Lee, Jaehwa, Kim, Jhoon, Lee, Hee Choon, Takemura, Toshihiko (2007). Classification of Aerosol Type from MODIS and OMI over East Asia. ASIA-PACIFIC JOURNAL OF ATMOSPHERIC, 43(4), 343-357.
Aerosols affect the earth's climate by scattering and absorbing radiation, and by altering the cloud microphysics. Since the effects of aerosols are different from the type to the other, aerosol type classification from satellite remote sensing is challenging. In this study, types of aerosol are classified into dust, sea salt, smoke plume, and sulfate by using aerosol optical thickness (AOT) and Angstrom exponent (AE) of Moderate Resolution Imaging Spectroradiometer ( MODIS) and aerosol index ( AI) of Ozone Monitoring Instrument (OMI) data. The optical properties and types of aerosol are analyzed over the East Asia, one of the heavily polluted regions on the globe. AOTs significantly decrease from near-continent ocean to remote ocean in monthly climatology, and the maximum AOT is represented in June over the Northeast China, Korean Peninsula, and Japan. The distribution of AEs indicates that aerosols of anthropogenic as well as natural origin exist over both land and ocean in East Asia. The seasonal variation of AI is represented over the regions with latitude between 30 degrees N and 40 degrees N where the influence of Asian Dust is significant in dry season. Retrieved aerosol types by MODIS-OMI algorithm showed that the aerosols are considerably mixed, and represented seasonal variation over East Asia. The evidence of pollution transport is detected by aerosol classification, that is, the anthropogenic aerosol types are detected over ocean not only over land. Based on retrieved aerosol types, the frequency distributions of each aerosol type are analyzed in Beijing, Seoul, and Tokyo. Dust type aerosol is most frequently detected followed by mixtures of sea salt/sulfate at Beijing, whereas sea salt/sulfate mixture is most frequently detected at Seoul and Tokyo. The frequency distributions of aerosol types are similar for both in 2005 and 2006, while the distribution showed considerable seasonal variations.