Skip to main content
Loading Explore guides…
History of Earth's Climate · Acedra
Home
Explore
Geography
History of Earth's Climate
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
46 segments
Copy
Hide video
0:01
Well thank you very much for being here today. It's a real pleasure to be part of this event. Sarah's just covered some things on a galactic scale in 15 minutes. So I should have no trouble covering 4 and 1/2 billion years in 15 minutes. So I'm here to talk about the history of Earth's climate. With trying to understand the climate system, we of course start with direct observations. So just south of here is the oldest continuously operating meteorological observatory in America.
0:26
It celebrates its 131st birthday this week. And over the world, we have pretty good density of direct measurements of things like atmosphere temperatures, and winds, and precipitation, sea surface temperatures, spanning roughly the last 100 years. The so-called observational or instrumental era. But Earth's history, of course, is 4 and 1/2 billion years. And so our direct observations cover but a tiny fraction of the history of Earth's climate. And just as they cover a tiny fraction in time,
0:59
they also cover a tiny fraction of the possible variability in Earth's climate. So we've seen a small change relatively-- by geological standards. Small change in CO2 levels. A relatively small change in Earth's surface temperature. In Earth's history we can look back at dramatic changes in boundary conditions and forcings. And really test our understanding of how the climate system responds under very different conditions. Looking at paleoclimate also allows
1:27
us to ask really important questions. Like, what was the climate like during the evolution of the earliest life between 3 billion and 4 billion years ago? Or what was the role of climate in the rise of complex life a half billion years ago? Or, similarly, what was the role of climate-- what was climate like during the evolution of anatomically modern humans and their dispersal out of Africa? A little bit later we'll hear a little bit about the interplay between climate and mass extinctions
1:55
in Lindy Elkins-Tanton's talk. So to unlock this record, of course, we don't have direct observations. And so what we have to do is to go into natural archives. These are deposits that somehow encode information about the climate in which they formed, and that we can come and unlock many millions or billions of years later. So in short time scales, like centuries or millennia, we have deposits like tree rings or tropical corals. On longer time scales, tens and hundreds and thousands
2:26
of years, and we have ice cores, and stalagmites, and lake sediments. Still, longer time scales, tens of millions of years, we have marine sediments. And for everything else, there's rocks. Now we can go into these archives and make increasingly sophisticated measurements of their isotopic and chemical properties to unlock information about past surface temperatures, for example. But we can also make very, very basic observations. Like the presence of rocks that indicate the presence
2:57
of ice sheets when they formed. So ice sheets leave behind very characteristic deposits-- really jumbled sediments that are a mix of different grain sizes, or these characteristic striations in bedrock. And going back in time, one of the first order things that we can do is to map the occurrence of these ice sheet indicating strata versus space and time. And we can use magnetic measurements to understand at which latitude the continent was when these ice sheets were there.
3:28
And compiling decades of research, we get a plot like this. Now this is just spanning the last 400 million years with time moving to the right here. And on the right hand vertical axis here, what you're seeing plotted is the latitude to which ice sheets extend. And so we see two periods where we have substantial ice cover on the Earth. The last 30 million years, where there was first an Antarctic ice sheet and later northern hemisphere ice sheets.
3:53
And then this period around 300 million years ago. So one of the things that jumps out to me about this plot is that greenhouse climate studies climates with very little ice cover, no evidence of ice cover, have really been the norm for the last several million years. The other thing that's plotted here are reconstructions by proxies and proxy-driven models of atmospheric CO2 levels.
