Aerosol Optical Depth Retrieval Over Land from Particulate Observing Scanning Polarimeter (POSP) Using a New Look-Up Table (LUT) Method

Abstract

Accurate estimation of Land Surface Reflectance (LSR) is the key to Aerosol Optical Depth (AOD) retrievals. However, the current band-specific LSRs retrieval using Look-Up Tables (LUTs) are typically pseudo-LSRs obtained by atmospheric corrections to the AOD predetermined in the LUTs that do not match the surface constraints established by the true LSRs, leading to an error in modeling reflectance at the top of atmosphere (TOA) using pseudo-LSRs calculation by linear interpolation. This study proposed a new LUT search method to improve the AOD retrievals from the Particle Observing Scanning Polarimetry (POSP) sensor onboard the China GaoFen-5 (02) satellite. Atmospherically corrected LSR using ERA5 reanalysis data and POSP AOD products for the year 2022 were adopted to create a new full-spectrum LSR self-consistent surface constraint. Results showed that the retrieved POSP AOD in January 2023 using the new method agrees with the ground-truth AOD values from AErosol RObotic NETwork (AERONET) site observations with the correlation coefficient (R) at 0.703 and the root mean square error (RMSE) at 0.068. 76.77% of the values fell into the expected error (EE) envelope of range ± (0.05 + 0.15 AOD_AERONET), and 67.35% met the accuracy requirements of the Global Climate Observing System (GCOS).