l ~ l l - ~ ~ ! S l O i %
OF OCEANIC’ HEAF AND MOMENTUM
369
divided by the kinematic viscosity. The Reynolds number must exceed 18.8/a
where Q is the angle of divergence.
Estimating the Reynolds number of the present flow as VL/v, where V is
the orbital velocity and L is the diameter of the orbit gives R - lo5. If
a = L/R, where R is the wavelength of the wave, Rcri,ics, lo4 which shows
that the mechanism is consistent with the apparent observation of marginally transitional turbulence. It also indicates the difficulty of ever observing such a phenomenon in laboratory waves.
Other mechanisms which could function include interaction of the surface
waves with current shear, internal waves or variation in the local density.
Since little is known of these phenomena for the present measurements, and
since the parameters characterizing these mechanisms are not clear, little
more can be said about them.
5. SUMMARY AND CONCLUSIONS
Attempts have been described to measure turbulence and turbulent temperature in the ocean and to relate the results to the vertical diffusivity of
momentum and heat. Using millimeter scale sensors it was possible to estimate power spectra of velocity and temperature fields to high wave numbers
in the viscous and thermal diffusive ranges, and to make use of universal
spectrum functions to estimate the normalizing parameters of universal similarity theories; namely, the viscous and diffusive dissipation rates E and x.
Consistency between different ci and xi values necessary to fit various i, j
portions of the measured u and T spectra was found in some cases, and is
taken as an indication that the fluid is indeed turbulent, and characterized
by the E~ -+ c and xi 4 x values that permit the fit. It was found that turbulent diffusivity parameters vr and D, estimated from E and x in the Croniwell
current were consistent with those estimated by independent methods. The
ratio VJD, was dependent on the Richardson number, ranging from I to
25.
Measurements of turbulence in the mixed layer have not been fully
analysed, but seem to show evidence of a small scale turbulence generation
mechanism in phase with the surface waves. Similar evidence apparently
exists in the data published by Grant et a/. (1968a) although in both cases
the possibility of an alternative explanation of the effect exists and further
measurements under more controlled conditions are called for.
ACKNOWLEDGMENTS
The authors iirc ymtcfiil lor the assistance and advice of a number of their colleagues at
UCSD in tltc course of this worh. particularly John Clay and Carl Friehe who participated in
both the FLIP iind Ownostofu nicasurements.
OF OCEANIC’ HEAF AND MOMENTUM
369
divided by the kinematic viscosity. The Reynolds number must exceed 18.8/a
where Q is the angle of divergence.
Estimating the Reynolds number of the present flow as VL/v, where V is
the orbital velocity and L is the diameter of the orbit gives R - lo5. If
a = L/R, where R is the wavelength of the wave, Rcri,ics, lo4 which shows
that the mechanism is consistent with the apparent observation of marginally transitional turbulence. It also indicates the difficulty of ever observing such a phenomenon in laboratory waves.
Other mechanisms which could function include interaction of the surface
waves with current shear, internal waves or variation in the local density.
Since little is known of these phenomena for the present measurements, and
since the parameters characterizing these mechanisms are not clear, little
more can be said about them.
5. SUMMARY AND CONCLUSIONS
Attempts have been described to measure turbulence and turbulent temperature in the ocean and to relate the results to the vertical diffusivity of
momentum and heat. Using millimeter scale sensors it was possible to estimate power spectra of velocity and temperature fields to high wave numbers
in the viscous and thermal diffusive ranges, and to make use of universal
spectrum functions to estimate the normalizing parameters of universal similarity theories; namely, the viscous and diffusive dissipation rates E and x.
Consistency between different ci and xi values necessary to fit various i, j
portions of the measured u and T spectra was found in some cases, and is
taken as an indication that the fluid is indeed turbulent, and characterized
by the E~ -+ c and xi 4 x values that permit the fit. It was found that turbulent diffusivity parameters vr and D, estimated from E and x in the Croniwell
current were consistent with those estimated by independent methods. The
ratio VJD, was dependent on the Richardson number, ranging from I to
25.
Measurements of turbulence in the mixed layer have not been fully
analysed, but seem to show evidence of a small scale turbulence generation
mechanism in phase with the surface waves. Similar evidence apparently
exists in the data published by Grant et a/. (1968a) although in both cases
the possibility of an alternative explanation of the effect exists and further
measurements under more controlled conditions are called for.
ACKNOWLEDGMENTS
The authors iirc ymtcfiil lor the assistance and advice of a number of their colleagues at
UCSD in tltc course of this worh. particularly John Clay and Carl Friehe who participated in
both the FLIP iind Ownostofu nicasurements.