Australian Antarctic Territory, Territory of Heard Island and McDonald Islands, and observations on Macquarie Island Tasmania
Australian National Committee on Antarctic Research
prepared for the 2006 Australian State of the Environment Committee, 2006
External pressures of global warming and other environmental changes can be expected to have a major direct effect on the physical systems of the Antarctic region. Antarctica and the surrounding Southern Ocean are dominated and shaped by snow and ice which, while controlled by the climatic regime and very sensitive to climate change, also influence and provide major feedbacks to the global oceanic and climate system.
Many globally significant processes are driven by the unique climate and geography of the Antarctic. These include the uptake of carbon dioxide by the Southern Ocean; the overturning circulation of the deep ocean ; the balance between storage and discharge of water in the continental ice sheet; modification of surface energy, mass and momentum exchange by ice masses; and energy transfer between all levels of the atmosphere to space.
Variations in the winter extent of sea-ice play a key role in global oceanic circulation and are thought to have profound effects on algal growth under the ice and the reproduction of Antarctic krill, a small, highly abundant crustacean which is a major component of the Antarctic food chain.
For Antarctic and high southerly latitudes, a number of indicators show changes in atmospheric and ice-related parameters. Many of the changes are broadly consistent with overall expectations for a warming climate. Temperatures at Australian Antarctic stations reflect the broader Antarctic picture, which shows little warming across most of the continent except for the Antarctic Peninsula that does not form part of Australia’s Antarctic Territory. Some stations report weak cooling, and none of the trends, either positive or negative, are statistically significant. These and other Antarctic temperature trends have been associated with atmospheric circulation changes and may be linked with stratospheric ozone depletion in recent decades.
Ultraviolet radiation values at Australian stations are elevated, relative to expectations for sites at comparable northerly latitudes, and this is directly attributed to stratospheric ozone depletion (the ‘ozone hole’ ).
Satellite data have shown significant ecosystem-level effects of exposure to ultraviolet B (UVB) radiation. This occurred below a threshold stratospheric ozone concentration of 300 Dobson Units (DU), reducing the seasonal increase of chlorophyll by around 60 per cent at less than 300 DU. Inhibition occurs over time scales greater than one day, suggesting this is due to indirect effects of exposure to enhanced UVB radiation. Results indicate that ozone depletion can substantially reduce phytoplankton biomass , with consequences for productivity, trophodynamics and biogeochemistry in Antarctic waters. Exposure to increased levels of UVB, through ozone depletion appears to be detrimental to endemic species of moss in the Australian Antarctic Territory (Robinson et al 2005).
Glacier and ice sheet changes are occurring and consistent with expectations for a warming climate. The sub-Antarctic glaciers of Heard Island are retreating rapidly and creating new areas for vegetation colonisation and expansion. On the Antarctic Peninsula and over large parts of West Antarctica, total ice volume is decreasing. Ice volume in East Antarctica is increasing, apparently from increased snow accumulation. The ice volume of Antarctica as a whole is increasing, with a small offset (that is, decreasing influence) on sea-level rise. Historical and proxy data suggest a decrease in total sea-ice cover through the latter part of the 20th century.
There have been no significant continent-wide trends in air temperature change over the Antarctic, excluding the region of the Antarctic Peninsula, over the past 50 years. No clear conclusion can be drawn about any trends in inland Australia Antarctic Territory air temperatures, in contrast with data from the Antarctic Peninsula where the average warming over the last 40 years is greater than 0.4 °C per decade.
Nine of twelve Antarctic coastal stations (including the Australian stations of Casey and Davis) show a slight warming within this period, and three (including Australia’s Mawson) show a slight cooling. On average there is a warming trend around the coast of less than 0.1°C per decade (Jacka et al 2004).
The situation over the inland ice sheet is less clear because of the dearth of long-term observations. Weak cooling over much of the interior is widely reported in longer term instrumental records. Jacka and colleagues (2004) reported an average warming of 0.1 °C per decade for the interior, but with very high variability between sites; this includes data from less reliable and more recent automatic weather stations. No clear conclusion can be drawn about any trends in inland Antarctic air temperatures.
The overall pattern of Antarctic temperature changes is consistent with global atmospheric circulation changes in the latter part of the twentieth century. Specifically, the atmospheric pattern known as the Southern Annular Mode (SAM) has strengthened in recent decades. This has been associated with the strong warming in the Peninsula and the tendency to cooling elsewhere across the continent. Thompson and Solomon (2002) linked this strengthened SAM with stratospheric ozone loss.
