Observed effects of sastrugi on CERES top-of-atmosphere clear-sky reflected shortwave flux over Antarctica



[1] Determining the clear-sky top-of-atmosphere (TOA) albedo over snow from space requires knowledge of the bi-directional reflectance distribution function (BRDF), which itself is strongly influenced by the surface roughness of the snow. Sastrugi, a common element of surface roughness on Antarctica, tend to have a preferred azimuth direction, meaning the BRDF depends on the location and time of sampling. In this study we demonstrate that a sastrugi signal is present in the Clouds and the Earth's Radiant Energy System (CERES) reflectance measurements and TOA albedo estimates, leading to a spurious variation in instantaneous albedo as a function of solar azimuth of up to 0.08. By using the difference in flux between oblique and nadir views, we estimate the biases in monthly- and annual-mean 24-hour energy weighted clear-sky reflected TOA fluxes caused by sastrugi over Antarctica. At the grid box level, statistically significant monthly-mean biases of between ±15 Wm−2are found. For the entire Antarctic continent, monthly-mean biases are between 0.2 ± 0.9 Wm−2 to −1.7 ± 1.1 Wm−2 where a negative bias indicates the reflected flux is being underestimated. On an annual basis, the Antarctic bias is between −0.9 ± 1.1 Wm−2 and −1.0 ± 1.1 Wm−2. For the global annual mean clear-sky TOA flux, the bias caused by the presence of sastrugi is insignificant, −0.01 ± 0.02 Wm−2. By examining the anisotropy and the wind direction we infer that the negative TOA flux biases are likely to caused by sastrugi perpendicular to the solar azimuth whereas the positive TOA flux biases are likely to be caused by sastrugi parallel to the solar azimuth.