Estimating the cloudy-sky albedo of sea ice and snow from space

While satellites provide the means to monitor the temporal and spatial variability of surface albedo, their use has been limited to clear-sky areas because clouds obscure the surface at wavelengths in the solar spectrum. However, the effect of clouds on the surface albedo, especially that of snow and ice, is significant and should be considered in satellite retrievals. In this paper theoretical and observational evidence is given that shows the snow/ice albedo to be on the average 4-6% (absolute) higher under cloud cover than for clear skies, with a range of slightly less than 0 to approximately 15%. A method for retrieving the clear-sky broadband albedo of snow/ice from the advanced very high resolution radiometer is presented, and an adjustment for cloud optical depth is proposed. The cloudy-sky adjustment is independent of sensor type and could also be used with nonsatellite data sets. An application of the algorithm to data from the Surface Heat Budget of the Arctic Ocean experiment demonstrates that clear- and cloudy-sky snow surface albedo can be obtained from space with an uncertainty of approximately 7% absolute. While it may be sufficient to adjust a monthly clear-sky surface albedo climatology for clouds by incorporating the mean cloud effect of approximately 5%, adjustments for cloud optical depth should be performed with instantaneous retrievals.