Articles | Volume 26, issue 20
https://doi.org/10.5194/acp-26-15003-2026
https://doi.org/10.5194/acp-26-15003-2026
Research article
 | 
26 Oct 2026
Research article |  | 26 Oct 2026

Classification and quantification of low-visibility events using deep learning over eastern China

Yuting Liang, Shitong Zhao, Dantong Liu, Changhao Wu, Li Yi, Xingcan Jia, Honghui Xu, and Jiankan Wang

Data sets

ERA5 hourly data on single levels from 1940 to present Copernicus Climate Change Service https://doi.org/10.24381/cds.adbb2d47

Download
Short summary
We identified and quantified low-visibility events over land and sea in eastern China using satellite, weather and pollution data. Our model distinguishes polluted fog from clean fog. Model tests show that a 20% cut in fine particle pollution reduces winter polluted fog area by 43%, but population exposure by only 15%, as the most polluted areas are also densely populated. These findings guide air-quality policies to improve visibility, protect health and support transport safety.
Share
Altmetrics
Final-revised paper
Preprint