According to an atlas I have published in 1939, the North Magnetic Pole has migrated about sixty miles northward since that year. Does that mean it will change the earth’s magnetic field, for the worse? What instruments are used by scientists/geologists/whatever to establish the location of the magnetic poles? (The old atlas didn’t have any information about where the South Magnetic Pole was in 1939.)
I heard once that the Van Allen Belts are kept aloft by the earth’s magnetic field; if the magnetic poles shift inordinately, the field would collapse, and so would the Van Allen radiation belts–thus fulfilling a prophecy in the book of a “minor prophet” about tongues rotting in mouths" (ewww!)–Zechariah 14:12.
I hope I am just unduly alarmed…
dougie_monty,
No need to be alarmed (just yet, anyway ;)). The drift of the north magnetic pole is called the secular variation of the Earth’s magnetic field. (That simply means that the pole position varies with time. Incidentally, the south magnetic pole position changes in step with the north magnetic pole). If you look at the bottom of a topographic map sheet, you’ll see a little symbol telling you the difference between the heading for the north geographic pole and the north magnetic pole for the area shown on the map. This difference is called magnetic declination, and it varies depending on where you are on the Earth’s surface. The magnetic declination is given in degrees and is dated, and the amount of drift per year is also noted. That’s so that when you’re standing out in the middle of nowhere with your map and trusty compass 16 years after the map was printed, you’ll be able to make the appropriate adjustments to your compass and find your way back to camp. (A real compass like a Brunton, by the way, not the little things they put on key chains & such.) As you’ve noticed, secular variation can really add up in a relatively short time, so without info on declination drift you can get seriously lost if you’re relying on a compass to determine direction. If you average the locations of the magnetic pole over a long-enough interval of time (say, a few tens of thousands of years or more), you’ll find that the time-averaged magnetic pole position is basically the same as the geographic pole position. (BTW, magnetic field strength is measured with… a magnetometer. No joke. Different instruments make the measurements in different ways; I can point you toward a reference or two, if you’d like.)
Why does the position of the magnetic pole shift at all? On short time scales (a few thousand years), it has to do with the fact that the Earth’s magnetic field (EMF from now on) is a somewhat complex beast. To first order, we can envision the EMF as having a geometry very much like the magnetic field around an ordinary bar magnet (i.e., a dipole), where magnetic lines of force emerge from one end of the magnet and loop around back into the other end. For the present-day EMF, that portion of the field which behaves like a dipole (about 90% of the field strength) is aligned approximately with the earth’s axis of rotation, so the magnetic poles lie close to the geographic poles. The remainder of the EMF is non-dipole in geometry (there are several orders of higher frequency spherical harmonics involved – ack!) which perturbs the overall EMF geometry. Some non-dipole components of the EMF drift westward with time, and are the principal causes of short-term secular variation.
On longer time scales, the Earth’s magnetic poles do swap position; this event is known as a reversal. Reversals have happened many times in the past; in fact, the pattern of reversals going back as far as 200 million years ago is frequently used to help constrain the timing of geologic events. (Reversals are known from even more distant periods, but the record is not continuous and so is less useful as a dating tool.) At present, we have no real explanation for while field reversals happen, although some researchers have been successful in modeling small bits of field-like behavior. The difficulty lies in understanding the complex magnetohydrodynamics of the Earth’s liquid outer core, where most of the EMF is generated.
Just what happens during a reversal of the Earth’s field? We don’t really know that yet either, because field reversals are believed to occur over short intervals (3-10,000 years) and the geologic record seldom has a high-enough resolution to record such events. IIRC, there is some thought that the dipole component of the EMF weakens, allowing the non-dipole components to temporarily come to the fore (meaning that your trusty compass would be relatively useless). The overall strength of the magnetic field probably weakens as well.
This last bit gets to the heart of your worries, I think. The EMF does not support the Van Allen belts, but it does shield us from some cosmic radiation. With a weakened field, some additional radiation would make it to the Earth’s surface; but since reversals have happened before without accompanying mass extinctions, I don’t think that we’d be looking at a problem that much more serious than losing the ozone layer, say. Now if your tongue rots in your mouth, you’ll have to find something else to blame it on. 
BTW, Unca Cecil fielded a similar question but answered it by discussing (sort of) a phenomenon called true polar wander, which is distinct from what I wrote about above. In true polar wander, the thought is that the Earth’s entire crust and mantle shift rapidly with respect to the orientation of the dipole component of the EMF, so that it looks as though the magnetic pole has taken a sudden hike across the Earth’s surface. (The reasons why that might happen are a little too long-winded to go into here – I’ve taken up enough space. J) There are a few times in Earth history when this is thought to have occurred, but the jury is still out (the data are a bit sparse).
