→East Antarctic ice sheet: adding ref
|
Merging material from ice sheet dynamics to here, as per talk page. See page history for attribution. Will edit material in place to condense and standartize it.
|
||
Line 15: | Line 15: | ||
== Common properties == |
== Common properties == |
||
[[File:Carbon stores and fluxes in present day ice sheets.webp|thumb|upright=1.7| Carbon stores and fluxes in present-day ice sheets (2019), and the predicted impact on carbon dioxide (where data exists). <br /> Estimated carbon fluxes are measured in Tg C a<sup>−1</sup> (megatonnes of carbon per year) and estimated sizes of carbon stores are measured in Pg C (thousands of megatonnes of carbon). DOC = [[dissolved organic carbon]], POC = [[particulate organic carbon]].<ref name="Wadham2019">Wadham, J.L., Hawkings, J.R., Tarasov, L., Gregoire, L.J., Spencer, R.G.M., Gutjahr, M., Ridgwell, A. and Kohfeld, K.E. (2019) "Ice sheets matter for the global [[carbon cycle]]". ''Nature communications'', '''10'''(1): 1–17. {{doi|10.1038/s41467-019-11394-4}}. [[File:CC-BY_icon.svg|50x50px]] Material was copied from this source, which is available under a [[creativecommons:by/4.0/|Creative Commons Attribution 4.0 International License]].</ref>]] |
[[File:Carbon stores and fluxes in present day ice sheets.webp|thumb|upright=1.7| Carbon stores and fluxes in present-day ice sheets (2019), and the predicted impact on carbon dioxide (where data exists). <br /> Estimated carbon fluxes are measured in Tg C a<sup>−1</sup> (megatonnes of carbon per year) and estimated sizes of carbon stores are measured in Pg C (thousands of megatonnes of carbon). DOC = [[dissolved organic carbon]], POC = [[particulate organic carbon]].<ref name="Wadham2019">Wadham, J.L., Hawkings, J.R., Tarasov, L., Gregoire, L.J., Spencer, R.G.M., Gutjahr, M., Ridgwell, A. and Kohfeld, K.E. (2019) "Ice sheets matter for the global [[carbon cycle]]". ''Nature communications'', '''10'''(1): 1–17. {{doi|10.1038/s41467-019-11394-4}}. [[File:CC-BY_icon.svg|50x50px]] Material was copied from this source, which is available under a [[creativecommons:by/4.0/|Creative Commons Attribution 4.0 International License]].</ref>]] |
||
{{Further|Ice-sheet dynamics}} |
|||
Ice sheets have the following properties: "An ice sheet flows outward from a high central ice plateau with a small average surface slope. The margins usually slope more steeply, and most ice is discharged through fast-flowing ice streams or [[Outlet glacier|outlet glaciers]], often into the sea or into [[Ice shelf|ice shelves]] floating on the sea."<ref name=":2" />{{Rp|page=2234}} |
Ice sheets have the following properties: "An ice sheet flows outward from a high central ice plateau with a small average surface slope. The margins usually slope more steeply, and most ice is discharged through fast-flowing ice streams or [[Outlet glacier|outlet glaciers]], often into the sea or into [[Ice shelf|ice shelves]] floating on the sea."<ref name=":2" />{{Rp|page=2234}} |
||
Line 22: | Line 21: | ||
Until recently, ice sheets were viewed as inert components of the [[Marine carbon cycle|carbon cycle]] and were largely disregarded in global models. Research in the past decade has transformed this view, demonstrating the existence of uniquely adapted [[Marine microorganisms|microbial communities]], high rates of [[Marine biogeochemical cycles|biogeochemical]]/physical weathering in ice sheets and storage and cycling of organic carbon in excess of 100 billion tonnes, as well as nutrients (see diagram).<ref name="Wadham2019" /> |
Until recently, ice sheets were viewed as inert components of the [[Marine carbon cycle|carbon cycle]] and were largely disregarded in global models. Research in the past decade has transformed this view, demonstrating the existence of uniquely adapted [[Marine microorganisms|microbial communities]], high rates of [[Marine biogeochemical cycles|biogeochemical]]/physical weathering in ice sheets and storage and cycling of organic carbon in excess of 100 billion tonnes, as well as nutrients (see diagram).<ref name="Wadham2019" /> |
||
[[Image:Antarctica glacier flow rate.jpg|300 px|thumb|Glacial flow rate in the Antarctic ice sheet.]] |
|||
[[File:Flow of Ice Across Antarctica.ogv|thumb|upright=1.35|The motion of ice in Antarctica]] |
|||
== Dynamics == |
|||
[[File:West Antarctic Glacier Ice Flows and Elevation Change.ogv|thumb|Animation showing glacier changes.]] |
|||
[[File:Mass Balance Change over India from GRACE.ogv|thumb|This animation shows the average yearly change in mass, in cm of water, during 2003–2010, over the Indian subcontinent. The yellow circles mark locations of glaciers. There is significant mass loss in this region (denoted by the blue and purple colors), but it is concentrated over the plains south of the glaciers, and is caused by [[Overdrafting|groundwater depletion]]. A color-bar overlay shows the range of values displayed.]] |
|||
The motion of ice sheets is driven by [[gravity]] but is controlled by temperature and the strength of individual glacier bases. A number of processes alter these two factors, resulting in cyclic surges of activity interspersed with longer periods of inactivity, on time scales ranging from hourly to the [[wikt:centennial|centennial]]. |
|||
===Boundary conditions=== |
|||
The interface between an ice stream and the ocean is a significant control of the rate of flow. |
|||
[[Image:Larsen B collapse.jpg|thumb|upright=1.2|The collapse of the [[Larsen B]] ice shelf had profound effects on the velocities of its feeder glaciers.]] |
|||
[[Ice shelves]] are thick layers of ice floating on the sea – can stabilise the glaciers that feed them. These tend to have accumulation on their tops, may experience melting on their bases, and [[Ice calving|calve]] icebergs at their periphery. The catastrophic collapse of the [[Larsen B]] ice shelf in the space of three weeks during February 2002 yielded some unexpected observations. The glaciers that had fed the ice sheet ([[Crane Glacier|Crane]], [[Jorum Glacier|Jorum]], [[Green Glacier|Green]], [[Hektoria Glacier|Hektoria]] – see image) increased substantially in velocity. This cannot have been due to seasonal variability, as glaciers flowing into the remnants of the ice shelf (Flask, Leppard) did not accelerate.<ref name=Scambos2004>{{cite journal |last1=Scambos |first1=T. A. |title=Glacier acceleration and thinning after ice shelf collapse in the Larsen B embayment, Antarctica |journal=Geophysical Research Letters |date=2004 |volume=31 |issue=18 |pages=L18402 |doi=10.1029/2004GL020670 |bibcode=2004GeoRL..3118402S |s2cid=36917564 |doi-access=free |hdl=11603/24296 |hdl-access=free }}</ref> |
|||
Ice shelves exert a dominant control in Antarctica, but are less important in Greenland, where the ice sheet meets the sea in [[fjord]]s. Here, melting is the dominant ice removal process,<ref name=IPCCc4/> resulting in predominant mass loss occurring towards the edges of the ice sheet, where icebergs are calved in the fjords and surface meltwater runs into the ocean. |
|||
'''Tidal effects''' are also important; the influence of a 1 m tidal oscillation can be felt as much as 100 km from the sea.<ref name=Clarke2005/> On an hour-to-hour basis, surges of ice motion can be modulated by tidal activity. During larger [[spring tide]]s, an ice stream will remain almost stationary for hours at a time, before a surge of around a foot in under an hour, just after the peak high tide; a stationary period then takes hold until another surge towards the middle or end of the falling tide.<ref name=Bindschalder2003>{{cite journal |last1=Bindschadler |first1=Robert A. |last2=King |first2=Matt A. |last3=Alley |first3=Richard B. |last4=Anandakrishnan |first4=Sridhar |last5=Padman |first5=Laurence |title=Tidally Controlled Stick-Slip Discharge of a West Antarctic Ice |journal=Science |date=22 August 2003 |volume=301 |issue=5636 |pages=1087–1089 |doi=10.1126/science.1087231 |pmid=12934005 |s2cid=37375591 |url=https://zenodo.org/record/1230832 }}</ref><ref name=Anandakrishnan2003>{{cite journal |last1=Anandakrishnan |first1=S. |last2=Voigt |first2=D. E. |last3=Alley |first3=R. B. |last4=King |first4=M. A. |title=Ice stream D flow speed is strongly modulated by the tide beneath the Ross Ice Shelf |journal=Geophysical Research Letters |date=April 2003 |volume=30 |issue=7 |page=1361 |doi=10.1029/2002GL016329 |bibcode=2003GeoRL..30.1361A |s2cid=53347069 |doi-access=free }}</ref> At neap tides, this interaction is less pronounced, without tides surges would occur more randomly, approximately every 12 hours.<ref name=Bindschalder2003/> |
|||
Ice shelves are also sensitive to basal melting. In Antarctica, this is driven by heat fed to the shelf by the [[circumpolar deep water]] current, which is 3 °C above the ice's melting point.<ref name=Walker2007>{{cite journal |last1=Walker |first1=Dziga P. |last2=Brandon |first2=Mark A. |last3=Jenkins |first3=Adrian |last4=Allen |first4=John T. |last5=Dowdeswell |first5=Julian A. |last6=Evans |first6=Jeff |title=Oceanic heat transport onto the Amundsen Sea shelf through a submarine glacial trough |journal=Geophysical Research Letters |date=16 January 2007 |volume=34 |issue=2 |pages=L02602 |doi=10.1029/2006GL028154 |bibcode=2007GeoRL..34.2602W |s2cid=30646727 |url=http://nora.nerc.ac.uk/id/eprint/1199/1/grl22452.pdf }}</ref> |
|||
As well as heat, the sea can also exchange salt with the oceans. The effect of latent heat, resulting from melting of ice or freezing of sea water, also has a role to play. The effects of these, and variability in snowfall and base sea level combined, account for around 80 mm annual variability in ice shelf thickness. |
|||
===Long-term changes=== |
|||
Over long time scales, ice sheet mass balance is governed by the amount of sunlight reaching the Earth. This variation in sunlight reaching the Earth, or [[insolation]], over geologic time is in turn determined by the angle of the Earth to the Sun and shape of the Earth's orbit, as it is pulled on by neighboring planets; these variations occur in predictable patterns called [[Milankovitch cycles]]. Milankovitch cycles dominate climate on the glacial–interglacial timescale, but there exist variations in ice sheet extent that are not linked directly with insolation. |
|||