4:19
There is a wide variety of estimates that there's pretty good evidence that atmospheric CO2 levels were quite high in the realm of 1,000 PPM or more for much of this time period, consistent with the lack of ice on Earth. So one of the things that this means is that the rock record is really well suited to teach us about climates that are warmer than today. If we can only unlock the record. So the time when we have the best chance of doing this, of course, is the most recent time
4:47
when the Earth was substantially warmer than present. And that's a time called the Eocene. So here we're zooming in the last 65 million years. And this is a compilation of atmospheric CO2 concentration estimates, with a lot of scatter in the estimates, but generally some agreement that it's between, say, 600 and 1,500 PPM. And this is a proxy-based estimate of high latitude temperatures up here, suggesting that the poles were much warmer than today. So when we go back to these times
5:19
we can start to make measurements. For example, the composition of fossil membrane lipids preserved in marine sediments. And these give us a quantitative means of estimating temperatures this time. And so we can compile these temperature estimates. These are all sea surface temperature estimates. Plotted versus latitude, for two different time slices in the Eocene. And so the sea surface temperature estimates are on the vertical axis here. And then the black line is the modern distribution
5:51
of sea surface temperatures peaking around 28 degrees C in the tropics. So what we see, as we might expect, is that things were much warmer in the Eocene. But what's particularly surprising about these data is just how much warmer it was at very high latitudes. Something like 20 degrees C warmer in many high latitude regions of the ocean. And this warming doesn't just extend to the ocean or to coastal regions. It also appears to extend to continental interiors.
6:20
Seen perhaps most strikingly by fossils, like fossil crocodilians at 55 degrees north in interior North America during the Eocene. And what's interesting about this and the way that the paleoclimate record has the capacity to surprise, is that if you take the general circulation model back to Eocene boundary conditions and reasonable estimates of CO2, it's very difficult to reconstruct these warm conditions, and sea surface temperatures warm conditions in continental interiors.
6:51
And so this is an active area of research. Both from the proxy side, making sure that we really have our story right from the data side, but also from the dynamics side. So Tim Cronin, Kerry Emanuel, and others in EAPS are working to try to understand what are some of the processes that allow you such high temperatures at the poles, given Earth's radiation budget. And the answer most probably is clouds. But Dan will tell us about that. If we look still further back in time,
7:21
so this is just an expanded version of that same plot. Again just looking at where the ice sheets extend. I mean, ignore these vertical red bars here. They're mass extinctions. You can hear about those a little bit later. But we see these two striking periods where we are at the opposite end of Earth's temperature spectrum. These two times when ice appears, at least from this schematic, to extend from pole to pole. These are so-called snowball Earth events, around 630 million years ago.
7:47
And a little bit more than 700 million years ago. And it helps with a very simple schematic like that to really ask, well what's the evidence for that? This is a reconstruction of where the continents were 635 million years ago in the most recent of these snowball events. And with the colored dots, you see places where we have rock record that can be confidently dated to that time. Now there is different little symbols here that are very difficult to see from where you're sitting.
8:17
It's very difficult to see even when you're reading the paper. But I'm going to do now is put stars in all of these deposits that are related to ice sheets. The really neat thing we see here is not just ice sheets near the poles, but evidence of ice sheets right at the equator. And these are not tropical mountain glaciers, like the glaciers of Kilimanjaro, the tropical Andes. Evidence in the rocks suggest that these deposits were formed right at sea level, with marine terminating ice
8:45
sheets like we see in Antarctica and Greenland today, requiring, essentially, a near global extent of ice cover. Now, of course, when ice forms on Earth, it starts to reflect more and more incoming solar radiation, making the Earth as a whole more reflective. Lowering surface temperatures, leading to more ice growth. And essentially forming an amplifying feedback. So perhaps it's not surprising that we have these periods where Earth went to near global ice cover.
9:15
In fact, what's really surprising about these, at least to me, is that these periods-- these so-called snowball Earth events- occupied not more than a few percent of the Earth's history that we've reconstructed from the rock record. The real question here is not how do you get into it. You have an amplifying feedback to get you into it. But how do you get out of it once you're in the state of a highly reflective, ice-covered, cue ball Earth. Well the answer to this, we think,
9:44
has to do with stabilizing feedbacks within the carbon cycle itself. That is, on geological time scales, the removal of carbon dioxide from the atmosphere is principally accomplished by burial of calcium carbonate in deep ocean sediments, followed by the subduction of that calcium carbonate into the deep mantle. And we believe that this burial is related to Earth's surface temperature. Because of the dependence of weathering and the supply of new calcium in soils,
10:14
the dependence of the [? atom ?] temperature. And so, in a snowball world, you would decrease weathering, decrease calcium carbonate burial, decrease the removal of carbon dioxide from the atmosphere. And you would get a slow build up of carbon dioxide from volcanism, that eventually get you to high enough CO2 levels, perhaps 50,000 PPM, to pull you out of this snowball glaciation. And then the same amplifying feedbacks that got you into it could get you out of it.