Changes in stratospheric ozone over the Antarctic have a direct impact on the solar ultraviolet radiation (UVR) levels at the earth’s surface, with an increase in the biologically active UVB radiation expected due to ozone reduction.
Australia’s Antarctic stations have annual UVR totals significantly above those expected for their latitude and higher than those recorded at Macquarie Island and at many locations in Europe and the United Kingdom (Gies et al 2004a). This is believed to be due to the reduction of ozone, allowing increased transmission of UVB radiation.
The differences in yearly UV Index totals at Australia’s Antarctic stations are attributed to weather, cloud cover and ozone variations. Variability from year to year for Casey and Davis, which have long data records, is approximately seven per cent and nine per cent respectively, contrasting with two per cent to four per cent for Australian mainland sites. Maximum daily UV Index values for Australia’s Antarctic stations in spring are comparable to those measured in the southern capital cities of Australia (Gies et al 2004b).
There is, on average, a 60 per cent loss of total column ozone above Antarctica in spring. This is the major cause of summertime ozone losses at mid-latitudes in the southern hemisphere, including southern Australia (Ajtic et al 2004). Ozone loss over Antarctica appears to have stabilised during the 1990s, although there is no direct evidence of long-term ozone recovery .
Changes in greenhouse gas concentration at Mawson and other Antarctic stations are similar to those recorded at mid–high-latitude southern hemisphere regions such as Cape Grim, Tasmania. Data from Antarctic stations show an increase in the levels of carbon dioxide, nitrous oxide and methane through the industrial period; variations in the growth rates, particularly interannual changes in the direct record; and decadal changes in the lower resolution ice record.
There has been long-term growth in atmospheric carbon dioxide levels; the highs and lows in atmospheric carbon dioxide levels are correlated with El Niño events. There has been long-term growth in atmospheric methane levels, although the rate of increase has declined over the past 20 years. Atmospheric nitrous oxide levels have increased.
Aside from the implication for climate change, increased atmospheric concentrations of greenhouse gases are believed to be related to observed changes in marine organisms and seawater acidity. Laboratory experiments have indicated that elevated carbon dioxide levels produces changes in the thickness and morphology of calcium carbonate tests of a key planktonic organism (Emiliania huxleyi) as a result of increasing the acidity of seawater (Cubillos 2005). Ocean acidification is also expected to have a major impact on populations of pteropods (small planktonic snails), an important prey in the Southern Ocean, and also plays an important role in the uptake of carbon dioxide (Royal Society 2005).
If increasing acidity decreases the thickness and integrity of the skeletal structures of plankton organisms that are key to the marine food web and are also responsible for the removal of carbon dioxide from the marine environment of the Southern Ocean, then the loss of these species will not only have a significant impact on the food chain but will also reduce the ocean’s ability to absorb carbon dioxide from the atmosphere (Royal Society 2005).
No significant changes are reported in overall Antarctic sea-ice extent through the period of satellite data (that is, since the 1970s), although significant regional changes are evident, particularly around the western Antarctic Peninsula. It appears from historical and palaeo-climate data that decreases have occurred before the satellite period (de la Mare 1997).
Total water storage in the Antarctic ice sheet appears to have increased over the period 1992–2003, reducing potential sea-level rise over the same period by around 0.1 mm per year (out of the overall increase in sea level of 1.8 mm per year). West Antarctica is showing overall mass loss and east Antarctica mass gain. These changes are consistent with modelled expectations in a warming climate.
There have been significant changes in dates of maximum fast-ice thickness and fast-ice breakout at Davis and Mawson stations.
Heard Island’s glaciers, in particular Brown Glacier, are all in retreat, consistent with warming in the southern Indian Ocean.
Antarctic sea-ice is a potentially sensitive indicator of climate change. Observations of the overall extent of Antarctic sea-ice show large interannual variability but show no statistically significant trend through the period of satellite observations since the 1970s (Parkinson 2004). Considerable regional variability is apparent, however, with a statistically significant decrease in the west Antarctic Peninsula region but an increase in the Ross Sea sector (Zwally et al 2002). Work is currently underway to relate interannual through decadal variability in sea-ice extent to large-scale modes of atmospheric circulation.
Despite the lack of a clear trend in the extent of Antarctic sea-ice in the record from 30 years of satellite observations, there are suggestions of a decline in its extent over the longer term. Australian research has found a link between sea-ice extent and biochemical markers in ice cores from the continent. This provides a longer record that suggests a reduction of about 20 per cent in sea-ice in the 80°E–140°E longitude sector over the latter half of the 20th century (Curran et al 2003). A similar decline has been inferred from historical whaling records (de la Mare 1997).
