LiDAR active remote sensing provides vertical measurement (10s of meters) profiles of aerosols and clouds, which can be applied for air quality analysis. LiDAR sensors generate their own energy, and measurements occur during both the day and night. This study uses volume data collected by the CALIPSO LiDAR satellite. CALIPSO stands for Cloud–Aerosol LiDAR and Infrared Pathfinder Satellite Observations and was launched by NASA and CNES to measure the vertical distribution of aerosols and clouds. The measurements contain narrow curtains that are scanned to determine their wavelengths in the Electromagnetic spectrum.

The main objective of this study consists of analyzing the air quality using LiDAR data, such that the information included in the curtain is interpreted to discern the cloud phase and aerosol type. To this end, we present a data-driven framework protocol. First, an analysis of the total attenuated backscatter at 532 nm, known as the level-1 product, is provided. Second, the level-2 product Cloud–Aerosol Discrimination is used to capture the characteristics of the aerosols and clouds. Third, the aerosol type is classified as polluted or clean.

The dataset considered in the experiments was acquired by CALIPSO on 04 July 2020. The images include two sets of the total attenuated backscatter at wavelengths of 532 nm and 1064 nm. The analysis of the results is presented as follows. The LiDAR curtain contains several features that differ between lower latitudes (left) and higher latitudes (right). At latitudes ranging from ~ 30 to 36°N, high-intensity features are noticed between altitudes of 5km and 10km, which are classified as desert dust. However, from latitudes of 5.97 to -0.16°N and at altitudes lower than 5km, the aerosols are classified as desert and polluted dust, which correspond to biomass burning. To conclude, the results presented in this study are confirmed by the pixel intensities of various features present in the LiDAR curtain.