Knowable Magazine · The history of climate change offers clues to Earth’s future


Listen on: Amazon Music | Apple Podcasts | Deezer | Google Podcasts | LibSyn | Player.FM | Soundcloud | Spotify | Stitcher


How can we study a past about which no history was written? What can clues and relics reveal about the time they were created? And can this picture of our past help us prepare for an uncertain future?

This is Knowable, and I’m Adam Levy.

Earth’s past was not like its present. We now know that the Earth has been variously far hotter than it is today, and far colder. And we also understand that today people are changing the climate: heating the planet and creating an unfamiliar future. But how did we arrive at this knowledge? Well, in their quest to uncover the secrets of yesterday, researchers have given us unique insights into tomorrow.

Sidney Hemming: “It’s sobering, because we have to worry about the future. But from a purely intellectual point of view, it’s extremely exciting and interesting. It’s a most fun sleuthing project to really try to put together what happened in the past and why it happened.” 

This is Sidney Hemming, a historical geologist at Columbia University.

It’s hard to pin down when people first suggested that immense ice sheets once extended far further over the Northern Hemisphere than they do today. But through the 19th century, the evidence mounted. Once you look for them, clues of these past conditions are everywhere: the mysterious scouring of rock surfaces, boulders that only a glacier could have carried and left behind, or fossils of mammals adapted to the cold. Take woolly mammoths, for a particularly striking example. 

And, in fact, by the 20th century it was clear that there hadn’t been just one ice age, but several. What remained unclear, though, was why.

Here’s geoscientist Richard Alley, of Pennsylvania State University.

Richard Alley: “So ice ages proved to be one of the most complicated pieces of the Earth history. And there’s so many moving parts going on that it’s been really tough to unpack. We know the ice got bigger and smaller: When? How many times?”

One theory — perhaps you’ve heard of it — revolves around what are called Milankovitch cycles, named after the Serbian scientist Milutin Milankovitch, who refined the key calculations around a century ago. The theory proposed that subtle shifts in Earth’s orbit around the Sun could lead to periods of time where less sunlight hit the Northern Hemisphere, causing ice to build up over the years. In other words, causing an ice age. But in the middle of the 20th century, many scientists were still skeptical about whether such a fantastical theory really explained ice ages. As one researcher put it in 1952:

The theory cannot account for past changes. The effects are too small and the chronology of … occurrence of glaciation is so uncertain that any correspondence ... appears fortuitous.

In other words, it appeared to be a nice theory, but there was no reason to think Earth’s orbit could drive the planet’s cooling. And it didn’t help that researchers were still struggling to pin down the timing of the ice ages. But that was about to change. Here’s Richard again, who began his career peering into Earth’s past in the 1970s:

Richard Alley: “When I came into the field, it was just really exploding with the ability to find out what happened, and when it happened.”

To do this, researchers needed to learn to read the clues that the past has left behind. This, of course, requires the tools and insights to understand what you’re looking at. But before even that, scientists need to find out where to look.

Richard Alley: “People who study the history of climate desperately need a record.”

The problem for records of the ice ages was, well, there had been more than one ice age. Looking where ice had grown then retreated could only reveal so much, since each new ice age would have trampled over the history of its predecessors. Researchers needed a continuous record that captured the details and timing of the last ice age and those that came before it —from somewhere where climate history built up and up over time.

Richard Alley: “In the ocean, almost everywhere, things are accumulating. They’re not being eroded. And things live in the surface water and they sink and they pile up on the bottom. And you can go out in a ship. And you take a glorified drill — it’s got a pipe that spins. And you can spin it into the mud and then pull it up. And then you can get a record from this.”

Extracting these messy cores from the depths of the ocean without disturbing the layers and so destroying the record presented a huge technical challenge. But the reward was a detailed history book of climate. Each layer of sediment represented a chapter of Earth’s past — and the deeper the layer, the further back in time it was deposited. But researchers needed to learn to read the pages of these books.

One trick came from radioactive dating. Certain atoms decay, and always do so at the same rate. So working out the fraction of atoms that have decayed enables researchers to date the sample. And so, applying this technique, a chronology could be calculated for a core of sediment from the sea floor.

