Effect of snow grain shape and impurities on snow albedo and its parameterization: China, North America, and the Arctic

Abstract

Snow cover is important for Earth’s surface energy budget, primarily because the surface reflectance (albedo) is greatly increased if covered by a layer of snow. A correct estimation of snow albedo is therefore crucial for studying climate and snow hydrology. Snow albedo is influenced by many factors including snow depth, snow grain size, solar zenith angle, cloud cover, and light-absorbing particles (LAPs, black carbon, organic carbon, and mineral dust) in snow, and snow albedo is usually obtained from radiative transfer calculations. Therefore, an error in those factors and assumptions used in radiative transfer calculations could introduce errors in the computed snow albedo. This dissertation presents three separate works related to snow albedo, aiming towards a better understanding and quantification on the impact of different factors on snow albedo. First, a new parameterization method for narrowband and broadband albedo of pure snow and snow containing black carbon and mineral dust is introduced. Spectral albedo of snowpacks with different grain radii (5 – 2500 μm), containing a wide range of black carbon or mineral dust amounts (mixing ratio of 0 to 1) were calculated using radiative transfer models (DISORT). For each case, three broadband albedos (visible, near-IR, and all-wave) and twelve narrowband albedos (RRTM bands 2-13) are calculated and parameterized as functions of snow grain radius and LAP concentrations. This method can be incorporated into climate models to calculate snow albedo or study the impact of LAPs on snow albedo. Second, the effect of snow grain shape on snow albedo is studied. Radiative transfer calculations on snow albedo have usually assumed a spherical shape for snow grains, using Mie theory to calculate the single-scattering properties of ice spheres. The scattering by more realistic non-spherical grains is less in the forward direction and more to the sides. Incident sunlight scattered in the forward direction travels a longer path in the snowpack before reemerging from the snowpacks, and is therefore more likely to be absorbed. For snowpacks with the same area-to-mass ratio, albedo of snowpacks consisting of non-spherical grains is higher than the albedo of snowpacks consisting of spherical grains; the albedo reduction caused by the same amount of black carbon is smaller for snow consisting of non-spherical grains. Third, the albedo reduction by LAPs in snow over large areas is calculated using field observations and radiative-transfer modeling. Black carbon, mineral dust, and organic carbon may deposit on snowpacks and reduce the very high albedo of snow. As a large fraction of LAPs (especially black carbon) is emitted from anthropogenic sources, the direct albedo reduction induced by LAPs exerts a positive radiative forcing on snow-covered regions. This direct forcing may trigger a series of feedback processes that lead to an amplified radiative forcing, and is therefore crucial for climate and snow hydrology studies. Large-area surveys of LAPs in snow have been carried out by our group and collaborators in the Arctic, China, and North America, and based on these datasets, the albedo reduction caused by black carbon and other LAPs are calculated, and the impact of different factors on derived albedo reductions are also estimated.