Articles | Volume 25, issue 19
https://doi.org/10.5194/acp-25-12549-2025
https://doi.org/10.5194/acp-25-12549-2025
Research article
 | 
09 Oct 2025
Research article |  | 09 Oct 2025

Aerosol type classification with machine learning techniques applied to multiwavelength lidar data from EARLINET

Ana del Águila, Pablo Ortiz-Amezcua, Siham Tabik, Juan Antonio Bravo-Aranda, Sol Fernández-Carvelo, and Lucas Alados-Arboledas

Data sets

Manually Labelled Aerosol Typing from the Multiwavelength Raman Lidar MULHACEN at ACTRIS/EARLINET UGR Station for Machine Learning Applications Ana del Águila et al. https://doi.org/10.5281/zenodo.16925786

Download
Short summary
This study applies machine learning (ML) techniques to classify aerosols using high-resolution multiwavelength lidar data from EARLINET network. We developed a reference dataset and evaluated six ML models, with LightGBM achieving over 90 % accuracy. Depolarization data proved critical for improving dust classification. Validated against independent datasets, our approach improves aerosol classification and may help refine lidar-based processing strategies.
Share
Altmetrics
Final-revised paper
Preprint