The Larsen C Ice Shelf growing rift

Antarctic Peninsula ice shelves | Ice shelf collapse on the Antarctic Peninsula | Rifting on Larsen C | Impact of calving the large iceberg | Sea level rise following ice-shelf collapse | References | Comments |

Antarctic Peninsula ice shelves

The Antarctic Peninsula is fringed by floating ice shelves. They are floating extensions of the glaciers on land, and receive mass by snowfall and marine freeze-on. They lose mass by melting at their base and by calving icebergs. Larsen C Ice Shelf, the largest ice shelf on the Antarctic Peninsula, is currently being closely watched. Following a series of high-profile ice-shelf collapse events on the Antarctic Peninsula over the last few decades, all eyes are watching Larsen C and wondering when, and if, it will collapse.

A growing rift on Larsen C Ice Shelf

Those concerns are growing more acute as a large rift on Larsen C Ice Shelf is growing rapidly, threatening to soon calve a huge iceberg, equivalent to losing 10% of the area of the ice shelf. This could destabilise the ice shelf, making it more susceptible to a total collapse.

The ice shelf, which is approximately the size of Wales, can be seen in the embedded Google Maps below.

Larsen C Ice Shelf is around 51,000 km² in size. It has been receding for some decades, but much more slowly than the ice shelves to the north of it ¹. It shrank by 5000 km² between the first satellite observations, in the 1960s, and 2009. Throughout the last few decades, it has shown little evidence of climate-driven retreat, with variations in the ice front following normal fluctuations.

However, the ice shelf has been thinning, with surface lowering observed by analysis of satellite images², likely due to basal melting and enhanced melting of snow on the ice-shelf surface. A large and growing rift at its front suggests it will soon calve the largest ever recorded iceberg.

Ice shelf collapse on the Antarctic Peninsula

This seemingly stable behaviour contrasts with the ice shelves to the north, the Prince Gustav, Larsen A and Larsen B ice shelves, which have disintegrated over the last few decades. Prince Gustav Ice Shelf retreated throughout the second half of the 20^th Century, before rapidly disintegrating in 1995¹. Larsen A disintegrated in 1995, following a previous period of stability from the 1980s. 2000 km² of ice were lost in just a few weeks, breaking up into thousands of ice bergs that were discharged into the Weddell Sea.

Larsen B collapsed even more dramatically in 2002, with only 2400 km² of its original (in 1963) 12000 km² remaining¹. This ice shelf was likely present for the entire last 10,000 years, and this is probably the first time it has collapsed³.

Ice shelves buttress and support glaciers on land

Ice-shelf collapse on the Antarctic Peninsula does not directly contribute to sea level, because the ice is already floating. However, ice shelves play an important role in controlling the balance between ice lost and gained (the mass balance) in Antarctica. Basal melting of ice shelves is one of the principle mechanisms by which ice mass is lost in Antarctica⁴. The rate at which ice shelves on the Antarctic Peninsula melt is increasing, resulting in ice-shelf thinning⁵.

Importantly, ice shelves play an important role in supporting and buttressing the glaciers on land that flow into them. When the ice shelf is removed, the glaciers respond by flowing faster, receding and thinning⁶. This is unsustainable, so the glaciers discharge large amounts of ice to the oceans^7,8. This is why ice shelves have been called the “Gatekeepers of Antarctica’s Ice”.

What causes ice shelf collapse?

These past ice-shelf collapse events have been attributed to climate change. They are a visual and dramatic sign of a changing world. Increased upwelling of warm, deep waters onto the Antarctic continental shelf is melting the ice shelves from below, causing them to thin⁹. This increased upwelling has been related to changes in the Southern Hemisphere Westerly winds¹⁰.

These thinned ice shelves are more vulnerable to collapse, which occurs when warm summers promote increased meltwater ponding on the ice-shelf surface. This seems to occur when the mean annual air temperature reaches -9°C^1,11,12. These meltwater ponds infill crevasses within the ice, causing them to melt downwards and deepen due to the increased pressure¹³. If the ice-shelf is thin enough and there is enough meltwater, the crevasses can penetrate the entire depth of the ice shelf, inducing catastrophic iceberg calving^14-16.

The Larsen B Ice Shelf disintegrated along structurally controlled lines, with softer, thinner, suture-zone ice from tributary glaciers calving more easily. In essence, the ice shelf was ripped apart along structural lines¹⁷ following a period of rifting and large iceberg calving over several years prior to collapse.  The stress field in Larsen B Ice Shelf was changed following the loss of a stabilising frontal portion in 1995. After this stabilising frontal portion was removed through calving, the ice shelf rapidly accelerated and ultimately collapsed¹⁸.

Rifting on Larsen C Ice Shelf

The long period of stability on Larsen C might be about to change. A rift has grown dramatically across Larsen C Ice Shelf. The rift reached soft suture-zone ice in 2012, and progressed in 2013 into a region that previously resisted transverse fractures such as this¹⁹. Since then, the rift has grown rapidly. The ice shelf is being monitored by Project Midas, and as of 31^st May 2017 the rift was only 13 km from the ice front. The iceberg, which is about to calve imminently, will be one of the largest-ever recorded icebergs at 5000 km². Ice flow velocity beyond the rift has increased markedly, and the rift has now reached the softer, suture-zone ice. Calving is therefore likely to happen soon.

