ALBERT

Description: The surface reflection of solar radiation comprises an important boundary condition for solar radiative transfer simulations. In polar regions above snow surfaces, the surface reflection is particularly anisotropic due to low Sun elevations and the highly anisotropic scattering phase function of the snow crystals. The characterization of this surface reflection anisotropy is essential for satellite remote sensing over both the Arctic and Antarctica. To quantify the angular snow reflection properties, the hemispherical-directional reflectance factor (HDRF) of snow surfaces was derived from airborne measurements in Antarctica during austral summer in 2013/14. For this purpose, a digital 180∘ fish-eye camera (green channel, 490–585 nm wavelength band) was used. The HDRF was measured for different surface roughness conditions, optical-equivalent snow grain sizes, and solar zenith angles. The airborne observations covered an area of around 1000 km × 1000 km in the vicinity of Kohnen Station (75∘0′ S, 0∘4′ E) at the outer part of the East Antarctic Plateau. The observations include regions with higher (coastal areas) and lower (inner Antarctica) precipitation amounts and frequencies. The digital camera provided upward, angular-dependent radiance measurements from the lower hemisphere. The comparison of the measured HDRF derived for smooth and rough snow surfaces (sastrugi) showed significant differences, which are superimposed on the diurnal cycle. By inverting a semi-empirical kernel-driven bidirectional reflectance distribution function (BRDF) model, the measured HDRF of snow surfaces was parameterized as a function of solar zenith angle, surface roughness, and optical-equivalent snow grain size. This allows a direct comparison of the HDRF measurements with the BRDF derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite product MCD43. For the analyzed cases, MODIS observations (545–565 nm wavelength band) generally underestimated the anisotropy of the surface reflection. The largest deviations were found for the volumetric model weight fvol (average underestimation by a factor of 10). These deviations are likely linked to short-term changes in snow properties.

Description: The synergy between airborne lidar, radar, passive microwave, and passive imaging spectrometer measurements was used to characterize the vertical and small-scale (down to 10 m) horizontal distribution of the cloud thermodynamic phase. Two case studies of low-level Arctic clouds in a cold air outbreak and a warm air advection observed during the Arctic CLoud Observations Using airborne measurements during polar Day (ACLOUD) were investigated. Both clouds exhibited the typical vertical mixed-phase structure with mostly liquid water droplets at cloud top and ice crystals in lower layers. The cloud top horizontal small-scale variability observed during the cold air outbreak is dominated by the liquid water close to the cloud top and shows no indication of ice in lower cloud layers. Contrastingly, the cloud top variability of the case observed during a warm air advection showed some ice in areas of low reflectivity or cloud holes. Radiative transfer simulations considering homogeneous mixtures of liquid water droplets and ice crystals were able to reproduce the horizontal variability of this warm air advection. To account for more realistic vertical distributions of the thermodynamic phase, large eddy simulations (LES) were performed to reconstruct the observed cloud properties and were used as input for radiative transfer simulations. The simulations of the cloud observed during the cold air outbreak, with mostly liquid water at cloud top, realistically reproduced the observations. For the warm air advection case, the simulated cloud field underestimated the ice water content (IWC). Nevertheless, it revealed the presence of ice particles close to the cloud top and confirmed the observed horizontal variability of the cloud field. It is concluded that the cloud top small-scale horizontal variability reacts to changes in the vertical distribution of the cloud thermodynamic phase. Passive satellite-borne imaging spectrometer observations with pixel sizes larger than 100 m miss the small-scale cloud top structures, which limits their capabilities to provide indications about the cloud vertical structure.