Assessing Superspheroids in Modeling the Scattering Matrices of Dust Aerosols

Abstract

Atmospheric dust particles are known to have diverse and irregular morphologies. In order to account for nonsphericity, the spheroidal model with an aspect ratio distribution has been extensively used for modeling the optical properties of dust. The spheroidal model is superior to the spherical shape assumption, but it requires further improvement. In this study, superspheroids' modeling capabilities were systematically examined by comprehensively comparing the spheroid's and superspheroid's scattering matrices. Superspheroids have one more degree of freedom than spheroids and can be nonspherical at an aspect ratio of unity. The invariant imbedding T-matrix and the improved geometrical optics methods were employed to compute superspheroids' single-scattering properties with a wide distribution of aspect ratios and a number of roundness parameters. We then assessed the spheroidal and superspheroidal models' applicability for simulating the scattering matrices of 25 dust samples from the Amsterdam-Granada Light Scattering Database. It was found that extreme aspect ratios for spheroids in reproducing the measurements were unnecessary if superspheroids were used. Even with equi-probable aspect ratio distribution, superspheroids with constrained roundness parameters (from 2.4 to 3.0) could achieve better performances in concurrently matching six nonzero scattering matrix elements from the laboratory measurements. Moreover, superspheroids demonstrated better performances than spheroids in achieving spectral consistency for modeling dust scattering matrices. Therefore, superspheroids appear to be highly promising for atmospheric radiative transfer and remote sensing applications.