TEOTWAWKI 10: Earthquakes (2017 Podcast Transcript)

Hello, and welcome to physical attraction — the show that usually explains physics one chat-up line at a time. But on the side, we’ll be exploring the end of the world… one apocalyptic scenario at the time.

Today, I’m going to be starting a special series of episodes that I’ve wanted to do — probably since I was thirteen years old and first had unrestricted access to Wikipedia. I guess we all remember those heady days, right? You’d start on a simple article about Henry VIII that you needed for your homework, and fourteen hours later, with red eyes, you’d be reading about… the principle of falsifiability in science, or some rare and bizarre languages. Or MICRONATIONS, which are amazingly fun to read about. (Long live the glorious kingdom of seaworld.) But one area I always came back to, time and time again, like a moth to a flame, was the end of the world. People say to me “lighten up!” — or they used to, at any rate — but really, I am perfectly light already. It’s just very simple; sometimes you feel like something is too specific or too mundane to be worthy of spending much time on. It’s that classic moment when you catch yourself engaging in small-talk with someone you barely know; maybe you’re in a lift and obliged to chat to each other for as long as you’re confined together in space… and it maybe goes along these lines:

Them: I see they still haven’t fixed that chair.
Me: Yes, I hope they fix that chair soon. It squeaks when you sit on it.
Them: It’s a health and safety hazard, really.
Me: I’m always worried that it’s going to collapse when I sit down on it.
Them: That probably wouldn’t happen, but someone should come and fix it.
Me: Do you know who’s in charge of fixing it?
Them: I’m not too sure, but I guess you could ask around.
Me: I think I’ll do that.


Yes, I understand that chair dialogue has its place, and yes, I know that small-talk isn’t really about the content of what you’re saying but communicating the message “You are a human and I validate you! Well done for being a human! We are both humans aren’t we, yay!” and not what you’re actually saying… but even so… wouldn’t life be just a little better if some of these conversations got replaced with talking about something more interesting? Something that affects everyone? Something that you can have quite dramatic opinions about? Something that’ll maybe kick off a little bit of imagination? Something like the end of the world as we know it?

So, in classic clickbaity fashion, I wanted to produce a top ten list of potential apocalpyses. And I’m going to rank them in the order of the threat posed. This is a tricky calculation that will include how probable I think the event is to happen, how dangerous it would be if it did happen, and other factors that I’ll come up with on the fly. I’m going to describe what each apocalypse might look like, how we can avoid it (if it’s avoidable), and why I think it deserves its particular place on the list. Now the important caveat is that I’m far, far from an expert in any of these fields; even the ones that are fundamentally to do with physics — we spent way too little time studying the end of the world in my degree as we should have done. So if I get things wrong, it’s just because I’m a huge fan, and you’ll have to forgive me. And, as ever in this show, if you know more, let me know! If there’s an apocalypse you think should be featured, then tell me about it, and I might recalculate my list.

The final bit of housekeeping to point out is around the rather shaky definition of “The End of the World As We Know It”. So I don’t think, for an apocalyptic scenario, you need to satisfy particularly specific conditions. There are some apocalypses, like a massive gamma-ray burst in the nearby galaxy, or — for example — if all that rubbish about the LHC producing black holes was true — that could probably destroy the Earth, or render it completely incapable of hosting any life. Meanwhile, there are some apocalypses that aren’t nearly as bad. For example, consider an electromagnetic pulse. People are kind of concerned that North Korea might develop the capacity to set off an electromagnetic pulse, maybe by detonating one of their nuclear weapons at high altitude in the atmosphere. At the moment, it’s not considered a credible threat — and I should also point out that what we know of nuclear weapons in the atmosphere producing EMPs is limited, but doesn’t suggest they’d be that powerful. Certainly, at present, if you want to cause damage, you’re better off actually dropping the nuclear bomb directly on the city. More likely, an electromagnetic pulse could come from a nearby region of outer space — there are all kinds of processes that can produce them. EMPs have the potential to damage or destroy electronic devices; for example, one of the ones generated on Earth did run through some power-lines in Kazakhstan and caused the local power plant to catch fire. So you could imagine an apocalyptic EMP that destroys or disables most electronic devices. This wouldn’t kill everyone immediately — you’d imagine anyone not on life support, or in an aeroplane, would be okay, although millennials like myself would probably have nothing to live for without our smartphones — but the medium-term consequences would probably be apocalyptic. If every power plant failed overnight, and there was suddenly no electricity at the plugs — if every computer and television failed — civilization would almost certainly collapse. It would be the end of the world as we know it, even though in some ways the Earth would be just as habitable as it was before. This is an equal opportunities apocalypse series; so anything that messes with civilization enough that people would refer to it as “THE EVENT” afterwards is fine by me.

