Greenland ice sheet

Greenland ice sheet
Grønlands indlandsis
TypeIce sheet
Coordinates76°42′N 41°12′W / 76.7°N 41.2°W / 76.7; -41.2
Area1,710,000 km2 (660,000 sq mi)
Length2,400 km (1,500 mi)
Width1,100 km (680 mi)
Thickness1.67 km (1.0 mi) (average), ~3.5 km (2.2 mi) (maximum)
Greenland ice sheet as seen from space

The Greenland ice sheet (Danish: Grønlands indlandsis, Greenlandic: Sermersuaq) is an ice sheet about 1.67 km (1.0 mi) thick on average, and almost 3.5 km (2.2 mi) at its thickest point. It is almost 2,900 kilometres (1,800 mi) long in a north–south direction, with the greatest width of 1,100 kilometres (680 mi) at a latitude of 77°N, near its northern margin. It covers 1,710,000 square kilometres (660,000 sq mi), around 80% of the surface of Greenland, and is the second largest body of ice in the world, after the East Antarctic ice sheet. It is sometimes referred to as an ice cap, or inland ice or its Danish equivalent, indlandsis. The acronyms GIS or GrIS are also frequently used in the scientific literature.

While Greenland has had major glaciers and ice caps for at least 18 million years, the ice sheet first emerged as a single entity covering most of the island some 2.6 million years ago – a process which required significant changes to Greenland's orography millions of years earlier, as well as the low temperatures due to a reduction in atmospheric carbon dioxide levels. Since then it has both grown, sometimes to a significantly larger size than now, and shrunk, to less than 10% of its volume on at least one occasion around 400,000 years ago, when the temperatures were somewhat warmer than today. Its oldest known ice is about 1 million years old.

Normally, the Greenland ice sheet loses some mass from its ice cliffs every year in a process known as ice calving, and parts of its surface melt every summer, but then it recovers a similar amount of ice from snow accumulation during the winter. However, due to greenhouse gas emissions by humans, the ice sheet is now the warmest it's been in at least the past 1000 years, and is melting two to five times as fast in the summer as it did before 1850, so winter gains have not been offsetting ice losses since 1996. This causes the ice sheet to shrink, and to contribute to sea level rise, at rates which will soon become the fastest seen in at least the past 12,000 years.

In 2021, the Intergovernmental Panel on Climate Change estimated that if the less stringent Paris Agreement goal of staying below 2 °C (3.6 °F) is achieved, than the melting of Greenland ice adds around 6 cm (2+12 in) to global sea level rise on its own by the end of the century, while it would add around 13 cm (5 in) (and potentially up to 23 cm (9 in)) if there are no efforts to reduce emissions.: 1302  For comparison, it has so far contributed 13.7 mm since 1972, while the sea level rise from all sources was 15–25 cm (6–10 in)) between 1901 and 2018.: 5 

If the ice sheet's entire volume of 2,900,000 cubic kilometres (696,000 cu mi) were to melt, it would increase the global sea levels by ~7.42 m (24 ft) all on its own. This kind of melting is expected to take millennia to fully play out, with a significant fraction of the ice sheet remaining in 2300 under any climate change scenario. However, the ice sheet has a prolonged inertia in its response to temperature changes, and it is already anticipated to eventually lose at least 27 cm (10+12 in) of sea level rise equivalent irrespective of any future temperature change. Paleoclimate evidence suggests that in the long run, global warming of around 1.5 °C (2.7 °F) causes Greenland to lose enough ice to increase the sea levels by 1.4 m (4+12 ft).

2012 estimates suggested that the global warming threshold which would commit the ice sheet to melting entirely is uncertain, and may lie anywhere between 0.8 °C (1.4 °F) and 3.2 °C (5.8 °F). More recent research has narrowed this threshold to a 1.7 °C (3.1 °F)-2.3 °C (4.1 °F) range. It also indicates that reducing the global temperature to 1.5 °C (2.7 °F) or lower (i.e. through large-scale carbon dioxide removal) will prevent the loss of the entire ice sheet, yet also result in a greater sea level rise contribution than if the threshold was never breached in the first place.


A narrated tour about Greenland's ice sheet.

Ice sheets are formed through a process of glaciation, when the local climate is so cold that snow regularly falls yet never melts entirely, causing its layers to pile up onto each other, the pressure of this steadily growing weight compressing snow into solid ice over thousands of years.

The pattern of ice flows at the Greenland ice sheet, with arrows pointing to its outlet glaciers where ice calving occurs

Once the ice sheet was formed, it generally remained of a similar size to where it is now for most of its history. The massive combined weight of causes it to slowly "flow", unless it is stopped by a sufficiently large obstacle, such as a mountain. The terrain of Greenland has many mountains near its coastline, which usually prevent the ice sheet from flowing out into the Arctic Ocean across most of its periphery. However, there were 11 times in its history where the ice had gotten large to push past them, extending up to 120 km (75 mi) past its current boundaries, before it seemingly lost the capability to do that around 1 million years ago.

Nowadays, the exceptions where the ice sheet still reaches the ocean mainly occur in its northwest and southeast, primarily in the form of the so-called outlet glaciers, which flow out through gaps in the mountains and regularly shed ice in what is known as ice calving. Some of that calved ice sinks into the sediment, and can be preserved for a very long time, with sediment cores from places such as the Fram Strait providing some of the longest record of glaciation at Greenland.

