Surveying True Azimuths using British Geological Survey Magnetic Declination Calculator

[andrewjsmith]

Hi Folk... I am interested in surveying azimuths with a precision and accuracy of about 1 minute of arc. I have been finding true north using a sunshot technique (theodolite, clock, nautical almanac) and have been getting good reproducible results which have been satisfactory. But the technique is complicated and time consuming. So I have fitted my theodolite with a tubular compass (see pic) and have been evaluating this as an alternative way forward. Of course I need to know the magnetic declination of the survey site, but this online calculator provides the answer to that -

http://www.geomag.bgs.ac.uk/data_service/models_compass/wmm_calc.html

In a recent series of test surveys, I have determined the local magnetic declination by finding the difference between the sunshot azimuths and the magnetic azimuths, and I have compared the declination found in this way with that calculated using the on-line calculator (link above). The two values of declination differ by about 0.07deg, which is outside my experimental error. The obvious place to look for the cause of this discrepancy is the presence of magnetic materials near the tubular compass, but I have rigorously excluded this possibility, so I am now questioning the accuracy of the on-line calculator. Is there anyone out there who has some experience of finding true azimuths magnetically, and who would like to comment on this ?

 

[IRstuff]

Did you look at the reference WMM cited in the calculator description? The WMM's equation provides a limited spatial efficacy. My recollection is that it's based on a 360-term harmonic equation.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[andrewjsmith]

Thanks for the comment. I have received a reply from the BGS. It is not good news...

"The accuracy that we assign to magnetic field model values is rather difficult to assess. This is because the observed field is made up of signals from a number of sources, primarily from flow of molten material in the Earth’s core (by far the largest and varies slowly in time), magnetic material in the underlying rocks in the crust (varies only on geological timescales but can vary dramatically over short distances), and the ionosphere and magnetosphere above us (which can vary a lot over short timescales).

The models only capture a certain part of the observed field, and for the World Magnetic Model (assuming this is the calculator you are using) this is only the core field signal. There is an assessment of errors associated with the World Magnetic Model in the report here (Figure 12) but a ball-park figure for your location is 0.35°.

I suspect it is the crustal field and the time-varying external field that will be the main reason for the differences you are observing but, as the World Magnetic Model is only revised every 5 years, there will be a difference arising from incorrect prediction of the core field as well. The difference will also depend on the instrument you are using to sense the field."

 

[IRstuff]

I had arrived a similar value in the past. The issue is that things the size of a car can affect your compass readings, but 360 terms in the equation can only resolve so far...

Your best bet, in my view, is to stick with theodolites or transits use RTK surveyed monuments. If your RTK errors are under 2 cm, you only need about 200-m baselines to meet your accuracy requirement

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[andrewjsmith]

Yes - thanks again. One of the problems is that here in the UK we don't get that much sunshine !