THE NEAR-SURFACE LAYER OF THE OCEAN
sudden change of the gas transfer velocity due to the direct influence of the
steep capillary waves on the aqueous molecular sublayer.
Soloviev and
(1998) proposed an alternate explanation of the
Wu (1996) results: The change of surface roughness due to capillary waves
could directly influence the flow on the airside of the interface, thus
modifying the wind stress. The sudden change could be merely a reaction to
enhanced roughness modifying the wind field rather than a direct impact of
the ripples on the aqueous molecular sublayer. The change of roughness
could result in a sudden increase of the friction velocity and hence in the
intensification of the air-water gas exchange. Another interpretation of the
Wu (1996) results points to the connection between microscale wave
breaking (rollers) and the parasitic capillaries (Soloviev and
1998). The rollers, if present in the Wu (1996) experiment, would result in a
significant effect on the air-sea exchange. In any case, this is a rather indirect
influence of capillary waves on the diffusion sublayer.
There is nevertheless evidence of a direct impact of the capillary waves
on air-sea exchange. Saylor and Handler (1997) experimented in a small
laboratory tank with capillary waves from 2.62 to 3.62 mm wavelength
(which corresponds to 400 to 200 Hz frequencies) and found an almost two
orders of magnitude increase in the interfacial gas transport rate as the wave
slope was increased from zero to 0.2 m m
-1 . In this work, small vertical
vibration of the tank generated capillary waves via the Faraday instability.
The Saylor and Handler (1997) experiment provides remarkable evidence
that capillary waves can greatly increase fluxes across the air-water
interface. Applicability of these results to the real ocean, however, is not
completely clear since the Faraday waves differ from the parasitic capillaries
observed in a wind/wave tank or on the open ocean surface. In natural
conditions, the capillary waves of these frequencies (from 200 to 400 Hz)
dissipate quickly and may only cover a very small percentage of the sea
surface, while in the tank waves excited via the Faraday instability
completely covered the water surface.
As seen on the ocean surface, capillary waves indeed appear suddenly
when the wind speed exceeds some threshold level. The wind speed has not
only a mean but also a variance that makes the sea surface patchy with
respect to the coverage with capillary waves (the so-called “cats paws”). As
the wind speed increases, the area covered by ripples gradually increases so
that the surface averaged over a larger area should demonstrate a smooth
transition from no capillary waves to full coverage without an obvious
“jump”. This is relevant to the mean gas transfer (i.e., averaged over some
space and time intervals). The sudden increase should only be observed on a
small scale that might be relevant to fluctuations but not to the mean
exchange.
The sudden increase in gas transfer has been observed mainly (if not
exclusively) in laboratory studies (see for instance Fedorov and Ginzburg,
84
Schl ssel
ü
Schl ssel
ü
sudden change of the gas transfer velocity due to the direct influence of the
steep capillary waves on the aqueous molecular sublayer.
Soloviev and
(1998) proposed an alternate explanation of the
Wu (1996) results: The change of surface roughness due to capillary waves
could directly influence the flow on the airside of the interface, thus
modifying the wind stress. The sudden change could be merely a reaction to
enhanced roughness modifying the wind field rather than a direct impact of
the ripples on the aqueous molecular sublayer. The change of roughness
could result in a sudden increase of the friction velocity and hence in the
intensification of the air-water gas exchange. Another interpretation of the
Wu (1996) results points to the connection between microscale wave
breaking (rollers) and the parasitic capillaries (Soloviev and
1998). The rollers, if present in the Wu (1996) experiment, would result in a
significant effect on the air-sea exchange. In any case, this is a rather indirect
influence of capillary waves on the diffusion sublayer.
There is nevertheless evidence of a direct impact of the capillary waves
on air-sea exchange. Saylor and Handler (1997) experimented in a small
laboratory tank with capillary waves from 2.62 to 3.62 mm wavelength
(which corresponds to 400 to 200 Hz frequencies) and found an almost two
orders of magnitude increase in the interfacial gas transport rate as the wave
slope was increased from zero to 0.2 m m
-1 . In this work, small vertical
vibration of the tank generated capillary waves via the Faraday instability.
The Saylor and Handler (1997) experiment provides remarkable evidence
that capillary waves can greatly increase fluxes across the air-water
interface. Applicability of these results to the real ocean, however, is not
completely clear since the Faraday waves differ from the parasitic capillaries
observed in a wind/wave tank or on the open ocean surface. In natural
conditions, the capillary waves of these frequencies (from 200 to 400 Hz)
dissipate quickly and may only cover a very small percentage of the sea
surface, while in the tank waves excited via the Faraday instability
completely covered the water surface.
As seen on the ocean surface, capillary waves indeed appear suddenly
when the wind speed exceeds some threshold level. The wind speed has not
only a mean but also a variance that makes the sea surface patchy with
respect to the coverage with capillary waves (the so-called “cats paws”). As
the wind speed increases, the area covered by ripples gradually increases so
that the surface averaged over a larger area should demonstrate a smooth
transition from no capillary waves to full coverage without an obvious
“jump”. This is relevant to the mean gas transfer (i.e., averaged over some
space and time intervals). The sudden increase should only be observed on a
small scale that might be relevant to fluctuations but not to the mean
exchange.
The sudden increase in gas transfer has been observed mainly (if not
exclusively) in laboratory studies (see for instance Fedorov and Ginzburg,
84
Schl ssel
ü
Schl ssel
ü