Aerosol type classification analysis using EARLINET multiwavelength and depolarization lidar observations

Abstract. We introduce an automated aerosol type classification method, called Source Classification Analysis (SCAN). SCAN is based on predefined and characterized source regions, the time that air parcel spends above each geographical region, a number of additional criteria. The output compared with two independent methods, which use intensive optical parameters from lidar data: (1) Mahalanobis distance automatic (MD) (2) neural network typing algorithm (NATALI). In this paper, data European Aerosol Research Lidar Network (EARLINET) have been used. A total 97 free tropospheric layers four typical EARLINET stations (i.e., Bucharest, Kuopio, Leipzig, Potenza) in period 2014–2018 were classified 3?+2?+1? configuration. found SCAN, as method properties, not affected by overlapping values different types. Furthermore, has no limitations concerning its ability to classify mixtures. Additionally, it valuable tool even single (elastic) signals case cannot provide full set (3?+2?+1?) properties; therefore, can work independently capabilities system. Finally, our results show NATALI lower percentage unclassified (4 %), while MD higher (50 %) cases mixtures (5 %).