The east coast of Greenland, observed from the air
Glaciologist at work

Besides providing crucial information about the past states of the ice sheet and its impact on sea level rise, ice cores are invaluable for other kinds of paleoclimate research as well. The subtle differences in isotope distributions of ice core's water molecules can reveal important information about the water cycle at the time, and air bubbles frozen within the ice core provide a snapshot of the lower atmosphere, detailing the gas and particulate composition it used to have.

When properly analyzed, they provide a wealth of proxies suitable for reconstructing the past temperature record, precipitation patterns, volcanic eruptions, solar variation, ocean primary production, and even changes in soil vegetation cover and the associated wildfire frequency. The ice cores from Greenland also record human impact, such as lead production during the time of Ancient Greece and the Roman Empire.

In geologic timescales

Timeline of the ice sheet's formation from 2.9 to 2.6 million years ago

While there is evidence of large glaciers in Greenland for most of the past 18 million years, they were more similar to various smaller modern formations, 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. The conditions in Greenland were not initially suitable to enable the presence of a single cohesive ice sheet, 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 had experienced uplift for the first time, which ultimately formed the Upper Planation Surface at a 2000 to 3000 meter height above mean sea level.

Later, during the Pliocene, a Lower Planation Surface, with the 500 to 1000 meter height above sea level, was formed during the second stage of uplift 5 million years ago, and the third stage had created multiple valleys and fjords below the planation surfaces. These increases in height had intensified glaciation due to increased orographic precipitation and cooler surface temperatures, which made it easier for ice to accumulate during colder periods and persist through higher temperature fluctuations. While as recently as 3 million years ago, during the Pliocene warm period, Greenland's ice was limited to the highest peaks in the east and the south, ice cover had gradually expanded since then, until the atmospheric CO2 levels dropped to between 280 and 320 ppm 2.7–2.6 million years ago, which had reduced the temperatures sufficiently for the disparate ice caps build up in the meantime to connect and cover most of the island.

For much of the past 120,000 years, the climate in and around Greenland had been colder than in the last few millennia of recorded history (upper half), allowing the ice sheet to become considerably larger than it is now (lower half).

Often, the base of ice sheet is warm enough due to geothermal activity to have some liquid water beneath it. This liquid water, subject to great pressure from the continued movement of massive layers of ice above it, becomes a tool of intense water erosion, which eventually leaves nothing but bedrock below the ice sheet. However, there are parts of the Greenland ice sheet, near the summit, where the upper layers of the ice sheet slide above the lowest layer of ice which had frozen solid to the ground, preserving ancient soil, which can then be discovered when scientists drill ice cores, up to 4 kilometres (2.5 mi) deep. The oldest such soil had been continuously covered by ice for around 2.7 million years, while another, 3 kilometres (1.9 mi) deep ice core from the summit reveals ice that is around ~1,000,000 years old.

On the other hand, ocean sediment samples from the Labrador Sea provide evidence that nearly all of south Greenland had melted around 400,000 years ago, during the Marine Isotope Stage 11, and other ice core samples, taken from Camp Century in northwestern Greenland at a depth of 1.4 km (0.87 mi), demonstrate that the ice there melted at least once during the past 1.4 million years, during the Pleistocene, and that it did not return for at least 280,000 years. Taken together, these findings suggest less than 10% of the current ice sheet's volume was left during those geologically recent periods, when the temperatures were less than 2.5 °C (4.5 °F) warmer than preindustrial, which contradicts how climate models typically simulate continuous presence of solid ice under those conditions.

Recent melting

Arctic temperature trend, 1981–2007

In the earlier decades, an area in the North Atlantic which included southern Greenland was one of the only locations in the world which showed cooling rather than warming, as it was already unusually warm in the 1930s and 1940s than it was in the decades immediately before and after. However, later and more complete data sets have established both a trend of warming and ice loss starting from 1900(well after the start of the Industrial Revolution and its impact on global temperatures) and a trend of strong warming starting around 1979, in line with the concurrently observed Arctic sea ice decline and its role in arctic amplification due to ice-albedo feedback. Consistent with this warming, 1970s were the last decade when the Greenland ice sheet had grown, gaining about 47 gigatonnes per year, while the 1980–1990 period already had an average annual mass loss of ~51 Gt/y. 1990–2000 period had smaller average annual loss of 41 Gt/y, as 1996 was the last time when the Greenland ice sheet saw net mass gain. As of 2022, it had been losing ice for 26 years in a row, and the temperatures there had been the highest in the entire past last millennium – about 1.5 °C (2.7 °F) warmer than the average of the 20th century.

Several factors determine the net rate of ice sheet growth or decline. These are:

  • Accumulation and melting rates of snow in and around the centre
  • Melting of ice along the sheet's margins, where it then runs off into the sea before it can refreeze during the winter
  • Ice calving into the sea from outlet glaciers also along the sheet's edges
Until 2007, rate of decrease in ice sheet height in cm per year
Trends of ice loss between 2002 and 2019

When the IPCC Third Assessment Report was published in 2001, its analysis of observations to date had shown that the ice accumulation of 520 ± 26 gigatonnes per year was getting offset by the runoff and bottom melting equivalent to ice losses of 297±32 Gt/yr and 32±3 Gt/yr, as well as iceberg production of 235±33 Gt/yr, with the net loss of −44 ± 53 gigatonnes per year.

