Dust Layer Height Retrieval Over Bright Surface from TROPOMI Oxygen A and B Absorption bands

The scattering and absorption of solar and infrared radiation by dust particles influence the Earth’s energy budget and global climate, but these radiative impact estimates are still uncertain. In addition to dust composition and particle size distribution, the vertical profile of dust plumes also influences their direct radiative forcing estimation significantly. Therefore, the measurements of global three-dimensional distribution of dust properties are necessary to improve our understanding of dust climate impact. Spaceborne lidar provides effective aerosol extinction profile measurements, but its application is limited due to sparse spatial coverage. Passive satellite measurements in oxygen (O₂) absorption bands are sensitive to the altitude of absorbing aerosol layer, making it possible to derive the optical centroid height (AOCH) of smoke/dust layer by assuming a quasi-Gaussian distribution of aerosol extinction profile. Based on the TROPOMI measurements in O₂ B (688 nm) and A (760 nm) absorption band, we have developed an algorithm to retrieve AOD and AOCH for smoke and dust over dark surface (Chen et al, 2021). Given the much lower surface reflectance in O₂ B band than in A band for vegetated surface, the AOCH retrievals by applying O₂ B band have better agreement with CALIOP measurements. The main purpose of this study is to extend our algorithm to bright surface to get dust layer height near the source region of dust plumes. The thresholds used in cloud mask tests are adjusted for bright surface classified by the NDVI from MODIS MAIAC BRDF climatology. The single scattering albedo and phase function are derived from AERONET climatology considering the non-spherical shape of dust particles, and are used in the radiative transfer simulation to create the look-up-table (LUT) of top-of-atmosphere (TOA) reflectance for different AOD and AOCH. A few dust cases over Saharan Desert are processed for retrieval test and validated with CALIOP measurements. ~1 km retrieval uncertainty is found for these retrievals of AOCH, larger than the retrieval uncertainty we found over dark surfaces (Chen et al, 2021), mainly due to the less sensitivity of O₂ absorption measurements to AOCH over higher surface reflectance. Even though, these dust AOCH retrievals still show good consistency of ~0.6 correlation coefficient with CALIOP extinction weighted AOCH. The dust plume height retrieved from IASI thermal infrared measurements is also compared.