Key points are not available for this paper at this time.
Accurate identification of aerosol types is crucial for understanding aerosol microphysical and optical properties and quantifying their radiative impacts. We propose an aerosol types identification algorithm, belonging to the XBAER (eXtensiBle Atmospheric and surfacE parameters Retrieval) algorithm family, using the Geostationary Environment Monitoring Spectrometer (GEMS) data. The key concept of the XBAER algorithm is to combine aerosol optical properties (an UV–Vis radiance slope index, Aerosol Optical Depth and Single Scattering Albedo) and spectral information observed by GEMS. First, a multi-parameter threshold method based on the UV–Vis radiance slope index (340–440 nm), Aerosol Optical Depth, and Single Scattering Albedo provides initial classification into high absorbing, non-absorbing, and dust aerosols. Subsequently, the radiance slope between 400 and 500 nm refines the classification by capturing subtle spectral variations that differentiate dust from high absorbing aerosols. Qualitative validation with VIIRS true-color imagery demonstrates improved spatial consistency and accurate depiction of dust and smoke plumes. Quantitative evaluation using Aerosol Robotic Network (AERONET) observations across four representative months (January, April, July, and October 2023) yields an overall accuracy of 86.53% (κ = 0.66), outperforming the official GEMS product (78.99%, κ = 0.53). The results highlight that integrating radiance spectral slope information substantially enhances aerosol discrimination and supports operational aerosol monitoring, climate modeling, and radiative forcing assessments over East Asia.
Wang et al. (Thu,) studied this question.