The Day Earth Almost Died

OlaOkay, so get this. 252 million years ago, Earth had this gorgeous stable ocean world, 50 million years of rugose corals, crinoids, brachiopods, and then, in geological terms, an absolute blink, it was just gone.

AmaraWelcome to Fault Lines, everyone. I'm here with Ola, and today we are going full deep time. We're talking about the Great Dying. The Permian-Triassic Extinction, the single worst thing that has ever happened to life on this planet.

OlaAnd I mean that literally. According to Encyclopedia MDPI, up to 96% of all marine species wiped out. 96%! Right?

AmaraAh, that number never gets less shocking.

OlaSo today, Amara, we are walking through the whole thing. thing what the late Permian world actually looked like, what broke it.

AmaraAnd the culprit is this absolutely unhinged volcanic event in Siberia, two million square kilometers of lava. We're talking 100,000 billion metric tons of CO2 pumped into the atmosphere per Frontiers in Earth Sciences.

OlaWait for it, because the kill chain after that, the ocean temperatures, the anoxia, the acidification, acidification is where Amara and I are going to have some very polite but very real disagreements.

AmaraVery polite. Totally.

OlaAnd then, get this, a January 2026 study in NPJ Biodiversity flips the whole recovery narrative on its head, not the slow, stepwise comeback the models predicted.

AmaraFast but broken, top heavy, predators bouncing back before the prey base could even support them. support them.

OlaWhich raises the question we're going to sit with all episode: was any of this actually recovery or just restructuring into something new?

AmaraYeah, that's the one. Okay, let's set the scene. Late Permian. What did this world look like before everything fell apart?

OlaOkay, so picture this: two hundred and fifty two million years ago, no Atlantic Ocean, no Pacific-one continent, Pangea, wrapping around a single global ocean called Panthalassa, and the seafloor, Amara-the seafloor is alive!

AmaraOkay; alive how; like give me specifics.

OlaRugose corals building reefs, brachiopods, little filter feeding shells carpeting the bottom in the millions. crinoids waving around like underwater flowers, and trilobites still hanging on after 270 million years.

Amara270 million years at that point?

Ola270 million years!

AmaraWow!

OlaThey'd survived every prior extinction, every single one! Overachievers, right? And the ocean supporting all this is, by comparison, pretty stable. Wikipedia cites pre-extinction global average surface temperature Its temperatures around eighteen degrees Celsius (tropical seas somewhere between twenty two and twenty five) deep water still well oxygenated.

AmaraSo not some hellscape just a functioning world.

OlaFor fifty million years, that's the thing; the Permian marine world wasn't a brief flash, it was fifty million years of stability-reefs built on reefs, food webs stacked and tested.

AmaraHmm, and according to Nature's January piece The Nicholls et al study? This was the most ecologically complex marine system in the entire Phanerozoic up to that point.

OlaUp to ninety per cent of marine animal species-gone-that's the number nature puts on it: not fifty, not seventy-ninety.

AmaraNinety per cent!

OlaAnd the Permian biosphere had no reason to expect it; CO2 was relatively stable, oxygen in the deep ocean was normal, the system had stress tested itself across millions of years. of years!

AmaraSo what actually broke a world that stable-because something had to hit it hard and hit it fast?

OlaYeah, and here's the thing: the answer is geological; it came from underground. The question is just what exactly did that look like when it started? So the culprit is the Siberian Traps, and the scale of it, Amara, is just-OK, two million years, two million square kilometers of Siberia buried under basaltic lava!

AmaraThat's the size of Saudi Arabia, just smothered!

OlaAnd according to MIT research and the wider geological record, the estimated CO2 injection is around one hundred thousand billion metric tons. Into the atmosphere total

AmaraA hundred thousand billion? I need a minute with that number.

OlaTake your time-but here's the thing that actually keeps me up at night: the eruptions weren't one long steady pour. Paleomagnetic data from Norilsk shows discrete pulses-short, violent bursts-each one probably lasting under ten thousand years.

AmaraSo not a slow bleed.

OlaNo-hammer blows-and between pulses enough recovery time to love the Love a system-then another hit. Hmm,

Amaraokay, but I want to push back a little, because here's where I think the eruptions alone don't tell the full story.

OlaGo on.

AmaraWell, the Siberian Traps didn't just sit on open rock-the magma burned through massive carbon-rich sediments, coals, organic deposits. The Wikipedia entry on the extinction event actually flags this when the eruption style shifted to sill dominated emplacement: and emplacement, forcing magma sideways through sedimentary layers, the thermogenic carbon and CO2 release spiked. That's when the main extinction pulse hits.

OlaUh-huh. And the isotopic record backs that up. Science Advances published work showing the thermogenic carbon phase came first, then the direct volcanic CO2 ramped up. Two phases, two punches.

AmaraSo

OlaRight.

Amarathe volcano was kind of the trigger, but the sediments are the

Speaker 3THE ACCELERANT

OlaThat's a decent analogy, though I'd say the Frontiers research puts it bluntly: the CO2 emission rate during the Permian extinction was roughly five times faster than the Capitanian extinction event just eight million years earlier—five times.

