Spectral and broadband albedo of Antarctic sea-ice types


Stephen Warren and Maria Zatko

Spectral albedo across the ultraviolet, visible, and near-infrared wavelengths will be measured for a variety of sea-ice types.  The project continues a collaboration between the University of Washington and the Australian Antarctic Division, which has involved expeditions in 1988, 1996, and 2000.

In the prior work spectral radiometers were used that covered the wavelengths 300-1060 nm.  Computation of broadband albedos, used in assessing the solar energy budget of the Antarctic sea-ice zone, and computation of albedos for partial solar bands used in general circulation models (GCMs) and for remote sensing, were accomplished by multiplying clear-sky and cloudy-sky solar spectra with the measured spectral albedos, making assumptions about the behavior of the albedo spectrum at the wavelengths not measured, 1060-2500 nm.

The instrument to be deployed in the proposed expedition covers the entire solar spectrum at high spectral resolution, out to 2500 nm, so the earlier assumptions can be verified or corrected.  The region 1060-2500 nm includes some important absorption bands of ice, where the spectral albedo of snow undergoes large variations.  These variations are expected to be muted in snow-free sea ice.  Specular reflection may cause the albedo of ice to exceed that of snow at the strongly-absorbing wavelengths near 1500 and 2000 nm.

The surface types expected to be encountered in September and October are open water, grease ice, nilas, pancakes, young grey ice, young grey-white ice, and first-year ice, both bare and with varying thicknesses of snow cover.  Their albedos from prior work spanned a wide range, from 0.07 for open water to 0.87 for thick snow-covered ice under cloud.  The voyage will provide the opportunity to measure some ice types that were not sampled in the prior work.

The frequency-distribution of ice types and snow thickness in all seasons, from the project on Antarctic Sea Ice Processes and Climate (ASPeCt), will be used together with the measured albedos for each ice type to update the climatology of area-averaged albedo.

Light-absorbing impurities will be measured for snow on Antarctic sea ice in an attempt to explain some measurements of lower-than-expected visible and UV albedos.  The black-carbon content is expected to be considerably lower than on Arctic sea ice, because the Antarctic sea ice is remote from sources of soot.  However, the absorption by brown carbon may be more significant than in the Arctic because of the greater biological productivity of the Antarctic Ocean.  Snow samples will be melted and filtered, and the filters will be analyzed in a laboratory spectrophotometer as was done for the Arctic.

Stephen Warren <sgw@uw.edu>
Maria Zatko <mzatko@uw.edu>