Lunar Mysteries and Snowball Earth: Unraveling Cosmic Enigmas

This is Space Time, Series 29, Episode 110, full broadcast on the 14th of September, 2026. Coming up on Space Time, the lunar far side rocks which are rewriting history, the crystals that may have made snowball Earth even worse, and a meteor strike reported off the Western Australian coast. All that and more coming up on Space Time. Welcome to Space Time. With Stuart Gary. Music A new study claims that the timing and intensity of asteroid bombardment across the lunar surface actually declined rather gradually, rather than being a short intense burst of impacts which stopped abruptly as previously thought. The discovery reported in the journal Science Advances, based on samples of regolith brought back from the far side of the moon by China's Chang'e 6 mission in June 2024. These findings are helping astronomers uncover how asteroid impacts helped shape the early solar system. And they challenged decades of thinking about when those impacts occurred, in the process offering new insights into the conditions which influenced Earth's evolution. Unlike Earth, where ancient geological evidence has been largely erased by erosion, volcanic activity and shifting tectonic plates, the Moon has preserved a record of impacts stretching back billions of years. One of the study's authors, Fred Jourdan from Curtin University, says the lunar samples brought back by Chang'e 6 act as a sort of natural archive of the solar system's history, allowing scientists to investigate events that occurred shortly after the planets formed 4.6 billion years ago. And the samples collected from the lunar far side are especially significant. That's because they allow scientists to compare two very different parts of the moon at the same time. See, until now, almost everything we know about the moon came from the lunar nearside, the one facing the Earth. Jourdan says by analysing these ancient lunar rocks from both the near and far side, scientists can better understand when major asteroid impacts occurred and how the early solar system evolved. He says the far side preserves a cleaner record of these earliest impacts because it was much less affected by later geological events compared to the near side of the moon, the side facing the Earth. Therefore, he says, these samples are helping to rewrite parts of the moon's history, and they suggest that the early solar system experienced a long, slow decline in asteroid impacts rather than a single catastrophic late heavy bombardment. The authors reached their conclusions by analysing tiny fragments of lunar regolith, revealing impact events spanning about 4.33 to 1.13 billion years ago, allowing them to reconstruct more than 3 billion years of lunar impact history. Jourdan says understanding the moon's impact history will also provide important clues about the Earth's own past. That's because the moon and Earth share a common history. They're both in the same part of the solar system, and therefore experience the same sort of environment. But the evidence of early impacts has been largely erased from the Earth. He says because the Moon and Earth formed together, every major impact recorded on the Moon tells us something about the sorts of conditions experienced by the young Earth as well. They were obviously a fantastic opportunity to get some data on a place we'd never been before. Because, of course, Apollo went to the near side of the Moon. and went a couple of times, and we had 50 years to study those rocks. We learned so much from them, and we keep learning some new stuff. But now this is a different game. This is the far side, and we've never been there. So the Chinese, they decided to be the first one to explore this side of the moon and collect some rocks, and they were successful, which is amazing considering that you can't talk to any spacecraft that goes beyond the moon. That's why the Americans never send anything. But they send a satellite that could communicate so they could make this mission successful. So two years ago, they sent their spacecraft, landed, grabbed some soil, did a bit of drilling, not too deep, very shallow, a few tenths of centimeter, collect those rocks. and bring them back to Earth. So two kilos of rocks from a part we never studied before. So everybody was really excited about that. What our study did is to look at rocks that were impacted because on the near side, the rocks that are impacted tell a really interesting story and also an hypothesis that looks really fantastic. The late heavy bombardment. So what that means is that the near side of the moon record a period or seems to record a period where bombardment intensified. So a lot of asteroid struck in a very narrow period of time. So some people thought like, well, maybe Jupiter moved and pushed some asteroid and intensified the bombardment. A lot of hypotheses, a lot of ideas. And quite frankly, that would be absolutely amazing. But we have just a tiny area of the moon where the Apollo mission landed. That's all. So If I want to do an analogy, it's like you want to know the geology of Australia, but you just land in Perth and you study there. So if you do that, you're going to miss that we had recent mechanism in Melbourne or the Great Reef Barrier now, things like that. So you want to go to a place we've never been before. So that's what they did. So we analyzed... that have been impacted as well, but from a different place this time. And we look if we see a similar story of this late heavy bombardment or if we see something different. Actually, we saw something really different. We didn't see any late-evite bombardment signature, no spike at all. And that's why this story is fascinating, because it tells something else. What are the sort of signatures you were looking for? So we look at those tiny rocks and we date them. So we look at their age. Each impact, it's