The physical environment includes both the nonliving factors that characterise an ecosystem (e.g. weather patterns, ice coverage, the atmosphere) and the processes that drive them (e.g. weathering of rocks, anthropogenic emissions that deplete ozone in the atmosphere).
The physical environment: The cryosphere
The physical environment: The cryosphere
The cryosphere—Antarctic ice sheet and glaciers
The Antarctic ice sheet consists of 3 topographically different regions:
- the Antarctic Peninsula, which reaches further north than any other area in Antarctica
- the West Antarctic Ice Sheet
- the East Antarctic Ice Sheet, which is by far the largest component, extending from about 30°W to about 165°E.
The ice mass budget of the Antarctic continental ice sheet is the balance between mass gained from snowfall, and mass lost by melt from the ice shelves and formation (discharge) of icebergs at the coast. The net mass balance is complex to assess, because changes in snowfall and iceberg discharge vary by region. Abrupt changes have been observed in some coastal regions, including the rapid disintegration of floating ice shelves. This has raised questions about the potential for rapid ice discharge from Antarctica into the sea. Changes in global sea levels and in the freshwater input to the Southern Ocean are the major environmental consequences of changes in the Antarctic ice sheet. There are also possible flow-on effects on global ocean circulation and marine ecosystems.
Methods for measuring ice mass changes for Antarctica fall into 3 main categories:
- The mass budget method uses measured snowfall (input), combined with losses across the periphery (from measured velocity and thickness—output), to compute gains or losses over time.
- A second method monitors surface elevation changes to determine losses (lowering elevation) or gains (rising elevation) and infer mass changes.
- A third method uses satellite measurements of gravitational pull as the instruments pass over the ice to ‘weigh’ the ice sheet directly.
Each method has advantages and disadvantages, and relies on different data sources. Consequently, the magnitudes of estimates vary; however, most studies now broadly agree within their error estimates.
The Fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change (Vaughan et al. 2013) provides a synthesis of several studies and methods. For Antarctica, the loss of mass totalled around 2 gigatonnes (Gt) from 1991 to the end of 2011 (Figure ANT8). It is worthwhile noting that 360 Gt of ice mass converts to around 1 millimetre of sea level rise.
The average loss of ice in 2002–11 was 147 (72–221) Gt per year (Gt/y). This is an increase from a loss of 30 Gt/y in 1992–2001. The average loss during the 2 decades was 88 ± 35 Gt/y, which is consistent with an independent assessment combining several detection methods that found a loss of 71 ± 53 Gt/y during the same period (Shepherd et al. 2012). Shepherd et al. (2012) give a regional separation showing:
- West Antarctic loss of 65 ± 26 Gt/y
- Antarctic Peninsula loss of 20 ± 14 Gt/y
- East Antarctica gain (within errors) of 14 ± 43 Gt/y.
Studies since IPCC AR5 (McMillan et al. 2014, Harig & Simons 2015) give varied results from a range of methods, underscoring considerable remaining uncertainties, including those driven by the impact of interannual variability on studies with limited temporal scope (Gorodetskaya et al. 2014, Paolo et al. 2015).
The broad picture remains one of overall ice loss from Antarctica, dominated by the West Antarctic Ice Sheet and the Antarctic Peninsula, with accelerating losses likely regionally in West Antarctica (Sutterley et al. 2014, Harig & Simons 2015). The situation for the East Antarctic Ice Sheet is more ambiguous, ranging from close to balance to potentially gaining considerable net mass (King et al. 2012, Harig & Simons 2015). However, even mass gains are not uniform, and some parts of the sheet are likely to lose mass.
Change is also seen in the ice shelves that fringe the continent. Ice shelves consist of floating ice where the continental ice discharges to the ocean. More than 80 per cent of the continental ice drains through such floating ice (Pritchard et al. 2012), which impedes the flow of ice discharge from the grounded ice behind. Although removal of floating ice has no direct impact on sea level, the removal of ice shelves allows for accelerated discharge from the continent, with a consequent impact on sea level.