For instance, during at least the last 100,000 years, portions of the ice sheet covering much of North America, the [[Laurentide Ice Sheet]] broke apart sending large flotillas of icebergs into the North Atlantic. When these icebergs melted they dropped the boulders and other continental rocks they carried, leaving layers known as [[ice rafted debris]]. These so-called [[Heinrich events]], named after their discoverer [[Hartmut Heinrich]], appear to have a 7,000–10,000-year [[Periodic function|periodicity]], and occur during cold periods within the last interglacial.<ref>{{cite journal |last1=Heinrich |first1=Hartmut |title=Origin and Consequences of Cyclic Ice Rafting in the Northeast Atlantic Ocean During the Past 130,000 Years |journal=Quaternary Research |date=March 1988 |volume=29 |issue=2 |pages=142–152 |doi=10.1016/0033-5894(88)90057-9 |bibcode=1988QuRes..29..142H |s2cid=129842509 }}</ref> |
|||
Internal ice sheet "binge-purge" cycles may be responsible for the observed effects, where the ice builds to unstable levels, then a portion of the ice sheet collapses. External factors might also play a role in forcing ice sheets. [[Dansgaard–Oeschger event]]s are abrupt warmings of the northern hemisphere occurring over the space of perhaps 40 years. While these D–O events occur directly after each Heinrich event, they also occur more frequently – around every 1500 years; from this evidence, paleoclimatologists surmise that the same forcings may drive both Heinrich and D–O events.<ref>{{cite book |doi=10.1029/GM112p0035 |chapter=The North Atlantic's 1–2 kyr climate rhythm: Relation to Heinrich events, Dansgaard/Oeschger cycles and the Little Ice Age |title=Mechanisms of Global Climate Change at Millennial Time Scales |series=Geophysical Monograph Series |year=1999 |last1=Bond |first1=Gerard C. |last2=Showers |first2=William |last3=Elliot |first3=Mary |last4=Evans |first4=Michael |last5=Lotti |first5=Rusty |last6=Hajdas |first6=Irka |last7=Bonani |first7=Georges |last8=Johnson |first8=Sigfus |volume=112 |pages=35–58 |isbn=978-0-87590-095-7 }}</ref> |
|||
'''Hemispheric asynchrony in ice sheet behavior''' has been observed by linking short-term spikes of methane in Greenland ice cores and Antarctic ice cores. During [[Dansgaard–Oeschger event]]s, the northern hemisphere warmed considerably, dramatically increasing the release of methane from wetlands, that were otherwise tundra during glacial times. This methane quickly distributes evenly across the globe, becoming incorporated in Antarctic and Greenland ice. With this tie, paleoclimatologists have been able to say that the ice sheets on Greenland only began to warm after the Antarctic ice sheet had been warming for several thousand years. Why this pattern occurs is still open for debate.<ref>{{cite journal |last1=Turney |first1=Chris S. M. |last2=Fogwill |first2=Christopher J. |last3=Golledge |first3=Nicholas R. |last4=McKay |first4=Nicholas P. |last5=Sebille |first5=Erik van |last6=Jones |first6=Richard T. |last7=Etheridge |first7=David |last8=Rubino |first8=Mauro |last9=Thornton |first9=David P. |last10=Davies |first10=Siwan M. |last11=Ramsey |first11=Christopher Bronk |last12=Thomas |first12=Zoë A. |last13=Bird |first13=Michael I. |last14=Munksgaard |first14=Niels C. |last15=Kohno |first15=Mika |last16=Woodward |first16=John |last17=Winter |first17=Kate |last18=Weyrich |first18=Laura S. |last19=Rootes |first19=Camilla M. |last20=Millman |first20=Helen |last21=Albert |first21=Paul G. |last22=Rivera |first22=Andres |last23=Ommen |first23=Tas van |last24=Curran |first24=Mark |last25=Moy |first25=Andrew |last26=Rahmstorf |first26=Stefan |last27=Kawamura |first27=Kenji |last28=Hillenbrand |first28=Claus-Dieter |last29=Weber |first29=Michael E. |last30=Manning |first30=Christina J. |last31=Young |first31=Jennifer |last32=Cooper |first32=Alan |title=Early Last Interglacial ocean warming drove substantial ice mass loss from Antarctica |journal=Proceedings of the National Academy of Sciences |date=25 February 2020 |volume=117 |issue=8 |pages=3996–4006 |doi=10.1073/pnas.1902469117 |pmid=32047039 |pmc=7049167 |bibcode=2020PNAS..117.3996T |doi-access=free }}</ref><ref>{{cite journal |last1=Crémière |first1=Antoine |last2=Lepland |first2=Aivo |last3=Chand |first3=Shyam |last4=Sahy |first4=Diana |last5=Condon |first5=Daniel J. |last6=Noble |first6=Stephen R. |last7=Martma |first7=Tõnu |last8=Thorsnes |first8=Terje |last9=Sauer |first9=Simone |last10=Brunstad |first10=Harald |title=Timescales of methane seepage on the Norwegian margin following collapse of the Scandinavian Ice Sheet |journal=Nature Communications |date=11 May 2016 |volume=7 |issue=1 |pages=11509 |doi=10.1038/ncomms11509 |pmid=27167635 |pmc=4865861 |bibcode=2016NatCo...711509C }}</ref> |
|||
===Glacier flows=== |
|||
{{Redirect|Ice flow|floating ice|Ice floe}} |
|||
[[File:Aerial Photo of Monte Rosa Massif - Wallis - Switzerland (cropped).jpg|thumb|[[Aerial photograph]] of the [[Gorner Glacier]] (l.) and the [[Grenzgletscher]] (r.) flowing (in the image downwards) around the [[Monte Rosa]] massif (middle) in the Swiss [[Alps]]]] |
|||
[[Image:Stress-strain1.svg|thumb|upright=1.2|The stress–strain relationship of plastic flow (teal section): a small increase in stress creates an exponentially greater increase in strain, which equates to deformation speed.]] |
|||
The main cause of flow within glaciers can be attributed to an increase in the surface slope, brought upon by an imbalance between the amounts of accumulation vs. [[ablation]]. This imbalance increases the [[shear stress]] on a glacier until it begins to flow. The flow velocity and deformation will increase as the equilibrium line between these two processes is approached, but are also affected by the slope of the ice, the ice thickness and temperature.<ref name="Easterbrook">Easterbrook, Don J., Surface Processes and Landforms, 2nd Edition, Prentice-Hall Inc., 1999{{page needed|date=February 2014}}</ref><ref name="GreveBlatter2009">{{cite book|author1=Greve, R. |author2=Blatter, H. |year=2009|title=Dynamics of Ice Sheets and Glaciers|publisher=Springer|doi=10.1007/978-3-642-03415-2|isbn=978-3-642-03414-5}}{{pn|date=October 2021}}</ref> |
|||
When the amount of strain (deformation) is proportional to the stress being applied, ice will act as an elastic solid. Ice will not flow until it has reached a thickness of 30 meters (98 ft), but after 50 meters (164 ft), small amounts of stress can result in a large amount of strain, causing the deformation to become a [[Plasticity (physics)|plastic flow]] rather than elastic. At this point the glacier will begin to deform under its own weight and flow across the landscape. According to the [[Glen–Nye flow law]], the relationship between stress and strain, and thus the rate of internal flow, can be modeled as follows:<ref name="Easterbrook" /><ref name="GreveBlatter2009" /> |
|||
:<math> |
|||
\Sigma = k \tau^n,\, |
|||
</math> |
|||
where: |
|||
:<math>\Sigma\,</math> = shear strain (flow) rate |
|||
:<math>\tau\,</math> = stress |
|||
:<math>n\,</math> = a constant between 2–4 (typically 3 for most glaciers) that increases with lower temperature |
|||
:<math>k\,</math> = a temperature-dependent constant |
|||
The lowest velocities are near the base of the glacier and along valley sides where friction acts against flow, causing the most deformation. Velocity increases inward toward the center line and upward, as the amount of deformation decreases. The highest flow velocities are found at the surface, representing the sum of the velocities of all the layers below.<ref name="Easterbrook" /><ref name="GreveBlatter2009" /> |
|||
Glaciers may also move by [[basal sliding]], where the base of the glacier is lubricated by meltwater, allowing the glacier to slide over the terrain on which it sits. Meltwater may be produced by pressure-induced melting, friction or geothermal heat. The more variable the amount of melting at surface of the glacier, the faster the ice will flow.<ref name="Schoof2010">{{Cite journal | last1 = Schoof | first1 = C. | title = Ice-sheet acceleration driven by melt supply variability | journal = Nature | volume = 468 | pages = 803–806 | year = 2010 | pmid = 21150994 | doi = 10.1038/nature09618|bibcode = 2010Natur.468..803S | issue=7325| s2cid = 4353234 }}</ref> |
|||
The top 50 meters of the glacier form the fracture zone, where ice moves as a single unit. Cracks form as the glacier moves over irregular terrain, which may penetrate the full depth of the fracture zone. |
|||
===Subglacial processes=== |
|||
[[File:Glacier cross-section.jpg|thumb|upright|A cross-section through a glacier. The base of the glacier is more transparent as a result of melting.]] |
|||
Most of the important processes controlling glacial motion occur in the ice-bed contact—even though it is only a few meters thick.<ref name=Clarke2005>{{cite journal |
|||
| author = Clarke, G. K. C. |
|||
| title = Subglacial processes |
|||
| journal = Annual Review of Earth and Planetary Sciences |
|||
| volume = 33 |
|||
| issue = 1 |
|||
| pages = 247–276 |
|||
| year = 2005 |
|||
| doi = 10.1146/annurev.earth.33.092203.122621 |
|||
| bibcode = 2005AREPS..33..247C |
|||
}}</ref> Glaciers will move by sliding when the basal shear stress drops below the shear resulting from the glacier's weight.{{Clarify|date=February 2009}} |
|||
:τ<sub>D</sub> = ρgh sin α |
|||
:where τ<sub>D</sub> is the driving stress, and α the ice surface slope in radians.<ref name=Clarke2005/> |
|||
:τ<sub>B</sub> is the basal shear stress, a function of bed temperature and softness.<ref name=Clarke2005/> |
|||
:τ<sub>F</sub>, the shear stress, is the lower of τ<sub>B</sub> and τ<sub>D</sub>. It controls the rate of plastic flow, as per the figure (inset, right). |
|||
For a given glacier, the two variables are τ<sub>D</sub>, which varies with h, the depth of the glacier, and τ<sub>B</sub>, the basal shear stress.{{Clarify|date=February 2009}} |
|||
====Basal shear stress==== |
|||
The basal shear stress is a function of three factors: the bed's temperature, roughness and softness.<ref name=Clarke2005/> |
|||
Whether a bed is hard or soft depends on the porosity and pore pressure; higher porosity decreases the sediment strength (thus increases the shear stress τ<sub>B</sub>).<ref name=Clarke2005/> If the sediment strength falls far below τ<sub>D</sub>, movement of the glacier will be accommodated by motion in the sediments, as opposed to sliding. |
|||
'''Porosity''' may vary through a range of methods. |
|||
*Movement of the overlying glacier may cause the bed to undergo [[wikt:dilatancy|dilatancy]]; the resulting shape change reorganises blocks. This reorganises closely packed blocks (a little like neatly folded, tightly packed clothes in a suitcase) into a messy jumble (just as clothes never fit back in when thrown in <!--this sentence only makes sense if the word "in" is repeated--> in a disordered fashion). This increases the porosity. Unless water is added, this will necessarily reduce the pore pressure (as the pore fluids have more space to occupy).<ref name=Clarke2005/> |
