Big Compute - Tsimulating Tsunamis

Episode Date: December 8, 2020

In 1908, the largest earthquake ever recorded in Europe hit Southern Italy, wiping out the entire coastal town of Messina. Once the shaking had stopped, survivors thought they were safe until a massive tsunami followed minutes later. Even today, the exact cause of the tsunami is debated in the scientific community. In this episode, we talk to Dr. Lauren Schambach from the University of Rhode Island about what her computational simulations of the Messina tsunami have told her, and what that means for people living along the coastlines around the world.

And make sure you get the last name pronunciation right, too. Oh, F. Shambach? Shambach. That's why I said it. Figured it out. Shambach. Shambach, yep.

Sachambach is probably what I would have guessed. Hi, everyone. I'm Jolie Hales. And I'm Ernest DeLeon. And welcome to the Big Compute Podcast. Here we celebrate innovation in a world of virtually unlimited compute. And we do it one important story at a time. We're talking about the stories behind scientists and engineers who are embracing the power of high-performance computing to better

the lives of all of us. From the products we use every day to the technology of tomorrow, high-performance computing plays a direct role in making it all happen, whether people know it or not. Hey, Ernest, guess what? What? Want to know what this episode is not about? Theoretical physics. Well, yes, it's not about theoretical physics. It's also not about... COVID. It's not about COVID.

Yeah, so the last three Big Compute podcast episodes have been about how supercomputing is fighting the coronavirus, which you know. And that's definitely important and we'll probably have more on the subject in the future, especially given the recent spikes in confirmed cases, as well as the exciting vaccine development. So there's much more to tell on this subject. In fact, my mom has the Rona. She's home recuperating from coronavirus right now. But I mean, thankfully, she hasn't had a need to be hospitalized. And then I have a brother-in-law who has it and a few uncles and aunts who have it. So it suddenly really hit my at least my extended

family pretty hard. But I mean, a few uncles and aunts isn't very statistically significant because I come from an LDS family and we have like a thousand relatives. So like the Romneys. Yeah, like the Romneys. Interesting how it's following a similar pattern to the 1918 pandemic, isn't it? It is interesting. I mean, the thing that's nice about this particular pandemic, if there's anything nice about it when comparing to the 1918 pandemic, is that even though cases are going up, the death toll is not going up at the same rate, like not even close. And there were still significantly more deaths in the spring. You know, every death is tragic, but there are less of them now.

Whereas in 1918, you couldn't say that. And part of that is because of science and supercomputing. We're learning how to treat it effectively as opposed to just trying to keep people alive. Yes. We're learning how to deal with it more and more. And it sounds like it's just going to get better as we have the vaccines, hopefully. Yep. But it's so funny.

I tell you that this episode isn't about COVID and then I start talking about COVID. Well, I can tell you that from now on, everything is going to get compared to COVID until our generation passes. I know. So that's pretty much the rest of our history. I think you're right. And it affects everything. My brother's wife had a baby in February and they named her Cove. Oh, no.

And then like a few days later, the pandemic hit. And so my family has a running joke that on her 19th birthday, we're going to have a very large celebration. Yeah, pretty much. But today, we're going to give the COVID subject a breather, and we're going to take on a completely different topic. I figured now is a nice time to do that. And to start us out, I wanted to put on my filmmaker hat, okay, so bear with me. And even though this is all audio, I want to try to paint a picture for you. Okay. Okay. So I want you, Ernest, as well as our listeners to try to use your

imaginations as I paint you into this scene. Okay. Okay. So this is going to be like a verbal Bob Ross. What does he say? Fluffy clouds or paint the trees or happy, happy trees. Happy trees. Happy trees. Yeah. There's no happy trees in this, though, but maybe I'll throw one in at the end. Sounds good.

OK, so here we go. So imagine you live in a coastal town and it's too warm there to see snow during the winter. So it's a couple days, we'll say after Christmas, in the very early morning while the sky is still dark. And you lie in bed, you're awake for some reason, lying in bed listening to heavy rain outside pounding on the window. We'll say that you live on the third floor of an old apartment building, like one of those really old, ornate stone apartment buildings we see in Europe. So not like a grungy apartment, more classic or historical, if you know what I'm saying. Yeah, I lived in San Francisco for a short while, so I can definitely picture this, minus the rain. Yeah, not as much rain up there. So you're in that apartment building and the rain is coming down and you hear some faint thunder in the background. But then instead of dying out,

the thunder grows louder and louder. And right when you're starting to wonder if it's even thunder at all, your bed starts to shake. Before you know it, everything is shaking and shaking loudly. So you grip your bed, you tuck your head for protection, as sounds a crashing echo all around you. A sudden crash is so loud that it hurts your ears and a cold cloud of dust suddenly absorbs what small amount of natural light had existed in the roof. After what feels like an eternal 35 seconds or so, the shaking finally stops. You pause and you do a mental check. It seems like you're okay.