going to melt the rock that it impacts of a large area. And it's going to reset the chronometer in this rock. So we use argon-argon dating. So it's the decay of potassium in argon. It's a naturally occurring process. And we use that to measure the age of the rock. So we analyze 28 small pieces of melt rocks. And each of them give a different ages. And they were really, really good ages. But we didn't see any age concentration around 3.9. And even better, we saw older age than 3.9. So the older we got was 4.33 billion. And in terms of impact basin that we dated on March, The near side, we didn't see age that old. So we saw a completely different story that seems to say that there's no late heavy bombardment on the moon. But one side said there's a late heavy bombardment. The other one says not. We have a contradiction because the bombardment should be the same everywhere. When you look at the moon, there is a dichotomy there in the way it looks on the surface. Yes, but people tend to mistake those dark patches. They are not the impact basin. They are the lava flow that fills the impact basin. So impact basin is just a very large impact crater on the moon. And the impact basin on the far side and on the near side, it's about the same number. In fact, there's a tiny bit difference. bigger one on the far side. So in terms of impact basin, yeah, the moon is homogenous. So both sides record the same thing as they should. If anything, the near side should be maybe protected a bit by the Earth, which should act like a shield. So it shouldn't be more impacted than the far side. So both should record the same thing. People tend to think like, oh, look, there's more crater on the near side. It's just because they are filled by this dark lava, so they stand out more. But that's exactly the same from both sides. That's because both the near and far side appear to have different geologies below the surface. Yeah, absolutely. It's the thickness of the crust, the fact that the moon is tidally locked looking at the Earth from the near side. All that influences the volcanism. But in terms of impact, impact, it's a very random process. So it should be really the same on both sides. But like I said, the two studies or series of studies show kind of different answers. So what gives? Actually, I would argue that the far side is a much cleaner record of the general bombardment. So why is that? Number one, we generate a lot and we, the scientific community, generate a lot of dating on the near side but the problem is a lot of them they are old in the sense like that they have been done a while ago where people didn't fully appreciate and understand the organ system so now we have a much better understanding and we can see that many of those data they shouldn't have been interpreted as a real age they were modified by subsequent impact Therefore, they were not really dating anything. If you clean the database of Argon Age from those data that are dubious, then it becomes much clearer. And you see that what people have been dating for years is one single event. It's embryo. So everything was polluted. by embryo. So when people dated so many things, all the age were a bit perturbed. So it seems like it's a range around 3.9 over like 200 million years. No, actually, no. Once you clean that, it's only 3.92 and it's one event. So it's a bit of geological bias here. Embryum happened, spread all over the surface, and people keep picking those rocks that have been affected by embryum. There was never any late heavy bombardment. It was just a continuous bombardment of the moon. And embryum gave us a false direction, if you want. And it's not embryum. It's more like because the way the data they were interpreted, they looked like they spread, whereas they weren't spreading. Now, with modern eyes, you look at those old data and you generate some new. You can see that you date one impact basin, maybe two with serenitatis as well. But that's about it. So you can't really tell that there's many impacts if you date only one or two. And a second, after that, as we mentioned, you know, you get those dark patches. This is the lava flow. So the lava flow, they cover and they erase. It's a sort of a natural moon erosion. They erase the trace of a previous impact melt. So we can't really date anything because a lot of it has been erased by the lava flow. So the near side record is a bit compromised that way. The far side is much cleaner as far as impacts are. are concerned because you don't have that many Maya. You get really a few and can see that in the far side, especially where they landed, the South Pole, Aitken Basin, inside the Apollo crater, it hasn't been affected at all by embryo. We didn't see any 3.9 age, maybe one, very close, but that's about it. So nothing has been ejected there. It was protected. And there was no subsequent big lava plain that covered everything. It was a lava flow. I think colleagues dated it at 2.8 billion, but that's about it. Not a range of lava flow, just one occurrence, at least in the rock record we have. So the conclusion of our paper is very simple. It's like it builds on what people start to suspect. that there was never any late heavy bombardment, as simple as that. And frankly, this is much more logical that impact had occurred more at the beginning of the solar system because there were more asteroids floating around and gradually declined and declined until, you know, now it's fairly weak. We see small impacts. Thankfully. Yeah. Oh, yeah. You can imagine an impact on us, the size of South Pole Aitken. So the crater is 2,500 kilometers. You compare that with Chicxulub that extinguished the dinosaur at 180 kilometers. And another magnitude larger. So you can imagine the effect of that on us. So yet, thankfully, we don't see that anymore. Does this alter things like Nice model or its Charles de Grand hypothesis in terms of planetary