A huge crater in Western Australia was created by an asteroid strike 3 billion years ago, according to a mineral-dating technique. This would make it the oldest impact crater on Earth – but other researchers have questioned its proposed age.

The North Pole Dome crater, also known as the Miralga impact structure, was first described by at Curtin University in Perth, Australia, and his colleagues in 2025. They estimated that it could be up to 100 kilometres wide.

Kirkland and his team found a layer of rock containing cone-shaped features called shatter cones, which form only after a high-impact event, such as an asteroid strike. Their original study didn’t directly date this rock, but based on correlations with dated rocks in the layers above and below, they proposed that the impact .

This would make it more than 1.2 billion years older than the Yarrabubba crater in the south of the state, which is regarded as the oldest reliably dated asteroid-strike crater on Earth.

It would also make it the only known impact from the Archaean aeon, a time when the entire planet was a giant but inhospitable water world.

However, another team, including , also at Curtin University, strongly contested the 3.47-billion-year date. Based on their own analysis of rocks in the area, the team members argue that the impact took place no earlier than .

Free newsletter

Sign up to The Earth Edition

Sign up to newsletter

Now, Kirkland and his colleagues say they have successfully dated recrystallised minerals at the crater site, which contain shatter cones. “We have now actually looked inside the rocks and tried to find minerals that directly responded to impact in the rock itself, rather than making correlations,” says Kirkland.

Using the rate of the decay of uranium into lead, the team dated zircons within the shatter cones, which recrystallised as a result of the force of the asteroid strike.

They also dated the mineral apatite, which would have formed in the hydrothermal system created by the heat of the impact.

Both the apatite and the zircons returned dates of around 3.02 billion years old, says Kirkland. “So now we’ve got evidence for very hot water percolating through the rocks 3 billion years ago and also evidence for this really unusual heating and recrystallisation process,” he says.

Kirkland says no other known process, such as mountain building or regional metamorphism, easily explains the mineral changes inside the shocked rocks, because there is no evidence that the area was heated or deformed by those processes at about 3.02 billion years ago.

“The only process really that we can link to these mineralogical changes is an impact,” he says. “So that means the best evidence now is a 3-billion-year-old impact, and that by far is the oldest impact crater on the planet.”

Cavosie welcomes the fact that the age of the crater has been revised significantly, but he thinks Kirkland’s team is still overestimating the crater’s age.

“While I’m relieved these authors have backed off their 2025 ‘3.5-billion-year impact’ hypothesis, I don’t think they’ve presented a compelling case for a [3.02-billion-year] impact either,” says Cavosie. “The slow march of science towards the truth thus continues.”

Cavosie says there are clearly shatter cones in younger rocks that are only 2.77 billion years old, which means the impact must have happened after this date.

at Yale University, who was also part of the group that critiqued the original study, agrees with Cavosie that the rocks must be younger than 2.77 billion years.

“While the new study dismisses this observation because these rocks ‘have not been dated’, they are straightforwardly correlated to nearby rocks that have been dated,” says Brenner.

Kirkland says the key difference is that his team has now directly dated minerals inside the shocked rocks. “The younger age argument still depends on long-distance correlation of undated rocks, largely from satellite-based mapping rather than direct geochemistry or geochronology,” he says. “We now have two mineral clocks from the impact rocks themselves giving the same age. That is why direct dating matters.”

The asteroid strike that wiped out the dinosaurs hit with such force that it took at least 8 million years for the impact site to cool down, creating a warm underground ecosystem where microscopic life thrived.

The Chicxulub asteroid, which collided with Earth 66 million years ago at what is now Mexico, is thought to have been as large as 15 kilometres in diameter. The strike caused so much climate chaos that it wiped out three-quarters of species on Earth. All the dinosaurs except the ancestors of birds became extinct and a nuclear winter gripped the planet for at least 15 years.

Its effects were also felt deep underground. “The Chicxulub impact was big enough to cause deformation at least 35 kilometres under the surface of the Earth, detectable using geophysical surveys,” says at the University of Glasgow, UK.

The impact melted about 10,000 cubic kilometres of rock, she says. The combination of melted rock and seawater created porous material filled with tiny pockets of hot water, known as a hydrothermal system.

Because of the presence of minerals that only form where there is liquid water and heat, we know that the asteroid would have created hydrothermal environments to depths of several kilometres. But the scale and lifespan of the heating and resulting hydrothermal system has, it seems, been massively underestimated.

Previously, it was thought it took only 2 million years for the impact site to cool down. Now, Pickersgill and her colleagues say it may have taken at least four times longer, giving hydrothermal life much more time to thrive.

Free newsletter

Sign up to The Earth Edition

Sign up to newsletter

“One of the biggest unknowns about all impact-generated hydrothermal systems, and Chicxulub in particular, is how long the heat keeps water circulating through the structure,” says Pickersgill.

