Australia: The Land Where Time Began

A biography of the Australian continent 

Antarctic Sea Ice

There is a long distance separating Antarctic from any other landmasses, which differs from the case of the Arctic sea ice which is only hundreds of kilometre from the surrounding landmasses of Eurasia and North America. The shortest distance between Antarctica and another land mass is 1,200 miles to South America across the Drake Passage. There are at least 2 reasons why this difference between Antarctica and the Arctic matter.

The feedback effect from the retreat of snow and ice must be computed for the entire planet. It is known that sea ice in the Arctic is retreating at a speed that greatly exceeds the projections made by most climate models, which predict a slower retreat that is in tune with general global warming. Arctic sea ice is an anomaly in this respect. Antarctic sea ice is an even greater anomaly as [see 2016] it is advancing. Sea ice has been expanding in spite of an overall warming over the Antarctic continent. Wadhams suggests that if Antarctic sea ice is advancing this will help offset the reduction of global albedo that is due to the retreat of Arctic snowline and sea ice. Also, advancing Antarctic sea ice is as much a challenge to global climatic models as is the retreat if the sea ice in the Arctic – a slow retreat is predicted for both polar regions by computer models, so they are wrong at both ends. The extent of Antarctic sea ice reached a record maximum of 19.47 km2 in September 2012, according to the US Snow and Ice Center (NSIDC) in Boulder. This is approximately 30,000 km2 larger than the previous record that was set in 2012, and is 2.6 % higher than the average for 1981-2010. The area covered by sea ice has declined somewhat in more recent years to 18.83 million km2 in 2015, which is suggested by Wadhams to possibly be due to the onset of an El Niño atmospheric pattern in the Southern Hemisphere (see website www.climate,nasa,gov/news/), though it still shows a slow increasing trend.

It is know from passive microwave instruments on satellites that overall Antarctic sea ice is advancing. This advance is occurring, in spite of at least 1 part of the Antarctic - the Antarctic Peninsula – warming very fast (Rignot et al., 2008), leading in 2002 to the spectacular collapse of the Larsen B ice shelf, an event in which an area of 3,250 km2, that was 200 m or more thick, broke up into many icebergs which drifted away, and left the coastline and islands available to shipping for the first time in recorded history on the eastern side of the Antarctic Peninsula.

So why is the Antarctic sea ice advancing in the face of a climate that is warming and the loss of the area of ice shelf? It is necessary to understand the difference between the Antarctic sea ice and the sea ice in the Arctic before that question can be answered. A difference is that the Arctic is an ocean surrounded by land while Antarctica is a huge landmass at the Pole that is surrounded by a vast ocean, though a similarity is that the size and shape of the Arctic Ocean is very similar to the size and shape of Antarctica. There is a tendency for Antarctica to be isolated by wind patterns and ocean currents, so it can exhibit trends that are uncoupled from those of the Arctic.

Why Antarctic ice differs from Arctic ice

The difference between Antarctic sea ice and Arctic sea ice is that Antarctic sea ice forms in a different way, and its properties and appearance differ from those of Arctic Ice. Early in the Southern Hemisphere winter sea ice begins forming close to the coast of Antarctica, and as winter progresses the edge of the ice advances northwards into the great Southern Ocean where it is exposed to all the power of the biggest ocean in the world. The mechanism of the sea ice formation was not understood until an expedition was able to work in the pack ice zone in winter at the time of advance of the ice edge. The first time this took place was in 1986, when the German research ship FS Polarstern was used for the Winter Weddell Sea Project. Careful study of the ice conditions and characteristics was made as the ship traversed the region of the ice margin, and frazil-pancake cycle was identified, and named, as the source of most of the first-year ice seen deeper in the pack (Wadhams, Lange & Ackley, 1987).

