Karbe, Fritz Richard (2008) On the estimation of turbulent mixing from high frequency lowered ADCP data in the tropical Atlantic. (Diploma thesis), Christian-Albrechts-Universität zu Kiel, Kiel, Germany, 69 pp.

Text Dipl. 2008 Krabe, F.-R.pdf - Reprinted Version Restricted to Registered users only Available under License German copyright act UrhG. Download (12MB) | Contact

Abstract

Data sampled with a high frequency 1200 kHz lowered Acoustic Doppler CmTent Profiler system attached to a CTD rosette during a research cruise in the tropical Atlantic in 2006 are analyzed. The motive is to examine wether it can be used to obtain information associated with turbulence in the ocean. The ability of doing so would be striking, since lADCP measurements are common and numerously. Furthermore a CTD system can easily be extended to an CTD/IADCP system, and if one could infer on parameters of turbulence using such measurements, it could lead to savings of ship time and thus money, since microstructure measurements are time consuming. Besides that, microstructure datasets are still sparse and mostly cover only a small portion of the watercolumn. Several approaches are pursued in order to analyze the data extensively in respect to the possibility to infer on turbulence. Vertical gradients of horizontal velocity, as well as variance of velocity fluctuations are compared to direct estimates of turbulent kinetic energy dissipation rates from microstructure measurements. Since this comparison yields no significant results, reasons to explain this are shown to be associated with scales of turbulence, as observed by microstructure measurements. These scales seem to be not sufficiently large enough to be observed with the used vertical resolution of lADCP measurements. Furthermore the noise level of the high frequency IADCP system is examined in detail, analyzing different setups used during the trials of the lADCP system at the beginning of the cruise. Noise levels of high resolving setups of the lADCP system are shown to be quite large, of 0(2 cms-1) for the setup of a binsize of 0.5 m and a sample rate of 2 Hz used during most of the observations. Although an enhanced resolution with a bin size of 0.25 m and a sample rate of 2.5 Hz would allow to resolve the scales of turbulence found, associated noise levels are shown to be of 0(8crn s-1 ), making it difficult to obtain reliable data. Thus using a 1200 kHz ADCP in conventional forms of lADCP profiling is unlikely to be able to quantify the strength of turbulence. A possible solution could be averaging IADCP data over longer periods, stopping the instrument in the watercolumn for a certain time. In a last approach the gradient Richardson and Froude numbers are derived using CTD and IADCP data and compared to microstructure measurements. They are shown to have a slight connection to the observed turbulence, but give no reliable basis to measure turbulence with such data only. However, a more sophisticated derivation of parameters from CTD and lADCP data for the derivation of Richardson numbers could probably improve the indicated relationship to some degree.

Actions (login required)

View Item