Aerosol Types Classification in the Tropical Climate of Singapore by Integrating Ground-Based Remote Sensing and Reanalysis Datasets

AERONET Version 3 inversion products were used to classify optically defined aerosol regimes and examine their temporal distribution over Singapore during 2007–2023 using particle linear depolarization ratio (PLDR), dust ratio (R d ), fine-mode fraction (FMF), and single scattering albedo (SSA). Lower PLDR values were associated with fine-mode, relatively spherical particles characteristic of pollution-related aerosols, whereas higher PLDR values indicated non-spherical coarse particles associated with dust. The joint FMF-R d distribution showed a predominance of fine-mode, low-dust aerosol conditions. Among the quality-assured AERONET inversion retrievals, dust-free pollution regimes, comprising strongly absorbing (SA), moderately absorbing (MA), weakly absorbing (WA), and non-absorbing (NA) regimes, accounted for 74%, while dust-related regimes, including dust-dominated mixtures (DDM), pollution-dominated mixtures (PDM), and pure dust (PD), accounted for 26%. The occurrence fractions were approximately 7% (SA), 11% (MA), 21% (WA), 36% (NA), 5% (DDM), 20% (PDM), and 1% (PD), with NA being the most prevalent regime. Seasonal variations further indicated the persistent dominance of NA, while PDM showed enhanced occurrence during the April–May inter-monsoon period. CAMS and MERRA-2 showed qualitative differences in aerosol representation relative to the AERONET-derived optical regimes; therefore, these datasets were compared qualitatively rather than as direct quantitative validation. Overall, the results demonstrate the dominance of pollution-related aerosol regimes over Singapore and highlight the value of combining ground-based observations with reanalysis datasets for aerosol characterization and air-quality assessment in tropical urban environments.