PubTalk — Tsunamis

Video Transcript
Download Video
Right-click and save to download

Detailed Description

Lessons and Questions from the Indian Ocean Disaster

By Eric L. Geist, geophysicist, Bruce E. Jaffe, oceanographer, and Brian F. Atwater, geologist

  • What do computer animations reveal about transoceanic tsunamis?
  • What varied marks of its force and height did the December 26 tsunami leave in the coa stal environment?
  • Why did waves 100 feet high strike norhtern Sumatra in December, while low-lying Bangl adesh was spared?
  • Can a Sumatra-size tsunami occur on the U.S. Pacific coast?
  • What is the tsunami threat to central California, and how is this region getting prepa red?


Date Taken:

Length: 02:00:27

Location Taken: Menlo Park, CA, US


- Good evening, ladies and gentlemen. It’s just a little after 7:00. I’d like to welcome you all to the U.S. Geological Survey Evening Public Lecture Series. For those of you who are used to seeing Leslie up here, I am not Leslie. She is doing some interesting training this week. I’m Terry Bruns. I’m with the Coastal and Marine Geology team, as are two of our speakers tonight. Before we go into the speaking part of the program, I would like to let you know, we are getting more of these little pamphlets. Many of you may have missed it as they came in. We’re trying to get some more of them in. They’re out on the table in the lobby. And, secondly, I’d like to remind you all that we have this public lecture every month. And next month we’re going to be viewing and touring the San Andreas Fault System in 3D with Phil Stauffer – California Rock and Roll. Now, with this lecture, there is a field trip. And I would encourage any of you who are interested to maybe read the brochure. And if you would like to sign up for the field trip, there is a sign-up list outside on the table also. There is also a sign-up list out there for any who would like to be on our mailing list. Our lecture tonight is entitled, Tsunamis: Lessons and Questions from the Indian Ocean Disaster. This is going to be a team performance. Eric Geist, our first speaker, is a member of the U.S. Geological Survey Coastal and Marine Geology team, with a primary interest in modeling tsunamis and understanding why and under what conditions they propagate. And how they interact with the coastal zone. He will show us computer simulations for the Indian Ocean tsunami and for a possible Pacific Northwest tsunami. Bruce Jaffe, also with the Coastal and Marine Geology team, is project chief of the Tsunami Risk Assessment Project. His specialty is examining sedimentary deposits left by tsunamis to estimate flow regimes and tsunami history. He has made field studies of tsunamis in Peru, Papua New Guinea, and most recently has been to both Sri Lanka and Sumatra as part of international teams studying the Indian Ocean tsunami. Finally, Brian Atwater is a geologist with the Western Earthquake Hazards team and an affiliate professor at the University of Washington. His research concerns earthquakes and tsunamis on the Pacific Rim, with special emphasis on the Pacific Northwest. His work has been fundamental to our understanding of tsunami and earthquake risk in the Pacific Northwest, and he was listed, actually, in the April 18, 2000, Time magazine as one of the 100 most influential people in the world and was the only true scientist included. It’s kind of astounding that Time magazine would recognize that the kind of work that’s being done on tsunamis may be valuable to this nation. So, with that, we’ll let Eric start.