Several studies analysing satellite measurements of the surface height of ice shelves have established that a large percentage of Antarctica’s ice shelves are thinning (Pritchard et al. 2012), with large interannual variability (Paolo et al. 2015). The total Antarctic ice-shelf volume for 2003 to the end of 2011 decreased on average by 310 ± 74 cubic kilometres per year. This compares with just 25 ± 64 cubic kilometres per year for 1994–2002. The dominant driver of ice-shelf thinning in most cases is believed to be increased ocean melting (Pritchard et al. 2012, Dutrieux et al. 2014). However, atmospheric processes may also be important, particularly for lower-latitude ice shelves in rapidly warming regions such as the Antarctic Peninsula (Cook & Vaughan 2010).
The largest ice-shelf thinning is currently seen in West Antarctica, with more modest changes in East Antarctica. For 1994–2002, the volume of the East Antarctic ice shelf increased by 148 ± 45 cubic kilometres per year. This reversed for 2003–11, with a volume loss of 56 ± 37 cubic kilometres per year. West Antarctica showed accelerating loss for the 18 years to the end of 2011 (Paolo et al. 2015).
In some regions, ice-shelf thinning and loss carry additional significance for future mass loss because of potential instability of the ice sheet. In areas where ice is grounded below sea level on a bed that deepens inland, initial ice retreat, once triggered, can lead to accelerated discharge and further retreat, which is irreversible (Schoof 2007). Some studies indicate that this process is already under way in regions of West Antarctica (Rignot et al. 2014), and reflects the connection between ice-shelf and ice-sheet losses in West Antarctica.
Although this process is yet to be seen in East Antarctica, recent work mapping the bed beneath the East Antarctic Ice Sheet reveals extensive areas grounded below sea level that may be similarly vulnerable to loss (Roberts et al. 2011, Young et al. 2011, Fretwell et al. 2013). In East Antarctica, 2 regions have been identified where future large-scale retreat could occur. The Wilkes Subglacial Basin and the Aurora Subglacial Basin each cover extensive areas of ice grounded more than a kilometre below sea level. Ice thinning and loss in the margins of these regions have the potential to lead to large-scale retreat on centennial timescales (Golledge et al. 2015, Pollard et al. 2015).
The Totten Glacier ice shelf and Moscow University ice shelf are of interest because they drain a large region of the interior (the largest by volume in East Antarctica), which includes large areas grounded below sea level that could be subject to unstable retreat (Greenbaum et al. 2015). The changes to the East Antarctic ice shelf noted above (Paolo et al. 2015) are modest, and do not yet indicate that processes of loss seen in West Antarctica are under way. Studies indicate that large interannual and decadal variability in the Totten Glacier ice shelf is a response to increases in ocean temperature (Gwyther et al. 2014). Further detailed investigation of the bed beneath Totten Glacier is required to establish potential future response and vulnerability, but recent work suggests that large-scale ice retreat in East Antarctica has occurred in the past (Mackintosh et al. 2014).
The influences of climate change on Antarctica are also illustrated by events in the Antarctic Peninsula region. The Antarctic Peninsula has experienced one of the highest regional temperature increases on the planet (2.8 °C in 50 years). Several floating ice shelves in that region have recently collapsed abruptly—for example, the Larsen B ice shelf collapsed in March 2002, and the Wilkins ice shelf started to disintegrate in March 2008 (Steig et al. 2009, Humbert et al. 2010). By 2009, the Antarctic Peninsula had lost about 28,100 km2 from the 152,200 km2 of ice shelves present in the 1950s (Cook & Vaughan 2010). With the buttressing effect of grounded ice shelves gone, glaciers adjacent to the collapsing ice shelves now flow around 3 to 4 times faster into the ocean (Scambos et al. 2003, Rignot 2008). This increase in the discharge of grounded ice from the ice sheet to the ocean is contributing to sea level rise.
Glaciers on Heard Island are continuing to retreat. There have been 5 complete aerial inventories of Heard Island glaciers since 1947. In the 1940s, the island had a total glaciated area of 288 km2 (Ruddell 2006). This decreased to 256 km2 in the late 1980s (Ruddell 2006)—an 11 per cent loss. By the late 2000s, the glaciated area had decreased further to 231 km2 (Harris 2009, Lucieer et al. 2009).
