Monitoring PM2.5 Distributions over China from Geostationary Satellite Observations

摘要

Satellite-derived aerosol optical depth (AOD) has been widely used to estimate surface fine particulate matter (PM₂.₅) concentrations. To understand the temporal and spatial variation of PM₂.₅ distribution at various scales, we developed techniques for estimating hourly PM_2.5 concentration from visible and infrared imagers onboard Himawari-8, GK-2A and FY-4A satellites. During daytime, satellite observed reflectance at the top of atmosphere (TOA) is directly converted to PM₂.₅ through a machine-learning algorithm. At night, the thermal channel data from satellites are tested for PM estimation through using the aerosol absorbing properties. It is found that the PM concentration is higher in the morning and lower in the afternoon in east central China. PM concentrations also exhibit a north-to-south decreasing gradient with the highest one in winter and the lowest in summer. From AHI PM data collected in the past five years, we found a decreasing trend of PM₂.₅ concentration in most of areas of China.