How Is Magnetic Declination Calculated?

[+Kordite]

My father called me up the other day. It seems that he was experimenting with the accuracy of his GPS by working with several waypoints, the corner of his back porch and a water tower on a distant hill.

 

Using the GPS to calculate the true bearing from one point to the other gave him one reading. Setting the compass to magnetic north gave him a number 9 degrees further west, which is the expected declination for SW Pennsylvania. These all worked out with bearings worked out on a map. When he got out his magnetic compass, it was west by another 4 degrees.

 

Now, I'm tendant to trust the math of the GPS over the compass given that the local magnetic field could be affected by any number of minor anomalies. He should have known this as he was a draftsman and surveyor but the GPS is a new instrument for him and he needed some reassurance.

 

This got me wondering about how the GPS come up with the difference between true and magnetic north.

 

The USGS map from the National Geographic Topo software for my Dad's home shows a declination of 9.5 degrees. While the map is dated 1979, the declination appears to be more current.

 

The NOAA Geophysical Data Center website at http://www.ngdc.noaa.gov/seg/geomag/jsp/Declination.jsp has the detailed model online. For that location it calculated the declination as 9.17 degrees currently.

 

I know that the satellites broadcast the ephemeris tables (the precise orbital information of the satellite) every 5 minutes or so but what about the calculation of magnetic declination? It's a pretty complex model. Is that also broadcast as part of the GPS signal? I assume so. And if my assumption is correct, is it a high-resolution table or is it simplified?

[+Miragee]

[reidster]

This is mostly pure conjecture, but… I’ve never read anything about magnetic declination being part of the GPS data stream and I’m inclined (pun intended) to believe the GPSr does it all internally.

 

It’s probably computed in a matter very similar to the way it does for ionosphere-delay corrections by using a fixed grid. However rather than receiving a set of dynamic grid-point corrections from satellites, static declination corrections would be stored in internal flash memory. The GPSr would interpolate its specific declination from the surrounding four grid-point values.

 

But I’m just guessing.

[+EraSeek]

Hmm. Don't really know. I assume it is part of the firmware and not part of the almanac. Just as a GPS thinks in terms of UTC and not your local time zone and thus has to be offset to adjust to your time zone, so the gps thinks in terms on TRUE north, Magnetic must be offset to local declination depenant upon your coordinates. How and to what degree of accuracy it does it, I don't know. I imagine it works much like the tidal calulations in my Map76.

[+RumJungle]

This is an interesting question. I know the magnitic pole changes. You can actually find out the position by visitiong a few sites like: Geological Survey of Canada or Woods Hole Marine Magnetism Group .

 

I suppose you could mark one of these coordinates as a waypoint and perform a goto to verify the calculation.

 

Could this be a virural cache??

 

LOL!

[Mittellegi]

In the software upgrade for the Garmin 12XL from version 4.57 to 4.58, the following is listed

 

"Updated magnetic variation tables based on the new IGRF 2000 model"

 

See Here

[+GSVNoFixedAbode]

'S'funny - a few weeks ago I sent an email to Garmin asking this exact same question. Their answer.....nothing so far.

[+NightPilot]

Magnetic variation is not broadcast by the satellites. It's calculated from tables in the firmware. Variation changes over time, and by different rates in different locations. A declination diagram from 1979 might be very far off or it might still be accurate - it depends on the location. Thus the satellites can't broadcast it, because they don't know where a receiver might be on the ground. The receiver has to find where it is, and then use the tables to apply the appropriate variation for that location.

[reidster]

In the software upgrade for the Garmin 12XL from version 4.57 to 4.58, the following is listed

 

"Updated magnetic variation tables based on the new IGRF 2000 model"

bexybear - Thanks for the info. This pretty much confirms how I thought it worked. Since these values change, but very slowly over time, there's no need to receive them constantly from satellites. However since they do change, this is another reason to periodically check for and install firmware updates.

 

Kordite - Sorry , my ionosphere-delay correction grid idea probably doesn't make sense if you're not familiar with WAAS and how it works. Visit Dale's DGPS site for a very good explanation of WAAS and where this correction grid idea came from...Dale's DGPS Info site

 

reidster

[+Kordite]

OK. That pretty well establishes that the table is stored in the firmware.

 

I'm curious as to how detailed the table is. And, based on bexybear's comment, how current the table is. The "new IGRF 2000 model" is already 5 years old.

[Dan Vull]

The Garmin unit in question is old. The Mag Dec from one year to another changes in such a small way that the only people that would notice such a small value would be surveyors and scientists. The IRTF or IGRF changes in such a small way every time an adjustment is run that you would hardly notice.

[+NightPilot]

The declination changes more rapidly in some locations than in others. In some places it only changes a degree every 10 years or so, and in some places it changes by more than a degree per year. This change is rather predictable, though, at least in the short run (a few years) so it's not difficult to know both the declination and what it will be in the future, and program that into the GPS. However, there is evidence that the earth's magnetic field may be beginning a reversal (there have been many over the last few billion years) and things could change, so your tables may be inaccurate in a few decades. But by that time you'll need a new GPS anyway, right?

[+apersson850]

reidster is correct above, in his assumption about how it's done in the GPS.