← all systems · see it in The System
Ice sheet
A mass of glacial ice that covers surrounding terrain of continental size, greater than 50,000 km2; the largest glacial bodies on Earth, flowing outward from domes and drained by ice streams and outlet glaciers. Today only the Antarctic and Greenland ice sheets exist.
Properties published scienceour proposal
Sourced fields, and our own readings.
| aliases | Ice: sea ice, sea glaciers and tropical ice sheets |
| born at | Antarctica ~34.4-33.7 Ma; Greenland 2.7-2.6 Ma; earliest ice sheets not dated in the sources read |
| born precision | ±0.35 Myr (EOT window); ±0.05 Myr (Greenland) |
| boundary | Inside: the grounded ice of continental size, from its domes and divides to its margins and grounding lines, its firn, its drainage basins, ice streams and outlet glaciers, the water at its bed (subglacial lakes, channels, wet sediment) and its surface melt and lakes. Its floating ice shelves are treated as its edge: they are fed by it and buttress it, but in the mass-balance sense they are counted apart (IMBIE counts only grounded ice that changes sea level). Outside: the bed rock and sediment, the mantle under it, the ocean and its warm deep water, the atmosphere, sea ice, the icebergs once calved, and the climate that feeds and starves it. |
| nested level | One component of the cryosphere and of Earth's climate system; the largest fresh-water store; built from snow turned to firn and ice, carried by gravity on a bed of rock and sediment, much of it below sea level; held up and pushed back by its shelves, lapped by the ocean, pressing on the mantle. Its growth is the signature of glacial periods; its loss is the largest uncertain term in future sea level. |
| scale | planetary |
Measures (77) published science
| Antarctic ice volume including ice shelves | 27.17 million km3 |
| Antarctic potential sea-level equivalent | 58 m |
| Antarctic area including ice shelves | 13.63 million km2 |
| Antarctic mean ice thickness excluding ice shelves | 2148 m |
| thickest Antarctic ice | 4757 m |
| share of Antarctic grounded area with bed below sea level | 46.7 % |
| deepest Antarctic bed below sea level (Byrd Glacier) | -2973 m |
| Greenland ice sheet area | 1,710,000 km2 |
| Greenland ice sheet mean thickness | 1,673 m |
| Greenland ice volume | 2,900,000 km3 |
| sea-level equivalent, Greenland Ice Sheet | 7.4 m |
| sea-level equivalent, West Antarctic Ice Sheet | 5.3 m |
| sea-level equivalent, East Antarctic Ice Sheet | 52.2 m |
| share of Earth's fresh water in the Antarctic ice | 61 % |
| combined ice sheet mass loss 1979-2023 | 11,309 ± 565 Gt |
| combined sea-level contribution 1979-2023 | 31.4 ± 1.6 mm |
| share of 1979-2023 ice-sheet loss from ice dynamics | 84 % |
| Antarctic mass loss 1979-2023 | 4,780 ± 513 Gt (13.3 ± 1.4 mm sea level) |
| Greenland mass loss 1972-2023 | 6,215 ± 467 Gt (119 ± 9 Gt/yr; 67% discharge, 33% SMB) |
| Greenland mass loss rate 2010-2019 | 264 ± 18 Gt/yr |
| Greenland mass loss rate 2020-2023 | 199 ± 29 Gt/yr |
| West Antarctic mass loss rate 2010-2019 | 163 ± 12 Gt/yr |
| East Antarctic mass balance 2020-2023 | +92 ± 63 Gt/yr |
| ice mass per mm of sea level | 360 Gt/mm |
| Antarctic mass loss rate 2009-2017 | 252 ± 26 Gt/yr |
| Antarctic mass loss rate 1979-1990 | 40 ± 9 Gt/yr |
| share of ice crossing the grounding line that melts in the ocean | nearly half (the other half calves) fraction |
| ice-sheet loss rate 1992-1996 to 2016-2020 | 105 to 372 Gt/yr |
| share of mass loss through ice streams, Antarctica | about 90 % |
| share of mass loss through ice streams, Greenland | about 50 % |
| ice stream speed | upwards of 1,000 m/yr |
