Abstract: Objective　As a precursor of atmospheric aerosols and an indicator of human activity, nitrogen dioxide (NO2) plays a critical role in air quality and climate change. Although extensive research has focused on national or urban agglomeration scales, few studies have systematically compared the vertical column density (VCD) characteristics of NO2 across regions with distinct functional types. This study investigates the spatiotemporal variations of NO2 VCD in Anhui Province, China, from 2017 to 2022, focusing on three representative city types: agricultural region (Fuyang city), industrial concentration region (Hefei city), and ecotourism region (Huangshan city). The objective is to elucidate long-term trends and seasonal cycles of NO2 VCD changes, as well as its episodic responses to human activities such as the Lunar New Year, crop straw burning, and the COVID-19 epidemic control, so as to provide a scientific basis for region-specific air pollution control policies. Methods　This study utilized the secondary data product of tropospheric NO2 column concentration (VCD) derived from the Ozone Monitoring Instrument (OMI) onboard the Aura satellite. The research scope covered the entire administrative region of Anhui Province, China, with targeted sampling conducted in Fuyang city, Hefei city, and Huangshan city. Firstly, the annual and monthly mean values of NO2 VCD from 2017 to 2022 were calculated to analyze interannual trends. Secondly, typical temporal windows were selected to examine episodic variations: the Spring Festival period (February), representing reduced anthropogenic emissions, and the post-harvest season (June), representing biomass burning. Thirdly, the OMI satellite retrievals were validated against the ground-based near-surface NO2 mass concentration observations from the China National Environmental Monitoring Centre using Pearson correlation analysis. Finally, the relative change rate and anomaly analysis methods were employed to quantify the impact of the COVID-19 pandemic control in January 2020. Results and Discussion　Significant spatial heterogeneity and temporal variability in NO2 VCD were observed across the three different functional regions. The annual mean NO2 VCD in the agricultural region (Fuyang) and the industrial region (Hefei) exhibited a consistent trend, reaching its peak in 2019 and lowest point in 2022. In contrast, the ecotourism region (Huangshan) maintained a consistently low and stable annual mean throughout the six-year period, reflecting the minimal anthropogenic interference. During typical episodic periods, distinctive response patterns emerged. In February, coinciding with the Lunar New Year holiday of China, NO2 VCD in all three regions decreased to varying degrees due to industrial shutdowns and reduced traffic. In June, NO2 VCD in the agricultural region experienced a dramatic surge, with an increase of 81.89% compared to non-burning periods. This sharp increase is directly attributable to the extensive straw burning activities following wheat harvest, which is not observed in industrial or ecotourism regions. The validation analysis demonstrated a robust consistency between the satellite-derived column concentrations and the ground-based monitoring data, with correlation coefficients (R) of 0.80, 0.75, and 0.68 for Fuyang, Hefei, and Huangshan, respectively, which confirmed the reliability of OMI data in capturing near-surface NO2 variations across diverse underlying surfaces. Regarding seasonal characteristics, NO2 VCD in Anhui Province displayed a typical "U-shaped" pattern, with the highest values in winter (averaging 12.9 × 1015 molec/cm²) and the lowest in summer. The mean value of NO2 VCD in winter was 3.07 times that of summer, mainly due to the enhanced fossil fuel combustion for heating, coupled with unfavorable meteorological conditions such as shallow planetary boundary layers and low precipitation, which suppress the diffusion of pollutants. The relative change rate curve further elucidated the fluctuating characteristics of NO2 over the six years. Notably, in the early stage of the COVID-19 pandemic in January 2020, NO2 VCD in the three regions plummeted by more than 45% compared to the same period in previous years. This confirmed that stringent control on population mobility and industrial production directly suppressed NO2 emissions, underscoring the high sensitivity of NO2 to human activity intensity. Conclusions　This study demonstrates that NO2 VCD in Anhui Province are highly responsive to both regional functional types and specific human activities. Industrial and agricultural activities serve as the dominant drivers of NO2 pollution, so the changes of NO2 VCD in agricultural and industrial concentration areas are closely related to these activities, while that in ecotourism areas remains relatively stable. As a natural experiment, COVID-19 control confirmed that rapid emission reduction is achievable through controlled human intervention. In addition, the pronounced seasonal disparity highlights the need for winter-specific emission reduction measures. The strong correlation between OMI satellite data and ground observations validates the utility of satellite remote sensing in supplementing ground networks, especially in regions with sparse monitoring stations. These findings provide critical insights for formulating differentiated air quality management strategies in Chinese provinces undergoing rapid urbanization and agricultural intensification.

Key words: troposphere, NO2 column concentration, spatial-temporal distribution, ozone monitoring instrument satellites; Anhui Province