|||
*Pressure may cause compaction and consolidation of underlying sediments.<ref name=Clarke2005/> Since water is relatively incompressible, this is easier when the pore space is filled with vapour; any water must be removed to permit compression. In soils, this is an irreversible process.<ref name=Clarke2005/> |
|||
*Sediment degradation by abrasion and fracture decreases the size of particles, which tends to decrease pore space, although the motion of the particles may disorder the sediment, with the opposite effect.<ref name=Clarke2005/> These processes also generate heat, whose importance will be discussed later. |
|||
[[Image:Ice flow controls.jpg|thumb|upright=1.2|Factors controlling the flow of ice]] |
|||
A soft bed, with high porosity and low pore fluid pressure, allows the glacier to move by sediment sliding: the base of the glacier may even remain frozen to the bed, where the underlying sediment slips underneath it like a tube of toothpaste. A hard bed cannot deform in this way; therefore the only way for hard-based glaciers to move is by basal sliding, where meltwater forms between the ice and the bed itself.<ref name=Boulton2006>{{cite book |doi=10.1002/9780470750636.ch2 |chapter=Glaciers and their Coupling with Hydraulic and Sedimentary Processes |title=Glacier Science and Environmental Change |year=2006 |last1=Boulton |first1=Geoffrey S. |pages=2–22 |isbn=978-0-470-75063-6 }}</ref> |
|||
Bed softness may vary in space or time, and changes dramatically from glacier to glacier. An important factor is the underlying |
|||
geology; glacial speeds tend to differ more when they change bedrock than when the gradient changes.<ref name=Boulton2006/> |
|||
As well as affecting the sediment stress, fluid pressure (p<sub>w</sub>) can affect the friction between the glacier and the bed. High fluid pressure provides a buoyancy force upwards on the glacier, reducing the friction at its base. The fluid pressure is compared to the ice overburden pressure, p<sub>i</sub>, given by ρgh. Under fast-flowing ice streams, these two pressures will be approximately equal, with an effective pressure (p<sub>i</sub> – p<sub>w</sub>) of 30 kPa; i.e. all of the weight of the ice is supported by the underlying water, and the glacier is afloat.<ref name=Clarke2005/> |
|||
====Basal melt==== |
|||
A number of factors can affect bed temperature, which is intimately associated with basal meltwater. |
|||
The melting point of water decreases under pressure, meaning that water melts at a lower temperature under thicker glaciers.<ref name=Clarke2005/> This acts as a "double whammy", because thicker glaciers have a lower heat conductance, meaning that the basal temperature is also likely to be higher.<ref name=Boulton2006/> |
|||
Bed temperature tends to vary in a cyclic fashion. A cool bed has a high strength, reducing the speed of the glacier. This increases the rate of accumulation, since newly fallen snow is not transported away. Consequently, the glacier thickens, with three consequences: firstly, the bed is better insulated, allowing greater retention of geothermal heat. Secondly, the increased pressure can facilitate melting. Most importantly, τ<sub>D</sub> is increased. These factors will combine to accelerate the glacier. As friction increases with the square of velocity, faster motion will greatly increase frictional heating, with ensuing melting – which causes a positive feedback, increasing ice speed to a faster flow rate still: west Antarctic glaciers are known to reach velocities of up to a kilometre per year.<ref name=Clarke2005/> |
|||
Eventually, the ice will be surging fast enough that it begins to thin, as accumulation cannot keep up with the transport. This thinning will increase the conductive heat loss, slowing the glacier and causing freezing. This freezing will slow the glacier further, often until it is stationary, whence the cycle can begin again.<ref name=Boulton2006/> |
|||
[[Supraglacial lake]]s represent another possible supply of liquid water to the base of glaciers, so they can play an important role in accelerating glacial motion. |
|||
Lakes of a diameter greater than ~300 m are capable of creating a fluid-filled crevasse to the glacier/bed interface. |
|||
When these crevasses form, the entirety of the lake's (relatively warm) contents can reach the base of the glacier in as little as 2–18 hours – lubricating the bed and causing the glacier to [[surge (glacier)|surge]].<ref name=Krawczynski2007>{{cite conference |last1=Krawczynski |first1=M. J. |last2=Behn |first2=M. D. |last3=Das |first3=S. B. |last4=Joughin |first4=I. |title=Constraints on melt-water flux through the West Greenland ice-sheet: modeling of hydro- fracture drainage of supraglacial lakes |date=1 December 2007 |pages=C41B–0474 |bibcode=2007AGUFM.C41B0474K |url=http://www.agu.org/cgi-bin/wais?jj=C41B-0474 |
|||
|archive-url = https://archive.today/20121228013531/http://www.agu.org/cgi-bin/wais?jj=C41B-0474 |
|||
|url-status = dead |
|||
|archive-date = 2012-12-28 |
|||
|access-date = 2008-03-04 |
|||
|book-title = Eos Trans. AGU |
|||
|volume = 88 |
|||
|issue = 52 |
|||
}}</ref> Water that reaches the bed of a glacier may freeze there, increasing the thickness of the glacier by pushing it up from below.<ref name="Bell2011">{{Cite journal | last1 = Bell | first1 = R. E. | last2 = Ferraccioli | first2 = F. | last3 = Creyts | first3 = T. T. | last4 = Braaten | first4 = D. | last5 = Corr | first5 = H. | last6 = Das | first6 = I. | last7 = Damaske | first7 = D. | last8 = Frearson | first8 = N. | last9 = Jordan | first9 = T. | last10 = Rose | doi = 10.1126/science.1200109 | first10 = K. | last11 = Studinger | first11 = M. | last12 = Wolovick | first12 = M. | title = Widespread Persistent Thickening of the East Antarctic Ice Sheet by Freezing from the Base | journal = Science | volume = 331 | issue = 6024 | pages = 1592–1595 | year = 2011 | pmid = 21385719| bibcode = 2011Sci...331.1592B | s2cid = 45110037 }}</ref> |
|||
Finally, '''bed roughness''' can act to slow glacial motion. The roughness of the bed is a measure of how many boulders and obstacles protrude into the overlying ice. Ice flows around these obstacles by melting under the high pressure on their [[stoss (geography)|stoss side]]; the resultant meltwater is then forced into the cavity arising in their [[lee side]], where it re-freezes.<ref name=Clarke2005/> |
|||
====Pipe and sheet flow==== |
|||
The flow of water under the glacial surface can have a large effect on the motion of the glacier itself. Subglacial lakes contain significant amounts of water, which can move fast: cubic kilometres can be transported between lakes over the course of a couple of years.<ref name=Fricker2007>{{cite journal| first1 = A.| last3 = Bindschadler| first2 = T.| last2 = Scambos| first3 = R.| first4 = L. | title = An Active Subglacial Water System in West Antarctica Mapped from Space| last1 = Fricker | journal = Science| last4 = Padman | volume = 315 | issue = 5818 | pages = 1544–1548 | date=Mar 2007 | issn = 0036-8075| pmid = 17303716 | doi = 10.1126/science.1136897| bibcode = 2007Sci...315.1544F| s2cid = 35995169}}</ref> |
|||
This motion is thought to occur in two main modes: '''pipe flow''' involves liquid water moving through pipe-like conduits, like a sub-glacial river; '''sheet flow''' involves motion of water in a thin layer. A switch between the two flow conditions may be associated with surging behaviour. Indeed, the loss of sub-glacial water supply has been linked with the shut-down of ice movement in the Kamb ice stream.<ref name=Fricker2007/> The subglacial motion of water is expressed in the surface topography of ice sheets, which slump down into vacated subglacial lakes.<ref name=Fricker2007/> |
|||
{{Further|Ice shelf basal channels}} |
|||
===Climate change=== |
|||
{{See also|Effects of climate change#Glaciers and ice sheets decline|Marine ice sheet instability}} |
|||
[[Image:Greenland ice sheet thinning rate.png|thumb|Rates of ice-sheet thinning in Greenland (2003).]] |
|||
The implications of the current climate change on ice sheets are difficult to ascertain. It is clear that increasing temperatures are resulting in reduced ice volumes globally.<ref name=IPCCc4>Sections 4.5 and 4.6 of {{IPCC4/wg1/4}}</ref> (Due to increased precipitation, the mass of parts of the Antarctic ice sheet may currently be increasing, but the total mass balance is unclear.<ref name=IPCCc4/>) |
|||
Rising sea levels will reduce the stability of ice shelves, which have a key role in reducing glacial motion. Some Antarctic ice shelves are currently thinning by tens of metres per year, and the collapse of the Larsen B shelf was preceded by thinning of just 1 metre per year.<ref name=IPCCc4/> Further, increased ocean temperatures of 1 °C may lead to up to 10 metres per year of basal melting.<ref name=IPCCc4/> Ice shelves are always stable under mean annual temperatures of −9 °C, but never stable above −5 °C; this places regional warming of 1.5 °C, as preceded the collapse of Larsen B, in context.<ref name=IPCCc4/> |
|||
[[Image:Geirangerfjord (6-2007).jpg|thumb|upright|Differential erosion enhances relief, as clear in this incredibly steep-sided Norwegian [[fjord]].]] |
|||
Increasing global air temperatures take around 10,000 years to directly propagate through the ice before they influence bed temperatures, but may have an effect through increased surfacal melting, producing more supraglacial lakes, which may feed warm water to glacial bases and facilitate glacial motion.<ref name=IPCCc4/> In areas of increased precipitation, such as Antarctica, the addition of mass will increase rate of glacial motion, hence the turnover in the ice sheet. Observations, while currently limited in scope, do agree with these predictions of an increasing rate of ice loss from both Greenland and Antarctica.<ref name=IPCCc4/> A possible positive feedback may result from shrinking ice caps, in volcanically active Iceland at least. Isostatic rebound may lead to increased volcanic activity, causing basal warming – and, through {{co2}} release, further climate change.<ref>{{cite journal| first1 = C.| first2 = F.| title = Will present day glacier retreat increase volcanic activity? Stress induced by recent glacier retreat and its effect on magmatism at the Vatnajökull ice cap, Iceland| last2 = Sigmundsson | url = http://eprints.whiterose.ac.uk/4094/1/pagli.pdf| journal = Geophysical Research Letters| volume = 35| issue = 9 | pages = L09304| year = 2008| last1 = Pagli| doi = 10.1029/2008GL033510 | bibcode=2008GeoRL..35.9304P| doi-access = free}}</ref> |
|||
