Due to the influence of multiple factors such as the physical and chemical properties of the atmosphere, the limitations of data sources, and the assumptions of inversion methods, there are many difficulties in inverting the concentration distribution with high temporal and spatial resolution over a large area near the ground. In this study, the GEOS-Chem chemical transport model was adopted. Through dynamic constraints of emission sources, meteorological fields, and chemical mechanisms, combined with the optimization output of radial basis functions, a 1 km × 1 km hourly near-surface nitrogen dioxide concentration distribution dataset in China from 2015 to 2024 was generated. Based on the analysis of spatial differences and temporal fluctuations, the concentration changes of nitrogen dioxide are closely related to human activities, climate change, and seasonal variations. Thanks to China’s implementation of a large number of proactive pollution control measures, the average annual concentration of nitrogen dioxide has dropped from 19.7 μg/m3 in 2015 to 14.1 μg/m3 in 2024, with a cumulative reduction of 28.43%. The phenomenon of the one-hour average concentration exceeding the limit of 200 μg/m3 has been basically eliminated across the country.
Zhan et al. (Sat,) studied this question.