But then you feel water on your back and you realize that the rain is much louder than it had been before. And there's this unexpected breeze that chills you. You put your head up from your bed and you look around to realize that the front apartment wall and a portion of the ceiling are completely gone opened up to the rain around you and this is where we cue in prince's purple rain i don't have the rights to that earnest our budget isn't that big yet we're a podcast if it's just a small like 10 second clip it's fair use there's more so let's say that you're in that room right the wall is down the ceiling's gone so you sit up from your bed and carefully survey the scene around you. And after a moment, you gain the courage to stand up and then you carefully step over crumbled debris

to walk over to that open wall and you look outside while the rain is falling on your face. Through the heavy rain and dust in the faint early morning light, you look out at the silhouette of city buildings that you've gotten so used to, but it's suddenly completely unfamiliar with open air where there used to be structures. And after a few minutes of trying to understand reality and maybe hearing voices calling out into the morning air around you, another rumble starts in the distance. This one also growing louder and louder, but it's different somehow. Your eyes have started to adjust to the darkness, and before you can react, a dark river engulfs the street below and quickly rises,

consuming lower levels of your damaged apartment building and carrying debris of the city with it at incredible speeds. The Kraken. The Kraken! Or... A tsunami. A tsunami, which might as well be the Kraken because both would be completely terrifying to me. I thought we were going to talk about earthquakes in this episode. Okay, well, earthquakes are definitely a part of this, but today's focus is going to be on that moving water, so the tsunami. In fact, that scene I just tried to paint for you is actually based on a real event and real accounts. So the year was 1908 and the town was Messina, Italy, which is on the island of Sicily, just off the toe of the Italian boot, as we say. And I did not know this until creating this episode,

but in 1908, there was a 7.1-ish earthquake near Messina, Italy, and it tragically basically just leveled the city. It's apparently the strongest earthquake to ever have hit Europe, at least that we have record of. I happen to know a little bit about earthquakes, spending some time in the Bay Area of California, so I can definitely relate. Maybe not a 7.1, but I know a little bit about earthquakes, spending some time in the Bay Area of California. So I can definitely relate. Maybe not a 7.1, but I know a little bit about earthquakes. Yeah, earthquakes are kind of the norm here in California, for sure. I was actually really surprised to hear that this Messina earthquake was 7.1 in magnitude because, believe it or not, I actually have been in an earthquake of that exact same size. And it was also in the early morning hours. This time it was it was 2 a.m. for me when I was visiting Guatemala a couple years ago.

And I remember the shaking being incredibly intense. I was standing up and there was nothing to duck and hover under. So I went to the corner of this cement room and I just held on for dear life in the corner as the building was shaking around me. But in that situation, the buildings managed to stay up, thank goodness. And very few structures had more damage than some cracking. Here in Quetzaltenango, and we just had a 7.0 earthquake that scared the crap out of me. But we're good. We just had another aftershock. Scary.

And now the power's out. But in this Messina earthquake, almost every single structure was either leveled completely or damaged beyond repair. And the saddest thing is that around 75,000 to 100,000 people, including half the population of Messina, did not survive. Most of them who died, which, I mean, if you think about it, it included a large number of their police force, their medical professionals, their political leadership, all these people who would have helped otherwise in the aftermath.

A lot of them died in their beds during the actual earthquake when their homes caved in on them. That's incredibly sad. Makes my heart so sad for these people back in 1908. They say that about 2,000 or so people died in the tsunami alone. So that would have been, oh my gosh, I can't even imagine. It's terrible. I know, just the worst, right? And apparently at the time in 1908,

building materials in this particular area of Italy were just really incredibly weak and they were susceptible to earthquakes and they were built on this really soft sandy sort of coastal soil which easily shifted below the buildings during the shaking and so that combination from what I read is a big reason as to why this earthquake was so devastating when compared to maybe the one that I experienced in Guatemala where apparently the cement coated rebar structures that are built on more solid ground were just much better equipped to withstand that kind of shaking. I'm sure that maybe the type of shaking had something to do with it maybe as well but the building construction

was a big player there. And then to add to that devastation, like I said, Messina is a coastal town. And just a few minutes after the earthquake, this large tsunami raged across the destruction and added to the already horrific situation. That's amazing. I'm not sure how tied the United States and Europe were at that point. But remember, this is two years after the major earthquake that essentially leveled San Francisco for very similar reasons. Construction, obviously not properly retrofit for earthquakes because that didn't really exist back then. Not very stable soil. All these kind of things kind of add. And in San Francisco's case, there was also natural gas leaks that sparked fires.

Oh, yeah. And they had that problem in Messina, too. A lot of fires after the tsunami kind of leveled out. Right. So very, very similar scenario. The difference is obviously there wasn't a tsunami in the 1906 one. But this is interesting to study because Sicily is not an island in the open ocean like those in the Gulf that we hear about getting you know, getting tsunamis all the time. It's enclosed within the Mediterranean Sea. Oh, I completely agree with you. So I always think of tsunamis as being associated with the oceanic coastline. You know, I mean, that's what we hear about. And it looks like that is where most tsunamis occur. In fact, I read 80 percent-ish

or so. But apparently tsunamis can also form in other bodies of water, like lakes, if the circumstances call for it. And I've just really never considered that. And this particular Italy tsunami came from what is known as the Messina Strait, which at its narrowest point is only about two miles wide. And that's not very wide. And so I wouldn't personally have expected a tsunami to come from there. It's interesting to us because we don't quite understand exactly how everything that happened worked. I'm guessing that was the voice of our tsunami expert. You are correct. Meet Lauren Schambach. And Lauren is fresh out of Ph.D. graduation land. She received hers in August.