| tidal modulation of ice stream D speed | factor of 3 over a tidal cycle (~24 h) ratio |
| reach of tidal influence inland | up to 100 km |
| time for surface warming to reach the bed by conduction | around 10,000 years |
| supraglacial lake drainage to the bed | 2-18 hours |
| ice accumulated over a 1942 crash site in 50 years | 81 m |
| Heinrich event spacing | 7,000-10,000 years |
| obliquity period pacing glacial cycles | ~41 kyr |
| regrowth time against retreat time | tens of thousands of years against centuries to millennia |
| active subglacial lakes newly detected | 85 (a 58% increase on 146) lakes |
| Antarctic meltwater generation | 1100-1500 Gt/yr |
| organic carbon stored and cycled in ice sheets | >10^4 (printed '>104' in the text extraction) Pg C |
| particulate organic carbon in sediments beneath the Antarctic Ice Sheet | 6000-21,000 Pg C |
| carbon beneath the Greenland ice sheet | 0.5-27 Pg C (= billion tonnes) |
| sea-level fall as the East Antarctic Ice Sheet formed | ~70 m (within ~700 kyr, 34.4-33.7 Ma) |
| CO2 threshold for Antarctic glaciation | 600 ppm |
| CO2 when Greenland's ice caps joined into a sheet | 280-320 ppm (2.7-2.6 Ma) |
| Antarctic cooling since 6 Ma | 12 ± 2 °C |
| oldest continuous ice core | 800 kyr |
| longest continuous ice cover over soil, Greenland summit | around 2.7 million years |
| peak sea-level rise rate, Meltwater Pulse 1A | 3.5-4 m per century |
| peak sea-level rise rate, termination IV | up to ~5 m per century |
| temperature thresholds (best estimates): GrIS and WAIS | +1.5 °C above pre-industrial |
| temperature thresholds: marine-based EAIS sectors | +2 to +3 °C above pre-industrial |
| first Antarctic threshold (collapse of ~40% of West Antarctic marine ice) | as low as 1-2 °C above pre-industrial |
| hypothesised safe limit for ice sheets | at or below +1.0 °C above pre-industrial |
| threshold lowering by rate-induced tipping | up to 0.5 K |
| GIA slowing of ice loss over centuries (RCP2.6) | up to 40 % |
| ISMIP6 Antarctic contribution by 2300 (SSP5-8.5) | up to +4.4 m |
| extreme Greenland melt events after 2000 | 7 of the 10 most extreme since 1950 count |
| projected increase in extreme Greenland meltwater anomalies by 2100 (SSP5-8.5) | up to +372 % |
| global mean warming offset by Antarctic meltwater by 2100 | ~0.3 °C lower |
| Arctic warming rate relative to global, since 1979 | nearly four times ratio |
| people living below 10 m above sea level | about one billion people |
| Jakobshavn Isbræ mean annual speed, 2012 against mid-1990s | nearly 3 (peak summer more than 4) ratio |
| Heinrich layer deposition time | approximately 500 ± 250 years |
| organic carbon in englacial ice (DOC + POC), both ice sheets | <10 Pg C |
| sub-ice-sheet methane hydrate reserve at the Last Glacial Maximum | >500 Pg C (20% of present marine hydrate stocks) |
| decline in annual ice speed, K-transect 1991-2007 | 10 % |
| slowdown within ~50 km of the margin, west Greenland, 2007-2014 against 1985-1994 | 12 % slower (only ~30% explained by thickness and slope change) |
| median subglacial lake drainage time, Antarctica | 2.2 (upper quartile 4.4) years |
| elevation where Greenland albedo feedback turns from positive to negative | 1500 m |
| albedo feedback averaged over the ablation area (anomaly-gauged) | +8.3 W m-2 K-1 |
| albedo feedback averaged over the accumulation area (anomaly-gauged) | -0.4 W m-2 K-1 |
| rise of Greenland maximum runoff elevation 1985-2020 | 58-329 m |
| growth of Greenland runoff area 1985-2020 | 29 (-8/+6) % |
| share of recent runoff from the expanded runoff area | 5-10 % |
Open contentions (13) open contention
Where sources or readings disagree. Both stay on the map.