Cold meltwater provides cooling of the ocean's surface layer, acting like a lid, and also affecting deeper waters by increasing subsurface [[ocean warming]] and thus facilitating ice melt. |
|||
{{Quote|Our "pure freshwater" experiments show that the low-density lid causes deep-ocean warming, especially at depths of ice shelf grounding lines that provide most of the restraining force limiting ice sheet discharge.<ref name="Hansen2016">{{cite journal|title=Ice melt, sea level rise and superstorms: evidence from paleoclimate data, climate modeling, and modern observations that 2 °C global warming could be dangerous|author1=J. Hansen |author2=M. Sato |author3=P. Hearty |author4=R. Ruedy |author5=M. Kelley |author6=V. Masson-Delmotte |author7=G. Russell |author8=G. Tselioudis |author9=J. Cao |author10=E. Rignot |author11=I. Velicogna |author12=E. Kandiano |author13=K. von Schuckmann |author14=P. Kharecha |author15=A. N. Legrande |author16=M. Bauer |author17=K.-W. Lo |year=2016|journal=Atmospheric Chemistry and Physics|doi=10.5194/acp-16-3761-2016|volume=16|issue=6|pages=3761–3812|arxiv=1602.01393 |bibcode=2016ACP....16.3761H |s2cid=9410444 |doi-access=free }}</ref>}} |
|||
===Erosion=== |
|||
Because ice can flow faster where it is thicker, the rate of glacier-induced erosion is directly proportional to the thickness of overlying ice. Consequently, pre-glacial low hollows will be deepened and pre-existing topography will be amplified by glacial action, while [[nunatak]]s, which protrude above ice sheets, barely erode at all – erosion has been estimated as 5 m per 1.2 million years.<ref name=ngeo2008>{{cite journal | author = Kessler, Mark A.| year = 2008| doi = 10.1038/ngeo201| title = Fjord insertion into continental margins driven by topographic steering of ice| journal = Nature Geoscience | volume = 1 | pages = 365 | last2 = Anderson | first2 = Robert S. | last3 = Briner | first3 = Jason P. | issue=6 | bibcode=2008NatGe...1..365K}} Non-technical summary: {{cite journal | author = Kleman, John | year = 2008 | doi = 10.1038/ngeo210 | title = Geomorphology: Where glaciers cut deep | journal = Nature Geoscience | volume = 1 | pages = 343 | issue=6|bibcode = 2008NatGe...1..343K }}</ref> This explains, for example, the deep profile of [[fjord]]s, which can reach a kilometer in depth as ice is topographically steered into them. The extension of fjords inland increases the rate of ice sheet thinning since they are the principal conduits for draining ice sheets. It also makes the ice sheets more sensitive to changes in climate and the ocean.<ref name=ngeo2008/> |
|||
===Marine ice sheet instability=== |
|||
[[File:West Antarctic Collapse.ogv|thumb|A collage of footage and animation to explain the changes that are occurring on the West Antarctic Ice Sheet, narrated by glaciologist [[Eric Rignot]]]] |
|||
In the 1970s, [[Johannes Weertman]] proposed that because [[seawater]] is denser than ice, then any ice sheets grounded below [[sea level]] inherently become less stable as they melt due to [[Archimedes' principle]].<ref name="Weertman1974" /> Effectively, these marine ice sheets must have enough mass to exceed the mass of the seawater displaced by the ice, which requires excess thickness. As the ice sheet melts and becomes thinner, the weight of the overlying ice decreases. At a certain point, sea water could force itself into the gaps which form at the base of the ice sheet, and ''marine ice sheet instability'' (MISI) would occur.<ref name="Weertman1974">{{Cite journal|last=Weertman|first=J.|date=1974|title=Stability of the Junction of an Ice Sheet and an Ice Shelf|journal=Journal of Glaciology|language=en|volume=13|issue=67|pages=3–11|doi=10.3189/S0022143000023327|issn=0022-1430|doi-access=free}}</ref><ref name="Pollard2015" /> |
|||
Even if the ice sheet is grounded below the sea level, MISI cannot occur as long as there is a stable ice shelf in front of it.<ref name="Pattyn 2018" /> The boundary between the ice sheet and the ice shelf, known as the ''grounding line'', is particularly stable if it is constrained in an [[Bay|embayment]].<ref name="Pattyn 2018" /> In that case, the ice sheet may not be thinning at all, as the amount of ice flowing over the grounding line would be likely to match the annual accumulation of ice from snow upstream.<ref name="Pollard2015" /> Otherwise, ocean warming at the base of an ice shelf tends to thin it through basal melting. As the ice shelf becomes thinner, it exerts less of an buttressing effect on the ice sheet, the so-called back stress increases and the grounding line is pushed backwards.<ref name="Pollard2015" /> The ice sheet is likely to start losing more ice from the new location of the grounding line and so become lighter and less capable of displacing seawater. This eventually pushes the grounding line back even further, creating a [[Positive feedback|self-reinforcing mechanism]].<ref name="Pollard2015">{{cite journal|journal=Nature|volume=412|pages=112–121|year=2015|title=Potential Antarctic Ice Sheet retreat driven by hydrofracturing and ice cliff failure |author=David Pollard |author2=Robert M. DeConto |author3=Richard B. Alley |doi=10.1016/j.epsl.2014.12.035|doi-access=free|bibcode=2015E&PSL.412..112P}}</ref><ref>{{cite web|url=https://blogs.egu.eu/divisions/cr/2016/06/22/marine-ice-sheet-instability-for-dummies-2/|title=Marine Ice Sheet Instability "For Dummies"|work=EGU|year=2016|author=David Docquier}}</ref> |
|||
Because the entire West Antarctic Ice Sheet is grounded below the sea level, it would be vulnerable to geologically rapid ice loss in this scenario.<ref>{{Cite journal|last=Mercer|first=J. H.|date=1978|title=West Antarctic ice sheet and CO2 greenhouse effect: a threat of disaster|journal=Nature|language=En|volume=271|issue=5643|pages=321–325|doi=10.1038/271321a0|issn=0028-0836|bibcode=1978Natur.271..321M|s2cid=4149290}}</ref><ref>{{Cite journal|last=Vaughan|first=David G.|date=2008-08-20|title=West Antarctic Ice Sheet collapse – the fall and rise of a paradigm|journal=Climatic Change|language=en|volume=91|issue=1–2|pages=65–79|doi=10.1007/s10584-008-9448-3|bibcode=2008ClCh...91...65V|s2cid=154732005|issn=0165-0009|url=http://nora.nerc.ac.uk/id/eprint/769/1/The_return_of_a_paradigm_16_-_nora.pdf}}</ref> Sea level rise from the ice sheet could be accelerated by tens of centimeters within the 21st century alone.<ref name="IPCC AR6 WG1 Ch.9">{{Cite journal |last1=Fox-Kemper |first1=B. |last2=Hewitt |first2=H.T.|author2-link=Helene Hewitt |last3=Xiao |first3=C. |last4=Aðalgeirsdóttir |first4=G. |last5=Drijfhout |first5=S.S. |last6=Edwards |first6=T.L. |last7=Golledge |first7=N.R. |last8=Hemer |first8=M. |last9=Kopp |first9=R.E. |last10=Krinner |first10=G. |last11=Mix |first11=A. |date=2021 |editor-last=Masson-Delmotte |editor-first=V. |editor2-last=Zhai |editor2-first=P. |editor3-last=Pirani |editor3-first=A. |editor4-last=Connors |editor4-first=S.L. |editor5-last=Péan |editor5-first=C. |editor6-last=Berger |editor6-first=S. |editor7-last=Caud |editor7-first=N. |editor8-last=Chen |editor8-first=Y. |editor9-last=Goldfarb |editor9-first=L. |title=Chapter 9: Ocean, Cryosphere and Sea Level Change |journal=Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change |url=https://www.ipcc.ch/report/ar6/wg1/downloads/report/IPCC_AR6_WGI_Chapter09.pdf |publisher=Cambridge University Press, Cambridge, UK and New York, NY, USA |pages=1270–1272 }}</ref> The majority of the East Antarctic Ice Sheet would not be affected, but its subglacial basins such as [[Wilkes Basin]] and the [[Aurora Subglacial Basin]] are also grounded below sea level and so subject to MISI. However, even geologically rapid sea level rise would still most likely require several millennia for the entirety of these ice masses to be lost.<ref name="ArmstrongMcKay2022">{{Cite journal |last1=Armstrong McKay |first1=David|last2=Abrams |first2=Jesse |last3=Winkelmann |first3=Ricarda |last4=Sakschewski |first4=Boris |last5=Loriani |first5=Sina |last6=Fetzer |first6=Ingo|last7=Cornell|first7=Sarah |last8=Rockström |first8=Johan |last9=Staal |first9=Arie |last10=Lenton |first10=Timothy |date=9 September 2022 |title=Exceeding 1.5°C global warming could trigger multiple climate tipping points |url=https://www.science.org/doi/10.1126/science.abn7950 |journal=Science |language=en |volume=377 |issue=6611 |pages=eabn7950 |doi=10.1126/science.abn7950 |pmid=36074831 |hdl=10871/131584 |s2cid=252161375 |issn=0036-8075|hdl-access=free }}</ref><ref name="Explainer">{{Cite web |last=Armstrong McKay |first=David |date=9 September 2022 |title=Exceeding 1.5°C global warming could trigger multiple climate tipping points – paper explainer |url=https://climatetippingpoints.info/2022/09/09/climate-tipping-points-reassessment-explainer/ |access-date=2 October 2022 |website=climatetippingpoints.info |language=en}}</ref> |
|||
==== Marine Ice Cliff Instability ==== |
|||
A related process known as ''Marine Ice Cliff Instability'' (MICI) posits that due to the physical characteristics of ice, [[subaerial]] ice cliffs exceeding ~90 meters in height are likely to collapse under their own weight, and this could lead to runaway ice sheet retreat in a fashion similar to MISI.<ref name="Pollard2015" /> For an ice sheet grounded below sea level with an inland-sloping bed, ice cliff failure removes peripheral ice, which then exposes taller, more unstable ice cliffs, further perpetuating the cycle of ice front failure and retreat. Surface melt can further enhance MICI through ponding and hydrofracture.<ref name="Pattyn 2018">{{Cite journal |last=Pattyn |first=Frank |author-link=Frank Pattyn |date=2018 |title=The paradigm shift in Antarctic ice sheet modelling |journal=Nature Communications |language=En |volume=9 |issue=1 |page=2728 |bibcode=2018NatCo...9.2728P |doi=10.1038/s41467-018-05003-z |issn=2041-1723 |pmc=6048022 |pmid=30013142}}</ref><ref>{{Cite journal|last1=Dow|first1=Christine F.|last2=Lee|first2=Won Sang|last3=Greenbaum|first3=Jamin S.|last4=Greene|first4=Chad A.|last5=Blankenship|first5=Donald D.|last6=Poinar|first6=Kristin|last7=Forrest|first7=Alexander L.|last8=Young|first8=Duncan A.|last9=Zappa|first9=Christopher J.|date=2018-06-01|title=Basal channels drive active surface hydrology and transverse ice shelf fracture|journal=Science Advances|language=en|volume=4|issue=6|pages=eaao7212|doi=10.1126/sciadv.aao7212|issn=2375-2548|pmc=6007161|pmid=29928691|bibcode=2018SciA....4.7212D}}</ref> However, this process is considered more speculative than MISI, as it has never been observed at any scale. Some of the more detailed modelling has ruled it out.<ref>{{cite news|url=https://www.sciencenews.org/article/climate-marine-ice-cliffs-sheets-collapse-not-inevitable-sea-level|title=Collapse may not always be inevitable for marine ice cliffs|last1=Perkins|first1=Sid|date=June 17, 2021|access-date=9 January 2023|agency=ScienceNews}}</ref> |
