Introduction to glacial landforms

Glaciers are one of the most powerful forces shaping our local landscape. As glaciers flow downhill from mountains to the lowlands, they erode, transport, and deposit materials, forming a great array of glacial landforms. They can erode mountains, and change their morphology. Large glaciers and ice sheets can deposit great swathes of sands and gravels, forming swarms of hills called drumlins. Ice sheets deposit great thicknesses of glacial tills, and glaciers and ice sheets form moraines at their terminus. These pages will explain these concepts in more detail.

Erosional glacial landforms

In their upper reaches, glaciers can erode bedrock by quarrying, pucking, abrasion and polish. Rocks and debris embedded in the ice scratches the rock below.

Polished and striated bedrock in Utah. Photo credit: Bethan Davies

The photograph below shows ice-scoured, smoothed bedrock in Greenland. The passage of the ice, with lots of debris embedded in it, has scratched and abraded the rocks, making them smooth. Over time, roche moutonnees develop, which are smooth on one side but have a blunted downstream face.

Bethan Davies standing on ice-scoured bedrock in Greenland. She is pointing in the direction of former ice flow.
Polished and striated bedrock in Utah. Note the plucked face on the down-stream end. Photo credit: Bethan Davies

This erosion creates deep hollows in mountain sides, called cirques. Multiple cirques on a mountain may cause a pyramidal peak as they form back-to-back. Cirques are one of the most visual and characteristic glacial landforms of glaciated mountains.

Classic glacial cirque basin. Cwm Clyd in the Glyderau mountains of Snowdonia. Image from GoogleEarth.

Larger glaciers can excavate a glacial trough, which has a parabolic, or U-shape.

The parabolic glaciated valley of Glen Coe, Scotland. Photo credit: Bethan Davies

Trimlines on the valley side mark out the former ice surface. The area below the trimline was smoothed by the passage of glacier ice. In the photo below, the trimline was formed during the “Little Ice Age”, when the glacier reached the moraines visible in the bottom of the photograph. The valley side above the trimline was not glaciated at this time, and so is more vegetated and weathered.

Valley side trimlines (labelled with white arrows) marking the former thickness of the Callequeo Glacier, Monte San Lorenzo. Photo credit: J. Martin.

Depositional glacial landforms

Lower down, in the ablation zone, deposition becomes more important and shear stress lessens. This can lead to the deposition of vast thicknesses of unsorted sediments called tills.

Two tills
Two tills rest on top of Magnesian Limestone bedrock at Whitburn Bay, overlain by deformed glaciofluvial sands (sands deposited by a proglacial river). Note the large, faceted boulders at the boundary between the two tills.

Around the margins of the glacier, lateral moraines may develop in the ablation zone, and terminal moraines may form at the end of the glacier. In front of the ice margin, there may be small scale streamlined ridges called flutes.

The ridges in the forefield of this glacier are moraines. They are made up of debris carried by the glacier, and deposited in ridges at its terminus.
Push moraines and flutes in a recently deglaciated glacier forefield.
Esmeralda Moraine. A subaerial terminal push moraine with symmetrical sides. Photo credit: Bethan Davies

Under larger ice masses such as ice sheets, drumlins may form. These are elongated hills made up of glacial sediments (sands, gravels, boulders, unsorted muds) that form in the direction of ice flow.

Drumlin at Holwick, Teesdale. Photo credit: Bethan Davies

Under faster-flowing ice streams, mega-scale glacial lineations may form. These landforms are important for telling us about directions and dynamics of ice flow under former ice sheets.

Belgica Trough Mega Scale Glacial Lineations, Antarctic Peninsula.

Glaciofluvial landforms

Glaciers are wet. Temperate glaciers, which are those in more moderate climates and that have meltwater at their base, produce huge volumes of water each melt season. This results in a characteristic suite of glaciofluvial landforms.

Meltwater stream on Mendenhall Glacier, Alaska. From: Gillfoto, Wikimedia Commons

All this water produces a whole suite of glacial landforms. These include eskers (ridges of sediments that form underneath a glacier), kame terraces, sandar (braided gravel-rich outwash streams), and meltwater channels.

A sinuous esker ridge, and several smaller eskers, mapped from satellite imagery in the Lago Cochrane-Pueyrredón valley. Copyright: J. Bendle.

Meltwater may cut meltwater channels underneath and around the margins of the glacier. Ice-marginal meltwater channels usually form around sub-polar glaciers where water cannot get underneath the glacier, which is frozen to its bed. They therefore form in a lateral position, between the glacier and the valley flank, or around the snout. These meltwater channels can therefore mark out the position of the former ice margin.

Lateral meltwater channel in Lunedale, Pennines. Image credit: Bethan Davies.

The glaciofluvial rivers that drain away from glaciers are typically very laden with sediment. If the valley floor is quite low-angle, as is common in glacial valleys, then the river tends to form a braided pattern, with bars of gravelley sand forming. These rivers are very active and their form changes regularly. These features are called sandars.

Glaciofluvial outwash from Nef Glacier, Patagonia

Glaciolacustrine landforms

Many temperate glaciers terminate in glacial lakes, which results again in a characteristic suite of glacial landforms. Lakes may form in front of glaciers, occupying the glacial overdeepening, and may be dammed by moraines, by the ice itself, or by bedrock.

An ice-dammed lake on the northern margin of the Russell Glacier, in western Greenland.

Moraine-dammed and ice-dammed lakes may be susceptible to hazardous Glacial Lake Outburst Floods.

Rapid growth of glacial lakes in the Bhutan-Himalaya in response to retreating glacier termini. Photo: NASA/USGS, Wikimedia Commons.

Sediments in glacial lakes may be varved, with winter and summer layers being laid down each year.

Varves in a glacial lake. Photo credit: Jacob Bendle.

The location of former glacial lakes may be marked out by shorelines, raised deltas, beaches, and grounding line fans or morainal banks.

Raised delta in Patagonia. The high flat delta top formed when the lake was higher, due to ice damming the outflow. The lower delta is forming in the lake today.

Glacial landsystems

The types of glacial landforms that are generated are particular to glacier flow, basal processes, the substrate (soft and deformable? Hard crystalline bedrock?), the basal driving stress and thermal regime, and the ice temperature.

There are diagnostic landforms associated with wet-based sheet flow, ice streams, and surging ice. These diagnostic suites of landforms are called glacial landsystems.

Ice streamsSurging ice Sheet flow Cold-based ice
Mega scale glacial lineations (MSGLs)Looped medial morainesMarginal / subglacial / glaciofluvial domainsMay be very little modification of previous landforms
Progressive elongation of landforms down-iceThrusted end morainesPush, dump, squeeze morainesSmall glaciotectonic structures
Trough mouth fansConcertina eskersSubglacial till, flutes, drumlins, overridden morainesSome deposits with a coarse, sandy to boulder-gravel texture.
Till, glaciotectonised sedimentsTill, glaciotectonite, complex till stratigraphiesRoche moutonnees, striated and polished bedrockLittle evidence of fluvial reworking, but aeolian reworking may be common.
Drumlins, meltwater channels, terminal moraines, grounding linesCrevasse-squeeze ridges; flutingsTill, glaciotectonite 
“Sticky spots” (bedrock bumps/cold-based ice/dry bed)Hummocky moraineSandur, eskers, kame terraces,  proglacial lakes, braided channels, pitted outwash 

In these pages, you can learn more about glacier erosional and depositional landforms. There are case studies to illustrate the key points.

Once you have a grounding in the different kinds of glacial landform, take a look at the Glacial Landsystems pages, where the different suites of landforms that make up characteristic glacial landscapes are highlighted.

You can learn about the techniques that researchers use to understand these landforms, including geomorphological mapping and chronostratigraphy (dating glacial landforms).

There are sections of the website here on the characteristic landforms associated with the last Antarctic Ice Sheet, British-Irish Ice Sheet and the Patagonian Ice Sheet, and even glaciers on Mars.

Thwaites Glacier

Thwaites Glacier, Antarctica, is of particular concern to scientists. Here, warm water is pushed up onto the continental shelf, where it flows along the bottom until it reaches the floating ice shelf in front of Thwaites Glacier.

Thwaites Glacier today is rapidly losing mass in response to changing atmospheric and oceanic conditions.

Thwaites Glacier
Continue reading

Ice stream structural glaciology

The pattern of velocity across the surface of an ice stream is complex, and varies between ice streams. It is captured by surficial structures, such as crevasses and longitudinal flow structures (also known as flow stripes, flow lines, and streak lines)[20]. In places, they can be traced for > 100 km. They form on valley glaciers, outlet glaciers and ice streams, all flowing at a variety of velocities. Longitudinal structures are typically developed parallel to the margins of glacier flow units.

Surface Structures

Glasser and Gudmundsson (2012)[20] have mapped surface structures on a number of Antarctic glaciers.

There are three key hypotheses for longitudinal surface structure formation[20]:

  1. They form as a result of lateral compression in topographic situations where glaciers flow from wide accumulation basins into a narrow tongue.;
  2. They form where two glacier tributaries converge, and are associated with shear margins between flow units;
  3. They are the surface expression of subglacial bed perturbations created during rapid basal sliding.

Glasser and Gudmundsson 2012[20] make the following key observations regarding longitudinal surface structures on Antarctic glaciers:

  1. They are common features on Antarctic glaciers and ice streams, forming at a variety of scales from entire glacier catchments to individual small valley glaciers;
  2. At confluences, larger glaciers “pinch out” structures where they meet smaller tributary glaciers;
  3. The structures can be followed from cirque headwalls to glacier snouts and are continuous;
  4. Longitudinal surface structures sometimes intensify in zones of lateral compression;
  5. Longitudinal structures are more closely spaced at flow-unit boundaries than away from these boundaries.
  6. They are most prominent where ice flow is convergent, but can also be maintained where flow diverges;
  7. They start abruptly, particularly behind bedrock obstacles and nunataks;
  8. They can turn more than 90° without interruption or increased lateral compression;
  9. They are sometimes, but not always, associated with surface debris.

How do they develop?

Glasser and Gudmundsson suggest that these surface structures can develop in two main situations: within glacier flow units, and where there is convergent flow around nunataks of glacier confluence zones. Development within flow units is relatively well understood and derives from basal perturbations on the ice-stream surface[21].

Where these structures start abruptly, they form in areas with rapid longitudinal extension, suggesting that extensional flow can explain these structures. The confluence of two glaciers or flow units, for example, results in strong transverse convergence and longitudinal extension.

What is the ice volume of Thwaites Glacier?

Thwaites Glacier in West Antarctica is currently the focus of a major scientific campaign. Why is Thwaites Glacier of so much interest, however? How much ice is there, and how much would sea levels rise if it all melted?

Thwaites Glacier is roughly the size of UK (176 x103 km2). The glacier terminus is nearly 120 km wide, and the bed of the glacier reaches to >1000 m below sea level. Pine Island Glacier and Thwaites Glacier together account for 3% of grounded ice-sheet area, but they receive 7% of Antarctica’s snowfall1.

Continue reading

Palaeo-ice stream landsystem

Ice streams are corridors of fast-flowing ice within ice sheets that are flanked on either side by slowly moving ice1. Palaeo-ice streams are ice streams that existed in former ice sheets2,3, such as the continental ice sheets that grew during the last Ice Age. Glaciologists know that these palaeo-ice streams existed as they left a clear imprint on the landscape over large parts of North America4, Scandinavia5, and Britain6.

Landsat 7 ETM+ satellite image of Byrd Glacier, an ice stream in West Antarctica. Ice flow is towards the top of the image. Note how flow converges into the main ice stream trunk. Also, note the sharp boundary between fast- and slow-flowing ice. Image: NASA.

Why are ice streams important?

Ice streams in Greenland and Antarctica are the main control on ice sheet mass balance and discharge to the world’s oceans7. Understanding how ice streams behave and change over time, therefore, is important for predicting and managing the impacts of future climate change.

But this is easier said than done…

Firstly, records of modern-day ice stream activity only cover the most recent ~50 years (the length of the satellite record), which is not enough to confidently predict how they may change in the future.

