Crane Glacier photographed on 24 February 2004 (Photo courtesy of Pedro Skvarca, Glaciology Division, Instituto Antártico Argentino)

Antarctic glaciers respond rapidly to climate change, according to new evidence found by NSIDC, NASA, and IAA scientists. In the wake of the Larsen B Ice Shelf disintegration in 2002, glaciers in the Antarctic Peninsula have both accelerated and thinned en route to the Weddell Sea. The findings indicate that ice shelf breakup may rapidly lead to sea level rise.

In a paper published in Geophysical Research Letters, Ted Scambos and Jennifer Bohlander of NSIDC, Chris Shuman of the Oceans and Ice Branch at NASA's Goddard Space Flight Center, and Pedro Skvarca of the Instituto Antártico Argentino describe two- to six-fold increases in centerline speed of four glaciers feeding the now-collapsed section of the Larsen B Ice Shelf. They also describe elevation losses in three glaciers in the collapse area. The researchers used both Landsat 7 and ICESat satellite imagery in this study.

In the same issue of GRL, Eric Rignot of NASA's Jet Propulsion Laboratory and collaborators describe the same acceleration using Interferometric SAR from RADARSAT. Using their map of the flowspeed, they estimate that the glaciers ought to be thinning by tens of meters. ICESat elevation measurements by the Scambos team corroborate their prediction.

Ice shelves are thick platforms of ice that are fed by glaciers and float on the ocean. Ice shelf disintegration has no direct effect on sea level because the ice shelf has already displaced its own volume in seawater. In the wake of an ice shelf collapse, however, the resulting glacier acceleration can raise sea level by introducing a new ice mass into the ocean. Although glaciers on the Antarctic Peninsula are too small to significantly raise sea level, this research provides a glimpse of what could happen on a larger scale if other large ice shelves in Antarctica — for example, the Ross Ice Shelf — experienced similar warming. "This gives us a chance to watch an experiment before the consequences get serious," said Scambos. "Even though we don't see immediate evidence of ice shelf breakup on the Ross Ice Shelf, everything we've seen up to this point has occurred faster than we expected. The larger ice shelves in other parts of Antarctica could have real effects on the rate of sea level rise."

Diagram by Ted Scambos and Michon Scott, NSIDC.

Stress on the high ice cliff enhances fracturing of the Hektoria Glacier near the grounding line in this March 2002 aerial photo. (Photo courtesy of Pedro Skvarca, Glaciology Division, Instituto Antártico Argentino)

In Antarctica, glaciers flowing to the coast form ice shelves — thick platforms of ice that float on the ocean. Together, the glacier and ice shelf form a stable system, but this system can lose its stability in response to warmer temperatures.

Warmer summer temperatures sometimes result in glacier acceleration as melt water percolates through the glacier to its base. Here the water lowers the friction between the glacier and the underlying rock. This effect is seasonal, and with the ice shelf in place, the glacier returns to a lower flow speed once summer (and surface melting) ends.

Warmer summer temperatures can also lead to rapid ice shelf disintegration. As temperature rises, melt water accumulates on the shelf surface. Although only a tiny fraction of the ice shelf melts, the water infiltrates the shelf through small cracks in the ice. Over time, the weight of the melt water in the cracks shatters the shelf. This happened in the Antarctic Peninsula in 1995 and again in 2002. To read more about these events, see Larsen Ice Shelf Breakup Events.

Removal of the ice shelf causes much more dramatic glacier acceleration by reducing two forces that counteract glacier flow. One counteracting force is "backstress" produced by islands or coastline underlying the original shelf. Another is the buoyant (hydrostatic) force of the seawater against the front of the shelf or glacier. A full explanation will require numerical modeling of glacier flow, but observations to date suggest that ice shelves act as "braking" systems on the glaciers behind them.

This satellite image from NASA's MODIS sensor aboard the Terra spacecraft shows the Larsen B Ice Shelf region on 1 November 2003. Red dots indicate sites where ice flow speed was measured using more detailed Landsat 7 images. The colored lines track the retreat of the Larsen B Ice Shelf during the past 6 years, and the black line shows the coastline, or "grounding line," where the thick ice begins to float off the sea floor. Blue lines on the glaciers show the location of laser elevation profiles from ICESat. A weather station location marked in the upper right of the image map ("Matienzo AWS") has tracked atmospheric warming in summers over the past 30+ years in the region. (Image derived from Scambos et al.: Larsen B Glacier Acceleration, 2004)

In all images, the black contour line indicates the grounding line for the Larsen B Ice Shelf and the red arrow indicates the flow of the Hektoria Glacier system. (Images supplied by Ted Scambos and Jennifer Bohlander, NSIDC)

27 January 2000: The Hektoria Glacier system is stable, but increased summer melting from climate warming in the 1980s and 1990s affected the glacier system in two ways: (1) a seasonal speedup from summer melt water percolating through the glacier ice to its base, and (2) initial retreat of the Larsen Ice Shelf due to the effects of melt ponds (downstream from this image).

6 April 2002: Record-high temperatures and melting in the austral summer earlier this year caused significant glacier acceleration. While the March 2002 Larsen Ice Shelf disintegration likely resulted from rapid local climate warming, sea level is not affected by the breakup of ice that is already afloat.

18 December 2002: As ice shelf retreat continues past the grounding line, the lower portion of the glacier accelerates rapidly. Thousands of new crevasses form in the glacier trunk due to major changes in the forces ("backstresses") acting on the glacier. These changes occur during the Antarctic winter, so they are not related to melt percolation.

20 February 2003: As the glacier acceleration continues and propagates upstream, the lower portion of the glacier loses tens of meters in elevation. The mass of ice delivered to the ocean increases, contributing to a rise in sea level. Larsen B glaciers are too small to significantly affect sea level, but the processes that acted on this area could play out on other, bigger ice shelves.

Hektoria: The initial speedup between 2000 and 2002 is probably due to the very warm summer during which the Larsen B disintegrated. Immediately after breakup, glacier speed rapidly increased. Because the downstream part of the glacier accelerated more than the upper glacier, the ice was "stretched." New crevasses formed throughout the lower glacier as a result, and the stretching caused the glaciers to drastically thin.

Green: As with the Hektoria Glaicer, the initial speedup between 2000 and 2002 is probably due to the very warm summer during which the Larsen B disintegrated. Immediately after breakup, glacier speed rapidly increased. Because the downstream part of the glacier accelerated more than the upper glacier, the ice was "stretched." New crevasses formed throughout the lower glacier as a result, and the stretching caused the glaciers to drastically thin.

Crane and Jorum: Retreat of the ice shelf near these glaciers was prolonged; after the main collapse occurred, some shelf ice remained. Even a small amount of remaining shelf ice appears to reduce the amount of stretching and acceleration — particularly for Jorum Glacier, which slowed during the first winter period after the breakup.

Flask and Leppard: Because the shelf collapse did not extend to these glaciers, a substantial area of shelf ice remained in front of these areas for the entire period, so only the seasonal changes in flow speed appear. This indicates that ice shelf loss, not some other process, led to speedup and potential increases to Earth's sea level.

Reference

Scambos, T. A., J. A. Bohlander, C. A. Shuman, and P. Skvarca. 2004. Glacier acceleration and thinning after ice shelf collapse in the Larsen B embayment, Antarctica. Geophysical Research Letters. doi:10.1029/2004GL020670.
Visit AGU Publications for online access.

Return to News Room