Geology Reference
In-Depth Information
It is important to mention here that the formation of air
bubbles in sea ice takes a somewhat different path than
that in freshwater ice. Air bubbles in young and FY sea ice
are usually not directly visible because they are small and
often incorporated within the brine pockets as illustrated
in Figure 2.26. During directional solidification in seawa-
ter, rejection of air occurs simultaneously with the rejec-
tion of dissolved salts. Both the salts and the air are pushed
to the dendritic or platelet boundaries in the skeletal layer
as the freezing front moves downward in the melt. However,
subsequent desalination processes, to be addressed later,
do not involve the migration of air pockets. The air pock-
ets may coalesce with each other and thereby increase in
their sizes, but they remain trapped along the subgrain
boundaries or inside the ice lattice as the case may be.
out both in the field and laboratories.
Weeks and Lee
[1962] found that the average bulk salinity of new sea ice
(off Labrador Sea, Canada) loses roughly 20‰ of the
original water salinity during the initial formation and
about 5‰ more during the first few weeks after forma-
tion.
Nakawo and Sinha
[1981] conducted perhaps the
most comprehensive investigations on the growth and
salinity profile of FY sea ice in Eclipse Sound, Baffin
Island near Pond Inlet (lat. 72.7°N., long. 78°W) during
the entire winter season of 1977-1978 after the freezing
started on 30 November 1977. They found the salinity of
freshly solidified ice to drop rapidly and immediately after
initial freezing. On 2 December, the salinity at the bottom
of the ice sheet was found to be 25‰. As this level was
incorporated within the bulk ice, the salinity decreased to
8.5‰ on 9 December. Of course, the former measurement
was rather dubious because it was impossible to remove
the ice core from the ice sheet without some drainage
during the coring process. Thus the actual drop in salinity
was even higher. Nonetheless, the authors provided con-
vincing arguments with a series of weekly measurements
that major reduction in the brine content would occur
within a week after ice formation. They also presented, as
described in section 3.2, a method for quantitative assess-
ment of salinity loss at various depths of ice with time
as the ice thickness increased. In a recent study
Notz and
Worster
[2008] measured bulk salinity profiles of young
ice in the Arctic from the onset of freezing until the ice
had reached a thickness of 200 mm with high vertical and
temporal resolutions. They found that salt fluxes emanat-
ing from the ice were as high as 90 g/m
-2
· h during the first
few hours of new ice formation and were roughly half as
large during later stages of the experiments.
2.3.3. Salinity Loss During Ice Growth
The low salinity of freshly frozen sea ice compared to
that of seawater has been reported in numerous obser-
vations since the popularly known studies of
Malmgren
[1927] in the Arctic Basin. Salt rejection from growing
sea ice has a controlling influence on the buoyancy
forces and ecology of the water in the polar oceans. The
salt that remains in the ice in the form of brine is a key
factor in determining the physical properties of sea ice,
as will be discussed in Chapters 3 and 4. It is usually
expressed in terms of brine volume fraction. Salt is
rejected from the ice into the underlying seawater
through two major processes, one is rapid and the other
is slower. A review of these processes is presented in sev-
eral recent publications including
Petrich and Eicken
[2009],
Weeks
[2010], and
Hunke et al.
[2011]. In quanti-
tative manners, the terms “rapid” and “slow” salt rejec-
tion were introduced by
Nakawo and Sinha
[1981] and
will be used in this topic.
The rapid process is the salt rejection at the ice‐water
interface as seawater freezes while rejecting impurities.
This process takes place within the vicinity of the skeletal
layer, schematically depicted in Figure 2.24, and is usu-
ally active during the first few hours to few days of freez-
ing at the interface, depending upon the growth rate. The
slow process of salt rejection, on the other hand, entails
the subsequent brine drainage from the bulk of the ice
sheet as freezing advances. While most of the drainage
takes place toward the ice‐water interface, some brine
may move upward to the surface by diffusion to snow on
the surface if precipitation occurs or is affected by subli-
mation processes at the surface.
Field measurements in the Arctic indicated that the
salinity of freshly frozen sea ice decreases rapidly during
the period after initial freezing. One of the most well‐
known series of investigation was carried out by
Malmgren
[1927]. Subsequently, numerous studies have been carried
2.3.3.1. Initial Rapid Salt Rejection at the Ice‐Water
Interface
At the ice‐water interface, salt is rejected from seawater
as it freezes. The salt rejection depends on the ice growth
rate and the salinity of the underlying seawater. Its amount
can be determined from equation (2.24), which can be
rewritten as
C
C
K
KK
Vx
D
(2.31)
0
(
)exp
This equation gives the ratio of the concentration of
the salt retained in the ice to that in the seawater at depth
x
under the ice‐water interface. According to the defini-
tion of
K
given in equation (2.24), this factor equals to
C
0
/
C
when the growth velocity (interface propagation)
V
= 0 (or extremely low growth rate). For that reason
K
in the above equation is denoted
K
0
. The ratio of the