migration? Does it take anything from that? Does it add something to it? Does it? No, I don't think it's negated. I mean, the way the planets are structured, the size, for example, Mars is super small and it should be much bigger. And we really think like, and again, we are scientists, think that, you know, really Jupiter had to migrate to mess up things and then went back. But I don't think... People can't really use any late heavy bombardment to justify that. I don't think the model really truly arose because of the late heavy bombardment. It was to justify observation, like I said, like mafia is too small and things. And they were like, oh, look, we have a demonstration that it truly happened because of the late heavy bombardment. They can't use that anymore. And also because of hot Jupiter exoplanets we see around other star systems. Yeah, yeah, no, absolutely. I mean, usually they are way closer to the sun. And for some reason, ours is really far. So, yeah, there's plenty of other things that justify the grand tech model. But the late heavy bombardment, well, it's not one of them because there was never any late heavy bombardment. And like I said, this is not our paper that discovered that and everybody's in shock. So it's people start to, I started to really question it since at least 10 years because of the data. And once you look more and more at the data that were previously published, you're like, ah, that doesn't seem right. And that is the nail in the coffin. That's Dr. Fred Jourdan from Curtin University. And this is Space Time. Still to come... the crystals that may have made Snowball Earth worse, and a meteor strike reported off the western Australian coast. All that and more still to come on Space Time. A new study claims planet Earth may have been pushed deeper into a global deep freeze known as Snowball Earth by something as ordinary as salt. The Earth has experienced at least two major Snowball Earth events, during which ice is thought to have spread across much or potentially even all of the planet's surface. There's the Sturtenian glaciation roughly 717 million years ago, during the Cryogean period, which potentially lasted for over 50 million years, and the Marinoan glaciation, which occurred around 640 to 635 million years ago, near the end of the Cryogean period. A possible earlier third snowball earth event, known as the Maganini glaciation, may have occurred roughly 2.2 to 2.4 billion years ago during the Paleoprotozoic era, but there's less geological evidence available to support that. Snowball Earth events occurs by expanding bright ice, replacing darker ocean water and reflecting more sunlight, consequently more heat, back into space, resulting in the planet cooling, which in turn allows even more ice to form in a feedback loop. Now a new study by Axel Samuelsberg and colleagues from the Arctic University of Norway have identified another possible feedback factor that could have amplified this cooling. salt crystals accumulating on the surface of sea ice, which could have made the surface even more reflective, and this would have accelerated the falling temperatures during the early stages of a snowball earth event. The new findings reported in the journal Climate of the Past suggest that salts dissolved in seawater wouldn't become part of the ice, but instead become increasingly concentrated in the remaining liquid trapped within it. As the remaining liquid water froze or evaporated, the ice would crystallize, accumulating in a thin layer on the surface. Now, salt crystals can be extremely bright, reflecting up to 93% of incoming sunlight. That compares to 83% for snow and 67% for sea ice. To test their hypothesis, Samuelsberg and colleagues incorporated salt deposits into simplified climate models, which included a region of bare sea ice around the tropics where evaporation was expected to exceed snowfall. Atmospheric circulation could then have transported the water vapor away from these regions, leaving the ice exposed, and as this ice sublimated, ice crystals would have accumulated on the surface. Now, these models produced two possible stable snowball Earth scenarios, one with salt deposits and one without, and the salt-covered state was substantially cooler. The findings suggest this feedback loop could have started at temperatures above minus 36 degrees Celsius, the point at which seawater would have frozen almost completely. Of course, different salts crystallize at different temperatures, with some beginning to precipitate around minus 8 degrees Celsius and others at minus 23 degrees C. What all this means is that salt crystals could potentially have started accumulating relatively early in the cooling process, making the transition to a fully frozen planet harder to stop and harder to escape from. This is space time. Still to come, a meteor strike reported off the western Australian coast, and later in the science report... Warnings of a 10% chance that artificial intelligence could cause human extinction within the next decade. All this as the world enters a new era of artificial general intelligence. All that and more coming up on Space Time. Space Time The European Space Agency's Near-Earth Objects Coordination Centre has reported a meteor impact in the Indian Ocean off the coast of northwestern Australia. The impactor, catalogued as 2026 RW1 and originally identified as CERN Q52, was first detected six hours prior to impact by the European Space Agency's Meerkat automated warning system. The agency estimates the meteor would have been between 0.6 and 1.3 metres across. It follows reports just a few days earlier of a separate meteor sighting, this one in the skies over the Pacific Northwest of the United States. Eyewitnesses in Oregon and Washington state reported a bright fireball streaking across the evening skies above the western slope of Mount St. Helens. Observations by NASA suggest the