10:39
On the whole, we think this is a stabilizing feedback. It keeps CO2 levels from getting too high by increasing weathering as the Earth warms. It keeps the Earth from getting into a prolonged snowball state. But it operates on time scales of hundreds of thousands to millions of years. So this isn't to say that the carbon cycle can't have surprises for us on shorter time scales. One important example, or sort of poster child example of this, is the time called the Paleocene-Eocene Thermal Maximum.
11:09
I'm not going to be able to really summarize the evidence today. But look it up on Wikipedia. It's a very interesting time. It appears to be a time when something like several thousand gigatons of carbon were added to the ocean atmosphere system over a relatively short amount of time. How short? We don't know, but on the time scale of hundreds to thousands of years. From a reservoir that is as yet unidentified. And this was this happened several times during the Eocene.
11:35
It lead, as we would expect, to isotopic evidence of pronounced warming. Something like 6 degrees C on a global basis. And to pronounce ocean acidification, a signal pronounced seafloor carbonate dissolution. So this is sort of the closest thing we have to a rapid increase in atmospheric CO2 in a greenhouse world. And again, the real question is where those carbon come from. And then in much more recent time scales, just zooming in the last million years, or here the last 800,000
12:11
years, we have, of course, the well known ice age cycles of the late Pleistocene. And here we have direct measurements of the CO2 content of the atmosphere from air trapped in ice cores. So we have these swings between lower and higher CO2 over the last 800,000 years, matched by changes in mean global temperature. These are Antarctic temperatures so they're amplified. But the mean global temperature changes are probably something like 4 to 5 degrees C. And ice cover going from near today
12:40
to much more extensive ice cover, about 120 meters of sea level equivalent. Now some of us are trying to use this as a way of understanding how carbon moves back and forth between the ocean and the atmosphere. As it almost certainly did. Asking the question of, well, why? Why do you have these oscillations? They seem to be paced by orbital forcing. By slow changes in the Earth's orbit. But certainly you need amplifying feedback, say, from CO2. What brings the carbon cycle into it?
13:10
But another way in which they are really useful is they provide a natural experiment about how other parts of the Earth's climate system respond to warming and cooling, and to changes in the spatial distribution of temperatures on Earth. So in my lab, we asked the question of how have precipitation changes responded to these past climate changes? Just taking them as a given, how do things like the tropical rain belt, or continental monsoons, or the world's dry regions respond
13:41
when Earth's orbit changes? Or when mean global temperature changes? Or when you put big ice sheets or remove them? And so we do this really the main way that you have-- is to look into these natural archives. Archives like stalagmites, or deep sea sediments, or lake deposits. And by doing precise uranium series dating of them, to put them on a confident chronological time scale. We can start to map precipitation distributions in past worlds. And what we find are some really fascinating changes
14:12
that speak to the sensitivity of precipitation to changes in boundary conditions and forcings. So time of the green Sahara 6,000 years ago. where there was widespread human settlement of the Sahara. Or a time 17,000 years ago, where there was a lake as because Lake Michigan, where the Great Salt Lake was as big as Lake Michigan. We can look at these really big signals and ask, well, what were the dynamics that led to those? And what was the spatial distribution of precipitation
14:41
that that accompanied these changes? So with that I'll close with just some summary points, which is just to point out that the paleoclimate record is this rich natural laboratory for understanding the Earth's climate at the broader scale. So really testing our understanding of how the climate system works when it's pushed well outside of the conditions that we've been able to directly observe. And the thing that I really appreciate of the paleoclimate record is, like with exoplanets,
15:10
it contains surprises. There are times when we go back and we see changes that we don't expect. And I think this is a fantastic learning opportunity, and certainly an exciting place of discovery. So thank you very much. [APPLAUSE]
15:32
Thank you very much, David. We have ample time for questions. Question in the back. In that Paleocene-Eocene boundary with the increase of ocean acidification, is there any fossil evidence of a, say, a slowing of the growth of coral reefs, or a reduction in the amount of fossils, of calcium sequestering animals? Yes. So I'll repeat the question. The question is, during the Paleocene-Eocene Thermal Maximum, or this evidence of ocean acidification, whether there is any evidence for impacts
16:09
on coral reefs or other calcifying organisms. The evidence that I do know about is that there is a widespread extinction event for benthic calcifiers, that is calcifiers living on the sea surface in the deep sea. So these are foreign for us. A small zooplankton So there is this extinction event. And the question, I guess, is whether it's related to the rapid warming of the deep sea, or whether it's the acidification.
