flow during the summer melt season
29
(see Figure 6A.6 sient acceleration. Another important consideration is
in previous section). However, recent work
30
has shown that the West Antarctic Ice sheet is grounded below cur-
the flow rate of at least two of these glaciers has recently rent sea level. As the ice sheet thins and starts to float,
decreased to near their earlier rates, suggesting that there warm ocean water can penetrate beneath and enhance
is significant short-term variability in glacier flow rates. melting at the base.
Another potential factor is the role of ice shelves in re- All of these dynamic ice-sheet processes, in both Green-
straining the flow of outlet glaciers. The rapid break up land and West Antarctica, could lead to a greater rate of
of the Larsen B Ice Shelf in the Antarctic Peninsula (Fig- sea-level rise than in current projections. However, the
ure 6A.4 in previous section) was followed by a signifi- processes are inadequately understood and are therefore
cant increase in the flow rate of the glaciers previously not included in the current generation of ice-sheet and
feeding this ice shelf
31
, suggesting that the ice shelves climate models. It is therefore not possible to make ro-
played a role in restraining the flow of outlet glaciers. bust quantitative estimates of their long-term contribu-
However, some modelling studies suggest this is a tran- tion to the rate of sea-level rise.
Level
(m)
2.3
2.2
After 1950
2.1
Before 1950
2
0.1 1 10 100
Average recurrence interval (years)
Figure 6C.7: Average Recurrence Interval for sea-level events
of a given height at Sydney, Australia. For the second half of
the 20th century (red line), the average recurrence interval for
a sea-level height of a given value is less than half the value for
Greenland ice sheet.
the first half of the 20th century (blue line).
Photo: Konrad Steffen Sources: Based on Church and others 2006
42
162 GLOBAL OUTLOOK FOR ICE AND SNOW