Exploring the spatial heterogeneity and temporal homogeneity of ambient PM(10) in nine core cities of China

We focus on the causes of fluctuations in wintertime PM(10) in nine regional core cities of China using two machine learning models, Random Forest (RF) and Recurrent Neural Network (RNN). RF and RNN both show high performance in predicting hourly PM(10) using only gaseous air pollutants (SO(2), NO(2) and CO) as inputs, showing the predominance of the secondary inorganic aerosol and implying the existence of thermodynamic equilibrium between gaseous air pollutants and PM(10). Also, we find the following results. The correlation of gaseous air pollutants and PM(10) were more relevant than that of meteorological conditions and PM(10). CO was the predominant factor for PM(10) in the Beijing-Tianjin-Hebei Plain and the Yangtze River Delta while SO(2) and NO(2) were also important features for PM(10) in the Pearl River Delta and Sichuan Basin. The spatial heterogeneity and temporal homogeneity of PM(10) in China are revealed. The long-range transported PM(10) was substantiated to be insignificant, except in the sandstorms. The severity of PM(10) was attributable to the lopsided shift of thermodynamic equilibrium and the phenology of indigenous flora.