Premium
Capability of Superspheroids for Modeling PARASOL Observations Under Dusty‐Sky Conditions
Author(s) -
Lin W.,
Bi L.,
Weng F.,
Li Z.,
Dubovik O.
Publication year - 2021
Publication title -
journal of geophysical research: atmospheres
Language(s) - English
Resource type - Journals
eISSN - 2169-8996
pISSN - 2169-897X
DOI - 10.1029/2020jd033310
Subject(s) - radiance , sky , radiative transfer , polarization (electrochemistry) , physics , atmospheric model , linear polarization , scattering , anisotropy , atmospheric radiative transfer codes , computational physics , remote sensing , optics , meteorology , geology , laser , chemistry
Abstract A comprehensive dust‐particle geometry model is highly required for accurate computations of optical parameters in radiative transfer simulations and relevant remote sensing applications. In this study, a superspheroidal model is proposed for simulating polarized radiation at the top of the atmosphere (TOA) under dusty‐sky conditions. The superspheroidal model has one more degree of freedom than the spheroidal model. Sensitivity studies are first conducted to investigate how the additional freedom in the superspheroidal dust model affects the polarized signals at the TOA followed by an examination of the impact of particle size, complex refractive index, and surface properties on these polarized signals. The applicability of the superspheroidal model is then assessed for 11 selected dust events in three main dust source regions (namely, North Africa, Middle East, and the Taklamakan Desert). Specifically, the normalized polarized radiance as observed by the Polarization and Anisotropy of Reflectances for Atmospheric Sciences coupled with Observations from a Lidar (PARASOL) are compared with simulations from an adding‐doubling vector radiative transfer model. It is found that the concave superspheroidal model with large roundness parameters achieves favorable performances in fitting the angular distribution of the PARASOL polarized radiance. The optimal roundness parameter is found to be 2.6–3.0 and is consistent with recent comparison of the simulated scattering matrices and their laboratory measurement counterparts. These findings support the potential applicability of the superspheroidal model for polarized remote sensing applications.