- C-3411 Do the Greenland and Antarctic ice sheets count as glaciers?
booked: The map keeps the competing readings; none is ruled.
the other reading: broad sense: ice sheets are a kind of glacier; mass-balance sense: glaciers exclude the ice sheets - C-3574 Does a darker, melting ice surface always speed the melt, or does extra snowfall higher up slow it?
booked: The map keeps the competing readings; none is ruled.
the other reading: positive albedo feedback over 97% of the ice sheet, strongest in the ablation area (Box 2012); second-order negative feedback over 63% of the accumulation area from snowfall rising with temperature (Box 2012); transition at about 1500 m; the regimes overlap in elevation - C-3616 Is East Antarctica gaining or losing ice?
booked: The map keeps the competing readings; none is ruled.
the other reading: input-output: East Antarctic loss in 2021 (-210 ± 73 Gt/yr); altimetry and gravimetry: gain (99 ± 124 and 166 ± 50 Gt/yr); IMBIE notes the single input-output estimate is over-weighted; grounding-line thickness data missing within 5 km for over 60% of the grounding line in several regions - C-3617 What made Antarctica freeze over 34 million years ago: falling carbon dioxide or a newly opened ocean passage?
booked: The map keeps the competing readings; none is ruled.
the other reading: the opening of the Drake Passage played a role (encyclopedia) - C-3629 Did central Greenland lose its ice within the last million years?
booked: The map keeps the competing readings; none is ruled.
the other reading: the Summit bed is a long-lived stable land surface; central Greenland was tundra-covered in the Pleistocene (Bierman 2024); near the summit a frozen basal layer has kept soil under ice for ~2.7 Myr and ice ~1 Myr old exists (encyclopedia); climate models typically simulate continuous solid ice under warm interglacials, contradicted by Camp Century and Labrador Sea evidence (encyclopedia) - C-3632 What set off the great iceberg discharges of the last ice age?
booked: The map keeps the competing readings; none is ruled.
the other reading: internal binge-purge oscillation (early theories); binge-purge preconditioned, with the Dansgaard-Oeschger interstadial warming as the final trigger (Schannwell 2024); external forcings may drive both Heinrich and D-O events (encyclopedia); jökulhlaup activity from a Hudson Bay lake (Hemming 2004, abstract); an ice shelf buildup and collapse fed by Hudson Strait (Hemming 2004, abstract) - C-3634 How much warming can the polar ice sheets take?
booked: The map keeps the competing readings; none is ruled.
the other reading: GrIS and WAIS best estimates +1.5 °C; marine EAIS +2 to +3 °C (Stokes 2025, citing recent work); a safe limit at or below +1.0 °C, hypothesised (Stokes 2025); a first Antarctic threshold as low as 1-2 °C for ~40% of West Antarctic marine ice; marine East Antarctica at 2-5 °C (Winkelmann 2026); rate-induced tipping lowers thresholds by up to 0.5 K (Swierczek-Jereczek 2026); central-western Greenland already close to a critical transition (Boers & Rypdal 2021, abstract); Thwaites/Pine Island, Ronne and Siple Coast basins at risk below +1 °C; may already have passed a critical point at present warming (Winkelmann 2026, results read) - C-3636 How high did the sea stand in the last warm period, and which ice sheet supplied the water?
booked: The map keeps the competing readings; none is ruled.
the other reading: Last Interglacial peak GMSL +6 to +9 m (earlier assessment); +1 to +4 m (two Bahamas studies); ~+6 m from Antarctica alone (northern Europe coastal deposits); the North American ice sheets may have persisted into warm interglacials, so Antarctic sensitivity is underestimated (Creel 2026) - C-3644 Does meltwater reaching the bed speed an ice sheet up or slow it down?
booked: The map keeps the competing readings; none is ruled.
the other reading: more runoff to the bed speeds flow (early melt season; inland where efficient drainage cannot form) (Davison 2019); efficient drainage slows annual flow in the ablation area despite more melt: self-regulation (Davison 2019); the 2000s hypothesis of melt-driven acceleration as a reinforcing loop (specialist, unsourced) - C-3645 Does Antarctic meltwater slow further melting or speed it?
booked: The map keeps the competing readings; none is ruled.
the other reading: balancing: meltwater lowers global surface warming (~0.3 °C by 2100) and slows Antarctica's contribution (Sadai 2025, one coupled model); reinforcing: meltwater-induced subsurface ocean warming round the coast could drive further ice-shelf and ice-sheet melt (Bronselaer 2018, abstract only) - C-3646 Can tall ice cliffs collapse fast enough to drive rapid sea level rise?