Secondly, it is almost impossible for glaciologists to study the processes that occur at ice stream beds – which control fast ice flow and, ultimately, ice stream discharge to the oceans1 – owing to the great thickness (up to ~3 kilometres) of ice sheets.

Fast-flowing ice streams (blue and white areas) drain the interior of the East and West Antarctic ice sheets, controlling ice sheet mass balance and discharge to the oceans. Image: Jonathan Bamber

Why study palaeo-ice streams?

Therefore, the landforms and sediments left behind by palaeo–ice streams in areas like North America, Scandinavia, and Britain, are very important.

Firstly, they allow glaciologists to study how ice streams have evolved over thousands to tens of thousands of years, through important stages, such as ice sheet build-up, at a glacial maximum, and during deglaciation2,3,8,9.

Secondly, the landform record offers a window into the processes that occurred at former ice stream beds, allowing researchers study how they flowed, shifted, turned ‘on’ and ‘off’, and interacted with the landscape2.

The palaeo–ice stream record, therefore, can be used to better understand how ice streams change over long timescales and under different climate conditions, in order to improve predictions of future ice sheet change.

The palaeo–ice stream landsystem

Ice streams have three important characteristics that are reflected in the landforms they create10,11,12. First, they flow very rapidly – orders of magnitude faster than a typical valley glacier13 – by a combination of internal deformation, sliding, and subglacial deformation1,10. Second, they have convergent onset zones1,10 (onset zones are areas where ice flow changes from slow- [sheet flow] to fast-moving [stream flow] at the head of an ice stream). Third, their lateral margins are very sharp1,10.

Characteristics of an ice stream (fast-flowing ice, a convergent onset, and sharp lateral margins) displayed at Byrd Glacier, West Antarctica. Image: NASA.

Fast ice flow

Mega-scale glacial lineations are the most striking landforms created by fast ice flow in palaeo–ice streams14,15. They are streamlined sediment ridges formed at the bed in the main ice stream trunk zone16. You can think of these landforms as ‘stretched’ out flutes or drumlins, as they are similar in shape, but much larger and more elongate14,15.

In size, mega-scale glacial lineations are between 10–100 kilometres long and 200–1300 metres wide11, making it difficult to identify them on the ground. Instead, they are most easily mapped from satellite images (see below). When viewed from space, it is also obvious that mega-scale glacial lineations are not isolated features, but occur together in large groups. Within these groups, they run parallel to one another over great distances11,14,15.

Mega-scale glacial lineations formed at the bed of the Duawnt Lake palaeo-ice stream in Canada (see ref. 23). Note how individual lineations are highly elongate and closely parallel each other. In this example, the palaeo-ice stream flowed from right to left. Image: Google Earth.

Convergent onset zones

Shorter subglacial bedforms, such as flutes and drumlins, form in palaeo–ice stream onset zones, where ice velocity is lower than in the ice stream trunk zone11,17. These landforms are arranged in a fan-like pattern that flows in toward (or converges on) a narrower corridor of fast-flow landforms that include mega-scale glacial lineations.

Convergent flow in the onset zone of the Transition Bay palaeo-ice stream, Arctic Canada (see ref. 17). Also, note how this ice stream flow path (white) crosscuts an older ice stream flow path (black). Image: Google Earth.

Sharp ice stream margins

In modern ice streams, shear zones – areas of intense deformation several kilometres wide, marked by crevassing at the ice-surface18 – develop at the margins of ice streams, where fast- and slow-moving ice meet19.

Surface crevasses in a shear zone at Recovery Glacier ice stream in East Antarctica Image: NASA.

Ice stream shear margin moraines are sediment ridges deposited subglacially in the shear zone20. At first glance, they look similar to mega-scale glacial lineations, but they are generally wider and longer20. Shear margin moraines can be used to identify the edges (and thus lateral extent) of palaeo–ice streams11,12.

Shear margin moraine (arrowed) with a fast-flow assemblage (e.g. drumlins, mega-scale glacial lineations) ‘inside’ the palaeo-ice stream flow path (right of shear moraine) and ice-stagnation landforms ‘outside’ the ice stream flow path (left of shear moraine). Example from the M’Clintock Channel palaeo-ice stream in Arctic Canada (see ref. 20). Image: Google Earth.

Flow-direction changes

Ice streams do not always follow the same flow pathway; they are capable of switching flow-direction over time owing to glaciological (e.g. ice thickness) or topographic (e.g. basin infilling) changes9,21.

In the palaeo–ice stream landsystem, flow-direction changes can be mapped where one group of flow assemblages (e.g. drumlins) crosscuts another11,12,14. It is usually possible to work out the relative order of flow changes by studying the pattern of crosscutting (see the Transition Bay palaeo-ice stream diagram above).

Ice stream shutdown

While the palaeo–ice stream landsystem is dominated by features relating to fast ice-flow (e.g. mega-scale glacial lineations), these may be overprinted by other landform assemblages. For example, during deglaciation, moraine ridges and ice-stagnation landforms may be deposited over the top of fast-flow landforms as the active ice-front moves back2,11,12.

Similarly, ribbed moraines (transverse sediment ridges) may form over the top of glacial lineations22. Ribbed moraines are thought to form where ice-flow changes from an extensional (ice streaming) to a compressional regime. Where they lie on top of glacial lineations, therefore, they may record the slowing or shutdown of palaeo-ice streams during ice sheet deglaciation22.

Ribbed moraines lying on top of glacial lineations at the bed of the former Dubawnt Lake palaeo-ice stream. This ordering of landform assemblages records ice stream shutdown during deglaciation (see ref. 22).


Ice streams shape the land surface they flow over, leaving behind a distinctive landsystem11 that includes mega-scale glacial lineations, which record the passage of fast-moving ice14, convergent bedforms in onset zones, and shear margin moraines that mark their sharp lateral margins20. In addition, the palaeo–ice stream landsystem often displays evidence of dynamic ice sheet changes5,6, such as switches in flow-direction9,21 (crosscutting landforms) and velocity.

Related content

Professor Chris Clark’s Sheffield University webpages also host a wealth of information on mega-scale glacial lineations, drumlins, and ribbed moraines!


1. Bennett, M.R., 2003. Ice streams as the arteries of an ice sheet: their mechanics, stability and significance. Earth-Science Reviews61, 309-339.

2. Stokes, C.R. and Clark, C.D., 2001. Palaeo-ice streams. Quaternary Science Reviews20, 1437-1457.

3. Livingstone, S.J., Cofaigh, C.Ó., Stokes, C.R., Hillenbrand, C.D., Vieli, A. and Jamieson, S.S., 2012. Antarctic palaeo-ice streams. Earth-Science Reviews111, 90-128.

4. Margold, M., Stokes, C.R., Clark, C.D. and Kleman, J., 2015. Ice streams in the Laurentide Ice Sheet: a new mapping inventory. Journal of Maps11, 380-395.

5. Kleman, J., Hättestrand, C., Borgström, I. and Stroeven, A., 1997. Fennoscandian palaeoglaciology reconstructed using a glacial geological inversion model. Journal of glaciology43, 283-299.

6. Hughes, A.L., Clark, C.D. and Jordan, C.J., 2014. Flow-pattern evolution of the last British Ice Sheet. Quaternary Science Reviews89, 148-168.

7. Rignot, E., Velicogna, I., van den Broeke, M.R., Monaghan, A. and Lenaerts, J.T., 2011. Acceleration of the contribution of the Greenland and Antarctic ice sheets to sea level rise. Geophysical Research Letters38 (5).

8. Stokes, C.R., Margold, M., Clark, C.D. and Tarasov, L., 2016. Ice stream activity scaled to ice sheet volume during Laurentide Ice Sheet deglaciation. Nature530, 322-326.

9. Ó Cofaigh, C., Evans, D.J. and Smith, I.R., 2010. Large-scale reorganization and sedimentation of terrestrial ice streams during late Wisconsinan Laurentide Ice Sheet deglaciation. GSA Bulletin122, 743-756.

10. Clark, C.D., 1999. Glaciodynamic context of subglacial bedform generation and preservation. Annals of Glaciology28, 23-32.

11. Clark, C.D and Stokes, C.R. 2003. Palaeo-ice stream landsystem. In Evans, D.J.A. (Ed.) Glacial Landsystems. Hodder–Arnold, UK.

12. Stokes, C.R. and Clark, C.D., 1999. Geomorphological criteria for identifying Pleistocene ice streams. Annals of Glaciology28, 67-74.

13. Rignot, E., Mouginot, J. and Scheuchl, B., 2011. Ice flow of the Antarctic ice sheet. Science333, 1427-1430.

14. Clark, C.D., 1993. Mega‐scale glacial lineations and cross‐cutting ice‐flow landforms. Earth Surface Processes and Landforms18, 1-29.

15. Stokes, C.R. and Clark, C.D., 2002. Are long subglacial bedforms indicative of fast ice flow? Boreas31, 239-249.

16. King, E.C., Hindmarsh, R.C. and Stokes, C.R., 2009. Formation of mega-scale glacial lineations observed beneath a West Antarctic ice stream. Nature Geoscience2, 585-588.

17. Angelis, H.D. and Kleman, J., 2008. Palaeo‐ice‐stream onsets: examples from the north‐eastern Laurentide Ice Sheet. Earth Surface Processes and Landforms, 33, 560-572.

18. Raymond, C., 1996. Shear margins in glaciers and ice sheets. Journal of Glaciology42, 90-102.

19. Schoof, C. 2004. On the mechanics of ice-stream shear margins. Journal of Glaciology50, 208-218.

20. Stokes, C.R. and Clark, C.D., 2002. Ice stream shear margin moraines. Earth Surface Processes and Landforms27, 547-558.

21. Winsborrow, M.C., Stokes, C.R. and Andreassen, K., 2012. Ice-stream flow switching during deglaciation of the southwestern Barents Sea. GSA Bulletin124, 275-290.

22. Stokes, C.R., Lian, O.B., Tulaczyk, S. and Clark, C.D., 2008. Superimposition of ribbed moraines on a palaeo‐ice‐stream bed: implications for ice stream dynamics and shutdown. Earth Surface Processes and Landforms33, 593-609.

23. Stokes, C.R. and Clark, C.D., 2003. The Dubawnt Lake palaeo‐ice stream: evidence for dynamic ice sheet behaviour on the Canadian Shield and insights regarding the controls on ice‐stream location and vigour. Boreas32, 263-279.

Ice stream initiation on the northern Antarctic Peninsula

Please see the following work:

Glasser, N.F., B.J. Davies, J.L. Carrivick, A. Ròdes, M.J. Hambrey, J.L. Smellie and E. Domack (2014). Ice-stream initiation, duration and thinning on James Ross Island, northern Antarctic Peninsula. Quaternary Science Reviews 86, 78-88.

Download the Glasser et al. 2014 Preprint. The following is a shorter, simpler version of the paper.

Reconstructing ice stream changes in Antarctica | Why study ancient, long-gone ice streams? | Prince Gustav Ice Stream | Living on James Ross Island | What did we find? | Dynamic ice-sheet change | Further reading | References | Comments |

Reconstructing ice stream changes in Antarctica

Map of the Antarctic Peninsula, after Davies et al., 2012 (Quaternary Science Reviews)

Map of the Antarctic Peninsula, after Davies et al., 2012 (Quaternary Science Reviews). James Ross Island is located at the northern tip of the Antarctic Peninsula. Inset shows predominant ocean currents.

Antarctica is an area that is changing very rapidly. Ice streams are thinning, receding and shrinking. How will these ice streams change in the future?

If we want to understand this, we must look to the past; specifically, to another recent period of rapid climatic change. The Last Glacial-Interglacial Transition, a time of rapid warming not dissimilar to the present day, is particularly relevant.

Our work from James Ross Island, northern Antarctic Peninsula, reveals how the Antarctic Peninsula Ice Sheet changed during the Last Glacial-Interglacial Tranistion, from a thicker, cold, slow-moving ice sheet at around 18,000 years ago to a thinner, warmer, more dynamic ice sheet drained by fast-flowing ice streams after 12,000 years ago.

Why study ancient, long-gone ice streams?

Schematic figure of geomorphology on the continental shelf around the Antarctic Peninsula. From Davies et al., 2012.

