Hi Fred!

This can be a confusing issue. As a result I generally avoid setting the value for magnetic declination before deployment, because then doing the correction in post-processing is much clearer. From my recollection the "headingBias" is applied during recording, so that the headings recorded by the ADCP compass are not the "raw" headings. This should mean that the final ENU velocities are then relative to True North, provided the compass calibration was accurate and the magnetic declination set correctly (e.g. for long deployments you may want to consider a changing declination value).

In the `oce` package that I use with the R language, we made the decision to "correct" the heading values back to the original if a nonzero "headingBias" is detected, and spit out a warning.

Not that this doesn't apply to the "headingAlignment" variable, which describes the offset between the heading source and beam 3. This is only applied during the final ENU step when processed with RDI software. This should only really apply to shipboard ADCPs though.

Cheers,

Clark

Hi Clark,

you know what? I was thinking that you would be the one answering!

You confirm what I thought and what other colleagues told me after my post, i.e., that the correction is made internally when transferring data into Earth Coordinates system. As you suggest, it is thus a good idea to avoid setting the declination to non-zero value. I'll check this for the future.

Thanks for the detailed answer.

Cheers!

Fred

No problem Fred!

And just to be clear (for others who may find this post), the correction is applied as the data is collected, rather than just when transforming to Earth coordinates. That means that even if you collect data in Beam coordinates, the heading as recorded by the instrument will already have had the heading bias value applied.

Clark

Ok...

But there are two things here. 'Heading Alignment' (EA command) and 'Heading Bias' (EB command), see attached document, section 2.4.1.

While the first command is for mounting an ADCP on a ship (physical alignment with the shipline), the second is for magnetic declination. What you say is that if only the magnetic bias is entered (e.g. a moored ADCP), and you record in beam coordinate, the velocities will be off? I thought in this case the correction was made during transformation into Earth coordinates.

Not quite. Velocities recorded in Beam coordinates do not use the compass information, so they will be correct. However, the raw binary file will have heading values recorded in it, which correspond to:

H = HC + EB

Where HC is the raw compass reading and EB is the Heading Bias (as described in Section 2.4.1). The "correction" is made at the time the data are recorded, regardless of what coordinate system the unit is set to record in, rather than just at the ENU transformation stage.

All I meant to point out is that if EB has a nonzero value, the heading as recorded in the `.000` files is not really the "raw" heading. This confused me for some time, as I had been told that if you record in beam coordinates, the heading values are just the raw compass readings.

And yes, the EA is another issue. If you ever have to worry about it I suggest you be very careful -- I had to spend some time chatting with an RDI engineer last fall trying to sort out an issue related to when EA is applied relative to EB (hint: EA is only applied during the instrument-to-earth transformation when it is done using RDI's software).

Sorry to confuse things a bit, I realized you got your answer but I wanted to warn others of the details of the heading bias correction in case someone else stumbles across this thread.

Cheers,

Clark

Thanks Clark,

Good answer for everyone. Since I never used EB before I was confused a bit, but for the future I'll be careful (for both EB and EA).

Good luck and hope to see you soon!

Fred