A Blended Approach Toward Simulating Spectral Snow Reflectivity and Transmissivity Using Monte Carlo Photon-Tracking and X-ray Microtomography Surface Rendering

The interaction between visible and near-infrared radiation and individual snow particles is well described by geometric optics theory. However, this interaction is complicated enormously when considering a snowpack in aggregate due to the intricate microscopic details of the snow in ways that are not sufficiently understood to accurately model and predict radiative transfer through a snowpack in all circumstances. Monte Carlo ray-tracing and photon-tracking models provide advantages in radiative transfer modelling, especially in their ability to simulate radiative transfer through 3D renderings of snow microstructure derived from X-ray microtomography (microCT). Here we present a Monte Carlo radiative transfer model that combines several different ray-tracing methods to determine medium optical properties from closed surface 3D microCT snow renderings and simulate spectral albedo and transmissivity in the visible and NIR. This blended approach aims to relax common assumptions regarding snow grain size and shape, while limiting the computational and observational constraints of explicit photon-tracking through large snow samples. Notably, the model expands ray-tracing techniques applied to sub-1 cm3 snow samples to snowpacks of arbitrary depths. An initial evaluation of the model is presented through a comparison with measured spectral reflectivity and transmissivity of a snowpack in Vermont. For this evaluation, spectral reflectance measurements were collected with an ASD spectroradiometer alongside snow samples, which were measured with the microCT. Finally, we compare simulated optical properties to derived physical properties from the microCT samples and assess the impacts of snow microstructure on simulated snow transmissivity. This analysis reveals that snowpacks with low specific surface area (SSA) and low density are the most optically transparent. We discuss ongoing model developments and future improvements.