|||
===Ocean warming=== |
|||
{{See also|Ocean heat content}} |
|||
[[File:Schematic-of-stratification-and-precipitation-amplifying-feedbacks.jpg|thumb|Schematic of stratification and precipitation amplifying feedbacks. Stratification: increased freshwater flux reduces surface water density, thus reducing AABW formation, trapping NADW heat, and increasing ice shelf melt. Precipitation: increased freshwater flux cools ocean mixed layer, increases sea ice area, causing precipitation to fall before it reaches Antarctica, reducing ice sheet growth and increasing ocean surface freshening. Ice in West Antarctica and the Wilkes Basin, East Antarctica, is most vulnerable because of the instability of retrograde beds.]] |
|||
According to a 2016 published study, cold [[meltwater]] provides cooling of the ocean's surface layer, acting like a lid, and also affecting deeper waters by increasing subsurface [[ocean warming]] and thus facilitating ice melt. |
|||
{{Quote|Our "pure freshwater" experiments show that the low-density lid causes deep-ocean warming, especially at depths of ice shelf grounding lines that provide most of the restraining force limiting ice sheet discharge.<ref name="Hansen2016"/>}} |
|||
Another theory discussed in 2007 for increasing warm bottom water is that changes in air circulation patterns have led to increased upwelling of warm, deep ocean water along the coast of Antarctica and that this warm water has increased melting of floating ice shelves.<ref name="auto">{{cite web|url=http://www.jsg.utexas.edu/news/2007/05/statement-thinning-of-west-antarctic-ice-sheet-demands-improved-monitoring-to-reduce-uncertainty-over-potential-sea-level-rise/|title=Statement: Thinning of West Antarctic Ice Sheet Demands Improved Monitoring to Reduce Uncertainty over Potential Sea-Level Rise|website=Jsg.utexas.edu|access-date=26 October 2017}}</ref> An ocean model has shown how changes in winds can help channel the water along deep troughs on the sea floor, toward the ice shelves of outlet glaciers.<ref name="ThomaJenkins2008">{{cite journal| doi = 10.1029/2008GL034939| last1 = Thoma | first1 = M.| last2 = Jenkins | first2 = A.| last3 = Holland | first3 = D.| last4 = Jacobs | first4 = S.| year = 2008| title = Modelling Circumpolar Deep Water intrusions on the Amundsen Sea continental shelf, Antarctica| journal = [[Geophysical Research Letters]]| volume = 35| issue = 18| page = L18602| bibcode = 2008GeoRL..3518602T| s2cid = 55937812 | url = http://epic.awi.de/25479/1/2008_Modelling_Circumpolar_Deep_Water_intrusions_on_the_Amundsen_Sea_continental_shelf_Antarctica.pdf }}</ref> |
|||
===Observations=== |
|||
{{See also|West Antarctic Ice Sheet#Potential collapse}} |
|||
In West Antarctica, the [[Thwaites glacier|Thwaites]] and [[Pine Island glacier|Pine Island]] glaciers have been identified to be potentially prone to MISI, and both glaciers have been rapidly thinning and accelerating in recent decades.<ref>{{cite web|url=https://www.theatlantic.com/science/archive/2018/06/after-decades-of-ice-loss-antarctica-is-now-hemorrhaging-mass/562748/|work=The Atlantic|year=2018|title=After Decades of Losing Ice, Antarctica Is Now Hemorrhaging It}}</ref><ref>{{cite web|url=http://www.antarcticglaciers.org/glaciers-and-climate/ice-ocean-interactions/marine-ice-sheets/|work=AntarcticGlaciers.org|year=2014|title=Marine ice sheet instability}}</ref><ref name="Gardner 2018">{{Cite journal|last1=Gardner|first1=A. S.|last2=Moholdt|first2=G.|last3=Scambos|first3=T.|last4=Fahnstock|first4=M.|last5=Ligtenberg|first5=S.|last6=van den Broeke|first6=M.|last7=Nilsson|first7=J.|date=2018-02-13|title=Increased West Antarctic and unchanged East Antarctic ice discharge over the last 7 years|journal=The Cryosphere|volume=12|issue=2|pages=521–547|doi=10.5194/tc-12-521-2018|bibcode=2018TCry...12..521G|issn=1994-0424|doi-access=free}}</ref><ref>{{Cite journal|author1=IMBIE team|date=2018|title=Mass balance of the Antarctic Ice Sheet from 1992 to 2017|journal=Nature|language=En|volume=558|issue=7709|pages=219–222|doi=10.1038/s41586-018-0179-y|issn=0028-0836|pmid=29899482|url=https://orbi.uliege.be/handle/2268/225208|bibcode=2018Natur.558..219I|hdl=2268/225208|s2cid=49188002}}</ref> In East Antarctica, [[Totten Glacier]] is the largest glacier known to be subject to MISI,<ref>{{Cite journal|last1=Young|first1=Duncan A.|last2=Wright|first2=Andrew P.|last3=Roberts|first3=Jason L.|last4=Warner|first4=Roland C.|last5=Young|first5=Neal W.|last6=Greenbaum|first6=Jamin S.|last7=Schroeder|first7=Dustin M.|last8=Holt|first8=John W.|last9=Sugden|first9=David E.|date=2011-06-02|title=A dynamic early East Antarctic Ice Sheet suggested by ice-covered fjord landscapes|journal=Nature|language=En|volume=474|issue=7349|pages=72–75|doi=10.1038/nature10114|pmid=21637255|issn=0028-0836|bibcode=2011Natur.474...72Y|s2cid=4425075}}</ref> and its potential contribution to sea level rise is comparable to that of the entire West Antarctic Ice Sheet. Totten Glacier has been losing mass nearly monotonically in recent decades,<ref>{{Cite journal|last=Mohajerani|first=Yara|date=2018|title=Mass Loss of Totten and Moscow University Glaciers, East Antarctica, Using Regionally Optimized GRACE Mascons|journal=Geophysical Research Letters|volume=45|issue=14|pages=7010–7018|doi=10.1029/2018GL078173|bibcode=2018GeoRL..45.7010M|s2cid=134054176 |url=https://escholarship.org/uc/item/21c3r9dv|doi-access=free}}</ref> suggesting rapid retreat is possible in the near future, although the dynamic behavior of Totten Ice Shelf is known to vary on seasonal to interannual timescales.<ref>{{Cite journal|last1=Greene|first1=Chad A.|last2=Young|first2=Duncan A.|last3=Gwyther|first3=David E.|last4=Galton-Fenzi|first4=Benjamin K.|last5=Blankenship|first5=Donald D.|date=2018|title=Seasonal dynamics of Totten Ice Shelf controlled by sea ice buttressing|journal=The Cryosphere|language=en|volume=12|issue=9|pages=2869–2882|doi=10.5194/tc-12-2869-2018|issn=1994-0416|doi-access=free|bibcode=2018TCry...12.2869G}}</ref><ref>{{Cite journal|last1=Roberts|first1=Jason|last2=Galton-Fenzi|first2=Benjamin K.|last3=Paolo|first3=Fernando S.|last4=Donnelly|first4=Claire|last5=Gwyther|first5=David E.|last6=Padman|first6=Laurie|last7=Young|first7=Duncan|last8=Warner|first8=Roland|last9=Greenbaum|first9=Jamin|date=2017-08-23|title=Ocean forced variability of Totten Glacier mass loss|journal=Geological Society, London, Special Publications|volume=461|issue=1|pages=175–186|doi=10.1144/sp461.6|issn=0305-8719|url=https://eprints.utas.edu.au/25611/1/SP461.6.full.pdf|bibcode=2018GSLSP.461..175R|doi-access=free}}</ref><ref>{{Cite journal|last1=Greene|first1=Chad A.|last2=Blankenship|first2=Donald D.|last3=Gwyther|first3=David E.|last4=Silvano|first4=Alessandro|last5=Wijk|first5=Esmee van|date=2017-11-01|title=Wind causes Totten Ice Shelf melt and acceleration|journal=Science Advances|language=en|volume=3|issue=11|pages=e1701681|doi=10.1126/sciadv.1701681|issn=2375-2548|pmc=5665591|pmid=29109976|bibcode=2017SciA....3E1681G}}</ref> The Wilkes Basin is the only major submarine basin in Antarctica that is not thought to be sensitive to warming.<ref name="Gardner 2018" /> |
|||
In October 2023, a study published in ''[[Nature Climate Change]]'' projected that ocean warming at about triple the historical rate is likely unavoidable in the 21st century, with no significant difference between mid-range emissions scenarios versus achieving the most ambitious targets of the Paris Agreement—suggesting that [[Climate change mitigation#Needed emissions cuts|greenhouse gas mitigation]] has limited ability to prevent collapse of the [[West Antarctic Ice Sheet]].<ref name=NatureClimateChange_20231023>{{cite journal |last1=Naughten |first1=Kaitlin A. |last2=Holland |first2=Paul R. |last3=De Rydt |first3=Jan |title=Unavoidable future increase in West Antarctic ice-shelf melting over the twenty-first century |journal=Nature Climate Change |date=23 October 2023 |volume=13 |issue=11 |pages=1222–1228 |doi=10.1038/s41558-023-01818-x |s2cid=264476246 |doi-access=free |bibcode=2023NatCC..13.1222N }}</ref> |
|||
== Earth's current two ice sheets == |
== Earth's current two ice sheets == |
||
Line 70: | Line 235: | ||
*[[Snowball Earth]] |
*[[Snowball Earth]] |
||
*[[Wisconsin glaciation]] |
*[[Wisconsin glaciation]] |
||
* [[Ice-sheet model]] |
|||
==References== |
==References== |
||
Line 77: | Line 243: | ||
* [https://web.archive.org/web/20070608011925/http://www.unep.org/geo/geo_ice/ United Nations Environment Programme: Global Outlook for Ice and Snow] |
* [https://web.archive.org/web/20070608011925/http://www.unep.org/geo/geo_ice/ United Nations Environment Programme: Global Outlook for Ice and Snow] |
||
* http://www.nasa.gov/vision/earth/environment/ice_sheets.html {{Webarchive|url=https://web.archive.org/web/20120916074032/http://www.nasa.gov/vision/earth/environment/ice_sheets.html |date=2012-09-16 }} |
* http://www.nasa.gov/vision/earth/environment/ice_sheets.html {{Webarchive|url=https://web.archive.org/web/20120916074032/http://www.nasa.gov/vision/earth/environment/ice_sheets.html |date=2012-09-16 }} |
||
*{{cite journal|last=Barber|first=D.G. |author2=McCullough, G. |author3=Babb, D. |author4=Komarov, A. S. |author5=Candlish, L. M. |author6=Lukovich, J. V. |author7=Asplin, M. |author8=Prinsenberg, S. |author9=Dmitrenko, I. |author10=Rysgaard, S.|year=2014|title=Climate change and ice hazards in the Beaufort Sea|journal=Elementa|volume=2 |pages=000025 |doi=10.12952/journal.elementa.000025 |bibcode=2014EleSA...2.0025B |doi-access=free |url=https://pure.au.dk/portal/files/79272460/journal.elementa.000025.pdf }} |
|||
*[https://blogs.egu.eu/divisions/cr/2016/06/22/marine-ice-sheet-instability-for-dummies-2/ Marine Ice Sheet Instability "For Dummies"] |
|||
{{Global warming}} |
{{Global warming}} |
||
Line 89: | Line 258: | ||
[[Category:Effects of climate change]] |
[[Category:Effects of climate change]] |
||
[[Category:Cryosphere]] |
[[Category:Cryosphere]] |
||
[[Category:Articles containing video clips]] |
Inglaciology, an ice sheet, also known as a continental glacier,[2] is a mass of glacial ice that covers surrounding terrain and is greater than 50,000 km2 (19,000 sq mi).[3] The only current ice sheets are the Antarctic ice sheet and the Greenland ice sheet. Ice sheets are bigger than ice shelves or alpine glaciers. Masses of ice covering less than 50,000 km2 are termed an ice cap. An ice cap will typically feed a series of glaciers around its periphery.