Annual ice losses from the Greenland ice sheet had more than quadrupled in 2000s, going from 41 Gt/y in the 1980–1990 to ~187 Gt/y in 2000–2010. The losses worsened at a slower rate in 2010s: the average mass loss during 2010–2018 of 286 Gt per year – which meant that half of the ice sheet's observed net loss (3,902 gigatons (Gt) of ice between 1992 and 2018, or approximately 0.13% of its total mass) happened during those 8 years. When these losses are converted to sea level rise equivalent, Greenland Ice Sheet contributed about 13.7 mm since 1972, with 4.4 mm from its northwest, 3 mm from its southeast and 2 mm from central west.

Between 2012 and 2017, it had contributed 0.68 mm per year, as opposed to 0.07 mm per year between 1992 and 1997. Its net contribution for the 2012–2016 period was also equivalent to 37% of sea level rise from land ice sources (excluding thermal expansion). These melt rates are comparable to the largest experienced by the ice sheet over the past 12,000 years of its Holocene history, and they will inevitably be exceeded later in this century.

Currently, the Greenland ice sheet is losing more mass every year than the Antarctic ice sheet, because of its position in the Arctic, where it is subject to far more intense regional amplification of warming. However, ice losses from the West Antarctic Ice Sheet have been accelerating at a greater rate due to its uniquely vulnerable Thwaites and Pine Island Glaciers, and its contribution to sea level rise is expected to overtake that of Greenland later this century.

Observed glacier retreat

This narrated animation shows the overall change in the elevation of the Greenland ice sheet between 2003 and 2012. It can be seen that the coastal areas of the ice sheet had lost far more height, or "thinned", compared to the more inland regions.
Greenland ice sheet has 215 marine-terminating glaciers whose retreat directly impacts sea level rise. As of 2021, 115 accounted for 79% of ice flow and could be simulated with good accuracy, 25 had their retreat underestimated and accounted for 13%, 67 lacked sufficient bathymetry surveys while accounting for 5% of the flow, and 8 had their retreat overestimated, accounting for the remaining 3%.

Retreat of outlet glaciers as they shed more and more ice into the arctic waters is a large, or even dominant, factor in the decline of Greenland's ice sheet. Some analyses estimate that losses from glaciers explain 66.8% of the observed ice loss since the 1980s, but others place it at 49%, with the rest accounted for by surface melting. Net loss of ice was already observed across 70% of the ice sheet's coasts in the 1990s: scientific literature commonly described it as "thinning", since the glaciers started to lose height, and thus form a thinner layer over the bedrock. Between 1998 and 2006, thinning occurred four times faster for coastal glaciers compared to the early 1990s, falling at rates between 1 m (3+12 ft) and 10 m (33 ft) per year, while the landlocked glaciers experienced almost no such acceleration.

One of the most dramatic examples of such thinning took place in the southeast, at Kangerlussuaq Glacier. It is over 20 mi (32 km) long, 4.5 mi (7 km) wide and around 1 km (12 mi) thick, which makes it the third largest glacier in Greenland. Between 1993 and 1998, parts of the glacier within 5 km (3 mi) of the coast lost 50 m (164 ft) in height. Later, its observed ice flow speed went from 3.1–3.7 mi (5–6 km) per year for 1988-1995 to 8.7 mi (14 km) In 2005, which was then the fastest known flow of any glacier. The retreat of Kangerlussuaq slowed down by 2008, and its position even experienced some recovery until 2016-2018, when even more rapid ice loss had occurred.

Greenland's other major outlet glaciers had also experienced rapid and dramatic change in the recent decades. Its single largest outlet glacier is Jakobshavn Isbræ (Greenlandic: Sermeq Kujalleq) in west Greenland, which has been observed by glaciologists for many decades, as it historically sheds ice outflow from 6.5% of the ice sheet (compared to 4% for Kangerlussuaq), at speeds of ~20 metres (66 ft) per day. While it had already lost enough ice to retreat around 30 km (19 mi) between 1850 and 1964, its mass gain increased sufficiently to keep it in balance for the next 35 years, only to switch to rapid mass loss after 1997. By 2003, its average annual ice flow speed had almost doubled since 1997, as the ice tongue in front of the glacier which used to impede ice flows had disintegrated,, and the glacier shed 94 square kilometres (36 sq mi) of ice between 2001 and 2005. The ice flow hit a record of 45 metres (148 ft) per day in 2012, but slowed down substantially afterwards, to the point of experiencing mass gain between 2016 and 2019.

On the other hand, northern Greenland's Petermann Glacier is smaller in absolute terms, yet it had experienced some of the most rapid degradation in recent decades: a loss of 85 square kilometres (33 sq mi) of floating ice in 2000-2001, followed by a 28-square-kilometre (11 sq mi) iceberg breaking off in 2008, and then a 260 square kilometres (100 sq mi) iceberg calving from ice shelf in August 2010, which became the largest Arctic iceberg since 1962, and amounted to a quarter of the shelf's size. In July 2012, Petermann glacier had experienced the loss of another major iceberg measuring 120 square kilometres (46 sq mi), or twice the area of Manhattan. As of 2023, the glacier's ice shelf lost around 40% of its pre-2010 state, and it is considered unlikely to recover from further ice loss.

