Aerosol layer height (ALH) retrievals from oxygen absorption bands: Intercomparison and validation among different satellite platforms, GEMS, EPIC, and TROPOMI

Abstract

Abstract. Although containing only single piece of information, aerosol layer height (ALH) indicates the altitude of aerosol layer in vertical coordinate which is essential for assessment of surface air quality and aerosol climate impact. Passive remote sensing measurements in oxygen (O₂) absorption bands are sensitive to ALH, providing an opportunity to derive global or regional ALH information from satellite observations. In this study, we compare ALH products retrieved from near-infrared O₂ absorption measurements from multiple satellite platforms including Geostationary Environment Monitoring Spectrometer (GEMS) focusing on Asia, Earth Polychromatic Imaging Camera (EPIC) in deep space, and polar orbiting satellite TROPOspheric Monitoring Instrument (TROPOMI), and validate them using spaceborne lidar (CALIOP) measurements for typical dust and smoke plumes. Adjustments have been made to account for the inherent variations in the definitions of ALH among different products, ensuring an apple-to-apple comparison. In comparison with CALIOP ALH, both EPIC and TROPOMI ALH display a high correlation coefficient (R) higher than 0.7 and an overestimation by ~ 0.8 km, whereas GEMS ALH exhibits minimal bias (0.1 km) but a slightly lower correlation with R of 0.64. Categorizing GEMS retrievals with UVAI ≥ 3 improves the agreement with CALIOP. GEMS ALH demonstrates a narrower range and lower mean value compared to EPIC and TROPOMI, and their correlation is further improved when UVAI ≥ 3. Furthermore, diurnal variation of GEMS and EPIC ALH, especially for UVAI ≥ 3, aligns with boundary layer development. Considering the important role of AOD in ALH retrieval, we found GEMS AOD at 680 nm correlates well with AERONET AOD (R ~ 0.9) but features a negative bias of -0.2. EPIC and TROPOMI tend to overestimate ALH by 0.33 km and 0.23 km, respectively, in dust cases. Finally, a dust and a smoke case are analysed in detail to explore the variation of ALH during plume transport from multiple data.