Spatial-temporal distribution and variation of atmospheric NO₂ dry deposition in the Yellow River Basin from 2015 to...

Nitrogen dioxide (NO₂) is a major atmospheric pollutant that threatens human health and environmental quality amid rapid urbanization and industrialization. The Yellow River Basin is a heavily populated and economically significant area that is essential to China's industrial and agricultural sectors. For this reason, it is especially critical to accurately measure NO₂ concentrations and related dry deposition fluxes. To estimate near-surface NO₂ concentrations throughout the Yellow River Basin from 2015 to 2023, a Random Forest (RF) machine learning model was created using tropospheric NO₂ column data from the Ozone Monitoring Instrument (OMI), ground-based station observations, and auxiliary environmental variables. With an R² (coefficient of determination) of 0.884 and RMSE (root mean square error) of 4.626 µg/m³ for the training set and an R² of 0.777 with RMSE of 6.447 µg/m³ for the test set, the model demonstrated strong predictive ability. NO₂ levels showed a little downward trend from 2015 to 2021, followed by a modest uptick in 2021-2023, according to spatial and temporal analysis. Seasonally, there was a clear U-shaped pattern with winter peaks and summer troughs in NO₂ concentrations and the associated dry deposition fluxes. Upstream regions like Sichuan and Qinghai had consistently low levels, while industrialized downstream provinces like Shandong, Henan, and Shanxi had high rates. These results offer scientific support for nitrogen load mitigation and air quality management in the Yellow River Basin, as well as crucial insights into the spatial dynamics of NO₂ pollution.