Hourly PM2.5 Estimates from Different Measurements of a Geostationary Satellite Using an Ensemble Learning Algorithm

Most studies have focused on making daily PM_2.5estimations using polar-orbiting satellite retrieved aerosol optical depth (AOD) (e.g., MODIS), which are inadequate for understanding the evolution of PM_2.5. This study developed an ensemble learning model to estimate hourly PM_2.5concentrations from both Himawari AOD and top-of-atmosphere reflectance with someauxiliary parameters. The estimated PM_2.5concentrations from AOD and reflectance are comparable each other and agree well with ground measurements. Satellite-estimated PM_2.5concentrations over central East China display a north-to-south decreasing gradient with the highest concentration in winter and the lowest concentration in summer. Diurnally, concentrations are higher in the morning and lower in the afternoon. PM_2.5concentrations exhibit a significant spatial agglomeration effect in central East China. The errors in AOD do not necessarily affect the retrieval accuracy of PM_2.5proportionally, especially if the error is systematic. High-frequency spatiotemporal PM_2.5variations can improve our understanding of the formation and transportation processes of regional pollution episodes.