In the early 2010s, some estimates suggested that tracking the largest glaciers would be suffficient to account for most of the ice loss. However, their dynamics can be hard to predict, like with the ice sheet's second largest glacier, Helheim Glacier. Its ice loss culminated in rapid retreat in 2005, and was also associated with a marked increase in glacial earthquakes between 1993 and 2005.Since then, it had remained comparatively stable near its 2005 position and lost relatively little mass in comparison to Jakobshavn and Kangerlussuaq, although it might have eroded sufficiently by 2021 to experience another rapid retreat in the near future. Meanwhile, smaller glaciers had been consistently losing mass at an accelerating rate, and later research concluded that total glacier retreat is underestimated when the dynamics of largest glaciers are extrapolated without explicitly calculating the smaller glaciers. By 2023, the rate of ice loss across Greenland's coasts had doubled in the two decades since 2000, in large part due to the accelerated losses from smaller glaciers.

Processes accelerating glacier retreat

Petermann Glacier experiences notable shifts from year to year not just at its calving front, but also at its grounding line, which renders it less stable. If such behaviour turns out to be widespread at other glaciers, this could potentially double their rates of ice loss.

Starting from the early 2000s, glaciologists have concluded that glacier retreat in Greenland was accelerating too quickly to be explained by a linear increase in melting in response to greater surface temperatures alone, and that additional non-linear mechanisms must also be at work. The rapid calving events at the largest glaciers match what was first described as the "Jakobshavn effect" in 1986: thinning causes the glacier to be more buoyant, reducing friction that would otherwise impede its retreat, and also results in a force imbalance at the calving front, with the increase in velocity spread across the mass of the glacier. The overall acceleration of Jakobshavn Isbrae and other glaciers from 1997 onwards had been attributed to the warming of North Atlantic waters which melt the glacier fronts from underneath: while this warming had been going on since the 1950s, 1997 also saw a shift in circulation which brought relatively warmer currents from the Irminger Sea into closer contact with the glaciers of West Greenland. By 2016, waters across much of West Greenland's coastline had warmed by 1.6 °C (2.9 °F) relative to 1990s, and some of the smaller glaciers were losing more ice to such melting than normal calving processes, leading to rapid retreat.

Conversely, Jakobshavn Isbrae is sensitive to changes in ocean temperature as it experiences elevated exposure through a deep subglacial trench, yet this sensitivity also meant that a sudden influx of cooler currents to its location had been responsible for its equally sudden slowdown after 2015, in large part because the sea ice and icebergs immediately off-shore were able to survive for longer and thus help to stabilize the glacier. Likewise, rapid retreat, then slowdown of Helheim in northwest and Kangerdlugssuaq in the east had also been connected to the respective warming and cooling of nearby currents. At Petermann Glacier, its rapid rate of retreat had been linked to the topography of its grounding line, which appears to shift back and forth by around a kilometer with the tide: it has been suggested that if similar processes can occur at the other glaciers, then their eventual rate of mass loss could be doubled.

Meltwater rivers may flow down into moulins and reach the base of the ice sheet

Research has shown that there are also several ways in which increased melting at the surface of the ice sheet can also accelerate lateral retreat of outlet glaciers. Firstly, the increase in meltwater at the surface causes larger amounts to flow all the way through the ice sheet down to bedrock via moulins. There, its presence lubricates the base of glaciers and generates higher basal pressure, which collectively reduces friction and accelerates glacial motion, including the rate of ice calving. This mechanism was observed at Sermeq Kujalleq in 1998 and 1999, where its flow was accelerated by up to 20% for two-three months. However, subsequent research had shown that this mechanism only applies to certain small glaciers, rather than to the largest outlet glaciers, and has only a "marginal" impact on ice loss trends.

An illustration of how meltwater forms a plume once it flows out into the ocean

Secondly, once meltwater flows into the ocean, it can still impact the glaciers by interacting with ocean water and altering its local circulation - even in the absence of any ocean warming. In certain fjords, large meltwater flows from beneath the ice may mix with ocean water to create turbulent plumes that can be very damaging to the calving front. While the models generally consider such the impact from meltwater run-off deeply secondary to ocean warming, observations of 13 glaciers found that meltwater plumes play a greater role for glaciers with shallow grounding lines. Further, 2022 research suggests that the warming from plumes had a greater impact on underwater melting across the entire northwest Greenland, with only south Greenland definitely affected by changes in ocean currents more than by the impact of local warming on its own meltwater.

Finally, it has been shown that in addition to major moulins, meltwater can also flow through a large number of cracks that are too small to be picked up by most research tools - only 2 cm (1 in) wide. Such cracks do not connect to bedrock through the entire ice sheet but may still reach several hundred meters down from the surface. Their presence is important, as it weakens the ice sheet, and the meltwater inside them also conducts more heat directly through the ice, making it more viscous and thus allowing it to flow faster. As this research is recent, it is not currently captured in the models. One of the scientists behind these findings, Alun Hubbard, described finding moulins where "current scientific understanding doesn’t accommodate" their presence, because it disregards how they may evolve from such hairline cracks even in the absence of existing large crevasses that are normally thought to be necessary for their formation.