booked: The map keeps the competing readings; none is ruled.
the other reading: MICI: cliffs above ~90 m fail once shelves are gone and drive rapid retreat; helps reconcile past high sea levels (Pattyn 2018; Stokes 2025); MICI may not act this century: three models show Thwaites would not retreat further, rapid thinning and speed-up stabilise the cliff (Morlighem 2024); mélange may slow or stop it (encyclopedia); unconfirmed either way at Crane Glacier 2002-2004 (encyclopedia); Provenance: the hypothesis entered projections mainly through DeConto & Pollard's 2016 model work (specialist, not opened); DeConto co-authors Stokes 2025 (author list read), and Sadai 2025's model (PSU3D) simulates hydrofracturing and ice-cliff calving (discussion read): those 'for' readings are not independent confirmation.; Morlighem 2024 discussion: calving rate would need to rise 25-fold to trigger unstoppable retreat at Thwaites; the result does not suggest WAIS is stable and does not preclude MISI - C-3647 Is an ice sheet resting on a bed that slopes down inland always unstable?
booked: The map keeps the competing readings; none is ruled.
the other reading: steady grounding lines cannot be stable on reverse bed slopes (Schoof 2007, abstract; a two-dimensional sheet-shelf flowline); retrograde slopes do not per se imply an instability; marine ice sheets are conditionally stable in two horizontal dimensions, with buttressing (Gudmundsson 2012, read: abstract, discussion, conclusions); Gudmundsson: the two are compatible; 1HD and 2HD problems differ fundamentally - C-3648 Has West Antarctica's irreversible retreat already begun?
booked: The map keeps the competing readings; none is ruled.
the other reading: MISI may already be underway in the Amundsen Sea; WAIS partial collapse 'almost inevitable this century' (Stokes 2025, reviewing); present-day grounding-line retreat is not yet irreversible, but sustained forcing commits it ('committed tipping') (Stokes 2025, citing a stability analysis); WAIS very likely to disappear even without further warming (encyclopedia); Thwaites/Pine Island basin committed to ~70% of its sea-level-relevant volume once past its threshold (0.9 m), mainly by MISI (Winkelmann 2026); the encyclopedia's 'very likely to disappear even if the warming does not progress' has no paper behind it in this walk (specialist; note on o-wais-loss)
Cross-scale patterns seen here (19) our proposal
Shapes that recur when the same kind of problem is solved at different scales, and how each shows in Ice sheet. Each links to every other system that shows it.
- conservation law plus closure (structural)
Mass balance is a balance law that does not close by itself: altimetry measures volume change and needs a firn density model to turn it into mass; gravimetry measures mass but needs a model of glacial isostatic adjustment to remove the solid Earth's own signal; the input-output method needs ice thic - lightning (structural)
Stored ice or water released at once, changing what follows: Heinrich purges sending flotillas of icebergs across the North Atlantic; the Larsen B shelf collapsing in three weeks and its glaciers speeding up; surface lakes emptying to the bed in 2-18 hours and the ice surging; Meltwater Pulse 1A at - runaway from an unstable balance (structural)
Three self-reinforcing runaways around a balance point: the melt-elevation feedback (a lowered surface sits in warmer air and melts faster); marine ice sheet instability (on a bed deepening inland, a grounding line off its balance point retreats into thicker ice and more flux); and the contested ice - same role, different builds (structural)
One job, taking mass off the ice sheet, done by different builds: solid discharge at the grounding line, melt at the bed and at the ice-ocean contact, surface runoff, sublimation, evaporation and blowing-snow erosion (IMBIE). One job, weighing the ice sheet, done three ways: input-output (velocity t - same shape, opposite sign (structural)
One process with opposite net signs by place or timescale: meltwater reaching the bed speeds the ice early in the season and, once drainage turns efficient, slows it over years in the ablation area while inland, where efficient drainage cannot form, a small net speed-up remains (Davison 2019). Secon - The carry (structural)
Overflow passed on: at the marine margin excess ice is discharged through ice streams and outlet glaciers, and shelves calve icebergs when they hold an excess (encyclopedia). The wait: surface warming takes about 10,000 years to reach the bed by conduction, and the response lag grows with the size o - borrowed-then-held (partial)