From the University of Rhode Island in ocean engineering. And she has a very unique expertise. I basically study tsunamis how we model them on the computer. And the reason that Lauren got involved in tsunamis is because she grew up at a famous beach. It was back in 2009 when MTV's pop cultural phenomenon the Jersey Shore drew in as many as eight million viewers. Yes, the Jersey Shore. But her experience was a little bit more real life. The Jersey Shore has had 19 weekends of clear weather with temperatures rising above 70 degrees. The only time it rains is at nighttime and then it gets sunny in the daytime.

So it's been a good summer. And you can see just a number of people who are out here on the boardwalk this evening. So I love surfing. My family's really big into surfing. We just go to the beach every day during the summer. That's what we do during summer vacation. And while she's never seen a tsunami in person, she's seen her share of big waves. Waves are terrifying. But I think that that's part of what makes it so cool.

That's what I'm so interested in. On the East Coast, we have hurricanes that come up the coast, and you'll get these huge swells that'll come in. And, you know, you have the crazy surfers who go out. I mean, I love surfing, but I'm not about to paddle out in some of those hurricane waves. And the Jersey Shore has definitely seen its share of hurricanes that have brought destructive waves with them. Most notably, as I'm sure you remember, in 2012,

Hurricane Sandy ripped across the East Coast, damaging 346,000 homes and pretty much wiping out the casino pier with its roller coaster and other amusement park rides, all of which were icons for the Jersey Shore. This morning, New Jersey's Seaside Heights, a city synonymous with summer fun, is now a city completely submerged. So much of that iconic beach town of Seaside Heights is underwater, full of debris there, parts of that famous boardwalk wiped out. This is an unbelievable scene, block after block, mile after mile of this kind of ruins. Was Lauren living at the Jersey Shore when the hurricane hit? So at the time that Hurricane Sandy hit her hometown, Lauren was finishing her last year of undergrad in Rhode Island. And even though she wasn't present for the hurricane, she, I mean, as you can imagine, she was just sitting on pins and needles

watching the news coverage of her childhood playground being ripped apart. I've never seen anything like it. We had waves as high as the light poles down on the boardwalk. Without the ability to even get in touch with her family and her friends because of all the power outages that were going on. In the end, her parents' home was lucky to scrape by without heavy damage. But a lot of her friends and neighbors suffered extensive losses and images of the destroyed boardwalk and the pier were devastating to Lauren. And even after rebuilding, Lauren says she still gets nervous whenever she hears reports about a hurricane about to hit anywhere. We begin with the long-awaited reopening of the Jersey Shore, almost seven months to the day after the beaches, boardwalks, and seaside towns were blasted by Superstorm Sandy.

But obviously, hurricanes and tsunamis are quite different. Hurricanes are these big storms with winds that consequently create large waves. And while they're much more common than tsunamis and they cause more damage over the long term, they're also much easier to predict and much easier to prepare for. Tsunamis, on the other hand, they usually strike with very little warning, often within a few short minutes of an event such as an earthquake or maybe a volcanic eruption. And the reason they strike so quickly is that these waves move at an average speed of 450 miles per hour, some even faster than that, which to me is insanely fast.

Have you seen the movie San Andreas? I have seen clips of it when I was working for Disney. It was on in a break room and I couldn't keep watching. It's such a bad movie that it's great. The Rock, I'm a big fan of The Rock. I like The Rock. He really nailed it in this movie.

It's one of those where you can tell he didn't take it seriously. He didn't take himself seriously, but he played the part. I love it. It's such a those where you can tell like he didn't take it seriously. He didn't take himself seriously, but he played the part. I love it. Such a good, bad movie. OK, maybe I need to give it another shot. And that's funny that you mentioned San Andreas, because while I was doing research for this episode, I ended up falling down the rabbit hole of watching movie scenes that depict tsunamis. And I'll be honest, it's no wonder society is a little bit confused as to

what tsunamis look like. I mean, instead of these tragic scenes that they're trying to portray, these movie scenes came off more like comedies. And maybe that's because I was just watching the scene. I was not involved in like the characters' emotions at all, but they were so ridiculously over the top. Like, for instance, they would show this wall of water, maybe the height of a skyscraper, that would move in slow motion and it would cast this menacing shadow on some city and then it would crash into it and level everything as people ran screaming for their lives and it was just, it was just hilariously ridiculous. Definitely a lot of times movies take a lot of creative license when it comes to how they portray these waves.

I watched movie tsunamis topple cruise ships, aircraft carriers, skyscrapers, the Golden Gate Bridge. Um, your apartment was pretty much gone in one of these. The whole city of Dubai. I mean, everything was taken over by these tsunamis. And obviously none of this would be funny if it could possibly be true. But thankfully it can't, not at these ridiculous levels. I mean, Hollywood definitely takes some liberties. And I gotta hand it to them. Some of their visual effects were a few steps up from Sharknado. Though, I do have to say, and I don't know if you've seen the movie called The Impossible. I think it's on

Netflix right now. It's about the 2004 Indian Ocean tsunami. And I actually thought that movie was really good. I liked it. I haven't seen that, but I am a huge fan of the Sharknado franchise, as well as the less well-regarded spinoff, Lava Lantula. Oh my gosh. I haven't even heard of La Valencia. It's a hidden gem. Let's just put it that way.

