First retrieval of absorbing aerosol height over dark target using TROPOMI...

Chen, X., J. Wang, R. Xu, Meng Zhou, H. Zhang, L. C. García, P. R. Colarco, S. Janz, J. Yorks, M. McGill, J. Reid, M. de Graaf, and S. Kondragunta (2021), First retrieval of absorbing aerosol height over dark target using TROPOMI oxygen B band: Algorithm development and application for surface particulate matter estimates, Remote Sensing of Environment, 265, 112674, doi:10.1016/j.rse.2021.112674.

Constraint of the vertical distribution of aerosol particles is crucial for the study of aerosol plume structure, aerosol radiative effects, and ultimately monitoring surface air pollution. We developed an algorithm to retrieve the aerosol optical central height (AOCH) of absorbing aerosols by using, for the first time, the oxygen (O2) A and B absorption band measurements from the TROPOspheric Monitoring Instrument (TROPOMI) over dark targets. For the retrieval, narrow band radiance at seven channels ranging from ultraviolet (UV) to shortwave infrared (SWIR) are convolved from TROPOMI hyperspectral measurements. Subsequently, cloudy pixels are screened out by using the slope of spectral reflectance, while aerosol types (dust and smoke) are classified by the wavelength dependence of aerosol path radiance in conjunction with UV aerosol index. Surface reflectance over land is derived from the MODIS surface bi-directional reflectance climatology, and over water from the GOME-2 surface Lambert-equivalent reflectivity (LER) database. The aerosol optical depth (AOD) and AOCH are retrieved through an approach of look-up-table accounting for AERONET-based dust and smoke optical properties. For multiple smoke and dust plume events around the world, our retrieved AOCH values agree with space-borne lidar CALIOP counterparts, with a mean bias of <0.15 km and a correlation coefficient of 0.85–0.87. Due in part to adding the O2 B band, our retrieval represents an aerosol extinction peak height better than the TROPOMI operational Level 2 aerosol layer height retrieved from only the O2 A band. The latter shows 0.5–2 km low bias, especially over land. Finally, the high potential of AOCH for improving surface PM2.5 estimates is also illustrated with a case study in which the high bias of surface PM2.5 in MERRA-2 data is corrected after being scaled by the retrieved AOCH.

PDF of Publication: 
Download from publisher's website.
Research Program: 
Applied Sciences Program (ASP)
Modeling Analysis and Prediction Program (MAP)
Atmospheric Composition
Atmospheric Composition Modeling and Analysis Program (ACMAP)
Tropospheric Composition Program (TCP)