Current warming and likely future impacts - Climate Change: Biological and Human Aspects - page 335

In 2009 the Scientific Committee on Antarctic Research (based at the Scott Polar
Research Institute, Cambridge) of the International Council for Science expressed
concern about possibly large contributions to sea level from the dynamic instability
of Antarctic ice sheets during the 21st century. These were not included by the IPCC's
2007 assessment. They recommend that we prepare for a maximum sea-level rise of
2 m (part of which will come from Antarctica) by 2100 on the basis that it may be
less than this.
Where does this leave us with regards to surprises? The IPCC's 2007 best estimate
for sea-level rise from 1990 to the end of the 21st century is 37 cm (A2 median value),
in the range 23-51 cm. The 2007 IPCC's A1F1 scenario gives a higher estimate of
59 cm (see Table 5.1). For the IPCC, and hence policy-makers, the surprise would
be if we were taken outside this window. Is this likely? The reasons to suggest that it
might be are 3-fold.

1. Some of Greenland's margins and Antarctica's ice shelves appear to be disinteg-
rating faster than anticipated.
2. Antarctic and Greenland interior ice is now moving faster to the coast.
3. Mass additions to past 20th-century sea-level rise appear to be greater than thought.

First, with regard to points 1 and 2, even though the potential sea-level rise from
Antarctica greatly exceeds that from northern hemisphere ice caps (see Table 6.3),
recent research suggests that during the last glacial the sea-level rise of 30 m (when
the sea was much lower than now) over the period 65 000-35 000 years ago was
due to equal contributions (

10%) of melt from northern hemisphere ice caps and
Antarctica, and that these were pulsed with 5000-8000-year spacing between events.
The suggestion is that while the then glacial Earth warmed 2-3 ◦ C the sea-level rise of
30 m took place at a rate of 2 m per century (Rohling et al., 2004). This is over three
times the IPCC's best estimate (IPCC 2001a, 2001b) of the rate of rise forecasted for
the 21st century.
Furthermore, many had assumed that just 25 m of the total 120 m glacial-
interglacial difference in sea level came from Antarctica. Consequently this more
recent research suggests much of the past 25 m Antarctic (100 000-year glacial-
interglacial) melt was released in episodes within a 30 000-year period in the depth
of the ice age. Whereas past melt is not necessarily a guide to future melt (the 21st
century does not have large Laurentide or Fennoscandian ice sheets, which accoun-
ted for much of the rest of the 120 m glacial-interglacial rise), it does suggest that
past melt was not smooth, and so future melt might not be as linear or smooth as
the IPCC predict. In 2011 an international team, including Europeans but primarily
from Canada, led by Alex Gardner and Geir Moholdt noted that the Canadian Arctic
Archipelago, located off the north-western shore of Greenland, contains one-third of
the global volume of land ice outside the ice sheets, but its contribution to sea-level
change remains largely unknown. Their estimates were based on three independ-
ent approaches: surface mass-budget modelling plus an estimate of ice discharge,
repeat satellite laser altimetry (ICESat) and repeat satellite gravimetry (GRACE).
All three approaches showed consistent and large mass-loss estimates. They showed
that the Canadian Arctic Archipelago (the two largest islands of which are Baffin and
Ellesmere) has recently lost 61

±

7 Gt year − 1 ± of ice, contributing 0.17 ± 0.02 mm