Nature
PUBLISHED
1
Earth’s magnetic poles have flipped many times since the planet was formed, so that half the time what we think of as the magnetic north pole was actually south, and vice versa. These flips are not evenly spaced through the geologic record, occurring at a rate of around once every 100,000 years in the late Jurassic, but about half as often recently. They were even rarer for much of the time in between. However, new analysis indicates we have probably missed some flips, particularly in eras when flips have been thought to be rare.
The rest of this article is behind a paywall. Please sign in or subscribe to access the full content.We have a very incomplete understanding of why the magnetic poles reverse, which makes it very hard to predict when future flips will occur. Although claims that past geomagnetic reversals have been associated with an influx of radiation and mass extinctions appear to be low on evidence, our technological society might not cope so well with being caught in one. Consequently, it would be comforting to be able to predict the next flip ahead of time, but it’s not clear we even know when all the past ones have happened; you’d expect it to be a much easier task.
Researchers have used techniques from statistics known as adaptive bandwidth kernel density estimation (AKDE) to seek examples of polar flips that haven’t previously been identified, and therefore left out of timelines of the Earth.
We know about ancient geomagnetic reversals because when iron-rich rocks cool out of lava, they align with the local magnetic field. Around mid-ocean ridges, the rocks form “zebra stripes”, their magnetic fields pointing alternately north and south depending on when they were formed. Sometimes, however, reading the magnetic patterns is not so easy, particularly from eras when little oceanic crust survives.
Three years ago, a team led by Dr Yutaka Yoshimura of Kyushu University identified flips in basalts produced by Ethiopian volcanoes around 30 million years ago that have been missed in measurements taken elsewhere on Earth. If we’ve overlooked reversals in relatively recent rocks, by geological standards, it’s easy to believe there are other ones, particularly earlier, that we don’t know about.
There are two reasons why geologists are keen to identify past reversals. Firstly, they are a useful – often the only – way to date important events, whether they be movements of tectonic plates, or the deposition of significant fossils. Secondly, it is hoped that tracking their frequency could offer crucial clues to explain why these flips happen at all.
Attempts to make a timeline of geomagnetic reversals have identified a period lasting around 37 million years where reversals apparently stopped, known as the Cretaceous Normal Superchron (CNS). If any dinosaurs navigated by magnetic fields, they apparently had a long time without having to learn to go the opposite way.
In contrast, within the period we have been able to measure, reversals were most frequent around 155 million years ago, with the spaces between growing longer until 121 million years ago, when the CNS began. After the CNS finished, the reversals did not suddenly resume their Jurassic frequency; instead, they initially appeared quite seldom, and gradually increased in tempo.
Using the AKDE, however, Yoshimura and co-authors have now identified four periods following the CNS at intervals of around 14 million years where reversals appear to be rare. Rather than simply accepting that polar flips stopped during these periods, the authors think there may have been reversals we have overlooked. If they include these suspected flips, a graph of reversal frequency becomes much smoother.
Besides possibly helping with dating rock formations, a timeline with the extra flips also makes reversals more consistent with phenomena with underlying physical causes, rather than random noise. That, in turn, makes the idea of finding such a cause look much more achievable.
Although the basis for reversals remains a mystery, Yoshimura and co-authors endorse a 40-year-old theory that it is influenced by the rate of flow of heat across the boundary between the Earth’s core and its mantle.
The findings are still tentative – the authors identified times when they think additional flips are likely to have occurred, rather than finding hard evidence of them. However, the work gives scope for future research to try to find those flips.
The study is published in Geophysical Research Letters.