Physiographic and orographic effects on Heard Island have resulted in the glaciers on the leeward (eastern) and windward (western) side reacting differently to changes in the climate. The most recent surveys indicate that the glaciers on the leeward side of the island continue to retreat at a more rapid rate than those on the windward side. Brown Glacier, on the eastern side of the island, decreased from approximately 6.2 km2 in 1947 to 4.4 km2 in 2004—a 29 per cent loss in area (Thost & Truffer 2008). Recent studies have shown that the area of Brown Glacier decreased to 3.6 km2 in 2008 (Harris 2009, Lucieer et al. 2009), and had decreased even more by 2014 (Donoghue 2016). The nearby Stephenson Glacier has also shown recent ice loss (Box ANT4).
The cryosphere—sea ice
Sea ice is the frozen surface of the ocean. It covers, on average, approximately 3 million km2 of the Southern Ocean each summer and about 18.5 million km2 each winter (or 0.8–5.2 per cent of the global ocean’s surface area) (Comiso 2010). This cover forms a crucially important component of the global cryosphere and climate system. Sea ice and its snow cover insulate the ocean from heat loss to the atmosphere and significantly raise the ocean surface albedo (reflectivity). Thus, incoming solar radiation is effectively reflected. Sea ice also provides a barrier to the exchange of momentum and gases, such as CO2 and water vapour, between the ocean and the atmosphere. Moreover, brine (salty) water rejected by growing sea ice modifies the ocean density structure, which plays a key role in driving global ocean circulation (see Global importance of Antarctica).
Sea ice also dominates the seasonal physical and chemical dynamics of the high-latitude Southern Ocean, and plays a crucial role in marine ecosystem structure and function. Plants and animals at all trophic levels are highly dependent on sea ice for a variety of reasons. For example, in shallow waters, sea ice controls the amount of light that is available to photosynthesising organisms (Clark et al. 2015a), while marine mammals and seabirds use it as a habitat (Ainley et al. 2003). The pulse of fresh water into the ocean from its seasonal melt is a major driver of intense algal blooms around the high-latitude Southern Ocean each late spring to summer.
Any substantial changes in sea ice coverage and properties therefore have wide-ranging and lasting climatic, meteorological, physical, ecological and human impacts.
Given its close association with patterns of atmospheric and oceanic temperature and circulation, sea ice is a sensitive passive indicator of climate change and climate variability. It is also a key modulator of such change and variability through complex feedback processes within the atmosphere, termed the ocean–sea ice interaction system.
Sea ice itself is made up of 2 main components. The most extensive is ‘pack ice’, which is made up of individual pieces called ‘floes’, and is in constant motion in response to winds and ocean currents. The other main component is ‘fast ice’, which forms as a narrow band of stationary sea ice (up to about 150 kilometres wide) that is confined to coastal margins, where it is held in place by grounded icebergs and/or in sheltered embayments.
The ratio of fast ice to overall sea ice changes with season. In summer, the percentage of fast ice can be higher than in winter and the proportion of pack ice is significantly reduced. Variability in the extent of fast ice is higher in summer than in winter (Fraser et al. 2012). Narrow fast ice is of key significance as a stabilising influence on the floating ice sheets in certain locations (Massom et al. 2010) and is an important habitat for many organisms (Massom & Stammerjohn 2010). For Antarctic operations, sea ice can create significant shipping and logistical challenges. For example, sea ice can be an aid or impediment to station resupply in the AAT, depending on its extent, duration and thickness.
Sea ice in the Arctic and Southern oceans has different characteristics, based on differences in the geographical settings and the processes affecting them. During the past few decades, the sea ice in the 2 oceans has displayed dissimilar changes with time. Whereas sea ice in the Arctic Ocean is largely enclosed by land masses, Antarctic sea ice surrounds the continent. It is largely unconstrained and highly dynamic because it is exposed to Southern Ocean wind and waves.
Based on satellite data extending back to 1979, annual sea ice extent in the Arctic decreased by 3.8 ± 0.3 per cent (or 0.48 million km2) per decade from 1979 to 2012 (Vaughan et al. 2013), with particularly rapid loss in summer (about 30 per cent has been lost during the summers since the late 1970s) (Stroeve et al. 2012). This change has wide-ranging climatic and ecological consequences. In contrast and during the same period, there has been a small but significant net increase in the overall Antarctic sea ice extent of 1.5 ± 0.2 per cent (or about 0.17 million km2) per decade (see Box ANT4 for a discussion of recent annual records). Note that this slight net increase is the sum of strong regional differences—that is, substantial decreases in the Bellingshausen–Amundsen seas sector and a larger increase in the Ross Sea, which dominate the overall trend (Comiso et al. 2011, Parkinson & Cavalieri 2012, Hobbs et al. 2016a). Proxy information from ice-core and historical whaling records suggests that Antarctic sea ice coverage declined significantly in the decades before the late 1950s to early 1960s (Curran et al. 2003, Hobbs et al. 2016b).
