Most of us remember that the terrestrial magnetic field has something to do with swirling currents of conducting, iron-rich fluid in the Earth's core, but after that it gets pretty vague. It turns out we aren't alone in this: while computer models are inching closer to something that will generate a realistic field, they still aren't quite there yet.
When they do get there, though, it will make the magnetic field and its variations a powerful probe of the "interior geography" at the base of the Earth's mantle, where enormous structures influence the circulation in the outer core.
The curious thing about the Earth's field is that it exists at all, because when you think about it, swirling a conducting fluid doesn't do anything much. Put some salt water in a glass and whoosh is around and no sparks fly, compasses are not deflected. If you're like me, the only thing that really happens is the floor gets wet.
Why would a circulating, conducting fluid in the Earth's core be different? For one, scale matters a lot: phenomena that are undetectable at a scale of a few centimetres can become dominant over the Earths' radius of 6800 km. But there's more to the story.
The Earth has four major layers due to how temperature and pressure increase with depth. Low levels of radioactive material--mostly uranium and thorium--are present throughout the Earth, and the heat they give off raises the temperature to almost 6000 C at the centre (sorry, Jules Verne fans!). At the same time, pressure increases with depth, which increases the melting point of rock. Competition between these two effects results in the layering we observe, with solid crust, plastic mantle, liquid outer core, and solid inner core.
The crust is the cold outer layer, fifteen to twenty kilometres thick, which floats on top of the next layer down: the mantle. In the mantle, rock is hot enough to flow slowly, but isn't really a liquid. It has a lot of structure, much of which we are still learning about, from half-melted remnants of long-subducted continents to lobes that influence everything from volcanoes to the position of the magnetic poles.
Beneath the mantle lies the outer core, which is iron-rich, electrically conductive, liquid rock, and below that the pressure is high enough to create the solid inner core, despite the 6000 C temperature. The inner core is almost all iron, but it's too hot to hold much of a magnetic field: the Earth's field instead is due to the convective flow of the outer core, but how this "dynamo" works is not immediately obvious. As our experiment with swirling salt water shows, moving a conducting fluid does not automatically generate a magnetic field.
But... moving a conducting fluid in a magnetic field does generate an electric current. This is how electric generators work: we move a conducting loop in a magnetic field and the electric field that is induced causes a current to flow around the loop. The thing about electric currents, though, is that they cause magnetic fields, so if there is a magnetic field with a conducting fluid moving in it, there will be a secondary magnetic field created by the current generated in the conductor by the primary magnetic field... and it turns out there are self-sustaining geometries where the secondary magnetic field created by this process is capable of regenerating itself, at which point the field that got it all started no longer matters.
Without a continual influx of energy from the core's heat, the field would die away in about twenty thousand years, which gives a sense of how much power is going into this. When I swirl a glass of water and then stop, the swirling dies away in seconds, not hundreds of centuries.
In a sense, then, the magnetic field of the Earth is its own cause: once it exists, it's capable of sustaining itself so long as there's heat from the core to drive it. But where did it come from in the first place?
We're not totally sure, but the most likely source of the original magnetic field is the sun. Sun-like stars typically go through an early phase where they have very strong magnetic fields, many thousands of times greater than the modern value. This would be more than enough to get the geomagnetic dynamo going on Earth, and once going, it would sustain itself. So the Earth's magnetic field is quite probably the self-sustaining remnant of the primordial solar magnetic field.
Once it gets started the field strength is determined by the internal dynamics, not the primordial field. Computer models of those internal dynamics have yet to match the observed field using realistic parameters, though, so there there is more going on than we yet understand.
When we do finally figure it all out, there is a chance that we can use such modelling as a kind of "computational telescope" pointed at the Earth's core, especially the underside of the mantle. The magnetic field isn't a simple dipole, but has global variations (as well as local anomalies due to magnetic ore deposits), and it changes with time in both strength and direction. Irregularities can form in the convective currents, and these result in everything from the field's deviation from a perfect dipole, to the north pole’s recent wandering toward Siberia, to periodic short-term "excursions" of a few centuries, to longer-term reversals that last tens of thousands of years and are typically spaced by a few hundred thousand years or more.
Since these non-uniformities and fluctuations are in part a result of the structure of the underside of the mantle, a detailed understanding of the Earth's field may allow us to map that surface, because its shape will be an input into our computer models. That will allow us to generate a topography of hidden mountains and canyons that deflect the currents of stone, thousands of kilometres beneath our feet, that no human eye will ever see.
[Thanks for subscribing! This is the VERY FIRST “World of Wonders”, and next week we’re going to jump from beneath the Earth to beyond the sky, possibly via warp drive! I’m happy to take questions and any feedback on what people would like to see more (or less) of. As well as the comment section (if you’re a substack member) this e-mail address is set up so you can respond to it and I’ll get the message! —TJ]