tive contribution to sea-level rise. For the Antarctic Ice expansion is likely to be the dominant contribution to
Sheet, the uncertainty is greater. There are insufficient 21st century sea-level rise, with the next largest contribu-
data to make direct estimates for the preceding decades. tion coming from the melting of glaciers and ice caps.
At present, the mass gain of the Antarctic Ice Sheet due
to increased thickening of the East Antarctic Ice Sheet Recent estimates indicate that non-polar glaciers and ice
does not appear to compensate for the mass loss due caps may contain only enough water to raise sea level
to the increased glacier flow on the Antarctic Peninsula by 15 to 37 cm
26
. Melting of glaciers at lower altitude
and the West Antarctic Ice Sheet
20,21
. Modelling studies and latitude in a warming climate will eventually result
suggest that the Antarctic Ice Sheet is still responding to in significant reduction of the sizes of the glaciers and
changes since the last ice age and that this may also be reductions in their contribution to the rate of sea-level
contributing to sea-level rise. rise. The most important impact is from large glaciers
in regions with heavy precipitation, such as the coastal
The difference between the sum of the contributions mountains around the Gulf of Alaska (Figure 6C.6), or
to sea-level rise and the observed rise from 1993 to the Patagonia and Tierra del Fuego in South America. Many
present is smaller than the estimated errors. However of these glaciers flow into the sea or large lakes and melt
during the 1961 to 2003 period, ocean thermal expan- quickly because the ice is close to melting temperature
sion along with the melting of glaciers and ice caps and (see also Section 6B).
a reasonable allowance for an ice sheet contribution do
not adequately explain the observed rise. Possible rea- For Greenland, both glacier calving and surface melting
sons for this discrepancy include the inadequate ocean contribute to mass loss. Over the last few decades sur-
database, particularly for the deep and Southern Hemi- face melting has increased
27
and now dominates over in-
sphere oceans, leading to an underestimate of ocean creased snowfall, leading to a positive contribution to sea
thermal expansion, and inadequate measurements of level during the 21st century. For the majority of Antarc-
the cryosphere. tica, present and projected surface temperatures during
the 21st century are too cold for significant melting to oc-
Changes in the storage of water on land, including cur and precipitation is balanced by glacier flow into the
changes in lakes, building of dams (both large and small), ocean. In climate change scenarios for the

21st century,
seepage into aquifers, and mining of ground water, may climate models project an increase in snowfall, resulting
also be important – but the extent of these contributions in increased storage of ice in Antarctica, partially offset-
is unclear. Model studies suggest significant variability ting other contributions to sea-level rise. However, an in-
from year to year of the climate-related components of crease in precipitation has not been observed to date
28
.
terrestrial water storage, but little long-term trend
22
.
In addition to these surface processes, there are sugges-
tions of a potential dynamical response of the Greenland
Outlook for sea-level change and Antarctic ice sheets (see also Section 6A). In Green-
land, there was a significant increase in the flow rate of
During the 21st century, sea level will continue to rise many of the outlet glaciers during the early 21st century
19
.
due to warming from both past (20th century and earlier) One potential reason for this is increasing surface melt
and 21st century greenhouse gas emissions (see box on making its way to the base of the glaciers, lubricating their
projections of 21st century sea-level rise). Ocean thermal flow over the bed rock, consistent with increased glacier
160 GLOBAL OUTLOOK FOR ICE AND SNOW