This rift has always been present, but in the last few years and in 2017 particularly, it has deepened to full ice-shelf thickness and grown wider. When the iceberg calves, the ice shelf will be at its smallest in recorded history, having lost 10% of its area.

This gif, created by Prof. Adrian Luckman and reposted with permission, shows the increases in velocity of the Larsen C iceberg beyond the rift.

Impact of Larsen C Ice Shelf calving the large iceberg

Thinning and recession?

This process of meltwater ponding infilling crevasses does not seem to be the process ongoing on Larsen C Ice Shelf. Rather, a large rift is propagating towards the margin. However, we do know that this mechanism occurred on Larsen B several years prior to the eventual demise of the ice shelf¹⁷. We also note that the first stage of ice-shelf collapse, thinning by basal melting, does seem to be underway^2,5. Across Larsen B, C and D, the regional rate of ice-shelf thinning is 3.8±1.1 m per decade, but the highest thinning rates on Larsen C Ice Shelf reach 16.6 ±8.1 m per decade⁵. Ice-shelf volume loss on the Antarctic Peninsula is accelerating.

A critical question is whether the calving of this huge iceberg will destabilise the ice shelf, in a similar way to the rifting observed on Larsen B ice shelf prior to collapse. Several computer simulation studies suggest that the ice shelf may become more unstable once the iceberg is calved. Larsen C Ice Shelf has a stress regime very similar to Larsen B before its collapse¹⁸ (‘Stress’ is a measure of how hard a material is pushed or pulled as a result of external forces).

An unstable pattern of principle stresses?

The stress-flow angles of the ice are critical to inducing calving. This is illustrated below. When the direction of the first tensile stress (a pulling-apart stress) is aligned close to the direction of ice flow, the stress-flow angle is low. In this situation, calving and rifting occurs along pre-existing weaknesses inherited from tributary glaciers. This situation is considered unstable, with frequent small calving (Part A in the figure below).

Currently, the stress-flow angles at the ice front of Larsen C are high (part B in the figure above). Calving events are therefore rare, and the ice front is stable^18,19. This creates a margin of stable ice at the ice front. However, a large calving event would narrow this safety margin¹⁹. A very large calving event would result in a very narrow buffer of high stress-flow ice, and a significant section of the ice front could have low stress-flow angles. This could induce rapid calving, resulting in ice-shelf collapse.

Passive ice vs active ice

Another modelling study suggests that the frontal portion of Larsen C is ‘passive ice’, which could be removed without significant dynamic implications²⁰. This ‘passive ice’ has limited buttressing effect on the ice shelf, so its removal would not affect the ability of the ice shelf to ‘buttress’ the glaciers on land. This means that when Larsen C Ice Shelf calves its large glacier, it is unlikely that the glaciers on land will accelerate. Whether the ice shelf collapses quickly following the calving event will be critical to its influence on the glaciers on the Antarctic Peninsula.

Soft squashy marine ice

The rift on Larsen C Ice Shelf is currently within a zone of softer marine ice, which may slow rift lengthening²¹; boundaries between flow units, inherited from the tributary glaciers on land or from ice flowing around capes and islands, contain anomalously soft ice. This soft ice may reduce stress intensities and reduce rifting. Glasser et al. 2009 conclude that,

“The presence of flow units containing anomalously soft meteoric ice limits rates of rift propagation and thus contributes to the stability of this ice shelf.”

Glasser et al. 2009 argue that, based on a structural glaciological analysis, the ice shelf has been stable for more than 500 years. If the large rift passes through the soft marine ice and does calve a large iceberg, it seems that we will be in a new, unprecedented era for Larsen C Ice Shelf.

If Larsen C Ice Shelf has been thickening by freeze on of softer marine ice, this could stabilise the ice shelf and counteract the impact of the calving event on altering the stress regime on the ice shelf. The impact of marine ice depends on how much thickening or thinning has been occurring, and whether this is cyclical or a progressive event. Data on possible thickening are as yet unpublished.

Sea level rise following ice-shelf collapse

Loss of the Larsen C Ice Shelf would result in significant recession and acceleration of the tributary glaciers that flow into it, which would result in sea level rise. However, the volume of sea level rise is likely to be equivocal, as the sea level equivalent of the Antarctic Peninsula north of 70°S as a whole is 69±5 mm sea level equivalent²². The volume of the glaciers on the SE part of the Peninsula, which flow into Larsen C Ice Shelf, is 22.4±1.3 mm sea level equivalent. The collapse of Larsen C Ice Shelf would therefore only raise global sea levels by a few millimetres.

However, Larsen C ice-shelf collapse would be a sign that the threshold for ice-shelf collapse is moving southwards, towards West Antarctica. Ice shelf collapse and glacier recession here, in front of the large ice streams such as Pine Island Glacier and Thwaites Glacier, would have potential to raise sea levels by tens of centimetres to a metre, through the process of marine ice sheet instability²³.

Links

• Project Midas
• NASA Earth Observatory
• Antarctica’s contribution to global sea level rise
• Marine Ice Sheet Instability
• Ice shelf collapse