Core conundrum: how old is Earth's magnetic field?


Without its iron core's geodynamo, Earth's magnetic field wouldn't exist. But exactly how long it's been ticking away – and how much heat it's pumping into the mantle – is far from a simple calculation. Belinda Smith reports.


Convection currents in the liquid iron outer core (yellow) transport heat to the mantle (orange) and produce the planet's magnetic field. But how long has it been this way?
ANDRZEJ WOJCICKI / GETTY IMAGES
Earth's magnetic field is our shield against radiation flung by the Sun. Without it, life on the planet simply wouldn't exist as we know it. But exactly how long it's existed has geoscientists split.

Two studies published today recreated the crushing pressures and tremendous temperatures in the centre of the planet but came to different conclusions: one suggests the magnetic field's been around for as little as 700 million years – about the same age as complex life – while the other calculated the magnetic field has existed for three billion years.

Both papers were published in Nature.

Earth's iron core – a solid lump in the centre surrounded by a liquid outer layer – is the magnetic field's driver.

Heat thrown off as the inner core crystallises forces the molten outer core to rise until it butts against the thick overlying mantle layer.

There, the iron cools, dropping back to the inner core again to reheat. These "geodynamo" convection currents generate the planet-wide magnetic field.

Because we can't drop a camera to the centre of the Earth to see what's going on down there, scientists model and simulate the action of the outer and inner core. But a big problem is we don't really know exactly how iron acts when it's heated more than 4,000 °C at a million times the pressure at sea level.

So two teams – one led by Kenji Ohta from the Tokyo Institute of Technology in Japan and another headed by Zuzana Konôpková at DESY Photon Science in Germany – attempted to measure what does happen at the boundary between outer core and mantle, but in the lab.

Both used laser-heated diamond anvil cells, where a piece of iron foil smaller than a pinhead was placed between two diamonds and squashed to epic pressures and heated via lasers.

Despite the disparity, he hailed both studies as 'experimental feats'.

Ohta's team used wires to measure the hot crushed iron's electrical conductivity. Electrical conductivity is related to thermal conductivity, and the more heat conductive a substance is, the faster it can cool.

Their experiments yielded a thermal conductivity of 90 watts per metre per kelvin. At that rate of heat conduction, they calculated the inner core is no more than 700 million years old.

Konôpková and colleagues more directly measured the iron's thermal conduction by seeing how long it took for a second pulse of heat to travel through the sample.The Conversation

This "thermal rate of diffusion" is also closely related to thermal conductivity – but they calculated a figure of 18-44 watts per metre per kelvin.

At this substantially slower heat transfer rate, the inner core was been present for three-quarters of Earth's age, or around three billion years.

So why the 2.3-billion-year discrepancy?

University College London geophysicist David Dobson, in a News and Views article, suggests the extra blasts of heat in the Konôpková study may have imperceptibly melted part of the iron sample. This melting acted as a thermal buffer, slowing the transfer of heat.

On the other hand, Ohta and colleagues may have misjudged the amount heat drawn away from the sample down the electrode wires.

There may even be new physics at play, given the extreme pressures and temperatures. But despite the disparity, he hailed both studies as "experimental feats".

The papers can agree on one thing: future work should also examine the role of impurities in iron. The core also contains traces of elements such as nickel, oxygen and hydrogen, all of which could affect its thermal conductivity.

  1. http://nature.com/articles/doi:10.1038/nature17957
  2. http://nature.com/articles/doi:10.1038/nature18009
  3. http://www.nature.com/nature/journal/v534/n7605/full/534045a.html
Latest Stories
MoreMore Articles