Breakup of the Larsen B Ice Shelf: 15 February 1998 to 18 March 1999

Back to Larsen Ice Shelf Breakup Events

thumbnail image of larsen B, links to animation of 1998-99 breakup

Satellite image series of Larsen B Ice Shelf, 1998-1999. View image series

7 April 1999

Below are four images showing the evolution of the Larsen B Ice Shelf from winter 1998 through spring 1999. Images are in either the thermal band (dark areas are warmer) or the visible band. Note the presence of melt ponds in the February 15 and possibly in the November 20 images.

Total shelf area lost over the four scenes is approximately 1,839 square kilometers (710square miles); between March 23, 1998 and November 20, 1998 about 1037.5 square kilometers (400 square miles); between November 20, 1998 and March 18, 1999 approximately 676 square kilometers (261 square miles).

Note the differences between the northern and southern half of the calving front. The northern half of the front is curved, with myriad tabular bergs in front of it. This is very similar to what was seen across the entire front of the Larsen A during breakup. The southern half is also retreating, but is not "embayed," or curved inward. In the earlier images from February and March, the northern half again led the way in terms of breakup. Note also that the ice front on the November 20 image appears to have followed a crack barely visible on the February 15 image.

Larsen B 980323Larsen B 980323Larsen B 981120Larsen B 990318

Summertime images of this area consistently show melt ponding present in the northern half of the shelf, and absent in the southern half, implying a link between extensive surface melting and breakup.

Recent models by Christina Hulbe indicate that the shelf ice stiffness suggests a low internal temperature. This implies that most of the shelf ice comes from cold glacial catchment areas in the mountains of the Antarctic Peninsula. Cold ice is stiff ice. This stiffness is a key factor in determining its stability. For the southern Larsen B, the glacial catchments are considerably further south, and presumably cooler. Downslope winds, following the glacier troughs, are probably also cooler on average and therefore would cause less surface melt. Thus, the southern section might be more stable than the north.

Surface flowstripes on both the Larsen A and Larsen B cross the entire former extent of those shelves, at least in the vicinity of Robertson Island (visible in SAR, but not the above AVHRR scenes). These features are inferred to form at the grounding line and above as part of glacier flow. Using this assumption and the ice flow speed provides an estimate of the minimum age of the shelves: about 400 years (if the flow speed observed at present has not changed much over time). In other words, it appears that the shelves were stable for centuries prior to the present events.

  • NSIDC maintains an extensive archive of polar satellite images with contributions from Scripps Arctic and Antarctic Research Center, Eros Data Center, and NOAA's Satellite Active Archive.