Ice streams around the Antarctic Peninsula at the Last Glacial Maximum. Red lines schematically indicate mega scale glacial lineations. After Davies et al., 2012

Ice streams around Antarctica are undergoing rapid change. In recent centuries, the Siple Coast ice streams have had rapid and dynamic fluctuations in flow, including stopping flowing, whilst Pine Island Glacier is currently accelerating, thinning and receding. Predicting the wider future response of the Antarctic Ice Sheet to change requires a detailed understanding of the ice streams that dominate its dynamics.

We know that, at the Last Glacial Maximum (about 18,000 years ago), the Antarctic Peninsula Ice Sheet was drained by ice streams (see map to the right). Much of our evidence of these ice streams comes from the marine geological record; from swath bathymetry images of the sea floor, providing images of moraines and mega-scale glacial lineations, and from radiocarbon ages from microfossils from marine muds. However, these data provide only a snapshot of ice-stream behaviour during deglaciation, and don’t provide information on ice-stream initiation and dynamics.

Prince Gustav Ice Stream

In order to understand how ice streams begin, we must look to the terrestrial record. A three-person team (Neil Glasser, Bethan Davies and Jonathan Carrivick) therefore spent seven weeks on Ulu Peninsula, James Ross Island, on the Antarctic Peninsula, camping in a tiny tent, carrying around huge sacks of rocks, trying to decipher a complicated story from fragmentary evidence left behind. Ulu Peninsula is one of the largest ice-free areas of the Antarctic Peninsula, so it’s ideal for our kind of glacial geology. We wanted to use cosmogenic nuclide dating of boulders (surface exposure age dating) to define the evolution of Last Glacial Maximum ice in Prince Gustav Channel, the region between Trinity Peninsula (northern Antarctic Peninsula) and James Ross Island.

Several lines of evidence, including mega-scale glacial lineations (which are mapped on the figure below), suggest that, during the Last Glacial Maximum, Prince Gustav Channel was occupied by an ice stream, Prince Gustav Ice Stream, which flowed north and south around James Ross island, with an ice divide halfway along Prince Gustav Channel. Marine radiocarbon ages document the final recession of this ice stream, but cannot tell us whether it was a short-lived feature, or a more permanent feature of the Antarctic Peninsula Ice Sheet during the last glaciation.

James Ross Island, northern Antarctic Peninsula. Red indicates granite bedrock, the source of granite boulders on James Ross Island. Ulu Peninsula is shown by the red box.

James Ross Island, northern Antarctic Peninsula. Red indicates granite bedrock, the source of granite boulders on James Ross Island. Ulu Peninsula is shown by the red box. From Glasser et al., 2014

Living on James Ross Island

There is no way around it: if you want to collect high-quality cosmogenic nuclide samples, you have to spend time on the ground. You need to go there, count pebbles to determine sediment transport histories, sketch landforms like moraines, collect samples for particle size analysis, and mark geomorphological features like ridges and breaks in slope in great detail on your map. We were joined by Alan Hill, our mountain safety expert whose job it was to stop the clumsy scientists setting fire to their tent, getting lost in a blizzard or cutting off their foot while sampling a hard granite boulder with a hammer and chisel. We were completely cut off from society, with only a radio link to the main British research station, Rothera, 330 miles to the south by direct flight (which, I hasten to add, wasn’t possible as the ski planes couldn’t land on our rocky ground). We were dropped off, and picked up, by ship. Until our uplift date, some seven weeks hence, we were on our own.

You can use Google Maps to explore James Ross Island for yourself. Ulu Peninsula is the large ice-free peninsula projecting north of the island. If you zoom in on Ulu Peninsula, you will be able to see the flat-topped hills (basalt volcanic floods), with small, cold-based ice domes and small cirque glaciers.

View Larger Map

What did we find?

By sampling granite boulders derived from the nearby mainland Antarctic Peninsula, we were able to reconstruct ice-sheet behaviour through the last glacial cycle. Those granite boulders situated on tops of the ‘mesas’ (a flat-topped hill; that was a hike that day I can tell you!) proved to have been exposed at the Earth’s surface for over 17,700 years. This indicates that the earliest ice-sheet thinning occurred just after the Last Glacial Maximum.

We also found a drift rich in granite erratics around the margins of Prince Gustav Channel. We interpreted this drift, rich in exotic pebbles and boulders, as representing the lateral margins of Prince Gustav Ice Stream, impinging on Ulu Peninsula. Boulders associated with this drift proved to have exposure ages of around 12,000 years in the north to 6,000 years in the south of the study region. One boulder, associated with a lateral moraine from Prince Gustav Ice Stream, had a height of 144 m above sea level and an exposure age of 7,600 years.

Ulu Peninsula on James Ross Island. Glacial drift rich in granite erratics from the mainland Antarctic Peninsula is denoted by cross-hatching. Our cosmogenic nuclide ages are marked with green triangles and stars.

Ulu Peninsula on James Ross Island. Glacial drift rich in granite erratics from the mainland Antarctic Peninsula is denoted by cross-hatching. Our cosmogenic nuclide ages are marked with green triangles and stars. From Glasser et al., 2014

These exposure ages told us that, at the Last Glacial Maximum, Ulu Peninsula was overwhelmed by a  large ice sheet that originated on Trinity Peninsula. This thick, colm, slow-moving Antarctic Peninsula Ice Sheet deposited granite boulders all across James Ross Island and on nearby Seymour Island. When it became warmer in the Early Holocene (with temperatures recorded in the Mount Haddington Ice Core, see figure below), the ice sheet began to thin and recede from the continental shelf edge. This surface lowering was coincident with a dynamic change with the onset of Prince Gustav Ice Stream, which occurred after 12,000 years ago but before 18,000 years ago. The ice surface lowered in excess of 230 m during this time. The ice stream continued to impinge on the edges of Ulu Peninsula until 7,000 years ago, after which time it shrank back away from the study area. Local ice from the Mount Haddington Ice Cap remained on the inner parts of Ulu Peninsula until around 6,000 years ago.

Ice sheet thinning occurred during a period of rapid warming (from the Mount Haddington ice core) and rapid regional sea level rise.

Ice sheet thinning occurred during a period of rapid warming (from the Mount Haddington ice core) and rapid regional sea level rise. From Glasser et al., 2014.

Dynamic ice-sheet change

These results indicate that a dynamic change occurred during deglaciation, with the ice sheet switching from a thicker, cold-based style of glaciation at the Last Glacial Maximum to a warm-based, thinner, more dynamic ice sheet with ice streaming and a lower surface profile. This occurred during a period of rapid sea level rise and warming recorded in the Mount Haddington Ice Core. Oxygen isotope data from marine sediments indicate that ice streaming and rapid deglaciation also occurred from 13,000 to 12,000 years ago on the western Antarctic Peninsula, suggesting that ice-stream response was synchronous on both the western and eastern Antarctic Peninsula. This region-wide recession was coincident with increased upwelling of Circumpolar Deep Water onto the continental shelf edge – something we are again seeing around the Antarctic’s marine margins.

The Antarctic Peninsula Ice Sheet is a dynamic environment, sensitive to small changes in oceanic and atmospheric circulation. Our results have important implications for future ice dynamics, as temperatures approach those last seen during the Early Holocene and Mid-Holocene Climatic Optimum – both periods of rapid ice-sheet change. And it’s projected to get even warmer than that.

Further Reading

Pine Island Glacier

Investigating Pine Island Glacier | Why is Pine Island Glacier important? | Pine Island Glacier ice shelf | Pine Island Glacier: the longer term view | Conclusions | References | Comments |

Investigating Pine Island Glacier

A fast-flowing ice stream

Ice streams of Antarctica with Pine Island Glacier and Thwaites glacier highlighted.

Ice streams of Antarctica with Pine Island Glacier and Thwaites glacier highlighted.

Pine Island Glacier is one of the largest ice streams in Antarctica. It flows, together with Thwaites Ice Stream, into the Amundsen Sea embayment in West Antarctica, and the two ice streams together drain ~5% of the Antarctic Ice Sheet1. Pine Island Glacier flows at rates of up to 4000 m per year2. It is of interest to scientists because it is changing rapidly; it is thinning, accelerating and receding3, all of which contribute directly to sea level, and its future under a warming climate is uncertain. Pine Island Glacier is buttressed by a large, floating ice shelf, which helps to stabilise the glacier, but this ice shelf is itself thinning and recently calved a huge iceberg.

Just watch how fast the ice flows in the video below, and notice especially how the ice speeds up when it reaches the floating ice shelf.

Caption: Visualisation of ice flow in the Antarctic ice sheet model PISM-PIK. The white dots show how particles move with the ice which are initially randomly distributed over the ice surface. Colours in addition show the flow speed. By Youtube user pikff1.

An inaccessible location

Autosub near the ice (from

Autosub near the ice (from

Despite this interest, Pine Island Glacier is difficult to access. It is remote from any research bases, so flying there means making multiple short flights, making fuel depots to allow scientists to hop to the location. Low lying cloud often makes flying hazardous. The ice stream is heavily-crevassed and dangerous, so walking on it is difficult. Sea ice keeps ships away, making it difficult to access the ice stream from the ocean. However, scientists have several ingenious ways in which they can observe changes to this fragile, important ice stream. They can measure changes in ice extent and thinning from satellites4,5, and they have fired javelins loaded with sensors onto the ice surface, into places with too many crevasses for people to travel. Finally, scientists on board ships have deployed ‘Autosub’ beneath the very ice shelf, to make observations where no man can go.

Exploring Pine Island Glacier

You can use Google Earth below to explore the ice stream. Can you identify the ice shelf? If you zoom in far enough, you’ll be able to see the huge crack in the ice shelf. You can also see how the surface of both the ice stream and ice shelf is heavily crevassed, making it difficult to walk on the surface of the ice.

View Pine Island Glacier in a larger map

Why is Pine Island Glacier important?

Simplified cartoon of a tributary glacier feeding into an ice shelf, showing the grounding line (where the glacier begins to float).

Simplified cartoon of a tributary glacier feeding into an ice shelf, showing the grounding line (where the glacier begins to float).

Pine Island Glacier drains much of the marine-based West Antarctic Ice Sheet, and it has a configuration susceptible to rapid disintegration and recession. The ice sheet in this area is grounded up to 2000 m below sea level, making it intrinsically unstable6 and susceptible to rapid melting at its base, and to rapid migration of the grounding line up the ice stream7 (see Marine Ice Sheet Instability). The images below show how much of the West Antarctic Ice Sheet, especially around Pine Island Glacier, is grounded well below sea level.

Pine Island Glacier is one of the most dynamic features of the Antarctic Ice Sheet. It is buttressed by a large ice shelf that is currently thinning8, and the ice stream itself has a negative mass balance (the melting is not replaced by snowfall)3, it is flowing faster9,  and the grounding line is retreating further and further up into the bay. The grounding line receded by more than 20 km from 1996 to 20092. The ice stream is steepening, which increases the gravitational driving stress, helping it to flow faster, and there is no indication that the glacier is approaching a steady state10.

Possible future collapse?

A1B warming scenarios from the IPCC. A1B is the "Business as Usual" scenario, with emissions continuing to increase in line with present-day rates of increase.

A1B warming scenarios from the IPCC. A1B is the “Business as Usual” scenario, with emissions continuing to increase in line with present-day rates of increase. The grey bars at the right indicate the best estimate and likely range of temperatures.

Pine Island Glacier could collapse – stagnate and retreat far up into the bay, resulting in rapid sea level rise – within the next few centuries, raising global sea levels by 1.5 m11,12, out of a total of 3.3 m from the entire West Antarctic Ice Sheet13. Some studies have suggested that the entire main trunk of Pine Island Glacier could unground and become afloat within 100 years14, but more recent modelling efforts suggest that much longer timescales are needed to unground the entire trunk2. These numerical computer models indicate that annual rates of sea level rise from Pine Island Glacier could reach 2.7 cm per 100 years2. Under the A1B “Business as Usual” emissions scenario from the IPCC (2.6°C warming by 2100), Gladstone et al. (2012) predict recession over the next 200 years with huge uncertainty over the rate of retreat, and full collapse of the trunk of Pine Island Glacier during the 22nd Century remains a possibility15.

