Aerosol monitoring in Siberia using an 808 nm automatic compact lidar

Abstract

Abstract. Our study provides new information on aerosol-type seasonal variability and sources in Siberia using observations (ground-based lidar and sun photometer combined with satellite measurements). A micropulse lidar emitting at 808 nm provided almost continuous aerosol backscatter measurements for 18 months (April 2015 to September 2016) in Siberia, near the city of Tomsk (56∘ N, 85∘ E). A total of 540 vertical profiles (300 daytime and 240 night-time) of backscatter ratio and aerosol extinction have been retrieved over periods of 30 min, after a careful calibration factor analysis. Lidar ratio and extinction profiles are constrained with sun-photometer aerosol optical depth at 808 nm (AOD808) for 70 % of the daytime lidar measurements, while 26 % of the night-time lidar ratio and AOD808 greater than 0.04 are constrained by direct lidar measurements at an altitude greater than 7.5 km and where a low aerosol concentration is found. An aerosol source apportionment using the Lagrangian FLEXPART model is used in order to determine the lidar ratio of the remaining 48 % of the lidar database. Backscatter ratio vertical profile, aerosol type and AOD808 derived from micropulse lidar data are compared with sun-photometer AOD808 and satellite observations (CALIOP space-borne lidar backscatter and extinction profiles, Moderate Resolution Imaging Spectroradiometer (MODIS) AOD550 and Infrared Atmospheric Sounding Interferometer (IASI) CO column) for three case studies corresponding to the main aerosol sources with AOD808>0.2 in Siberia. Aerosol typing using the FLEXPART model is consistent with the detailed analysis of the three case studies. According to the analysis of aerosol sources, the occurrence of layers linked to natural emissions (vegetation, forest fires and dust) is high (56 %), but anthropogenic emissions still contribute to 44 % of the detected layers (one-third from flaring and two-thirds from urban emissions). The frequency of dust events is very low (5 %). When only looking at AOD808>0.1, contributions from taiga emissions, forest fires and urban pollution become equivalent (25 %), while those from flaring and dust are lower (10 %–13 %). The lidar data can also be used to assess the contribution of different altitude ranges to the large AOD. For example, aerosols related to the urban and flaring emissions remain confined below 2.5 km, while aerosols from dust events are mainly observed above 2.5 km. Aerosols from forest fire emissions are observed to be the opposite, both within and above the planetary boundary layer (PBL).