- Thanks, Terry. As you can tell from the crowd here tonight, many people are trying to understand what tsunamis are all about. That’s what we’re trying to do, is to impart what happened after the horrendous devastation in the Indian Ocean last December, and also what’s going to happen here closer to home, in the Pacific Northwest and even central California. So, to do this, we’re going to take, as Terry mentioned, several different perspectives. First, I’m going to start with a tsunamis 101. I hope to impart some kind of basic understanding of what happens during a tsunami. From there, I’m going to take the geophysical perspective. And simply put, this is just a bird’s eye view, or a space-based view, of tsunamis as they propagate across ocean basins. Then, we’ll bring this down to Earth and talk to some geologists about the subject. And they’ll show you what happens to the tsunamis as they impact shorelines and how they change from scales of landforms all the way down to grain sizes. And to do that, we’ll have Bruce Jaffe explain some of these exciting results that are coming back from the Indian Ocean, and Brian Atwater, in his elegant fashion, will explain the tsunamis in the Pacific Northwest. So, to start with, to understand what tsunamis are all about, we often break the process up into three parts. They include generation, propagation, and runup. Now, what do you need to generate a tsunami? It’s something that displaces the entire water column very quickly. Now, there’s fortunately very few natural phenomena that can do this. One is very large earthquakes. They have to be larger than about magnitude 6-1/2. Very, very, large underwater landslides. Volcanic eruptions such as the 1883 Krakatoa event. And even asteroids generate tsunamis. A deep impact scenario. So, here’s the case of an earthquake-generated tsunami. This is the most common generating mechanism for tsunamis. As soon as this earthquake – it generates this massive gravitational instability in the water. And of course, that big mound of water is not going to want to stay in one place. It’s going to start spreading out. And a little bit of terminology here. The tsunami that’s going towards this local shoreline – so here’s the seafloor in brown – this is called the local tsunami. The one going out into the middle of the ocean, we call that the transoceanic tsunami, or sometimes the far-field tsunami. Now as it propagates over varying water depths, it undergoes these gradual transformations, as shown here. And then finally, once it impacts the shore, it undergoes a radical transformation called this runup process. And Bruce will kind of explain what happened during runup in the Indian Ocean tsunami. So, okay, so it takes large earthquakes. And the earthquakes have to be underwater to generate landslides. What kind of plate tectonic boundaries need to be in place for this to happen? And this is an important point to get across when we start talking about the relative hazard of California – central California – and places like the Pacific Northwest or the Indian Ocean. What we have in our backyard is a transform boundary, probably more commonly known as the San Andreas Fault. Well, this has got a lot of horizontal motion, not much vertical to push up the water. It’s on land for most of – most of its length. So, this is an inefficient way to generate tsunamis. Some small ones of historical note can be generated. Most of the tsunamis – most of the major tsunamis are generated along subduction zones. This is where one plate is being pulled beneath another. And we’ll see two examples of subduction zones in Indonesia and the Pacific Northwest. Now, most of us are familiar with earthquakes. Earthquakes are strong ground shaking of the Earth, right? Well, that’s not what causes tsunamis. Tsunamis are caused by this permanent deformation, or coseismic deformation, we call it, of the Earth. And this is pretty subtle. Like, with the Loma Prieta earthquake, it was really only detectable over tens of hundreds of kilometers by sensitive geodetic instruments. Sometimes faults can rupture up to the surface and create nice fault scarps like this. So, this is an example of permanent deformation. Well, these same processes that occur onshore are going on for earthquakes, of course, underwater. In this case, they’re lifting up and dropping down the seafloor. For a typical subduction zone earthquake – so here’s our cross-section of a subduction zone, here’s our downgoing plate, here’s the overriding plate, and part of this massive fault will suddenly slip. So, what happens to the seafloor when that slip occurs along the fault? Well, the seafloor is going to be pushed up, and it’s kind of this thrusting motion, so that’s kind of obvious. Now, what’s a little less obvious, and it’s very important in understanding tsunamis, is on the landward extent of rupture, it’s going to be pulled down. The seafloor and the coast is going to be pulled down. So, you get this effect, what we call coastal subsidence. If it’s out in the water, it’s going to create this initial tsunami waveform. We call this an N-wave because it looks like the letter N. But it has particular hydrodynamic properties. Recall the local tsunami is propagating in this direction. So, what a person on the beach is going to see is the trough phase. But they’re not going to see the trough phase, they’re going to see the ocean receding first, before this peak wave coming ashore. Now, the transoceanic tsunami – remember, the one propagating over to the left – they might not get that natural warning. They might see the first elevation of the wave. And so, they might not see the recession of the ocean. So, if you see that, it’s a pretty good indicator that a tsunami is going to come. But it’s not always going to come. So, I have a series of tips that I throw in through these slides, I’m not going to test anybody on these, but useful background information. Tip number one is, natural warning signs include strong ground shaking and a receding ocean. And even in California, where we don’t have a big subduction zone like this, if you feel strong ground shaking and you’re at the coast, there’s always a remote possibility that a big landslide could have been triggered. So, it’s always a good idea to leave the coast if really strong ground shaking occurs. Now, tsunami propagation is a rich field, and I’m only going to touch on a couple different aspects. I’m going to compare the way tsunamis move through the ocean in this middle panel with how typical – you know, if you’re out in a fishing boat, these swell waves move through the ocean. Now, if you’re a scuba diver or snorkeler you know you don’t have to dive very deep to get out of this orbital wave action. It dies off exponentially. Well, tsunamis are a lot different. They, again, the wavelength of these waves is derived from this permanent deformation of the Earth that can be hundreds of kilometers long – 150 kilometers long in the case of Sumatra. That’s much longer than the ocean is deep. The ocean is only about 4 kilometers deep. So, it might sound silly because we’re modeling these waves in the deep ocean, but these behave as shallow-water waves. The main thing is these elliptical orbits are extremely elongate and, more importantly, these wave motions go all the way down to the seafloor. Everything you want to know about tsunami propagation is dictated by how deep the ocean is. And, as an example of that, we can see how a deep-water tsunami gets transformed as it propagates into shallow water. What happens is the velocity in the deep water is moving about the speed as a jet airliner. As it goes to shallower and shallower water, it’s slowing up. The wavelength is starting to compress, and, in order to conserve this massive volume of water, the amplitude has to increase. So as the tsunami goes from deep water to shallow water, it increases in amplitude. We have satellite measurements of the tsunami in the Indian Ocean, they’re only a meter high. Less than a meter high. But we know from field observations from Bruce that – and hopefully I’m not giving away part of the story, but they’re over 100 feet high when they hit shore. So, this amplification factor is pretty extreme. Now, just a little bit about this complex process we call tsunami runup. Again, contrasting this with wind-generated waves and tsunamis down here. We’re all familiar – we’ve gone to the beach and seen wind-generated waves come in – we’re familiar with how far ashore they wash. They come in as these nice, curling breakers a lot of the time that people like to surf. Tsunamis aren’t anything like that. Most of the time, tsunamis don’t even break. They come in as a very fast and strong-moving tide that’s going much farther inland than a typical tide or a wind-generated wave will. When they do break, they form these vertical walls of turbulent water we call a tsunami bore. So, if you’re standing on the coast, they actually might be hard to see. Because it’s not a breaking wave at all, it’s, like, a gradual elevation of the wave. But they’re still moving pretty – they’re moving slower than in deep water, but they’re still moving at 20 to 30 miles per hour. This leads us to tip number two. The rule of thumb is, if you can see a tsunami at the coast, it’s too late to outrun it. [laughter] Now, shortly after the devastating tsunami, the first thing scientists need to do is look at the regional earthquake history to see if the history of earthquakes in the region had something to tell us, as whether or not this event was expected. So, this is Sumatra here. Malaysia. The earthquake started at this epicenter – and I’ll show some images of how it propagated upward – but this whole green area is the part of the surface – the surface trace of the fault that ruptured during this massive magnitude 9 earthquake. Well, what happened before that is, the only earthquakes greater than magnitude 7 is this one that occurred in 1881. The Car Nicobar earthquake that did generate a small tsunami, 1 meter high, that was observed in India. And then a fairly enigmatic earthquake in 1941 up in the Andaman Islands. Now, most of the history – the earthquake history along this subduction zone – I should – this subduction zone involves this Indo-Australian Plate moving at a very oblique angle to a large conglomeration of plates. We’ll call this the Eurasian Plate. So, most of the earthquake history is written in the region in central Sumatra and further south from work – fascinating work done by Kerry Sieh. He’s uncovered evidence for a magnitude 9 earthquake that occurred in 1833 that was tsunamogenic, right across from Padang, population 860,000. This portion of the fault has not ruptured yet. This