It remains difficult to assess how soon a collapse of Pine Island Glacier could occur, but a new paper by Bamber and Aspinall (2013) suggest that there is a growing view that the West Antarctic Ice Sheet could become unstable over the next 100 years16. The largest contibution to global sea level rise from the Greenland and Antarctic ice sheets combined is around 16.9 mm per year, but is more likely to be around 5.4 mm per year by 2100. This gives a total of 33 to 132 cm of global total sea level rise by 2100. Uncertainty over the future behaviour of Pine Island Glacier in West Antarctica is one of the largest constraints on accurately predicting future sea level rise16.

Current behaviour

Reprinted by permission from Macmillan Publishers Ltd: Nature (Rignot et al., 2008), copyright 2008

Reprinted by permission from Macmillan Publishers Ltd: Nature
(Rignot et al., 2008), copyright 2008

Pine Island Glacier is currently flowing very quickly and it is accelerating, causing thinning. The velocity is well above that required to maintain mass balance – so the ice stretches longitudinally, and thins vertically3. In the figure right, from Rignot et al. 2008, you can see that mass losses from Pine Island Glacier and Thwaites Glacier dominate Antarctic Ice Sheet ice losses. Mass loss from this basin doubled from 1996 to 2006, and it is the largest ice loss in Antarctica.

Pine Island Glacier ice shelf

Pine Island Glacier has a large ice shelf, which supports the glacier. Removal of the ice shelf would likely result in rapid acceleration, thinning and recession as the glacier adjusts to new boundary conditions; these reactions have been observed following ice shelf collapse around the Antarctic Peninsula17-21. The ice shelf around Pine Island Glacier is currently thinning, and it is warmed from below by Circumpolar Deep Water that flows onto the continental shelf22,23. This melts the ice shelf from below24, and this melting is probably the cause of the observed ice stream thinning, acceleration and grounding line recession25, which is contributing to a sea level rise of 1.2 mm per decade3.

Calving Icebergs

NASA’s DC-8 flies across the crack forming across the Pine Island Glacier ice shelf on Oct. 26, 2011. The ice shelf is in the midst of a natural process of calving a large iceberg, which it hasn’t done since 2001. Credit: Jefferson Beck/NASA

NASA’s DC-8 flies across the crack forming across the Pine Island Glacier ice shelf on Oct. 26, 2011. The ice shelf is in the midst of a natural process of calving a large iceberg, which it hasn’t done since 2001. Credit: Jefferson Beck/NASA

Pine Island Glacier ice shelf periodically calves huge icebergs. The ice shelf currently loses around 62.3 ± 5 Gigatonnes per year of ice through calving, and loses 101.2 ± 8 Gigatonnes per year through basal melting24. It calved a large iceberg in 2001, and in 2011 a huge rift developed on the ice shelf.  This iceberg was finally calved in July 2013. It’s about eight times the size of New York, or half the size of Greater London, at 720 km2. However, this iceberg calving event is a natural process, part of how the ice shelf regularly calves – this ice shelf spawns huge icebergs every 6-10 years. Releasing a huge iceberg, by itself, is a normal process, unrelated to warming, but increased calving may occur in the future if the ice shelf continues to thin, which would make it susceptible to plate bending and hydrofracture processes21. This threshold has yet to be passed.

Current melting, thinning and acceleration

What is concerning is the current intense melting, thinning and glacier acceleration observed on Pine Island Glacier ice shelf22. Measurements from the British Antarctic Survey’s Autosub, the intrepid sub-ice shelf explorer, help scientists understand sub-ice conditions. Autosub is a remotely operated vehicle, loaded with sensors that measure temperature, salinity, pressure and so on, and it can map the sea bed using downward-pointing swath bathymetry. It can dive to 1600 m and travel 400 km, and it has a clever collision avoidance system.  It’s a dangerous business; several iterations of Autosub have been lost under the ice. However, data from Autosubs that did return indicates that more warm Circumpolar Deep Water has been in Pine Island Bay in recent summers22. Meltwater production underneath the ice shelf increased by 50% from 1994 to 2011; this increased melting results from stronger sub-ice-shelf circulation. As the ice shelf thins, more water is able to circulate beneath it22, exacerbating the problem and encouraging further melting.

Warm ocean waters are melting a cavity beneath Pine Island Glacier

Warm ocean waters are melting a cavity beneath Pine Island Glacier. After Schoof, 2010, Nature Geosci, 3, 450-451.

Pine Island Glacier ice shelf now has one of the fastest rates of ice-shelf thinning in Antarctica24,25.

Antarctic ice shelf thickness changes. Note the rapid thinning of Pine Island Glacier ice shelf in West Antarctica. From Pritchard et al., 2012, Nature. Reprinted by permission from Macmillan Publishers Ltd: Nature (Pritchard et al. 2012), copyright (2012).

Antarctic ice shelf thickness changes. Note the rapid thinning of Pine Island Glacier ice shelf in West Antarctica. From Pritchard et al., 2012, Nature. Reprinted by permission from Macmillan Publishers Ltd: Nature
(Pritchard et al. 2012), copyright (2012).

Pine Island Glacier: the longer term view

It is important that we take a longer-term perspective of the current changes observed on Pine Island Glacier. Are these on-going changes unprecedented, or are they part of the normal behaviour for the glacier? Marine sediment cores and swath bathymetry from ships can image the sea floor and detect and date the former behaviour of this ice stream. These data suggest that the recession of this ice stream was largely controlled by sea level rise, with a 55 m in sea level rise during deglaciation resulting in 225 km of grounding-line recession26. At the Last Glacial Maximum, circa 18,000 years ago, the ice stream was at the continental shelf edge27. It rapidly shrank back from around 16400 years ago, when rising sea levels made this ice stream more buoyant, causing lift-off, decoupling from the ice sheet’s bed, and recession.  The ice stream continued to recede from 16400 to 12300 years ago, controlled by global sea level rise. It reached its current position around 10,000 years ago27.

The recession of the ice stream was also controlled by the presence or absence of ice shelves. From 12300 to 10600 years ago, there was a large ice shelf throughout the Amundsen Sea Embayment. This ice shelf collapsed after 10600 years ago28, when warmer waters flowed onto the continental shelf. The grounding line of the ice stream retreated rapidly following ice-shelf collapse26.

It seems that the glacier is capable of very rapid recession within millennial timescales27, and that the dynamics between ice shelf and ice stream are intrinsically linked.  More work at a higher resolution, combined with modelling studies, is required to fine-tune and better understand the longer-term history of Pine Island Glacier.


Pine Island Glacier is a cause for concern, because it’s thinning rapidly, steepening, accelerating and receding. It is out of balance. Huge amounts of meltwater are generated in a large cavity beneath the ice shelf. It periodically, every 10 or so years, calves large icebergs – but on their own, they are not worrisome. The recently calved iceberg may be 720 km2, but that’s the least of this ice stream’s worries. This ice stream is unlikely to collapse in our lifetime – but the same cannot be said for future generations.

Pine Island Glacier is one of the largest ice streams in Antarctica, and drains much of the West Antarctic Ice Sheet. Because it is grounded in ever deeper sea water, it is vulnerable to melting at its base and rapid grounding line migration. A collapse of Pine Island Glacier could occur within 1000-2000 years, raising sea levels by up to 1.5 m, but it is unlikely to contribute to more than 2.7 cm of sea level rise over the next 100 years.

Wider Reading

Go to top or jump to Marine Ice Sheet Instability.


1.            Vaughan, D.G., Smith, A.M., Corr, H.F.J., Jenkins, A., Bentley, C.R., Stenoien, M.D., Jacobs, S.S., Kellogg, T.B., Rignot, E. & Lucchitta, B.K. A Review of Pine Island Glacier, West Antarctica: Hypotheses of Instability Vs. Observations of Change. in The West Antarctic Ice Sheet: Behavior and Environment 237-256 (American Geophysical Union, 2001).

2.            Joughin, I., Smith, B.E. & Holland, D.M. Sensitivity of 21st century sea level to ocean-induced thinning of Pine Island Glacier, Antarctica. Geophysical Research Letters 37, L20502 (2010).

3.            Rignot, E., Bamber, J.L., van den Broeke, M.R., Davis, C., Li, Y., van de Berg, W.J. & van Meijgaard, E. Recent Antarctic ice mass loss from radar interferometry and regional climate modelling. Nature Geosci 1, 106-110 (2008).

4.            Rignot, E., Mouginot, J. & Scheuchl, B. Ice Flow of the Antarctic Ice Sheet. Science (2011).

5.            Shepherd, A., Wingham, D.J., Mansley, J.A.D. & Corr, H.F.J. Inland thinning of Pine Island Glacier, West Antarctica. Science 291, 862-864 (2001).

6.            Schoof, C. Ice sheet grounding line dynamics: Steady states, stability, and hysteresis. Journal of Geophysical Research-Earth Surface 112(2007).

7.            Mercer, J.H. West Antarctic Ice Sheet and CO2 Greenhouse effect – threat of disaster. Nature 271, 321-325 (1978).

8.            Rignot, E. Ice-shelf changes in Pine Island Bay, Antarctica, 1947-2000. Journal of Glaciology 48, 247-256 (2002).

9.            Rignot, E. Changes in ice dynamics and mass balance of the Antarctic ice sheet. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 364, 1637-1655 (2006).

10.          Scott, J.B.T., Gudmundsson, G.H., Smith, A.M., Bingham, R.G., Pritchard, H.D. & Vaughan, D.G. Increased rate of acceleration on Pine Island Glacier strongly coupled to changes in gravitational driving stress. Cryosphere 3, 125-131 (2009).

11.          Hughes, T. A simple holistic hypothesis for the self-destruction of ice sheets. Quaternary Science Reviews 30, 1829-1845 (2011).

12.          Vaughan, D.G. West Antarctic Ice Sheet collapse – the fall and rise of a paradigm. Climatic Change 91, 65-79 (2008).

13.          Bamber, J.L., Riva, R.E.M., Vermeersen, B.L.A. & Le Brocq, A.M. Reassessment of the potential sea-level rise from a collapse of the West Antarctic Ice Sheet. Science 324, 901-903 (2009).

14.          Wingham, D.J., Wallis, D.W. & Shepherd, A. Spatial and temporal evolution of Pine Island Glacier thinning, 1995-2006. Geophysical Research Letters 36, L17501 (2009).

15.        Gladstone, R.M., Lee, V., Rougier, J., Payne, A.J., Hellmer, H., Le Brocq, A., Shepherd, A., Edwards, T.L., Gregory, J. & Cornford, S.L. Calibrated prediction of Pine Island Glacier retreat during the 21st and 22nd centuries with a coupled flowline model. Earth and Planetary Science Letters 333–334, 191-199 (2012).

16.          Bamber, J. L., and Aspinall, W. P. (2013). An expert judgement assessment of future sea level rise from the ice sheets. Nature Clim. Change 3, 424-427.

17.         Scambos, T.A., Bohlander, J.A., Shuman, C.A. & Skvarca, P. Glacier acceleration and thinning after ice shelf collapse in the Larsen B embayment, Antarctica. Geophysical Research Letters 31, L18402 (2004).

18.          De Angelis, H. & Skvarca, P. Glacier surge after ice shelf collapse. Science 299, 1560-1562 (2003).

19.          Rott, H., Rack, W., Skvarca, P. & De Angelis, H. Northern Larsen Ice Shelf, Antarctica: further retreat after collapse. Annals of Glaciology 34, 277-282 (2002).

20.          Glasser, N.F., Scambos, T.A., Bohlander, J.A., Truffer, M., Pettit, E.C. & Davies, B.J. From ice-shelf tributary to tidewater glacier: continued rapid glacier recession, acceleration and thinning of Röhss Glacier following the 1995 collapse of the Prince Gustav Ice Shelf on the Antarctic Peninsula. Journal of Glaciology 57, 397-406 (2011).

21.          Scambos, T., Fricker, H.A., Liu, C.-C., Bohlander, J., Fastook, J., Sargent, A., Massom, R. & Wu, A.-M. Ice shelf disintegration by plate bending and hydro-fracture: Satellite observations and model results of the 2008 Wilkins ice shelf break-ups. Earth and Planetary Science Letters 280, 51-60 (2009).