Observed surface melting

Satellite measurements of Greenland's ice cover from 1979 to 2009 reveals a trend of increased melting.
NASA's MODIS and QuikSCAT satellite data from 2007 were compared to confirm the precision of different melt observations.

Currently, the total accumulation of ice on the surface of Greenland ice sheet remains larger than either outlet glacier losses individually or surface melting during the summer, and it is the combination of both which causes net annual loss. Every summer, a so-called snow line separates the ice sheet's surface into areas above it, where snow continues to accumulate even then, with the areas below the line where summer melting occurs. Notably, the exact position of the snow line moves around every summer, and if it moves away from some areas it covered the previous year, then those tend to experience substantially greater melt as their darker ice is exposed. In this way, the uncertainty about the snow line is one of the factors making it hard to predict each melting season in advance.

Satellite image of dark melt ponds

A notable example of ice accumulation rates above the snow line is provided by Glacier Girl, a Lockheed P-38 Lightning fighter plane which had crashed early in World War II and was recovered in 1992, by which point it had been buried under 268 ft (81+12 m) of ice. Another example occurred in 2017, when an Airbus A380 had to make an emergency landing in Canada after one of its jet engines exploded while it was above Greenland; the engine's massive air intake fan was recovered from the ice sheet two years later, when it was already buried beneath 4 ft (1 m)of ice and snow.

While summer surface melting had been increasing, it is still expected that it'll be decades before it will consistently exceed snow accumulation on its own. It had also been hypothesized that the increase in global precipitation associated with the effects of climate change on the water cycle would also increase snowfall over Greenland, and thus further delay this transition. This hypothethis had been difficult to test in the 2000s due to the poor state of long-term precipitation records over the ice sheet. By 2019, it was found that while there was an increase in snowfall over southwest Greenland, there had been a substantial decrease in precipitation over western Greenland as a whole. Further, more precipitation in the northwest had been falling as rain (which is warmer and forms darker and less thermally insulating ice layer once it freezes) instead of snow, with a fourfold increase since 1980. Rain is particularly damaging to the ice sheet when it falls due to late-summer cyclones, whose increasing occurrence had been overlooked by the earlier models. There had also been an increase in water vapor, which paradoxically increases melting by making it easier for heat to radiate downwards through moist, as opposed to dry, air.

NASA graphics show the extent of the then-record melting event in July 2012.

Altogether, the melt zone below the snow line, where summer warmth turns snow and ice into slush and melt ponds, has been expanding at an accelerating rate since the beginning of detailed measurements in 1979. By 2002, its area was found to have increased by 16% since 1979, and the annual melting season broke all previous records. Another record was set in July 2012, when the melt zone extended to 97% of the ice sheet's cover, and the ice sheet lost approximately 0.1% of its total mass (2900 Gt) during that year's melting season, with the net loss (464 Gt) setting another record. It became the first directly observed example of a "massive melting event", when the melting took place across practically the entire ice sheet surface, rather than specific areas. That event led to the counterintuitive discovery that cloud cover, which normally results in cooler temperature due to their albedo, actually interferes with meltwater refreezing in the firn layer at night, which can increase total meltwater runoff by over 30%. Thin, water-rich clouds have the worst impact, and they were the most prominent in July 2012.

Rivers of meltwater flowing on 21 July 2012.

Ice cores had also shown that the last time a melting event of the same magnitude as in 2012 took place was in 1889, and some glaciologists had expressed hope that 2012 was part of a 150-year cycle. This was disproven in 2019, when a combination of high temperatures and unsuitable cloud cover led to an even larger mass melting event over both June and July, which ultimately covered over 300,000 mi (482,803.2 km) at its greatest extent. Predictably, 2019 set a new record of 586 Gt net mass loss. In July 2021, another record mass melting event occurred. At its peak, it covered 340,000 mi (547,177.0 km), and led to daily ice losses of 88 Gt across several days. High temperatures continued in August 2021, with the melt extent staying at 337,000 mi (542,348.9 km). At that time, rain fell for 13 hours at Greenland's Summit Station, located at 10,551 ft (3,215.9 m) elevation. Researchers had no rain gauges to measure the rainfall, because temperatures at the summit have risen above freezing only three times since 1989 and it had never rained there before.

Due to the enormous thickness of the central Greenland ice sheet, even the most extensive melting event can only affect a small fraction of it before the start of the freezing season, and so they are considered "short-term variability" in the scientific literature. Nevertheless, their existence is important: the fact that the current models underestimate the extent and frequency of such events is considered to be one of the main reasons why the observed ice sheet decline in Greenland and Antarctica tracks the worst-case rather than the moderate scenarios of the IPCC Fifth Assessment Report's sea-level rise projections. Some of the most recent scientific projections of Greenland melt now include an extreme scenario where a massive melting event occurs every year across the studied period (i.e. every year between now and 2100 or between now and 2300), to illustrate that such a hypothetical future would greatly increase ice loss, but still wouldn't melt the entire ice sheet within the study period.

