User Guide

MSCART

Introduction

The Multiple-Scaling-based Cloudy Atmospheric Radiative Transfer (MSCART) model is a universal Monte Carlo code for radiative transfer in three-dimensional cloudy atmospheres, serving as a useful tool for passive and active remote sensing purpose (Wang et al 2017 and 2019, Emde et al 2018, Zhang et al 2020). It supports both the forward and backward Monte Carlo photon-transport simulations. In forward mode, it can simulate average radiances over horizontal area in specified directions for solar source and range-resolved signals for lidar source; In backward mode, it not only simulate average radiances over horizontal area in specified directions for solar source or terrestrial source, but also give radiances at specified locations in specified directions for solar source. It was initially developed for scalar radiative transfer simulation. Recently, it has been added with polarization feature, i.e. vector radiative transfer simulation, and validated by the IPRT intercomparisons.

The sophisticated variance reduction techniques are applied to accelerate the convergence rate of radiance simulations for cloudy atmosphere with highly forward-peaked scattering. In previous studies, the variance reduction is achieved either by using the scattering phase function forward truncation technique or the target directional importance sampling technique. We now present a novel scattering order-dependent variance reduction method to combine them and a new scattering order sampling algorithm to achieve an order-dependent tuning parameters optimization strategy (see Wang et al 2017). The IPRT intercomparison of Phase B (3D test cases) demonstrates that the variance reduction technique can greatly improve the trade-off between numerical efficiency and accuracy order by order (see Emde et al 2018).