Retrieval of aerosol optical properties over a vegetation surface using multi-angular, multi-spectral, and polarized data

Abstract

An algorithm to retrieve aerosol optical properties using multi-angular, multi-spectral, and polarized data without a priori knowledge of the land surface was developed. In the algorithm, the surface polarized reflectance was estimated by eliminating the atmospheric scattering from measured polarized reflectance at 1640 nm. A lookup table (LUT) and an iterative method were adopted in the algorithm to retrieve the aerosol optical thickness (AOT, at 665 nm) and the Angstrom exponent (computed between the AOTs at 665 and 865 nm). Experiments were performed in Tianjin to verify the algorithm. Data were provided by a newly developed airborne instrument, the Advanced Atmosphere Multi-angle Polarization Radiometer (AMPR). The AMPR measurements over the target field agreed well with the nearby ground-based sun photometer. An algorithm based on Research Scanning Polarimeter (RSP) measurements was introduced to validate the observational measurements along a flight path over Tianjin. The retrievals were consistent between the two algorithms. The AMPR algorithm shows potential in retrieving aerosol optical properties over a vegetation surface.