Okay, but I will watch at least one minute. A big public misconception is that tsunamis are these walls of water that basically crash down on people. But in reality, they don't actually look like that at all. When you think of a wave, you think of like an ocean wave, right? So you have like, you know, the up and down that you can see. So a tsunami is actually technically a really long wave. So a better way for someone to imagine it would be to think of, for example, a tide and how the tide kind of like slowly comes in. You can't really tell that it's a wave, but it technically is. So a tsunami is really similar in that you basically just have a rush of water coming

inland and it just keeps going and going and going and going like a tide coming in that doesn't stop. In fact, there is this rockin' video by NOAA, or the National Oceanic and Atmospheric Administration, which is part of the United States government. And this video, which has more than 16 million views on YouTube, which is pretty good for like a science video. I just want to add a note here that it's really sad that we consider 16 million views a lot for a science video in this country. I know! Like, there should be 330 plus million views on that thing. But unfortunately, it's... It's science.

Baby shark, that I think is number one. Baby shark, do-do-do-do-do-do, baby shark. But this video really helps you understand how a tsunami can start and what it can look like. And I want to show this to you, Ernest, and our listeners will be able to hear it, and we'll also make the video available to view on bigcompute.org on the notes page for this episode. In fact, you can go there right now if you're listening and watch it along with us if you want.

Okay, so we see the land underneath the water. There's an earthquake and the land, what do you say, displaces, goes up? Yep, goes up, displaces, yep. And then on the surface of the water, it causes this kind of big moving hump wave that's now traveling across the ocean, right? It's not like a rip curl style wave. And now we're, let's see, it's a shot of the coastline. So this is like a city and neighborhood coastline. So right now it looks like the tide going out, which is not actually what's happening.

But it's that portion of the tsunami where the water pulls back from the shore. Before it hits. Yeah. In the distance, you can barely make out that there's a wave coming because it just looks like the ocean is higher. But then here it comes. I think, and it's moving, it's heading to the city and the village incredibly fast. Like, crazy fast. But it's not this weird curl. It's like a tide that's moving fast. And then, boom!

It smashes these 3D model homes and buildings. But yeah, it hits the shoreline and it obviously goes past the shoreline and just envelops everything it can along its way. Yeah. Okay, cool. What did you think of that video, Ernest? That was pretty good. Not at all what I saw in San Andreas, but probably much more scientifically accurate. Probably. So as for our tsunami expert, Lauren, she was totally entranced by the way that water can move with such an impact. For me, it's so awe-inspiring because you know how powerful it is. Water just has this great destructive power, but it could also be really beautiful. Lauren has studied tsunamis from a couple of different angles, but she specifically got

involved in looking at the Messina tsunami when a PhD student studying in Italy was working on that project and then graduated. So Lauren picked up the project from there. This was a pretty significant event for the area at the time. People were paying attention to it. So scientists went to the area within weeks of it happening, and they saw the importance of collecting data and collecting interviews from the people who were witnesses so that we could study this event. So there's four main guides, Mercalli, Omori, Plotania, and Barada. So each of these four guys, they went to the area, did some measurements. So they basically measured, you know, in this specific area, like how high did the water go? They talked to people who live there. They said, you know, did you see it? What did you see? You know, how long was it between

the earthquake shaking and the tsunami arriving? And then they took all this data and they put it into their own report, which they shared with each other and shared with the scientific community. So we actually have a lot of information about this event, even though it happened over 100 years ago. This is why data collection is so important in events like these. You may not have the scientific technology to do much with it at the time, but it is a good bet that someone in the not too distant future will. Yes. It kind of reminds me of forensic investigations decades ago where investigators collected DNA samples at crime scenes before DNA technology really existed. And then later, as technology advanced, many of those DNA samples have been used to solve cold cases and and put bad guys in jail.

Yeah. So what size tsunami are we talking about here in Messina? So that kind of depends on where exactly you were located on the coast. But in some places it was up to 12 meters, which is around 40 feet. That's huge. That's really significant. Not only was there this devastating earthquake. So say, for example, you know, you managed to get yourself out of your house and you live near the coastline and you, you know, ran outside. And the next thing you would see would be this massive wave coming towards you. It would be absolutely terrifying. Although that doesn't necessarily mean that the wave was 40 feet tall like we see in the movies.

So when I say 12 meters, so it's actually, that's a measured run-up. So basically what that means is if you're at the coastline, the coastline is, for example, zero, your zero elevation line. The run-up is actually on land, the highest point that water reached. So you're not necessarily talking about a depth of 12 meters, but the wave itself reached up to 12 meters high on the ground. Right. So it's the elevation of the highest point. So just imagine if you drew a triangle where the top two points, there was a flat line and that was the level to which the ocean had risen. And the lower point that makes the triangle would be where sea level was before the thing happened.

That's what she's saying there, right? So it's 12 meters high at its highest point once it's on the ground, but it wasn't a 12 meter wall coming at you. Yes. And to put that into perspective, that horrible 2011 tsunami in Japan reached 40 meters, which is 131 feet elevation. So it was quite a bit bigger than the one in Messina. Yes, I still remember watching the videos coming out of Japan as the tsunami washed ashore. Cars, boats, buildings, all floating along gently as the water came in and then went out.

