Himawari-8-Derived Aerosol Optical Depth Using an Improved Time Series Algorithm Over Eastern China

Himawari-8 (H8), as a new generation geostationary meteorological satellite, has great
potential for monitoring the spatial–temporal variation of aerosol properties. However, the large
amount of spectral data with diering observation geometries require re-formulation of the surface
reflectance correction to utilize this new satellite data. This is achieved by using an improved version
of the time series (TS) technique proposed by Mei et al., (2012) based on the assumption that the
ratio of the surface reflectance in dierent spectral bands does not change between any two scan
times within an hour. In addition, more suitable aerosol models were adopted, based on cluster
analysis of local Aerosol Robotic Network (AERONET) data. The improved TS algorithm (ITS) was
applied to retrieve the Aerosol Optical Depth (AOD) over eastern China and the results compare
favorably with collocated reference AOD data at eleven sun photometer sites (R > 0.8, Root Mean
Square Error (RMSE) < 0.2). Comparison with the H8 ocial AOD product and with MODIS Dark
Target (DT)–Deep Blue (DB) combined AOD data shows the good performance of the ITS method for
AOD retrieval with dierent observation angles.