Sea ice exhibits a marked transition in its fluid transport properties at a critical brine volume fraction p c of about 5 percent, or temperature T c of about -5°C for salinity of 5 parts per thousand. For temperatures warmer than T c , brine carrying heat and nutrients can move through the ice, whereas for colder temperatures the ice is impermeable. This transition plays a key role in the geophysics, biology, and remote sensing of sea ice. Percolation theory can be used to understand this critical behavior of transport in sea ice. The similarity of sea ice microstructure to compressed powders is used to theoretically predict p c of about 5 percent.Sea ice is a complex, composite material consisting of pure ice with brine and air inclusions, whose size and geometry depend on the ice crystal structure, as well as the temperature and bulk salinity. It is distinguished from many other porous composites, such as sandstones or bone, in that its microstructure and bulk material properties vary dramatically over a small temperature range. For brine volume fractions p below a critical value p c Ϸ 5%, columnar sea ice is effectively impermeable to fluid transport, whereas for p above p c (Ͼ5%), brine or sea water can move through the ice. The relation of brine volume to temperature T and salinity S (1) implies p c corresponds to a critical temperature T c Ϸ -5°C for S ϭ 5 ppt; we refer to this critical behavior as the "law of fives." Perhaps the most direct observations of this are that the time rate of change of sea ice salinity dS/dt due to gravity drainage vanishes for brine volumes below 5% (2, 3) and that the permeability of thin sea ice decreases by more than two orders of magnitude as the surface temperature is lowered, in a small critical region around -5°C (4).Brine transport is fundamental to such processes as sea ice production through freezing of flooded ice surfaces, sea ice heat fluxes, and nutrient replenishment for sea ice algal communities, as well as being an important factor for remote sensing. However, the basic transition controlling brine transport has received little attention. Percolation theory (5, 6) has been developed to analyze the properties of materials where connectedness of a given component determines the bulk behavior. We show that it provides a natural framework to understand the critical behavior of sea ice. In particular, we apply a compressed powder percolation model to sea ice microstructure that explains the law of fives, the observed behavior (4) of the fluid permeability in the critical temperature regime, as well as data on surface flooding collected recently on sea ice in the Weddell Sea and East Antarctic regions. It was observed in the Arctic (7) that a snow storm and its resultant loading on a sea ice layer can induce a complete upward flushing of the brine network. In the Antarctic, it was observed that the freezing of a surface slush layer, with resultant brine drainage, induced convection within the ice, whereby rejected dense brine is replaced by nutrient-rich sea water ...

Abstract. Snow on Antarctic sea ice plays a complex and highly variable role in air-sea-ice interaction processes and the Earth's climate system. Using data collected mostly during the past 10 years, this paper reviews the following topics: snow thickness and snow type and their geographical and seasonal variations; snow grain size, density, and salinity; frequency of occurrence of slush; thermal conductivity, snow surface temperature, and temperature gradients within snow; and the effect of snow thickness on albedo. Major findings include large regional and seasonal differences in snow properties and thicknesses; the consequences of thicker snow and thinner ice in the Antarctic relative to the Arctic (e.g., the importance of flooding and snow-ice formation); the potential impact of increasing snowfall resulting from global climate change; lower observed values of snow thermal conductivity than those typically used in models; periodic large-scale melt in winter; and the contrast in summer melt processes between the Arctic and the Antarctic. Both climate modeling and remote sensing would benefit by taking account of the differences between the two polar regions. INTRODUCTIONAt maximum extent each year (September-October), sea ice covers a vast area of the Southern Ocean (---19 million km2), attaining latitudes as far north as ---55øS [Gloersen et al., 1992]. In so doing, it profoundly alters the exchange of energy and mass between ocean and atmosphere and forms an integral part of the global climate system. These effects are significantly amplified by the presence of an insulative snow cover which is itself highly variable in thickness and properties. Persistently strong winds redistribute the snow, and its properties [Gordon and Huber, 1990] on snow distribution and properties have only been conducted in the past 5-10 years. These studies are beginning to establish the full significance of snow on Antarctic sea ice as a key component of the global climate system. In this paper we review the major findings. Section 2 is a summary of snow data from five Antarctic sectors (designated by Gloersen et al. [1992]), namely, the Weddell Sea (20øE-60øW), the Indian Ocean (20øE-90øE), the western Pacific Ocean (90øE-160øE), the Ross Sea (160øE-140øW), and the Bellingshausen and Amundsen Seas (140øW-60øW), as shown in Figure 1. The Indian and western Pacific Ocean sectors are collectively referred to as the East Antarctic sector. Section 3 assesses the significance of snow in the air-sea-ice interaction system. New findings have significant implications for modeling (both physical and biological) and remotesensing studies of Antarctic sea ice. Gaps in our current knowledge are identified. Finally, the possible enhanced role of snow under global warming conditions is examined. Throughout, snow is described using the combined morphological and process-oriented classification of snow types of Colbeck et al. [1990] As a result, thickness may not be directly related to either the frequency or duration of snowfall.Mean snow thi...