Okay, without further ado: the final countdown.

10. Mega-Earthquake/Mega-Tsunami

9. Supervolcano eruption

8. Cyber-warfare

7. Malthus’ Revenge

6. Death From Above (Meteorites, GRBs)

5. Peak Oil

4. Supervirus!

3. Climate Change / Ecosystem Collapse

2. The Singularity — AI or Nanotechnology

1. Nuclear war/error

10. Mega-Earthquake/Mega-Tsunami

Let’s talk about earthquakes. As we know, the Earth’s crust is made up of tectonic plates — vast plates of rock that shift around due to convection and the motion of fluid in Earth’s mantle. They, roughly speaking “join together” at fault-lines. But in reality, the plates can’t slide past each other frictionlessly; what tends to happen is that they “catch” on each other, stick for a while — with pressure building, because whatever predominant force is pushing them along is building up pressure and elastic strain energy. Then, eventually, it all becomes too much, and they slip dramatically, releasing all that built-up energy quickly and causing vibrational waves to propagate, which we feel as Earthquakes. For this reason, they occur predominantly around faultlines, like the San Andreas fault that runs through California, and others in Japan, South America, and the middle east.

The Richter scale is actually one of the few examples of a logarithmic scale that most people are familiar with, but this comes up all the time in physics. It’s a logarithm to base 10 scale. That means that a magnitude 9 earthquake is 10x more powerful than a magnitude 8 earthquake.

The Richter magnitude of an earthquake can actually be calculated as a ratio concerning the amplitude of the shaking/vibrations of the earthquake. But it wasn’t always explicitly calculable; its precursor, the Mercalli scale, was based pretty much on human reactions to the earthquake: so a 1 was barely noticeable, a 5 was enough to wake people up and break some windows, a 6 would “frighten many”, a 10 would destroy lots of structures, and when you get up to a twelve, you can actually see the waves of the earthquake on the surface of the ground and objects are being thrown up into the air. Now although this scale might seem a bit wooly and less quantitative than the Richter scale, it’s important to note that the Richter scale alone isn’t really enough to tell you about the impact of the earthquake on humans. It depends on the depth at which the earthquake occurs. So, for example, a 0.7 magnitude quake was put in category three (vibrations similar to a passing truck) when it occurred in California, because it was only 4km from the ground level. Meanwhile, a 4.5 magnitude quake — which is as you can see over 1000 times more powerful — was a 1 on the Mercalli scale, because it occurred 165km down and was felt by nobody — even though the vibrations were much greater than the baseline. And, of course, the Mercalli scale is biased against earthquakes that occur away from centres of population density. So the most powerful earthquake ever recorded to hit the UK was magnitude 6.1, in 1931, the Dogger Bank Earthquake — but it occurred way off in the North Sea, and so only ended up being a 3–4 on the Mercalli scale. Reportedly some buildings were twisted up, and the head of a Madame Tussaud’s waxwork fell off, but there was no apocalyptic damage.