part, that last ruptured in 1861 in a magnitude 8.5 event, re-ruptured almost three months to the day after the December event, in a magnitude 8.6 event. And Bruce will be discussing some of the results from that. So, what does the tsunami look like from this December event? I’ll try to walk you through that. This is a northwest perspective. Here, you can kind of see the bathymetry in the background. Here’s a deep oceanic trench which is the surface expression of this subduction zone. These are volcanoes from the subduction zone along the Sumatra island. So let’s get the tsunami started. Here’s the initial tsunami wavefield. You can still kind of make out that N-wave shape. Here’s the peak of the N-wave and the trough. So again, this trough is propagating towards Banda Aceh and towards Phuket. And only about 15 to 20 minutes after the earthquake, the first waves you can see hit the west coast of northern Sumatra and Banda Aceh – a very short amount of time. As we propagate this through, you can see the waves propagating across the Andaman Sea. At about an hour, you can see this large wave coming towards Phuket right here. At about two hours, a satellite will pass – there it goes. And then [laughter] hits Sri Lanka and then India. I’ve been doing this presentation too many times. [laughs] Another interesting factor to look at is what we call trapping of tsunami waves. And waves will be trapped along island arcs or ridges, especially if there’s a long bathymetric ridge like you can see one right here. So, the waves will preferentially be amplified along these ridges. And what happened in this case, it amplified these ridges towards – these waves towards Myanmar. Conversely, along the trench, the waves will be deflected. And because of the curvature of this arc, there is, in effect, a shadow zone of tsunami energy. And this is right where Bangladesh sits. It’s a very low-lying, coastal plain that did not receive, fortunately, much tsunami damage. I think we can trace back why, because of this kind of shadow zone effect of the tsunami. So, okay, tip number three. At the end you still saw a lot of wave activity around all these coastal regions. A tsunami is not a single wave. The tsunami waves can last for hours. We have a tide gauge record here at the Visakhapatnam station in India where we can measure the tsunami waves, and they last for days. So, a tsunami is not a single wave. It’s very important to wait until all-clear is issued before you return to the coast. So, in a talk like this one has to adopt some sense of reductionism to set things out in its real basic terms. However, I really – in this slide only – I’d like to emphasize the rich field of complexity associated with earthquake rupture. And this is exemplified. Now these aren’t tsunami waves. These are seismic waves in Southern California. You can see an earthquake starting on the southern end of the San Andreas Fault here, and propagate northward into the Los Angeles Basin. So you can see how complex these waves are. On the right-hand side, what looks like fractal displays of landforms, like you can get with a lot of software programs these days, this is actually a fractal display of slip distribution. We know from USGS studies way back in the early 1980s that, after slip has occurred during an earthquake, it’s not just a simple, uniform slip. It’s very heterogeneous. So, what we have here is the length dimension of rupture and the width dimension of rupture. And you can see how strongly variable the slip distribution – and we can express this with fractal functions. So there’s many absolutely fascinating aspects coming out of this earthquake. And hopefully they’ll lead to a better understanding of protecting ourselves from earthquake hazards and tsunami hazards in the future. One of these studies that was recently published by Ishii and her colleagues in a recent issue of Nature is best described, I think, as an ultra – let’s see, let me get back to that – an ultrasound image from a very dense seismic network in Japan. And you can see some of this complexity also in the Sumatra-Andaman earthquake. It doesn’t slip uniformly. It goes through these pulses. You see one pulse away from the epicenter right here. And then another pulse up in the Nicobar Islands. So, again, we’re starting to fold in this kind of complexity into our tsunami generation models. So here is a model that encompasses the earthquake rupture. You can see how fast that rupture occurred in comparison to the way the tsunami waves are propagating outwards. This is, again, the island of Sumatra – a southeast view of the rupture, and Banda Aceh sits right there. So, again, we’re trying to fold in a lot of what earthquakes actually do into our tsunami generation models. Now, taking our space view of the tsunami, this was truly a global event. The first globally measured tsunami since the 1883 Krakatoa tsunami. This is a model of maximum tsunami amplitudes from our good colleague, Vasily Titov, at NOAA. They show where tsunami heights were essentially the largest during its propagation around the Earth. So, if you’re a tsunami scientist, you get these bulletins after a major event. And to get you some sense of the time sequence of this tsunami, of course we got all the reports back from India, Sri Lanka, and Somalia. But what the tsunami scientists got – we got our first report from the Japanese Syowa station in Antarctica. Kind of a harbinger of how bad this tsunami really was to propagate. I think it was the first tide gauge reading of a tsunami I’ve ever heard of coming from Antarctica. Well then, on its westward path, we got tide gauge readings of this tsunami in South Africa and then into Brazil, as it made its way into the Atlantic Ocean. It then split up into the middle of the Atlantic Ocean, and we got tide gauge reports of the tsunami in Nova Scotia. On its eastward path, the tsunami sped through this gap between Australia and Antarctica. And here’s the third – that second type of plate boundary was a mid-ocean spreading center. Well, this creates a big bathometric feature called the mid-East Pacific Rise. This acted as a wave guide during tsunami propagation and focused tsunami energy towards Central America. And we got a – especially for the distance this is from the earthquake – we got a reading of a meter high in Manzanillo, Mexico. So here’s a comparison of – we’re bringing this back home now to Brian Atwater’s backyard – here’s the Sumatra rupture. We have, again, our own subduction zone. This is where the oceanic Juan de Fuca and Gorda Plates are being subducted beneath the Pacific Northwest. And you can see it’s about the same dimensions. And what’s that kind of tsunami going to look like? This is one possibility. There’s different ways you can think of how this earthquake is going to rupture. You can see it started off very simple in our transoceanic tsunami that’s going to get picked up by this tsunami buoy. But as you can see, as it impacts the coast, it gets really complicated quickly. You see waves sweeping up and down the coast. And a disclaimer I have to make – and it’s going to sound silly – but these waves, I have to make these waves at a different scale than the topography. These waves aren’t as big as mountains. They’re big, but not that big. [laughter] So, what happened two weeks ago on the Gorda Plate? Well, this was not a subduction zone – the tsunami warning that we got two weeks ago – this was not a subduction zone event. So, we have our subduction zone starting from what the equivalent of a trench is, here. And so, the Gorda Plate is being pulled down beneath the North American Plate. Well, the magnitude 7.2 earthquake on June 14th occurred in the oceanic Gorda Plate. The Gorda Plate is really busted up, basically. And a strike-slip fault occurred here. These strike-slip faults, even though they’re in deep water, they’re very inefficient at generating tsunamis. You can see a small tsunami was generated from this. And it was recorded on the Crescent City tide gauge station. It’s about 3 centimeters. Very small. It was recorded – and that was about – a little over a half hour after the earthquake occurred. And on the tsunami buoy, it’s eventually recorded. But – and if you’re in the tsunami warning, in six minutes, all you know is the epicenter and the magnitude. You don’t know where – which tectonic regime, which fault it occurred on. So, they always have to be conservative, and so that’s why the warning was issued. So now what can we do in the Bay Area – and anywhere, really – to protect ourselves from tsunamis? The idea is a tsunami-resilient community. And it really involves three key elements. And you can’t have any one of these elements missing. Of course, what’s most obvious is the warning centers. And the warning centers get information from these DART buoys as well as the global distribution of seismic data as fed through the USGS. But, in addition to the warning, you need to do your homework and know where tsunami flooding areas might occur. And there’s a whole suite of inundation modeling studies that can be ground truth with tsunami deposit information like Bruce is collecting – and Brian. And then, probably the most important piece of this puzzle – and I’m glad so many people are here today – is the education component. To know the natural warning signs – to know not to go back too early. Of course, we have to work out all the pieces in terms of efficient evacuation planning so we don’t get any injuries or anything like that when an evacuation is necessary. The California – let me see if I can pull up one California slide here – central California. Okay, we have some – you know, we have these strike-slip faults – the San Andreas and the Hayward Faults. They do some odd things, and they split over. So they’ll generate small tsunamis. In fact, there was a 4-inch tsunami, for example, produced in the 1906 tsunamis and some fascinating stories inside the bay of possible tsunamis that were generated. But most of our risk is going to be from far-field transoceanic tsunamis. So here, very similar to that figure I showed from Vasily Titov, these are maximum tsunami amplitudes. And you can see, for this particular subduction zone in Alaska, where the 1964 magnitude 9.2 earthquake occurred, it’s focusing its tsunami energy towards Washington, Oregon, and central California. For this, we have several – many hours to get a warning out. But this is going to be – all these subduction zones around Alaska, Russia, and Japan can have a potential of sending transoceanic tsunamis that might hit central California. So, with that, I think I will switch over to Bruce, who, again, is going to discuss a lot of the – [laughter] not that.

