Physical and deep learning retrieved fine mode fraction (Phy-DL FMF)

Aerosol fine-mode fraction (FMF) is a key parameter for discriminating anthropogenic aerosols from natural ones, but it is highly uncertainty from satellite-based retrieval, especially over land. Here we integrated physical and deep learning (Phy-DL) methods to retrieval aerosol FMF in a global land scale based on MODIS data and produced the 20-year (2001-2020) Phy-DL FMF (500 nm) at daily temporal resolution and 1° spatial resolution. The Phy-DL FMF is uploaded as Geotiff format, stretched from -89.5° to 89.5° latitude and from -179.5° to 179.5° longitude. In some certain days Phy-DL FMF might be inavailable, due to the missing MODIS satellite data used for calculating Phy-DL FMF.