16:41
Is there any chance that these slides could be made available to us either online or hard copy later on, if not now? There will be. Yes. At the end of the day. Yeah. Question in the back. [INAUDIBLE] sophisticated cause to simulate [INAUDIBLE]. How do they do [INAUDIBLE] Some of your colleagues claim that they reproduced them very well. Knowing some of this [INAUDIBLE] personally skeptical. What is your comment? Well this is a very active area of work right
17:19
now-- is to build rich enough paleoclimate data sets, that we can provide sort of a testing ground for taking the same GCMs, the same models that are used to project forward, and trying them out, taking them for a spin in a paleoclimate setting. The major features of what we observe in the paleoclimate record are reproduced. What tends to be, or at least there's a couple of prominent examples, where in general the models get the right answer in the right direction
17:52
but underestimate the magnitude of the change. So for example, in this greenhouse climate example that I gave, underestimating the magnitude of the warming at high latitudes. Or for example the green Sahara problem of 6,000 years ago is an example where climate models classically get the right answer. They make the Sahara a little bit wetter, but not nearly as wet as the pollen and other evidence would suggest.
18:17
Any other questions for the David?
18:21
One in the back. Robin. Before you do, can I ask you and all the speakers that follow to quickly paraphrase and repeat the questions. Yes. Sorry. The question was how do models do in reproducing these changes. Robin. Do you have an estimate of how much natural CO2 is degassed from the planet compared to what human beings were producing? I believe the number is something like-- sorry? Oh yes. Question. I'm so excited to answer. So the question is how much CO2 is naturally
18:57
degassed from the planet, compared to what human beings are putting into the atmosphere. So the number for humans, at least a few years ago, was roughly 10 gigatons per year. It's a few bit more than that now I think-- of carbon. And I believe the number for volcanism is something like 0.1 gigatons per year.
19:21
But someone else in the audience could correct me on that. Ron? Yes. Roughly? 1 decimal point, yeah? OK. Close enough. Time for one more question.
19:35
[INAUDIBLE] have you been modeling the impact that that would have on climate change? So speaking from the paleoclimate perspective-- Could you ask-- Oh yes. So the question-- at least I'm consistent-- is have we been looking at how humans have perturbed the nitrogen cycle. And I can only answer from a paleoclimate perspective. What we're looking at the paleoclimate world is-- nitrogen is one of the parts of biogeochemistry that has a very short residence time in the ocean.
20:06
It's only a few thousand years. And so while other parts of the calcium content of the world's oceans is quite constant, it's got a really long residence time, nitrogen has the capacity for very fast changes. And so one of the things that people in the paleoclimate world are looking at is how stable has the nitrogen cycle been in past climate. So if you change, for example, the removal of fixed nitrogen by [? denitrification, ?] does nitrogen fixation compensate very rapidly?
20:36
Or are there times where you can have pronounced excursions in how much fixed nitrogen there is? And the jury is still out with that. One last question for David. I'll repeat this one. [INAUDIBLE] Sorry, say it again. Methane-- how does it fit in? Where does methane fit into this.
20:59
So the only time when we have direct estimates of how much methane there was in the world's atmosphere is during the ice core era of last 800 and 1,000 years. And it's basically moving along with CO2, at least at time scales of more than 1,000 years. So it's amplifying these glacial interglacial swings. It's lower during glacial periods. For previous times, we don't know how much methane there was in the world's atmosphere. Methane has to be called upon-- but we think, at least
21:27
prior to the rise of oxygen in the world's atmosphere 2.5 billion years ago-- that methane was probably a big part of keeping the Earth warm during the first 2 billion years of Earth's history, when there was less oxygen and when the sun was only about 20% less bright than it is today.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat