NOVA

So a magnitude 7 earthquake releases about 30 times the energy of a World War II era atomic explosion.

And we have about 20 magnitude 7 earthquakes on average that occur around the globe every year.

Holy cow.

A magnitude 9 earthquake releases 1,000 times more energy.

Those earthquakes are so large that the Earth rings like a bell for days.

Days?

Days.

And in fact, they actually change the rate of rotation of the Earth.

They actually change in a measurable way, not a lot, but in a measurable way, the length of the day and the length of the night.

Professor John Shaw, welcome to Particles of Thought.

It's wonderful to be here.

Yeah, I'm so happy to have you, man.

So humans cause earthquakes.

Now I know we're not talking about like Jurassic Park, right, where the dinosaur walks in and they're so heavy.

We're talking about, we're doing subsurface things or other things that cause actual real earthquakes.

Tell me more.

I thought it was a myth for a while.

Yes, my goodness.

We, we as a society have the ability to generate earthquakes.

We've done that through much of human history, but we're doing a lot more of it today for reasons that relate to the way we produce energy in the subsurface in particular.

So historically, one of the ways that humans have generated earthquakes is by building dams and building reservoirs, bodies of water behind those dams.

It's almost guaranteed that anywhere you build a very large dam and a reservoir, you're actually going to generate earthquakes.

And that's driven by 2 processes.

One is the weight of the water.

You're simply putting more force downward on the earth in that area, and that can actually cause faults that are close to rupture, close to failure, to sort of push them over the edge.

The other effect is that the water itself that's in that reservoir can trickle down into the subsurface.

It can increase the pressure of the water that's down in the earth, and that can have an effect of producing earthquakes.

Wow.

The way we're doing a lot more of this today, causing earthquakes— there's terms "induced" and "triggered" earthquakes, but they're both basically flavors of earthquakes, in a sense, sort of initiated by humans is by either producing large volumes of fluid out of the subsurface or more commonly injecting large volumes of fluid into the subsurface.

And that's a process that can destabilize faults in ways that generate earthquakes.

And this can occur in earthquake-prone regions, but it could also, we can generate earthquakes in areas that don't have many, if any, tectonic earthquakes by these subsurface operations.

So you mentioned both injecting fluids and removing fluids.

So the injection part is the thing that when it first— I first heard about it in the context of fracking causing earthquakes, I thought, oh, certainly not.

Is that real?

Could that be— could we do that?

But now that you mentioned extracting fluid, is it the case, you know, we've been extracting oil for a long time and there have been wells for water for a long time.

Have they been producing earthquakes?

In some cases, but not that often.

Got it.

And that's because when you actually produce fluids from the subsurface, what actually causes earthquakes in those circumstances is the rocks from which that fluid is produced.

You can sort of think about it a bit like a sponge.

If you move the fluid out of that, the shape of the sponge may change.

And as a result, it may cause changes in forces that could, for a given fault, cause it to generate an earthquake.

However, the injection of fluids is the way we produce earthquakes much more commonly.

I see.

And that's because while the injection of fluids also can change the shape of the sponge, the rock volume, causing there to be forces that could trigger earthquakes, the fluids themselves, the increased fluid pressure, is a key way in which you weaken faults.

That fluid that's injected actually unclamps the faults in ways that can actually make them easier to slip, to slide, to move, and to generate earthquakes.

So here's the question then.

So if we just go around Earth injecting fluid everywhere, we could just relieve all that pressure and we'll be done with earthquakes.

Okay.

No.

In concept, perhaps.

Interestingly, there was a study, a very informative study, probably one that can never be repeated again, in the 1970s, in an area called Rangely, Colorado.

Mm.

There was an oil field that was being developed by a large oil company, still operating today, and they were producing oil, but they were also producing water.

When you produce oil from the subsurface, it's typically not pure oil, there's also water.

And you often have to re-inject that water.

I see.

It's contaminated water.

You wouldn't wanna dump it into a surface body of water.

So you return it back to where it came from.

You wanna return it back to where it came from.

That makes sense.

But you often have wells that inject the fluid that are separate from the wells that produce it.

So you're not always putting it back exactly where you got it from.

And if you inject it in a place where you're actually gonna elevate that fluid pressure, right?

You hadn't already produced a lot of fluids out of that exact location.

The elevation of that fluid pressure is what can destabilize a fault.

And so this, in this field, they were actually generating small earthquakes and they wanted to understand this process.

Of course, everyone wanted to avoid generating a large earthquake.

There had been some activities in this general area of the United States earlier related to wastewater injection from weapons manufacture that had generated earthquakes.

At what depth, out of curiosity?

These earthquakes that are occurring in this region of this field were only at depths of, you know, 3 or 4 kilometers.

Okay.

So a couple of miles.

Yeah.

Not very deep, not as deep as the largest tectonic earthquakes that occur that cause ground shaking.

But the opportunity to understand the physics of earthquakes was present.