The cold that builds an ice sheet is first lent by its surroundings and then held by the sheet itself. Greenland: uplift of the land gave cooler surface temperatures and more snow, letting ice accumulate (borrowed); CO2 falling to 280-320 ppm let the caps join. Once built, the sheet's own height kee - causal ceiling (partial)
A limit on knowing set by the system's own slowness: because the ice responds with a lag, the warming already done may have committed future instability, and a threshold will probably not be apparent until it has been passed; the IPCC's 20-year-mean rule itself recognises exceedance only a decade la - holding / entrainment (partial)
The ice sheet's own internal oscillator (binge-purge) is set in time by an outside clock: in a coupled model the Dansgaard-Oeschger interstadial warming triggers the purge, explaining synchronous Heinrich events from different regions of the Laurentide sheet; the tide paces ice-stream speed through - same toolkit, independent builds (partial)
Ice sheets were built separately at different times from the same parts (land at high latitude, falling CO2, orbital timing, snow that outlasts summer): Antarctica's at the Eocene-Oligocene Transition (~34 Ma, CO2 threshold ~600 ppm), Greenland's at 2.7-2.6 Ma (280-320 ppm), and the Northern Hemisph - smooth to sharp in finite time (partial)
Smooth warming meets a sharp change at a finite threshold in some basins and not others: ice loss in some Antarctic basins unfolds gradually with warming, while others have a critical threshold beyond which large parts disintegrate (Winkelmann 2026); equilibrium runs show a bifurcation point, and fa - The Payment to Entropy (partial)
Thermal form: sliding generates frictional heat at the bed (with geothermal heat it melts basal ice); some of the toll is reused as lubricating water that speeds flow, so part of the loss is a hand-off back to the system itself (the dark twin). Material form: of the ice that crosses the grounding li - the whole sets the local reference (partial)
Whether ice is grounded or afloat is judged locally against the buoyant thickness, which depends on sea level, set by the whole ocean; and the ice sheet's loss raises sea level unevenly, so local sea level from Antarctic loss runs up to 0.9 m above the global mean in the Pacific by 2200 (RCP8.5): th - Umwelt: each system carves its own slice of one shared medium (partial)
The ice sheet takes its slice of the shared Earth surface: snow, sun, air, CO2 and orbit at its surface; the ocean's deep water and the tide (felt up to 100 km inland) at its marine edge; gravity, the bed, its heat and water, and the mantle's slow answer to its own weight below. A mountain glacier, - who opens the door (partial)
At its doors: the grounding line opens by plain buoyancy (open field); warm deep water reaches the shelves along submarine glacial troughs cut by earlier ice (an earlier epoch opened the door; named in a reference title, paper not read); a surface lake opens its own door, a water-filled crack its we - barrier-gates-complexity (surface)
Two barriers with the without case shown, but what they gate is not complexity: ice shelves hold back the grounded ice (without them, Larsen B's glaciers sped up and those still held did not); a frozen basal layer has sealed soil for ~2.7 Myr and kept a till with plant, insect and fungus fossils und - nested ladder (surface)
Size classes (ice sheet above ice caps, ice fields and glaciers) form an ordered list, not an implication ladder; hydrofracture is named as a precursor of the cliff instability, which 'facilitates and enhances' the marine ice sheet instability: a chain of mechanisms, not each rung implying the last. - partner-becomes-part (surface)
Separate ice bodies fuse into one: Greenland's disparate ice caps connected to cover most of the island 2.7-2.6 Ma; at the last glacial maximum the Laurentide and Cordilleran sheets merged with Greenland's. A merger of like bodies with no gain/gain loop or inheritance shown. - templating (surface)
An ice stream is 'a type of glacier' running inside the ice sheet: the glacier module reused as the sheet's fast lane. The modelling dark twin is also on record: the shallow-ice template fits the bulk of the sheet and breaks down at its edges (Pattyn 2018).
Valence: what it trades, and with whom (17) published science
Each exchange with a neighbour, read from both sides: who gains, who pays, and what carries it.