Changes in albedo

Albedo change in Greenland

On the ice sheet, annual temperatures are generally substantially lower than elsewhere in Greenland: about −20 °C (−4 °F) at the south dome (latitudes 63°65°N) and −31 °C (−24 °F) near the center of the north dome (latitude 72°N (the fourth highest "summit" of Greenland). On 22 December 1991, a temperature of −69.6 °C (−93.3 °F) was recorded at an automatic weather station near the topographic summit of the Greenland Ice Sheet, making it the lowest temperature ever recorded in the Northern Hemisphere. The record went unnoticed for more than 28 years and was finally recognized in 2020. These low temperatures are in part caused by the high albedo of the ice sheet, as its bright white surface is very effective at reflecting sunlight. Ice-albedo feedback means that as the temperatures increase, this causes more ice to melt and either reveal bare ground or even just to form darker melt ponds, both of which act to reduce albedo, which accelerates the warming and contributes to further melting. This is taken into account by the climate models, which estimate that a total loss of the ice sheet would increase global temperature by 0.13 °C (0.23 °F), while Greenland's local temperatures would increase by between 0.5 °C (0.90 °F) and 3 °C (5.4 °F).

Even incomplete melting already has some impact on the ice-albedo feedback. Besides the formation of darker melt ponds, warmer temperatures enable increasing growth of algae on the ice sheet's surface. Mats of algae are darker in colour than the surface of the ice, so they absorb more thermal radiation and increase the rate of ice melt. In 2018, it was found that the regions covered in dust, soot, and living microbes and algae altogether grew by 12% between 2000 and 2012. In 2020, it was demonstrated that the presence of algae, which is not accounted for by ice sheet models unlike soot and dust, had already been increasing annual melting by 10–13%. Additionally, as the ice sheet slowly gets lower due to melting, surface temperatures begin to increase and it becomes harder for snow to accumulate and turn to ice, in what is known as surface-elevation feedback.

Geophysical and biochemical role of Greenland's meltwater

Meltwater runoff has the greatest positive effect on phytoplankton when it can force nitrate-rich waters to the surface (image B), which will become more difficult as the glaciers retreat (image D).

Even in 1993, Greenland's melt resulted in 300 cubic kilometers of fresh meltwater entering the seas annually, which was substantially larger than the liquid meltwater input from the Antarctic ice sheet, and equivalent to 0.7% of freshwater entering the oceans from all of the world's rivers. This meltwater is not pure, and contains a range of elements - most notably iron, about half of which (around 0.3 million tons every year) is bioavailable as a nutrient for phytoplankton. Thus, meltwater from Greenland enhances ocean primary production, both in the local fjords, and further out in the Labrador Sea, where 40% of the total primary production had been attributed to nutrients from meltwater. Since the 1950s, the acceleration of Greenland melt caused by climate change has already been increasing productivity in waters off the North Icelandic Shelf, while productivity in Greenland's fjords is also higher than it had been at any point in the historical record, which spans from late 19th century to present. However, some research suggests that Greenland's meltwater mainly benefits marine productivity not by adding its carbon and iron, but through stirring up lower water layers that are rich in nitrates and thus bringing more of those crucial nutrients to phytoplankton on the surface. As the outlet glaciers retreat inland, their meltwater will be less able to impact the lower layers, which implies that benefit from their meltwater will diminish even as its volume will grow in absolute terms.

The impact of meltwater from Greenland goes beyond nutrient transport. For instance, meltwater also contains dissolved organic carbon, which comes from the microbial activity on the ice sheet's surface, and, to a lesser extent, from the remnants of ancient soil and vegetation beneath the ice. While the overall quantities of this carbon are relatively limited (between 0.5 and 27 billion tonnes of pure carbon underneath the entire ice sheet, and much less within it,, as opposed to 1400–1650 billion tonnes for the Arctic permafrost, or the annual anthropogenic emissions of around 40 billion tonnes of CO2: 1237 ) its release through meltwater can still lead to increased carbon dioxide emissions, thus acting as a climate change feedback. There is one known area, at Russell Glacier, where meltwater carbon is released into the atmosphere as methane, which has a much larger global warming potential than carbon dioxide: however, it also harbours large numbers of methanotrophic bacteria, which limit those emissions.

The cold blob visible on NASA's global mean temperatures for 2015, the warmest year on record up to 2015 (since 1880). Colors indicate temperature evolution (NASA/NOAA; 20 January 2016).

Further, there is a risk of toxic waste being released from Camp Century, formerly a United States military] site secretly built to carry nuclear weapons for the Project Iceworm. The project was cancelled, but the site was never cleaned up, and it now threatens to eventually pollute the meltwater with nuclear waste, 20,000 liters of chemical waste and 24 million liters of untreated sewage as the melt progresses. Finally, research in 2021 discovered that bedrock beneath southwest Greenland contains a lot of mercury, which is a highly toxic heavy metal, and that it is now getting released through meltwater run-off into the local fjords. Mercury concentrations in southwest Greenland meltwater are 165 times higher than in any Arctic river, and the combined quantity of mercury across that meltwater catchment may be equivalent to 5-10% of all mercury in all of the world's rivers.

