The origin of the Earth’s Magnetic Field

The current theory on the origin of the Earth’s magnetic field is that circulating equatorial currents in the core give rise to a magnetic field. This current keeps going for millennia, then suddenly stops, and on restarting may flow in the opposite direction. Virtually all papers suggest that the current is self-sustaining in some way, although there is no suggestion that the material of the Earth’s core is super-conducting. Further, there is no clear idea of why it stops, and why it sometimes reverses.

For reference, the inner part of the Earth’s core is believed to be solid and made up mainly of a nickel iron, and although this is at about 3,300 degrees Celsius, the tremendous pressure at the core keeps it solid. This inner core is surrounded by an outer core of molten iron, in turn surrounded by molten rock. The core loses heat at about 100 degrees Celsius every billion years. The inner core is 1220 km in radius, and the outer core is a layer 2260 km thick lying on top of it. The inner core appears to be spinning slightly faster than the spin of the rest of the Earth, although some say it is slower; most agree there is some difference in spin.

If the core’s magnetic field is caused by circulating electric currents it is difficult to see how they can continue for more than a few seconds, because the resistance of the iron will cause the currents to slow and stop very rapidly. This gives rise to the view that there must be some regenerative mechanism that keeps the currents circulating despite these resistive losses - that is, that the convection of the iron in a magnetic field causes currents to flow, leading to the magnetic field which - together with the convection - causes the current to flow. Such a mechanism has the flaw that resistive decay introduces a massive loss into the system. The supporting idea, that the rotation of the Earth somehow orders these currents, has the flaw that the material of which the core is made is in a highly disordered molecular state. I believe it is fair to say that at the moment the origin of the Earth’s magnetic field is something of an enigma.

I therefore want to suggest an alternative mechanism for the Earth’s magnetic field, so that just as a flowing electric current can sustain an electric field, so a flowing magnetic material can sustain a magnetic field. There are significant (but not insurmountable) issues to be dealt with on this approach, not least being that the iron is well above its Curie point, but I will try and tackle them at the end of the page.

The basic idea is that...

• A magnetic material can create a magnetic field by being in laminar flow, in the direction of that flow.
• The central core is a source of heat that drives convection currents in the liquid outer core.
• There is laminar flow in the liquid outer part of the core, from the solid central core to the surface of the outer core. This takes heat from the central core.
• The flow goes up in the north polar region, cooling down as the flow traverses the surface of the outer core, finally descending in the south polar region of the core, cooling the inner core as it reaches it.
• The current flows over the surface of the inner core back up to the north polar region of the core, picking up heat from the inner core, then rising again at the north pole.
• The descending current in the south pole carries extra spin with it that causes the inner core to rotate marginally faster than the rest of the Earth. This gives a slight extra spin to the rising current that helps stabilise the laminar flow up.

Note that I have assumed the flow is up at the north - it may in fact be the other way round, but I will assume this orientation for the purposes of this discussion.

The central core is hotter than the surrounding liquid. This gives rise to convection currents in the outer core which will attempt to rise. If they rose in equatorial regions they would need to pick up more and more angular speed as they rose through the molten iron, because of the rotation of the Earth - at greater radii they go faster. This means that it is thermodynamically easier for them to rise along the axes of rotation, in polar regions, where they simply rise without needing to pick up angular momentum.

The heat differential is very low and the mass of moving metal is immense, leading to low flow rates in the rising currents. This is conducive to laminar flow, further maintained by a slight extra rotation of the central core imparting a slight spin to the rising stream. The flow, once established as up on the north pole, will continue to flow in this direction.

At the top of the outer core there will be a slight “hummock” at the top of the flow, as it spreads out. It flows out from the north pole over the surface of the outer core, picking up its rotation as it does, and there will be strong mixing with the rest of the surface material as it flows across the equator and on down to the south pole. Here heat is lost to the molten rock that surrounds the outer core. As it travels south from the equator it will be shedding spin again, but if it has not shed all of it by the time it reaches the descending stream there will be a very shallow rotating vortex at the top of the descending flow.

The flow descends to the inner core, now cooler than it, and picks up heat from it as it flows around it on its way north. As it does so it sheds the excess angular momentum it picked up on the surface of the outer core, giving it a slight extra spin. As the flow reaches the north it will have become so hot that it will rise again at the north pole, and continue in the cooling cycle.

Rising magnetic material suffers a drop in pressure that changes its magnetic properties slightly, creating a North magnetic pole through the combination of motion and pressure drop. At the opposite pole of the Earth the molten iron falls towards the core, experiencing a pressure rise as it does. The combination of falling motion and pressure rise gives rise to a South magnetic pole. Hence the two magnetic poles are formed.

If that was all, the descending stream would cancel out the magnetic field of the ascending stream. However, there is a clear imbalance. The ascending stream is hotter at any given altitude from the centre of the Earth than the descending stream. The ascending stream therefore remains hotter up to lower pressure altitudes, while the descending stream remains cooler down to higher pressure altitudes. This imbalance could be the source of the Earth’s overall magnetic field.

That is the basic mechanism. But what is going on when the magnetic field decays and then even reverses?

The upward flow is inherently unstable, twitching backwards and forwards and causing the magnetic north pole to wander. Occasionally it twitches so much that flow cohesion is lost, and the laminar flow collapses into turbulent flow before simply falling apart so that instead of ordered flow, the liquid metal roils around the hot inner core, rising in myriad minor streams of short range. The outer core becomes turbulent throughout, and the temperature of the surface of the inner core rises. The Earth’s magnetic field loses its cohesion and disappears.

The myriad different rising flows compete with each other, and those near the equator lose out because they take more energy to drive. Hence flows at one or other pole will win out, leading to the re-establishment of a laminar flow which may rise randomly at either the north or the south pole, so may be associated with a reversal or a

So this is what it looks like, with the solid inner core in green, the outer molten core in blue, and the paths of the flow in red. The flow starts up going north, with the magnetic field generated by that flow, wiggles up through the molten outer core, then flows along the surface of the outer core to the south, cooling down and picking up spin as it flows. The cooled flow falls through the south pole, transferring some extra spin to the inner core, flows north over the surface of the inner core picking up heat again, and repeating the cycle...

So that is the whole mechanism. Now how about the fact that iron loses its magnetism at its Curie point at 1100 degrees Celsius?

The model described here requires a coherent laminar flow of iron above the Curie point to exhibit a (relatively tiny) magnetic field. First, bear in mind that the effect is very small, given the mass of iron at the centre of the Earth. Further to this, several recent papers now report magnetic effects above the Curie point. At least one of these papers suggests a relevance to the maintenance of the Earth’s magnetic field. Accordingly, given the weak nature of the field, the possibility of magnetic coherence above the Curie point, and the nature of fluid flow, it seems that this model is at least worthy of consideration.