Search our site

Custom Search

Friday, 5 December 2014

Collapsing Ice Shelves

In 2002 the Larsen B Ice Shelf on Antarctica collapsed spectacularly. An area of ice twice the size of Greater London was lost in less than a month. This occurred in the northernmost region of Antarctica - the Peninsula - which has warmed-up by more than five times the global average over the last century[1]. A result of this is that, in certain parts of the Peninsula, the surface of the ice is starting to melt. The water from this melting can accumulate to form lakes up to 4km long. Larsen B was covered in these lakes, but just before the ice shelf collapsed these lakes started to drain. First one drained, then those around it, then those around them in a chain reaction that is suggested to have been key to the sudden collapse of the ice shelf[2].

Photograph of Larsen B ice shelf collapse in February 2002 by NASA MODIS
In February 2002, satellite images of the area stunned scientists as they watched 3,250 square kilometers of ice sheet disintegrate within the space of a month. By the end of 2006 the Larsen A and B glaciers were losing 22-40 billion tonnes of ice per year. Image credit: MODIS, NASA's Earth Observatory (CC-BY 2.0)

We don’t know what caused the lakes to drain so suddenly, or exactly how this links to the ice shelf’s collapse. Various hypotheses have been suggested as triggering the collapse; including the forces that the weight of the lakes exert on the ice shelf[2], and the melting of the ice shelf from below by heating from the ocean[3]. However, it is clear from the sudden drainage of the lakes that their role needs investigation, especially as they are beginning to appear further south on the Antarctic Peninsula.

The loss of ice shelves such as Larsen B is worrying for many reasons. It can mean a loss of habitat for creatures such as penguins. It can also cause changes in ocean circulation, temperature and salt content, due to the addition of cold, fresh water that the ice from the collapsed ice shelf provides as it floats away into the ocean and melts. Additionally, ice shelves are key in regulating the speed of glaciers on Antarctica[5]. Ice shelves can act to hold glaciers back, but take the ice shelf away and the glaciers are free to speed up and flow straight into the ocean, and all the ice and water that they take with them will contribute to sea level rise.

Recently these lakes have been observed on the Larsen C Ice Shelf, which is situated next to where Larsen B used to be, and this is why my work as a PhD student at the University of Reading is focusing there. The reason for this area warming so rapidly, and therefore surface melt spreading further south on the Peninsula, isn’t certain; but it is thought that atmospheric circulation and human influences on the belt of westerly winds that circle Antarctica, such as greenhouse gas emissions and ozone depletion, have played a role[4].

map of Antarctica which highlights the locations of the 12 major ice shelves
The Larsen C Ice Shelf (yellow) is one of the 12 major ice shelves of Antarctica. Image courtesy of Ted Scambos, and the National Snow and Ice Data Center, University of Colorado, Boulder.

The snow on Larsen C that sits on top of the ice is many metres deep, and water from snow melting at the surface will just drain away into it. For a lake to form, the snow needs to be fully saturated with water, and this would take much more than would be provided just by the melting snow. This suggests that something else must be playing a part. In order to investigate the ways in which these lakes are forming, my work involves creating a mathematical model of the way that heat is transferred through the surface of the Larsen C Ice Shelf. Through these computer simulations I have been looking at whether the water accumulates in certain areas due to the shape of the ice shelf’s surface, providing enough water to make a lake, or if processes involving the refreezing of water are having an effect.

In addition to the difficulties in taking year-round measurements in a location as remote as Larsen C, we also have a lack of good satellite observations because the area is often very cloudy. Creating a mathematical model allows us to combine the information we do have, such as weather station data (which can be collected automatically once a weather station has been set up) and information from ice cores taken during one-off expeditions, with physical equations in order to simulate how these lakes form and change with time, without someone having to actually be there to observe this.

The results of this work will provide information that will allow ice shelf and climate models to be improved in the future. In particular, knowing how the lakes form will give new insights into the role of surface melting in ice shelf collapse, and therefore the fate of Larsen C - is it headed for a similar fate to Larsen B?

Sammie Buzzard is a PhD student in the department of Meteorology at the University of Reading. After completing a Master's degree in mathematics she is now researching surface melt on the Larsen C Ice Shelf. She blogs about polar science, life as a PhD student and cake (because cake is essential for research) at Ice and Icing and can be found tweeting at @treacherousbuzz.

why don't all references have links?

[1] Khazendar et al., Acceleration and spatial rheology of Larsen C Ice Shelf, Antarctic Peninsula, Geophysical Research Letters, Vol. 38 (2011). DOI:10.1029/2011GL046775
[2] A.Banwell et al., Breakup of the Larsen B Ice Shelf triggered by chain reaction drainage of supraglacial lakes, Geophysical Research Letters, Vol. 40, pages 1-5, 2013. DOI: 10.1002/2013GL057694
[3] Shepherd et al., Larsen ice shelf has progressively thinned, Science, 2003, (31 October), Vol. 302, 856-859
[4] IPCC Fourth Assessment Report: Climate Change 2007, Working Group II: Impacts, Adaptation and Vulnerability, 15.6.3.
[5] E. Rignot, et al., Accelerated ice discharge from the Antarctic peninsula following the collapse of Larsen B ice shelf. Geophysical Research Letters, 31(L18401), 2004. DOI:10.1029/2004GL020697