Finally, increased quantities of fresh meltwater can affect ocean circulation. Some scientists have connected this increased discharge from Greenland with the so-called cold blob in the North Atlantic, which is in turn connected to Atlantic meridional overturning circulation, or AMOC, and its apparent slowdown. In 2016, a study attempted to improve forecasts of future AMOC changes by incorporating better simulation of Greenland trends into projections from eight state-of-the-art climate models. That research found that by 2090–2100, the AMOC would weaken by around 18% (with a range of potential weakening between 3% and 34%) under Representative Concentration Pathway 4.5, which is most akin to the current trajectory, while it would weaken by 37% (with a range between 15% and 65%) under Representative Concentration Pathway 8.5, which assumes continually increasing emissions. If the two scenarios are extended past 2100, then the AMOC ultimately stabilizes under RCP 4.5, but it continues to decline under RCP 8.5: the average decline by 2290–2300 is 74%, and there is 44% likelihood of an outright collapse in that scenario, with a wide range of adverse effects.

Future ice loss


Greenland ice sheet's impact on sea level rise under the worst-case warming scenario, by 2300.
By the year 2300, enough of Greenland's ice would melt to add ~3 m (10 ft) to sea levels under RCP8.5, the worst possible climate change scenario. Currently, RCP8.5 is considered much less likely than RCP 4.5, which lies in between the worst-case and the Paris Agreement goals.
Sea level rise from all sources by the year 2300, under different climate scenarios.
If countries cut greenhouse gas emissions significantly (lowest trace), then sea level rise by 2100 can be limited to 0.3–0.6 m (1–2 ft). If the emissions instead accelerate rapidly (top trace), sea levels could rise 5 m (16+12 ft) by the year 2300, which would include ~3 m (10 ft) caused by the melting of the Greenland ice sheet shown on the left.

In 2021, the IPCC Sixth Assessment Report estimated that under SSP5-8.5, the scenario associated with the highest global warming, Greenland ice sheet melt would add around 13 cm (5 in) to the global sea levels (with a likely (17%–83%) range of 9–18 cm (3+12–7 in) and a very likely range (5–95% confidence level) of 5–23 cm (2–9 in)), while the "moderate" SSP2-4.5 scenario adds 8 cm (3 in) with a likely and very likely range of 4–13 cm (1+12–5 in) and 1–18 cm (12–7 in), respectively. The optimistic scenario which assumes that the Paris Agreement goals are largely fulfilled, SSP1-2.6, adds around 6 cm (2+12 in) and no more than 15 cm (6 in), with a small chance of the ice sheet gaining mass and thus reducing the sea levels by around 2 cm (1 in).: 1260 

There are a few scientists, mainly led by James Hansen, who have long claimed that the ice sheets can disintegrate substantially faster than estimated by the ice sheet models, but even their projections also have much of Greenland, whose total size amounts to 7.4 m (24 ft) of sea level rise, survive the 21st century. I.e. a 2016 paper from Hansen claimed that Greenland ice loss could add around 33 cm (13 in) by 2060, in addition to double that figure from the Antarctic ice sheet, if the CO2 concentration exceeded 600 parts per million, which was immediately controversial amongst the scientific community, while 2019 research from different scientists claimed a maximum of 33 cm (13 in) by 2100 under the worst-case climate change scenario.

Projections of 21st century retreat for Greenland's largest glaciers

As with the present losses, not all parts of the ice sheet would contribute to them equally. For instance, it is estimated that on its own, the Northeast Greenland Ice Stream would contribute 1.3–1.5 cm by 2100 under RCP 4.5 and RCP 8.5, respectively. On the other hand, the three largest glaciers - Jakobshavn, Helheim, and Kangerlussuaq - are all located in the southern half of the ice sheet, and just the three of them are expected to add 9.1–14.9 mm under RCP 8.5. Similarly, 2013 estimates suggested that by 2200, they and another large glacier would add 29 to 49 millimetres by 2200 under RCP 8.5, or 19 to 30 millimetres under RCP 4.5. Altogether, the single largest contribution to 21st century ice loss in Greenland is expected to be from the northwest and central west streams (the latter including Jakobshavn), and glacier retreat will be responsible for at least half of the total ice loss, as opposed to earlier studies which suggested that surface melting would become dominant later this century. If Greenland were to lose all of its coastal glaciers, though, then or not it will continue to shrink will be entirely determined by whether its surface melting in the summer consistently outweighs ice accumulation during winter. Under the highest-emission scenario, this could happen around 2055, well before the coastal glaciers are lost.

It should also be noted that the sea level rise from Greenland does not affect every coast equally. The south of the ice sheet is much more vulnerable than the other parts, and the quantities of ice involved mean that there is an impact on the deformation of Earth's crust and on Earth's rotation. While this effect is subtle, it already causes East Coast of the United States to experience faster sea level rise than the global average. At the same time, Greenland itself would experience isostatic rebound as its ice sheet shrinks and its ground pressure becomes lighter. Similarly, a reduced mass of ice would exert a lower gravitational pull on the coastal waters relative to the other land masses. These two processes would cause sea level around Greenland's own coasts to fall, even as it rises elsewhere. The opposite of this phenomenon happened when the ice sheet gained mass during the Little Ice Age: increased weight attracted more water and flooded certain Viking settlements, likely playing a large role in the Viking abandonment soon afterwards.


