80
GORDON A. RILEY
is not much more than the diameter of the disk that was used for
measurement. I n order to make a rough estimate of the rate of coilvective circulation, the downdraft is assumed to be exactly 20 cm wide
with a vertical velocity of 6 cmisec. The wind speed is assumed to be
6 misec, and according to Faller and Woodcock (1964) the distance
between convergences would be about 30 m, and the depth of the mixed
layer would be about 24m in temperate latitudes. If the downdraft
went all the way to the bottom of the mixed layer, the time required
would be 400 seconds. Between the convergences there would be a
return movement toward the surface of a more diffuse nature, and if
it were uniform throughout the 30 m distance between convergences
the speed would be 1.4 rn/hour. The time required for complete overturn
would be about three-quarters of a day.
This rate of upward movement is many times faster than observed
sinking rates, and particles caught in the upward movement would be
carried back toward the surface. Stommel (1949) developed a mathematical model of the behaviour of sinking particles in these convection
cells. The model possibly is somewhat idealized in that he postulated
a symmetrical rotation, and the actual cells appear to be very assymmetrical, with strong downdrafts and much more diffuse upward
motion, but the general principles should hold.
Stommel showed that when the velocity of upward motion exceeds
the sinking rate there is an " area of retention " in the ceiitre of the
cell, in which the upward motion of water is sufficient to overcome
natural sinking tendencies, and particles within this area will be
trapped. Theoretically they would remain there permanently. In
actual fact, as Stommel pointed out, there would be turbulent exchanges between this area and surrounding regions. However, the
important point is that the rate of loss by sinking from the surface
layer would not depend simply on the sinking rate of the particles but
would be determined by all the variables involved in the convective
system.
The smaller the sinking rate is in relation to the velocity of the
water, the larger the area of retention will be, and in the present case
it would be quite large. The net effect is presumed to be that only a
small proportion of the particle content of the mixed layer will be
subject to loss at any one time, and this loss will occur as water reaches
the bottom of the convective layer and spreads laterally and before it
begins to rise again.
The form of the convective movement can be traced more easily in
the columnar cells that occur in calm weather than in Langmuir cells.
These are often seen in the Sargasso Sea, where the Sargasso weed
GORDON A. RILEY
is not much more than the diameter of the disk that was used for
measurement. I n order to make a rough estimate of the rate of coilvective circulation, the downdraft is assumed to be exactly 20 cm wide
with a vertical velocity of 6 cmisec. The wind speed is assumed to be
6 misec, and according to Faller and Woodcock (1964) the distance
between convergences would be about 30 m, and the depth of the mixed
layer would be about 24m in temperate latitudes. If the downdraft
went all the way to the bottom of the mixed layer, the time required
would be 400 seconds. Between the convergences there would be a
return movement toward the surface of a more diffuse nature, and if
it were uniform throughout the 30 m distance between convergences
the speed would be 1.4 rn/hour. The time required for complete overturn
would be about three-quarters of a day.
This rate of upward movement is many times faster than observed
sinking rates, and particles caught in the upward movement would be
carried back toward the surface. Stommel (1949) developed a mathematical model of the behaviour of sinking particles in these convection
cells. The model possibly is somewhat idealized in that he postulated
a symmetrical rotation, and the actual cells appear to be very assymmetrical, with strong downdrafts and much more diffuse upward
motion, but the general principles should hold.
Stommel showed that when the velocity of upward motion exceeds
the sinking rate there is an " area of retention " in the ceiitre of the
cell, in which the upward motion of water is sufficient to overcome
natural sinking tendencies, and particles within this area will be
trapped. Theoretically they would remain there permanently. In
actual fact, as Stommel pointed out, there would be turbulent exchanges between this area and surrounding regions. However, the
important point is that the rate of loss by sinking from the surface
layer would not depend simply on the sinking rate of the particles but
would be determined by all the variables involved in the convective
system.
The smaller the sinking rate is in relation to the velocity of the
water, the larger the area of retention will be, and in the present case
it would be quite large. The net effect is presumed to be that only a
small proportion of the particle content of the mixed layer will be
subject to loss at any one time, and this loss will occur as water reaches
the bottom of the convective layer and spreads laterally and before it
begins to rise again.
The form of the convective movement can be traced more easily in
the columnar cells that occur in calm weather than in Langmuir cells.
These are often seen in the Sargasso Sea, where the Sargasso weed