Autumn sea ice thickness, ridging and heat flux variability in and adjacent to Terra Nova Bay, Ross Sea, Antarctica

Abstract

The variability of sea ice thickness and snow depth observed from the R/V Nathaniel B. Palmer in late May 1998 in and adjacent to Terra Nova Bay (TNB), Ross Sea, Antarctica, is described. Conductive heat fluxes are estimated on the basis of the observed snow and ice thicknesses. There was only 6% areal coverage of open water in TNB and 87% of the areal coverage of unridged ice was ≤0.3 m thick. Between the western and eastern margins of TNB there were increases in the area‐weighted unridged ice thickness (0.05–0.25 m) and the areal extent of ridging (2.5–18.5%). Ridged ice accounted for 44–66% of the total ice mass in TNB and was as vital to thickening the ice cover as it was outside the bay, where 30–60% of the total ice mass was contained within the 5–35% of ice that was ridged. Satellite passive microwave (Special Sensor Microwave Imager SSM/I) data show that the extensive ice cover in TNB was normal for this time of year probably because of the blocking effect of thick, ridged, 100% concentration pack ice outside the bay. Despite the extensive ice cover in TNB, there was an order of magnitude difference in area‐weighted mean heat flux between the bay and the pack ice. For example, depending on the choice of bulk snow thermal conductivity value ks (0.14 or 0.31 W m−1 K−1), the heat flux values for unridged ice were 199 or 216 W m−2 in TNB versus 35 or 49 W m−2 on the inbound pack ice leg and 20 or 39 W m−2 on the outbound pack ice leg. Factoring in the contribution of ridges to the ice thickness distribution decreased the heat flux in the bay and the pack ice by 25 or 36%, depending on ks. The greatest proportion of the heat flux in TNB occurred through the ice; no more than 30% of the heat loss was from open water. Since the ice conditions observed in TNB were apparently normal for the time of year, the heat fluxes also must have been quite normal.