- We'll take questions later, or ...

- Yeah, I think we’ll take questions later.

- Yeah, I think it would be best after you hear all of us. I better plug in the power into this.

- Oh.

- So we don’t run out of power in the middle. Thank you, Eric. I’m going to talk about my experiences going, after the December 26th tsunami, to Sri Lanka and to Sumatra. Although, I’m going to really be focusing in on Sumatra. And then, after I talk, Brian Atwater is going to be bringing this closer to home, talking about tsunamis in the Pacific Northwest. So, after the December 26th tsunami, there were a number of international teams that were formed by tsunami scientists. The idea was to get as much data as possible on this unique event in order – before it was destroyed by bulldozers, weather, etc. The December 26th tsunami was very directional. It went out and hit – here’s Sri Lanka here. It hit this fairly strongly a little bit over two hours after the earthquake. It also hit the Maldives island. But much of the – much of the damage, and much of the large waves were in Sumatra. The USGS participated on international teams in four locations. In Sri Lanka, since I’m not going to be able to go into detail on there, the tsunami waves there were up to 7 meters high. In the Maldives, the tsunami waves were up to 3 meters high. And then I will be talking about two teams that we participated as part of, and actually co-led, in northern Sumatra. So the purpose of these tsunami field surveys is to collect data on wave height, the runup, and the inundation distances. This is important data to be able to improve models that predict tsunami heights from earthquakes and other sources. To investigate modern and paleotsunami deposits. And, whenever possible, we collect bathymetry and topographic data so that we can have realistic topography and bathymetry for the models to propagate over. And, in cases when we go to an area where there’s a locally generated tsunami, we take measurements of the magnitude of the coseismic subsidence and uplift. So, as I said, the water measurements are very important for improving the models of how the tsunami inundates. They also are important in that they help with improving the understanding of how a tsunami loses energy as it goes inland. It’s – as a tsunami goes inland, it does decrease in height. And the rate at which it decreases will determine how large the zone of destruction is, and the zones of high casualties. The Japanese have studied what tsunami heights and flow velocities are deadly. And they’ve come up with a tsunami greater than half a meter high or moving faster than 2 meters a second is deadly. The studies that we do on these international teams of the sand deposits are important in that, when the tsunami leaves an area, when the water is long gone, the sand that the tsunami brings up tells a story about the tsunami. So, we can use these tsunami deposits – these tsunami sand deposits – to extend the record of tsunamis. When we have a series of these deposits, we can – and we can date them, we can estimate recurrence intervals for tsunamis. And we can also learn about the tsunamis from the characteristics of the deposits. We can learn about the height and the flow of the tsunami. So I’m going to show you a few slides from Sumatra. I’m showing the slides with before and after pictures because, unless you see the before picture, you don’t really get the full effect of what the tsunami did. In many places that we visited, all we saw were foundations where houses were totally destroyed and washed away. And this upper two frames shows that. And then another effect from the tsunami is that it flooded rice fields. And if you see this color here – it’s kind of a tannish/brown color – that’s actually sediment that was brought in. Sand and some silt and mud that is left when the tsunami leaves. Probably many of you have seen pictures of Banda Aceh, an area that was hard-hit by the tsunami. This is a mosque in Banda Aceh that wasn’t destroyed. Many of the buildings were destroyed. The reason why it wasn’t destroyed is – well there are several reasons. One, it was constructed very well. But also, the mosques have an open area on the ground floor so that people can go in and out of the mosque easily. And the tsunami wave just went underneath. The tsunami here was up to this level. Having been there, I still don’t comprehend exactly how large the tsunami was. And I can imagine it’s difficult for you, having not been there. One of the conundrums about tsunamis is that they are capable of moving very large objects. They obviously destroy large buildings, and well-built buildings, as well. But what you see, when you look in the aftermath of tsunamis, you see what is most prevalent, which is fine sand that’s on the beach. So, all this sand here was brought in by the tsunami, but it was obviously capable of moving that large barge. The tsunami in Sumatra surprised tsunami scientists in just how large it was. This area is about 50 miles, or slightly less, south of Banda Aceh. And the tsunami came in from the ocean here, and then it was strong enough that it went around and it went into these lows and was still very devastating, in this case, over 2 kilometers from the coast. Also, extremely powerful, in this shot here, these are I-beams – steel I-beams that are bent around a tree. When you’re on the ground, you don’t really get the scale of the tsunami inundation. As a matter of fact, we could – the tsunami inundates so far inland that we often couldn’t get to the limit of inundation. We chose to get the longshore variability of the tsunami in most places, and then, in selected places, went inland and got the cross-shore variability. But this is a before and after. This shot is from maybe 10 miles south of Banda Aceh. And you can see before, very green, there’s houses in this area. And then afterwards, it’s that same tan color, where the sand has been brought by the tsunami, and the tsunami has completely destroyed all the houses. So, the types of data that we collect during these surveys are eyewitness reports – you have to be careful that you don’t lead the eyewitnesses into giving answers that you want. And you also have to have a number of eyewitnesses giving similar accounts to verify that they’re accurate. Water levels. Flow directions. Water level and flow directions are very helpful for verifying the inundation models. Water and sand inundation. And then the topography and bathymetry profiles, again, helping with the modeling of the tsunami. And then, when there’s coastal – when there's coseismic subsidence and uplift, we measure that. And then we also look at the morphology. There were places where there were reefs offshore. And it’s difficult to separate out what the effects of the near-shore bathymetry focusing and defocusing the wave into areas from what’s in front of an area of coastline, but we’re going to attempt to do that. And then one area – the area of research that I’m involved with is looking at the deposits. And we measure the thickness, and we take many, many sediment samples to learn the properties – characteristics of the deposits. A little terminology will give meaning to some of the measurements that I’ll be reporting on later. The tsunami height is just the height of the tsunami above the ground level. As you can see, it typically decreases inland. And then, the inundation distance is the distance from the shoreline to the limit of tsunami flooding. And then runup elevation is a measurement that’s fairly easy to take. It’s just the elevation of the land at the limit of inundation. So, first, some eyewitness reports. In Banda Aceh, one of the surprising things was that the earthquake shaking wasn’t destroying the buildings – that the tsunami was causing the destruction. Because I think one of the thoughts before this earthquake was that there would be a lot more earthquake damage. There was some earthquake damage, but it wasn’t extensive. And then, one key point is that the tsunami arrived 15 to 20 minutes after the earthquake. The warning in Banda Aceh and in northern Sumatra was the shaking. People who knew that there was potentially a tsunami after that shaking, and were able to get inland, survived. And, an even more important observation that was relayed by eyewitnesses was that the tsunami did not move in and then come out and have – each of the tsunami waves wasn’t independent of the others. The second wave rode over the first, and the third rode over the second. The reason why that’s so important is because you have additive effects. And, right now, that’s not something that’s included in tsunami models. Well, one of the things that – or, one of the mantras of these tsunami field survey teams is to get in quickly after the event. We wait until there’s been adequate time for disaster response. We don’t want to get in the way of the response – the initial response. But afterwards, we need to get in quickly, because the evidence becomes altered with time. This is January 22nd, and you can see this is about – hard to tell scale here – but that’s about 20 meters high – a cliff where all the vegetation is removed. The tsunami got up to that level. And then this is a picture just a few months later where there isn’t the same thing on this cliff that’s further back from the ocean. But it already has grass growing on it, and you can see that the grass is already growing on the area where the tsunami left the sand. So, the bread and butter of the tsunami field teams is water levels. And fairly simple – once you have been on a tsunami survey, you know what to look for. Look for water marks on buildings if any buildings are surviving. And if you’re fortunate enough to have trees that survived – and in Sumatra, even though the waves heights were extremely large, there were trees that survived. You look for broken branches, stripped bark, debris on the trees, and get water levels. Or you can do the same on smaller trees and shrubs. And to give you an idea of what it looks like on a tree up close, the tsunami level went up to this – the tsunami went up to that level. There’s a debris pile here that was piled up. And the bark is stripped here up to that level and broken branches up to that level. So fairly easy to tell how high the tsunami was. So one of the things that we focused on on these international tsunami survey teams, and something that’s a recent development, is that geologists have become more involved and are focusing on the tsunami deposits trying to build relations between modern tsunami deposits and the flow, so that we can then use that information to interpret paleotsunami deposits. So this is all a tsunami deposit – it’s very extensive – about 100 kilometers south of Banda Aceh. And the types of things we do with tsunami deposits – we do a lot of digging to be able to see the deposits. We do sampling. Some of our sampling is very detailed to the sub-centimeter level to get the vertical variation and the grain size. And that gives information about the flow. And then we also need to describe the deposits. And we can bring home the deposits, taking a peel of it. And there’s a peel there that – thank you, Brian – several peels. These are two peels from Sumatra. And I’m going to be talking about this peel in detail a little bit more later. Let me put those back. Otherwise, I can’t … [laughter] I only have two hands. I’m not going to use my nose. So, how large was the tsunami in Sumatra? I don’t think it’s really sunk in to me, and I don’t think many people know how large the tsunami was in Sumatra. This is Banda Aceh. This is where we saw a lot of footage. And there, the tsunami was up to 10 meters high – slightly over 10 meters high. The tsunami had to come around the corner, around this headland, and so it’s smaller there. What we didn’t see in any footage was the west coast here, where the tsunami had a more direct impact. And these levels are up to 30 meters high – so 100 feet high. And that’s not just in a localized area. It’s pretty consistent. There is variability, but it’s consistent over a fairly long stretch of coast. Okay, I’m going to – and then, in a later field survey, we extended down into areas where we didn’t have measurements from the first survey. And, for scale, this is 5 meters here – this bar. So, you can see that there’s a large stretch of coast, perhaps 100 kilometers, where the tsunami was 20 meters high or larger. And then, it did decrease, but they were still large on the island of Simeulue, and by, the time it got to Nias, it wasn’t as large. That’s actually pretty close to what your animation shows, so that’s good. I should point that it’s very appropriate that Eric and I are both talking on this, because I think the – and Brian as well – I think that the real future in tsunami research and in understanding tsunamis is going to be collaboration between modelers, geologists studying tsunami deposits, geologists studying tectonics. So, a bit about the water levels. These are places that are about 50 and 70 kilometers south of Banda Aceh. And, at Jantang, here’s a situation. It’s hard to see, but this is the profile. There’s a cliff here, and the tsunami was still moving fast enough that it actually was piled up against this cliff. And it’s at about 15 meters-plus, above the ground level, so extremely large. That’s about 600 meters inland. At Kuala Meurisi, that was – it was even – it already – it had started dropping down in height. But at a mile inland, it was still 30 feet high, so just huge. Much larger than anyone expected. And, I’ll show you a bit about the tsunami deposits. This is, again, a satellite image. You can see that this tan area here is where the tsunami deposited sand. And, in cartoon view, there’s – what we typically saw was a zone of erosion, which is what people typically think of a tsunami being, is an erosive agent. And there is that zone here – the coast, where there is erosion. And then the deposit thickens. And it is fairly complicated. The simple model for a tsunami deposit is, it thins landward. It’s very much a sand sheet. And our observations from the December 26th tsunami is that it’s a more complicated picture than that. It’s going to be very challenging – I think also very rewarding – getting into the complexities of how the deposit changes. So, it’s 19 centimeters here, 250 meters inland. About 350 meters inland, it was 3 centimeters. And it was up to 33 centimeters back at around half a kilometer inland. And when we planned this trip – the second trip was planned back in the end of February. And we planned to be there the 30th of March. And we didn’t know that there was going to be another earthquake, which actually did generate a tsunami. So our efforts were redirected into documenting that tsunami. It was probably a – definitely a once-in-a-lifetime opportunity. He had a boat, we had all of the scientists there. There were 10 scientists from Indonesia and the United States – and did very rapid measurements of the tsunami. Very fresh evidence from March 28th. This was a much smaller tsunami. Here’s your 5-meter scale again. It got up to 4 meters in southern Nias. This is still a large tsunami. It’s a deadly tsunami. Silver lining to the tragic December 26th tsunami, is that the awareness was raised so high in Indonesia – and I think in the world, in general – that once the shaking stopped, people on these islands and on the mainland went inland very quickly. And there were no deaths from this tsunami. And I’m sure there would have been many deaths if there wasn’t that awareness. And there’s the deposit – the peel that was taken from an area that where actually two tsunamis hit. And you can come up afterwards and look at that. It’s pretty obvious that there’s two tsunami deposits. They look different. There’s a layer of vegetation in between. And, the interpretation is December 26th tsunami and March 28th tsunami. So why am I excited about that? Well, this is the first time that I’ve ever heard of or seen two deposits on top of each other. And really, studying this will help us look at the paleo record, where – looking for older deposits where you need to do coring and look at outcrops. But it’s going the direction that I think we want to go with linking the modern to the ancient. And learning more about the tsunami risk in areas where there is a short written record of tsunamis, which is pretty much worldwide, except Japan. And so, one thing that we did on this last trip to Sumatra is that we did reconnaissance looking for paleotsunamis in these promising areas that Indonesian scientists are going to go back and see if they can determine whether or not they’re tsunami deposits or not. So, a quick summary. From the December 26th, 2004, tsunami, the tsunami heights near the coast were much larger than expected. There were maximums of 30 meters on northern Sumatra and greater than 15 meters for more than 100 kilometers of coast. Tsunami heights were greater than 10 meters at 1700 meters inland. Very large tsunami. And extensive tsunami deposits were formed. And now I’m going to pass this off to Brian, who’s going to talk about the Pacific Northwest tsunami. And … There you go.