- Water cycle (hydrologic cycle) via sys:precipitation · Ice sheet gain
- Earth's crust (primordial, oceanic and continental) via sys:earth-crust · Ice sheet mixed
- Atmosphere of Earth via sys:atmosphere-of-earth · Ice sheet mixed
- Sun via sys:electromagnetic-field · Ice sheet mixed
- Human (Homo sapiens) via sys:electromagnetic-field · Ice sheet loss
- Ocean via sys:ocean · Ice sheet loss
- Climate system (climate) via sys:climate-system · Ice sheet loss
- Carbon cycle via sys:carbon-cycle · Ice sheet mixed
- Groundwater and aquifers via sys:water · Ice sheet mixed
- Glacier via sys:ocean · Ice sheet even
- Biosphere via sys:biosphere · Ice sheet loss
- Earthquake via sys:earth-crust · Ice sheet loss
- Milankovitch cycles via sys:electromagnetic-field · Ice sheet mixed
- Mantle (geology): Earth's mantle via sys:gravitational-field · Ice sheet gain
- Ice age via sys:climate-system · Ice sheet gain
- Sea level via sys:ocean · Ice sheet loss
- Volcano via sys:gravitational-field · Ice sheet loss
Partners and parts published science
Built from / builds (14)
- Water cycle (hydrologic cycle)
- Cryogenian Snowball Earth (Sturtian and Marinoan glaciations, and the thaw)
- Ice
- Snow
- Sediment
- Biosphere
- Meltwater (supraglacial, englacial, subglacial)
- Ice stream
- Ice shelf (and ice tongue)
- Subglacial lakes and water
- Ice vs photosynthesisers for the sunlit tropical ocean (hard-snowball reading)
- Ice vs photosynthesisers for the sunlit tropical ocean (slushball / waterbelt reading)
- Greenhouse vs ice albedo for absorbed sunlight (termination; hypotheses kept: volcanic CO2, permafrost methane, dust and cryoconite, methanogens)
- Ice sheet against ocean for the marine bed (the grounding zone)
Sits inside / holds (4)
expressed by (7)
- Generation/Synthesis
- Recombination/Composition
- Niche Construction (alter environment)
- Internal Coordination/Regulation
- Physical/Virtual Barrier
- Quality Control & Error Correction
- Bridging Supply Gaps (Store & Dormancy)
floor filled by (32)
- Water cycle (hydrologic cycle)
- Gravitational field
- Earth's crust (primordial, oceanic and continental)
- Atmosphere of Earth
- Sun
- Ice
- Ocean
- Climate system (climate)
- Carbon cycle
- Ice core
- Precipitation
- Snow
- Sediment
- Antarctic ice sheet (kept by the ACC; meltwater back)
- Biosphere
- Milankovitch cycles
- Greenland Ice Sheet
- Mantle (geology): Earth's mantle
- Meltwater (supraglacial, englacial, subglacial)
- Ice stream
- Ice shelf (and ice tongue)
- Ice age
- Subglacial lakes and water
- Laurentide ice sheet
- Outlet glacier
- Marine ice sheet instability (MISI)
- Marine ice cliff instability (MICI)
- Surface elevation and melt feedback
- Sea level
- Firn
- Paleoclimatology
- History (the field)
Kinds (11)
- Glacier
- Geological archive
- Antarctic ice sheet (kept by the ACC; meltwater back)
- Greenland Ice Sheet
- West Antarctic Ice Sheet
- East Antarctic Ice Sheet
- Antarctic Peninsula ice sheet
- Laurentide ice sheet
- Cordilleran ice sheet
- Weichselian (Fennoscandian) ice sheet
- Patagonian ice sheet
wired (4)
- Atmosphere of Earth
- Hydrosphere (Earth's planetary water, above all the early ocean)
- Cyanobacterium (Cyanobacteria, phylum Cyanobacteriota)
- Precipitation
Notes published science
- Under the tide beneath the Ross Ice Shelf, a West Antarctic ice stream changes speed threefold every day; the pulses had never been seen because they average to zero over the longer surveys. (src:anandakrishnan-2003)
- The East Antarctic Ice Sheet grew to almost its modern size in about 700,000 years, 34 million years ago, and the world's sea level fell about 70 m as it did. (src:delira-mota-2023)
- Under 3 km of ice at Greenland's Summit lies till with willow wood, a poppy seed, spike-moss spores, fungus and insect parts: the middle of Greenland was tundra, ice-free at least once in the last 1.1 million years. (src:bierman-2024)
- Meltwater reaching the bed of the Greenland ice sheet speeds the ice up for days, yet over years the ablation area has slowed (10% along the K-transect, 1991-2007) as the drainage under it grew efficient: more melt, slower ice. (src:davison-2019)