Ice that grows on calm water forms an initial skim which solidifies as a thin transparent layer called nilas; the ice is then extended downwards by water molecules freezing to the bottom of the nilas, with crystals with horizontal c-axes being favoured, eventually yielding a first-year ice sheet. At the extreme edge of Antarctic ice it is unable to grow into a continuous sheet of nilas like this because of the high energy of the wave field in the Southern Ocean, which maintains the new ice as a dense frazil ice suspension. As a result of the particle orbits in the wave field this suspension undergoes cyclic compression, and during this compression phase the crystals can freeze together, eventually forming small coherent cakes of slush, which then increase in size by accretion from the frazil ice and continued freezing between the crystals increase the solidity of the ice. This forms what is known as pancake ice as collisions between the cakes pump a suspension of frazil ice onto the edges of the cakes, then a raised rim of frazil ice is left as the water drains away and this gives the cakes the appearance of a pancake. Pancakes are only a few centimetres in diameter at the ice edge, but their diameter and thickness gradually increase with increasing distance from the ice edge:  they may reach a diameter of 3-5 m and a thickness of 50-70 cm. As the surface of the water is not completely closed off by ice and there is a large ocean-atmosphere heat flux is still possible which can dispose of latent heat, the surrounding frazil continues to grow and supply material to the growing pancakes. This is the exact mechanism by which the Odden ice tongue forms in the Greenland Sea, though there are differences that occur as the ice develops further.

At some point at a greater distance deeper within the ice edge some protection from the waves is afforded by the loss of wave energy at the edge, the pancakes begin to freeze together in groups, though during the winter of 1986 when the research was carried out the wave field was enough to prevent overall freezing until a distance of about 270 km from the ice edge was reached. The pancakes then coalesced to form large flows, and eventually a large continuous sheet of first-year ice formed. At this point the rate of growth dropped to a very low level, that was estimated to be 0.4 cm/day (Wadhams et al., 1987), as the open water surface cut off, and the thickness of the first-year was only a few centimetres more than when the pancakes consolidated (Ibid).

First-year ice that has formed in this way is called consolidated pancake ice and the shape of the underside of this ice differs from that of Arctic ice. At the time of consolidation the pancakes are jumbled together and rafted over one another, freezing together in this configuration with frazil ice acting as ‘glue’. As a result the shape of the lower surface of the ice is rough and jagged, with the normal ice thickness being doubled or tripled, and with the edges of pancakes protruding above the upper surface of the ice to give a shape that Wadhams described as a ‘stony field’ because it had the appearance of a landscape resembling tiny fields surrounded by drystone walls.

The rafted lower surface of consolidated pancake ice provides a large surface area per unit area of sea surface, which provides an excellent substrate for the growth of algae as well as a refuge for krill. Plenty of light penetrates the thin ice which allows the phytoplankton to photosynthesise and live on the underside of the ice. This results in a fertile winter ice ecosystem which is believed to contribute about 30 % to the total biological production of the Southern Ocean.

According to Wadhams 30 years after the first research expedition in the midwinter Antarctic pack ice not much more research has been carried out in the pack ice at midwinter in Antarctic waters. There is not yet enough evidence to be sure if the frazil-pancake sequence of ice growth is followed all the way around the periphery of Antarctica, and Wadhams suggests that if it is the area occupied by Antarctic pancake ice in early winter could cover a total area of up to 6 million km2, which would make it an important, though seldom seen, component of the surface of the Earth.

Snow on the ice

The proximity of the vast Southern Ocean results in more moisture, and so more precipitation, with the result that the annual snowfall onto the Antarctic ice being much greater than in the Arctic Ocean, As well as in coastal regions snow also being blown onto the sea ice by katabatic winds (winds blowing down the slope of the Antarctic ice sheet from its summit) off the tops of the ice shelves. A mean snow thickness of 14-16 cm was found on the surface of first year ice during the cruise of the Polarstern in July-September of 1986. This was sufficient to push the surface of the ice below sea level in 15-20 % of the holes that were drilled, as the ice itself was so thin, which leads to the infiltration of sea water into the overlying snow and the formation of either of a layer that was wet and slushy on top of the ice, or in the case of freezing, a layer of ‘snow ice’ between the original upper surface of the ice and the wetted snow. The snow was even thicker in September-October of 1989, and in the Weddell Sea, over multi-year ice in the western part of the sea. This was enough to push the surface of the ice below sea level in almost every case. The ice and its slushy layer is insulated by the thick snow and as a result of its slushy wetness satellite radar methods for mapping the thickness of the ice do not work very well as the wet snow reflects the radar beam. Wadhams suggests there is no doubt that a much bigger role is played by the snow, and the slushy layer that is formed when water infiltrates into the snow (‘meteoric ice’), in the Antarctic sea ice than it does in the Arctic (Massom et al., 2001). See Antarctic Sea Ice Expansion - Important role of Ocean Warming and Increased Ice-Shelf Melt