Although the surface is cold, the base of an ice sheet is generally warmer due to geothermal heat. In places, melting occurs and the melt-water lubricates the ice sheet so that it flows more rapidly. This process produces fast-flowing channels in the ice sheet — these are ice streams.
In previous geologic time spans (glacial periods) there were other ice sheets: during the Last Glacial PeriodatLast Glacial Maximum, the Laurentide Ice Sheet covered much of North America, the Weichselian ice sheet covered Northern Europe and the Patagonian Ice Sheet covered southern South America.
An ice sheet is "an ice body originating on land that covers an area of continental size, generally defined as covering >50,000 km2 , and that has formed over thousands of years through accumulation and compaction of snow".[4]: 2234
Ice sheets have the following properties: "An ice sheet flows outward from a high central ice plateau with a small average surface slope. The margins usually slope more steeply, and most ice is discharged through fast-flowing ice streams or outlet glaciers, often into the sea or into ice shelves floating on the sea."[4]: 2234
Ice movement is dominated by the motion of glaciers, whose activity is determined by a number of processes.[6] Their motion is the result of cyclic surges interspersed with longer periods of inactivity, on both hourly and centennial time scales.
Until recently, ice sheets were viewed as inert components of the carbon cycle and were largely disregarded in global models. Research in the past decade has transformed this view, demonstrating the existence of uniquely adapted microbial communities, high rates of biogeochemical/physical weathering in ice sheets and storage and cycling of organic carbon in excess of 100 billion tonnes, as well as nutrients (see diagram).[5]
The motion of ice sheets is driven by gravity but is controlled by temperature and the strength of individual glacier bases. A number of processes alter these two factors, resulting in cyclic surges of activity interspersed with longer periods of inactivity, on time scales ranging from hourly to the centennial.
The interface between an ice stream and the ocean is a significant control of the rate of flow.
Ice shelves are thick layers of ice floating on the sea – can stabilise the glaciers that feed them. These tend to have accumulation on their tops, may experience melting on their bases, and calve icebergs at their periphery. The catastrophic collapse of the Larsen B ice shelf in the space of three weeks during February 2002 yielded some unexpected observations. The glaciers that had fed the ice sheet (Crane, Jorum, Green, Hektoria – see image) increased substantially in velocity. This cannot have been due to seasonal variability, as glaciers flowing into the remnants of the ice shelf (Flask, Leppard) did not accelerate.[7]
Ice shelves exert a dominant control in Antarctica, but are less important in Greenland, where the ice sheet meets the sea in fjords. Here, melting is the dominant ice removal process,[8] resulting in predominant mass loss occurring towards the edges of the ice sheet, where icebergs are calved in the fjords and surface meltwater runs into the ocean.
Tidal effects are also important; the influence of a 1 m tidal oscillation can be felt as much as 100 km from the sea.[9] On an hour-to-hour basis, surges of ice motion can be modulated by tidal activity. During larger spring tides, an ice stream will remain almost stationary for hours at a time, before a surge of around a foot in under an hour, just after the peak high tide; a stationary period then takes hold until another surge towards the middle or end of the falling tide.[10][11] At neap tides, this interaction is less pronounced, without tides surges would occur more randomly, approximately every 12 hours.[10]
Ice shelves are also sensitive to basal melting. In Antarctica, this is driven by heat fed to the shelf by the circumpolar deep water current, which is 3 °C above the ice's melting point.[12]
As well as heat, the sea can also exchange salt with the oceans. The effect of latent heat, resulting from melting of ice or freezing of sea water, also has a role to play. The effects of these, and variability in snowfall and base sea level combined, account for around 80 mm annual variability in ice shelf thickness.
Over long time scales, ice sheet mass balance is governed by the amount of sunlight reaching the Earth. This variation in sunlight reaching the Earth, or insolation, over geologic time is in turn determined by the angle of the Earth to the Sun and shape of the Earth's orbit, as it is pulled on by neighboring planets; these variations occur in predictable patterns called Milankovitch cycles. Milankovitch cycles dominate climate on the glacial–interglacial timescale, but there exist variations in ice sheet extent that are not linked directly with insolation.
For instance, during at least the last 100,000 years, portions of the ice sheet covering much of North America, the Laurentide Ice Sheet broke apart sending large flotillas of icebergs into the North Atlantic. When these icebergs melted they dropped the boulders and other continental rocks they carried, leaving layers known as ice rafted debris. These so-called Heinrich events, named after their discoverer Hartmut Heinrich, appear to have a 7,000–10,000-year periodicity, and occur during cold periods within the last interglacial.[13]
Internal ice sheet "binge-purge" cycles may be responsible for the observed effects, where the ice builds to unstable levels, then a portion of the ice sheet collapses. External factors might also play a role in forcing ice sheets. Dansgaard–Oeschger events are abrupt warmings of the northern hemisphere occurring over the space of perhaps 40 years. While these D–O events occur directly after each Heinrich event, they also occur more frequently – around every 1500 years; from this evidence, paleoclimatologists surmise that the same forcings may drive both Heinrich and D–O events.[14]
Hemispheric asynchrony in ice sheet behavior has been observed by linking short-term spikes of methane in Greenland ice cores and Antarctic ice cores. During Dansgaard–Oeschger events, the northern hemisphere warmed considerably, dramatically increasing the release of methane from wetlands, that were otherwise tundra during glacial times. This methane quickly distributes evenly across the globe, becoming incorporated in Antarctic and Greenland ice. With this tie, paleoclimatologists have been able to say that the ice sheets on Greenland only began to warm after the Antarctic ice sheet had been warming for several thousand years. Why this pattern occurs is still open for debate.[15][16]
The main cause of flow within glaciers can be attributed to an increase in the surface slope, brought upon by an imbalance between the amounts of accumulation vs. ablation. This imbalance increases the shear stress on a glacier until it begins to flow. The flow velocity and deformation will increase as the equilibrium line between these two processes is approached, but are also affected by the slope of the ice, the ice thickness and temperature.[17][6]
When the amount of strain (deformation) is proportional to the stress being applied, ice will act as an elastic solid. Ice will not flow until it has reached a thickness of 30 meters (98 ft), but after 50 meters (164 ft), small amounts of stress can result in a large amount of strain, causing the deformation to become a plastic flow rather than elastic. At this point the glacier will begin to deform under its own weight and flow across the landscape. According to the Glen–Nye flow law, the relationship between stress and strain, and thus the rate of internal flow, can be modeled as follows:[17][6]
where:
The lowest velocities are near the base of the glacier and along valley sides where friction acts against flow, causing the most deformation. Velocity increases inward toward the center line and upward, as the amount of deformation decreases. The highest flow velocities are found at the surface, representing the sum of the velocities of all the layers below.[17][6]
Glaciers may also move by basal sliding, where the base of the glacier is lubricated by meltwater, allowing the glacier to slide over the terrain on which it sits. Meltwater may be produced by pressure-induced melting, friction or geothermal heat. The more variable the amount of melting at surface of the glacier, the faster the ice will flow.[18]
The top 50 meters of the glacier form the fracture zone, where ice moves as a single unit. Cracks form as the glacier moves over irregular terrain, which may penetrate the full depth of the fracture zone.
Most of the important processes controlling glacial motion occur in the ice-bed contact—even though it is only a few meters thick.[9] Glaciers will move by sliding when the basal shear stress drops below the shear resulting from the glacier's weight.[clarification needed]
For a given glacier, the two variables are τD, which varies with h, the depth of the glacier, and τB, the basal shear stress.[clarification needed]
The basal shear stress is a function of three factors: the bed's temperature, roughness and softness.[9]
Whether a bed is hard or soft depends on the porosity and pore pressure; higher porosity decreases the sediment strength (thus increases the shear stress τB).[9] If the sediment strength falls far below τD, movement of the glacier will be accommodated by motion in the sediments, as opposed to sliding. Porosity may vary through a range of methods.
