Reconstructing Sea Ice Conditions in the Arctic and Sub-Arctic Prior to Human Observations - Arctic Sea Ice Decline: Observations, Projections, Mechanisms, and Implications - page 32

Figure 4. Spatial distribution of the residuals (observed minus estimated sea ice cover). The solid line and the dashed line
correspond to the modern mean winter sea ice cover limit and the extreme sea ice limit, respectively.

document the natural variability and trends in sea ice cover
extent beyond the mid-1900s.
The variability of sea ice over the last centuries is diffi-
cult to document from sediment cores because it requires
particularly high sediment accumulation rates to achieve a
suitable resolution in time. Nevertheless, a few cores per-
mitted analyses with adequate resolution (Figure 6). Some
historical records provide additional evidence for secular
variations [e.g., Lamb , 1977; Ogilvie , 1984; Hill , 1998; Hill
et al. , 2002]. The examples illustrated in Figure 6 are from
the western Arctic and the northern Baffin Bay. Both re-
constructions were made using the same approach and both
show significant changes in sea ice on secular scale. How-
ever, the two records illustrate distinct trends. Whereas the
Chukchi Sea data show a denser sea ice cover during the
late 19th and 20th centuries, the northern Baffin Bay record
suggests a decrease in sea ice cover during the 20th century
after an interval of maximum areal extent at about 1750-
1900 A.D. This maximum in the sea ice cover might well
correspond to the episode of extremely cold winters, which
have marked the second half of the last millennium in the
subpolar North Atlantic and in western europe and which is
known as the “Little Ice Age” (LIA). The spreading of sea
ice along the coasts of Iceland from 1600 to 1900 A.D. (Fig-

ure 7) [ Lamb , 1977] is another example of the recording of
the lIA by this parameter. However, as pointed out by Jones
and Mann [2004], the lIA documented from historical or
proxy data is far from a synchronous and uniform climate
event over the Northern Hemisphere, and some discrepan-
cies may be expected. Nonetheless, although sea ice records
of the last millennium are still rare, the few data available
indicate important secular changes in sea ice margins with
some diachronous responses of sea ice extent from one re-
gion to another.

4. ISOTOPIC (d 18 O) COMPOSITION OF
MeSOPelAGIC FORAMINIFeR SHellS: A PROXY
FOR SeA ICe ACCReTION RATeS?

Seasonal duration in the sea ice cover at given sites pro-
vides information on the sea ice distribution but not on its rate
of formation. Other proxies are needed for estimating such
rates that are dependent upon many parameters including
air temperature, depth of the pycnocline, and stratification
and salinity of the surface layer of the ocean, as well as wind
strength and patterns. As briefly mentioned in section 2.3,
some proxy for sea ice formation may be obtained in the
isotopic (δ 18 O) composition of foraminifer shells, which pre-