And so what they actually did is they partnered with people from the U.S.

Geological Survey who came out and they installed a network of seismometers to record these very small microearthquakes.

And they actually worked with the operators of the field to change the volume and the rate at which they were injecting fluids, and they were able to turn the earthquakes on and off.

What?

Literally turn them on and off.

That sounds like useful information.

Critically, because what it allows you to do is really understand the process that triggers an earthquake to occur, what allows maybe an earthquake that's a small one to become a large one, how specifically the faults are stressed in the natural setting and how you're perturbing that stress or those forces in the subsurface by whatever you're doing, how much fluid you're injecting, under what pressure to actually cause earthquakes to occur.

So is it a case that they can characterize a location, find out its threshold, and then operate underneath that?

Exactly right.

And so that's a learning that has moved through decades in ways where we can actually, with proper science and proper physics, we can go to lots of different parts of the world where there is a risk of inducing earthquakes, and we can try to essentially moderate the operations, plan the operations of fluid injection in ways that will reduce the likelihood and the frequency of earthquakes that may occur.

And that's to everyone's benefit.

So can we learn about the natural earthquakes from these human-created earthquakes somehow?

Absolutely.

They are like our perfect laboratories in some sense, right, to study the tectonic earthquake process, because we simply know so much more about the subsurface in the areas where these induced earthquakes have occurred.

And we also know precisely what we did, right, to tip the balance, to actually generate the earthquake.

Yeah.

By whatever this, let's say, fluid injection operation may have been.

So again, one of the things that we don't know in a tectonic regime very well is the actual magnitude of the forces that are acting on the faults that are there.

In an environment— and how much more force is going to be required basically by motions of tectonic plates over long time periods to actually generate the earthquake on that fault.

In an area where we're inducing or triggering earthquakes through human activity, We actually have a lot of wells typically in that area already.

We may have made some measurements about what the state of stress or the forces were in the earth.

And we know sort of how much we've added to that system to get it to rupture.

Right.

That case I referenced before where they were actually able to turn on and off the earthquakes gave them a very precise measure of what it took to actually cause those earthquakes in a way that you could simply never do in a natural system.

But the knowledge transfer is there because it tells us, it confirms our knowledge of the strength of the faults, of how much the forces are acting on them before humans have an effect on them.

And again, what those added amounts of stress or force are that often require or cause that tipping, that initiation of an earthquake, whether it be natural or human-induced.

Particles of Thought is supported by Carlisle Companies, a manufacturer of innovative building envelope systems.

With buildings responsible for over a third of total energy use, Carlisle's energy-efficient solutions are built to reduce strain on the energy grid.

For example, Carlisle's UltraTouch Denim insulation, made from sustainable recycled cotton fibers, delivers energy efficiency while being safe to handle and easy to install.

Made with 80% recycled denim, UltraTouch diverts nearly 20 million pounds of textile waste from landfills each year.

Operating across North America, Carlisle is working towards a more sustainable future.

Learn more at carlisle.

com.

So let's do a little groundwork, pun intended, and lay down what exactly is an earthquake and, you know, if they come in different types and what characterizes each type.

Absolutely.

And maybe to go back a bit in human history, we have only developed our current modern understanding of what an earthquake is really in the last century.

If you go back to the time of Aristotle, people used to think earthquakes were the result of gases that were moving around in caverns within the Earth's interior.

Oh, wow.

And it wasn't until the late 1800s, the early 1900s, that people recognized that the cause of earthquakes was actually motion on faults or fractures within the Earth.

Yeah.

Where there was sliding of one block relative to the other.

Right.

And it was that process essentially that radiated the seismic waves that produced the ground shaking that we consider an earthquake.

Wow, that is amazing, man.

Here humans have been living with civilization for literally thousands of years going back to ancient Mesopotamia and Egypt and these societies, and it took us until 100 years ago, just over, to put that story together.

That is amazing.

So in terms of the different types of faults, so I know the one on the Pacific Coast, they're moving laterally relative to each other.

And I remember the big Indonesia earthquake, there was a part of the ocean surface— excuse me, ocean bottom that lifted, right?

So what are the motions associated with different types of earthquakes?

We learn about these faults as sources of earthquakes principally by looking at maps of tectonic plates and looking at these boundaries between tectonic plates.

And those are the regions of earthquakes, the Ring of Fire in the Pacific, for example.

Right, right.

But when you actually zoom in to these areas, it's not uncommon that a single plate boundary can have something like 500 faults that are actually accommodating the relative motion between those plates.

Wow.

And within that large inventory of hundreds of faults, there are different types.

Some of those faults, as you mentioned, are what we call strike-slip faults.

the blocks on either side slide laterally relative to one another.

The case of the Sumatra earthquake that you mentioned, that's a thrust or reverse fault where the fault itself is not vertical, it's inclined or dipping.

And the block that's above that fault actually moves up during the earthquake.

In the case of the Sumatra earthquake, the seafloor rose about 25 meters or 75 feet.

