Was the First Climate Mass Extinction Really Caused by Cooling?
A closer look at the Late Ordovician reveals a more complicated chain of events
Among all major mass extinctions in Earth’s past, for a long time, we used to think the Late Ordovician mass extinction was the simplest one to explain.
Among the “Big Five” extinctions, most are linked, in one way or another, to warming, oceans losing oxygen, and ecosystems collapsing under chemical stress. The Late Ordovician event, which happened around 445 million years ago, always seemed like the clean exception: a plunge into an ice age, a dramatic cooling of the planet, and a huge loss of marine life.
For years, that story felt tidy. Cooling came first. Ice sheets grew. Sea levels dropped. The habitat shrank. Life faded. End of mystery.
Then, we started paying attention to the details that don’t fit neatly into that sequence. Not because the classic explanation is wrong, but because the rock record is rarely that polite. The Earth system is perfectly capable of doing several disruptive things at once, and when it does, the “main cause” is often not a single switch that flips, but a chain of effects that reinforce each other.
That is why the Late Ordovician extinction is still one of the most interesting climate stories in deep time. It looks like an ice-age extinction. But the more you read, the more you wonder whether ice was the whole story.
The “cooling killed life” explanation is intuitive, and it is grounded in real geology.
In the Late Ordovician, most complex life lived in the oceans, and not in the deep sea the way we might picture modern marine life, but in shallow waters covering continental shelves. These broad, warm, sunlit seas were the biological hotspots of the planet. When conditions were stable, they supported rich communities of animals living on and in the seafloor.
Now imagine what happens when the planet moves toward an ice age.
As ice sheets grow on land, they lock up water that used to circulate through the ocean. Sea level drops, sometimes dramatically, because that water is no longer in the sea. Whole shallow seas disappear as coastlines retreat. If your species is tied to those shallow environments, you cannot simply “move deeper.” For many animals, the habitat vanishes faster than they can adapt.
That is why the Late Ordovician extinction has long been taught as the “cooling extinction.” It seems like a straightforward case of climate change working through geography: ice grows, sea level falls, habitat shrinks, life collapses.
But here’s the detail that complicates the picture.
The extinction didn’t happen in one smooth decline. It happened in two pulses: a first wave of losses, a pause, and then a second wave. If the key driver was simply cooling and sea-level fall, you would expect the biological crash to track the climate more continuously, not hit twice like a double punch.
So what caused the second punch?
That question is why scientists keep returning to the Late Ordovician event. It is not only a story about extinction. It is also a case study in how climate disruption cascades through living systems.
When paleontologists talk about “causes,” they are not looking for a dramatic headline like “Ice killed 85% of life.” They are trying to understand which environmental changes happened first, which ones were most stressful, and how those stresses interacted. Temperature is part of the answer, but it is rarely the whole answer because temperature changes do not act on animals directly in isolation. They act through sea level, ocean circulation, oxygen availability, and food supply.
To tease those apart, researchers rely on different kinds of evidence that each captures a different part of the system.
So, what did they actually look at? One major line of evidence comes from chemistry that acts like a temperature tracer.
Certain chemical signatures in fossils and sediments shift depending on temperature, especially forms of oxygen that become more or less common in the shells of marine organisms depending on the water they grew in. These measurements are not “thermometers” in the everyday sense, but they are reliable signals of change.
This is where classic studies, like Finnegan and colleagues (2011), are so important: they show that tropical seas were extremely warm for much of Late Ordovician time, and that a rapid cooling pulse occurred close to the extinction interval. In other words, yes, glaciation happened, and it likely happened fast enough to matter to ecosystems.
Another line of evidence comes from sea-level changes.
When sea level drops, the types of sediments laid down in coastal environments shift. Shorelines move. Shallow-water deposits retreat. You can track ancient regression and transgression like you track old flood lines in a landscape. Reviews such as Harper’s (2024) emphasize that the sea-level fall during the glacial maximum was very large, likely around 100 meters, which would have erased huge areas of shallow marine habitat worldwide.
So far, the cooling story holds up well: big climate change, big geographic shift, big habitat loss.
But the Late Ordovician record contains another recurring signal that cannot be ignored: Oxygen.
Oxygen loss in the ocean is one of the most common mechanisms behind marine extinctions. It doesn’t sound dramatic, but it is devastating. When oxygen drops below a certain level, many animals cannot survive, and entire communities can collapse.
