Deciphering wintertime air pollution upon the West Lake of Hangzhou, China

A severe air quality degradation event on the UNESCO-listed West Lake of Hangzhou, China, in January 2015, was found to be primarily driven by fluctuations in gaseous pollutants (SO₂, NO₂, and CO) rather than meteorological factors or regional transport. Machine learning models, specifically Random Forest (RF) and Recurrent Neural Networks (RNN), demonstrated high accuracy in predicting daily average concentrations of these pollutants during winter, offering a promising tool for future forecasting and early warning systems. The study also revealed that sulfate constituted a larger proportion in PM₁₀ than in PM₂.₅, while CO showed a strong positive correlation with particulate matter. Dew-point deficit emerged as the key meteorological factor influencing SO₂, NO₂, and surface O₃ levels.

Feng, R., Huang, C.-C., Luo, K., & Zheng, H.-J. (2021). Deciphering wintertime air pollution upon the West Lake of Hangzhou, China. Journal of Intelligent & Fuzzy Systems, 40(3), 5215–5223. https://doi.org/10.3233/jifs-201964