| Summary: | Ice Station Belgica was commenced in late winter 2007 in the Bellingshausen Sea as part of Sea Ice Mass Balance in Antarctica (SIMBA), an IPY 2007 cruise on the research vessel N.B. Palmer. A primary objective was to build on the work of previous Antarctic drift station experiments to geophysically characterize sea ice in terms of thickness, surface and ice bottom morphology, and ultimately area-unitized mass. A 24 day drift station was established at approximately 70°S and 93°W in mixed first-year and multi-year ice with three geophysical study sites selected on a 5 km 2 floe. A comprehensive time series assessment of elevation-surveyed transects ranging from 100 m to 300 m in length included snow surface elevation, snow depth, electromagnetic (EM) profiling, and direct drilling for ice draft and ice freeboard. Additional work included a snow surface morphology characterization of a 100 m x 300 m area between the primary time series EM transects. Correlation of EM ice thicknesses with collocated drilled ice thickness yielded equations for the correction of EM underestimation of thick deformed ice, particularly at pressure ridges. Mean ice thickness from corrected EM was compared to isostatic ice thickness calculated from surface elevation, snow depth, ice freeboard and respective snow, slush, ice, and sea water densities. Results were consistent, with mean ice thicknesses for multi-year ice of 2.35 m, 2.34 m, and 2.41 m, with similar variance, for corrected EM, drilling, and buoyancy methods respectively. Additionally, a mean ice thickness of 2.31 m was calculated from ASPeCt observations of the ice field associated with the floe, using the method incorporating mean sail heights and fractional coverage of surface deformities or ridging. Temporal series assessment of ice freeboard indicated a slightly negative mean ice freeboard (<0.04 m), with clear evidence of new snow-ice formation from the freezing of slush. The three distinct snow and ice regions assessed on the Belgica floe had mean corrected EM ice thickness of 0.52 +/- 0.04 m (+/- 1 std. deviation), 0.92 +/- 0.17 m, and 2.35 +/- 1.37 m, and mean snow depths of 0.08 +/- 0.03 m, 0.36 +/- 0.09 m, and 0.68 +/- 0.31 m respectively. Each ice type represented a sizable fraction of the floe's total area (similar to 20%, 40%, and 40% respectively from visual estimates) reflecting a complex dynamic and thermodynamic history of formation, as well as the difficulty in characterizing even a single floe by a single class or mean value for thickness and snow depth. Implications of these results are discussed with regards to the resolution of satellite-based altimetry and snow depth products and efforts to generate and validate satellite sea ice and snow thickness products.
|
|---|