Sixty-six million years ago, an estuary in what is now the US state of North Dakota was shaken by a violent earthquake – so strong it caused multi-metre-high walls of water to slosh up and downstream, stirring up massive amounts of sediment. Simultaneously, BB-sized (4–5 millimetre) glass spherules were raining from the sky. Within minutes, uncounted numbers of fish were buried alive.
For the fish it was a tragedy. But for palaeontologists, it’s a goldmine, because the earthquake and the glassy rain were caused by an asteroid strike 3,500 kilometres away in what is now Mexico’s Yucatan Peninsula: the asteroid strike that killed the dinosaurs.
And because the North Dakota fish all died within minutes of the strike, scientists have made an amazing discovery: the asteroid hit in the northern hemisphere spring.
The discovery began in 2017, when a young vertebrate palaeontologist named Melanie During, now at Uppsala University, Sweden, got a chance to visit the site.
There, she was amazed to find the fossil fish lying in layers, one layer pointing downstream as they fought a surge coming upriver, and the next pointing upstream as they fought a return surge. Some had even been hurled around so hard that their bodies were wrapped around tree branches. “It looks like the worst car crash you have ever seen, frozen in place,” she says.
During dug up six fossil fishes and took them back to Europe for examination.
One thing evident from the start was that the spherules had clogged the gills of the fish, but had not been ingested into their digestive systems.
That, During says, means the fish died before they’d had time to accidentally swallow any of the spherules, meaning that they died very soon after the deadly rainfall began. Since that would have been within 15 to 30 minutes of the impact (about the same time it would have taken for the earthquake shockwave to have travelled 3,500 km from Mexico), it puts their time of death to very soon after the impact. “It has to be within the hour,” During says.
Furthermore, it was very clear that it was the dino-killing impact that had created the catastrophe: draped right over the top of the deposit was an iridium-enriched layer that is known to be one of the signatures of that event. “So, we are very sure about the time constraint of this deposit,” says During’s colleague, Jeroen van der Lubbe, a palaeontologist at Vrije Universiteit Amsterdam, the Netherlands.
Some of that had been known before. But in a paper in today’s issue of Nature, During’s team carried it further by examining fin bones of these fish, looking for growth cycles similar to the growth rings of trees.
The fish, which were akin to modern paddlefish and sturgeon, are known to have grown bones in annual cycles: slowly in the winter and more quickly in the summer.
One way to detect these cycles is by taking tiny slices of the preserved bone material and looking at the ratio of carbon and oxygen isotopes in its minerals, which also vary with the availability of the fish’s primary food supply, plankton. These ratios, van der Lubbe says, show that peak plankton availability hadn’t yet happened when the fish died – meaning it was spring, not full summer.
But it’s also possible to look at the bones on a cell-by-cell level by using high-energy radiation from a particle accelerator to do something akin to 3D microscopy. Not that the cells are still there, says Dennis Voeten, the vertebrate palaeontologist at Uppsala who led this work, “[but] we can see the holes in which bone cells were once located.” Big cells mean fast growth (as in summer), small cells mean slow growth (as in winter).
Because the fish collected by During were about seven years old, the team was able to trace seven years of annual cycles, finding that the fish died just as they were starting to move into the next year’s faster growth cycle. Not late winter, and not full summer, but somewhere in the middle. “My guess is April,” During says. “Definitely not summer.”
It’s not just a cool discovery but an important one, because ever since the 1980 explosive eruption of America’s Mount St. Helens, ecologists have realised that when it comes to the effects of a cataclysm, timing can be everything. In the case of Mount St. Helens, it was in May, when patches of mountain snow were still on the ground, protecting the underlying plants from incineration.
In the case of the dino-killing impact (which also exterminated about three-quarters of all other animal species alive at the time), the timing may have played a major role in why the northern hemisphere appears to have been harder hit than the southern hemisphere.
That’s because spring would have been a particularly vulnerable time for northern hemisphere animals, just emerging from winter. Surviving animals’ food chains would also have been particularly vulnerable at that time of year, because the impact blew massive amounts of dust and smoke into the air, blocking sunlight and causing temperatures to plummet, severely affecting plant growth.
“I think we all know how certain plants can be extremely sensitive to sudden night frosts when just starting up,” During says. “They are so much more vulnerable [then] than [in autumn] when everything is shutting down.”
In the southern hemisphere it would have been the reverse. Not only would many plants have already been preparing for winter, but many animals would also have been doing so. Some might even have been hibernating, or preparing for it.
Not that survival would have been easy in any case, which is probably why the impact took out all of the dinosaurs. “But in order to survive [the climate change], you need to be able to survive the impact,” During says. “So, anybody in the southern hemisphere already sheltering had a much better chance.”