Geoscience Reference
In-Depth Information
kinson
, 2006], with greater losses of more than 7% per dec-
ade for the September minimum [
Lemke et al.
, 2007]. Since
2002, the Arctic Ocean has experienced an unprecedented
period of continuously low September ice cover, with larger
reductions in more recent years [
Stroeve et al.
, 2008, 2005;
Serreze et al.
, 2003]. As the September minimum roughly
denotes the ice cover that has survived the melt season, it
may be seen that the downward trend in September ice cover
corresponds to a reduction in thick, multiyear ice, which is
being replaced by a thinner, seasonal ice cover. Global, cou-
pled models used in prognostic simulations have also indi-
cated that the Arctic perennial ice cover is in jeopardy. The
most recent survey by the Intergovernmental Panel on Cli-
mate Change (IPCC) has indicated that many models project
a seasonally ice-free Arctic Ocean by the end of the 21st
century under the high-emission A2 scenario for anthropo-
genic forcing [
Meehl et al.
, 2007;
Zhang and Walsh
, 2006].
The manner in which an increasing anthropogenic forc-
ing becomes manifest in the climate system is the subject of
some discussion within the literature. The Arctic multiyear
ice pack may be abated either through an increase in ice ex-
port out of the basin, a decrease in wintertime ice growth,
an increase summertime melting, or some combination of
the three [
Tremblay et al.
, 2007]. These processes may be
mapped onto aspects of the Arctic climate system which in-
clude the atmospheric circulation, radiative processes, and
oceanic poleward energy transport.
The overlying atmosphere has a significant influence on
the sea ice cover by imparting a wind stress on the surface,
thus affecting the spatial distribution and export, and as a
medium for the poleward transport of energy from lower
latitudes. In documenting the September 2002 minimum,
at the time a record low for the satellite era,
Serreze et al.
[2003] noted anomalies in the summertime atmospheric cir-
culation including unusually low pressure over the central
Arctic Basin. Through ekman transport, low surface pres-
sure tends to cause the ice pack to diverge, which would
produce openings and increased melting via the ice-albedo
feedback mechanism after dispersal. Subsequent years have
not followed this pattern [
Stroeve et al.
, 2005]. rather, high
pressure at the surface has generally been noted in summer,
which would result in consolidation and the rapid removal of
ice from marginal sea embayments [
Ogi and Wallace
, 2007],
as well as a reduction in cloud cover, leading to an increased
shortwave radiative flux absorbed at the surface. In contrast
to the summertime focus of
Serreze et al.
[2003], there has
been considerable interest in the literature of the last decade
in changes to the mean wintertime atmospheric circulation
toward the positive index phase, or the “Greenland Below”
condition of the North Atlantic Oscillation (NAO) [
Hurrell
,
1995]. The NAO is an atmospheric teleconnection pattern
that is characterized by variability in surface westerlies over
the North Atlantic. This positive index phase was found to
result in a decreased flux of ice that traverses around the
Beaufort Gyre from the western to the eastern Arctic and an
increased ice export through Fram Strait [
Kwok
, 2000;
Rigor
et al.
, 2002]. In disrupting the Beaufort Gyre circulation,
thick multiyear ice in the pack is spatially confined to the
canadian Archipelago and away from the Alaskan and eur-
asian coastlines. The increased export through Fram Strait
produces a direct loss of multiyear ice from the basin. The
predominant wintertime mode of atmospheric circulation
reached extreme positive values in the late 1980s and early
1990s but has trended toward neutral in more recent years.
In this context, the recent reduction in summer ice cover
since 2002 is then seen as a dilatory impact of anomalous
wintertime atmospheric circulation [
Rigor et al.
, 2002].
An examination of the sea ice cover time series in com-
parison to observed radiative fluxes and other variables was
conducted by
Francis et al.
[2005]. Satellite-derived down-
welling longwave flux was found to account for 40% of
the variability in the September sea ice extent.
Francis et
al.
[2005] suggest that longwave radiatively driven melting
may have eroded the perennial sea ice cover in more recent
years after dynamically forced reductions in ice volume. The
result is consistent with observed increases in springtime
cloudiness over the Arctic [
Wang and Key
, 2005].
Finally, oceanographic research in the Arctic in the 1990s
was highlighted by the discovery of warming in the Atlantic
Layer, located at about 300 m in depth [
Quadfasel et al.
,
1991;
Carmack et al.
, 1995;
Grotenfendt et al.
, 1998]. This
layer of inflow water from relatively lower latitudes of the
North Atlantic is insulated from the surface sea ice pack by
a cold halocline layer, which itself was found to be in retreat
in the early 1990s [
Steele and Boyd
, 1998]. Large-scale ba-
sal melting of the ice pack would occur if the halocline was
significantly weakened or removed. The inflow of Atlantic
warm water into the Arctic Basin is, in turn, influenced by
atmospheric and oceanic processes local to the North At-
lantic, including trends in atmospheric circulation and the
oceanic meridional overturning circulation. Changes in the
temperature of the Atlantic layer have been found to be well
correlated with atmospheric indices such as the NAO [
Dick-
son et al.
, 2000;
Jones
, 2001]. Submarine data [
Boyd et al.
,
2002] have suggested that the cold halocline layer recovered
over the period 1998-2000 after declining in the early 1990s.
More recent studies have begun to comprehensively assess
the influence of the warm Atlantic layer on the overlying ice
pack, even in the presence of a robust cold halocline. For
example, observations in the Arctic using ice-ocean buoys
have indicated that weather systems can produce mixing that
extends through the halocline [
Yang et al.
, 2004]. A study of
Search WWH ::
Custom Search