Unraveling Street-Level Air Pollution Upon a Pivotal City of Yangtze River Delta, China

Zhejiang Academy of Ecological and Environmental Sciences,Gao Han,Wang Zhuo,Luo Kun,Fan Jian-ren,Zheng Hui-jun
DOI: https://doi.org/10.1007/s41810-021-00093-7
2021-01-01
Aerosol Science and Engineering
Abstract:We use two machine learning models—random forest (RF) and recurrent neural network (RNN)—to analyze and predict air pollutants in a pivotal city of YRD, China. We quantitatively show the determinants for the atmospheric pollutants, providing insights for air pollution control policies. We propose, test and verify a five-step avenue to forecast the main atmospheric pollutants (SO 2 , NO 2 , CO, O 3 , PM 2.5 and PM 10 ) in the future 24 h. Step one, WRF is used to generate the meteorological conditions in the next 24 h. Step two, SO 2 and CO are predicted by RNN using WRF-simulated meteorological conditions. Step three, NO 2 is predicted by RNN using WRF-simulated meteorological conditions and RNN-simulated CO. Step four, O 3 is predicted by RNN using WRF-simulated meteorological conditions and RNN-simulated CO and NO 2 . Step five, PM 2.5 and PM 10 are predicted by RNN using WRF-simulated meteorological conditions and RNN-simulated SO 2 , CO and NO 2 . The significant role that dew-point deficit plays in shaping SO 2 , NO 2 and O 3 is recognized. CO was strongly positively linked with NO 2 and PM 2.5 . Decrease of CO may trigger the crescendo of ground-level O 3 . Stratospheric downward transport played paltry role in shaping tropospheric O 3 at Hangzhou. We also identify that some illegal factories were surreptitiously emitting trichlorofluoromethane (CFCl 3 ), one of the strongest stratospheric ozone-depleter that should have been forbidden since 2010.
What problem does this paper attempt to address?