A soft bed, with high porosity and low pore fluid pressure, allows the glacier to move by sediment sliding: the base of the glacier may even remain frozen to the bed, where the underlying sediment slips underneath it like a tube of toothpaste. A hard bed cannot deform in this way; therefore the only way for hard-based glaciers to move is by basal sliding, where meltwater forms between the ice and the bed itself.[19]
Bed softness may vary in space or time, and changes dramatically from glacier to glacier. An important factor is the underlying geology; glacial speeds tend to differ more when they change bedrock than when the gradient changes.[19]
As well as affecting the sediment stress, fluid pressure (pw) can affect the friction between the glacier and the bed. High fluid pressure provides a buoyancy force upwards on the glacier, reducing the friction at its base. The fluid pressure is compared to the ice overburden pressure, pi, given by ρgh. Under fast-flowing ice streams, these two pressures will be approximately equal, with an effective pressure (pi – pw) of 30 kPa; i.e. all of the weight of the ice is supported by the underlying water, and the glacier is afloat.[9]
A number of factors can affect bed temperature, which is intimately associated with basal meltwater. The melting point of water decreases under pressure, meaning that water melts at a lower temperature under thicker glaciers.[9] This acts as a "double whammy", because thicker glaciers have a lower heat conductance, meaning that the basal temperature is also likely to be higher.[19]
Bed temperature tends to vary in a cyclic fashion. A cool bed has a high strength, reducing the speed of the glacier. This increases the rate of accumulation, since newly fallen snow is not transported away. Consequently, the glacier thickens, with three consequences: firstly, the bed is better insulated, allowing greater retention of geothermal heat. Secondly, the increased pressure can facilitate melting. Most importantly, τD is increased. These factors will combine to accelerate the glacier. As friction increases with the square of velocity, faster motion will greatly increase frictional heating, with ensuing melting – which causes a positive feedback, increasing ice speed to a faster flow rate still: west Antarctic glaciers are known to reach velocities of up to a kilometre per year.[9] Eventually, the ice will be surging fast enough that it begins to thin, as accumulation cannot keep up with the transport. This thinning will increase the conductive heat loss, slowing the glacier and causing freezing. This freezing will slow the glacier further, often until it is stationary, whence the cycle can begin again.[19]
Supraglacial lakes represent another possible supply of liquid water to the base of glaciers, so they can play an important role in accelerating glacial motion. Lakes of a diameter greater than ~300 m are capable of creating a fluid-filled crevasse to the glacier/bed interface. When these crevasses form, the entirety of the lake's (relatively warm) contents can reach the base of the glacier in as little as 2–18 hours – lubricating the bed and causing the glacier to surge.[20] Water that reaches the bed of a glacier may freeze there, increasing the thickness of the glacier by pushing it up from below.[21]
Finally, bed roughness can act to slow glacial motion. The roughness of the bed is a measure of how many boulders and obstacles protrude into the overlying ice. Ice flows around these obstacles by melting under the high pressure on their stoss side; the resultant meltwater is then forced into the cavity arising in their lee side, where it re-freezes.[9]
The flow of water under the glacial surface can have a large effect on the motion of the glacier itself. Subglacial lakes contain significant amounts of water, which can move fast: cubic kilometres can be transported between lakes over the course of a couple of years.[22]
This motion is thought to occur in two main modes: pipe flow involves liquid water moving through pipe-like conduits, like a sub-glacial river; sheet flow involves motion of water in a thin layer. A switch between the two flow conditions may be associated with surging behaviour. Indeed, the loss of sub-glacial water supply has been linked with the shut-down of ice movement in the Kamb ice stream.[22] The subglacial motion of water is expressed in the surface topography of ice sheets, which slump down into vacated subglacial lakes.[22]
The implications of the current climate change on ice sheets are difficult to ascertain. It is clear that increasing temperatures are resulting in reduced ice volumes globally.[8] (Due to increased precipitation, the mass of parts of the Antarctic ice sheet may currently be increasing, but the total mass balance is unclear.[8])
Rising sea levels will reduce the stability of ice shelves, which have a key role in reducing glacial motion. Some Antarctic ice shelves are currently thinning by tens of metres per year, and the collapse of the Larsen B shelf was preceded by thinning of just 1 metre per year.[8] Further, increased ocean temperatures of 1 °C may lead to up to 10 metres per year of basal melting.[8] Ice shelves are always stable under mean annual temperatures of −9 °C, but never stable above −5 °C; this places regional warming of 1.5 °C, as preceded the collapse of Larsen B, in context.[8]
Increasing global air temperatures take around 10,000 years to directly propagate through the ice before they influence bed temperatures, but may have an effect through increased surfacal melting, producing more supraglacial lakes, which may feed warm water to glacial bases and facilitate glacial motion.[8] In areas of increased precipitation, such as Antarctica, the addition of mass will increase rate of glacial motion, hence the turnover in the ice sheet. Observations, while currently limited in scope, do agree with these predictions of an increasing rate of ice loss from both Greenland and Antarctica.[8] A possible positive feedback may result from shrinking ice caps, in volcanically active Iceland at least. Isostatic rebound may lead to increased volcanic activity, causing basal warming – and, through CO2 release, further climate change.[23]
Cold meltwater provides cooling of the ocean's surface layer, acting like a lid, and also affecting deeper waters by increasing subsurface ocean warming and thus facilitating ice melt.
Our "pure freshwater" experiments show that the low-density lid causes deep-ocean warming, especially at depths of ice shelf grounding lines that provide most of the restraining force limiting ice sheet discharge.[24]
Because ice can flow faster where it is thicker, the rate of glacier-induced erosion is directly proportional to the thickness of overlying ice. Consequently, pre-glacial low hollows will be deepened and pre-existing topography will be amplified by glacial action, while nunataks, which protrude above ice sheets, barely erode at all – erosion has been estimated as 5 m per 1.2 million years.[25] This explains, for example, the deep profile of fjords, which can reach a kilometer in depth as ice is topographically steered into them. The extension of fjords inland increases the rate of ice sheet thinning since they are the principal conduits for draining ice sheets. It also makes the ice sheets more sensitive to changes in climate and the ocean.[25]
In the 1970s, Johannes Weertman proposed that because seawater is denser than ice, then any ice sheets grounded below sea level inherently become less stable as they melt due to Archimedes' principle.[26] Effectively, these marine ice sheets must have enough mass to exceed the mass of the seawater displaced by the ice, which requires excess thickness. As the ice sheet melts and becomes thinner, the weight of the overlying ice decreases. At a certain point, sea water could force itself into the gaps which form at the base of the ice sheet, and marine ice sheet instability (MISI) would occur.[26][27]
Even if the ice sheet is grounded below the sea level, MISI cannot occur as long as there is a stable ice shelf in front of it.[28] The boundary between the ice sheet and the ice shelf, known as the grounding line, is particularly stable if it is constrained in an embayment.[28] In that case, the ice sheet may not be thinning at all, as the amount of ice flowing over the grounding line would be likely to match the annual accumulation of ice from snow upstream.[27] Otherwise, ocean warming at the base of an ice shelf tends to thin it through basal melting. As the ice shelf becomes thinner, it exerts less of an buttressing effect on the ice sheet, the so-called back stress increases and the grounding line is pushed backwards.[27] The ice sheet is likely to start losing more ice from the new location of the grounding line and so become lighter and less capable of displacing seawater. This eventually pushes the grounding line back even further, creating a self-reinforcing mechanism.[27][29]
Because the entire West Antarctic Ice Sheet is grounded below the sea level, it would be vulnerable to geologically rapid ice loss in this scenario.[30][31] Sea level rise from the ice sheet could be accelerated by tens of centimeters within the 21st century alone.[32] The majority of the East Antarctic Ice Sheet would not be affected, but its subglacial basins such as Wilkes Basin and the Aurora Subglacial Basin are also grounded below sea level and so subject to MISI. However, even geologically rapid sea level rise would still most likely require several millennia for the entirety of these ice masses to be lost.[33][34]
A related process known as Marine Ice Cliff Instability (MICI) posits that due to the physical characteristics of ice, subaerial ice cliffs exceeding ~90 meters in height are likely to collapse under their own weight, and this could lead to runaway ice sheet retreat in a fashion similar to MISI.[27] For an ice sheet grounded below sea level with an inland-sloping bed, ice cliff failure removes peripheral ice, which then exposes taller, more unstable ice cliffs, further perpetuating the cycle of ice front failure and retreat. Surface melt can further enhance MICI through ponding and hydrofracture.[28][35] However, this process is considered more speculative than MISI, as it has never been observed at any scale. Some of the more detailed modelling has ruled it out.[36]
According to a 2016 published study, cold meltwater provides cooling of the ocean's surface layer, acting like a lid, and also affecting deeper waters by increasing subsurface ocean warming and thus facilitating ice melt.