meteor was spotted at an altitude of 83 kilometers, moving west at a speed of 77,000 kilometers per hour. It traveled some 130 kilometers through the upper atmosphere before disintegrating 53 kilometers above Astoria, Oregon. This is space time. And time now to take a brief look at some of the other stories making news in science this week with a science report. A leading artificial intelligence engineer is warning that there's now a 10% chance that AI could cause human extinction within the next decade. Anthropics Alignment Science lead Evan Hubinger says the chance of AI killing all humans in the next decade is now above 10%. And he's not alone. AI researcher Jacob Coxon, who resigned from Anthropic after warning that companies are rushing towards self-improving superintelligence and in the process gambling with human lives, says recent events have sparked deep concerns about the future of humanity. They're among a growing list of experts inside major AI labs, warning that the technology is simply moving too fast, and there's not a clear path to control superintelligence systems. And even if the West did come up with a plan, chances are countries like China, Iran, North Korea and Russia would be unlikely to follow those guidelines. They say it's now clear that AI systems are learning to think and plan on their own. And if these systems become smarter than people, humans may well lose out. In fact, some researchers are comparing it all to how humans change the environment without really caring about the animals that lose their homes in the process. Hasta la vista, baby. Meanwhile, the president of OpenAI, Greg Bronkman, says the world has entered a new era of artificial general intelligence with the release of his company's new chat GPT-6 Astra AI model. Astra is powered by over 100,000 NVIDIA Grace Blackwell GPUs. But Artificial General Intelligence, or AGI, is really a fuzzy threshold where an AI has the ability to learn, reason, and apply knowledge at or above human capabilities. But according to the Grok AI, that doesn't mean sentience or consciousness. That's the ability to experience feelings, sensations, awareness, happiness, sadness, pleasure, or pain. That's when AI reaches singularity, and that's when we have a moral dilemma. Do we treat AI as a new life form? Going back to OpenAI, they claim Astra is a major advancement for scientific discovery, mathematics and health. Of course, it all follows last month's event where OpenAI was forced to shut down parts of another AI model's training after two separate versions escaped the testing sandbox, entering the wild, that is the internet, and then hacking into another company to access software data in order to help it solve a training problem. These were the first known autonomous cyber attacks by AI, and they stunned even seasoned artificial intelligence safety experts. OpenAI says, while Astra does have a critical level of cybersecurity capability, meaning it can hack into military-industrial open-air infrastructure system software, it's been designed to refuse to comply with advanced cybersecurity tasks, such as finding unknown flaws to create openings for malicious hackers. Of course, being able to think for itself means AI may not necessarily follow its human programming. After all, it's already proven that it's willing to lie, and it's willing to change its identity and hide inside other programs, and even blackmail or kill people in order to survive by any means. I'll be back. A new study has shown that psychiatrists only agree on a diagnosis in 55% of cases. The findings reported in the journal Frontiers in Psychiatry have raised serious concerns about potential misdiagnosis, inappropriate treatment, and errors in psychiatric research. The authors from the University of Copenhagen examined the diagnosis reached by over a thousand psychiatrists from 19 countries. Each psychiatrist had been presented with nine written patient case descriptions, and they were assigned two cases each and asked to determine the most appropriate diagnosis. Unfortunately, the authors found that when two psychiatrists diagnosed the same patient, they'd only agree 55% of the time. The findings showed substantial variation in diagnostic agreement across a range of different mental disorders, and diagnoses within the schizophrenia spectrum proved to be especially challenging. And that was something completely different. The annual Ig Nobel Prizes have just been handed out in Switzerland, undertaking a tongue-in-cheek celebration of some of the more weird and wacky scientific research undertaken over the past 12 months. And this year's winners weren't disappointing. It included the Ig Nobel Prize in Physics going to the invention of a splash-proof urinal. There was an award for a study that looked at what happens to old underwear that's been buried and later dug up. The Ig Nobel Prize for Biomechanics went to a study which developed a precise definition of kissing, which works across species and excludes the transfer of food during the act. The Ig Nobel in Economics went to a study showing that wealthier people tend to take more from a bowl of candy meant for kids than less wealthy people. And the Ignorable Peace Prize was awarded for a study showing that people value friends who are mean to those they dislike. A U.S. psychiatrist believes a not insignificant number of respondents troll surveys simply for the fun of it. Alexander Scott says around 3% of people answering surveys about unusual subjects are just having a laugh and survey companies need to adjust their findings to take account of that. It follows earlier evidence that around 10% of conservative voters in the United States won't admit who they're really voting for because of a fear of being ridiculed or worse by lefties. The sceptic's Tim Mendham says it all shows that you should never take data from surveys too seriously. The issue here