Once they’d established the dates, scientists needed a way to identify signs of ice sheets growing and shrinking within the core. And again, this came down to an atomic trick. Water is, of course, the molecule H2O. But the O — oxygen — can be of different “isotopes.” There are two that particularly interest geologists, one slightly heavier than the other. The lighter evaporates more easily from the oceans. That means snowfall is also disproportionately made of the lighter isotopes, so when ice sheets grow, they effectively pull the lighter isotope out of the ocean, leaving the seas more concentrated in the heavier. 

This ratio between the heavier and lighter isotopes in the ocean finds its way into the shells of tiny plankton, which fall to the bottom of the ocean when they die, building up in the layers of sediment as the years go by, which researchers can dig up as cores, and ...

Richard Alley: “And you can look back through ice age and ice age and ice age and ice age, and you can see the ice getting bigger and smaller, and the ocean getting smaller and bigger, and the temperature changing when the ice is bigger that the whole world is colder.”

Such powerful evidence from a core from the bottom of the ocean was first published in 1973, and there was something remarkable about the timing. The ice ages appeared to line up with features of Earth’s orbit. Long dismissed as implausible by many, this added to a growing body of evidence that Milankovitch cycles were indeed behind the coming and going of Earth’s ice ages. Here’s Sidney again:

Sidney Hemming: “To find evidence in continuous records that could be demonstrated to be approximately paced as predicted based on orbital variations — there’s no question that that was a huge game changer.”

But the correct timing wasn’t enough to put doubts over Milankovitch’s theory entirely to rest. As a 1978 review pointed out:

However, the mechanism of the atmospheric­-cryospheric linkages with these variations of solar radiation remains to be shown in detail.

That review was titled “Glacial Inception and Disintegration During the Last Glaciation” and published in the Annual Review of Earth and Planetary Sciences. One of the most notable things about this review, though, is what it doesn’t mention — one of the things you probably first think of when you think of changes to the climate: carbon dioxide.

By the 1970s, carbon dioxide’s importance for the climate was well understood, and had been for decades. Without some of that gas in our atmosphere, it was clear the Earth would be far too cold for human comfort. But the idea of “global warming” due to the burning of fossil fuels and emissions of CO2 was not so widely discussed.

Sidney Hemming: “So, honestly, 50 years ago, that would have been before, certainly widespread recognition, that there was a problem.”

In fact, some researchers had already come to the conclusion — and warned politicians — that burning fossil fuels could dangerously heat the climate. But the complete absence of CO2 from the discussion of this 1978 review reveals that the importance of this greenhouse gas was not a focus, even by some climate scientists.

But another review published in the same year and journal, “Temporal Fluctuations of Atmospheric 14C: Causal Factors and Implications,” did note the key role of CO2, pointing out the importance of estimating …

… the consequences of the increase of atmospheric CO2 contents for the Earth’s future climate.

And this review had important reflections on how we could come to understand this future climate:

In earth science, prediction of the future depends heavily on knowledge of the past.

So, in the 1970s, carbon dioxide was understood to be important to present climate, but this understanding wasn’t necessarily joined up within climate science — never mind in wider society. And as the 1978 review about glaciation illustrates, this importance did not appear prominently in discussions of the last ice ages. After all, why should it? How could CO2 fit into the puzzle connecting Earth’s orbit and the ice ages?

In fact, to some researchers studying Earth’s cycles of cooling and warming, these dug-up cores hinted at a conclusion that was markedly different to our fears today.

Richard Alley: “The initial thought — and this came out in the 1970s, and if you have any listeners of my generation, they may remember little warnings — that maybe the next ice age was fairly close.”

Such warnings — which were further fueled by fears of the cooling effects of aerosols — were widely seized upon by journalists. But we should offer a couple of disclaimers here. Firstly, “fairly close” has a different meaning to geologists than it does to the rest of us — here referring to tens of thousands of years away. And, secondly, even in the 1970s, there was many times more research into the threat of global warming than the risk of an imminent ice age, indicating that global warming was increasingly on scientists’ minds.

But, while the timing of ice ages had now been convincingly cleared up, exactly what explained this timing was still a way off. How could these subtle changes in sunlight create such dramatic shifts? The plot thickened as it became clear that the entire planet cooled during ice ages. Milankovitch’s cycles suggested that when the Northern Hemisphere receives more sunlight, the South would receive less and vice versa. So, it would stand to reason that the South would warm as the North cooled. But, quite to the contrary, the data showed both the poles warming and cooling in parallel.

