD observation data were derived from the MODIS MAIAC MCD19A2 satellite product, and processed on the Google Earth Engine platform to generate annual and seasonal mean composites for the years from 2000 to 2022. Spatial clustering patterns of AOD were analyzed at the county level using the Getis-Ord Gi* hot spot analysis to detect statistically significant AOD "hot spots" and "cold spots". To explore the driving mechanisms behind the observed patterns, the Geographical Detector Model was applied to 16 influencing factors, which were divided into five categories: socioeconomic indicators (population density, GDP per unit area), atmospheric state variables (air temperature, air pressure, water vapor pressure), precipitation cleansing factors (precipitation amount and frequency), air pollution factors (haze days, dust storm days), and atmospheric horizontal diffusion indicators (wind speed and strong wind frequency). The model quantified the explanatory power (q-values) of each factor for different seasons and representative years (2010 and 2020), revealing both long-term changes and seasonal dependencies. Results and Discussion　AOD in the Fenwei Plain from 2000 to 2022 demonstrated a clear two-stage interannual trend. From 2000 to 2013, AOD showed a slight but statistically insignificant upward trend, with a multi-year mean of 0.38. In contrast, from 2014 to 2022, AOD declined significantly at an annual rate of 0.011, with the mean decreasing to 0.32, reflecting the effectiveness of China's air pollution control policies implemented after 2013. Spatially, AOD in the Fenwei Plain from 2000 to 2022 exhibited a persistent stepwise pattern decreasing from major urban agglomerations such as Xi'an, Xianyang, Luoyang, and Yuncheng toward surrounding rural and high-elevation regions. The three major hot spot regions, including the Weihe River valley, the Fenhe-Yellow River confluence, and the Luoyang Basin, remained stable for decades, indicating spatially locked pollution accumulation zones influenced by topography, population distribution, and industrial structure. Seasonal analysis revealed the substantial differences that during 2000 – 2013, AOD followed the order of summer > spring > winter > autumn, and during 2014–2022, the pattern shifted to the order of spring > summer > winter > autumn, with a pronounced downward trend for each season. Seasonal hot spot patterns were consistent with sources such as spring dust transport, summer agricultural burning and photochemical activity, and winter heating emissions. The results of Geographical Detector indicated that atmospheric state factors exerted the strongest and most consistent influence on AOD spatial heterogeneity, followed by socioeconomic factors whose influence increased markedly from 2010 to 2020. In addtion, precipitation cleansing effects and diffusion capacity played secondary but seasonally variable roles, while pollution-related factors such as haze days demonstrated increasingly importance in recent years. Conclusions　The findings highlight that AOD in the Fenwei Plain is determined by a combination of meteorological processes, socioeconomic development intensity, and seasonal pollution sources. Although regional AOD has decreased significantly since 2014 due to effective national and regional environmental policies, persistent spatial hot spots remain. These reflect deep structural challenges related to industrial activity, population density, and terrain-induced unfavorable conditions for atmospheric diffusion. Sustained improvement will require strengthening emission control, adjusting industrial restructure, enhancing the penetration of clean energy, and implementing season-specific pollution mitigation strategies for dust transportation, agricultural burning, and winter heating emissions.

Key words: Fenwei Plain, aerosol optical depth, spatio-temporal variation, hot spot analysis, geographical detector; atmospheric pollution, driving factors