Yeah, I was completely horrified. I just sat there. I remember watching it online, and I just felt so sad. I mean, talk about feeling powerless to help people. And from these videos, you can kind of see, right, like how complicated the water flow is because the water is coming and coming and coming in. It starts hitting buildings, starts interacting with buildings, hitting cars, hitting trees, basically ripping all these things up and then taking it with it. So you start to get this like really complicated mixture of all sorts of different stuff.

So you can imagine if you were there and you were witnessing it, that would be pretty horrifying. And the Japan tsunami traveled around 500 miles per hour, which reached the coastline within, I read anywhere between 10 to 30 to 45 minutes, depending on where you were after the earthquake hit. And while in the case of the Messina tsunami, since it originated from much more shallow water, it didn't actually move quite as fast. Basically, the speed of a tsunami is very dependent on the water depth. That's like the main thing that is driving the speed of the tsunami. You could say that the speed equals the square root of gravity times the depth of the water. So if you have really deep water, you have faster wave.

If you have really shallow water, it slows down. But it's, I mean, it's still, you're talking like two to three meters per second inundation on land, which is, that's fast. But in contrast to the deep water, it's slower. And depending where survivors were located in Messina, they reported that the wave reached them anywhere between two to ten minutes after the earthquake. That's really fast. It is really fast. And it makes me think.

So I go running at the beach every weekend, and I think I mentioned that on every podcast episode. At least once. At least once. It's like my highlight of my week when I get out of quarantine. Especially during COVID. Oh yes. It's like the only time I see the sun. So I have to talk about it on every episode until we can all see the sun more again. But for about an hour of that run that I take every single weekend, I'm usually on a peninsula. And I'm kind of a safety nut, admittedly. My husband calls me Safety Inspector Joe. And I've thought that if an earthquake were to hit while

I was running on that peninsula, the only way that I could really even get to safer ground would be to maybe run to the main road and then jump in the back of someone's moving pickup truck and then pray that California's notorious traffic is somehow non-existent at that moment, which, yeah, right. Now, thankfully, tsunamis in our part of the Pacific Ocean are extremely rare because our main fault lines down here are inland rather than in the ocean. And California faults tend to move horizontally more than vertically. And it's those, I guess, vertical movements in the ocean that tend to cause these tsunamis. Right. I'm on the other side of the mountain range here in the valley that Silicon built, so I'm not really worried about tsunamis. I am, however, very cognizant about the earthquakes as they happen so frequently here.

So then what did they find was the cause of the tsunami? Obviously, it had something to do with the earthquake. Well, yeah, that's the interesting thing. The most common way that tsunamis form, as we've alluded to, is during an earthquake when slabs of rock on the seafloor suddenly move vertically past each other, like what we saw in the NOAA video. And that can happen in land-based earthquakes too, with that earth moving that way. But when happens underwater suddenly the water has to like rush in and fill this new hole causing a massive wave to form and then that wave moves away from the site at a rapid pace until it hits a shoreline which is what happened with the japan tsunami in 2011 right but with messina over the years over you know the 1900s there's been a ton of different hypotheses of how the fault moved because we don't exactly know where the fault is.

That's super interesting that they can't find the fault. Yeah, they know that a fault exists and they know that a fault moved, but they don't know for sure where that fault line is or how it moved exactly. But because scientists recorded so many data points in detail at the time of the earthquake, researchers over the last 100 years have been able to use that data to hypothesize what could have been the cause of the tsunami. But even with the same data points, you had different scientists saying different things. Okay, well, we think that the fault's here. Well, we think that the fault's here. Well, we think that the fault's here. We think it's oriented like this. We think it's oriented like that.

So there's a lot of these different models and they all use, you know, accepted methods to come up with their model, but they're all different. And nobody could agree on what was right. And then computational simulation was born and suddenly the game changed. Someone had the idea, you know what? Since we can't figure out what's going on with the earthquake, what if we model these different earthquake faults that people have proposed,

model this tsunami based off it, and then we compare the tsunami data to what happened and whichever one matches the best, that's the earthquake, that's the correct fault. That person got it right. And that makes perfect sense. But my guess is that things aren't going to line up as cleanly as they might have thought. And I mean, the way she mentioned it feels like this kind of global science fair, but obviously quite a few steps above the baking soda volcanoes that I turned in every year as a kid. I was that person. But in this case, finally computers were introduced and they were going to give us the answers. Well, there was a little bit of a problem with that.

Uh-oh. So all these different scientists ran various simulations of different earthquake sources and scenarios, but... They found out that none of the earthquakes could actually predict the tsunami that was actually experienced by the people. So all the simulations based on people's predictions for the earthquake were not correct? Yep. What?

So then there's this question, well, why? What happened? So we still don't know. Well, in 2008, another group of Italian researchers published a paper with yet another hypothesis. They proposed that the earthquake in the Messina Straits actually triggered an underwater landslide. And the landslide was so massive that that actually was the main trigger of the tsunami. That's an interesting hypothesis to come up with. Right? A landslide? I never thought about that. I guess I always tended to picture the earth below water as being like relatively flat,

like in Finding Nemo or something, which doesn't really make any sense. I mean, of course, there'd be variations in elevation underwater, just like there is above water. I just hadn't really thought about that before, let alone whether or not a landslide was possible underwater. So it depends on a couple of things. So it's your sediment accumulation, how steep your slope is. And basically, if you have some sort of event that's enough shaking, for example, from an earthquake that could destabilize a large amount of sediment, you can have this underwater landslide. So these Italian researchers proposed this,

and then another group of researchers simulated it computationally to see how it stacked up against the historical data. And let me guess. It didn't really match. Surprise, surprise. Yeah, so researchers even went out on boats around Messina, and then they mapped out the seafloor as well as what's under the seafloor. And at one point they found remnants of an underwater landslide from the past. But then they eventually concluded that it was too old to have actually happened as late as 1908.

So it couldn't be credited with causing this particular tsunami. And that's where Lauren comes in. With her supercomputer. How did you guess? I wonder. And so the idea of my research with our co-authors and collaborators was we were re-looking at this case.

They wanted to give it a shot. So we basically went through all of the observation, all of the data. We looked at newer maps and information from the seafloor. We worked with a marine geologist who's the co-author on the paper. And he said, you know, I think in this particular location, I think there was a smaller landslide. Another landslide hypothesis, but a different size and a different location. So we simulated that on the computer. We simulated some different earthquake configurations in combination with this new landslide.

And then we compared our results to all of the things that happened before. Did it match the historical data? Well, I'll tell you what they found after the break. Are you really going to leave us hanging like that? I am. From supersonic jets to personalized medicine, industry leaders everywhere are accelerating innovation with unprecedented speed and efficiency by using Rescale, the intelligent control plane that allows you to run any app on any infrastructure, totally optimized.

As a solution for intelligent full-stack automation for big compute and R&D collaboration on hybrid cloud, Rescale empowers IT leaders to deliver high-performance computing as a service, with software automation with incredible security, architecture, and financial controls. And as a proud sponsor of the Big Compute podcast, Rescale would especially like to say thank you to all of the scientists and engineers out there who are working to make a difference for all of us. Rescale, powering science and engineering breakthroughs. Learn how you can modernize HPC at rescale.com slash bcpodcast. Okay, so where were we?

Lauren's team just ran simulations of their own Messina tsunami hypothesis. Ah, yes. I'm curious, did their results match historical data? Pretty darn close. By using higher resolution grids, and they also used more accurate topography than what was used in previous studies, Lorentz simulations matched pretty much

all of the historical observations and timelines, except for run-up data along part of the Messina Straits, which was still simulated to be too small and shallow than historical reports, which is kind of interesting. But it's closer to that right answer, and simulation is allowing us to get there. And while everything didn't perfectly match historical records, which is, I mean, frankly, to be expected in this kind of a situation, Lawrence Group has helped push the understanding of this case study even further, and they agree. We think that there actually was a landslide that was a contributor to the event as well. All right, so correct me if I'm wrong here,

but it's the city of Messina. Early in the morning, there's an earthquake that hits. That triggers a landslide, which then triggers a tsunami. How big of a landslide are we talking about here? So I asked Lauren the same question. So for this particular case, we modeled about a two cubic kilometer landslide, which is that's pretty big. Which is basically a cubic half mile. That is a reasonably sized landslide.

Yeah, that sounds scary to me. I mean, looking at a map, it looks like Lawrence Group's results put the landslide south of Messina, northeast of Mount Etna, at the foot of what's called the Fiumifredo Valley. I probably said that wrong. And that's on the east coast of the island of Sicily. So anybody who knows Italy, which is probably very few of our listeners, but there you go. You can go look it up on a map. What kind of narrowed us down to figuring out that this probably happened in this particular location was that the observations of the timing of when the wave arrived

was between two and three minutes directly on shore of where that happened. So further away, you had, you know, 10 minutes, eight minutes, but directly on shore, they were consistently like, oh, two minutes, three minutes, five minutes. So it makes sense that the wave would have hit that spot first. And underwater landslides can cause tsunamis because when the land under the water suddenly moves down a slope, where the land is moving away from, you get a depression on the surface because you've displaced material. And then the water rushes to fill that spot. Yep. And then basically what it does, it'll rebound because of gravity and that creates waves that propagate outward from that spot. And that's what

she thinks happened back in 1908. Which brings us to today. Why are scientists so interested in what happened in the Messina Strait over 100 years ago? So for example, if you're from the United States and you're going on vacation to Italy, you're going on vacation to the south coast of Italy, which is a pretty popular thing to do. Do you have any idea that there have been these events in the past? For example, not even just a tsunami, but an earthquake. And if you were in a coastal area and you experienced an earthquake, would you know that the kind of accepted protocol globally is if you're near a coast and there's an earthquake that you get to higher ground? And this is why research like this is so important.

Learning from the past to prepare us for the future. And Lauren says, honestly, there's no need to like avoid coastlines completely. But rather for her, it's just good to know that if you're near a coastline and an earthquake hits, it's a good idea to seek higher ground out of an abundance of caution. If people have that knowledge and know that that's kind of the protocol for what you should do, I don't think you need to be overly worried about tsunamis because they are quite rare. And this is coming from someone who has studied a variety of different tsunami scenarios. In addition to her work on the Messina case study, Lauren has also run extensive simulations on the 2018 Indonesia tsunami, if you heard about that one. And that one was caused by a volcanic collapse.