22.          Jacobs, S.S., Jenkins, A., Giulivi, C.F. & Dutrieux, P. Stronger ocean circulation and increased melting under Pine Island Glacier ice shelf. Nature Geoscience 4, 519-523 (2011).

23.          Jenkins, A., Dutrieux, P., Jacobs, S.S., McPhail, S.D., Perrett, J.R., Webb, A.T. & White, D. Observations beneath Pine Island Glacier in West Antarctica and implications for its retreat. Nature Geoscience 3, 468-472 (2010).

24.          Rignot, E., Jacobs, S., Mouginot, J. & Scheuchl, B. Ice Shelf Melting Around Antarctica. Science (2013).

25.          Pritchard, H.D., Ligtenberg, S.R.M., Fricker, H.A., Vaughan, D.G., van den Broeke, M.R. & Padman, L. Antarctic ice-sheet loss driven by basal melting of ice shelves. Nature 484, 502-505 (2012).

26.          Kirshner, A.E., Anderson, J.B., Jakobsson, M., O’Regan, M., Majewski, W. & Nitsche, F.O. Post-LGM deglaciation in Pine Island Bay, West Antarctica. Quaternary Science Reviews 38, 11-26 (2012).

27.          Lowe, A.L. & Anderson, J.B. Reconstruction of the West Antarctic ice sheet in Pine Island Bay during the Last Glacial Maximum and its subsequent retreat history. Quaternary Science Reviews 21, 1879-1897 (2002).

28.          Jakobsson, M., Anderson, J.B., Nitsche, F.O., Dowdeswell, J.A., Gyllencreutz, R., Kirchner, N., Mohammed, R., O’Regan, M., Alley, R.B., Andandakrishnan, S., Eriksson, B., Kirshner, A., Fernandez, R., Stolldorf, T., Minzoni, R. & Majewski, W. Geological record of ice shelf break-up and grounding line retreat, Pine Island Bay, West Antarctica. Geology 39, 691-694 (2011).

Go to top or jump to Marine Ice Sheet Instability.

Mega scale glacial lineations

Sea floor landforms | The formation of Mega Scale Glacial Lineations | Mega Scale Glacial Lineations in Antarctica | References | Comments |

Sea floor landforms

Schematic figure of geomorphology on the continental shelf around the Antarctic Peninsula. From Davies et al., 2012.

Schematic figure of geomorphology on the continental shelf around the Antarctic Peninsula. From Davies et al., 2012.

The sea floor around Antarctica holds a great deal of information about past glacial behaviour, and by analysing the glacial landforms here, it is possible to obtain a great deal of information.

By using multibeam swath bathymetry (where the ship emits sonar ‘pings’ and collects their reflection from the sea floor; see information from Woods Hole), it is possible to image the sea floor. The continental shelf around the Antarctic Peninsula is a veritable treasure trove of glacial landforms, formed during the last glaciation of the region[6]. Some of these landforms indicate that the continental shelf was criss-crossed by ice streams that lay in deeper bathymetric troughs during the last glacial maximum.

In the Amundsun Sea Embayment, into which Pine Island Glacier drains, multibeam swath bathymetry has captured beautiful images of moraines, mega-scale glacial lineations (see work at Sheffield University) and drumlins[7]. The geology of the continental shelf poses a considerable control on the formation of these glacial landforms. Hard crystalline bedrock on the inner shelf have been moulded into short drumlins and incised with meltwater channels. On the outer shelf, soft sedimentary strata have been moulded into long mega-scale glacial lineations.

The formation of Mega Scale Glacial Lineations

What are Mega Scale Glacial Lineations (MSGLs)? MSGLs are long, elongated landforms made typically in soft sediments that reflect fast ice flow of an ice sheet. Indeed, MSGLs are often thought to be indicative of ice streaming, and are found all over the world on previously glaciated areas.

In a seminal paper, Clark et al. 1 described MSGLs as linear forms 10,000 to 100,000 m in length, which are most easily observed from aerial photographs or satellite images. They characteristically have length:width ratios of greater than 15:1 and have a convergent flow pattern2. MSGLs typically have a large zone of convergence, feeding into a main trunk, which then diverges again near the ice margin. The elongation ratios of some MSGLs, such as those reported in the Dubawnt Ice Stream in Canada, are up to 13 km long and have elongation ratios of up to 43:1. Cross-cutting relationships with drumlins helps scientists to reconstruct complex ice-flow pathways, which is important for understanding ice-sheet evolution.

Map showing location of modern ice streams around Antarctica, made using velocity data from Rignot et al. 2011

Map showing location of modern ice streams around Antarctica, made using velocity data from Rignot et al. 2011

Mapping the location of former MSGLs is important, because it helps scientists to understand the former configuration of the great ice sheets during the last ice age. Ice streams are important because they drain large parts of the Antarctic Ice Sheet today, and are capable of changes in their velocity and drainage.

Mega Scale Glacial Lineations in Antarctica

The East Antarctic Ice Sheet

During the Last Glacial Maximum (LGM), the East Antarctic Ice Sheet was larger (but not necessarily thicker) ice sheet than present, and in many places it extended out onto the continental shelf. Studying East Antarctic Ice Sheet retreat at the end of the last ice age, when the Earth underwent a period of rapid warming, is important for understanding and modelling how ice sheets might react to a similar rapid increase in atmospheric and oceanic temperatures in the future. Mackintosh et al. (2012) investigated ice sheet retreat around Mac. Robertson Land, East Antarctica, where they found highly attenuated MSGLs in Nielson Basin and Iceberg Alley, reflecting the presence of former ice streams. The ice sheet was grounded > 1 km below modern sea level3.

Grounding Zone Wedges (GZWs) and MSGLs from Mac. Robertson Land, East Antarctica. From Mackintosh et al. 2012 (Nature Geoscience). Reprinted by permission from Macmillan Publishers Ltd: [Nature Geoscience]. (Mackintosh et al.), copyright (2012)

Grounding Zone Wedges (GZWs) and MSGLs from Mac. Robertson Land, East Antarctica. From Mackintosh et al. 2012 (Nature Geoscience). Reprinted by permission from Macmillan Publishers Ltd: Nature Geoscience. (Mackintosh et al. 2012), copyright 2012.

MSGLs in Iceberg Alley are confined to the deeper part of the trough, but in Nielson Basin, MSGLs and therefore ice streaming occurred throughout its trough. The maximum extent of the ice sheet is indicated by grounding zone wedges, large wedges of sediment that build up when the ice margin is at a stable position for a sustained period of time3.

These Mega Scale Glacial Lineations were imaged using multibeam swath bathymetry, which you can read more about at the Woods Hole Science Centre.

The Antarctic Peninsula

Belgica Trough Mega Scale Glacial Lineations, Antarctic Peninsula.

Belgica Trough Mega Scale Glacial Lineations, Antarctic Peninsula. Figure provided by Dr Stephen Livingstone.

Around the Antarctic Peninsula, extensive MSGLs record the presence of multiple ice streams during the LGM. On the Antarctic Peninsula continental shelf, the inner shelf is typified by irregular, erosional forms cut into crystalline bedrock; the mid-shelf is characterised by drumlins and more elongated forms, and the outer continental shelf is characterised by widespread, elongated MSGLs. These MSGLs are formed in deep, soft sediments deposited on the Antarctic Peninsula continental shelf during the last glacial cycle4. The MSGLs lie in topographic lows. These bathymetric troughs served to funnel ice during the LGM, with fast-flowing ice occupying the troughs and depositing large sediment trough-mouth fans at the continental shelf edge.

Mega scale glacial lineations are particularly well developed in Belgica Trough, as you can see in the picture below, kindly provided by Dr Stephen Livingstone from Sheffield University. This figure shows drumlinoid bedforms, which are shorter and more pronounced, feeding into elongated lineations behind. These bedforms were produced in soft sediments on the ocean floor by the Belgica Trough Ice Stream during the Last Glacial Maximum on the Antarctic Peninsula.

Further Reading

There is lots of information on drumlins and MSGLs on Chris Clark’s webpages at Sheffield University.

Go to top or jump to Dating Glacial Sediments.

Antarctic Peninsula Ice Sheet evolution

Pre-Quaternary Antarctic Peninsula Ice Sheet evolution | Quaternary glaciation of the Antarctic Peninsula | Last Glacial Maximum | Antarctic Peninsula Ice streams | References | Comments |

This section is largely taken from Davies et al. 2012 (Quaternary Science Reviews)[1], and summarises Antarctic Peninsula Ice Sheet evolution throughout the Cenozoic, Last Glacial Maximum and into the Holocene.

Pre-Quaternary Antarctic Peninsula Ice Sheet evolution

Palaeogene (65.5 to 23.03 Ma)

Evidence for pre-Quaternary (see Table 1) glaciations of the Antarctic Peninsula mostly come from offshore seismic and drilling campaigns. There are some terrestrial records from King George Island, South Shetland Islands, and on James Ross Island.  The continental shelf and slope, extending beyond the reach of later Quaternary ice sheets, preserves thick sedimentary strata deposited during glacials and interglacials over the last 35-40 million years (Ma)[2]. Pre-Quaternary sediments have been dated using biostratigraphy (dinoflagellate cysts)[2], isotopic dating of volcanic rocks[3-5] and strontium isotopes analysis on shells[6, 7].

Map of the Antarctic Peninsula, after Davies et al., 2012 (Quaternary Science Reviews)

The earliest ice sheets began to develop around the Palaeogene-Neogene boundary (see Table 1), circa 35 Ma[8]. Mean global temperatures were around ~4°C higher than today[9, 10]. Mountain glaciation around the Antarctic Peninsula was initiated 37-34 Ma, coinciding with the opening of the Drake Passage, the separation of the Andes and the Antarctic Peninsula, and the development of the Antarctic Circumpolar Current[11]. The Antarctic Circumpolar Current isolated Antarctica from other regimes, resulting in the development of a cooler polar climate[12].

These early ice sheets were thin and dynamic, fluctuating at 40,000 year cycles in response to variations in the earth’s orbit around the sun (Milankovitch cycles)[13, 14]. The relatively high mountains of the Antarctic Peninsula probably acted as a nucleus for glaciation, with cooler temperatures at higher altitudes encouraging glacierisation[1, 11].

The longest terrestrial record of glaciation comes from King George Island, South Shetland Islands, with glaciers developing from the Miocene.

Neogene (23.03 to 2.54 Ma)

Table 1. Timescale in the Antarctic Peninsula, showing glacial events from the Cenozoic to the present day. APIS – Antarctic Peninsula Ice Sheet. JRI – James Ross Island. WAIS – West Antarctic Ice Sheet. Small ice caps began to develop in the area about5 million years ago. Large continental wide ice sheets began to develop during the Quaternary, with oscillations at 100,000 year periodicities after about 400,000 years ago.

Sediments from the Pacific continental margin, ~9 Ma in age, have yielded a high-resolution history of multiple ice advances and erosional episodes, indicating a persistent Antarctic Peninsula Ice Sheet[1, 15, 16]. Sedimentary evidence suggests that Pliocene ice (<3 Ma) was probably relatively thin and did not inundate the topography[17].

During this period, the West Antarctic Ice Sheet and Antarctic Peninsula Ice Sheets together grew successively larger, with periodic collapses during interglacials. During periods of West Antarctic Ice Sheet absence, the Antarctic Peninsula Ice Sheet remained as a series of island ice caps, and was also a refuge for plants and animals[2, 18, 19]. The East Antarctic Ice Sheet remained relatively stable during this time.

James Ross Island has yielded an excellent record of Neogene glaciations, preserved in glaciovolcanic sediments. The volcanic rocks are dateable, and the sequence provides an excellent record of glacial activity[1]. The volcanic sequences were formed by repeated volcanic eruptions (>50) beneath glacier ice from 9.9 to 2.6 Ma, forming pillow lavas and hyaloclastites[4, 5, 20]. They form the Hobbs Glacier Formation, which lies between Cretaceous marine sediments and the younger James Ross Island Volcanic Group. See Subglacial Volcanoes for more information on this.