Annual ice cycle and its changes

A problem for climate change models is that the extent of Antarctic sea ice has been increasing slowly in recent years, though there is a high regional variability. There are only 2 substantial areas of sea ice that remain in summer, in the Western Weddell Sea and the western Ross Sea, therefore these are the only areas that can contain much multi-year ice, the ice type that dominated the Arctic until recently. There is little variation of the minimum from year to year. Now ice forms to the north of the ice edge and the limit of sea ice advances until reaching a maximum by the end of winter (August-September) at between 55oS and 65oS, at which point it begins retreating back to its starting point. In the Indian Ocean sector the northern limit is 55oS at about 15oE, though around most of the remainder Antarctica it is about 60oS, and off the Ross Sea it drops to 65oS. At 150oW the edge moves to 62oS, then off the Amundsen Sea it shifts to the south to 66oS, then finally shifts north to engulf the South Shetland Islands and the South Orkney Islands of the Antarctic Peninsula and complete the circle. The latitudinal variation of this winter maximum around the Antarctic continent therefore is about 11o.

The edge of the Antarctic Circumpolar Current is the absolute limit of the northwards advance of the ice, at which point the temperature of the surface water abruptly changes in the Polar front, or Antarctic Convergence. At this point everything changes as well – it is the point at which southward-bound ships encounter icebergs, penguins, albatrosses, skuas, as well as a profusion of other Antarctic birds, rich plankton, including the shrimp-lime krill, and the great whales that feed on them. The colour of the sea water changes to green and there is a smell of life in the air. The ice rarely reaches this natural boundary, as its advance is limited by ocean processes, such as storms and eddies, both of which break it up, and by the temperatures of the surface air: it was shown by Zwally et al. at NASA (Zwally et al., 1983) that the advance in winter of the ice edge closely follows the advance of the surface air which is colder than the freezing point of sea water (1.8oC) and almost coincides with this temperature line (or isotherm) at the time the advance reaches its maximum. Satellites can easily measure the magnitude of this annual cycle of the extent, which is defined as the area of the main ice edge, especially the NASA passive microwave satellites (SMMR, SSM/I and SSMIS), The results obtained by the NASA group at the Goddard Space Flight Center in Greenbelt, Virginia, Maryland for the period from 1978 to 2011 (Bromwich et al., 2013). For the period the average maximum and minimum extent of the ice extents were 18.5 million km2 and 3.1 million km2.

According to Wadhams there is clearly a slow upwards trend in the maximum extent of the sea ice around Antarctica as a whole of 17,100 km2 per year. However the trend conceals a substantial degree of variation regionally and seasonally. The Ross Sea sector has seen the most rapid growth (13,700 km2 per year) with lesser contributions from the Indian Ocean sector and the eastern Weddell Sea, whereas a retreat of the ice edge at a rate of 8,200 km2 per year has occurred in the Bellingshausen/Amundsen seas of West Antarctica. Air temperatures over the Pacific sector of the Antarctic continent (Antarctic Peninsula to the Ross Sea) have warmed twice as rapidly as over the remainder of the continent (Steig et al., 2009), and an analysis of the temperature record at Byrd Station, at 120oW, shows that it warmed by between 1.6 and 3.2oC from 1958 to 2010, which is a very large increase (Bromwich et al., 2013). A decrease in the ice-covered season (number of day per year for which a given location has a cover of ice) by 1 to 3 days per year between 1979 and 2010 (Maksym, Stammerjohn, Ackley & Massom, 2012) is a reflection of the rapid warming over the Pacific sector (West Antarctica), whereas there was a slow increase in the ice-covered season in the Atlantic-Indian Ocean sectors. There is a clear message from the ice cover: a wide swathe of East Antarctica has a slowly growing ice cover, while in West Antarctica the ice cover is shrinking more rapidly, the net effect being a very slow growth.

