Geology Reference
In-Depth Information
It should be noted, however, that efforts to incorporate sea
ice information in weather models are limited so far to
identification of ice regardless of its type. However, thick-
ness information (a proxy indicator of ice type) is also
desirable even at coarse resolution.
One of the challenges in representing sea ice informa-
tion in weather models is the spatial and temporal varia-
tion of the sea ice cover in terms of its physical parameters
(e.g., temperature and emissivity) and surface geometry
(leveled or deformed ice). While the surface temperature
of the Arctic ice is fairly homogeneous in winter (can be
as low as −40 °C), the open‐water areas in leads and pol-
ynyas or open‐water areas beyond the ice edge have a sur-
face temperature at approximately −1.8 °C (the freezing
point of seawater). This sharp contrast makes ice‐covered
oceanic regions among the largest spatially changing sur-
face temperature on Earth. Accurate characterization of
the spatial distribution of ice surface temperature is cru-
cial to improving weather forecast, particularly during
winter, in areas where leads become closely distributed
within the ice cover. The same applies in summer when
melt ponds with its substantially less albedo than sur-
rounding ice also become closely distributed.
The importance of sea ice in meteorology and weather
forecast can also be appreciated using the following simi-
larity. Global ocean‐atmosphere phenomena affect
regions very far away from their origin. El Niño, for exam-
ple, starts in the middle of the Pacific Ocean, yet it affects
temperature and precipitation in North America in win-
ter. Arctic sea ice, in turn, must have an effect on regions
far from the Arctic. This is yet to be understood using
regional weather models that focus on the Arctic, then
upscale the results using a global model. A project has
been undertaken in Environment Canada to develop a
weather forecast system called Polar‐GEM to improve the
representation of sea ice at northern latitudes by coupling
a detailed dynamical/thermodynamic sea ice model (GEM
is Global Environmental Multiscale; the operational
weather forecast model used by Environment Canada's
Canadian Meteorological Centre). Particular emphasis is
put on improving representation of surface processes such
as using detailed dynamic‐thermodynamic models cou-
pled with ocean currents over the Arctic basin. Better rep-
resentation of snow processes and air‐sea interactions are
also implemented using detailed snow models over sea ice.
A recent project has been undertaken by Environment
Canada to develop a regional “sea ice analysis” system.
It is primarily designed to satisfy the requirements for
planning of marine transportation and other marine
operations in ice‐rich waters around North America.
Additionally, it will satisfy the needs of regional numeri-
cal weather prediction models and regional sea ice
model initialization. The term “sea ice analysis” means
a forecast system using satellite‐ or ground‐based
observations incorporated in a data assimilation
approach. The analysis is produced using the commonly
known three‐dimensional variational (3D‐Var) data
assimilation approach. In its first version, the system
produces only ice concentration at approximately 5 km
resolution using a 6 h persistence forecast from the pre-
vious analysis as the background state. The assimilated
observations encompass measurements from a few satel-
lite sensors. In addition, ice concentrations estimates
from operational ice analysis of synthetic aperture radar
(SAR) at the Canadian Ice Service as well as estimates
from passive‐microwave sensors are also assimilated.
The system is described in
Buehner et al.
[2013]. Sea ice
information required for weather forecast can be tracked
at an appropriate scale using remote sensing observa-
tions, a sea ice model, or “sea ice analysis.” Remote
sensing is an important tool, though improvement of
accuracy and spatial and temporal resolutions are still
needed. Timely information is required, particularly in
highly dynamic regimes such as ice edge where ice can
drift up to 40 km in a day.
Growing scientific evidence has pointed to the influence
of the current decline in Arctic sea ice (as explained later
in this chapter) on atmospheric phenomena within and
beyond the Arctic.
Budikova
[2009] found that changes in
Arctic sea ice interact with processes related to atmos-
pheric wind and temperature fields as well as to thermody-
namic and radiative processes connected with water vapor,
clouds, and aerosol feedbacks.
Cassano et al.
[2013] used
climate model simulation to explore the effect of the
increasing open water in the Arctic on Earth's atmosphere.
They found that the increase in open water during the fall
of 2007 (this season featured the third lowest Arctic sea ice
extent on record) led to an increase in atmospheric tem-
perature. The higher temperatures in the Arctic thus
caused a decrease in the pole‐to‐equator temperature
gradient, which in turn created a weaker jet stream and less
storminess in the midlatitudes. To conclude this section, it
is worth repeating that sea ice conditions, while of interest
in their own right, are important information for improv-
ing weather forecast models, both regional and global.
It is worth noting that the term “meteorology” encom-
passes sea ice according to some definitions. For example,
the CIS, a division of the Meteorological Service of
Canada (MSC), is the leading authority for information
about ice in Canada's navigable waters. Weather and ice
services operate under the same federal Department of
Environment (DOE) in Canada.
1.4.5. Sea Ice in Oceanography
During the winter months of the Northern Hemisphere,
when the oceans in the Arctic region freeze, sea ice in the
Antarctic region melt. A reversal in this process occurs