To figure this out, the team drilled 1 kilometre into the crater to obtain rock cores. Because potassium in the rocks has decayed into argon gas over time, the researchers could measure the amount of argon trapped in the samples to find out their age.

“We got a range of ages from the time of impact at 66 million years ago to about 58 million years ago,” says Pickersgill. “That told us that hydrothermal activity was ongoing in at least part of the Chicxulub structure for 8 million years after the impact.”

Sulphur isotopes in the cores provide evidence that microbial life existed in the hydrothermal system and recovered rapidly after the impact.

The results mean that the very earliest impact craters on the young Earth – and perhaps other worlds – may also have had habitable hydrothermal systems for longer than previously known.

“This provides more opportunity for life to develop, evolve and spread,” says Pickersgill. “It supports the concept that early life on Earth may have found a long-term home in impact craters, and possibly even life on other planets where these massive impact craters are dominant surface features.”

at Curtin University in Perth, Australia, says while there is “not an entirely unambiguous record of continuous hydrothermal activity” at Chicxulub, the evidence is strong that the impact site stayed hot for millions of years.

“Large impacts do not simply destroy environments,” he says. “They can also create long-lived underground systems where hot fluids circulate through shattered rock. These chemically rich settings may provide sheltered habitats for microbes and perhaps even favourable conditions for some of the early chemical steps towards life.”

While some have argued that northern India and western Nepal are overdue for a massive earthquake, an analysis says this is a myth, as the area has been experiencing smaller earthquakes at random for millennia.

It is common for officials and media to call populated areas near fault lines – like Istanbul, Seattle and Tokyo – “overdue” for violent earthquakes. Because the central Himalaya fault segment in India and Nepal last had a major recorded earthquake in 1505, some has suggested that earthquakes there recur about every 500 years, making a great earthquake now imminent.

But scientists have now found at least 50 earthquakes of magnitude 6.5 or larger that have occurred in this area in the past 6000 years, . And these earthquakes have been occurring randomly, not at regular intervals.

“We have to stop discussing and having long debates over the periodicity of earthquakes in the Himalayas and come to an agreement that it’s a random process … and consider the risk within that framework,” says at the British Antarctic Survey, who led the study.

The collision of the Indian and Eurasian tectonic plates that cast up the Himalayan mountains continues to this day, forming one of the largest seismic zones on the planet. The 2400-kilometre fault under the mountain range generates violent earthquakes, such as the magnitude 7.8 disaster that killed nearly 9000 people in and around Kathmandu in 2015.

However, less evidence of earthquakes has been found on the central segment of the fault immediately to the west of the Nepali capital, leading to fears that pressure was building up in this “seismic gap” and would soon be released in a devastating earthquake of magnitude 8 or 9.

Free newsletter

Sign up to The Earth Edition

Sign up to newsletter

Ghazoui-Schaus argues this was a misconception based on a “knowledge gap” rather than a seismic gap. Researchers have typically looked for evidence of earthquakes in the Himalayas by digging trenches to find ruptures in what was the ground surface in the past. While this method was able to uncover large earthquakes, it missed smaller “shadow earthquakes” that didn’t break the surface.

“You are only going to have a very sparse record of the largest earthquakes” with traditional paleo-seismology methods, says , a retired seismologist from the British Geological Survey. “Whereas for historical earthquakes, then the catalogue can be good down to about magnitude 4 or so.”

Because the record was primarily populated with large earthquakes, it led to calculations of a long “interevent interval,” also known as a “return period,” which is the average time between earthquakes of a certain magnitude in an area.

To uncover a better earthquake record in the central Himalaya, Ghazoui-Schaus and his colleagues trekked to Lake Rara in western Nepal in 2013 and took a 4-metre sediment core from the lakebed with an inflatable raft.

They later analysed the core for turbidites, layers of fine sediments on top of coarser ones, which were deposited on the lakebed by underwater landslides triggered by earthquakes. The team has now identified 50 earthquakes of magnitude 6.5 or greater over the past 6000 years, dating each according to its depth in the core. These have likely released energy and lessened tension in the fault, Ghazoui-Schaus says.

Statistical analysis found that the earthquakes tended to come in clusters, but these clusters occurred randomly. While that is what most seismologists would now expect based on the modern instrument record, Ghazoui-Schaus says it is one of the first times a paleo-seismological record has also confirmed it.

“If I have to build a house in western Nepal, I would definitely be more cautious in the way that I would build,” he says. And even though earthquakes occur at random, calculating the average interval between them can still be useful as an indicator of seismic activity that could damage structures in an area like bridges or dams, according to Musson.

“If you’re planning for the next 100 years, you want to know how many earthquakes of a certain size are going to occur in that period,” he says. “And if you are prepared for that, then it doesn’t matter whether the earthquake happens next year or in 10 years’ time, because you’ve built your dam strong enough.”