Wow.

During that single earthquake.

And then finally, there's a type of fault called normal faults.

There are also cases where the fault is inclined, not vertical, but dipping, and the block that's above the fault actually moves down during an earthquake or a slip event.

And so all of those different types of faults are common, most common in certain tectonic settings, certain types of plate boundaries.

So let me understand the magnitude scale.

We say this all the time.

This is another one of those things that everybody says it, but we really don't know, right?

So I know magnitude 7, bad.

Yeah.

I know it's gotten really bad then.

5, I'm not, I don't really know.

So what does it mean?

Like if you can have a negative, it's sort of like star brightness, right?

So we have a system called magnitudes for star brightness.

And when they first created it, they were like, oh, these are the brightest stars, so the brightest one is gonna be magnitude 0.

And then we get instruments that can make these measurements very precisely, and we learn, oh no, the brightest stars are actually negative magnitudes, they're so bright, right?

So how do you set your zero point?

And then what do the numbers mean?

Right, this is exactly, of course, what has happened historically in earthquake science.

So the Richter magnitude that most people think about when they think about the scale or size of an earthquake has actually been out of date for nearly 75 years.

What it did was it said, we're going to measure one component of the ground shaking that comes in, the amplitude of a specific wave at a set distance from where the earthquake occurred in a seismometer.

And it was a specific type of seismometer that, again, we haven't been using for decades and decades and decades.

Ah.

Okay?

It was very limiting to us as a field because it got what's called saturated in the high end.

You know, the ability to tell a big earthquake from a really big earthquake was very limited.

And so it's been replaced by something called the moment magnitude scale, which is actually a measure of the energy that's released in an earthquake.

And we've modified it in a way that it sort of looks like the Richter magnitude because people got familiar with that intuition about a 5 being a moderate earthquake and a 7 being a bigger earthquake.

Yeah.

And just for reference in terms of the energy released in an earthquake, so a magnitude 7 earthquake releases about 30 times the energy of a World War II era atomic explosion.

Wow.

And we have about 20 magnitude 7 earthquakes on average that occur around the globe every year.

Holy cow.

A magnitude 9 earthquake releases 1,000 times more energy.

Those earthquakes are so large that the Earth rings like a bell for days.

Days?

Days.

Wow.

And in fact, they actually change the rate of rotation of the Earth.

They actually, change in a measurable way, not a lot, but in a measurable way, the length of the day and the length of the night.

Longer or shorter?

Yeah, it depends on the earthquake and it depends on where it is and the type of fault.

But even the concept that these earthquakes can release that much energy is really staggering.

So what— what else can we do as regards predicting earthquakes?

So can you make a prediction prior to any waves propagating?

Well, you know, earthquake prediction, of course, is a challenge.

We really don't have a functional capability to do that as much as we want it.

And maybe if I could just take a moment to contrast this with other natural disasters.

Think about hurricanes, for example, Atlantic hurricanes.

We can detect where they form, in many cases off of Western Africa.

Yep.

We can track them.

We can measure their intensity.

We make maps now that actually model where they might go and the probabilities associated with that.

We lack that ability essentially in weeks and days and hours and minutes to actually forecast where or predict where earthquakes will occur.

Yeah.

There's reasons for that that we can discuss, but part of it is that we simply can't observe things in the subsurface, in the ground, with the same fidelity that we can observe things in the atmosphere.

So in the past, we're in a situation now where we're doing more and more observations, right?

Seismographs being put all over the world.

And one thing about understanding patterns in data is that the more data you have, the statistics build up and help you out.

So how far back do we have data?

And can we learn anything from the, like you mentioned, the 1906 earthquake?

Did we have seismographs then?

And I'm assuming earlier than that, there were a lot of large earthquakes.

And added to that question is, is it a data problem?

If we had a whole bunch, if we had seismograph data for 1,000 years, could we put that into an AI, a machine learning algorithm of some sort?

Mm-hmm.

—be able to predict earthquakes from that day forward?

We would certainly do a lot better job of it than we do now.

Okay.

Part of the problem is exactly what you point out, is that many faults— when you think about a fault, you know, that fault may have a clock that's ticking down to determine when it generates its next earthquake.

And many faults only generate an earthquake every 1,000 years or 2,000 years.

And so we have about 100 years of recorded earthquakes through seismometers, the devices that we deploy.

And in fairness, we really only have complete catalogs that are about 50 years old.

Ah, okay.

In fact, one of the things that helped us globally build a network of seismometers was actually nuclear test ban verification, sort of trying to detect nuclear detonations and ensure that people were being consistent with agreements, of course, international agreements that have been struck, because you can use these same networks to detect nuclear accidents, nuclear explosions.

And so that actually helped us to densify the network.

And in more recent decades, we've been going into earthquake-prone regions and densifying the local networks to help us understand the faulting.

But what that means is for most faults, you only have maybe, you know, 10%, 5%, 10% of their earthquake sort of cycle recorded in those observations.

