Over most of the world ocean the water in the mixed layer is less
dense than the
water below and convection is suppressed. Notable exceptions
are the far North
Atlantic along the edge of the Arctic pack ice and in the Weddell Sea
along the
edge of the Antarctic pack ice. In these regions the surface
waters can become
sufficiently cold and saline to enable them to sink to the bottom of
the ocean.
This oceanic convection is upside down relative to the convection in
the
atmosphere. Instead of involving buoyant, rising plumes of moist
fluid, it
involves sinking plumes of cold, saline water that is denser than the
surrounding water.
How does the surface water in these regions become cold and saline enough
to
sink to the bottom? Two processes are responsible: evaporation
and freezing.
Evaporation systematically removes fresh water from the surface waters
of the
ocean, leaving the residual water more saline. In changing phase
from liquid to
vapor, the evaporating air molecules also remove heat from the residual
ocean
water that will later be realized as latent heat of condensation in
clouds. The
rate of evaporation is particularly high over the North Atlantic and
the Weddell
Sea, where cold polar air masses flow out over warm ocean waters, creating
conditions that favor vigorous (atmospheric) convection. This
atmospheric
convection efficiently removes heat and water vapor from the sea surface
by
carrying them upward in buoyant plumes. The surface air that
ascends in the
plumes is replaced by drier air from above, that is quickly moistened
by
conduction when it comes into contact with the water surface.
But evaporation
alone is rarely sufficient to make the surface water of the ocean dense
enough
to sink. It is through the process of freezing, as occurs during
winter along
the edge of the polar sea ice, that the surrounding water acquires
the
additional density that makes it negatively buoyant. Although
the ice forms
from salt water, the ice itself is composed entirely of fresh water.
In the
process of freezing, salt molecules are systematically rejected.
Through this
process of 'brine rejection', freezing increases the salinity of the
residual
sea water that isn't incorporated into the ice, thereby increasing
the density
of patches of water near the edge of the ice to the point at which
they sink to
the bottom. Oceanographers refer to this process as 'the formation
of bottom water'.
The newly formed bottom water carries with it the properties that it
had when it
was up at the surface. It is close to 4 degrees C (39 F), the
temperature at
which the density of water is highest (i.e., if water at 4 C is either
cooled or
warmed, it expands; water at the freezing point is less dense than
water at 4
C). Ocean chemists can estimate the date when bottom water in
various parts of
the ocean was formed by measuring the concentration of dissolved freon
(chlorofluorocarbon--CFC) gases), the first of which began to be manufactured
during the 1950's. From these measurements North Atlantic bottom
water has been
tracked over the past few decades as it makes its way southward in
the bottom
branch of the thermohaline circulation. To picture how this circulation
affects
the remainder of the world ocean, oceanographer Wallace Broecker has
used the
analogy of a global scale 'conveyor belt' as pictured in Fig. 5-14
of the text.
Bottom water doesn't form over the North Pacific because the water is
too fresh
to sink, even along the ice edge. Evidently, there's just a little
more
precipitation relative to evaporation over the Pacific basin and the
river
drainage basins flowing into it than there is over the Atlantic.
The strength of the thermohaline circulation is determined by the rate
of
sinking of negatively buoyant plumes in the far North Atlantic which,
in turn,
depends upon the density of the water in that region. If the
ocean in this
region were capped by a thin layer of water that was too fresh to become
dense
enough to sink, the thermohaline circulation would shut down.
Among the
processes that could freshen the surface waters in this region are
an increase
in runoff of fresh water from rivers (like the St. Lawrence) that flow
into this
region; enhanced melting of Arctic sea-ice and/or the Greenland continental
ice
sheet; or a local increase in precipitation or decrease in evaporation
due to a
change in atmospheric wind patterns.
How the Ocean Circulation Impacts Climate The wind driven subtropical
gyres
transport transport warm water poleward on the western sides of the
oceans and
warmer water equatorward on the eastern side. The exchange of
equal amounts of
water containing different amounts of heat, results in a net poleward
transport
of heat, which supplements the poleward heat transport by atmospheric
motions
and helps to even out the temperature contrast between the tropics
and the polar regions
The thermohaline circulation is most intense over the North Atlantic,
the
primary region of bottom water formation. Surface waters flow
northward,
cooling as they approach the Arctic. Eventually they sink in
negatively buoyant
plumes, and subsequently return to lower latitudes as bottom water.
At a
latitude of, say 40 N, relatively warm water is flowing northward in
the upper
branch of thermohaline circulation near the ocean surface and an equal
mass of
much colder bottom water is flowing southward. The northward
flowing surface
water carries with it more heat than the southward flowing bottom water.
This
exchange of equal masses of water results in a net northward flux of
energy.
In the North Atlantic the wind driven subtropical gyre and the thermohaline
circulation transport relatively comparable amounts of energy poleward
and
together they transport roughly half as much energy poleward as the
atmospheric
circulation does. The wind driven circulation accounts for the
shape of the
surface currents but the thermohaline circulation contributes substantially
to
the strength of the Gulfstream. If the thermohaline circulation
were to shut
down, the Gulfstream would weaken considerably, the poleward heat transport
would decrease, and the winters in Northern Europe, which are currently
remarkably mild for the latitude, would be much colder (as apparently
was the
case during the 'Younger Dryas' period that will be discussed by Group
D).