Additions to the SASKTRAN Radiative Transfer Framework: Air Mass Factors and Rotational Raman Scattering Using the Monte Carlo Method

Abstract:

SASKTRAN is a fully spherical radiative transfer framework developed at the University of Saskatchewan. While it was originally designed for limb applications, recent efforts have been directed at applying it to nadir viewing geometries, specifically the glancing viewing angles typical of measurements by TEMPO over Canada. To this end, two additions have been made to SASKTRAN’s Monte Carlo engine, which was originally developed to play a complimentary role to the primary successive orders engine. The first addition is an air mass factor calculation, which provides reliable verification for air mass factors calculated with other methods. The second addition is rotational Raman scattering, allowing for more accurate forward modelling as well as efficient generation of Ring spectra used for retrievals.

Air mass factor comparisons between the Monte Carlo method and the successive orders finite difference method show good agreement, even under extreme viewing geometries, with convergence as the resolutions of the successive orders engine are increased. Using this method to generate large sets of air mass factors is not practical, due to the large volume of samples required to reduce the variance, and due to its inability to analyze multiple viewing geometries simultaneously, but it excels as verification for the successive orders engine. The initial results of the rotational Raman implementation are promising, appearing consistent with examples in the literature. It provides reasonably efficient generation of the Ring spectra required for many air quality retrieval algorithms, and it will serve as an accurate baseline for future rotational Raman implementations in SASKTRAN’s more efficient engines.