Project Title

Rotating Exchange Flows Through Multiple Channels

Name of Group Leader

Dr David Smeed

Home Laboratory

Southampton Oceanography Centre

E-mail address

das@soc.soton.ac.uk

Telephone

+44 23 8059 6407

The deep basins of the world’s oceans are separated by ridges and coasts, and the exchange between basins occurs in flow over sills and through straits.  Dynamical processes in the straits and sills control the value of the exchanges.  Thus these topographic features have a considerable impact upon the large-scale thermohaline circulation of the ocean, and hence upon climate. Many wide straits, where rotation is expected to be important are composed of multiple channels  separated by islands.  The project objectives were: 1) To quantify the exchange flux through a flat-bottomed strait and compare with existing theory, 2) Determine the qualitative and quantitative effects of an island on the exchange flow

     In each experiment the tank was divided into two reservoirs each with a different but uniform density.  The reservoirs were connected by a constricted channel.  In some experiments an island divided the channel into two passages.  The experiments were initiated by removing a barrier at one end of the channel.  A gravity driven exchange flow then ensued.  Experiments were conducted for a wide range of Rossby numbers.  Nineteen experiments were completed covering all of our objectives.

     The velocity field in the channel was measured on nine horizontal levels using the Correlation Imaging Velocimetry (CIV) technique.  This gave good results, with the majority of the flow adequately resolved.  Integraton of the velocity has enabled accurate calculations of the exchange flux.  Some problems with the CIV arose where the velocity very low, near the side walls, and in areas of high vertical shear.  This was remedied to some extent by thickening the laser sheet and by carrying out one experiment with vertical laser sheets that were moved horizontally to measure vertical velocity.  The problem of laser intensity at a large distance (particularly after passing through perspex) was largely remedied by subtracting the average image background, although this leads to more bad vectors and thus greater data  processing effort.  Some problems with residual motion in the reservoirs was evident at high rotation rates.

       Density measurements were carried out using 10 stationary (ultrasonic) and 3 profiling (conductivity/temperature) probes, located just outside the CIV camera view.  The data are found to be consistent with the velocity field measured by CIV.