Two separate papers in this week's issue of the journal Nature provide complementary theories on how an internal dynamo in the lunar interior was stirred up – either by giant impacts or by tidal interactions between the Earth and Moon – to produce the Moon's surface magnetic anomalies.

How the Moon got its magnetized rocks has long been a mystery since Apollo astronauts walked the lunar surface – its internal core is too small to support the same type of dynamo that powers the Earth's magnetic field, a process that involves heat from our planet's inner core driving complex fluid motions in the molten iron of the outer core.

"The whole lunar surface possesses magnetic anomalies, but with large variations in amplitude it is difficult to tell the whole story of the way the surface acquired its present magnetic anomalies," explains Michael Le Bars of Aix-Marseille University and lead author of one of the papers. "In order to create a large magnetic anomaly, you need to make a dynamo in its core, and in order to record it, you must melt the crust, at least locally. What we see today is the superimposition of lots of events, with the locally recorded magnetic field coming from the last time the crust was locally melted, provided that it has not been perturbed by a subsequent event."

Le Bars and colleagues favour an impact model for the generation of intermittent, localised magnetism. In this scenario, impact-induced changes to the Moon's rotation rate lead to a knock-on effect of distortions in the Moon's interior that temporarily power the dynamo.

"We show in our paper that an impact driven dynamo would persist for about 10,000 years after each large [basin-forming] impact, which would allow the upper one kilometre of the crust in the impact basin to acquire a magnetization," he says.

Support for this model comes from the observation that six large impact basins older than four billion years in age possess magnetic anomalies, and from computer simulations that show that high volumes of melted rock are generated in such basin-forming events, which could acquire their observed magnetism as they cool in the presence of a magnetic field.

Meanwhile, Christina Dwyer of the University of California, Santa Cruz and colleagues suggest that the proximity of the Moon to the Earth early in its history provided tidal interactions that set up differential motion across the Moon's core-mantle boundary sufficient to stir up the core to power a permanent magnetic field for at least one billion years.

"The stirring itself happens at the outer boundary of the liquid core and then turbulence causes the energy of that motion to be transmitted through the entirety of the liquid core," she explains. "The maximum Earth-Moon distance that we predict a dynamo is 48 Earth radii [the Moon formed at a distance of around 3 Earth radii, moved rapidly out to 20-30 Earth radii and is now at approximately 60 Earth radii], at which point there is no longer enough power to generate a dynamo."

Dwyer adds that the impact-induced stirring Le Bars et al discuss could combine with the pre-existing stirring her team discusses to make something which is smoothly decaying as the Moon moves away from the Earth, but interrupted with occasional spikes of higher magnetic field during the basin-forming era.

"Both models now pose a great challenge to numerical modellers of dynamos in determining the properties of mechanically driven dynamos," adds Le Bars. "Future works may thus help to discriminate between the two models."