Characteristics and distribution patterns of snow and meteoric ice in the Weddell Sea and their contribution to the mass balance of sea ice

Abstract

Abstract. Based on snow- and ice-thickness measurements at >11 000 points augmented by snow- and icecore studies during 4 expeditions from 1986 - 92 in the Weddell Sea, we describe characteristics and distribution patterns of snow and meteoric ice and assess their importance for the mass balance of sea ice. For first-year ice (FY) in the central and eastern Weddell Sea, mean snow depth amounts to 0.16 m (mean ice thickness 0.75 m) compared to 0.53 m (mean ice thickness 1.70 m) for second-year ice (SY) in the northwestern Weddell Sea. Ridged ice retains a thicker snow cover than level ice, with ice thickness and snow depth negatively correlated for the latter, most likely due to aeolian redistribution. During the different expeditions, 8, 15, 17 and 40% of all drill holes exhibited negative freeboard. As a result of flooding and brine seepage into the snow pack, snow salinities averaged 4‰. Through 18O measurements the distribution of meteoric ice (i.e. precipitation) in the sea-ice cover was assessed. Roughly 4% of the total ice thickness consist of meteoric ice (FY 3%, SY 5%). With a mean density of 290 kg/m3, the snow cover itself contributes 8% to total ice mass (7% FY, 11% SY). Analysis of ∆18O in snow indicates a local maximum in accumulation in the 65 to 75°S latitude zone. Hydrogen peroxide in the snow has proven useful as a temporal tracer and for identification of second-year floes. Drawing on accumulation data from stations at the Weddell Sea coast, it becomes clear that the onset of ice growth is important for the evolution of ice thickness and the interaction between ice and snow. Loss of snow to leads due to wind drift may be considerable, yet is reduced owing to metamorphic processes in the snow column. This is confirmed by a comparison of accumulation data from coastal stations and from snow depths over sea ice. Temporal and spatial accumulation patterns of snow are shown to be important in controlling the sea-ice cover evolution.

Sea ice and snow thickness and physical properties of an ice floe in the western Weddell Sea and their changes during spring warming

“…More importantly, based on earlier observations by Lange and Eicken (1991), Eicken et al (1994), and Lytle and Ackley (1996), we were expecting the widespread occurrence of negative ice freeboard as a result of thick snow depressing the snow/ice interface below the seawater level. Drill holes would irreversibly cause seawater to flush onto the ice surface if the ice had not yet been flooded for natural reasons (as observed during ISPOL, see below).…”

“…While drilling provides the most accurate ice and snow thickness estimates (e.g., Lange and Eicken, 1991;Eicken et al, 1994), the method is limited by its slow progress, in particular on thick second-year ice, preventing the acquisition of long profiles spanning and demarcating regions of different ice types. More importantly, based on earlier observations by Lange and Eicken (1991), Eicken et al (1994), and Lytle and Ackley (1996), we were expecting the widespread occurrence of negative ice freeboard as a result of thick snow depressing the snow/ice interface below the seawater level.…”

Changes in snow distribution and surface topography following a snowstorm on Antarctic sea ice

“…At larger scales, estimation of snowfall amounts on sea ice have been inferred from atmospheric reanalysis products [e.g., Bromwich et al, 1995;Maksym and Markus, 2008], which have not been directly validated over sea ice. Such estimations also fail to account for the amount of blowing and drifting snow that is actually lost to open water between ice floes (leads), which can amount to a significant portion of the total yearly mass balance [Eicken et al, 1994;Leonard and Maksym, 2011]. Eicken et al [1994] estimated that as much as 100 mm yr À1 was lost to leads from data from the Weddell Sea; however, there are scarce observations to better address this question over larger scales.…”

“…Such estimations also fail to account for the amount of blowing and drifting snow that is actually lost to open water between ice floes (leads), which can amount to a significant portion of the total yearly mass balance [Eicken et al, 1994;Leonard and Maksym, 2011]. Eicken et al [1994] estimated that as much as 100 mm yr À1 was lost to leads from data from the Weddell Sea; however, there are scarce observations to better address this question over larger scales. Similarly, Dery and Tremblay [2004] estimated that as much as 30% of annual snowfall was lost through sublimation of blowing snow, based on their observations during the Surface Heat Budget of the Arctic Ocean (SHEBA) experiment.…”

Wintertime heat flux to the underside of East Antarctic pack ice

“…Several recent studies Sturm et al, 1998] have found that the constantly changing synoptic conditions and the redistri- , 1975]. The ocean heat flux can play a critical role in determining the thermodynamic limit to which the ice will grow, the amount of snow-ice that may form [Eicken et al, 1994], and the ice melt rate in the spring. In this study we report on calculations of ocean heat flux Qo and its influence on the sea ice thickness in the region.…”