A fast and accurate radiative transfer model for aerosol remote sensing. (November 2020)

• Record Type: Journal Article
• Authors: Mei, Linlu; Rozanov, Vladimir; Burrows, John P.
• Is Part Of: Journal of quantitative spectroscopy & radiative transfer. Volume 256(2020)
• Publication Date: 2020-11
• Subjects: Fast radiative transfer -- XBAER -- Aerosol -- Remote sensing; Spectrum analysis -- Periodicals; Radiation -- Periodicals; Analyse spectrale -- Périodiques; Rayonnement -- Périodiques; Radiation; Spectrum analysis; Periodicals; 543.0858
• DOI: 10.1016/j.jqsrt.2020.107270
• Languages: English
• ISSNs: 0022-4073

Abstract

Highlights: A fast radiative transfer model, FASMAR, for aerosol remote sensing was developed. The accuracy of FASMAR is better than 3% for GEO and LEO satellite observations. FASMAR is valid for all surface and atmospheric conditions. FASMAR is used to create 'NRT FIRSTLOOK' global aerosol products. Abstract: After several decades' development of retrieval techniques in aerosol remote sensing, no fast and accurate analytical Radiative Transfer Model (RTM) has been developed and applied to create global aerosol products for non-polarimetric instruments such as Ocean and Land Colour Instrument/Sentinel-3 (OLCI/Sentinel-3) and Meteosat Second Generation/Spinning Enhanced Visible and Infrared Imager (MSG/SEVIRI). Global aerosol retrieval algorithms are typically based on a Look-Up–Table (LUT) technique, requiring high-performance computers. The current eXtensible Bremen Aerosol/cloud and surfacE parameters Retrieval (XBAER) algorithm also utilizes the LUT method. In order to have a near-real time retrieval and achieve a quick and accurate "FIRST-LOOK" aerosol product without high-demand of computing resource, we have developed a Fast and Accurate Semi-analytical Model of Atmosphere-surface Reflectance (FASMAR) for aerosol remote sensing. The FASMAR is developed based on a successive order of scattering technique. In FASMAR, the first three orders of scattering are calculated exactly. The contribution of higher orders of scattering is estimated using an extrapolation technique and an additional correction function. The evaluation of FASMAR has been performed by comparing with radiative transfer model SCIATRAN for all typical observation/illumination geometries, surface/aerosol conditions, and wavelengths 412, 550, 670, 870, 1600, 2100 nm used for aerosol remote sensing. The selected observation/illumination conditions are based on the observations from both geostationary satellite (e.g. MSG/SEVIRI) and polar-orbit satellite (e.g. OLCI/Sentinel-3). The percentage error of the top of atmosphere reflectance calculated by FASMAR is within ± 3% for typical polar-orbit/geostationary satellites' observation/illumination geometries. The accuracy decreases for solar and viewing zenith angles larger than 70 ∘ . However, even in such cases, the error is within the range ± 5%. The evaluation of model performance also shows that FASMAR can be used for all typical surfaces with albedo in the interval [ 0 − 1 ] and aerosol with optical thickness in the range [ 0.01 − 1 ] .