Contrasting regional patterns of change are also apparent in the seasonality (annual duration) of Antarctic sea ice coverage. In the north-eastern and western Antarctic Peninsula, and southern Bellingshausen Sea region, later ice advance and earlier retreat led to a shortening of annual sea ice duration by 100 ± 31 days (a trend of –3.1 ± 1.0 days per year) from 1979–80 to 2010–11 (Stammerjohn et al. 2012). These changes have had impacts on the marine ecosystem, particularly on phytoplankton communities (Montes-Hugo et al. 2009). The opposite is true in the western Ross Sea, where the ice season has lengthened substantially—by 79 ± 12 days (at 2.5 ± 0.4 days per year) (Stammerjohn et al. 2012).
Across East Antarctica (within the AAT), a small increasing trend in sea ice extent (Cavalieri & Parkinson 2012) is accompanied by patterns of change in sea ice annual duration that are generally of a lower magnitude and zonally complex (Hobbs et al. 2016a). From 1979–80 to 2009–10, the annual sea ice duration changed by ±1–2 days per year in some regions of the AAT (Massom et al. 2013a).
Looking to the near-coastal region, the current satellite-derived timeseries of fast ice extent from 2000 to 2008 (Fraser et al. 2012) is too short to support a statement about long-term trends. However, significant changes have been observed during this short period, with a 4.1 ± 0.4 per cent increase each year in the Indian Ocean sector (20°E to 90°E) and a decrease of 0.4 ± 0.4 per cent each year in the sector from 90°E to 160°E. Greater persistence of more extensive coastal fast ice across the Indian Ocean sector from 2004 (Fraser et al. 2012) had a major impact on logistical operations around the resupply of both Mawson Station and the Japanese Syowa Station.
Based on satellite and submarine records, there is high confidence that the average Arctic winter sea ice thickness decreased between 1980 and 2008 (Stocker et al. 2013). In contrast, there is inadequate information to assess whether large-scale sea ice thickness and volume are changing around Antarctica (Vaughan et al. 2013). IPCC AR5 stated that anthropogenic influences are very likely to have contributed to Arctic sea ice loss since 1979 (Stocker et al. 2013). There is current uncertainty as to whether the smaller overall increase in Antarctic sea ice extent is meaningful as an indicator of climate change, because the extent varies so much from year to year and with location around the continent (Stocker et al. 2013) (see Box ANT5). Moreover:
There is low confidence in the scientific understanding of the small observed increase in Antarctic sea ice extent due to the incomplete and competing scientific explanations for the causes of change and low confidence in estimates of natural internal variability in that region. (Stocker et al. 2013)
Research published in 2012 found that changing wind speed and patterns, and associated sea ice drift accounted for some of the increase in overall sea ice extent in the Antarctic and its regional contrasts (Holland & Kwok 2012). These wind variations are associated with changes in atmospheric pressure patterns around Antarctica, which have been linked to sea surface temperature anomalies in the tropical Pacific Ocean (Yuan & Martinson 2001, Turner et al. 2009) and North Atlantic Ocean (Li et al. 2014). These findings imply that the recent overall slight growth of Antarctic sea ice coverage is consistent with an Earth that is generally warming (King 2014). Other studies also provide evidence that Antarctic sea ice thickness and volume are changing because of climate change, and suggest reasons, such as the:
- changing patterns of ocean circulation and heat content around Antarctica (Martinson 2012)
- decreasing salinity in the upper ocean because of increased ice-shelf basal melt (Bintanja et al. 2013) and snowfall (Liu & Curry 2010)
- feedback mechanisms involving increased upper-ocean heat storage in spring–summer, because it delays subsequent sea ice advance in autumn (Stammerjohn et al. 2012).