An observational account of research carried out in July-August 1999 shows that grounded iceberg and related fast-ice distributions, and periodic "break-outs" of fast ice (in winter as well as at other times), have an important impact on the size and behaviour of the Mertz Glacier polynya, East Antarctica, and a smaller polynya to the east. Analysis of satellite and in situ data shows that a semi-constant "stream" of thick broken-out fast ice and other large floes from the east extends westwards from north of the glacier terminus to form a compact barrier to the net west-northwesterly export of ice formed in the polynya. An annual fast-ice promontory to the west further narrows the outlet path. As a result of this and high ice-production rates, the polynya periodically "back-fills", significantly reducing the open-water area present. Intervening "flush-outs" by synoptic storm events clear the polynya region to some extent before it back-fills again. This cycle continued from mid-March until early October in 1999, when a significant change in the regional ice drift occurred. A preliminary comparison with data from 1998 indicates that the timing and magnitude of the processes may vary interannually. Similar morphological features were also observed in 1963 (on a declassified photoreconnaissance satellite image).

An autumn bloom of sea-ice algae was observed from February to June of 1992 within the upper 0.4 meter of multiyear ice in the Western Weddell Sea, Antarctica. The bloom was reliant on the freezing of porous areas within the ice that initiated a vertical exchange of nutrient-depleted brine with nutrient-rich seawater. This replenishment of nutrients to the algal community allowed the net production of 1760 milligrams of carbon and 200 milligrams of nitrogen per square meter of ice. The location of this autumn bloom is unlike that of spring blooms previously observed in both polar regions.

The heat flux through the snow and sea ice cover and at the ice/ocean interface were calculated at five sites in the western Weddell Sea during autumn and early winter 1992. The ocean heat flux averaged 7 ± 2 W/m2 from late February to early June, and average ice/air heat flux in the second‐year floes depended on the depth of the snow cover and ranged from 9 to 17 (±0.8) W/m2. In late February, three of the five sites had an ice surface which was depressed below sea level, resulting, at two of the sites, in a partially flooded snow cover and a slush layer at the snow/ice interface. As this slush layer froze to form snow ice, the dense brine which was rejected flowed out through brine drainage channels and was replaced by lower‐salinity, nutrient‐rich seawater from the ocean upper layer. We estimate that about half of the second‐year ice in the region was covered with this slush layer early in the winter. As the slush layer froze, over a 2‐ to 3‐week period, the convection within the ice transported salt from the ice to the upper ocean and increased total heat flux through the overlying ice and snow cover. On an area‐wide basis, approximately 10 cm of snow ice growth occurred within second‐year pack ice, primarily during a 2‐ to 3‐week period in February and March. This ice growth, near the surface of the ice, provides a salt flux to the upper ocean equivalent to 5 cm of ice growth, despite the thick (about 1 m) ice cover, in addition to the ice growth in the small (area less than 5%), open water regions.