Antarctic ice sheet mass balance is the net increase or decrease in total stored water in the polar ice cap: a globally important parameter, but difficult to quantify on a continental scale. The ice sheet contains an estimated 75 per cent of the world’s fresh water—equivalent to a sea-level rise of 61 metres. While there is no suggestion that the entire ice sheet will melt, even a small imbalance between the rate at which water is deposited on the ice sheet as snow and the rate of loss through melting and ice discharge has significant consequences for sea-level change.
Recent Australian research has focused on the Lambert Glacier Basin, one of the major east Antarctic drainage basins, and the Amery Ice Shelf, into which it discharges. This work has provided estimates of the total ice discharge and the net basal melt beneath the ice shelf (Fricker et al 2000).
Different parts of the continent appear to respond differently to the warming of global climate. Snow accumulation is measured by analysis of ice cores and by atmospheric modelling, but there is no single agreed measurement technique or definitive time series for overall mass balance changes.
Consistent with expectations for a warming climate, observations in East Antarctica suggest an increase in ice mass due to increased snow accumulation, while the Antarctic Peninsula and West Antarctica show acceleration in the ice flow from several major outlet glaciers, leading to a thinning and overall loss of ice mass in that region.
However, significant uncertainty surrounds ice losses due to basal melting at the glacier–ocean interface, particularly in response to warming. Increases of two metres per year of basal melting under some regions of floating ice shelves have been suggested as realistic responses to ocean warming in recent decades (Rignot and Jacobs 2002). The loss of ice shelves in some regions has resulted in the accelerated ice discharge from inland glaciers (Rignot et al 2004).
The present estimates for total Antarctic mass balance are:
- mass increasing at 33±8 Gt/a over the period 1992–2003 (Alley et al 2005, Davis et al 2005)
- equivalent sea-level change—increase in ice-sheet volume corresponds to a decrease in sea-level of approximately 0.1 mm per year (1992–2003), which is an offset of about 5 per cent against the rate of global sea-level rise over the same period (1.8 mm per year).
Sea-ice forms on coastal waters near the Australian Antarctic stations in March- and April, and at Mawson and Davis, after perhaps an early break-out when the ice is still thin, it grows in situ and remains land-fast until the next summer. The maximum thickness and extent of the fast-ice is usually reached in about October at Mawson, and in November at Davis. At Mawson, where the near-coastal water depth can be well over 200 metres (for example about 1400 metres in the Neilsen Basin), the ice growth is slowed by heat supplied from the ocean, and interannual variability in ice thickness is partly due to changes in this heat supply (Heil et al 1996). At Davis, the ocean is much shallower, and changes in fast-ice reflect changes in atmospheric forcing. Growth rate, persistence, extent and maximum thickness of fast-ice are related to the atmospheric and oceanic climate at the particular region or site.
At Mawson and Davis fast-ice observations have been made since the 1950s, sometimes intermittently. Annual maximum ice thickness does not show any long-term trend at either location; however, the interannual variability in thickness increased significantly at Mawson in the 1980s and at Davis in the 1990s. At Mawson there is a trend in the date of annual maximum ice thickness of +4.8 days per decade, with maxima occurring later in the 1990s than in the 1950s (Heil and Allison 2002).
At Davis the dates of annual maximum ice thickness and ﬁnal fast-ice breakout are both delayed (each by +4.3 days per decade), and the delayed ice breakout contributes to a prolonged persistence of the Davis fast ice (+6.7 days per decade). Surface air temperature measurements at Davis from 1969–2003 show a cooling trend for summer and autumn, but a warming in winter and in spring. The tendency towards later dates of annual maximum ice thickness is correlated with winter and spring warming (Heil in press).
All glaciers on Heard Island have retreated in extent since 1947. The total land area covered by glaciers has decreased from 288 km² in 1947 to 253 km² in 2000, that is, by 12 per cent (Ruddell 2006). The retreating ice has exposed 35 km² of new terrain, including several large lagoons, representing nearly 10 per cent of the total area of the island.
Brown Glacier showed surface elevation decreases of 5.9 to 11.7 metres between 2000 and 2003 (Thost and Truffer submitted). The glacier has decreased in area by about 35 per cent and thinned by 30.3 m, or around 0.5 m per year, in the period 1947–2004 (Thost and Truffer submitted). Ice loss in this period was more than twice the 1947–2004 average.
Changes in Heard Island glaciers correlate with underlying climate change in the southern Indian Ocean—temperature observations in the area show a warming of 0.9 °C over the same period (Thost and Allison 2005).