Richard Alley: “And this is very, very weird. You’re just moving sunshine around, but the whole world gets warmer, the whole world gets colder, why the heck is that?”

To get to an answer to this question, researchers would need another core, where — once again — the layers would trace a timeline into the past. But these weren’t layers of mud.

Sidney Hemming: “Well, the really crucial connection is the ice core evidence.”

Ice cores dug up from ice sheets or glaciers contain signals that allow researchers to date them and deduce historic temperatures, just like their ocean sediment counterparts. But they have a secret locked within them that sediments can’t match: air bubbles. As new snow falls on top of old, it gradually seals miniscule gaps, trapping air inside. By carefully digging up and defrosting this ice, scientists can directly measure ancient air.

Leading the way on this task was the Soviet Union’s Vostok Station, deep in Antarctica. This remote station battled extreme cold to drill kilometers deep into the East Antarctic ice sheet, revealing hundreds of thousands of years of Earth’s history. And by the mid-1980s, analysis of the air bubbles in these cores delivered a powerful message: carbon dioxide concentrations in Earth’s atmosphere rose and fell in lockstep with Earth’s temperatures.

This helped complete the picture of the driving force of ice ages: variations in the Earth’s orbit lead to seasonal sunlight changes in the Northern Hemisphere, causing it to cool. This can cause carbon dioxide concentrations in the atmosphere to drop, for example as the gas dissolves into the ocean. And this in turn reduces how well the atmosphere insulates the planet through the greenhouse effect, cooling the entire planet. And all these steps reverse to bring about the end of ice ages. Through the shells in sediment cores and other evidence, we already had indirect evidence of how temperature varied over time, known as “proxy data.” The discovery of these CO2 fluctuations provided the missing link, explaining why the Milankovitch cycles were so powerful and could cool the entire globe in parallel.

Sidney Hemming: “The records from Antarctica where the temperature proxy compared to the gas composition is phenomenal, right? It really shows the very strong coupling between global temperature and the greenhouse gas concentration. And further it sets the stage for just how extremely out of whack the carbon dioxide concentration is today.”

That’s because if carbon dioxide could play a crucial role in such dramatic climatic changes as the coming and going of the ice ages, then surely it could also play a role in the planet’s present and future too. And, as Sidney explained, it was clearly out of whack. Today CO2 levels are around 50 percent higher than they were at any point in these ice cores — during ice ages or the warm periods between them.

Now, as we’ve already explained, the fact that carbon dioxide trapped heat and so could control Earth’s temperature was already understood long before all this. But seeing levels of CO2 and temperatures rising and falling in parallel in Earth’s past provided a powerful image of how close this relationship was.

Richard Alley: “So I really do think that this understanding of the ice ages, the role of carbon dioxide, has been a key step in the full understanding of the role of carbon dioxide in our climate.”

You can see this in a seminal 1990 review titled “Energy, Greenhouse Gases, and Climate Change” in the Annual Review of Energy, which...

… aims to provide a comprehensive review of the relationship between energy use and climate change.

This review uses the record of past climates to illustrate just how exceptional human emissions — and their potential consequences — are. For example, evidence from ice cores ...

… illustrates that today's CO2 concentrations... are 20-25% higher than at any time in the past 160,000 years …

The review author suggested that paleoclimate data could provide one line of evidence to politicians, although he still feared political inaction. In fact, the review cautions that if left unchecked …

… warming may result in average global temperatures higher than any seen in the past 160,000 years.

The graphs showing CO2 and temperature dropping and rising together as ice ages arrive and depart have been widely reproduced and shared well beyond the realms of academic journals. Perhaps most notably in the 2006 film “An Inconvenient Truth,” presented by former Vice President Al Gore:

The relationship is actually very complicated, but there is one relationship that is far more powerful than all the others, and it is this: When there is more carbon dioxide, the temperature gets warmer, because it traps more heat from the sun inside.

But evidence from past climates provides scientists with more than just powerful evidence that carbon dioxide can control the planet’s temperatures. These data also allow researchers to evaluate one of the most crucial quantities when it comes to global warming: the often-discussed “climate sensitivity.”