Ah.

As you say, if we had thousands to tens of thousands of years— Right.

—of recordings.

Yes.

They would tell us much more about each fault, even the slower moving faults, right?

Whereas most of our data today is dominated only by records we make from the fastest moving faults or the ones that just happened to go off that generated an earthquake during this relatively short span of observation.

So we're in the year 2026.

So that means in the year 1,000, 200— wait, 1, 2, 0, 2, 6, 12,026.

We'll have that data.

I'd like to not wait that long.

And that's obviously what is a big part of the motivation for earthquakes.

You mentioned artificial intelligence and machine learning.

Of course, like so many fields, this is one that is— this is a set of techniques that's helping us advance our understanding of earthquakes.

One of the things that's been most profound in that space for us has been we can actually use these techniques to detect much smaller earthquakes than we ever were able to before.

So typically with these traditional networks of seismometers, we can maybe detect earthquakes down to maybe magnitude 2.5 or something like that, or 3, with a degree of completeness.

But these new machine learning and artificial intelligence techniques that have been pioneered by a number of investigators around the world actually allow us to detect earthquakes that are as small as like 0 magnitude, in some cases negative magnitude.

They're literally— faults are the size of centimeter-scale slip events.

And it turns out that these earthquakes are happening all the time.

We now measure one of these earthquakes on average occurring every few minutes in the state of California.

Wow.

And what they do is they enrich this database that we need to understand the network of active faults in a given region.

Because now we're seeing between the large earthquakes that we're focused on in terms of their potential destruction, we're actually seeing these faults that we may not have known about before, at least starting to show signals of their presence and their activity in ways that we can learn about them.

And then again, better forecast.

Not only are they capable of generating larger earthquakes, but are they affecting the faults that are around them that may also be capable of generating larger earthquakes?

Wow.

Wow.

So here we are in Boston.

So I don't think of Boston as an earthquake-prone area, but there is a big historic earthquake here, right?

Yes, the 1755 Cape Ann earthquake actually occurred about 10 kilometers to 15 kilometers east of Boston, offshore.

It was felt basically from Nova Scotia to Washington, D.C.

Wow.

Earthquake waves in the East Coast tend to travel a little bit farther than they do in like the West Coast or other earthquake-prone regions, 'cause our rocks here are relatively old and relatively cold.

The earthquake actually did a considerable amount of damage.

It knocked a number of stone and sort of masonry structures down in Boston.

A lot of things sort of fell down as a result of that.

And so that earthquake, of course, does signal that while we're not in a Ring of Fire type setting, we don't have earthquakes that occur frequently, we do have earthquakes.

And, you know, it just reminds us that if there's simple things that we can do to secure the environment that we live in, it's worth doing those things because again, an earthquake could occur in this region with, again, without much warning.

And if we're better prepared for it, obviously it will be to all of our interests.

So there was another earthquake that year, right?

In Europe.

There was an earthquake in Lisbon, Portugal, actually that occurred.

Now, remember at that time, the prevailing view of earthquakes was that they were sort of acts of a deity.

Right.

Yeah.

Right.

And they often were a reflection of how well or poorly people were behaving.

Right.

And so they were— Moral decay brings upon earthquakes.

Exactly right.

Okay.

And interestingly, the Great Lisbon Earthquake, which had a huge impact on the city of Lisbon and that part of Western Europe, was a much larger earthquake than the Cape Ann Earthquake.

It actually occurred before the Cape Ann Earthquake, but we didn't know about it yet because it took months for ships to get across the Atlantic to actually provide that information.

And so, you know, when the earthquake occurred in Cape Ann, many of the people writing about it, of course, took that approach that this reflected misbehavior on our part.

Right.

And this was sort of a response to that.

Someone at Harvard, John Winthrop, who obviously did indicate that the divine power had ultimate control over what occurred and where, But also started to think about it a little bit more scientifically.

What was it that actually generated this earthquake?

What forces were acting in the earth?

And how might we actually learn about those in ways that we could better understand earthquake phenomena and maybe even in the long term try to forecast earthquakes?

Is there a particular event that we wish we could have recorded with modern technology?

And if so, what would we have learned from it?

There's so much that we would like to understand better and do so through informed observations, informed data.

One of the things, for example, we've rarely, if ever, actually observed is the rupture of a fault system at the surface, actually progressing along the trace of the fault.

We see the ground motion, we see the waves, we can go out after the earthquake and see that the ground has ruptured.

Right.

But actually observing the rupture occurring during the earthquake is something that we've seen in a few places, glimpses of, but seeing it at the scale, the full scale of a large earthquake, would of course be very valuable to us because it would tell us how fast is that fault rupturing.

Wow.

And, you know, that will control to an extent how some of the waves are radiating from it.

I think the other thing we'd really just like to understand better is the stresses or the forces acting deep in the Earth.

I think that's our, in my view at least, it's our main limitation right now in terms of trying to predict and forecast earthquakes.

Imagine trying to predict wind in the atmosphere without knowing the pressure gradients in the atmosphere, it would be a near impossible thing to do.

We're sort of closer to that situation in the subsurface because of our inability to measure stress and forces very accurately.

So is there, are there any scenarios where you guys actually dig boreholes or something to make measurements down there?

Because when I think of seismographs, I think of just surface detectors, not any, you know, not "let's go into that fault a kilometer beneath the surface."

Is that a thing?

Absolutely.

So one of the things that we— we can infer something about forces acting deep in the earth from just observing the earthquakes themselves.

That's an important way that we do it.

But we can also put instruments down boreholes.

So drill a well, in some cases kilometers deep in the earth, and you can actually put devices down there that actually measure the forces or the stresses that are acting on those rocks.

In some cases, the way you do this is you actually drill down in the well, in the earth, and you actually make a series of observations of how the waves, sound waves, travel through that rock.

And then you actually take a piece of that rock, you bring it up to the surface, you bring it in a laboratory, and you try to replicate the same behavior—Oh.

By squeezing the rock in the laboratory and getting it to the point where the forces in the laboratory are producing the same behavior that was observed in the natural setting in the subsurface.

And that tells you what those forces must have been in the subsurface.

Are there degeneracies?

Is it like typically a single force answer or do you get multiple possibilities that fit the data?

There's multiple components of force that you can resolve.

And there are uncertainties.

So you sort of get, you would never want to base your forecasting or understanding on a single measurement.

You would use large numbers of these.

But the challenge is they're expensive to acquire.

To drill these wells is very expensive, to run these types of experiments is also expensive.

And so most of where— ironically, most of where we have some of that information right now is in areas where energy companies are drilling very deep wells and are willing to run these types of experiments.

A geothermal field, an oil and gas field, because of course they have the financial incentive, right, to drill those wells.

Right, they're doing it anyway.

They're making money on those activities.

And so opportunities that we have to partner with them to acquire some of this type of information is incredibly valuable to us.

So one thing, so when I was in college, sometime in the late '80s, early, it was the late '80s, like '88 I'm thinking, there was a prediction that the tectonic plate, or not— the fault in Missouri that had caused the Mississippi River to run upstream was gonna have a major earthquake.

So we were evacuated from the school.

Mm-hmm.

Turns out it did not occur.

Mm-hmm.

So is it the case now that if you hear such a prediction, the probability of it being correct is really low?

Or, you know, if you're near, for example, a volcano, and they warn you, oh, these earthquakes mean the volcano's gonna erupt soon, you should evacuate, you should listen to that, right?

There's a lot of precedent there.

But I'm not aware of any precedent for earthquake prediction other than this 1-minute type of prediction that we know one has occurred, the surface waves are coming.

So is there any such of a, you know, or you can, low probability of being correct on such a prediction?

It depends on what data, what information such a forecast or prediction is based.

In many cases, of course, you know, we don't have the ability to predict earthquakes like we do weather.

Right.

And so many of those types of predictions simply don't pan out.

What we are able to do as a society well, is actually predict earthquakes based on observed sequences of earthquakes.

So for example, when we have an earthquake occur, maybe a large, moderate or large earthquake, we have very good ways of predicting the number and the frequency of aftershocks that will occur.

Oh, interesting.

And so those are the types of warnings that people should really take to heart.

Recognize that if even after a large earthquake, the system is probably not done yet, there's likely to be more earthquakes.

And oftentimes those aftershocks are things one has to take very seriously.

Oh, absolutely.

Because your building may have already been damaged during a large earthquake and less strong than it was.

So it's weakened and the aftershock comes and knocks it down on top of you.

And so listening to those types of forecasts, I think, is critically important.

Yeah.

I think in terms of the longer-term earthquake forecasting, that's also an area where we're making great progress.

If you live in an earthquake-prone region, in a region where there are specific faults that are perhaps long overdue from generating an earthquake, that's something that should motivate you not to evacuate your house or evacuate your building, but to think about the built environment around you, and try to take simple steps to secure it so that when that earthquake eventually does occur— Right.

You're obviously going to hopefully be impacted by it less.

One of the things that is, you know, often most heartening after a large earthquake is earthquake engineers go around and they do a census of all the buildings essentially that have been damaged in the earthquake, the intensity of damage.

And the frustrating thing is you pretty much know the answer even before you go do the survey.

There are certain classes of buildings and structures that are simply much more prone to damage during earthquakes than others.

Right.

And so that's something at a high level that we need to think about.

That's why we have building codes in earthquake-prone regions.

And one should pay attention to those.

Where do I live?

Am I in an area where earthquakes may occur or may occur frequently or strong earthquakes may occur?

And what type of building do I live in?

Certain types of buildings, for example, wood-frame buildings, steel-frame buildings, are reasonably good in terms of the response to earthquakes because that building is able to flex, it's able to shake.

If you're in what's called an unreinforced masonry building, a brick building, a concrete building without rebar in the concrete, those tend to be brittle structures.

They tend to not fare as well during earthquakes.

They break rather than bend.

They break rather than bend.

And when things start to break, of course, you know, earthquakes themselves typically don't kill or harm people.

It's the actual built environment that collapses, that falls on top of them.

Yeah.

There's also a set of sort of things that we can do at an even smaller level.

Think about an area in your home that would be more secure during an earthquake.

And what that means is often sort of a door jamb, an area where you can get to quickly.

Those parts of buildings, homes are often better built, they're stronger, and so even though other parts of the building may become more damaged, those are often going to survive, or at least they have a better chance of that.

Yeah.

There are also areas where you can avoid things falling on top of you, and that's not a minor issue.

You know, a lot of damage and injuries that occur during earthquakes are because people's bookcases fall down, things fall off the walls, and things fall essentially off the ceilings that can do damage.

And so, thinking about where you can be to be safe and maybe also securing some of those elements as best you can is a way as an individual you can try to protect yourself and the people you care about in, in terms of the damage from future earthquakes.

So when we have a magnitude 7 earthquake, we can with some degree of confidence indicate the number of smaller earthquakes, the frequency of those earthquakes that are going to occur as aftershocks.

And aftershocks, some of them can be large in their own right, And you need to obviously be prepared for those.

Often also the built environment you're building, or the bridge, has been damaged by the big earthquake, and so it might not be as strong as it was prior to that big earthquake.

Right.

And so even a smaller aftershock could compromise it, could cause damage.

Right.

So one way, one thing to pay close attention to is aftershocks.

The other sort of category, though, to consider is, as you say, how do you actually live and work in an area that's been impacted by earthquakes?

Yeah.

You know, we tend to think about what will happen in terms of having to rebuild buildings and infrastructure.

Right.

But we are so dependent these days on what we'll call lifelines.

So energy transmission structures, electric lines, natural gas lines, telecommunications systems.

And of course, those can be damaged by earthquakes.

Those can be damaged by earthquakes that are local, that are experiencing— those systems are experiencing the same ground shaking you are.

But you could also actually have a more distant earthquake that ruptures across a telecommunications system or a gas pipeline or an electric infrastructure that actually could pose a problem for you.

So you have to start to think in a modern world of, given your dependencies on all these things, how an earthquake that isn't even, you know, right next to home could actually impact you in ways that you'd be prepared for it.

In many ways, the same way you would prepare for a hurricane, right?

I remember seeing a story that said the buildings in Las Vegas are built into these X shapes and these crossing perpendicular shapes to protect them from earthquakes.

So are there, I know there's things you can do at the base of the building, and then I guess that's some sort of structural reinforcement.

How many, what do builders have in their toolbox to overcome— these challenges and then the challenge of earthquakes?

And then is there something that goes from the single building level to the community level that can be implemented?

Yeah.

Fortunately, we have a lot of tools in the chest.

You know, earthquake engineers, structural engineers are many, in many cases, very good at their practice.

There's been a tremendous amount of research in this space.

We know how to build hardened buildings relatively well.

The challenge is it costs a lot to do it.

It costs more than to build a building maybe by the standard means.

And so one has to really think about when and where that investment is warranted.

Sometimes it's motivated by building codes that don't allow those buildings to be built unless they're built to a certain standard.

In other cases, it's simply builders and people who want to ultimately live and work in those buildings who place a high standard, right?

Right.

That if I'm going to be there, that building has to be up to snuff.

There's a lot of techniques to do this.

The simplest thing is shear bracing.

So when you have a building that is going to maybe shake side to side, if you think about it built in a sort of a very rectangular sort of post and beam structure, as soon as you start to shake that building, the joints between those verticals and horizontals are going to get a lot of stress.

Yeah.

They often crack, and ultimately that can lead to building failure.

If you put diagonal beams, right, across within that structure, it resists some of that shaking, it limits that shaking, and therefore it limits the cracking and the breaking that could occur.

So that's probably the first and foremost, the simplest engineering thing one can do, a design of a building, or you can go into an existing building and retrofit it and add those structures in ways that make it stronger.

That goes all the way up to things you mentioned like base isolation, where for large high-rise buildings in particular, you actually put structures underneath them that are going to allow them to shake a little bit, right, to free float during an earthquake so that they're not experiencing the full intensity of the way the ground is shaking.

Yeah.

The challenge— they're very expensive, but they're very good.

They're very sophisticated systems.

But you have to actually be able to predict the type of ground motion that's going to occur during the earthquake reasonably well to design those structures.

For example, It's one thing if a building is gonna shake side to side.

It's another if it's gonna move up at an acceleration that's greater than the gravitational acceleration and want to push things up in the air.

You've got to design different systems for those.

I remember seeing something about a building that had a big pendulum in the center or something along those lines.

So does that stop it from shaking at all?

It does.

It dampens— it doesn't stop it from shaking, but it dampens the shaking.

One of the things about buildings is that You had mentioned, you know, extending beyond just a single building.

How do you think of the environment?

Buildings of a certain design, let's say a steel frame construction building, they tend to resonate at a specific frequency, meaning they, they tend to be able to shake back and forth at a given frequency a lot.

Yeah.

And if the earthquake waves that are coming in are of the same sort of frequency, they can cause certain buildings to resonate more than others.

More and more.

Yeah.

And so oftentimes that resonance frequency is a function of the height of the building.

So literally you can have an earthquake that tends to shake or resonate the 20-story tall buildings but not the 10 or the 30.

Wow.

However, if you think about it, if you've got buildings that are next to each other, right next to each other, of different heights and they start to resonate differently, they also smash into one another and can do additional damage.

So you need to, you know, we hope we are starting to think as a society about a broader built environment, not just each individual building in isolation.

So is there a city on the planet that has done this, an earthquake-prone region that has made sure that they've built their buildings and they've illustrated that they can survive earthquakes without much damage?

There have been many cities in different parts of the world that have certainly made strides towards this.

No city is earthquake-proof at this point.

It's not clear that we will ever get to that stage.

So it might not be possible?

But we can do a lot to minimize the damage.

Yeah.

And where that is clearly manifest is looking at the same size earthquake that occurs in a region where those preparations, those investments have been made.

And then you look at that same earthquake occurring in a different part of the world where those efforts have not been made.

And oftentimes the loss of life is— the difference is staggering, right?

Yeah, absolutely.

You could have, you know, if you look at a city like Tokyo, Taipei, Taiwan.

Prepared.

You look at, to an extent, efforts that have been made in parts of Los Angeles, in San Francisco.

Those are areas where there will be damage and destruction from future large earthquakes, but it's gonna be far less, for example, than it may be in a nation or region without the same economic resources and commitment to invest.

There was an earthquake a little more than a decade ago in Haiti.

I remember.

You know, moderate magnitude earthquake.

I think it was a magnitude 7, you know, a large earthquake, but by far not the largest of earthquakes.

But, you know, the loss of life was simply staggering because the buildings were simply not built to these types of standards.

Yeah.

And unfortunately, of course, as soon as these buildings collapse, people need a place to live.

And so they're going to start to rebuild, like, with what they have.

It's hard to make that progress of advancing the designing of buildings unless you actually have the financial resources and the commitment to actually make that happen.

Yeah, yeah, that's a really tough one, man.

I mean, poverty really impacts what you can do.

So for research and monitoring, where do we need to put most of our effort into, you know, how we can do better surviving and beating earthquakes?

Right.

Well, earthquake early warning systems are something that is critically important.

And these are, again, in practice in parts of Japan, in parts of Taiwan, Southeast Asia, parts of the West Coast.

You can actually get an app on your phone from the state of California and you can be signed up for earthquake early warning.

And again, what earthquake early warning does is it detects the early stages of an earthquake occurring.

It does an analysis to figure out that this is probably a big earthquake, large enough that we should be concerned about it, and it sends out an early warning notification.

How long does that entire process take?

It depends on how far the fault is away from the urban environment.

I mean, if you're literally on top of the fault, these don't work very well.

You have matters of seconds, right?

Yeah.

But if you're in Los Angeles, California, tens of kilometers away from where the earthquake would occur on the San Andreas Fault, you may have a minute or something like that, or 90 seconds of early warning.

And while that's, again, not enough time often to get out of a building, it is enough time to get you into a secure environment within that building.

It's also enough time to start to think about things we could do in a more automated way to reduce these kinds of bad things that could happen as a result of the earthquake, like electric grids, you know, being ruptured and causing fires, gas pipelines breaking.

You know, the great earthquake, the Tohoku earthquake that occurred offshore Japan.

Right.

Their earthquake early warning systems actually recognized that earthquake was coming.

They started to shut down the Fukushima Daiichi nuclear reactor and other reactors, and that was of great value.

In that particular case, what they didn't anticipate was the size and scale of the tsunami that inundated the environment and didn't give them the ability to have power to continue to circulate the pumps to cool down the reactor.

But the concept worked very well.

So in terms of data, is there, you know, quite often I imagine that more data is typically good.

And so for my field in astrophysics, right, we deploy more telescopes or we have telescopes take way more data.

Then we have a problem that we have too much data.

So we have citizen scientists to, you know, analyze the data for us or with us.

And we also have citizen scientists who can deploy instruments for us.

Mm-hmm.

Do you guys have any plans of that nature where individuals can be proactive in contributing to the scientific enterprise of the professionals like yourself?

100%.

You know, data is still our lifeblood in terms of understanding earthquake phenomena, trying to develop basically ways in which we can better understand and forecast when earthquakes will occur and the type of damage that they will produce so that we can again protect ourselves as individuals and societies.