We cannot measure ancient oxygen directly, but we can infer it using chemical fingerprints in sediments. Certain trace elements behave differently when seafloors are oxygen-poor, and when those signals appear repeatedly around extinction layers, they strongly suggest that oxygen stress was widespread.
In the Late Ordovician, multiple sections show evidence for low-oxygen conditions spreading into areas that should normally have been well oxygenated. That matters because it adds another stressor: not just losing habitat, but losing the ocean’s basic life-support capacity.
Cooling can contribute to oxygen instability by reorganizing ocean circulation, and changes in glaciation can disrupt mixing patterns. So oxygen loss does not contradict an ice-age scenario. But it does open the door to a more complex driver, especially when paired with the two-pulse extinction pattern.
In recent years, that door widened.
Bond and Grasby (2020) argued that the Late Ordovician Mass Extinction Event may not be the “odd one out” after all. They propose that volcanism and warming-related ocean stress may have helped drive both extinction pulses, with oxygen loss playing a central role. One of their key lines of evidence is mercury spikes in sediment layers that coincide with extinction intervals, since mercury can be released in large amounts during volcanic episodes and preserved in geological archives.
Mercury alone does not prove causation. But combined with oxygen and productivity signals, it suggests the Earth system was being disturbed from multiple directions. Volcanism matters because it can push warming by releasing carbon dioxide, and it can alter ocean nutrient dynamics in ways that increase biological productivity. That sounds helpful at first, but increased productivity often means more organic matter sinks and decays, consuming oxygen in the process and expanding low-oxygen zones.
So instead of a story where ice alone drives extinction, we may be looking at a stressed planet receiving repeated hits: cooling and sea-level fall reshaping habitats, while ocean chemistry becomes unstable and oxygen stress spreads across marine ecosystems.
So, what does this change?
This doesn’t mean the classic “cooling killed life” narrative is wrong. It means it is incomplete.
Cooling and glaciation likely mattered enormously because the sea-level drop removed habitat on a global scale. But habitat loss alone does not fully explain the two-pulse structure, nor does it capture the physiological stress implied by widespread oxygen disruption.
A better framing is not “cold versus warm,” but “instability versus stability.”
When climate shifts rapidly, the world changes in more ways than temperature. Sea levels move, circulation patterns reorganize, and oxygen distribution can fail. Those cascading changes can make ecosystems fragile, so that even groups that survive the first shock become vulnerable to the next one.
This is why we keep coming back to the Late Ordovician extinction. It reminds us that Earth history is rarely clean, and that science improves not by replacing stories with new ones, but by refining stories until they match the complexity of evidence.
The exciting part is not deciding whether ice or volcanism deserves the headline. The exciting part is learning how linked the Earth system is, and how one change can unlock several others.
Understanding this event does not give us a simple villain. It gives us a better mental model: ecosystems do not only respond to warming or cooling. They respond to the full suite of environmental changes climate sets in motion.
And once you see that, the best question becomes harder, but also more useful. Not “Was it cooling or warming?” But “What else changes when climate changes?”
Thank you for supporting independent science writing.
Climate Ages has officially become a Substack Bestseller, and it’s thanks to a community of 100+ paid subscribers who believe science writing should stay careful, clear, and accessible.
Your support helps keep these stories free for everyone, while spreading science, trust, and action. I truly couldn’t do this without you.
If you enjoy my work and want to help me keep these stories free for everyone, I’d appreciate your support at any level that feels comfortable:
💡 $1/month or $10/year – A small boost with a big impact.
🌱 $2/month or $20/year – Supporting science communication.
🔬 $3/month or $30/year – Investing in evidence-based insights.
🌍 $4/month or $40/year – Expanding science for everyone.
🚀 $5/month or $50/year – Powering independent science writing.
Every contribution helps me continue creating content that makes science accessible, engaging, and impactful.
With Love,
Silvia P-M, PhD — Climate Ages
A beautifully clear and thoughtful account. "instability vs stability" is key point - it's not that (some) life can adapt to change and some not, but more the chaotic patterns of change that arise in periods of instability - meaning it's not just what kiinds of life can adapt, but how quickly can they adapt.
What was the order of cooling and volcanism? For example, was cooling itself due to a "Pinatubo winter", ie volcanic dust? Then you would have two events in succession, and linked.
Alternatively, does a two mile thick coating of ice cause increased faulting in thin areas of crust an thereby create new volcanism? Thus, cooling for long enough might give the one-two punch.
Unless all the Big Five were chance occurrences of multiple unrelated events happening near simultaneously. Thereby accounting for their rarity.