Funky Probe Check

Hi Guys,

Interesting one here. . .I'm presently working on a mobile bed model with high suspended sediment concentration (SG of the sediment is about 1.3). The velocities look great as they are being recorded, however the probe check deviates significantly from the norm.

Attached in Jpeg format is a screen-shot that I took today. There is good convergence of the beams at 50 mm, however there is return signal from throughout the water column (depth is about 16 cm). It's my thought that measurements made with a probe check such as this are OK for average velocities (say at 0.6D), but shouldn't be used to extract turbulent metrics.

Any thoughts?

Gavin.

Hey Gavin,

That is indeed a funky one… but it looks more like a grounding problem from the probe check output. Is your total depth 16 cm or is that the distance from the probe to the boundary?

Is the suspended sediment concentration really high in this flow, or is it just a moving bed?

What are you other data quality metrics looking like?

P.J.

Hi, you can get this kind of return from probe check if you have both strong scattering and strong attenuation.  I think this is real, it is just very confusing.

The interesting questions is whether the data is valid and the truth is that I am not quite sure.  Could you lower the transmit power level to see what happens?

Best regards, Atle Lohrmann

Thanks for the prompt replies,

P.J.: 16 cm was the distance from the transmitter to the boundary. Sediment concentration was high, and the bed was mobile, other metrics looked great (SNR 22+, COR 95%+, relatively smooth signal - no spikes).

Atle: I tried lowering the power and it worked well! While the amplitude dropped a bit (as I suspected) the probe check once again looks as it should. Interestingly, the magnitudes of coordinate velocities did not change appreciably between high and low powers.

That said, do you think measurements made with a probe check as shown in the provided Jpeg are OK (for mean values or turbulent metrics)? I plan on doing a bit of a sensitivity analysis after this next run, but just wondering what you're thoughts are.

Thanks again, and Best Regards.

Gavin.

Dear Gavin

We normally think of the sampling volume as the intersection between the transmit and receive beam.  The full truth is that the echo we measure comes from an ellipsoidal shell with the receiver and transmit elements as the foci and the shell thickness determined by the transmit pulse length.  This is then modified by the strength of the acoustical fields generated by transmit/receive ceramics so that the strongest (and relevant) return can be ascribed to the sampling volume.

In extreme scattering conditions, the return echo can be so strong that the receiver looses its ability to distinguish between echo coming from the sampling  volume and other parts of the shell.   The sampling volume will then grow in size until it (theoretically) extends to the full shell.  If this happens we loose the definition of direction that we rely on in order to resolve the 3D velocity.  I have only seen something like this happen once and it was in a cloud of micro bubbles.  In your case, I would simply suggest you keep an eye on any deviation between the two vertical velocity components since they should be the same as long as the sampling volume is well defined.

Good luck!

Best regards, Atle