Again, we are a relatively data-rich field in the sense that we have had decades of instruments that have been monitoring earthquakes around the globe.

And, you know, we can learn a lot.

This is really an international effort.

So is there a central database where a person can deploy their own seismograph and submit their data?

Yes, functionally, yes.

And in fact, these— you can go online basically and access raw earthquake information, processed earthquake information.

And analyze the data yourself?

And start to analyze it yourself as a scientist or a researcher, as an individual, if you want to to do that.

That's an effort that really has been really profoundly helpful.

Again, the recognition comes from the fact that you can't just have your blinders on and study the fault that's right next to where you live.

Right.

Because it may not generate its earthquake for a long time.

There may be no data emanating from the fault or around the fault to really help you answer these challenging questions.

So this is why looking around the world, at earthquakes that occur in other places teaches you lessons about the fault that's next to you that you might be more concerned about.

And so that's why it has really been a very international community that's engaged in collecting data for earthquakes.

Is that how you got to that information that faults influence each other?

Absolutely.

So recognizing that when an earthquake occurs on one fault, we actually see other faults often in the region light up a little bit.

Their activity level changes.

Sometimes it changes, you know, in the minutes to hours after a large earthquake on one fault.

We start to see smaller earthquakes occur on other faults.

Sometimes it's something that we see on a little bit longer timescale, but it still tells us that there's communication between the faults.

Mm-hmm.

For example, we had an earthquake that occurred in California called the Landers earthquake.

It occurred in a region where the faults didn't generate earthquakes that often, but since that time we've actually had a series of other earthquakes on adjacent faults, including one in 1999 called the Hector Mine earthquake, that very likely was influenced— its clock was advanced by the fact that the 1994 Northridge earthquake occurred.

So that communication between faults is something that, again, you can only decipher if you're stepping back and looking at the larger system.

If you're just focused on your one fault, you'd never understand that communication.

You'd never understand those signals.

So this whole science of fracking is— or the engineering problem or technique— this technique of fracking has been around for a long time now.

And people had a lot of concerns.

And, you know, once it got into the headlines, but now it's sort of calmed down.

So is it the case that we're much more aware now, and so it's not as concerning as it used to be?

Or is it just something that just doesn't get the attention anymore?

Well, anytime humans are operating in the subsurface, producing fluids, injecting fluids, it's important to be very thoughtful.

And everyone should have a valuable perspective to offer in that space, the people who work in those environments, use that technology to produce energy, people who would be interested in, you know, using those techniques to actually develop renewable energies like geothermal.

Right.

Also, again, the people who live and work in those areas who have to obviously be subject essentially to those industrial-scale operations occurring in their backyard.

One of the big misunderstandings that I often encounter is the view that fracking is what's directly associated with the risks of generating earthquakes.

Again, in the majority of cases, the fracking is really just a technology that enables the production of the fluids from the subsurface, like natural gas.

Yeah.

It generates earthquakes, but they're often fairly small microearthquakes.

It's the enablement of, let's say, the natural gas production and the coproduced water that presents the problem of what to do with all that water.

And having to re-inject it is one of the primary ways in which we could generate earthquakes.

I'm more of an optimist in this space.

Right.

Because I think that given the knowledge that we're gaining about, you know, the state of stress in the subsurface, the forces acting, our knowledge of the fault systems in these regions, part of which has been learned by instances where we have actually generated earthquakes in the subsurface.

If we're proactive about this, we can have, you know, private-public partnerships that actually do the right kind of work to analyze these sites, figure out what are the safe ways in which we can maybe develop that resource, or maybe it's sequestering the CO2 in the subsurface in a way that will avoid these kinds of outcomes.

And most importantly, be a bit humble about this and recognize that we might not get it right from the beginning.

Well, I think I— as you were chatting, I think I figured out a solution to one of the problems, is what to do with that water that is typically re-injected.

We use it to cool the data centers.

Okay.

Unfortunately, it's hot.

It's the wrong side of the temperature spectrum.

It's usually—That's right.

It's coming out hot.

That's right.

Dagnabbit.

Yeah.

So we can't even use it for Elon Musk's data centers in space?

Well, people can use it for heating.

They actually use it for heat generation.

It's often not hot enough if it's coming from a natural gas field to actually produce steam and generate—Right.

Run a turbine, but it may be hot enough for other types of heating applications that could actually benefit.

There's places in Iceland that you may have seen or others may have seen where they actually heat their sidewalks and they heat their roads with piping of geothermal water that's produced from the subsurface.

So yes, there's ingenious ways to use these resources to help benefit us.

That's a great word for your species, ingenious.

You guys are clever.

Like, how many other species on this planet understand plate tectonics?

None.

Sir, it has been a pleasure.

Thank you for educating me on the movements of our Earth's surface.

Yeah, thank you for the time.

Wonderful today.

And it's so important to communicate the value of science— Absolutely.

To the community.

Very much appreciate your efforts to do this.