Multi-method aerosol classification over the Eastern Mediterranean based on long-term AERONET observations and machine learning
Atmospheric Environment, cilt.382, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 382
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.atmosenv.2026.122242
- Dergi Adı: Atmospheric Environment
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Applied Science & Technology Source, Artic & Antarctic Regions, BIOSIS, Chemical Abstracts Core, Chimica, Compendex, EMBASE, Environment Index, Geobase, Greenfile, INSPEC, Public Affairs Index, Urban Studies Abstracts, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO)
- Anahtar Kelimeler: AERONET, Aerosol classification, Aerosol optical properties, Eastern Mediterranean, Mixed aerosols, Spectral clustering
- Hacettepe Üniversitesi Adresli: Evet
Özet
This study presents a multi-method framework for aerosol characterization and classification over the Eastern Mediterranean using a 22-year AERONET dataset from the IMS-METU-Erdemli station for the period 2000–2022. Aerosol variability was investigated using complementary optical classification, spectral AOD curvature, backward trajectory, and machine-learning-based clustering analyses. The results show that aerosol conditions over Erdemli are dominated by a persistent fine-mode background associated with anthropogenic pollution, secondary aerosol formation, and aged continental particles, whereas the highest aerosol-loading episodes are mainly linked to episodic coarse-mode dust intrusions from North Africa and the Middle East. The seasonal AOD–AE classification indicates that spring exhibits the clearest enhancement of dust-influenced coarse-mode aerosol, while summer is dominated by transition/mixed and processed fine-mode aerosol conditions rather than a purely dust-dominated regime. Spectral AOD analysis further shows that aerosol loading increases through two distinct pathways: fine-mode growth and aging under warm and stagnant summer conditions, and fine–coarse mode mixing during dust-transport episodes, particularly in spring. The trajectory–optical-property analysis confirms that spring air-mass pathways are associated with lower AE and enhanced AOD, while summer pathways are linked with high-AE and relatively scattering aerosol populations. Spectral clustering using FineAOD, EAE, AAE, SSA, and RRI establishes a baseline aerosol-type climatology, showing that fine-mode anthropogenic/secondary aerosol is the dominant regime, accounting for 56.8% of clustered observations, followed by fine-mode mixed/aged aerosol with 24.2%, dust-influenced mixed aerosol with 14.0%, and strongly absorbing coarse/dust-related aerosol with 5.0%. Overall, the proposed framework provides a physically consistent approach for identifying mixed aerosol regimes in complex receptor regions influenced by both natural and anthropogenic sources.