- Well, in a very understated way, Bruce has discussed the horrors of the December 26th event and its successor. The story I have to tell is one that’s quite a bit more fun at the outset. It’s a detective story. And for those of us who worked on this detective story, it was a great joy to discover that the Earth can do amazing things in northwest United States. Illustrated here by this computer simulation by Kenji Satake of a snapshot of our most recent big tsunami, in the year 1700, 10 hours after the earthquake that was associated with it. So, it really is – it really was fun to work on that. But we were insulated from a lot of things by the passage of time. Time had cleaned up the record of this event. We could see evidence that Native American fishing sites had been overrun. We even found a basket fragment at the position of a tsunami deposit and wondered, what became of the maker? But, it’s not the same as what Bruce has just shown you and what Eric has discussed earlier in simulations. It’s just astonishing. So, there’s a very – there’s a disturbing aspect about this. First, the one about something like this really happening on our coast. You know, in the lower 48 and adjacent Canada. And the kinds of losses that that could bring about. And then on a more global scale, there’s the aspect of this detective story being one that enables us to anticipate a future tsunami or earthquake of magnitude – say, an earthquake of magnitude 9, even, and the tsunami associated with it, so that we won’t be taken by surprise in the way that people around the Indian Ocean were. But that in itself is a disturbing thing. Because you can – it’s horrible that there are parts of the world where it hasn’t yet been possible to do this sort of detective work we’ve been able to do here to anticipate these disasters. So, it’s not just the Indian Ocean, you know? You could think of other coasts along South America, for instance, where a lot of people are ill-prepared for a tsunami, in large part, because they don’t know the full extent of the risk. Well, as for the fun part of the story. There’s a Japanese element to it, that you can see – the tsunami hitting Japan about 10 hours after it starts in this snapshot. Hence the Japanese translation of the title up there. The first four characters are phonetic ones. They say “minashigo,” which is a politically correct way of saying orphan. And then the next two characters say “Genroku,” which refers to – officially to time between 1688 and 1704. And it’s a sort of a cultural high point of Japan, of the Tokugawa shogun time of Japan, which is between about 1600 and our Civil War. And then the last few characters there are “tsunami.” Well, we’ve had a comparison earlier. I just want to emphasize it. The 2004 rupture area there – the area on the fault plane where the slip occurred. That’s approximated from the Iishi paper that Eric referred to. And then the – probably exaggerated in terms of its width – but I tried to take the 1906 rupture and turn it on its side. And so that gives you a fault rupture area roughly for 1906. And so, you can just get some sense of a difference right there. The average displacement on the 2004 rupture, what’s it estimated to be? 10 meters, something like that?

