Michael: Earth has been through a lot in the four and a half billion years since it formed. Most of Earth’s history has been shaped by plate tectonics, where continents slide around. But instead of skirting around each other neatly, the continents can interact in some pretty unexpected ways. Continents come together and burst apart while the rocks at their centers stay put.

Earth’s crust is flung upward by tectonics and weathered back down by the atmosphere. And that’s led to a lot of changes over the years.

[SciShow intro plays]

For the earliest part of Earth’s history, we don’t have any rocks at all -- meaning we can’t study it directly. For the first part of its life, the planet was a molten mess, constantly bombarded by asteroids and not stable enough to preserve much record of that time. So any geological history of Earth has to start when the continents started to stabilize, somewhere between 3.5 and 3 billion years ago.

These super-old rocks don’t exactly have dates written on them, but a quirk of chemistry can tell us how old they are. And that has led to an entire field called isotope geochemistry. See, each atom comes with a certain number of protons and neutrons in its nucleus. Different elements are defined by the number of protons in the nucleus, so the atoms of each element all have the same number of protons. But there’s usually some wiggle room in the number of neutrons.

For example, carbon has six protons, but natural carbon can have six, seven, or eight neutrons. Those carbon variants are called isotopes, and they’re named based on how many protons and neutrons they have. So carbon atoms with 6 protons and 6 neutrons are carbon-12, 6 protons and 7 neutrons makes carbon-13, and so on.

Because different isotopes have different numbers of neutrons in them, they weigh different amounts. It’s a tiny difference, but it’s enough that geochemists can separate them and calculate how much of each one there is. Some isotopes decay over time. Carbon-14, for example, loses a proton and turns into nitrogen-14. And atoms that start out as uranium-238 decay over and over again until they eventually become lead-206.

Because we know how long those decaying processes take, the ratio of decayed isotopes to their non-decayed precursors can give us a good estimate of an object’s age. Uranium decays to lead over a very long period of time, so the ratio of different lead isotopes can be a very useful resource in telling the age of a rock.

And that’s only one of the simpler things isotope geochemistry can do. Ratios of isotopes can change as a result of all sorts of natural processes, not just radioactivity. The ratio of carbon isotopes can show whether carbon trapped in a mineral was once used for photosynthesis, and therefore was part of something alive at some point.

And sulfur isotopes can be used to show whether a mineral was formed near the surface of the Earth or much farther down. Isotope geochemistry gives us all kinds of information about the world’s oldest rocks. And those ancient rocks give us a way to piece together the history of the continents — including the way plate tectonics has shifted them around.

Plate tectonics are a thing because Earth, like an onion, has layers. Some layers are solid and some are more fluid. The outermost layer is, obviously, solid. That’s the lithosphere, which makes up the plates that hold the continents and oceans. Underneath that is a layer it’s little bit more liquid-y, where rock is flowing. That movement pushes the continents around, which gives us plate tectonics.

But there’s a catch: not all lithosphere is created equal. The crust that holds up the continents is thicker and less dense than the crust beneath the oceans. That continental crust floats really well, if you can imagine a slab of rock thousands of kilometers across floating.

That means, when oceanic crust and continental crust meet up, the oceanic crust tends to get shoved underneath and melted in a process called subduction. The continental crust rides on top and survives to collide another day. And that means that certain chunks of every continent go back as much as three billion years or more.

These super-stable continental chunks are called cratons. They’re made of tough, floaty rock that often hasn’t been melted by plate tectonics for three billion years or so. The youngest ones clock in around half a billion.

Back when Earth was still a liquidy mess of molten rock, the denser elements slowly sank towards the core. And just like oil floats on top of water, the less dense elements rose toward the surface. So continental crust is mostly made of relatively light rocks rich in silica. As these light rocks started to cool and condense, they would have bobbed like a cork on the surface of the planet. Those little floaty corks would have bashed into each other and, instead of one subducting under the other, they would have stuck together. After a while, you would get.. bigger floaty corks.

These chunks really started to stabilize into continents during the Archean eon, 4 billion to 2.5 billion years ago. Archean cratons formed the nuclei of the first continents, and they’ve stuck around ever since. Geologists think most continent building happened way back then and was pretty much done after that. Sure, they’ve been rearranged a ton by plate tectonics, but much of the actual land is the same land that existed in the Archean. But some evidence suggests that lighter rock can still bubble up from inside Earth sometimes and add new bits and pieces.

You might’ve heard of Pangea, the supercontinent that existed around the time of the dinosaurs. Geologists think Pangea is only the latest supercontinent in a planetary boom and bust cycle, where supercontinents assemble and break up every so often. They aren’t totally sure why this happens, but the leading hypothesis is that those big blocks of thick continental crust trap lots of heat beneath them.