And also hitting pretty close to home, Lauren has also run simulations on what could happen if a tsunami were to hit the east coast of the United States. You could have a large earthquake in the Puerto Rico trench that could send a tsunami wave up the east coast. Or if there was a moderate earthquake off the east coast, which again, seismicity on the east coast is kind of rare, there is potential that it could cause an underwater landslide event. Or there could even be a volcanic flint collapse in the Canary Islands. So you have, you know, these volcanic islands. In some studies, they have discovered that there's potential that part of one of the islands could potentially collapse into the ocean and cause a large tsunami that would travel across the Atlantic. So that's another case that we look at as well. And it's important to stress that these are extremely rare potential scenarios that Lauren has run computational simulations of to see basically what would happen.

It doesn't mean that if you live on the Canary Islands, you have this high chance of ending up at the bottom of the ocean anytime soon or anything like that. I'm curious, in all her studies about the East Coast, were there any particular areas that seemed at higher risk of being hit by a tsunami than other areas? Actually, yes. A spot that's actually kind of near and dear to me. So basically with the tsunami waves, especially when they're traveling across the ocean, they're going to interact with the bathymetry, which is basically like the topography underwater. And you get wave focusing and defocusing. So depending on if you have like a ridge or if you have a canyon or something like that, that's gonna change how the wave energy focuses or defocuses on the coast.

And so we have a really large continental shelf off of the US East Coast. And so the bathymetry actually has a pretty significant impact on this focusing and defocusing. And through some of the simulations that I've run, I definitely find quite often that a lot of waves energy often gets focused at part of Ocean County, New Jersey, an area called Maniloking and Seaside Heights. So that's actually really

close to where I grew up. So a lot of my simulations, I'm just basically destroying these really awesome, quaint coastal communities that I like to spend time at. Do you actually watch the simulation after it's been run, like replay it and see your house just kind of, bye. Yeah, well, luckily my parents' house is not exactly in that particular area. But yeah, I mean, it's pretty devastating to watch. But then also, you know, for my master's, when I was doing some studies on the extreme storms and Hurricane Sandy, for example, it's a lot of similar areas that were heavily impacted by that are also impacted by tsunamis, which is interesting. And it's just because of the way that the wave physics works.

So we're back to the Jersey Shore. So what kind of software did Lauren use for simulations? She in particular runs two open source wave models that are built in Fortran. Wow, that brings me back. I haven't used Formula Translator in quite a while. I've never used it. But one of the models is called NHWave. Which stands for non-hydrostatic wave model. That's where we can basically simulate I've never used it. But one of the models is called NH-Wave. Which stands for non-hydrostatic wave model. That's where we can basically simulate land motion and the waves that are generated from that. And FunWave, which has an awesome name. Which stands for fully nonlinear Boussinesq wave model. And that model is primarily used to propagate the wave to the shore and to do studies on run up and risk and things like that.

And to run one of these models, you basically enter a bunch of specific data into a supercomputer and then it spits out solved equations or outputs that can be plotted to create a kind of picture of what's going on. First, you need a representation, like a gridded representation of the seafloor, which we call the basimetry, and the topography on land. So basically, you have a big grid matrix of seafloor data, and you let the model know what the resolution of that is. So, you know, is it a kilometer? Is it 30 meters? Is it seven meters? And then basically you need some sort of wave input. So if it's from an earthquake, what we typically do is we'll use a method to generate an initial wave based off of how the seafloor would move. So we take that initial wave and we would put that into the model on the same grid as your bathymetry was on. If you're doing, for example, a landslide case, you would use this NH wave model and you would say this area is allowed to move

based off of the physics of the landslides. And that would generate the water wave based off of the physics between how those things interact. Are these simulations pretty compute intensive? I mean, what kind of compute power are we talking about here? For these studies, Lauren's team had access to a couple different supercomputers through what's called the NSF EXCEED program, which stands for National Science Foundation Extreme Science and Engineering Discovery Environment, which is a resource hub for supercomputing basically. And they specifically used the Pittsburgh supercomputer bridges and then the San Diego supercomputer comet. And on average, she said they typically used about 100 to 200 CPUs per run and each simulation took about one to three hours to complete. Which doesn't sound massive.

In fact, it made me possible to run these particular studies on a desktop computer, say one of the high-end Threadrippers from AMD. But it would take a few days instead of a few hours. Right. So like something cool that, you know, the supercomputer lets you do is I don't have to wait for the landslide case to finish before I can start the earthquake case. I can run them at the same time. That's interesting. I wonder how they can run at the same time if one of them kind of begets the other. For Lauren, the biggest benefit to using supercomputing for these tsunami studies, to your point, is that she can run a number of different

scenarios simultaneously instead of like running them back to back over a long period of time. So it's about changing a factor and then seeing how it adjusts on the exact same kind of simulation with just that one number changed, if that makes any sense. Got it. Yeah. Yeah. So this allows her to test different model parameters. Like for instance, she specifically mentioned altering a parameter like the bottom friction and then seeing how it changes the outcome. So she can run multiple simulations with different numbers for that bottom friction and then see what the results are. And that helps her quickly move along to the next kind of simulation. If you wanted to do a sensitivity study on that, you could very easily

do it, you know, using these supercomputers. And it'd be a lot more difficult to do without it. Yeah. It appears that studies like this are really only possible with computational simulation. Otherwise, we could only study what happened in the past. We couldn't study it in the current time period because we would have to trigger a tsunami and then watch what happens. Which is probably a little bit of a pain, maybe a little dangerous, a little expensive. And slightly unethical. Slightly. As far as I can tell with tsunami research in particular, I mean, when you look at some of the early studies, they consider one case and it's because of the computational limitation.