- [inaudible]

- More like 5. In that case, that’s similar to 1906 San Francisco. But in any case, a very big difference in rupture area. And then there’s an estimate of the size of the 1700 Cascadia rupture. So you see we’re talking about some very, very big earthquakes. And I think many of you know that these don’t happen very often. In the 20th century, there were just three earthquakes of certified magnitude 9 or larger. They happened in the cold days of the Cold War in 1952 in Kamchatka, and 1960 Chile, 1964 Alaska. And then a great drought in such earthquakes until Sumatra-Andaman. Well, the kind of fault that gives rise to these kinds of very big earthquakes mentioned are these boundaries between tectonic plates where one plate slides under the other – a subduction zone. The northwestern U.S. is graced also with two other generalized categories of earthquake source. The one you’re familiar with from – perhaps from an earthquake in the year 2001 – February 28th, 2001 – the so-called Nisqually earthquake and its predecessors. Those earthquakes happened in the subducted oceanic plate at depths of – oh, something like 55 or 60 kilometers, underneath the largely metropolitan areas inland. They’re the things that people have been designing buildings for for years. They don’t tend to make much in the way of tsunamis unless they set off a landslide that splashes in the Puget Sound. About 1,000 years ago – 1,100 years ago – an earthquake on a fault that runs underneath downtown Seattle raised some parts of the landscape 6 meters and dropped others about a meter. Those raised and lowered areas crossed arms of Puget Sound and set off a tsunami inside Puget Sound. Quite an impressive one. So, that is another kind of tsunami source. And then the 1700 earthquake is just the most recent of the very big earthquakes on our plate boundary. And in the lower diagram, I’ve just removed the North America Plate so you can see down onto the fault rupture plane so that the –  and the red patches on this diagram here are equivalent to that kind of dark red thing that – where it says plate boundary. So, it’s that shallow part of the plate boundary that’s capable of fault rupture like this. And the estimate for 1700 is that the leading edge of North America, on average, lurched about 20 meters westward during the – during that earthquake. Along – you know, 60 feet along about 600 miles of coast. Well, you wouldn’t know about that from European written records. [laughter] And this map is made during the Enlightenment when people are paying attention to observables and not doing much in the way of conjecture when they’re making maps. So, here’s the king’s cartographer in France, compiling in 1720 this map that was made public in 1724. The paper copy is somehow in the University of Washington library. But anyways, you can see the dotted line in the upper left-hand corner. Quite a few blue dots missing up there. And, you know, there’s nothing known to the Europeans north of Cape Blanco. There’s the explorations of Cabrillo and Vizcaino – more recently in 1602 it would be with Vizcaino – going up about as far about as Cape Blanco, and north of that, nothing known. So, you have to look into either Native American legends, which say something about catastrophes in the past, but not in a very exact way. A lot of it couched in myth or obscured by changes in the telling over time. And then most of those stories probably lost to smallpox. Then, other than that, you’ve got to resort to, you know, organizations like United States Geological Survey. And the USGS geologists – well, at least, in my case – I’m just a mud geologist. I started out working here in Menlo Park in the early ’70s, and I – the person I was working for, a guy named Ed Helley. He said, well, go down to that warehouse. There are a bunch of bridge cores from the California Division of Bay Toll Crossings. Any of you remember that? And, in the – in the heyday of Bay Toll Crossings, they had proposed these things like the Southern Crossing. And the bridges would be proposed. And the engineers would do their foundation studies. And then the things would get shot down in referendum. And the geologists would have a good look at what the history was of San Francisco Bay. So, it’s a pretty good racket, you know? So, I was involved in that and got interested in using sea level changes. And Eric emphasized, with his N-wave and so on, that you have these areas that drop during these big subduction zone earthquakes. Well, that’s just like a sudden rise in the level of the sea. So, for a person trained in looking at sea level change in San Francisco Bay in the early ’70s, it was pretty straightforward for me to go and see this kind of thing in the northwest. But I needed a geophysicist – a seismologist – to tell me there was a problem to begin with. So, there were a number of seismologists, Tom Heaton in particular, with the USGS at the time down at – based at Caltech, working for the Nuclear Regulatory Commission as a reviewer of Washington Public Power Supply System reactor – or, construction proposals. And Tom said, you know, you get big earthquakes up there. And so he was among the people who inspired geologists like me to go out and see whether they’d happened. And the abrupt land level change – the abrupt lowering of coasts – with an icon off to the right there with – those are stumps and tall trunks of western red cedar that have been standing dead for 300 years in a salt marsh. And, you know, either the sea rose or the land dropped. And you can figure out, with some geology, whether it happened quickly and so on. And you end up with a story that, yeah, land repeatedly dropped abruptly in the Pacific Northwest. The tidal marsh geology there looks completely different than San Francisco Bay’s. Nothing comparable. And all those yellow dots are places where geologists of different stripes have found this kind of evidence. And we all have incentives to blow one another out of the water, you know? [laughter] And so I don’t think – there’s not much of a bandwagon effect here. The Oregonians – some of the Oregonians would – they would – they would have blown the whistle on us if we’d had it too wrong in Washington. Now, the yellow dots here are sand sheets. And Bruce has introduced these with his examples up front here with the peels. In fact, there’s a little narrow peel in the middle that – I get …