Neogene glaciovolcanic outcrops on James Ross Island. From: Davies et al. 2012

These sediments indicate that Antarctic Peninsula ice expanded as far as James Ross and Seymour islands, with a polythermal regime[4, 20, 21]. Sedimentary facies indicate a climatic regime similar to  that in Svalbard today[22].

Ice thickness data can be deduced from these glaciovolcanic rocks[23].  Maximum ice thicknesses are  now well known for the Antarctic Peninsula since 7.5 Ma[4, 5, 20, 24]. Ice thicknesses were generally around 250-300 m around James Ross Island, but occasionally reached 850 m. Ice thicknesses were increasing towards the end of the Pliocene. This is covered in more detail under Subglacial Volcanoes.

Quaternary glaciation of the Antarctic Peninsula

Early Pleistocene (2.54 to 1 Ma)

Excepting a few glaciovolcanic sequences on James Ross Island, there is little terrestrial evidence of Early Pleistocene glaciation. Most of the data is from seismic profiling and coring of sediment drifts on the continental rise[17]. These sediments are dated by magnetic stratigraphy, tuned to the marine isotope record. These continental rise sediments indicate an ice-sheet dominated environment developing through the Pliocene into the Pleistocene, with increasing grounding on the continental shelf[25].

Middle Pleistocene (1Ma to 200 ka)

Simplified cartoon of a tributary glacier feeding into an ice shelf, showing the grounding line (where the glacier begins to float).

After the Mid-Pleistocene Transition at 1 Ma, ice streams began to develop on the continental shelf during this period, leading to the development of trough-mouth fans at their termini[17]. After 1 Ma, lower global sea levels encouraged grounding line advance well into the Bellingshausen Sea continental shelf. A positive feedback loop was established, with cooling leading to ice sheet growth and sea level lowering, which encouraged further ice sheet growth and cooling[26, 27]. During the Middle Pleistocene, ice sheets were reaching the continental shelf for longer, with more distinct glacial-interglacial cyclicity[28]. Marine sediments from the continental slope have less ice-rafted debris, which suggests that the ice sheet was bound by sea ice and ice shelves, which inhibited iceberg transport of glacial debris[17].


Terrapin Hill, a tuff cone on James Ross Island

The tuff cone Terrapin Hill on James Ross Island has been dated to 0.66 Ma, with a base resting on glacial material. The tuff cone morphology and sedimentology, however, indicates open marine, interglacial conditions[4, 21].

Late Pleistocene (200 ka to Last Glacial Maximum)

Late Pleistocene interglacials were characterised by ice shelves in the Larsen embayment. During the last interglacial, a smaller-than-present or absent West Antarctic Ice Sheet may explain globally higher sea levels[29-31].

Ice-rafted debris occurs in greater abundances on the continental slope and rise during interglacial periods during the Late Pleistocene, with more varied stone lithologies[32-34]. This is because warmer conditions during interglacials encouraged the collapse of ice shelves. Combined with reduced sea ice, this allowed icebergs to transport debris to the continental shelf and slope. Warmer conditions would also have encouraged faster movement and increased bedrock erosion. This is therefore analogous to the situation during the Holocene.

During Late Pleistocene glacial periods, ice volumes increased markedly along the Antarctic Peninsula[1], possibly reaching 2350 m at Mount Jackson. Glacial cycles now had a dominant periodicity of 100,000 years, with around 120 m eustatic sea level change. The Antarctic Polar Front was located further north than during earlier periods, with enhanced sea ice and reduced iceberg transport of debris[17, 34]. Thick ice streams were abundant on the continental shelf, with warm-based ice grounded on the continental shelf during glacials[35].

Last Glacial Maximum (~18,000 years ago)

Schematic figure of geomorphology on the continental shelf around the Antarctic Peninsula. From Davies et al., 2012.

There was a significant increase in ice volume during the Last Glacial Maximum[36, 37], which peaked prior to 18,000 years ago (18 ka BP). The Antarctic Peninsula Ice Sheet had an increased volume relative to today of 1.7 m eustatic sea level equivalent.

Geomorphological landforms on the continental shelf are typified by irregular, short erosional forms on the inner shelf, drumlins on the middle shelf, and elongate forms on the outer shelf[38-43]. These elongated “Mega Scale Glacial Lineations are formed in thick off lapping sequences of deformable sedimentary strata, which were deposited on the continental shelf from the Miocene onwards[17].

These landforms indicate that, at the Last Glacial Maximum, ice streams occupied bathymetric troughs and flowed out across the continental shelf around the entire Antarctic continent[44].  These topographically-controlled ice streams scoured out their bathymetric troughs throughout Pleistocene glacials, each time leaving a record of their occurrence in the trough-mouth fans on the continental rise. These ice streams drained the Antarctic Peninsula Ice Sheet and confined its thickness to less than 400 m[45].

Schematic reconstruction of the Antarctic Peninsula Ice Sheet during the Last Glacial Maximum. From: Davies et al., 2012

The figure above, from Davies et al. (2012), is a schematic map with the likely extent, disposition and behaviour of the ice sheet around 18 ka BP.

Antarctic Peninsula ice streams

Palaeo-ice streams around the Antarctic Peninsula during and after the LGM, showing isochrones of recession. From: Davies et al., 2012 (Quaternary Science Reviews).

Reconstructing past (palaeo) ice streams provides an important context for understanding their recent behaviour, controls on this behaviour, and how ice streams might behave in the future[44]. Studying the basal characteristics of Antarctic palaeo ice streams means that the role of basal topography, bedrock geology and sediment erosion, transportation and deposition can be better understood. The diagnostic sediment-landform assemblages left behind by ice streams[46] has meant that a large number of ice streams have been identified around the Antarctic continent, from both marine and terrestrial settings.

At the Last Glacial Maximum, palaeo-ice streams extended to the shelf edge in West Antarctica and in the Antarctic Peninsula, but in East Antarctica they usually were restricted to the mid-outer shelf[44]. These palaeo-ice streams occupied bathymetric troughs, and are identified by the glacial bedforms (such as mega-scale glacial lineations) in these troughs, and trough-mouth fans at their termini.

The outer-shelf zones of these cross-shelf troughs are characterised by soft, unconsolidated sediments, in which mega-scale glacial lineations and grounding zone wedges are preserved[44]. The inner shelf, instead, is generally composed of crystalline bedrock and has a higher bed roughness. Drumlins, grooved bedrock and meltwater channels are often observed here.

Where there is detailed geomorphological data available, the retreat styles of various Antarctic ice streams can be better understood. Three styles of retreat have been identified around the Antarctic Peninsula. Rapid retreat with floatation and calving results in well-preserved subglacial bedforms on the continental shelf. These include mega-scale glacial lineations[47]. Marguerite Trough Ice stream is an example of an ice stream characterised by rapid recession.

Episodic retreat is recorded by mega-scale glacial lineations that are overprinted by transverse grounding-zone wedges, each recording a pause in ice stream retreat with a stationary grounding line. An example of this would be the ice stream that extended out of the Larsen A embayment on the Antarctic Peninsula[47].

Finally, slow and steady retreat is recorded by numerous closely-spaced moraines and intermittent grounding-zone wedges[47]. In the Western Ross Sea, there are six bathymetric troughs on the continental shelf. The palaeo-ice stream was about 370 km long, with a zone of glacial deposition on the outer shelf, and erosional landforms on the inner shelf. Transverse sedimentary ridges overprint mega-scale glacial lineations throughout.  They are grounding zone wedges, and are 3-12 m high, 180 m to 8 km apart. There are also smaller moraines, 1-2 m high and 10-100 m apart[47, 48].

The geomorphological record therefore suggests that retreat varies strongly between different troughs, with three principle styles of retreat recognised. This suggests that individual ice streams respond differently to external forcings during deglaciation[47], and instead are regulated by local factors, such as drainage basin size, bathymetry and sediment supply.

The western sector of the Ross Sea is fed from two drainage basins in East Antarctica measuring 1.6 million km2 and 265,000 km2. This huge drainage basin may have meant that the outlet glaciers may have responded more slowly to external forcing. The Marguerite Bay drainage basin, in contrast, would have been of the order of 10,000 to 100,000 km2 during the LGM, and the ice streams draining this basin would have responded more rapidly to changes in external forcing[47]. Constraints such as these are important for numerical models that attempt to replicate and predict the past and future behaviour of the Antarctic Ice Sheet.

By analysing retreat styles and rates of retreat around Antarctica, we can put more recent variations into context and determine their significance. The individual characteristics of each ice stream modulates its recession. Even under the same changes in environmental conditions and external forcings, ice streams will retreat at individual rates. Ice-stream behaviour and grounding line retreat is therefore unique to every ice stream. In order to constrain future ice stream behaviour, a detailed understanding of subglacial bed properties and bed geometry is required[44].

Deglaciation and recession of the Antarctic Peninsula Ice Sheet

Schematic map showing isochrones of ice sheet and ice stream recession around the Antarctic Peninsula. From: Davies et al. 2012

Around the Antarctic Peninsula, recession from the outer shelf began at about 17.5 ka BP, from the middle shelf around 14 ka BP and the inner shelf around 11 ka BP. Ice streams receded both rapidly and episodically, depositing grounding-line wedges during periods of stand-still[40]. Radiocarbon dates from marine sediment cores across the continental shelf provide an indication of ice marginal positions during recession.


Further reading

Go to top or jump to Subglacial Volcanoes.


1.            Davies, B.J., Hambrey, M.J., Smellie, J.L., Carrivick, J.L., and Glasser, N.F., 2012. Antarctic Peninsula Ice Sheet evolution during the Cenozoic Era. Quaternary Science Reviews, 2012. 31(0): p. 30-66.

2.            Anderson, J.B., Wamy, S., Askain, R.A., Wellner, J.S., Bohaty, S.M., Kirshner, A., Livsey, D.L., Simms, A.R., Smith, T.A., Ehrmann, W., Lawver, L.A., Barbeau, D.L., Wise, S.W., Kuhlhenek, D.K., Weaver, F.M., and Majewski, W., 2011. Progressive Cenozoic cooling and the demise of Antarctica’s last refugium. Proceedings of the National Academy of Sciences, 2011. 108: p. 11356-11360.

3.            Smellie, J.L., Hole, M.J., and Nell, P.A.R., 1993. Late Miocene valley-confined subglacial volcanism in northern Alexander Island, Antarctic Peninsula. Bulletin of Volcanology, 1993. 55: p. 273-288.

4.            Smellie, J.L., Johnson, J.S., McIntosh, W.C., Esser, R., Gudmundsson, M.T., Hambrey, M.J., and van Wyk de Vries, B., 2008. Six million years of glacial history recorded in volcanic lithofacies of the James Ross Island Volcanic Group, Antarctic Peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology, 2008. 260(1-2): p. 122-148.

5.            Smellie, J.L., McArthur, J.M., McIntosh, W.C., and Esser, R., 2006. Late Neogene interglacial events in the James Ross Island region, northern Antarctic Peninsula, dated by Ar/Ar and Sr-isotope stratigraphy. Palaeogeography, Palaeoclimatology, Palaeoecology, 2006. 242(3-4): p. 169-187.

6.            . 1931. Contributions to the Geology of Northumberland and Durham: Written for the Summer Field Meeting, 1931. Proceedings of the Geologists’ Association, 1931. 42(3): p. 217-296, IN1-IN2.

7.            McArthur, J.M., Rio, D., Marenssi, F., Castradori, D., Bailey, T.R., Thirlwall, M.F., Houghton, S., and Dingle, R.V., 2007. A revised Pliocene record for marine 87Sr/86Sr used to date an interglacial event, Cockburn Island Formation, northern Antarctic Peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology, 2007. 242: p. 126-136.

8.            Siegert, M.J. and Florindo, F., 2009. Antarctic climate evolution, in Antarctic Climate Evolution, F. Florindo and M.J. Siegert, Editors. Elsevier: Rotterdam. p. 2-11.

9.            DeConto, R.M. and Pollard, D., 2003. Rapid Cenozoic glaciation of Antarctica induced by declining atmospheric CO2. Nature, 2003. 421(6920): p. 245-249.

