# The strength of the Earth’s magnetic field is falling

One of the most fascinating questions in geomagnetism concerns the strength of the Earth’s magnetic field. It is well-known (by those who know it well) that the direction of the field varies with time. You can see an estimate of how it is changing locally by looking at an OS map. These give the magnetic variation and the rate at which it changes each year, which is important if you are navigating with a magnetic compass. The strength of the field also varies from place to place and changes over time too, but you won’t notice this because it is pretty tricky to actually measure it. So tricky in fact that it wasn’t until the 1830s that the great mathematician and physicist C.F. Gauss, came up with a method for measuring the field strength. He did this by a crafty arrangement of magnets , suspended by torsion threads. In recognition of this great achievement, the units of magnetic flux density are called Gauss. (Actually, these days we tend to use the SI equivalent , the Tesla, named after another giant of electrodynamics. No need to feel too sorry for Gauss – he still has the ubiquitous Gaussian distribution, a celebrated theorem in vector calculus and a whole host of other mathematical machinery named after him).

The really interesting thing is this: ever since Gauss first measured the strength of the field, it has been dropping. Over the 170 years or so since the first measurements, the main field has decayed by some 8 or 9% and this has led to all sorts of speculation as to whether we are heading for some big geomagnetic event, like a reversal. It is true that this is quite a large drop- if the geodynamo suddenly stopped the field would only decay quite slowly- about 1% per century- but it is also likely that this sort of variation is not uncommon. The big question is how long was this going on for before people were first able to measure it?

To try and estimate this we have to use things like archaeomagnetism. In my last post I talked about how it was possible to use the direction of the “fossil magnetism” in heated archaeological structures to estimate the date at which they were fired. Well, it is also possible to estimate the strength of the field at the time they were fired, although the method is fraught with difficulties. People often use sherds of pottery of known date for this sort of work. Quite a lot of data has been amassed over the years, with over 300 “archaeointensities” spanning the period 1600-1840. Most people think that there is little evidence to suggest that there was any great drop in the field strength over this period.

I’ve always found this a bit unlikely. It just seems to be too much of a coincidence that the long term trend in the behaviour of the field changed at just the time we started measuring it. I’ve been looking at all the available data again, and I think there is evidence of a trend showing that the present decay has been going on since before 1700. The results of my analysis have just been published in Earth and Planetary Science Letters. So, I would argue that it seems likely that the Earth’s magnetic field has dropped by about 15%  since 1700, but I think the arguments will go on for some time yet.

One resource that might help to resolve this issue with more certainty is the experiments carried out by Alexander von Humboldt, Major Edward Sabine and others. These people made a great many observations of the  strength of the field, over the period from 1790s to 1830s, by timing the oscillations of a magnetised needle. Unfortunately this only gives us relative field strength: they were able to show that the field is stronger at high latitudes than it is at the equator, but they were unable to put absolute values on their measurements. It was an incredible scientific endeavour, with such names as Captain Fitzroy of the Beagle, the botanist David Douglas and the Arctic explorer Captain James Ross all taking sets of magnetised needles all over the globe. Sabine’s own account of this project, published in 1838 is is available online.

Neil Suttie is a geophysicist at the University of Liverpool where he mainly researches the past behaviour of the Earth's magnetic field. In addition he is interested in archaeomagnetic dating, instrumental development and shallow geophysical prospection.
This entry was posted in Uncategorized. Bookmark the permalink.

### 5 Responses to The strength of the Earth’s magnetic field is falling

1. Andy says:

Interesting. In one of Feynman’s lectures, I think its in ‘The Character of Physical Law’ or something, he discusses Millikan’s oil drop experiment to measure the rest mass of the electron. Millikan got the value wrong. It was under by about 1%, which is well outside his margin of error. But what happened with the data after that, when subsequent measurements were made, is not that we get a scattering of values around what is now accepted as the correct value, but that the published results gradually approach the true value. The next published value after Millikan was a little bit higher, the next a bit higher than that….
Now, of course, no one thinks that the rest mass of the electron has gone up by 1% since Millikan’s experiment. But that is what the data alone suggests. In fact, the change is an artefact of scientific practice. If a result came up that was 1.2% above Millikan’s, the scientist would check her equipment and have another go. Then she might get a result which was 0.2% higher and publish that.
I’m not really suggesting that this is the explanation of the trend in observed values for the strength of the magnetic field. I don’t know enough about it, and on the face of it it seems like the change in the magnetic field compared to the margin of error is huge compared to the Millikan case. But I just wondered if it is something that a practising scientist considers.
Cheers
Andy

• NeilSuttie says:

An important difference is this: the electronic charge (which is what Millikan measured) is , we think, a constant shared by all electrons everywhere. The “strength of the magnetic field” , of which I wrote, is not a simple thing to define, because it changes from place to place in a rather complicated way. So when we talk about the strength of the field in 1860, say, we are not referring to a single measurement, but a great many determinations of both field strength and direction taken in observatories around the globe. When we compare the data from different places , but taken at the same time, the discrepancies with our moels are far smaller than the observed trend, so I don’t think increasing experimental precision could be a factor. Interesting point though.

2. Andy says:

doh. It was charge of course, not rest mass.

3. AndyB says:

The wider point of people chucking out data that disagree strongly with previous (respected) measurements is definitely a concern. Some recent work has produced technically very high quality measurements that suggest the field in the Middle Eastern region was outrageously high (2-3 times stronger than present) about 3000 years ago. A group I am working with in Turkey are also about to publish similarly high results from the same sort of time and place. However, neither of us are the first groups to be working in this general area. Did the previous workers just miss these apparent “spikes” or did they summarily reject the data because they seemed “too high” to be real? I favour the former explanation but the latter is not impossible if the measurements were anyway of low technical quality.