So they did the best that they could, but it was kind of like, here's our one case that we're publishing. Whereas now, if you don't have like your case with all these different parameters that you tested, the sensitivity of it, really, we can like learn a lot more about these different events. And with the results of all these different tsunami simulations, what happens next? That gets passed off onto emergency managers and that can help them inform how they're going to kind of set up their community if they need community preparedness regarding this, if they want to change zoning based off of, you know, potential risk. Right. And this is the same exact response we've gotten on several previous podcast episodes, right, in terms of what the science results in. And it typically results in handing off data to decision makers of some kind, whether they're health professionals, emergency managers, risk managers, for them to be able to make decisions based on the most current scientific data that they can possibly have in their hands.

Exactly. For the Messina 1908 case, this happened in an area, it could potentially happen again. Do the people who live there know the risk? Is there more information that we need from that particular case study to understand what happened better? So it can kind of drive research, like, you know, have people propose to go out in boats and collect more information or something like that. So there's really a lot of different ways that this research can have a real impact. And really, that's what drives Lauren. I really love these historical studies because I think they're really important to kind of understand what has happened in the past because that helps us predict what could possibly happen in the future. But it's also kind of the human aspect of, you know, remembering what happened in the past, how it

impacted different cultures. And so I think it's really exciting that we now have the capability to kind of put together databases of information from historical events. And then, for example, in 2018, there was the Palu Indonesia tsunami, which we have a ton of different data for lots of different data sets, lots of teams went out and collected data. So I think it's really cool that we're going to be able to kind of have that data more accessible to more people who can then look at it in different ways and different perspectives. And I think just having access for more people in general to these different kinds of systems is going to lead to just really awesome discoveries. And now it's just a matter of spreading the word. We all have this great data from the past and predictions for the future.

And so the next step that Lauren is focused on is making sure that that information is actually used, not just by like being published in journals and sharing information with the scientific community, which is important, but also by getting the word out there in a way that reaches the right audience, like those decision makers you were talking about, Ernest. So just like people in earthquake prone areas, like here in California, just like how we have duck and cover plans and how those in tornado zones have basements or shelter plans. Do people along the coast know what to do if an earthquake strikes? Do they know that they should seek higher ground in the rare possibility that there is a tsunami? And maybe most people in the United States do know this,

but like what about people in Indonesia or Guatemala or developing countries? I mean, how do we get the word to them? Right. Having all of this data is not enough if no one else knows about it or benefits from it. Science is for the world and its purpose is to benefit all of humanity. Yes. And that's actually one of the reasons why Lauren wanted to talk to us on this podcast. This is one step towards spreading the word for her. And we're happy to help out. Indeed. So if you listeners out there would like to learn more about Lauren's research, check out bigcompute.org where we will include links, pictures, video, anything we find that will help you go down the tsunami rabbit hole. And you can also find Lauren on LinkedIn.

Look for Lauren Schambach, which is spelled very fancily. So we'll have a link for this as well so that you can just click on the episode notes and be a LinkedIn friend. Or I guess it's not called friends on LinkedIn. Connection, right? I guess. I don't know. Now, is it pronounced Lauren or Lauren? Lauren.

Wait, that was like a mix of the two. Was that Lauren? Lauren. Lauren. Oh, I'm going to say it. Don't worry about it. Both are fine. Okay. You're probably just being nice. And Lauren doesn't know I'm going to say this, but if anyone out there is looking for a rocking tsunami wave simulation person to work for them,

now that she's graduated with her PhD, Lauren is... Fun employed, as we like to say. But I'm in that nice, fun time in between graduating and figuring out what I want to do next. Any employer would be lucky to have her. I'm just saying. Yep. And this is just another example of how supercomputing has changed the way we do research and how it's changed research itself. It's totally true. And we see this more and more. In the past, it was all about studying history. But now we can use those studies of history to predict the future in a much more numerical way through computational simulation. And then we can really accelerate that pace using supercomputing. So it's an exciting time to be alive in science and engineering.

Absolutely. And of course, special thanks to undercover superhero Lauren Schambach for her work in tsunami simulations. Yes, whenever I run by a tsunami warning sign at the beach, I will think of her,

which I actually did do over Thanksgiving weekend. I even took a selfie with the warning sign and I dedicate it to Lauren and her research. And I will sleep soundly in a mountain valley

knowing tsunamis can't touch me here. Yeah, cue the MC Hammer music. Until next time. See you later. I don't know what else to say. Have a good one. Stay safe.

Mask it or casket. That's so funny. Oh, that's funny. Thanks, buddy.