10.          Mayewski, P.A., Meredith, M.P., Summerhayes, C.P., Turner, J., Worby, A., Barrett, P.J., Casassa, G., Bertler, N.A.N., Bracegirdle, T., Naveira Garabato, A.C., Bromwich, D., Campell, H., Hamilton, G.S., Lyons, W.B., Maasch, K.A., Aoki, S., Xiao, C., and van Ommen, T., 2009. State of the Antarctic and Southern Ocean climate system. Reviews of Geophysics, 2009. 47(RG1003): p. 1-38.

11.          Siegert, M.J., 2008. Antarctic subglacial topography and ice-sheet evolution. Earth Surface Processes and Landforms, 2008. 33: p. 646-660.

12.          Eagles, G. and Livermore, R.A., 2002. Opening history of Powell Basin, Antarctic Peninsula. Marine Geology, 2002. 185(3-4): p. 195-205.

13.          Naish, T., Powell, R., Levy, R., Wilson, G., Scherer, R., Talarico, F., Krissek, L., Niessen, F., Pompilio, M., Wilson, T., Carter, L., DeConto, R., Huybers, P., McKay, R., Pollard, D., Ross, J., Winter, D., Barrett, P., Browne, G., Cody, R., Cowan, E., Crampton, J., Dunbar, G., Dunbar, N., Florindo, F., Gebhardt, C., Graham, I., Hannah, M., Hansaraj, D., Harwood, D., Helling, D., Henrys, S., Hinnov, L., Kuhn, G., Kyle, P., Laufer, A., Maffioli, P., Magens, D., Mandernack, K., McIntosh, W., Millan, C., Morin, R., Ohneiser, C., Paulsen, T., Persico, D., Raine, I., Reed, J., Riesselman, C., Sagnotti, L., Schmitt, D., Sjunneskog, C., Strong, P., Taviani, M., Vogel, S., Wilch, T., and Williams, T., 2009. Obliquity-paced Pliocene West Antarctic ice sheet oscillations. Nature, 2009. 458(7236): p. 322-U84.

14.          Naish, T.R., Woolfe, K.J., Barrett, P.J., Wilson, G.S., Atkins, C., Bohaty, S.M., Bucker, C.J., Claps, M., Davey, F.J., Dunbar, G.B., Dunn, A.G., Fielding, C.R., Florindo, F., Hannah, M.J., Harwood, D.M., Henrys, S.A., Krissek, L.A., Lavelle, M., van der Meer, J., McIntosh, W.C., Niessen, F., Passchier, S., Powell, R.D., Roberts, A.P., Sagnotti, L., Scherer, R.P., Strong, C.P., Talarico, F., Verosub, K.L., Villa, G., Watkins, D.K., Webb, P.N., and Wonik, T., 2001. Orbitally induced oscillations in the East Antarctic ice sheet at the Oligocene/Miocene boundary. Nature, 2001. 413(6857): p. 719-723.

15.          Bart, P.J. and Anderson, J.B., 2000. Relative temporal stability of the Antarctic ice sheets during the late Neogene based on the minimum frequency of outer shelf grounding events. Earth and Planetary Science Letters, 2000. 182(3-4): p. 259-272.

16.          Pudsey, C.J., 2002. Neogene record of Antarctic Peninsula glaciation in continental rise sediments: ODP Leg 178, Site 1095, in Ocean Drilling Program Scientific Results, Vol. 178, P.F. Barker, et al., Editors. Texas A&M University: College Station, Texas. p. 1-25 (CD-ROM).

17.          Cowan, E.A., Hillenbrand, C.-D., Hassler, L.E., and Ake, M.T., 2008. Coarse-grained terrigenous sediment deposition on continental rise drifts: A record of Plio-Pleistocene glaciation on the Antarctic Peninsula. Palaeogeography, Palaeoclimatology, Palaeoecology, 2008. 265(3-4): p. 275-291.

18.          Convey, P., Gibson, J.A.E., Hillenbrand, C.-D., Hodgson, D.A., Pugh, P.J.A., Smellie, J.L., and Stevens, M.I., 2008. Antarctic terrestrial life – challenging the history of the frozen continent? Biological Reviews, 2008. 83(2): p. 103-117.

19.          Convey, P., Stevens, M.I., Hodgson, D.A., Smellie, J.L., Hillenbrand, C.-D., Barnes, D.K.A., Clarke, A., Pugh, P.J.A., Linse, K., and Cary, S.C., 2009. Exploring biological constraints on the glacial history of Antarctica. Quaternary Science Reviews, 2009. 28(27-28): p. 3035-3048.

20.          Smellie, J.L., Haywood, A.M., Hillenbrand, C.-D., Lunt, D.J., and Valdes, P.J., 2009. Nature of the Antarctic Peninsula Ice Sheet during the Pliocene: Geological evidence and modelling results compared. Earth-Science Reviews, 2009. 94(1-4): p. 79-94.

21.          Hambrey, M.J., Smellie, J.L., Nelson, A.E., and Johnson, J.S., 2008. Late Cenozoic glacier-volcano interaction on James Ross Island and adjacent areas, Antarctic Peninsula region. Geological Society of America Bulletin, 2008. 120(5-6): p. 709-731.

22.          Glasser, N.F. and Hambrey, M.J., 2001. Styles of sedimentation beneath Svalbard valley glaciers under changing dynamic and thermal regimes. Journal of the Geological Society, London, 2001. 158: p. 697-707.

23.          Smellie, J.L., Rocchi, S., and Armienti, P., 2011. Late Miocene volcanic sequences in northern Victoria Land, Antarctica: products of glaciovolcanic eruptions under different thermal regimes. Bulletin of Volcanology, 2011. 73(1): p. 1-25.

24.          Smellie, J.L., McIntosh, W.C., and Esser, R., 2006. Eruptive environment of volcanism on Brabant Island: Evidence for thin wet-based ice in northern Antarctic Peninsula during the Late Quaternary. Palaeogeography, Palaeoclimatology, Palaeoecology, 2006. 231(1-2): p. 233-252.

25.          Smith, T.R. and Anderson, J.B., 2010. Ice-sheet evolution in James Ross Basin, Weddell Sea margin of the Antarctic Peninsula: The seismic stratigraphic record. GSA Bulletin, 2010. 122(5/6): p. 830-842.

26.          Huybrechts, P., 1990. A 3-D model for the Antarctic ice sheet: a sensitivity study on the glacial-interglacial contrast. Climate Dynamics, 1990. 5: p. 79-92.

27.          Barker, P.F., Barrett, P.J., Cooper, A.K., and Huybrechts, P., 1999. Antarctic glacial history from numerical models and continental margin sediments. Palaeogeography Palaeoclimatology Palaeoecology, 1999. 150(3-4): p. 247-267.

28.          Hillenbrand, C.-D. and Ehrmann, W., 2005. Late Neogene to Quaternary environmental changes in the Antarctic Peninsula region: evidence from drift sediments. Global and Planetary Change, 2005. 45(1-3): p. 165-191.

29.          Birkenmajer, K., 1982. Pliocene tillite-bearing sucession of King George Island (South Shetland Islands, Antarctica). Studia Geologica Polonica, 1982: p. 77-72.

30.          Mercer, J.H., 1978. West Antarctic Ice Sheet and CO2 Greenhouse effect – threat of disaster. Nature, 1978. 271(5643): p. 321-325.

31.          Overpeck, J.T., Otto-Bliesner, B., Miller, G.H., Muhs, D.R., Alley, R.B., and Kiehl, J.T., 2006. Palaeoclimatic evidence for future ice sheet instability and rapid sea level rise. Science, 2006. 311(no. 5768): p. 1747-1750.

32.          Pudsey, C.J., Barker, P.F., and Larter, R.D., 1994. Ice sheet retreat from the Antarctic Peninsula shelf. Continental Shelf Research, 1994. 14(15): p. 1647-1675.

33.          Pudsey, C.J., 2000. Sedimentation on the continental rise west of the Antarctic Peninsula over the last three glacial cycles. Marine Geology, 2000. 167: p. 313-338.

34.          Ó Cofaigh, C., Dowdeswell, J.A., and Pudsey, C.J., 2001. Late Quaternary Iceberg Rafting along the Antarctic Peninsula Continental Rise and in the Weddell and Scotia Seas. Quaternary Research, 2001. 56: p. 308-321.

35.          Reinardy, B.T.I., Pudsey, C.J., Hillenbrand, C.-D., Murray, T., and Evans, J., 2009. Contrasting sources for glacial and interglacial shelf sediments used to interpret changing ice flow directions in the Larsen Basin, Northern Antarctic Peninsula. Marine Geology, 2009. 266(1-4): p. 156-171.

36.          Huybrechts, P., 2009. GLOBAL CHANGE West-side story of Antarctic ice. Nature, 2009. 458(7236): p. 295-296.

37.          Huybrechts, P., 2002. Sea-level changes at the LGM from ice-dynamic reconstructions of the Greenland and Antarctic ice sheets during the glacial cycles. Quaternary Science Reviews, 2002. 21(1-3): p. 203-231.

38.          Wellner, J.S., Heroy, D.C., and Andersen, J.B., 2006. The death mask of the Antarctic ice sheet: comparison of glacial geomorphic features across the continental shelf. Geomorphology, 2006. 75: p. 157-171.

39.          Ó Cofaigh, C., Dowdeswell, J.A., Allen, C.S., Hiemstra, J.F., Pudsey, C.J., Evans, J., and Evans, D.J.A., 2005. Flow dynamics and till genesis associated with a marine-based Antarctic palaeo-ice stream. Quaternary Science Reviews, 2005. 24(5-6): p. 709-740.

40.          Ó Cofaigh, C., Dowdeswell, J.A., Evans, J., and Larter, R.D., 2008. Geological constraints on Antarctic palaeo-ice-stream retreat. Earth Surface Processes and Landforms, 2008. 33(4): p. 513-525.

41.          Ó Cofaigh, C., Justin, T., Julian A, D., and Carol J, P., 2003. Palaeo-ice streams, trough mouth fans and high-latitude continental slope sedimentation. Boreas, 2003. 32: p. 37-55.

42.          Ó Cofaigh, C., Larter, R.D., Dowdeswel, J.A., Hillenbrand, C.-D., Pudsey, C.J., Evans, J., and Morris, P., 2005. Flow of the West Antarctic Ice Sheet on the continental margin of the Bellingshausen Sea at the Last Glacial Maximum. Journal of Geophysical Research, 2005. 110: p. B11103.

43.          Ó Cofaigh, C., Pudsey, C.J., Dowdeswel, J.A., and Morris, P., 2002. Evolution of subglacial bedforms along a palaeo-ice stream, Antarctic Peninsula continental shelf. Geophysical Research Letters, 2002. 29: p. 41-1 – 41-4.

44.          Livingstone, S.J., O Cofaigh, C., Stokes, C.R., Hillenbrand, C.-D., Vieli, A., and Jamieson, S.S.R., 2012. Antarctic palaeo-ice streams. Earth-Science Reviews, 2012. 111(1-2): p. 90-128.

45.          Bindschadler, R., 2006. The environment and evolution of the West Antarctic ice sheet: setting the stage. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 2006. 364(1844): p. 1583-1605.

46            Stokes, C.R. and C.D. Clark, 1999. Geomorphological criteria for identifying Pleistocene ice streams. Annals of Glaciology, 28: 67-74.

47.           Ó Cofaigh, C., J.A. Dowdeswell, J. Evans, and R.D. Larter, 2008. Geological constraints on Antarctic palaeo-ice-stream retreat. Earth Surface Processes and Landforms, 33(4): 513-525.

48.          Shipp, S.S., J.S. Wellner, and J.B. Anderson, 2002. Retreat signature of a polar ice stream: subglacial geomorphic features and sediments from the Ross Sea, Antarctica. Geological Society of America Bulletin, 111: 1486-1516.

Ice streams

What is an ice stream? | Ice streams around Antarctica | Siple Coast ice streams | Ice stream structures | Ice stream geomorphology | References | Comments |

What is an ice stream?