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
the boardroom (shared floors)
- L3: The ice body and its mass balance: snowfall in; surface runoff, sublimation, evaporation and blowing snow, melt at bed and ice-ocean contact, and solid discharge at the grounding line out; read as surface mass balance plus ice dynamics (IMBIE: dynamics 84% of 1979-2023 loss)
- L4: Flow: outward from domes, slow shearing inland, fast ice streams and outlet glaciers at the edges (about 90% of Antarctic loss), the grounding line where ice floats, ice shelves buttressing from in front
- L5: Stability: thresholds, hysteresis and the self-reinforcing instabilities (melt-elevation, marine ice sheet instability, the contested marine ice cliff instability, fracture weakening), basin by basin; the balancing brakes (bedrock uplift, meltwater cooling)
- L6: The ice sheet in the Earth system: the largest fresh-water store and the largest uncertain term in sea-level rise; meltwater into the overturning ocean; nutrients and carbon; brightness in the climate; weight on the crust; calving quakes
- L7: Ice sheets across deep time: their growth the signature of glacial periods, with CO2 and ice feedbacks carrying most of the Quaternary temperature signal; born at thresholds (Antarctica ~34 Ma, Greenland 2.7-2.6 Ma), vanished sheets (Laurentide, Cordilleran, Weichselian, Patagonian), Heinrich purges and meltwater pulses; the archive of past air and climate in their cores
the ice sheet as glaciologists describe it (reader's view)
- L1: What reaches the ice sheet: snowfall, sunlight, warm air, warm ocean water at its shelves, the tide, gravity, the bed and its geothermal heat, the solid Earth under it, orbital rhythm and atmospheric CO2
- L2: Transformation at the surface and the bed: snow pressed into firn and ice; melt, runoff and darkening at the surface; geothermal and frictional melt, water and rock flour at the bed
- L3: The ice body and its mass balance: surface mass balance and ice dynamics; domes and drainage basins
- L4: Flow: outward from the domes, ice streams and outlet glaciers, the grounding line, ice shelves buttressing
- L5: Stability: thresholds, hysteresis and self-reinforcing instabilities; balancing brakes
- L6: The ice sheet in the Earth system: sea level, ocean overturning, nutrients and carbon, climate, the crust
- L7: Ice sheets across deep time: the signature of glacial periods, amplifiers of orbital pacing, born at CO2 thresholds, vanished sheets and their pulses; the ice-core archive
way in: Air temperature and weather (melt events)
- L1: Warm air masses over the ice; extreme melt events driven by circulation and warming
- L2: Surface melt and runoff; lower surfaces sit in warmer air
way in: Atmospheric CO2 (greenhouse level)
- L1: CO2 in the atmosphere: thresholds for building ice sheets (about 600 ppm for Antarctica, 280-320 ppm for Greenland)
- L2: Falling CO2 cools the land until separate ice caps join into a sheet
way in: Geothermal and frictional heat at the bed
- L1: Heat from below
- L2: Basal melting by geothermal and frictional heat and pressure melting
- L3: Meltwater lubricates the bed and makes fast-flowing ice streams
way in: Global warming and regional climate (temperature, winds)
- L1: Global mean temperature above pre-industrial (+1.2 °C today, per the review)
- L2: Polar amplification and wind patterns: the Arctic warming nearly four times the global rate; a more positive Southern Annular Mode cutting snowfall and pushing warm deep water
way in: Gravity (driving stress)
- L1: The pull on the ice mass
- L2: Surface slope from accumulation inland and ablation at the margins builds shear stress
- L3: The ice deforms by shearing near a near-frozen bed (shallow-ice flow), valid for the bulk but not the edges
way in: Ocean heat (circumpolar deep water, fjord water)
- L1: Warm, salty circumpolar deep water reaching the continental shelf, pushed by stronger westerlies
- L2: Melt under the floating shelves: more than half of the ice lost at Antarctica's margins
