The Impact of a Spherical Atmosphere on Polarimetric Aerosol Inversion

Low sun illumination is associated with a small fraction (~15%) of our daytime. However, significant aerosol events have been observed in the early morning and early evening. For example, the diurnal PM coarse mode aerosols arrive at a peak during the rush hours of morning (7-9am) and evening (5-7pm) in some high population living areas (Mora et al., 2017). A majority of wildfire occurrences in the European region extend into the early evening in summer and spring (Lighthart and Shaffer, 1995). To date, most operational aerosol inversion approaches deploy the forward models that compute radiative transfer (RT) in a plane-parallel atmosphere. In this model, the atmosphere is assumed to be composed of many parallel slabs, each extending to infinity in the horizontal direction and having optical homogeneity. At low sun illumination (e.g., close to sunrise and sunset), however, this assumption does not hold true, and the light scattered out of the atmosphere is strongly impacted by the curvature of the Earth's atmosphere. Consequently, aerosol retrievals for low sun elevation are not attempted or expected to have a reduced performance.

With a focus on aerosol and cloud remote sensing, a polarimetric imager is anticipated to be deployed by NASA's Atmosphere Observing System (AOS) mission and launch within the next decade. The AOS polarimeter will fly in a sun-synchronous polar orbit and will include a portion of observations made at low sun elevations. To evaluate the spherical atmosphere impact on aerosol inversion in this case, we have performed RT modeling for a set of equatorial crossing times. And the aerosol bands in the spectral range from ultraviolet to shortwave infrared are selected for simulation. Both forward and inversion modeling errors caused by the plane-parallel atmosphere assumption are evaluated. We then improve aerosol inversion accuracy and efficiency by deploying a pseudo-spherical atmospheric RT model and computing single scattering in a full spherical atmosphere. This work is based on our existing RT model for Titan’s spherical-shell atmosphere (Xu et al., 2013) and the aerosol inversion algorithm (Xu et al., 2017) for NASA's Multi-Angle Imager for Aerosols (MAIA, Diner et al., 2018).