For our purposes, we’re going to need an earthquake or series of earthquakes that’s very high in magnitude and also easily a 12, or possibly an imaginary 13 or 14, on the Mercalli scale. The highest magnitude earthquake ever recorded — that’s a matter of some debate. The lists often include estimate magnitudes for historical earthquakes, based on the writings of the people at the time. But, of course, they didn’t have seismographs, so you have to take that with a grain of salt.

Interestingly, one of the oldest ones you’ll see reference to is the Sanriku earthquake, which occurred in 869AD off the coast of Japan. They have to infer that this was around an 8.9 on the Richter scale via the written accounts:

On the 26th day of the 5th month (9 July 869 AD) a large earthquake occurred in Mutsu province with some strange light in the sky. People shouted and cried, lay down and could not stand up. Some were killed by the collapsed houses, others by the landslides. Horses and cattle got surprised, madly rushed around and injured the others. Enormous buildings, warehouses, gates and walls were destroyed. Then the sea began roaring like a big thunderstorm. The sea surface suddenly rose up and the huge waves attacked the land. They raged like nightmares, and immediately reached the city center. The waves spread thousands of yards from the beach, and we could not see how large the devastated area was. The fields and roads completely sank into the sea. About one thousand people drowned in the waves, because they failed to escape either offshore or uphill from the waves. The properties and crop seedlings were almost completely washed away.

Because this earthquake was accompanied by a tsunami, they can actually go through the deposits of that tsunami, and work out how far the flooding extended — that’s the main measurement that was used to reconstruct the Richter magnitude of this earthquake.

There have almost certainly been more powerful earthquakes in human history. Generally, though, it’s agreed that the most powerful recorded earthquake was the 1960 Valdiva earthquake in Chile, which was 9.3–9.5 on the Richter scale. I hadn’t heard of this earthquake until I came to research the show, and the reason is essentially because other earthquakes have proved more fatal, if lower in magnitude. The Valdiva earthquake was actually triggered by a sequence of previous earthquakes, of magnitude 8 or so, that acted as a foreshock triggering the main earthquake. This was powerful enough to permanently distort the landscape in the region — and eyewitnesses reported seeing, due to the changing water table, water rising up through the soil. It triggered landslides, and a tsunami with waves up to 25m tall battering the Chilean coast. And, a rather grisly detail — to ward off future earthquakes, a local shaman in a coastal village ordered a human sacrifice to ward off future earthquakes. The victim was a five-year-old boy. Amazingly, the men involved were charged with murder, but released from prison after two years.

This earthquake was especially bad in terms of magnitude, but it’s nowhere near the costliest in terms of property damage — the earthquake and tsunami you probably remember from Japan in 2011 is the costliest — or human life — that dubious honour goes to the Shaanxi earthquake in 1556 in China. Chinese casualty figures from the Imperial period are often sources of historical controversy, but hundreds of thousands of people certainly died in this earthquake.

Magnitude 9.1–3 earthquakes in Japan in 2011, and in the Indian Ocean in 2004, loom large in the memory as amongst the most devastating and deadliest of recent times. And there fears that a truly powerful earthquake, with magnitude 9 or above, might one day strike a major population centre — in Japan or California. Even smaller earthquakes that are placed in vulnerable spots, like Haiti, can cause massive disruption.

Predicting earthquakes is a very inexact science. Indeed, the 2011 Tohoku earthquake in Japan was predicted and the public were warned — by around a minute. That’s how difficult it is to say anything for certain. The issue arises because, although earthquakes are often preceded by increased seismic activity — just as often you get seismic activity that occurs and then dies away. And not all earthquakes are accompanied by “foreshocks” that are noticeable or beyond the norm. If you analyse the seismic patterns, you see that the little cluster of foreshocks that precede some major earthquakes occurs far more regularly than the major earthquakes do. Ordering evacuation of civilians is a risky business; if you did it every time there was some increased, albeit low, likelihood of earthquake — it becomes ineffective, people stop listening. It’s very difficult to get first-hand data on earthquakes and seismology; you might imagine that it’d be possible to measure if all of this elastic strain energy is really building up between two huge tectonic plates, but, quite simply, we have never sent probes down that far.