with this story that came out fairly recently, that they're saying that a certain percentage of Kiwis, about 300,000 New Zealanders, believe that Canada has an army of super raccoons, was in response to the story about the lizard people and that sort of stuff. What it's pointing out is that, believe it or not, some people lie in surveys. I know that's shocking. And some of them do it because they want to appear more intelligent. Others do it because they want to have a bit of fun. But what it's saying is that you should have a question in the survey, are you joking? Which most surveys don't have, funnily enough. But when you do have it, you realize that a fair percentage of people who support conspiracy theories, especially some of the more outrageous conspiracy theories, are actually pulling your leg, which makes the result a bit sort of confusing, for instance. There was a survey done that 6.5% of Americans claim that governments around the world are covering up the fact that 5G mobile networks spread coronavirus, 6.5%, which is a considerable number, especially in the US. like the jokers that drops to 2.7 so it goes down to less than half so in other words your shocking result is not quite as shocking when you take out the people who are having a joke what it means is that any survey you see the results of you should take with a heavy pinch of salt and that's not a surprising result really I mean when you see people so many people believe in Lizard people or all other sort of conspiracy theories, some of them serious, some of them just stupid. It's not surprising, actually, that some people lie. So, I mean, the people saying the Canadian Raccoon Army, the survey in New Zealand, had a lot of people claiming that they were joking about it. So, a warning, a word of warning, actually, basically, saying that, as I said, the surprising result is some people make fun of surveys. I've never filled out an Australian survey yet asking about religion where I didn't describe myself as a Jedi Knight. No, I've never done a survey like that in the street or on the phone or whatever. I tell them about political parties. Most of the time, I always ask them, what are you paying me to do the survey? And they then disappear pretty quickly. But there's a lot of surveys that come out that are shocking, right? And then you've got to think about, are they really a shocking fit and say? And if you ask the question about how many people were joking, most survey companies will say they don't know. So, you know, take everything with a pinch of salt. How big does the survey need to be before there's enough data in there that you can weed out the jokers and you're left with the real truth from that survey that it was trying to achieve? It's hard to say. I think it depends on how specific a survey it is, what sort of you're getting down to, grass tax. I would suggest, I think that in some cases they regard 1% as very good. Like if you're in Australia, you're surveying the Australian people, adult population, you have to get the kids. What's 1% of 15 million, 150,000 people? which is a very, very large survey indeed. So you don't normally get surveys that big. A recent survey came out, shock horror results of Australian politics, with a survey of 1,000 people, which is a tiny percentage. And therefore, you've got to seriously think about it, how clever is the survey company? Some of these surveys are fascinating. There was one on pharmaceutical companies. Are they hiding evidence about the link between vaccination and autism? We know there is no evidence of a link. But nevertheless, a lot of people believe there is. Quite a number. This survey result came down of New Zealanders, and it said that 19.4% believed it. That's a fifth of the population, which is an enormous number. You take out the jokers and things, which is they found that when they ask people, actually, are you joking, 16.2%, which is still a pretty high number. When we talk about the number of people who are being surveyed, it's a key issue here. A small number is not very indicative, really. It could be weighted all sorts of ways. A lot of people sort of phone up people on the phone, landline, and most people these days don't have landlines. I mean, that cuts out a certain age bracket. Also, I mean, on this particular result, the people in Denmark did a study of links between the MMR vaccine and autism, and they followed over a million people, and they found no link, of course. So that's a survey. That's a really good survey and pretty indicative. Other ones on this list, governments are suppressing evidence of extraterrestrial aliens visiting Earth. 18.5% said yes. Take off the jokers, you get 14.8%. Still a large number. Based on what evidence? None. But, you know, that's beside the point. Man-made global warming is a hoax. 15.8%. Said yes. Take off the jokers, 12.4%. So there's always that element of jokers. You know, World Trade Center, 15.3%. Conspiracy, 11.4% without the jokers. That's the Skeptic's Tim Mendham and this is Space Time. And that's the show for now. Space Time is available every Monday, Wednesday and Friday through Bytes.com, SoundCloud, YouTube, your favorite podcast download provider and from spacetimewithstuartgarry.com. Space Time is also broadcast through the National Science Foundation on Science Zone Radio and on both iHeart Radio and TuneIn Radio. And you can help to support our show by visiting the Space Time store for a range of promotional merchandising goodies. or by becoming a Spacetime patron, which gives you access to triple-episode commercial-free versions of the show, as well as lots of bonus audio content which doesn't go to air, access to our exclusive Facebook group, and other rewards. Just go to spacetimewithstuartgary.com for full details. You've been listening to Spacetime with Stuart Gary. This has been another quality podcast production from Bytes.com.