Abstract
Small highly absorbing particles, present in concentrations of only 1 part per million by weight (ppmw) or less, can lower snow albedo in the visible by 5–15% from the high values (96–99%) predicted for pure snow in Part I. These particles have, however, no effect on snow albedo beyond 0.9 μm wavelength where ice itself becomes a strong absorber. Thus we have an attractive explanation for the discrepancy between theory and observation described in Part I, a discrepancy which seemingly cannot be resolved on the basis of near-field scattering and nonsphericity effects.
Desert dust and carbon soot are the most likely contaminants. But careful measurements of spectral snow albedo in the Arctic and Antarctic paint to a “grey” absorber, one whose imaginary refractive index is nearly constant across the visible spectrum. Thus carbon soot, rather than the red iron oxide normally present in desert dust, is strongly indicated at these sites. Soot particles of radius 0.1 μm, in concentrations of only 0.3 ppmw, can explain the albedo measurements of Grenfell and Maykut on Arctic Ice Island T-3. This amount is consistent with some observations of soot in Arctic air masses. 1.5 ppmw of soot is required to explain the Antarctic observations of Kuhn and Siogas, which seemed an unrealistically large amount for the earth's most unpolluted continent until we learned that burning of camp heating fuel and aircraft exhaust indeed had contaminated the measurement site with soot.
Midlatitude snowfields are likely to contain larger absolute amounts of soot and dust than their polar counterparts, but the snowfall is also much larger, so that the ppmw contamination does not differ drastically until melting begins. Nevertheless, the variations in absorbing particle concentration which will exist can help to explain the wide range of visible snow albedos reported in the literature.
Longwave emissivity of snow is unaltered by its soot and dust content. Thus the depression of snow albedo in the visible is a systematic effect and always results in more energy being absorbed at a snow-covered surface than would be the case for pure snow. Thus man-made carbon soot aerosol may continue to exert a significant warming effect on the earth's climate even after it is removed from the atmosphere.
