SPLITSnow: A spectral light transport model for snow

Snow is a fundamental component of the climate system. It is also an important part of the planet's hydrological cycle. Accordingly, the investigation of its light scattering properties is essential for remote sensing applications employed in the estimation of changes in the current amount of snowpack. These wide-scale environmental changes are key indicators of future climate events affecting global sustainability. Viewed in this context, computational simulations of light interactions with snow can be used to increase the effectiveness-to-cost ratio of remote sensing initiatives in this area. More specifically, by enabling a controlled assessment of the effects of snow granular structure and composition parameters on its light reflection and transmission profiles, these simulations can be instrumental in the high-fidelity interpretation of data remotely acquired from snow-covered landscapes that pose sizable challenges for field work. In order to contribute to these interdisciplinary research efforts, this paper presents a novel light transport model for snow that can predictively simulate the spectral and spatial distributions of light interacting with this ubiquitous particulate material. While the former radiometric responses are quantified in terms of hyperspectral reflectance and transmittance, the latter are quantified in terms of BSDF (bidirectional scattering distribution function). The proposed model employs a first-principles simulation approach that accounts for the positional dependence of the scattered light in the quantification of its spatial distribution. Thus, this distribution can also be expressed in terms of BSSDF (bidirectional surface-scattering distribution function). The predictive capabilities of the proposed model are quantitatively and qualitatively evaluated by comparing modeled results with measured data obtained from in situ experiments and phenomenological traits reported in the literature, respectively.