Recent break-up events on Wilkins Ice Shelf (Antarctic Peninsula)
Matthias Braun(1) and Angelika Humbert(2)
(1) University of Bonn, Walter-Flex-Str. 3, D-53113 Bonn, Germany
(2) University of Münster, Corrensstr. 24, D-48149 Muenster, Germany
The disintegration of various ice shelves on the Antarctic Peninsula has demonstrated their vulnerability and impacts on tributary glaciers. Wilkins Ice Shelf (WIS), with a size of 13.000km², is located on the south-western coast of the Antarctic Peninsula. Crossed by the -9°C mean annual surface isotherm, it is located in an area of which several authors suppose to be the limit of ice shelf viability. Previous studies have reported break-up events on WIS in 1998 and 1993. The events of 28/29 February and 30 May 2008 hit the press and were very well documented by remote sensing data. We present an integrated study using remote sensing imagery to gain a better understanding of the causes for ice shelf instability and the mechanisms leading to ice shelf break-up, with a specific emphasis on the 2008 events.
Time series of satellite images from various sensors (ERS-1/2 SAR, ENVISAT ASAR, TERRA ASTER and LANDSAT TM/ETM+) have been analysed for the general ice shelf structure, changes in ice front positions and rift formation associated with break-up and calving events. An interferometric surface velocity field was computed from imagery of the ERS-1/2 tandem and ice phase for the southern ice shelf and main tributaries. Additionally, surface elevation data from the ICESat GLAS instrument has been incorporated to estimate ice thickness distribution on the ice shelf.
In advance of the February break-up an ALOS PALSAR image from July of 2007 revealed the formation of a new large new double fracture, accompanied by numerous small fractures on the connection of WIS. We show that bending stresses induced by buoyancy forces were responsible for fracture formation. Subsequently, on 28/29 February 2008, an area of about 425 km² broke up exactly at this location. An ASTER image of 28 February clearly shows that the break-up along these fractures already detected in the L-band ALOS image from the previous winter. In contrast to Larsen-B Ice Shelf, melt ponds that drain into crevasses played no role in this break-up process. A further break-up of 160 km² in the same area occurred on 30/31 May 2008 and documented for first time a break-up during austral winter. It leaves now only a 2.7 km wide connection to the stabilizing Charcot Island. High resolution TerraSAR-X images show that this remaining connection is already fractured and hence prone for disrupture. The May break-up is clearly uninfluenced by surface melt. Radar images reveal a frozen surface, which demonstrates that in this break-up melt water does not play any role. We conclude that ice shelves with strong thickness contrasts carry potential for disintegration. These events disclose that there are in general several reasons for disintegration of ice shelves.
The analysis also shows that continuous image acquisitions and time series analysis are mandatory in order to gain a better understanding of the underlying processes. Only the ALOS PALSAR image from July 2007 revealed the fracture formation obviously due to its high resolution and deeper penetration in the snow pack as does the high resolution TerraSAR-X image for the fractures on the remaining connection to Charcot Island.