Towed Microstructure Estimation of Diapycnal Diffusive Flux

Tim Duda
Chris Rehmann
AOPE Department
Woods Hole Oceanographic Institution
Woods Hole MA 02543

Two papers describe this work in detail.
1. Rehmann, C.R and T.F. Duda, Diapycnal diffusivity inferred from scalar microstucture measurements near the New England shelf/slope front, Journal of Physical Oceanography, 30, pp1354-1371, June 2000.

2. Duda, T. F., and C. R. Rehmann, Systematic microstructure variability in double-diffusively stable coastal waters of nonuniform density gradient, J. Geophys. Res., 107(C10), 3145, doi:10.1029/2001JC000844, 2002.

During the August 1997 Coastal Mixing and Optics (CMO) dye-injection diffusion experiments the dye-sampling towsled (Photo 1) (Photo 2) (Photo 3) (Chris, Tim, Towsled, Notebook, Socket Wrench, Rough Seas Photo) was fitted with an SBE-7 microscale conductivity probe from Seabird Electronics. This enabled measurement of conductivity gradients at spatial scales near the diffusive cutoff scale for thermal microstructure (the Batchelor scale). From these measurements the rate of rate of dissipation of thermal microstructure can be estimated, after properly accounting for salinity gradient contribution to the signals.

The thermal variance dissipation rate is closely related to turbulent buoyancy flux in stratified water.  This flux, caused by vertical movement of heat or salt,  alters the density structure and associated forces, and is therefore integral  to the circulation.  The flux rate is usually assumed to roughly follows Fick's law,  with the flux magnitude (of heat, salt whatever) given by an ``eddy diffusivity coeffecient'' K times the gradient (of heat, salt, whatever); e.g.  salt flux  < w' S' >= -K  dSo/dz, where w' is perturbation vertical velocity, and S'  is salinity  perturbation and  dSo/dz is the gradient of the mean field.

Large K would weaken the stratification, cause in large part by the summer sun heating the surface water, and small K would leave it unchanged.

In August 2004 the N-dependent K results of the JGR paper cited above were tested with a more comprehensive repeat study in the same area. 30 to 40 times more ueable data were collected than on the previous trip thanks to improvements to our system and to increased sampling time.

Terry Donoghue with the tow vehicle. terry and sled

Technical details: The probe assembly, differential amplifiers and anti-aliasing filter were powered from their own 3-W power supply. The signal was routed carefully to a commercial 16-bit A/D board in our PC/104 tow-sled computer assembly. The A/D operated in differential (8-channel) mode. The dynamic range of the system accepting the SBE 7 signals was measured to be 90.7 dB. The range was reduced to 77.4 db with the probe in air (zero conductivity gradient(?), 2.7 mv rms signal). This gives a minimum measurable dissipation of 7x10-12 K2/s in conditions of uniform salinity. Sampling rate was 400 Hz, giving 100 cpm resolution at 4 knots tow speed. The signal from the A/D was fed into the towsled PC/104 system for immediate transmission to the ship (Electronics System Block Diagram).


The dye sampling towsled looks at home in the water!

The emergency release float provides buoyancy for trim, counter balancing the tail. The tail provides flow stability at the nose, limiting attack angle fluctuations induced by ship heave and pitch, transmitted by the cable.

Two web pages show some basic 1997 results from
  1. Towing very near the bottom (70 m depth)  and

  3. A 30-km transect.

For mode comprehensive results please refer to the two journal publications cited above.

Coastal and Ocean Fluid Dyn. Lab Webpage
Ocean Acoustics Lab Webpage (tfd formal affiliation.)
CMO experiment Webpage