AmaraAnd that one already wiped out a significant chunk of marine

Speaker 3Wow!

Amaralife.

OlaSo we already had a rough baseline for what a bad

Speaker 4year looks like.

OlaBad volcanic episode looks like-and the Siberian Traps just blew past it.

AmaraBy a factor of five-unprecedented in the prior two hundred and fifty million years at the rate it happened.

OlaAnd all of that CO2 had to go somewhere-into the air, into the water-which is exactly where we're headed, because what one hundred thousand billion metric tons of CO2 does to an ocean,

Speaker 4it makes it even more acidic.

Olais a different kind of horror story entirely. So all that CO2 hits the ocean; what happens next?

AmaraHere's the sequence, and it's brutal: first, the warming-tropical sea surface temperatures spiked to somewhere around thirty five to forty degrees Celsius. That's not warm-that's a cooking pot.

OlaGrimacing Animals that had spent fifty million years in stable oxygenated water had zero time to adapt.

AmaraZero; and then the oxygen collapsed; warm water holds less dissolved oxygen anyway; but on top of that, deep ocean circulation basically shut down; the seafloor went euxinic.

OlaEuxinic, meaning-?

AmaraJumping in, rich in hydrogen sulphide, the kind of water that kills things on contact. Tacked." According to Nature's January study, the cascade hit warming, ocean anoxia and acidification in rapid succession, and each one compounded the others.

Speaker 3Mm-hmm.

OlaOkay, so here's where I want to push back a little. Everyone talks about acidification as the main marine killer, but ScienceDirect published a study on this and they note the exact timing of ocean acidification's onset is still contested; the geochemical proxies don't agree.

AmaraThere! I'd actually side with Anoxia as the bigger culprit; PNAS data showed roughly a hundred fold increase in seafloor Anoxia right at the extinction horizon, and it persisted for something like five million years after.

OlaA hundred times!

AmaraA hundred times! But acidifications still hit the calcifiers hard: the shell builders' boron isotope records show ocean pH dropped sharply. during the second extinction pulse, and that's when heavily calcified organisms disappeared.

OlaSo it's not one weapon.

AmaraIt's a firing squad-anoxia, heat, acid, then H2S venting from euxinic water into the atmosphere, and all of this within what researchers have resolved to roughly a fifty-five thousand year window.

OlaFifty-five thousand years!

AmaraWow.

OlaIn geological terms that's-I mean a blink.

AmaraLess than a blink: the last trilobites, gone. More than ninety-six percent of marine species, gone. Seventy percent of terrestrial vertebrates, gone.

OlaQuietly.

AmaraBut here's the thing, Ola; what came back and what it looked like-that's where it gets deeply strange. Because the early Triassic wasn't recovery in any normal sense; the forests that would have pulled CO2 back down-they collapsed too.

OlaSo the planets stayed hot.

AmaraStayed hot for millions of years.

OlaSo, after everything we described, the world just sat there, broken, for five million years.

AmaraFive million years of super greenhouse conditions-that's not a recovery period; that's a different planet.

OlaPicture what the seafloor looked like: no reefs, no coral, no swarms of brachiopods. Bacterial mats, burrowing worms-that's mostly it.

AmaraGrimly, the menu of life got down to the stress-tolerant. And the desperate.

OlaAnd on land, honestly, it might be worse: the coal swamps which had been everywhere in the Permian-gone, completely absent from the fossil record through the early Triassic. Surprised?

AmaraNo coal at all?

OlaNone: Encyclopedia MDPI actually flags this directly: no coal deposits are known from the early Triassic; the geological record just has this gap where peat and swamp forests used to be.

AmaraWhich is where it gets really unsettling, because Nature Communications published a study on this in late 2025, and they found that the forest collapse was not just a symptom of the warming, it made things worse.

OlaThe feedback loop.

AmaraYeah, tropical forests are normally a carbon sink, they pull CO2 down, but when those forests died, the sink disappeared, so the CO2 that should have been drawn back Back down, just stayed up.

OlaLocking the planet into heat it couldn't escape.

AmaraNature Communications calls it a runaway feedback in Earth's climate carbon system. The volcanism lit the match, but the forest collapse is what kept the fire burning for millions of years after that.

OlaSo you removed the carbon sink at exactly the moment you need it most.

AmaraExactly, and the ocean reflects that: the surviving fauna was reduced to a handful of hot

Speaker 4springs.

Amaraof hardy lineages-things that could handle hot, low oxygen water-the rich Permian food web we described at the start of this episode? Basically gone.

OlaFive million years of that world, hot, thin, quiet.

AmaraAnd here's what's strange: according to the January 2026 NPJ Biodiversity Study, life actually didn't wait patiently through all of that.

OlaMeaning what exactly?

AmaraMeaning-the recovery data is not what any one expected; multiple trophic levels showing up fast, but in a form the researchers describe as "top heavy and unstable.

OlaWith a short laugh, "Fast, but broken.

AmaraThat's the question for the next part; what counts as recovery when the ecosystem coming back looks nothing like what was there before?

