Sorry, no link - subscription service of the American Geophysical Union
Examining Holocene Stability of Antarctic Peninsula Ice Shelves
EDIT
While the geological record shows that recent retreat of the Larsen B Ice Shelf is unique in the Holocene, this is not a feature common to all of the other Antarctic Peninsula ice shelves so far studied. Specifically, there is evidence that ice shelves on the west and northeast of the Antarctic Peninsula have behaved differently, experiencing previous retreat events associated with warm periods in the middle and early Holocene. For example, a mid-Holocene retreat of the Prince Gustav Channel Ice Shelf has been attributed to a well-documented period of increased atmospheric temperature
. Thus, atmospheric temperature has been a key feature of some past and present ice shelf retreats through a series of well-understood decay mechanisms.
However, ongoing research indicates that Antarctic Peninsula ice shelves may also be sensitive to a combination of both atmospheric and ocean temperature change. Research on the George VI Ice Shelf, which is currently south of the climatic limit of ice shelf stability, has shown that it retreated in the early Holocene (c. 96500 - 7900 years B.P.) immediately following a period of early Holocene warmth detected in the ice cores . At the same time, ocean records from the Palmer Deep, a basin on the continental shelf of the Antarctic Peninsula, indicate a dramatic increase in the presence of warmer surface waters over the Antarctic Peninsula's continental shelf that are likely to have contributed to the retreat of the George VI Ice Shelf . Present-day radar data show that up to 2.78 +/- 0.08 meters of ice is lost from the base of the George VI Ice Shelf each year \ due to basal melting (where warm seawater melts the underside of the shelf). This makes the ice shelf unusually susceptible to ocean temperature change.
Collectively, the research indicates that ice shelves are vulnerable to melting from above (atmospheric temperature) as well as from below (warm ocean currents), and that both processes are likely to have caused previous retreats of different Antarctic Peninsula ice shelves during the Holocene. As a result, there is an emerging geographical pattern in the history of ice shelf collapse. While to the east of the Antarctic Peninsula the Larsen B Ice Shelf has been stable throughout the Holocene, ice shelves studied further north (Larsen A Ice Shelf and Prince Gustav Channel Ice Shelf) and on the west of the Peninsula (George VI Ice Shelf) have broken up before in the Holocene (Figure 1). So how can this difference in behavior be explained?
EDIT