 This paper proposes a new methodology to estimate ground-level PM2.5 concentrations using satellite-derived 500M AOD data. The proposed geometrically weighted regression, GWR, model combines the simplified high-resolution MODIS aerosol retrieval algorithm, CERA, AOD product, with meteorological variables such as PBLH, RH, WS, and temp to capture the spatio-temporal dynamics in the PM2.5 AOD relationship. The estimated ground-level PM2.5 concentration has 500M resolution at the MODIS satellites over past moments twice a day, which can be used for air quality monitoring and haze tracking at the urban and regional scale. This article was authored by Yang Bai, Lee Sin Woo, Kai Chen, and others.