OlaSo here's where it gets genuinely weird: the Standard Model said the recovery should be slow and stepwise: primary producers first, then herbivores, then eventually, ten million years out, predators reclaimed the Middle Triassic.

AmaraRight-rebuild from the bottom of the food web up-that's just ecology one oh one.

OlaAnd then a January, twenty twenty six study in npj Biodiversity Curiosity by Nichols, Wignall, Strong, and colleagues looks at the global fossil record and says "no." Multiple trophic levels bounced back fast.

AmaraWhich sounds like great news.

OlaUntil you read the next sentence:

AmaraOkay, what's the catch?

OlaThe catch is, they came back top heavy, predators recovering before a stable prey base even existed, according to the paper, "the empirical data

Speaker 4suggest

OlaData shows these early Triassic ecosystems were, quote, "top heavy and unstable.

AmaraSo you've got sharks circling a reef that's basically just bacterial mats.

OlaEssentially, think about what that means ecologically: predators need more energy input than the food web can actually sustain. The system kept collapsing under its own weight.

AmaraRecovery that kept collapsing!

OlaExactly, and there is a twenty to

Speaker 4thirty percent chance that such a thing could happen again.

OlaThe nineteen twenty five Science Advances study that makes this even stranger: surviving species expanded specifically into hot, low oxygen habitats, which explains something puzzling in the fossil record.

AmaraWhat's that?

OlaEarly Triassic marine communities look weirdly similar to each other worldwide, across latitudes. Science Advances argues that's not recovery diversity. It stress homogenation; the only things alive everywhere are the things that could tolerate the worst conditions.

AmaraSo the globe spanning similarity is a symptom of how bad things still were, not how well life was doing.

OlaRight; same stressed out generalists everywhere you look; that's not a healthy ecosystem, that's a global waiting room.

AmaraI love that-a global waiting room. Okay, but here's where I'd push back a bit, Ola. Isn't fast recovery at multiple levels still better than the step wise model predicted, like at least something was out there?

OlaSure, life showed up, but showing up and being stable are two completely different things, and that gap is exactly where the next question lives, because if the ecosystems kept crashing back down, were they actually recovering at all?

AmaraOr were they just life in a hurry to fill empty space?

OlaThat's the question, and the fossil record doesn't give a clean answer. So, here's the thing that's been sitting with me this whole episode: five million years to get species counts back, fifty million years before the food web was structurally stable.

AmaraThose are not the same number.

OlaNot even close. And according to the NPJ Biodiversity Study, that gap between species are back and the ecosystem actually works is

Speaker 4more than four hundred million years.

OlaIs the clearest example of that split in the entire fossil record.

AmaraWhich means we've been measuring the wrong thing, counting heads and calling it a comeback.

OlaRight; and the Science Advances data from the University of Leeds spells it out.

AmaraMm-hmm.

OlaTop down rebuilding was still under way fifty million years after the extinction. Fifty!

AmaraFifty million years is longer than the entire span of primate evolution. That's not a recovery window, that's a geological epoch.

OlaYeah! So, when the twenty twenty six data says recovery was faster than models predicted, I keep asking, faster by which measure?

AmaraTaxonomic, sure. Ecological, that's a completely different question.

OlaAnd here is where I can't land on an answer, Amara. AMARA-The twenty twenty six MDPI Biodiversity Paper asks exactly this: was it recovery or was it restructuring a new baseline that just looks different from what was there before?

AmaraLife filling empty space isn't the same as life healing.

OlaNo, it really isn't.

AmaraAnd that question doesn't just apply to two hundred and fifty two million years ago; if an ecosystem reef refills with species but runs on completely different structural logic, did the old one recover or did it just stop?

OlaStop, and something else began.

AmaraWhich means we might not even have the right vocabulary. Recovery implies you get back what you lost.

OlaAnd the fossil record is saying you

AmaraWow.

Oladon't-you get something, something that works, maybe; something alive, but not the same something.

AmaraSo was any of this recovery?

OlaThat's the question I think we've been building toward all episode. and I genuinely don't know the answer.

AmaraNeither do I, and I think that's the honest place to leave it.

Speaker 3Okay, so that was a lot to sit with.

AmaraRight? We started with trilobites that survived everything for two hundred and seventy million years and then just... didn't.

Speaker 3And the thing that stuck with me most? Counting species back is not the same as rebuilding a functioning ecosystem. Those are two completely different clocks.

AmaraThe taxonomy clock versus the ecology clock, and that gap might be fifty million years or more. That's the one I keep turning over.

Speaker 3Yeah, yeah. Recovery or restructuring, honestly, I'm still not sure the word recovery even fits here.

AmaraWe'll call it a cliff hanger.

Speaker 3A two hundred and fifty two million year old cliffhanger.

AmaraIf this episode cracked something open for you, subscribe and leave us a review. It genuinely helps. Got a theory you think we got wrong? Email us at hello at hey mato dot com.

Speaker 3Thanks for being here, Amara. And thanks to everyone listening.

AmaraSee you at the next Fault Line.