There are other ice variations related to local topographic factors that are more detailed and usually visible in spring and summer. Wadhams says this is a much attenuated version of a mysterious polynya that was detected in winter of 1974-1976 in the middle of the pack ice in this sector but has not been seen since that observation, at least not as a full open water feature. The Waddell Polynya, as it was known, was located over the Maud Rise, which a plateau of reduced water depth. In the winter of 1986 the area was investigated by FS Polarstern, when it was found that the region was part of an area, the Antarctic Divergence, where there can be upwelling of deep water that is warmer, which allows enough heat to reach the surface and hence keep the region free of ice in winter (Bagriantsev, Gordon & Huber, 1989). It presumed that the region is balanced on a knife edge of instability as far as its winter ice cover is concerned, as it has not occurred since 1976. The winter cover in 1986 was highly concentrated, though very thin (www.climate,nasa,gov/news/). There is also a recurrent open water region that appears in the Ross Sea that is shown by the December distribution, the so-called Ross Sea Polynya, which has ice still present to the north of it. A series of small coastal polynyas can be seen actively opening along the coast of East Antarctica in November and December, which are mostly driven by offshore (katabatic) winds which drive the ice away from the coast as fast as it can form.

What is happening to the ice?

In the winter pack ice much of the ice is of pancake origin and is quite thin, and the ice limit is advancing instead of retreating, though the climate of much of Antarctica is warming.

Wadhams suggests a simple explanation is that the expansion of circumpolar ice (the ice over all of Antarctica) was given by Jinlun Zhang of the University of Washington. He suggested that it resulted from strengthening of the winds around the Antarctic continent ( The great belt of circumpolar west wind is the key. The strength of these winds has been measured by satellites since the 1970s, and this has shown that they have been steadily increasing in strength. The wind is coming mainly from the west and the average wind speed is higher. When considering a typical ice floe, it is blown in an eastward direction by the direct force of the wind stress on the surface of the ice, though there is also a force that tends to turn it to the left, i.e., to add a component of movement to the north applied by the Coriolis force.

Because of the increasing speed of the wind there is an increasing northward-acting Coriolis force in the ice floe, as the Coriolis force is proportional to speed of the object relative to surface of the Earth, which moves the flow more rapidly to the north, as it moves to the east. Therefore it will reach a latitude at which it will be melted by the warmer atmosphere, though as a result of the increasing speed to the north it will travel further to the north before it melts. The entire pack ice zone is being pushed to the north into warmer water by the wind. Wadhams suggests this may be a view that is too simplistic. The mechanism would lead to an increase of the extent of the ice, not necessarily its area, as it deals only with the dynamics of already existing ice. This can be explained by open water being left behind the ice moving to the north in winter which would freeze quickly under winter conditions. Also the increase is bound to be temporary as global warming will eventually win out over increased speed of the winds and the ice will not reach the lower latitudes. The mechanism is, however, rooted in simple physics and the observed fact that wind speeds have indeed increased.

Wadhams bases his suggestion on frazil-pancake cycle, and its interaction with the strengthened winds. Bigger and longer ocean waves have formed as a result of the stronger winds that blow around Antarctica. These longer waves are able to penetrate deeper into the marginal ice zone which can maintain it as frazil-pancake ice to a greater distance from the ice edge. It is known that frazil-pancake ice can grow much faster than continuous ice as a result of the atmosphere not being cut off from the water below, which allows heat from the ocean to be lost more easily to the atmosphere, which permits more rapid growth of ice. Wadhams suggest it may be that at a time of stronger winds and bigger waves the frazil-pancake zone is wider, growing ice more rapidly.

Antarctic response to changes occurring elsewhere

A model is required which considers whether forcing from elsewhere is able to cause regional effects on the Antarctic ice to explain the regional nature of the sea ice trends.

The Antarctic ice sheet is an obvious cause, though the effects will be long-term, as it is beginning to lose mass (Jacobs et al., 2012), though more slowly than the Greenland ice sheet. According to an estimate that was presented at the May 2016 Living Planet Conference in Prague is that the ice loss is presently 84 Gt per year, compared to 300 Gt per year, at least, from the Greenland ice sheet. It has been projected that if there is an increase in the rate of ice loss from Antarctica the Filchner-Ronne and Ross ice shelves will disintegrate, which would allow Antarctic glaciers, such as those in the Transantarctic Mountains, to debouch directly into the ocean. If this occurs it will accelerate rapidly the mass loss rate from Antarctica, which would lead to acceleration of the global sea level rise, as well as impacting Antarctic sea ice, if any exists by that time. This should not occur for a few centuries according to the predictions, with the exception of the disintegration of the ice shelf around Pine Island Bay and a region of East Antarctica where it is believed the ice sheet is potentially unstable should a ‘plug’ of coastal ice decay (Mengel & Levemann, 2014).

