Megadunes

In 2003-04, Jeff Severinghaus and colleagues conducted a firn air study in a region of “megadunes” near Vostok Station, Antarctica, to test the hypothesis that a deep “convective zone” of vigorous wind-driven mixing can prevent gas fractionation in the upper one-third of the polar firn layer. In the dunes, ultralow snow accumulation rates plus vapor transport cause firn grain diameters and pore sizes to grow up to 10 mm compared to typical grain sizes of 0.5 mm in most polar firn. The large pores make the permeability of firn to air movement orders of magnitude higher than normal. Steady katabatic winds blowing across regularly undulating dune topography at this site make it very likely that the air in the top few tens of meters of the firn is well mixed with the free atmosphere in a convective zone.

The unknown thickness of the convective zone has hampered the interpretation of nitrogen and argon isotope ratios in ice cores as indicators of past firn thickness, which is a key constraint on the climatically important variables of temperature, accumulation rate, and gas age-ice age difference.

The megadune study involved pumping from 20 depths in the firn to test for the presence of a convective zone. Permeability measurements on the core and 2-D air flow modeling (in collaboration with Mary Albert, CRREL) will permit a more physically realistic interpretation of the isotope data and will relate eddy diffusivities to air velocities. A new proxy indicator of convective zone thickness will be tested on firn and bubble air, based on the principle that isotopes of slow-diffusing heavy noble gases should be more affected by convection than isotopes of fast-diffusing nitrogen. These tools will be applied to a test of the hypothesis that the megadunes and a deep convective zone existed at the Vostok site during glacial periods, which would explain the anomalously low nitrogen and argon ratios in Vostok ice cores from glacial periods.

A new 3-inch Eclipse drill (Icefield Instruments, Whitehorse, YK) was used by ICDS driller Lou Albershardt, with assistance from other field party members. Despite some problems with the new drill, drilling proceeded to 115 meters, with 95% core recovery. That hole depth was well past the firn-ice transition, which met the basic requirement of the project, even though the planned-for depth of 50 m was not attained.