Effects of vertical inhomogeneity on snow spectral albedo and its implication for optical remote sensing of snow

[1] The spectral albedo of dry, vertically heterogeneous snow is simulated using a multilayer discrete ordinate method radiative transfer code, with single snow grain scattering parameters calculated either from Mie theory for small grains or from a geometric optics code for large grains. In the near-infrared wavelengths the snowpack is effectively semi-infinite at a depth of around 5 cm; that is, the reflectance is the same as for a deep snowpack. In the visible wavelengths, however, the snow critical depth is more than 50 cm for coarse-grained snow, but the top 5-10 cm layer is the most important one. Two heterogeneous snowpacks with the same average grain size in the topmost layer but with different vertical distributions of grain sizes can have quite different spectral albedos because of greater penetration by visible than by near-infrared radiation. The spectral albedo computed from the multilayer model agrees better with measured albedo when the vertical variability of grain size and the measured composite grain size rather than the measured single grain size are considered.