Wadhams suggest teleconnections (long-distance links) which can exist with the lower latitude oceans and atmosphere, and even with the northern latitudes extending up to the Arctic, need to be investigated in order that effects that are more immediate, which are determining the regional variation in the retreat and advance of Antarctic sea ice. There are many candidates that are known of for the linking mechanism. The Antarctic Circumpolar Wave, a system of waves on the Antarctic Circumpolar Current which propagate slowly to the east, though westwards relative to the current, and that Wadhams suggests may interact with the tropical El Niño-Southern Oscillation (ENSO) system. The El Niño (Holy Child Current) is a warm ocean current that has a variable intensity and that develops in late December (hence the name) along the coast of Ecuador and Peru, sometimes causing weather conditions that are catastrophic, though its name doesn’t apply to the anomaly of winds and currents that are present across the entire South Pacific. The Southern Annular Mode (SAM) (Cosimo, Kwok, Martin & Gordon, 2011), another complex mode of variability in the atmospheric circulation at high latitudes, has been more focused on in recent studies. It has been suggested that an El Niño year leads to an increase of sea ice in the Weddell Sea and reduced sea ice in the Pacific, and the opposite for La Niña leads to an increase of sea ice in the Weddell Sea and reduced sea ice in the Pacific, and the opposite for a La Niña year (La Niña refers to a cooling of the surface of the ocean off the western coast of South America, which occurs periodically every 4-12 years, affecting the Pacific and other weather patterns; it is opposite to El Niño); though recent discoveries about the about El Niño events in the Central Pacific complicate the ENSO link (Wilson, Bromwich, Hines & Wang, 2014). It is suggested that latitudinal teleconnections, that are wider-ranging, could relate to the link between warming in the Arctic and weather extremes in lower latitudes due to distortions of the jet stream (Francis & Vavrus, 2012), which could possibly involve onward links with patterns in the tropics and the Southern Hemisphere.

Fundamental differences between sea ice in the Arctic and the Antarctic must be depended on in any complete explanation of why there is a difference in the behaviour of sea ice between the Arctic and the Antarctic. Because of the greater area of ocean with its high heat capacity and the way the Antarctic continent is isolated by the Antarctic Circumpolar Current from the warmer ocean to the north. The sea ice limits around Antarctica are set in a different way than what occurs in the Arctic: the ice retreats to the land in the Antarctic, a substantial mass of ice remaining only in bights like the Weddell Sea that are awkwardly shaped, while the limits in winter are thermodynamic and are set by conditions in the open ocean. The situation is the opposite in the Arctic: the limit in winter is set by the surrounding landmasses, while in summer the ice retreats to an ocean limit that is set thermodynamically and dynamically. In the Antarctic albedo feedback is also less important than in the Arctic, as in Antarctica by the time of maximum solar insolation in late December the sea ice has already retreated almost to the continent, while in the Arctic the retreat of the ice in summer at the maximum for solar radiation, June, there is still a long time to its minimum in September and is therefore susceptible to forcing changes.

In the Arctic the rapid rate of warming itself causes feedbacks which lead to a further increase in the rate of warming. As well as ice albedo feedback there is also more albedo feedback due to the retreat of the terrestrial snowline, and additional warming that is potentially very serious that may result from the release of methane from the shelves in an ice free Arctic (Whiteman, Hope & Wadhams, 2013). In the Antarctic the retreat of the snowline and methane feedbacks cannot occur – due to the lack of shallow shelves and the inflexible area of snow cover. The Arctic amplification and greater feedbacks in the Arctic have the effect that whatever the interactions between the sea ice in the Antarctic and temperate oceans, it will always be the case over the next few decades and the rate of global warming will be determined by the Arctic more than the Antarctic. In this sense the Arctic is a driver and the Antarctic can be thought of as a passive trailer in, as Wadhams says, “the global warming road race to oblivion”.

Sources & Further reading

  1. Wadhams, P., 2016, A Farewell to Ice, Penguin Books Ltd


Author: M. H. Monroe
Last updated 30/09/2016
Journey Back Through Time
Experience Australia
Aboriginal Australia
National Parks
Photo Galleries
Site Map
                                                                                           Author: M.H.Monroe  Email:     Sources & Further reading