Breakup of Larsen B Ice Shelf May Be Underway
The Larsen B appears to have begun breaking up, receding past its historical minimum extent, and past the point where recent modeling suggests it can maintain a stable ice front. The breakup appears closely associated with the areas over which melt ponding is observed during warmer summer seasons.
Below are two AVHRR thermal-channel images of the Larsen B area acquired from the National Snow and Ice Data Center's AVHRR Polar 1 km Level 1b Data Set. The first image is from February 15 of this year, showing melt ponding typical of warm summers. The ice front in this image represents the approximate appearance of the ice shelf since early 1995, when a large iceberg calved off the face of the ice shelf, and when the Larsen A Ice Shelf, formerly to the north of Robertson Island, disintegrated. Up until February 15, only minor modifications to the ice front had been observed.
An image from February 26 (not shown) revealed that the ocean in the vicinity of the Larsen B was free of sea ice, exposing the front to the ocean swell. Further, this image showed a new embayment in the ice front, and the presence of several 1 to 5 kilometer (0.6 to 3.2 mile) icebergs immediately in front of the shelf (presumably derived from the breakup).
The second image, from March 23, clearly shows the new embayment, and indicates a loss of roughly 200 square kilometers (77 square miles) of ice shelf. The front has retreated by about 5 kilometers (3 miles) along the northern 40 kilometers (25 miles) of ice front.
The new embayment is occurring along the seaward edge of the part of the ice shelf where melt ponding is most commonly observed. Monitoring of the Larsen ice shelves over the last few years has shown that melt ponding regularly occurs north of Cape Disappointment, but is seen much less frequently south of there. Melt ponds were also observed over the entire Larsen A Ice Shelf prior to its breakup, and are observed on the Wilkins and George VI Ice Shelves, both of which may be currently undergoing slower irreversible retreats.
A recent Nature paper by Chris Doake and others (Nature:391, 778- 780) indicates the importance of this event: in their model, the ice shelf position of roughly a year ago (very similar to that shown in the February 15 image) was close to the minimum limit that maintained an adequate back pressure due to compressive strain between the Jason Peninsula and Robertson Island. With this breakup, (I suspect) the shelf has retreated past this limit. If the scenario of Doake et al. is correct, the retreat will continue in the next season, perhaps rapidly if storms and sea ice conditions are appropriate. At present, the onset of the winter season, with sea ice extent increasing, probably precludes much additional evolution. Sea ice shields the shelf from wave action, which may be important for facilitating breakup.
NSIDC regularly monitors the ice shelves of the Peninsula and West Antarctica, using data from EROS Data Center's Global Land 1km AVHRR Data Set, and from the Colorado Center for Astrodynamics Research's DOMSAT receiver. The archive of AVHRR scenes of the ice shelves is collected and maintained by Jennifer Bohlander.