We present the first detailed maps of fast ice around East Antarctica (75°E–170°E), using an image correlation technique applied to RADARSAT ScanSAR images from November in 1997 and 1999. This method is based upon searching for, and distinguishing, correlated regions of the ice‐covered ocean which remain stationary, in contrast to adjacent moving pack ice. Within the overlapping longitudinal range of ∼86°E–150.6°E, the total fast‐ice area is 141,450 km2 in 1997 and 152,216 km2 in 1999. Calibrated radar backscatter data are also used to determine the distribution of two fast‐ice classes based on their surface roughness characteristics. These are “smooth” fast ice (−25.4 dB to −13.5 dB) and “rough” fast ice (−13.5 dB to −2.5 dB). The former comprises ∼67% of the total area, with rough fast ice making up the remaining ∼33%. An estimate is made of fast‐ice volume, on the basis of fast‐ice type as a proxy measure of ice thickness and area. Results suggest that although fast ice forms 2–16% of the total November sea ice area for this sector of East Antarctica in 1997 and 1999 (average 8.3% across maps), it may comprise 6–57% of the total ice volume (average ∼28% across maps). Grounded icebergs play a key role in fast‐ice distribution in all regions apart from 150°E–170°E. These are “snapshot” estimates only, and more work is required to determine longer‐term spatiotemporal variability.

A multilayer thermodynamic model is used to simulate sea ice growth for 12 years between 1958 and 1986 in the vicinity of the Australian station Mawson on the coast of East Antarctica. The atmospheric forcing data for the model are derived from radiosonde profiles and from surface measurements. Global radiation data are available for 4 years, and we use these measurements for comparison with the results of a Zillman‐type model for global radiation. Combining the thermodynamic model with sea ice thickness measurements for 12 years, we solve the energy balance equation for the oceanic heat flux. The oceanic heat flux is not constant but changes with time within the year and from year to year. The oceanic heat flux averages 7.9 W/m2, and the yearly means vary between 5 and 12 W/m2. Seasonal values of the oceanic heat flux range from 0 to 18 W/m2. From the yearly averaged values a decadal trend is evident: During the first years that were analyzed the yearly average lies well above 10 W/m2; then in the mid‐1970s a decrease to 9 W/m2 occurs, while for all later years the values are ∼6–8 W/m2. In general, the oceanic heat flux increases from the start of the fast ice formation season in early April until it breaks out in December or January. To compare the calculated oceanic heat fluxes for different years, we divide the total ice season into three characteristic time regimes of the sea ice growth and calculate the averaged oceanic heat fluxes for each regime. For the first regime (through August) the mean flux is 2.7 W/m2, for the middle regime (September) it is 8.4 W/m2, and for the final regime (October–January) it is 17 W/m2. We discuss the results of our model calculations in conjunction with current meter observations, which give evidence of seasonally varying intrusions of relatively warm Circumpolar Deep Water into Prydz Bay. Comparison of passive microwave data of sea ice extent and concentration (from the scanning multichannel microwave radiometer sensor) with the model results reveals a correlation between the magnitude of the oceanic heat flux and local features such as polynyas.

ABSTRACT. In October 2003 a campaign on board the Australian icebreaker Aurora Australis had the objective to validate standard Aqua Advanced Microwave Scanning Radiometer (AMSR-E) sea-ice products. Additionally, the satellite laser altimeter on the Ice, Cloud and land Elevation Satellite (ICESat) was in operation. To capture the large-scale information on the sea-ice conditions necessary for satellite validation, the measurement strategy was to obtain large-scale sea-ice statistics using extensive sea-ice measurements in a Lagrangian approach. A drifting buoy array, spanning initially 50 km × 100 km, was surveyed during the campaign. In situ measurements consisted of 12 transects, 50-500 m, with detailed snow and ice measurements as well as random snow depth sampling of floes within the buoy array using helicopters. In order to increase the amount of coincident in situ and satellite data an approach has been developed to extrapolate measurements in time and in space. Assuming no change in snow depth and freeboard occurred during the period of the campaign on the floes surveyed, we use buoy ice-drift information as well as daily estimates of thin-ice fraction and rough-ice vs smooth-ice fractions from AMSR-E and QuikSCAT, respectively, to estimate kilometer-scale snow depth and freeboard for other days. The results show that ICESat freeboard estimates have a mean difference of 1.8 cm when compared with the in situ data and a correlation coefficient of 0.6. Furthermore, incorporating ICESat roughness information into the AMSR-E snow depth algorithm significantly improves snow depth retrievals. Snow depth retrievals using a combination of AMSR-E and ICESat data agree with in situ data with a mean difference of 2.3 cm and a correlation coefficient of 0.84 with a negligible bias.