Larsen B Ice Shelf Collapses in Antarctica
Posted: 18 March 2002
Recent NASA Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery analyzed at the University of Colorado's National Snow and Ice Data Center revealed that the northern section of the Larsen B Ice Shelf, a large floating ice mass on the eastern side of the Antarctic Peninsula, has shattered and separated from the continent. The shattered ice formed a plume of thousands of icebergs adrift in the Weddell Sea. A total of about 3,250 square kilometers (1,255 square miles) of shelf area disintegrated in a 35-day period beginning on January 31, 2002. Over the last five years, the shelf has lost a total of 5,700 square kilometers (2,200 square miles), and is now about 40 percent the size of its previous minimum stable extent.
Ice shelves are thick plates of ice, fed by glaciers, that float on the ocean around much of Antarctica. The Larsen B Ice Shelf was about 220 meters (720 feet) thick. Based on studies of ice flow and sediment thickness beneath the ice shelf, scientists believe that it existed for at least 400 years prior to this event, and has likely existed since the end of the last major glaciation 12,000 years ago (For more information on this research, see Seafloor Evidence of Larsen Ice Shelf Breakup.
For reference, the area lost in this most recent event, 2,717 square kilometers (1,049 square miles), dwarfs Rhode Island. In terms of volume, the amount of ice released in this short time was 720 billion tons, enough ice to fill about 12 trillion 20-pound bags.
This is the largest single event in a series of retreats by ice shelves in the Antarctic Peninsula over the last 30 years. The retreats are attributed to a strong climate warming in the region. The rate of warming is approximately 0.5 degrees Celsius (0.9 degrees Fahrenheit) per decade, and the trend has been present since at least the late 1940s. Overall in the Antarctic Peninsula, extent of seven ice shelves has declined by a total of about 13,500 square kilometers (43,800 square miles) since 1974. This value excludes areas that would be expected to calve under stable conditions.
Ted Scambos, a researcher with the National Snow and Ice Data Center (NSIDC) at University of Colorado, and a team of collaborating investigators have developed a theory of how the ice disintegrates. The theory is based on the presence of ponded melt water on the ice shelf surface in late summer as the climate has warmed in the area. Meltwater acts to enhance fracturing of the shelf by filling smaller cracks and forcing them through the thickness of the ice due to the weight of the water. The idea was suggested in model form by other researchers in the past (Weertman, 1973; Hughes, 1983); satellite images have provided substantial observational proof that it is in fact the main process responsible for the Peninsula shelf disintegrations. Christina Hulbe of Portland State University and Mark Fahnestock of University of Maryland collaborated with Scambos on the research.
A number of international scientists have also cooperated in the general study of the demise of the shelves and the climatic trend in the Antarctic Peninsula. As early as November of last year, Pedro Skvarca, Head of the Glaciological Division of the Instituto Antártico Argentino, warned of a possible impending breakup, due to very warm spring temperatures and a dramatic 20 percent increase in the rate of flow of the ice shelf. He and his team were the last people to set foot on the northern portion of the shelf. Later in the summer, the Argentine group returned to their base at Marambio, near the shelf, to await what they anticipated would be the final disintegration event. They flew over the shelf repeatedly, measuring its extent with GPS during the course of the breakup event. (See Dr. Skvarca's photos of the ice front line towards Cape Foyn and the broken ice shelf south of the Seal Nunataks, above right.)
A British research vessel, the RRS James Clark Ross, was in the area just as the event was occurring and provided images from the ocean surface (above, right) in the region of the event. Keith Nicholls of British Antarctic Survey (BAS) provided the images.
In prior studies, Dr. David Vaughan and Chris Doake of BAS have reported extensively on the climate warming in the area, and have modeled shelf stresses and possible causes of breakup. They collaborated with Skvarca and with Austrian and German scientists, Dr. Helmut Rott and Dr. Wolfgang Rack, who conducted detailed satellite radar image studies and field studies in the area. The radar study also showed ice flow increase in the years leading to breakup and an increased velocity of the glaciers as the shelves disappeared. Radar images have provided very detailed views of the events of past ice shelf collapses. Dr. Rott is a professor at the Department of Meteorology and Geophysics at Innsbruck University; Dr. Rack is now at Alfred Wegener Institute in Bremerhaven, Germany.
The melt water fracturing theory fared well in this last event (See Christina Hulbe's Larsen Ice Shelf site). Sequential images from the MODIS sensor, a new satellite imager flying on NASA's Terra platform, showed extensive melt ponding over the Larsen B in late January, consistent with an unusually warm summer and extended melt season. In a series of images taken in February, several of the melt ponds disappeared, presumably as they drained through opening fractures in the ice. By February 23, 790 square kilometers (305 square miles) had shattered from the front. The next image from March 5 showed another 1,960 square kilometers (757 square miles) of ice gone. The event continued to March 7 with an additional loss of 525 square kilometers (203 square miles). The area lost by the shelf was was almost solely the region covered by melt ponds in late January. The timing of the event, at the end of a particularly warm summer, is consistent with the theory.
Scambos and other scientists continue to look for additional mechanisms that may contribute to the breakups. One idea is that meltwater seeping between ice crystals and warming of the shelf as a whole reduces the fracture toughness of the ice so that the shelf shatters under the same stresses imposed by local geography and the flow it used to tolerate. Another idea is that meltwater seeps into shallow cracks and expands the cracks as it refreezes during the winter. Ocean warming and sub-ice currents dragging on the underside of the ice have also been cited as possible contributors.
While the breakup of the ice shelves in the Peninsula has little consequence for sea level rise, the breakup of other shelves in the Antarctic could have a major effect on the rate of ice flow off the continent. Ice shelves act as a buttress, or braking system, for glaciers. Further, the shelves keep warmer marine air at a distance from the glaciers; therefore, they moderate the amount of melting that occurs on the glaciers' surfaces. Once their ice shelves are removed, the glaciers increase in speed due to meltwater percolation and/or a reduction of braking forces, and they may begin to dump more ice into the ocean than they gather as snow in their catchments. Glacier ice speed increases are already observed in Peninsula areas where ice shelves disintegrated in prior years.
With the Antarctic Peninsula ice shelf breakups as a guide, we can now reassess the stability of ice shelves around the rest of the Antarctic continent. Past assessments of stability were based primarily on mean annual temperature; with this guideline, most shelves outside the Peninsula were considered well within their climate limit. Given the success of the melt pond theory, we use the climate conditions and physical parameters of ice shelves at the point of ponding as a guide in this assessment. In particular, the next shelf to the south, the Larsen C, is very near the stability limit, and may start to recede in the coming decade if the warming trend continues. Melt ponds are occasionally observed in limited regions of the Larsen C shelf. More importantly, the warmest part of the giant Ross Ice Shelf is in fact only a few degrees too cool in summer presently to undergo the same kind of retreat process. The Ross Ice Shelf is the main outlet for several major glaciers draining the West Antarctic Ice Sheet, which contains the equivalent of 5 meters (16 feet) of sea level rise in its above-sea-level ice.
Although several recent large iceberg calving events have been observed on the Ross and elsewhere in Antarctica, none of these are thought to be related to ice shelf instability.
Learn About NSIDC
General public and data users:
Members of the press: email@example.com or