Post-correction for speed of sound errors

Posted by Tim Maddux at October 21. 2009

Hi,

Is it possible to post-correct Vectrino data for speed of sound errors? We have discovered that some of our probe heads are reporting incorrect temperatures and hence predicting the speed of sound in water incorrectly. Here is an example for 2 probes in well-mixed fresh water within 5 cm of each other in the vertical:

Incorrect:

% Temperature: -7.050 degC
% SpeedOfSound: 1375.700 m/s

Correct:

% Temperature: 8.520 degC
% SpeedOfSound: 1437.000 m/s

The problem is worse in summer months, as much as 10%. From what I've gotten in the literature, it looks like the estimated velocity is directly and linearly proportional to the speed of sound, so that I should be able to correct in the example above by multiplying by 1437/1375.7. Is this correct?

Thanks,

Re: Post-correction for speed of sound errors

Posted by P.J. Rusello at October 22. 2009

Hi Tim,

The speed of sound correction is pretty easy to do. Details are on page 21 of the old version of the users manual Revision C, October 2004. I didn't see this in the newest version you can download from the website, but the correction is very simple:

V_new = V_old * (c_new/c_old)

I would recommend having the instruments reporting negative temperatures examined as this is potentially a problem with the probe head's wiring and could cause some other problems with the data. Does the probe check look okay and the various data quality indicators appear normal?

P.J.

Re: Post-correction for speed of sound errors

Posted by Tim Maddux at October 22. 2009

Thanks P.J.!

The probes have been behaving normally, just the temperature seems to have faulted. Something I'll be in the habit of checking on more often now that I'm aware it can happen...

Re: Post-correction for speed of sound errors

Posted by P.J. Rusello at October 23. 2009

I'd check the malfunctioning probes for any physical damage just to make sure. I sent an email to Judah so don't be surprised if he gets in touch with you about this.

If you're working with largely constant temperature water you could just key in the speed of sound in the deployment planning like the old DOS Sontek ADV software required. Even with a couple degrees change in temperature your errors due to speed of sound should be fairly minimal.

P.J.

Sections

Post-correction for speed of sound errors