The deepest we’ve ever gone into the Earth’s crust is a bit of a bizarre relic of the Cold War. Back when the US and the USSR were competing over everything — who could be first into outer space, into the moon, etc. — the USSR decided, in a Guiness Book of Records type of way, to dig the deepest ever hole. So they did; it’s a hole called the Kola Superdeep Borehole, and it goes 12.3km into Earth’s crust. Although, if you’re a mafia type, it’s no good for evidence disposal, because it’s only nine inches in diameter. The issue is that the heat and pressure gets ridiculous down there; their goal was 15km, but that would have been hot enough at 300 celsius to melt the drill-bit they were using. The project was shut down in 2006 due to lack of funding, and now… the world’s deepest hole is just abandoned in the far northwest of Russia, near the Finnish border. Which is sad. And it just goes to show how far we are from having perfect measurements; because the distance to the centre of the earth is 6,400km and we’ve travelled a mere 12.3km of that distance.

Pretty much every major attempt at Earthquake prediction has been met by scepticism or failure; in the 1970s, it was believed that we’d be able to predict them, but people are far less certain now that it can be achieved. This does come after a number of high-profile failures, and a few cases where people retroactively claimed that they’d managed to predict earthquakes after they occurred. The issue with these cases is that people have an algorithm based on foreshocks and seismic activity. An earthquake happens, and it’s consistent with when the algorithm said an earthquake was likely to happen: take that in isolation, and it’s a prediction. But this discounts all the times the algorithm predicts earthquakes that failed to materialise. Even if we could accurately measure the strain build-up in areas under the earth’s crust, there’s still a degree of randomness to the event. A foreshock might trigger a bigger earthquake… or it might not. A certain amount of strain might lead to an earthquake… or it might not. So all we’ll ever be able to do, really, is make probabilistic statements. And that’s fine. Anyone who phrases a prediction about the future not in terms of probabilistic statements must have access to some knowledge that the rest of us don’t. The difference is when probabilistic statements are useful, and when they’re almost indistinguishable from noise.

So why is it that people say that, for example, San Francisco is “due” a big one? Or that there should be an earthquake every 20,000 years in such-and-such a place? This is the distinction between prediction, which refers to a specific event, and forecasting, which refers to the distribution of events overall. And it actually turns out that most earthquake regions obey a rather beautiful law. If you take the logarithm of the earthquake power — which, as we mentioned, is the magnitude — and plot against the logarithm of the frequency, you get a straight line in most cases, to pretty good agreement. In other words, by measuring the frequency of small, low-power earthquakes in a region, you can get an estimate for the average frequency of big earthquakes. It might tell you that a magnitude 7 earthquake should occur every hundred years, or a magnitude 9 every ten thousand years on average. But just because you’ve waited 10,000 years since the last one, doesn’t mean there’s one ‘due’ necessarily. It’s just a statement about the average frequency at which things occur. If you wait a billion years, you’ll find that the average gap between those earthquakes was 10,000 years, in the same way as flipping a coin enough times will give you fifty-fifty heads and tails, but that doesn’t mean you couldn’t have a 100,000 year period with no earthquakes at all. We can make statements about places that are likely to have earthquakes, and how often earthquakes of a certain size are likely to occur.

So the real reason that earthquakes are number ten on my list is that, while they can be devastating, they tend to be locally devastating. And there seem to be limits to how powerful an earthquake can reasonably be. For example, a magnitude 12 earthquake on the Richter scale would probably require a fault-line bigger than the entire circumference of the Earth. It can’t happen. And in fact, most researchers think the limit is a lot lower — maybe around a 9.6–10. The reasoning behind this is simple: there is a limit to how much pressure the rocks can be under before they simply break. That’s what earthquakes are: a pressure release valve. And the amount of pressure that can build up depends on the fault-line — the width, and the size of the slip that can occur. We can analyse fault lines and predict the maximum magnitude they might support — so the San Andreas fault is unlikely to unleash an earthquake of magnitude greater than 8.6, and there aren’t many faultlines that are large enough to support an earthquake of 9.6–10. The Cascadia faultline in the pacific northwest, which is fairly close to Seattle, is more likely to support an earthquake of magnitude 9; one occurred in 1700, and they’re predicted to occur on average every 243 years. There’s a rather famous New Yorker article about the consequences of such a disaster, which the Pacific Northwest is far less prepared for than the San Andreas earthquake because the fault-line is less famous. Here’s a quote from that brilliant article, via Kathryn Schulz:

