Geology & Geophysics News

Paleomagnetic Data Hint at Link from Earth’s Core to Continents

Earth’s magnetic field waxes and wanes as supercontinents form and break up, suggests a new study postulating a direct connection between our planet’s crust and its core.


The last time Earth had a supercontinent was about 200 million years ago. All the landmasses that form the present continents sat together like pieces of a puzzle in a conglomerate called Pangaea. Pangaea was preceded by a number of other supercontinents that seem to have formed and broken up again about once every 600 million years.

Pangea over time.
Animation of the breakup of Pangaea from a single supercontinent in the Triassic (208–245 million years ago) to today’s modern individual continents in their present positions. Credit: U.S. Geological Survey

Now researchers from Australia and California report that reconstructions of Earth’s magnetic field strength over time show a correlation with this cycle, all the way back to roughly 3.6 billion years ago. This suggests to them that the regular assembling of supercontinents influences the heat flow from the planet’s core to its surface.

Uwe Kirscher, a geophysicist from Curtin University in Perth, Australia, and colleagues from the University of Adelaide in South Australia and the California Institute of Technology in Pasadena analyzed the Absolute Paleointensity Database (PINT), a large collection of paleomagnetic measurements from across the globe, to look for regular changes in the strength of Earth’s magnetic field over hundreds of millions of years. PINT contains about 4000 determinations of the strength of Earth’s magnetic field at different times in the past, as deduced from the magnetic properties of rocks.

Andy Biggin, a professor of paleomagnetism at the University of Liverpool in the United Kingdom who was not involved with the study, assembled the database using measurements from the literature. These measurements are of varying precision, Kirscher told Eos during the 2017 General Assembly of the European Geosciences Union in Vienna, Austria, at which he gave a talk about the new work.

Searching for Cycles

Normally, researchers try to select the best paleomagnetic data and discern any interesting patterns. Kirscher’s approach, which he concedes is somewhat controversial, is to use all available data and subject them to a standard mathematical treatment called a fast Fourier transform, which looks for cycles in the data.

“We took computer code that was used to study ice ages and their relationship with changes in the Earth’s orbit and applied it to magnetic data,” Kirscher said. “We didn’t invent anything, but to our knowledge, nobody in the past had done this.”

According to Kirscher, the agreement between the 600-million-year cycle he and his colleagues found and the periodicity of supercontinent formation can be understood if the supercontinents influence the amount of heat that escapes from the core to the cooler mantle above it, traveling through the crust and eventually into space. In the core, heat is lost by convection when hot molten iron rises to the boundary between the core and the mantle and releases heat into the mantle. The cooled iron then sinks again. The movement of the iron, an electrical conductor, gives rise to Earth’s magnetic field, so stronger convection will give a stronger field.

In the mantle, too, convection is instrumental in transporting heat upward, and it also influences the movement of the continents. In particular, many geologists suspect that a supercontinent in some way gives rise to a “superplume” of hot material from deep within the mantle. This superplume in turn leads to the supercontinent’s breakup, perpetuating a cyclical rise and fall of supercontinents.

Superplumes and How They Work

There are different proposals for how this would work, Kirscher says. For instance, the superplume could be triggered by the descent of cooler material from the margins of the supercontinent. “Once you have a supercontinent, the oceans are still spreading and the ocean floor material has to go somewhere,” Kirscher said. “So you get a subduction girdle around the continent, which forms a pile of cold material under it.”

The cold material sinks to the core-mantle boundary, where it can’t descend farther, absorbs heat from the core, and rises powerfully again as a superplume that pushes apart the blocks of crust that make up the supercontinent. Meanwhile, the temporary increase in the loss of heat from the core increases the convection in the core and with it the intensity of the magnetic field.

“So the magnetic field goes up when the superplume is there; then the supercontinent is destroyed, and the superplume goes away again,” says Kirscher. Even if this mechanism for superplume formation turns out not to be what actually happens, he says that alternative hypotheses involve changes in the heat flow from the core as well.

Sketchy Data?

Kirscher’s hypothesis faces some deep skepticism. Aleksey Smirnov, for instance, a specialist in the long-term evolution of the Earth’s magnetic field at Michigan Technological University in Houghton, wrote in an email that he doubts Kirscher has enough data to back up his proposal. “The current database on the Earth’s magnetic field strength is very limited, with about 90 percent of the data representing the past 250 million years. So no matter how sophisticated their statistical analysis is, there is no way one can identify a ‘strong 600 million year cycle’ in the field strength.”

Kirscher concedes that most of the paleomagnetic data are indeed from relatively recent times. However, he contends that enough data points exist to discern a cycle with a 600-million-year period (in addition to a number of other cycles and a lot of random noise) with 95% confidence that the result is not due to chance.

To further bolster his claim to have found the supercontinent cycle in paleomagnetic data, Kirscher plans to give the same treatment to another database of paleomagnetic data at the Borok Geophysical Observatory in Russia and see whether a 600-million-year cycle is hidden in there as well.

—Bas den Hond (email: [email protected]), Freelance Science Journalist

Citation: den Hond, B. (2017), Paleomagnetic data hint at link from Earth’s core to continents, Eos, 98, Published on 15 June 2017.
© 2017. The authors. CC BY-NC-ND 3.0