Image: Jurik Peter/Shutterstock

Is it possible to find trace evidence of supernovae from millions of years ago in the sediment lining the ocean floor? One astrophysicist has spent the better part of a decade trying to find the proverbial smoking gun to prove that it is. And now, it seems, he has succeeded.


Shawn Bishop, an astrophysicist at the Technical University in Munich, Germany, has been investigating the fossils of ancient bacteria on the ocean floor for several years now, in hopes of finding traces of an iron isotope produced in a supernova explosion some 2.2 million years ago. He presented promising preliminary findings at a physics conference in 2013. He has now confirmed those early findings, according to a new paper in the Proceedings of the National Academy of Sciences (PNAS). “The signal is definitely there,” he told Gizmodo.

The isotope in question is iron-60 (60Fe), just one of the heavy elements produced by supernovae when they explode, scattering those elements into space. Because of 60Fe’s short half life, there shouldn’t be any of it on Earth, but traces have nonetheless been detected in the ferromanganese crust on the ocean floor. And that means there could be traces of the isotope elsewhere as well.

Image: Shawn Bishop

Bishop thought the best candidates were tiny microfossils of ancient bacteria that scientists have found in ocean floor core sediment samples. Back in 1963, Italian microbiologist Salvatore Bellini noticed these types of bacteria could orient themselves towards the North Pole—that is, they could sense magnetic fields to help them navigate their way to favorable low-oxygen environments.


In technical terminology, they are “magneto tactic,” thanks to the presence of chains of magnetite crystals inside the critters. The bacteria likely picked up the crystals from sediments along the sea bed. As I wrote at Scientific American in 2013:

Bishop thinks it’s possible that fine-grained debris from a supernova explosion could pass through Earth’s atmosphere, rapidly oxidizing in the process so that they are broken down into tiny nano-oxides.

These would rapidly dissolve in oxygen, form rust, and eventually settle in the sediment along the ocean floor, where the bacteria would suck them up for their crystal chains. When the bacteria eventually die, those chains remain behind in the sediment, and 60Fe would be locked inside. So any traces of 60Fe found in that sediment would constitute a kind of biogenic signature of a supernova event, preserved in the fossil record.

Detecting those traces is, needless to say, very tricky, but Bishop figured out a way to do so using accelerator mass spectrometry (AMS) to analyze the grains of bacteria contained inside core samples from the floor of the Pacific Ocean. And then he counted, one by one, each individual atom of 60Fe. In his preliminary results, he found strong peaks in the concentrations of 60Fe in one of the cores that could be linked to a supernova explosion some 2.2 million years ago. He even identified a couple of candidate stars in the Scorpio Centauri cluster.

It was pretty exciting stuff, inspiring Anton Wallner of The Australian National University to assemble an international team to conduct its own research, in hopes of finding even more traces of 60Fe. As described in a pair of papers published in Nature in April, they did indeed find evidence for a series of nearby supernovae explosions that lit up the sky several million years ago, raining radioactive particles down onto Earth. The nearest of the explosions probably occurred in an aging star cluster some 326 light years away.



Between that result and Bishop’s new confirmation of his preliminary 2013 findings, this constitutes some pretty strong evidence. And at least one of the events coincides with the onset of the Pleistocene, ushering in a period of major global cooling. Astronomers have long suspected that nearby supernovae can affect Earth’s climate, most notably by burning up the ozone layer. Maybe one such explosion triggered the Pleistocene.

“We don’t have any concrete evidence that any one event is tied to a supernovae,” astronomer Adrian Mellott of the University of Kansas told Gizmodo’s Maddie Stone back in April. “But the odds are, one or more are.”

[Proceedings of the National Academy of Sciences]