Our "pure freshwater" experiments show that the low-density lid causes deep-ocean warming, especially at depths of ice shelf grounding lines that provide most of the restraining force limiting ice sheet discharge.[24]
Another theory discussed in 2007 for increasing warm bottom water is that changes in air circulation patterns have led to increased upwelling of warm, deep ocean water along the coast of Antarctica and that this warm water has increased melting of floating ice shelves.[37] An ocean model has shown how changes in winds can help channel the water along deep troughs on the sea floor, toward the ice shelves of outlet glaciers.[38]
In West Antarctica, the Thwaites and Pine Island glaciers have been identified to be potentially prone to MISI, and both glaciers have been rapidly thinning and accelerating in recent decades.[39][40][41][42] In East Antarctica, Totten Glacier is the largest glacier known to be subject to MISI,[43] and its potential contribution to sea level rise is comparable to that of the entire West Antarctic Ice Sheet. Totten Glacier has been losing mass nearly monotonically in recent decades,[44] suggesting rapid retreat is possible in the near future, although the dynamic behavior of Totten Ice Shelf is known to vary on seasonal to interannual timescales.[45][46][47] The Wilkes Basin is the only major submarine basin in Antarctica that is not thought to be sensitive to warming.[41]
In October 2023, a study published in Nature Climate Change projected that ocean warming at about triple the historical rate is likely unavoidable in the 21st century, with no significant difference between mid-range emissions scenarios versus achieving the most ambitious targets of the Paris Agreement—suggesting that greenhouse gas mitigation has limited ability to prevent collapse of the West Antarctic Ice Sheet.[48]
West Antarctic ice sheet | |
---|---|
![]() | |
Type | Ice sheet |
Area | <1,970,000 km2 (760,000 sq mi)[49] |
Thickness | ~1.05 km (0.7 mi) (average),[50] ~2 km (1.2 mi) (maximum)[49] |
Status | Receding |
The West Antarctic Ice Sheet (WAIS) is the segment of the continental ice sheet that covers West Antarctica, the portion of Antarctica on the side of the Transantarctic Mountains that lies in the Western Hemisphere. It is classified as a marine-based ice sheet, meaning that its bed lies well below sea level and its edges flow into floating ice shelves. The WAIS is bounded by the Ross Ice Shelf, the Ronne Ice Shelf, and outlet glaciers that drain into the Amundsen Sea.[51]
As a smaller part of Antarctica, WAIS is also more strongly affected by climate change. There has been warming over the ice sheet since the 1950s,[52][53] and a substantial retreat of its coastal glaciers since at least the 1990s.[54] Estimates suggest it added around 7.6 ± 3.9 mm (19⁄64 ± 5⁄32 in) to the global sea level rise between 1992 and 2017,[55] and has been losing ice in the 2010s at a rate equivalent to 0.4 millimetres (0.016 inches) of annual sea level rise.[56] While some of its losses are offset by the growth of the East Antarctic ice sheet, Antarctica as a whole will most likely lose enough ice by 2100 to add 11 cm (4.3 in) to sea levels. Further, marine ice sheet instability may increase this amount by tens of centimeters, particularly under high warming.[57] Fresh meltwater from WAIS also contributes to ocean stratification and dilutes the formation of salty Antarctic bottom water, which destabilizes Southern Ocean overturning circulation.[57][58][59]
In the long term, the West Antarctic Ice Sheet is likely to disappear due to the warming which has already occurred.[60] Paleoclimate evidence suggests that this has already happened during the Eemian period, when the global temperatures were similar to the early 21st century.[61][62] It is believed that the loss of the ice sheet would take place between 2,000 and 13,000 years in the future,[63][64] although several centuries of high emissions may shorten this to 500 years.[65] 3.3 m (10 ft 10 in) of sea level rise would occur if the ice sheet collapses but leaves ice caps on the mountains behind. Total sea level rise from West Antarctica increases to 4.3 m (14 ft 1 in) if they melt as well,[66] but this would require a higher level of warming.[67] Isostatic rebound of ice-free land may also add around 1 m (3 ft 3 in) to the global sea levels over another 1,000 years.[65]
The preservation of WAIS may require a persistent reduction of global temperatures to 1 °C (1.8 °F) below the preindustrial level, or to 2 °C (3.6 °F) below the temperature of 2020.[68] Because the collapse of the ice sheet would be preceded by the loss of Thwaites Glacier and Pine Island Glacier, some have instead proposed interventions to preserve them. In theory, adding thousands of gigatonnes of artificially created snow could stabilize them,[69] but it would be extraordinarily difficult and may not account for the ongoing acceleration of ocean warming in the area.[60] Others suggest that building obstacles to warm water flows beneath glaciers would be able to delay the disappearance of the ice sheet by many centuries, but it would still require one of the largest civil engineering interventions in history.East Antarctic ice sheet | |
---|---|
![]() | |
Type | Ice sheet |
Thickness | ~2.2 km (1.4 mi) (average),[70] ~4.9 km (3.0 mi) (maximum) [71] |
The East Antarctic Ice Sheet (EAIS) lies between 45° west and 168° east longitudinally. It was first formed around 34 million years ago,[72] and it is the largest ice sheet on the entire planet, with far greater volume than the Greenland ice sheet or the West Antarctic Ice Sheet (WAIS), from which it is separated by the Transantarctic Mountains. The ice sheet is around 2.2 km (1.4 mi) thick on average and is 4,897 m (16,066 ft) at its thickest point.[73] It is also home to the geographic South Pole, South Magnetic Pole and the Amundsen–Scott South Pole Station.
The surface of the EAIS is the driest, windiest, and coldest place on Earth. Lack of moisture in the air, high albedo from the snow as well as the surface's consistently high elevation[74] results in the reported cold temperature records of nearly −100 °C (−148 °F).[75][76] It is the only place on Earth cold enough for atmospheric temperature inversion to occur consistently. That is, while the atmosphere is typically warmest near the surface and becomes cooler at greater elevation, atmosphere during the Antarctic winter is cooler at the surface than in its middle layers. Consequently, greenhouse gases actually trap heat in the middle atmosphere and reduce its flow towards the surface while the temperature inversion lasts.[74]
Due to these factors, East Antarctica had experienced slight cooling for decades while the rest of the world warmed as the result of climate change. Clear warming over East Antarctica only started to occur since the year 2000, and was not conclusively detected until the 2020s.[77][78] In the early 2000s, cooling over East Antarctica seemingly outweighing warming over the rest of the continent was frequently misinterpreted by the media and occasionally used as an argument for climate change denial.[79][80][81] After 2009, improvements in Antarctica's instrumental temperature record have proven that the warming over West Antarctica resulted in consistent net warming across the continent since the 1957.[82]
Because the East Antarctic ice sheet has barely warmed, it is still gaining ice on average.[83][84] for instance, GRACE satellite data indicated East Antarctica mass gain of 60 ± 13 billion tons per year between 2002 and 2010.[85] It is most likely to first see sustained losses of ice at its most vulnerable locations such as Totten Glacier and Wilkes Basin. Those areas are sometimes collectively described as East Antarctica's subglacial basins, and it is believed that once the warming reaches around 3 °C (5.4 °F), then they would start to collapse over a period of around 2,000 years,[86][87] This collapse would ultimately add between 1.4 m (4 ft 7 in) and 6.4 m (21 ft 0 in) to sea levels, depending on the ice sheet model used.[88] The EAIS as a whole holds enough ice to raise global sea levels by 53.3 m (175 ft).[73] However, it would take global warming in a range between 5 °C (9.0 °F) and 10 °C (18 °F), and a minimum of 10,000 years for the entire ice sheet to be lost.[86][87]The Greenland ice sheet is an ice sheet which forms the second largest body of ice in the world. It is an average of 1.67 km (1.0 mi) thick, and over 3 km (1.9 mi) thick at its maximum.[89] It is almost 2,900 kilometres (1,800 mi) long in a north–south direction, with a maximum width of 1,100 kilometres (680 mi) at a latitude of 77°N, near its northern edge.[90] The ice sheet covers 1,710,000 square kilometres (660,000 sq mi), around 80% of the surface of Greenland, or about 12% of the area of the Antarctic ice sheet.[89] The term 'Greenland ice sheet' is often shortened to GIS or GrIS in scientific literature.[91][92][93][94]
Greenland has had major glaciers and ice caps for at least 18 million years,[95] but a single ice sheet first covered most of the island some 2.6 million years ago.[96] Since then, it has both grown[97][98] and contracted significantly.[99][100][101] The oldest known ice on Greenland is about 1 million years old.[102] Due to anthropogenic greenhouse gas emissions, the ice sheet is now the warmest it has been in the past 1000 years,[103] and is losing ice at the fastest rate in at least the past 12,000 years.[104]
Every summer, parts of the surface melt and ice cliffs calve into the sea. Normally the ice sheet would be replenished by winter snowfall,[92] but due to global warming the ice sheet is melting two to five times faster than before 1850,[105] and snowfall has not kept up since 1996.[106] If the Paris Agreement goal of staying below 2 °C (3.6 °F) is achieved, melting of Greenland ice alone would still add around 6 cm (2+1⁄2 in) to global sea level rise by the end of the century. If there are no reductions in emissions, melting would add around 13 cm (5 in) by 2100,[107]: 1302 with a worst-case of about 33 cm (13 in).[108] For comparison, melting has so far contributed 1.4 cm (1⁄2 in) since 1972,[109] while sea level rise from all sources was 15–25 cm (6–10 in)) between 1901 and 2018.[110]: 5
The melting of the Greenland and West Antarctic ice sheets will continue to contribute to sea level rise over long time-scales. The Greenland ice sheet loss is mainly driven by melt from the top. Antarctic ice loss is driven by warm ocean water melting the outlet glaciers.[114]: 1215
Future melt of the West Antarctic ice sheet is potentially abrupt under a high emission scenario, as a consequence of a partial collapse.[115]: 595–596 Part of the ice sheet is grounded on bedrock below sea level. This makes it possibly vulnerable to the self-enhancing process of marine ice sheet instability. Marine ice cliff instability could also contribute to a partial collapse. But there is limited evidence for its importance.[114]: 1269–1270 A partial collapse of the ice sheet would lead to rapid sea level rise and a local decrease in ocean salinity. It would be irreversible for decades and possibly even millennia.[115]: 595–596 The complete loss of the West Antarctic ice sheet would cause over 5 metres (16 ft) of sea level rise.[116]
In contrast to the West Antarctic ice sheet, melt of the Greenland ice sheet is projected to take place more gradually over millennia.[115]: 595–596 Sustained warming between 1 °C (1.8 °F) (low confidence) and 4 °C (7.2 °F) (medium confidence) would lead to a complete loss of the ice sheet. This would contribute 7 m (23 ft) to sea levels globally.[117]: 363 The ice loss could become irreversible due to a further self-enhancing feedback. This is called the elevation-surface mass balance feedback. When ice melts on top of the ice sheet, the elevation drops. Air temperature is higher at lower altitudes, so this promotes further melting.[117]: 362The icing of Antarctica began in the Late Palaeocene or middle Eocene between 60[118] and 45.5 million years ago[119] and escalated during the Eocene–Oligocene extinction event about 34 million years ago. CO2 levels were then about 760 ppm[120] and had been decreasing from earlier levels in the thousands of ppm. Carbon dioxide decrease, with a tipping point of 600 ppm, was the primary agent forcing Antarctic glaciation.[121] The glaciation was favored by an interval when the Earth's orbit favored cool summers but oxygen isotope ratio cycle marker changes were too large to be explained by Antarctic ice-sheet growth alone indicating an ice age of some size.[122] The opening of the Drake Passage may have played a role as well[123] though models of the changes suggest declining CO2 levels to have been more important.[124]
While there is evidence of large glaciersinGreenland for most of the past 18 million years,[95] these ice bodies were probably similar to various smaller modern examples, such as Maniitsoq and Flade Isblink, which cover 76,000 and 100,000 square kilometres (29,000 and 39,000 sq mi) around the periphery. Conditions in Greenland were not initially suitable for a single coherent ice sheet to develop, but this began to change around 10 million years ago, during the middle Miocene, when the two passive continental margins which now form the uplands of West and East Greenland experienced uplift, and ultimately formed the upper planation surface at a height of 2000 to 3000 meter above sea level.[127][128]
At first glance this seems to contradict the idea of "global" warming, but one needs to be careful before jumping to this conclusion. A rise in the global mean temperature does not imply universal warming. Dynamical effects (changes in the winds and ocean circulation) can have just as large an impact, locally as the radiative forcing from greenhouse gases. The temperature change in any particular region will in fact be a combination of radiation-related changes (through greenhouse gases, aerosols, ozone and the like) and dynamical effects. Since the winds tend to only move heat from one place to another, their impact will tend to cancel out in the global mean.
Although their methods of interpolation or extrapolation for areas with unobserved output velocities have an insufficient description for the evaluation of associated errors, such errors in previous results (Rignot and others, 2008) caused large overestimates of the mass losses as detailed in Zwally and Giovinetto (Zwally and Giovinetto, 2011).
| |||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| |||||||||||||
| |||||||||||||
| |||||||||||||
| |||||||||||||
| |||||||||||||
| |||||||||||||
| |||||||||||||
|
| |
---|---|
Mountainous |
|
Continental plain |
|
Fluvial |
|
Glacial |
|
Oceanic and coastal landforms |
|
Volcanic |
|
Aeolian |
|
Artificial |
|
|