- L3: Thinner, fractured shelves lose buttressing; the glaciers behind accelerate
way in: Ocean tides
- L1: The tide under the ice shelf
- L2: Felt at and inland of the grounding line, up to 100 km from the sea
- L3: Ice-stream speed varies up to threefold over a tidal day; at spring tides stick-slip surges after high tide
way in: Orbital forcing (Milankovitch)
- L1: Changes in Earth's orbit and tilt: precession ~23 kyr, obliquity ~41 kyr, eccentricity ~96 and 400 kyr
- L2: Insolation patterns that set glacial and interglacial transitions
way in: Snowfall (accumulation)
- L1: Snow falling on the ice sheet, its main gain; record snowfall can swing a whole sheet (East Antarctica 2022-2023)
- L2: Snow that never melts, compressed by newer snow into firn and ice
way in: Sunlight
- L1: Sunlight on the surface
- L2: Absorbed or reflected by the surface: bright snow reflects, darker bare ice, melt ponds and algae absorb and melt
way in: Surface life and dust (algae darkening)
- L1: Glacier ice algae and dust on bare ice
- L2: Dark algal blooms lower albedo and speed melt; algae colonise newly bared ice without added nutrients
way in: The bed (rock, sediment, topography)
- L1: Bedrock and sediment under 2 km of ice on average; nearly half Antarctica's grounded area below sea level
- L2: The ice-bed contact: hard ridges resist flow; water-filled sediments let it slide; soft sediment raises speed
- L3: Bed slope at the grounding line: a bed deepening inland makes grounding-line position unstable
way in: The solid Earth under the load (glacial isostatic adjustment)
- L1: Lithosphere and mantle under the ice load
- L2: Bedrock uplift after unloading: tens of mm a year over the low-viscosity Amundsen Sea mantle, slow (mm to cm a year) under Greenland
- L3: A rising bed shallows the grounding zone and slows retreat (up to 40% over centuries under low emissions)
- L4: The race of rates: when warming outpaces slow uplift, grounding lines reach retrograde slopes and retreat sustains itself below the equilibrium threshold
way in: Water at the bed (surface melt routed down, subglacial lakes)
- L1: Surface lakes that crack through to the bed; basal water in lakes, swamps, channels and aquifers
- L2: Active subglacial lakes fill and drain in linked networks
- L3: Grounding-zone coupling, two-way: lakes near the grounding line can raise discharge; retreat can trigger drainage
way out: Calving quakes
- L1: Seismic waves of about magnitude 5, about 90% in Greenland, mostly July-September
- L4: Calving at marine fronts
way out: Icebergs (calving)
- L1: Icebergs in the ocean carry rock (ice-rafted debris, Heinrich layers) and iron
- L4: Half the ice crossing the grounding line breaks off as icebergs
way out: Mass to the ocean: sea-level rise
- L1: Water added to the ocean; regionally uneven (up to 0.9 m above the global mean in the Pacific by 2200 under RCP8.5)
- L4: Discharge across the grounding line, surface runoff and basal melt
- L6: Ice lost to the ocean counted as sea-level contribution (360 Gt per mm)
way out: Meltwater and ice in the climate (temperature, albedo)
- L1: Air temperature and radiation over the hemispheres
- L6: Antarctic meltwater lowers global mean warming by ~0.3 °C by 2100 in a coupled model, warming parts of the north up to 1.5 °C
- L7: Ice-sheet feedbacks amplify orbital pacing into glacial cycles
way out: Meltwater into the overturning ocean
- L1: Freshened Antarctic Bottom Water; a weaker lower overturning cell
- L6: Antarctic meltwater, 1100-1500 Gt a year
way out: Nutrients and carbon (iron, silicon, rock flour, organic carbon)
- L1: Iron- and silicon-rich meltwater and iron-bearing icebergs fertilise the ocean
- L2: Rock flour weathered at the bed; microbes release Fe, Si and P
- L6: Ice sheets store and cycle organic carbon and nutrients
way out: Weight on the crust (isostatic depression and rebound)
- L1: Mantle flow and bedrock rise; uplift after collapse may add about 1 m of sea level over 1,000 years (encyclopedia)
- L3: The ice load; its loss unloads the lithosphere