Introduction - Sea Ice: Physics and Remote Sensing - page 14

It should be taken into consideration in regional and
global climate modeling.
The motion of the ice sheet may cause its divergence
and therefore the appearance of openings within the pack
ice in the form of cracks, leads or, polynya (for definition
see Section 2.6.2). The winter flux of heat from these
openings to the atmosphere can be two orders of magni-
tude larger than the heat flux through an adjacent thick
ice cover [ Maykut , 1982]. However, this sudden increase
may not last for a long time as water in these openings
usually freezes quickly. Lüpkes et al. [2008] found that
even a few percent of open water or refrozen thin ice in
the Arctic may increase the heat flux between the ocean
and the atmosphere drastically and increase air tempera-
tures within them by several degrees. In general, leads
and polynya are considered to be important climatic fea-
tures, and their distribution, along with the distribution
of thin, ice is particularly important to the regional and
perhaps global climate scenarios.
The primary optical property of sea ice that affects
weather and climate systems is albedo (the ratio of
reflected to incident optical radiation). Albedo from
snow‐covered ice (between 0.4 and 0.8) is significantly
higher that albedo from seawater (≈0.06). This causes
more sunlight to be reflected in the presence of ice
and therefore the ocean will not be as warm as it would
be in the absence of ice. That is how the ice‐covered areas
in the polar regions are climatologically cold. A possible
decrease in the area of polar sea ice, caused by global
warming, will cause more sunlight to be absorbed. This
will trigger a positive feedback loop that amplifies the
absorption and ice melting. In the past 30 years, air tem-
perature in the Arctic has increased at twice the rate of the
average increase around the globe. In relative terms, more
increase in winter temperature has been observed com-
pared to increase in summer temperature. Climatologists
are therefore interested in information about the ice extent
in the polar regions and the frequency and distribution of
melt ponds at the surfaces.
Using daily ice concentration maps retrieved from sat-
ellite passive microwave a few operational centers track
the ice extent in the Arctic region. Updated information
is available, for example, from the Arctic Sea‐Ice Monitor
System maintained by the Japan Aerospace Exploration
Agency (JAXA) in the IARC‐JAXA information system
(IJIS) located at the International Arctic Research Center
(IARC) in Fairbanks, Alaska. The minimum ice extent
usually occurs in mid‐September. The lowest ice extent so
far occurred on 5 September, 2012 (falling below 4 mil-
lion km 2 ) while the second and third lowest occurred in
2007 and 2011, respectively. The ice extent in the Arctic
has recently been reduced by 40%-50% of its average
value in the 1980s. This has also been accompanied by
reduction in ice thickness. A significant drop of about

0.6 m in multiyear (MY) ice thickness (and consequently
overall ice thickness) is observed since 2005. The decrease
in thickness has occurred at a rate of about 0.17 m per
year [ Kwok and Sulsky , 2010]. It should be mentioned
that the Antarctic ice extent has increased slightly in the
past three decades.
Early global climate models predicted that the polar
regions would undergo the largest greenhouse warming
[ Kattenberg et al. , 1996]. However, current climate models
underestimate the rate of ice retreat and thinning in the
Arctic. The ice extent and thickness estimated from satel-
lite remote sensing observations is four times faster than
model calculations [ Stroeve et al. , 2007]. Modelers indicate
that this is mainly due to a lack of accurate representation
of ice conditions in the models. For example, Rampal et al.
[2011] argue that the underestimation is caused by the
poor representation of the sea ice drift out of the Arctic
Basin. In general, climate models still show deficits in
reproducing the underlying processes for observed Arctic
sea ice decline [Stroeve et al., 2007]. More utilization of ice
information in climate models and further understanding
of sea ice processes in relation to the climate system should
ultimately improve regional and global climate modeling.

1.4.4. Sea Ice in Meteorology

In order to restore a temperature balance between the
cold polar atmosphere and the ocean, the latter always
tries to transport heat toward the poles. This process trig-
gers what is known as atmospheric circulation, which is
responsible for generating most weather systems over
north‐latitude areas. This includes the recurring low‐pres-
sure systems associated with storms and precipitation.
Obviously, the presence or absence of sea ice in the polar
regions affects these regional weather systems, directly or
indirectly. The effect is forecasted using weather models.
The models integrate a collection of physical processes
into a consistent framework that can be used to forecast
weather and project future climate conditions.
The traditional representation of sea ice information in
the model is achieved by feeding ice information directly
to the model at certain time steps. This approach was used
in the early 1970s to initialize weather forecast models
[ Stensrud , 2007]. At a basic level, the required information
at each grid cell should include ice concentration, but
additional data such as ice thickness, surface roughness,
and snow cover are also recommended. Other ice data that
can be parameterized such as albedo, surface emissivity,
and surface fluxes can be used as well. However, it should
be noted that parameterization of surface fluxes is a chal-
lenging task because of the heterogeneity and the rapidly
changing ice surface. Recent weather models incorporate
a sea ice module (or a coupled ice‐ocean model) to repre-
sent fundamental dynamic and thermodynamic processes.