 |
| Titel |
Comparison of Ellison and Thorpe scales from Eulerian ocean temperature observations. |
| VerfasserIn |
Andrea Cimatoribus, Hans van Haren, Louis Gostiaux |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250086501
|
| Publikation (Nr.) |
 EGU/EGU2014-382.pdf |
|
|
|
|
|
| Zusammenfassung |
| The geographical distribution of turbulent dissipation rate in the ocean is poorly understood at
best. Turbulent dissipation rate is measured either by means of microstructure shear
measurements, seldom performed in the open and deep ocean, or by means of adiabatically
reordering vertical profiles of density. The latter technique leads to the estimation of the
typical overturn size, the Thorpe scale, which can used to estimate average turbulent
dissipation rate by using empirical relations linking the Thorpe scale to the Ozmidov scale of
turbulence (Dillon, 1982). However, estimation of the Thorpe scale from temperature
measurements can be difficult if a reliable determination of the vertical density
profile is hindered by lack of resolution, salinity intrusions or low signal to noise
ratio.
We present here a method to estimate the typical overturn size by measuring the Ellison
length scale using frequency spectra of temperature. We apply the method to high resolution
temperature data from three moorings deployed at different locations around the Josephine
sea mount (North Eastern Atlantic Ocean). It is shown that the variance of the temperature
time series at the higher end of the internal wave frequency band and above is well correlated
with the overturn size. The method is based on a time frequency decomposition
using the “maximum overlap discrete wavelet transform”. This method can be a
viable alternative for indirectly estimating turbulent dissipation rate in the ocean
when limited vertical information is available but time resolution is sufficiently
high.
A major result is the indication that fine structure contaminated temperature
measurements can in fact provide reliable information on turbulence intensity. This method
could thus contribute to extending our knowledge of the turbulent dissipation rate
distribution, enabling the estimation of the latter from datasets where Thorpe scales cannot be
reliably determined. |
|
|
|
|
|
|