Last comment – no, magnetic field reversals don’t “funnel cold air down from the poles” to usher in new ice ages. Just thought I’d mention that, in case any Gordon Michael Scallion fans read this.
The Van Halen belts, now, that’s a different thing altogether.
I’m a loner, Dottie … a rebel.
Maybe they would catch fire, as happened in the Irwin Allen classic “Voyage to the Bottom of the Sea.”
fillet,
You mentioned that magnetic reversals have been identified as far back as 200 million years. I recall reading that the frequency of magnetic reversals was much, much lower during the Mesozoic and picked up around the beginning of the Tertiary. This lead me to jump to the conclusion that the frequency of magnetic reversals was some how related to periods of increased tectonic movement as most of the Mesozoic was comparatively tectonicly stable. (The proof is left to the reader.) If the actual generation of the magnetic field is due to convection currents in the mantle or the core, wouldn’t reverals be a result of a drastic change in those currents. Those changes would be the result of changing tectonic plates interfering with an existing convection regime.
mipsman,
We actually have records of reversals occurring much earlier than the Mesozoic; my own thesis research identified reversals in rocks roughly 600 million years old. I don’t recall offhand, but I don’t think these are the oldest reversals ever found, either. The record of reversals is very well documented up to ~180 million years ago because that’s the age of the oldest ocean crust that has not yet been subducted. There are some intervals in the Triassic that are also well-known from the Newark Supergroup rift basin lake deposits. The record prior to then is pretty spotty.
After taking another look at my geologic time scale, I see that the frequency of reversals during the Mesozoic Era as a whole is not that low. I think you’re thinking specifically of the Cretaceous quiet zone, an extremely long (~40 million years) magnetic normal interval. Further back at the end of the Paleozoic Era, there’s also seems to be an extremely long magnetic reverse interval, called the Kiaman superchron, which lasted about 70 million years.
The link between the magnetic field, outer core convection and mantle convection puzzles geomagneticists and paleomagneticists to no end. The pattern of polarity reversals is apparently a completely stochastic process, one of the few major Earth processes that can’t be explained by some sort of periodic activity. People have suggested that long intervals of one polarity, like the Cretaceous quiet zone, might be related to superplume activity in the mantle (which might manifest itself eventually through increased crustal plate tectonics and/or volcanic activity). This explanation seems to work okay in some instances like the Cretaceous case, but then we are left with a timing problem: magnetic polarity changes happen and are completed in the few-thousand-year time frame, whereas mantle convection operates on the millions-of-years-plus time frame.
Outer core convection, on the other hand, seems to operate within the short time frame needed to change field polarity. So, is there any sort of “communication” between outer core, mantle and magnetic field? Does there really need to be, or are there just some interesting coincidences between intervals of tectonic activity and magnetic field behavior?
I don’t think the geologic record is going to help much on this one. Computer models may be the best chance we have at understanding what’s going on.
I’ve found some interesting links in case anyone would like to do some exploring:
The Geomagnetic Field – FAQ
Gary Glatzmaier’s home page and related links
I bit my lip almost hard enough to make it bleed to keep myself from posting that. I’ll have to go pull that out of my tape library. Barbara Eden was such an absolute babe. Still is really.
Fillet,
Very interesting and informative post, thank you. 
Wish you had come to speak at one of the geology classes I took in college. Unfortunately for us, the proff we had could not keep from putting people to sleep, he read straight from the book.
Fillet,
Very interesting and informative post, thank you.
Wish you had come to speak at one of the geology classes I took in college. Unfortunately for us, the proff we had could not keep from putting people to sleep, he read straight from the book.
fillet, I agree, very informative (although I will only say it one time).
sorry for the double post. I’m apparently fat-fingered today.
Grazie, folks - glad to have been helpful in this case.
aenea, I’m sorry to hear you had such an uninspired prof.; they seem to be much too common. Which is a real pity, because you don’t have to get a degree in geology to gain a new appreciation for the world around you.
Bravo fillet, a university geology prof who can’t make an introductory geology class fascinating should be stripped of his Brunton and have his geology pick ceremoniously broken. I am non-practicing now but I still never fail to slow down at road cuts (alas, all too few in North Texas) and see the scenery with a more practiced and observant eye because of my geo degree.
I made a mistake in the OP: according to my atlases, the North Magnetic Pole was on the west coast of the Boothia Peninsula in the Northwest Territory; now it is in the waters off King Christian Island–600 miles to the north, not 60.
fillet,
it’s posts like that keep me coming back to the Straight Dope (okay, the wit & wiseacres also attract). thanks.
and the stars o’erhead were dancing heel to toe