Eventually, the trapped heat bubbles over in the form of magma plumes and blows the supercontinent apart, like billiard balls breaking up in slow motion. Then the pieces bounce around until they meet up again, with continental crust sticking together instead of getting subducted. And the cycle begins again.

Geologists have a pretty good understanding of how Earth’s continents have moved around over time, which they’ve figured out by essentially matching up rocks like puzzle pieces. Plate tectonics was first proposed in part based on how neatly South America and Africa fit together, suggesting they were once part of the same landmass -- Pangea. Hunting for older supercontinents is like that too, but with the difficulty ramped up to 11.

The main tool geologists use for this is called paleomagnetism, which is based on the fact that Earth’s magnetic field regularly reverses itself. When a rock forms, any magnetic bits in it will line up with Earth’s magnetic field at the time. That means a rock containing magnetic particles will reflect where Earth’s magnetic field was pointing when the rock formed.

And since the magnetic field reverses on the order of thousands of years, that provides a ton of data when we’re looking at things that happened over billions of years. Using math, we can pinpoint pretty accurately where on Earth the rock was. So the question of where the continents have been throughout Earth’s history is a challenging puzzle, and one that’s far from being solved. But we do have some tools that we can used to figure it out, and geologists have some ideas about the supercontinents that assembled before Pangea.

We call the very first continent Ur, and it was the first big bit of land to form from small islands. Ur goes back about three billion years, and was made up of bits of what is now Africa, Australia, India, and maybe Antarctica. In fact, Ur only broke up recently, when Pangea did. A continent that can last nearly three billion years is one heck of a continent!

Some evidence points to a landmass even older than Ur, known as Vaalbara, as far back as 3.6 billion years ago, but the evidence for its existence isn’t conclusive. Ur was eventually joined by more brand new continents, like Arctica, Atlantica, and Nena. By the way, these continents might have odd-sounding names, but if you pick them apart you’ll realize that a lot of them are smashed-together words to represent the smashed-together landmasses. Nena, for example, gets its name from the first letters of Northern Europe and North America.

All these continents are thought to have joined up about 1.9 billion years ago to form the first supercontinent that can be identified with some degree of confidence, called Columbia. Columbia lasted until about 1.5 billion years ago, when it broke up. The pieces then rebounded and joined up to form the supercontinent of Rodinia about 1.1 billion years ago.

After Rodinia, there may have been a very short-lived supercontinent called Pannotia, but things were definitely on their way to becoming Pangea by 450 million years ago, and at maximum scrunchiness around 250 million years ago. Then, between 170 and 100 million years ago, Pangea broke up into the continents we know today. We’re on track for another supercontinent in about 250 million years, give or take.

If North and South America continue to drift westward across the Pacific, they’ll meet up with Russia and form the supercontinent of Amasia. Now, all those continents shoving each other around doesn’t come without consequences for our world on the surface. That’s actually how we get mountains — the boundaries between tectonic plates often produce mountain ranges.

When oceanic crust is subducted under continental crust, the continental crust is shoved upward to compensate. The Andes in South America are a good example of that. When two pieces of more-resilient continental crust meet up, the results can be even more dramatic.

Continental crust doesn’t tend to subduct, so instead it just kind of... folds upward. And even when mountains form through the same basic process, they can still look hugely different from one another. One key difference is their age. The Appalachian mountains in North America aren’t much more than hills at this point -- basically scenic, rolling slopes compared to some other mountain ranges, like the Himalayas in Asia, the greatest mountain range on the planet and home to Mt. Everest. But the Appalachians were once even taller than the Himalayas.

The Appalachians and the Himalayas were formed in similar ways: The Himalayas came from the Indian subcontinent crashing into Asia. India had to cross the ocean to get there, meaning the oceanic crust north of it was subducting under the Tibetan plateau. But then the subcontinent hit, with its thick, cratonic continental crust. India and the Tibetan plateau crunched directly into one another and folded up like an accordion. All this happened relatively recently, within the last 40 million years or so. In fact, it’s still happening, and the Himalayas are still growing -- although that growth might be matched by weathering and erosion.

Given enough time, weathering can shrink mountain ranges by a lot — which is what happened to the Appalachians. The Appalachian mountains formed when the North American and African plates collided during the formation of Pangea. They might have been even taller and more impressive as the Himalayas are now, with two continental plates colliding and refusing to give way. But that was almost 500 million years ago, and 500 million years is enough time for a lot of rain and wind to wear the Appalachians down. So, huge continents smash together and break apart. Mountain ranges form and wear back down.

And even though we weren’t around to see those things happen, we can learn about them just by studying rocks. Thanks for watching this episode of SciShow, which was brought to you by our patrons on Patreon. If you want to help support this show, just go to Patreon.com/SciShow. And don’t forget to go to YouTube.com/SciShow and subscribe!