“The Pacific Northwest has no early-warning system. When the Cascadia earthquake begins, there will be, instead, a cacophony of barking dogs and a long, suspended, what-was-that moment before the surface waves arrive. Surface waves are slower, lower-frequency waves that move the ground both up and down and side to side: the shaking, starting in earnest.

Soon after that shaking begins, the electrical grid will fail, likely everywhere west of the Cascades and possibly well beyond. If it happens at night, the ensuing catastrophe will unfold in darkness. In theory, those who are at home when it hits should be safest; it is easy and relatively inexpensive to seismically safeguard a private dwelling. But, lulled into nonchalance by their seemingly benign environment, most people in the Pacific Northwest have not done so. That nonchalance will shatter instantly. So will everything made of glass. Refrigerators will walk out of kitchens, unplugging themselves and toppling over. Water heaters will fall and smash interior gas lines. Houses that are not bolted to their foundations will slide off — or, rather, they will stay put, obeying inertia, while the foundations, together with the rest of the Northwest, jolt westward. Unmoored on the undulating ground, the homes will begin to collapse.”

This sounds terrifying, but as horrible as it would be, even this is starting to strain at the upper limits, and it’s not yet a global catastrophe. We don’t know of any scenarios that would allow the rock to magically withstand enough strain to build up the energy release in a magnitude 10 or 11 earthquake, without releasing it via a smaller earthquake first. As far as we know, it can’t happen.

For a truly apocalyptic scenario, one would have to imagine lots of earthquakes simultaneously occurring, along all of the fault-lines, in very quick succession. The reality is that Earth is made of very solid, very inelastic rock; the vibrations that travel through it can only propagate so far before they lose all of their energy and dissipate. So earthquakes might be able to trigger aftershocks for faultlines within a few hundred miles of the original location. But it seems unlikely that, say, a really powerful earthquake in Chile could have a reasonably large effect on the fault-line in Japan. Without that kind of long-distance influence, the really apocalyptic scenario would have to be several magnitude 9.5 earthquakes occurring together practically by coincidence. And that seems really, really unlikely. At least, not with the plate tectonics that we know at the moment. What we know of the Earth’s core and mantle is limited, however: I’m sure seismologists and geologists could come up with catastrophic, if unlikely scenarios, where entire plates suddenly move due to a massive shift in the currents of Earth’s mantle. But it would be unlike anything we’ve ever seen.

However, one of the more recent deadly earthquakes has illustrated that — in a globalised world — you can have chains of events that might be more deadly than the initial earthquake itself. I’m talking, of course, about the Fukushima power plant meltdown that had the world on edge back in 2011. I remember following it live on the news, as I’m sure many of you do — the efforts to inundate the reactor core with salt water, and worrying about all those brave first responders getting irradiated, and the idea that — you never know — something catastrophic could occur at any minute. But if your scenario for earthquakes causing the end of the world relies on them triggering something else, like accidentally causing a nuclear meltdown, then it’s debatable whether that’s really an argument for us to be afraid of earthquakes, or for us to be afraid of nuclear power. And *even if Fukushima had somehow detonated* the effects would still hardly be in the realm of the kind of apocalypse we’re looking for. So, for that reason, earthquakes scrape onto the list, but they won’t cause me to lose as much sleep as the next nine entries.

Thanks for listening to this TEOTWAWKI episode of Physical Attraction.