Climate sensitivity tells us how much the planet warms for a particular increase of carbon dioxide. Knowing this is essential to knowing what the future of our planet could look like. Researchers use a range of approaches to estimate the climate sensitivity, from evaluating how climate processes will amplify or suppress the effects of extra CO2 to studying the warming we’ve actually observed over the past few decades. 

But ancient evidence tying together CO2 and temperature changes also provides a powerful tool to calculate how sensitive the climate is to CO2.

Still, it’s not straightforward to get useful estimates of climate sensitivity from past climates. It’s one thing to know that the temperature went up and down in past climates, but it’s another to know exactly what the temperatures were. Or, as Sidney puts it: 

Sidney Hemming: “The hard thing about using the paleoclimate record to get at that, is that we definitely can’t take a thermometer back in the paleoclimate record.”

But even though our measurements — whether that’s of temperature or timing — of past climates are still uncertain, they still provide researchers with invaluable information, presenting another line of evidence to decode the climate. As a 2018 review titled “Comparing Climate Sensitivity, Past and Present” in the Annual Review of Marine Science explains:

… climate sensitivity estimates from paleoclimate data have the merits of being based on real data …

And investigating past climates remains one of the core methods scientists use to predict how hot Earth’s future will be. Beyond climate sensitivity, though, past climates have a lot to teach us about our path. Building up a more detailed picture of these past climates could help shed light on the future that awaits us as we continue to burn fossil fuels.

Sidney Hemming: “As we collect more and more different kinds of data, we’re getting a closer and closer picture of the kinds of changes that can happen and how fast they can change. The real gains are going to be made by creating global perspectives. So if we can find enough records, like, for example, in time periods that had CO2 concentrations like we have today. And to see what was going on globally in that time frame.” 

Richard agrees that this evidence could have profound implications on our predictions.

Richard Alley: “So there’s this vast world of understanding history that will inform understanding: what’s going to happen where you live depending on what we decide to do, what does it mean for living things, including other species, as well as us.” 

We’ve come a long way from just a few decades ago, when the Earth’s climate past was shrouded in mystery. It has taken countless researchers digging up clues from other ages, and countless more to learn how to read these remains.

Richard Alley: “For me, it’s just been glorious. I’ve played a very small role — I’ve gotten to be in it and I’ve gotten to see what other people are doing. And it’s just brilliant work has been done: concentrated, focused, international, interdisciplinary, people who are leading in so many different ways, people who are working to diversify science to bring in more views so we can understand better what’s going on. And I just sit here now with this big smile on as to what has happened and how good the colleagues have been.”

But as thrilling as this journey into the past has been, it has also helped build a bleak understanding of the future. Today evidence indicates that carbon dioxide levels are higher in the atmosphere than they have been for millions of years. As expected, this greenhouse gas is driving up the planet’s temperatures, ramping up extreme weather, raising our oceans, and destabilizing human and natural systems across the planet. And our emissions continue to rise, just as scientists caution they need to rapidly fall.

So what will the climate of tomorrow look like? Digging up Earth’s past has given us some clues. But ultimately, it will be down to humanity to decide its path.

Sidney Hemming: “It really is a sobering picture because I think the collective wisdom that we’ve developed over these years has led to an inescapable recognition that humans are making a huge difference on the planet. While I still really enjoy studying the past, I do really worry about what this means for the future.”

If you enjoyed this episode of the Knowable Podcast, then why not let others know about it too? We’d also love to hear from you, write a review wherever you listen to your podcasts, tweet us @KnowableMag, and email us on [email protected]. And don’t forget to subscribe so you don’t miss the rest of this season and beyond — the history of particle accelerators, and the battle to treat depression are still to come.

In this episode you heard from Sidney Hemming and Richard Alley. The episode also featured quotes from four articles published by Annual Reviews. They are: Andrews and Barry, 1978; Damon et al., 1978; Mintzer, 1990; and Rohling et al., 2018. You can find links to those papers and more in the show notes on our website:

This podcast was produced by Knowable Magazine, a nonprofit publication that seeks to make scientific knowledge accessible to all. Knowable Magazine is an editorially independent initiative from Annual Reviews. Explore more sound science and smart stories at

I’m Adam Levy and this has been Knowable.