These graphs indicate the switch of peripheral glaciers to a dynamic state of sustained mass loss after the widespread retreat in 2000–2005, making their disappearance inevitable.
2023 projections of how much the Greenland ice sheet may shrink from its present extent by the year 2300 under the worst possible climate change scenario (upper half) and of how much faster its remaining ice will be flowing in that case (lower half)

Notably, the ice sheet's massive size simultaneously makes it insensitive to temperature changes in the short run, yet also commits it to enormous changes down the line, as demonstrated by paleoclimate evidence. Polar amplification causes the Arctic, including Greenland, to warm three to four times more than the global average: thus, while a period like the Eemian interglacial 130,000–115,000 years ago was not much warmer than today globally, the ice sheet was 8 °C (14 °F) warmer, and its northwest part was 130 ± 300 meters lower than it is at present. Some estimates suggest that the most vulnerable and fastest-receding parts of the ice sheet have already passed "a point of no return" around 1997, and will be committed to disappearance even if the temperature stops rising.

A 2022 paper found that the 2000–2019 climate would already result in the loss of ~3.3% volume of the entire ice sheet in the future, committing it to an eventual 27 cm (10+12 in) of SLR, independent of any future temperature change. They have additionally estimated that if the then-record melting seen on the ice sheet in 2012 were to become its new normal, then the ice sheet would be committed to around 78 cm (30+12 in) SLR. Another paper suggested that paleoclimate evidence from 400,000 years ago is consistent with ice losses from Greenland equivalent to at least 1.4 m (4+12 ft) of sea level rise in a climate with temperatures close to 1.5 °C (2.7 °F), which are now inevitable at least in the near future.

It is also known that at a certain level of global warming, effectively the entirety of the Greenland's ice sheet will eventually melt. Its volume was initially estimated to amount to ~2,850,000 km3 (684,000 cu mi), which would increase the global sea levels by 7.2 m (24 ft), but later estimates increased its size to ~2,900,000 km3 (696,000 cu mi), leading to ~7.4 m (24 ft) of sea level rise.

Thresholds for total ice sheet loss

In 2006, it was estimated that the ice sheet is most likely to be committed to disappearance at 3.1 °C (5.6 °F), with a plausible range between 1.9 °C (3.4 °F) and 5.1 °C (9.2 °F). However, these estimates were drastically reduced in 2012, with the suggestion that the threshold may lie anywhere between 0.8 °C (1.4 °F) and 3.2 °C (5.8 °F), with 1.6 °C (2.9 °F) the most plausible global temperature for the ice sheet's disappearance. That lowered temperature range had been widely used in the subsequent literature, and in the year 2015, prominent NASA glaciologist Eric Rignot claimed that "even the most conservative people in our community" will agree that "Greenland’s ice is gone" after 2 °C (3.6 °F) or 3 °C (5.4 °F) of global warming.

In 2022, a major review of scientific literature on tipping points in the climate system had barely modified these values: it suggested that the threshold would be most likely be at 1.5 °C (2.7 °F), with the upper level at 3 °C (5.4 °F) and the worst-case threshold of 0.8 °C (1.4 °F) remained unchanged. At the same time, it noted that the fastest plausible timeline for the ice sheet disintegration is 1000 years, which is based on research assuming the worst-case scenario of global temperatures exceeding 10 °C (18 °F) by 2500, while its ice loss otherwise takes place over around 10,000 years after the threshold is crossed; the longest possible estimate is 15,000 years.

Potential equilibrium states of the ice sheet in response to different equilibrium carbon dioxide concentrations in parts per million. Second and third states would result in 1.8 m (6 ft) and 2.4 m (8 ft) of sea level rise, while the fourth state is equivalent to 6.9 m (23 ft).

Model-based projections published in the year 2023 had indicated that the Greenland ice sheet could be a little more stable than suggested by the earlier estimates. One paper found that the threshold for ice sheet disintegration is more likely to lie between 1.7 °C (3.1 °F) and 2.3 °C (4.1 °F). It also indicated that the ice sheet could still be saved, and its sustained collapse averted, if the warming were reduced to below 1.5 °C (2.7 °F), up to a few centuries after the threshold was first breached. However, while that would avert the loss of the entire ice sheet, it would increase the overall sea level rise by up to several meters, as opposed to a scenario where the warming threshold was not breached in the first place.

Another paper used a more complex ice sheet model with more detailed calculations than the previous, more abstracted studies, and it found that ever since the warming passed 0.6 °C (1.1 °F) degrees, ~26 cm (10 in) of sea level rise became inevitable, which closely matched the estimate derived from direct observation in 2022. However, it had also found that 1.6 °C (2.9 °F) would likely only commit the ice sheet to 2.4 m (8 ft) of long-term sea level rise, while near-complete melting of 6.9 m (23 ft) worth of sea level rise would occur if the temperatures consistently stay above 2 °C (3.6 °F). The paper also suggested that ice losses from Greenland may be reversed by reducing temperature to 0.6 °C (1.1 °F) or lower, up until the entirety of South Greenland ice melts, which would cause 1.8 m (6 ft) of sea level rise and prevent any regrowth unless CO2 concentrations is reduced to 300 ppm. If the entire ice sheet were to melt, it would not begin to regrow until temperatures fall to below the preindustrial levels.

See also

This page was last updated at 2023-12-25 04:01 UTC. Update now. View original page.

All our content comes from Wikipedia and under the Creative Commons Attribution-ShareAlike License.


If mathematical, chemical, physical and other formulas are not displayed correctly on this page, please useFirefox or Safari