Ice streams are corridors of fast flow within an ice sheet (ca. 800 metres per year). They discharge most of the ice and sediment from these ice sheets, flowing orders of magnitude faster than their surrounding ice. Their behaviour and stability is therefore essentially important to overall ice sheet dynamics and mass balance[1].  The Antarctic Ice Sheet currently discharges 90% of ice and sediment through ice streams. Antarctic Ice Streams are fed by complex tributaries that extend up to 1000 km into the interior of the ice sheet[2]. These can be seen beautifully in the video below, released by NASA:

Ice streams are typically large features (> 20 km in width, >150 km in length), with a convergent onset zone feeding in to a main channel[3]. Modern ice streams are associated with pervasively deformed till and offshore trough-mouth fans, depo-centres for the large volumes of sediment that are transported from the interior of the ice sheet outwards to the continental shelves.

Ice streams can be constrained by topography or by areas of slow moving ice. They are called topographic ice streams or pure ice streams respectively. Both types show variations in behaviour (both through time and space), which indicates potential for instability and are therefore particularly interesting[1]. Their discharge of ice into ocean basins effects thermal and saline ocean circulation. Ice streams have therefore been a focus for research worldwide over the last 30 years.

Ice streams tend to occupy topographic lows, because:

  • thicker ice leads to greater driving stress at the bed and faster velocity, because internal deformation of ice is controlled by basal shear stress[1, 4];
  • thicker ice is better insulated and has greater basal temperatures, enhancing rates of ice deformation and bed slip from basal melting;
  • Meltwater flows towards and accumulates in topographic lows, and melt rate is greater beneath thicker ice, both of which encourage basal sliding.
  • This positive feedback system, with enhanced flow increasing temperature and basal lubrication, which in turn increases flow, leads to ice stream development in topographic corridors.

Ice streams can also develop in areas with weaker ice, or with a lubricated bed to aid basal motion[1]. Some ice streams are a combination of topographic and pure, bounded by both ice and topography. There is growing evidence that soft deformable sediments are a pre-requisite for fast ice flow; subglacial geology therefore is essential in determining ice stream location[5].

Ice streams lower surface topography, with greater ice-sheet drawdown for pure ice streams, which tend to have greater ice flow volumes. Pure ice streams are also likely to be variable through time and space, shifting location and switching on and off.

The flow velocity, thickness and grounding lines of ice streams are variable over decadal timescales, with observations in Antarctica of thinning, acceleration, deceleration, stagnation and lateral migration[6-9]. However, mechanisms controlling this fast and variable flow are complex and poorly understood[10]. There are a number of potential forcings, which include ocean temperatures, sea level changes, air temperatures, ocean tides, subglacial bathymetry, subglacial geomorphology, topographic pinning points, meltwater beneath the ice stream, thermodynamics and the size of the drainage basin[6].

 Ice streams around Antarctica

Map showing location of modern ice streams around Antarctica, made using velocity data from Rignot et al. 2011

The velocity map by Eric Rignot[11], showing ice velocities in 2007-2009, shows how the Antarctic continent today is drained by ice streams, with tributary glaciers reaching hundreds to thousands of kilometres inland. These dendritic drainage systems pass ice from the interior, near the ice divide, and flow into the ocean or ice shelves.

Siple Coast ice streams

The ice streams around Siple Coast in West Antarctica (Ice Streams A to F) discharge 40% of the ice from the entire West Antarctic Ice Sheet[12]. The behaviour of these ice streams is of particular interest, because they may be important to the stability of the West Antarctic Ice Sheet (see Marine Ice Sheet Instability)[1, 13]. These ice streams are the world’s only current pure ice streams (except perhaps in NE Greenland). Other glaciers draining into the Ross Ice Shelf are topographically constrained[1].

Ice streams around Siple Coast, using velocity data from Rignot et al. 2011

These ice streams are 50 km wide, 300-500 km long, with ice thicknesses ~1 km. Ice velocities are between 0.1 and 0.8 km per year[1]. There are lateral shear zones along the margins of each ice stream. There are many crevasses near the shear zone as a result of intense deformation.  In between the ice streams the glacier ice is cold and frozen to the bed[14]. Deformable subglacial sediments seem to be a requirement for ice-stream formation on the Siple Coast, with continuous sedimentary basins below the accumulation areas of Ice Streams C and D[15]. The adjacent non-streaming areas overlie harder bedrock, with thin or no basal sediments[5].

The velocity of these ice streams is variable. For example, there is evidence of deceleration on Ice Stream B (Whillans)[16]. Ice Stream C shut down ~150 years ago[7, 17]. Ice Stream D, which currently flows rapidly, shut down ~450 years ago[18]. This is because these wide, pure ice streams are inherently unstable. The glaciers are currently thinning, which may reduce driving stress, thus explaining some of the deceleration[16]. However, in general, the accumulation areas of these ice streams are thickening[17]. Ice Stream C has a strongly positive mass balance because of its negative outflow, and it is the stoppage of this ice stream that has contributed to the positive mass balances[17]. The positive imbalance is therefore driven by internal ice-stream dynamics. Ice flow in the area that once discharged into Ice Stream C now drains into Ice Stream B (Whillans), following thinning of Ice Stream B[18]. During these rapid changes, the Siple Coast grounding line has remained static, rather than undergoing continuous change[19]. These grounding lines may be prone to rapid, rather than continuous recession – see Marine Ice Sheet Instability.

These ice streams are highly variable over short timescales, which makes it difficult to draw meaningful conclusions for short-term observations. Analysis and ice sheet reconstructions over centennial to millennial timescales are therefore very important in analysing cryospheric response to modern environmental change.

Ice stream geomorphology

A classic paper by Chris Stokes and Chris Clark from 1999[3] suggests that the geomorphological record provides diagnostic criteria for identifying palaeo-ice streams. Understanding the locations and dynamics of palaeo-ice streams is important for understanding palaeo-ice sheets. This is because their large ice flux would have effected ice-sheet configurations; investigations on former ice-streams helps understand glacial processes; their interactions with climate help reconstruct past climate change, as well as predicting the response of contemporary ice sheets to future climatic perturbations; their sedimentary flux is comparable with the largest fluvial basins[3].

Palaeo-ice streams leave characteristic features in the sedimentological and geomorphological record, which are summarised in the table below (after Stokes and Clark, 1999).

Contemporary ice stream characteristic Geomorphological signature
Characteristic shape and dimensions
  • Characteristic shape and dimensions
  • Convergent flow patterns
Rapid velocity
  • Highly attenuated bedforms (length to width ratio of 10:1)
  • Boothia-type erratic dispersal trains
Sharply delineated shear margin
  • Abrupt lateral margins
  • Lateral shear margins
Deformable bed conditions
  • Glaciotectonic and geotechnical evidence of pervasively deformed till
  • Submarine till delta, sediment fan or trough-mouth fan

Hypothetical ice stream and Boothia-type erratic dispersal. After: Stokes and Clark, 2001.

Ice stream landsystem

The palaeo-landsystem left behind by an ice stream includes mega-scale glacial lineations (MSGLs) and highly attenuated drumlins. Ice-stagnation features may overprint these landforms as an ice stream switches off or recedes[22].

On the sea floor, grounding zone wedges indicate past pauses in ice stream recession, and scours made by icebergs document the travel of icebergs across the shallow continental shelf.

Further Reading

Go to top or jump to Glacier Hydrology.


1.            Bennett, M.R., 2003. Ice streams as the arteries of an ice sheet: their mechanics, stability and significance. Earth-Science Reviews, 61(3-4): 309-339.

2.            Bamber, J.L., D.G. Vaughan, and I. Joughin, 2000. Widespread Complex Flow in the Interior of the Antarctic Ice Sheet. Science, 287(5456): 1248-1250.

3.            Stokes, C.R. and C.D. Clark, 1999. Geomorphological criteria for identifying Pleistocene ice streams. Annals of Glaciology, 28: 67-74.

4.            Cuffey, K.M. and W.S.B. Paterson, 2010. The Physics of Glaciers, 4th edition: Academic Press. 704.

5.            Winsborrow, M.C.M., C.D. Clark, and C.R. Stokes, 2010. What controls the location of ice streams? Earth-Science Reviews, 103: 45-59.

6.            Livingstone, S.J., C. O Cofaigh, C.R. Stokes, C.-D. Hillenbrand, A. Vieli, and S.S.R. Jamieson, 2012. Antarctic palaeo-ice streams. Earth-Science Reviews, 111(1-2): 90-128.

7.            Retzlaff, R. and C.R. Bentley, 1993. Timing of stagnation of Ice Stream C, West Antarctica, from short-pulse-radar studies of buried crevasses. Journal of Glaciology, 39: 553-561.

8.            Rignot, E., 2008. Changes in West Antarctic ice stream dynamics observed with ALOS PALSAR data. Geophysical Research Letters, 35(12).

9.            Joughin, I. and S. Tulaczyk, 2003. Basal melt beneath Whillans Ice Stream and Ice Streams A and C, West Antarctica. Annals of Glaciology, 36: 257-262.

10.          Vaughan, D.G. and R. Arthern, 2007. Why is it had to predict the future of ice sheets? Science, 315: 1503-1504.

11.          Rignot, E., J. Mouginot, and B. Scheuchl, 2011. Ice Flow of the Antarctic Ice Sheet. Science.

12.          Price, S.F., R.A. Bindschadler, C.L. Hulbe, and I.R. Joughin, 2001. Post-stagnation behaviour in the upstream regions of Ice Stream C, West Antarctica. Journal of Glaciology, 47: 283-294.

13.          Alley, R.D. and R.A. Bindschadler, The West Antarctic Ice Sheet and sea-level change, in The West Antarctic Ice Sheet: Behaviour and Environment. Antarctic Research Series, vol. 77, R.D. Alley and R. Bindschadler, Editors. 2001, American Geophysical Union: Washington, DC. 1-11.

14.          Bentley, C.R., N. Lord, and C.H. Liu, 1998. Radar reflections reveal a wet bed beneath stagnat Ice Stream C and a frozen bed beneath ridge BC, West Antarctica. Journal of Glaciology, 44: 149-156.

15.          Peters, L.E., S. Anandakrishnan, R.B. Alley, J.P. Winberry, D.E. Voigt, A.M. Smith, and D.L. Morse, 2006. Subglacial sediments as a control on the onset and location of two Siple Coast ice streams, West Antarctica. J. Geophys. Res., 111(B1): B01302.

16.          Joughin, I., S. Tulaczyk, R. Bindschadler, and S.F. Price, 2002. Changes in west Antarctic ice stream velocities: Observation and analysis. J. Geophys. Res., 107(B11): 2289.

17.          Joughin, I. and S. Tulaczyk, 2002. Positive Mass Balance of the Ross Ice Streams, West Antarctica. Science, 295(5554): 476-480.

18.          Conway, H., G. Catania, C.F. Raymond, A.M. Gades, T. Scambos, and H. Englehardt, 2002. Switch of flow direction in an Antarctic ice stream. Nature, 419: 465-467.

19.          Horgan, H.J. and S. Anandakrishnan, 2006. Static grounding lines and dynamic ice streams: Evidence from the Siple Coast, West Antarctica. Geophys. Res. Lett., 33(18): L18502.

20.          Glasser, N.F. and G.H. Gudmundsson, 2012. Longitudinal surface structures (flowstripes) on Antarctic glaciers. The Cryosphere, 6: 383-391.

21.          Gudmundsson, G.H., C.F. Raymond, and R. Bindschadler, 1998. The origin and longevity of flow stripes on Antarctic ice streams. Annals of Glaciology, 27: 145-152.

22.          Stokes, C.R. and C.D. Clark, 2001. Palaeo-ice streams. Quaternary Science Reviews, 20(13): 1437-1457.

23.          Ó Cofaigh, C., J.A. Dowdeswell, J. Evans, and R.D. Larter, 2008. Geological constraints on Antarctic palaeo-ice-stream retreat. Earth Surface Processes and Landforms, 33(4): 513-525.

24.          Shipp, S.S., J.S. Wellner, and J.B. Anderson, 2002. Retreat signature of a polar ice stream: subglacial geomorphic features and sediments from the Ross Sea, Antarctica. Geological Society of America Bulletin, 111: 1486-1516.