- [inaudible]?


- It’s okay. So, this was a – this was a Cascadia peel that Bruce brought along from the place with the red cedar snags, actually. Not from this one. This place here in Oregon, you can see a firepit. This is an archeological site, a Native American fishing camp overrun by a tsunami. That gives you the sand layer there above the firepit. And the people had been living on sand dunes. And then the tides afterwards, because the land dropped, brought in the mud. Okay? And made the site continuously unusable after that, okay? So you have there both – you have  the evidence for subsidence and, at the time of subsidence, before the first of the mud could be laid down, the sand sheet was brought in. So that’s an argument that that’s a tsunami and not a storm. So anyways, that kind of evidence – not necessarily with the Indian firepits – but up and down the coast. Now, by the mid-1990s, thinking of this as a detective story … [laughter] Let’s see. We want to start this detective story about 1980, at which time the idea of a Sumatra kind of earthquake in the northwestern U.S. would have been scientific lunacy. It was. At that time. Just inconceivable. And Tom Heaton met a lot of abuse by even proposing it. And then, as geologists started to find evidence that they happened, then sort of the next line of retreat was, well, okay, they happened, but the whole thing can’t possibly break at once. And it was really hard to tell. And geologically, we couldn’t tell because we were using tree death as a signature of earthquakes,  so we couldn’t distinguish between tree death happening from an event that lowers an entire coast in five minutes, from tree death that lowers a coast piecemeal, through a series of a couple of events, as in the breakfast link alternative there. Or that was dubbed by one scientist, the Decades of Terror hypothesis. [laughter] So, you know, you have northern California, say, go first. And then Oregon and Washington wait for the shoe to drop. And, you know, you can think of the December 26th/March 28th sequence in regards to these. So, it turned out to be evidence from a somewhat surprising source, but, in hindsight, it’s not very  surprising. After all, the tide gauge at Fort Point recorded the very first – it provides the very first instrumental recordings of an earthquake in the form of its tsunami. From 1854 Japan – little waves that were recorded on the – here at San Francisco,  also in San Diego and at Astoria, by these newly installed,  self-registering tide gauges. So, you know, trans-oceanic tsunamis happen. And Japanese scientists learned – well, let’s see, I need one more prop. Okay, so this is just a root of a Sitka spruce that was a victim of the land dropping at Mad River Slough, Humboldt Bay, California, about 300 years ago. And this is a coping saw cut, and there’s another coping saw cut. And  these were – those were samples given to a radiocarbon lab. And in the end, what the lab was able to tell us was that this – the tree had died sometime between 1680 and 1720. That’s as much as we knew about the time of our most recent earthquake in the northwest. Now, from Japanese written records, it appears instead that this happened on Tuesday evening, the 26th of January, 1700. And it’s remarkable to have that kind of – that’s okay – precision. But, that’s part of the evidence for the dinner sausage model working in 1700. Now, previous events may have well been breakfast links. We don’t know. But in 1700, with the subduction zone, clearly dinner sausage. And so that came from these kinds of records. Really amazing. And, in the example shown up at the top, there’s a map of Japan with Hokkaido shaded a slightly darker color because it was then in the hands of a native people, the Ainu. The blue dots are not democratic strongholds, but … [laughter] But places where the tsunami was –  where written records of the tsunami are recorded. Oh, and then, the top  blue dot is actually three blue dots. And the political leader you see there, next to him, his name is – Nambu was his family name – Nobunao. And he was sent up there in exchange for control over the land. His job was to pacify the local farmers and assess their taxable land. And you see around his belt, he has demonstration that he knows how to do his job. So, you know, and for Eric and for Bruce and me – you know, I published this little booklet here. And it’s got my name on it. It’s my reprint. Well, it’s like – It’s like that  guy’s head, right? [laughter] It’s his way of showing he does his job. But anyways, he pacified that territory, and so the rest of Japan was pacified under the – finally – under the Tokugawa shogun – the Ieyasu. And a period of stability ensued that included Genroku time around 1700. And everybody in – all these feudal lords, like the Nambu family, they emulated the Tokugawa people in Edo, now Tokyo, and kept good records. If they didn’t, they might lose their lands. And so, in a library attic, there you see [inaudible] volumes of the written records from the Nambu domain. These are administrative records kept by out-of-work samurai who had nothing to do but be bureaucrats. [laughter] And – because it was pacified, you know? And, now, at bottom is one of the people who then collected this stuff. And it’s not like, in the mid-’90s, Japanese scientists learned about a – our evidence for a very big earthquake, or a series of earthquakes. We didn’t know which, whether it was breakfast links or dinner sausage. And then they started looking for this stuff. No, they’d been working on this for decades. And before World War II, this high school – or, yeah, I guess he’s a high school geography and English teacher named – family name Musha – was working as a volunteer for the Earthquake Research Institute of the Imperial University of Tokyo, now the University of Tokyo. And the three green volumes at right, he issued during the war as mimeographed forms – handwritten things, mimeographed. The blue volume survived the fire-bombing of Tokyo in a galvanized steel box lowered three meters below ground in a hole dug in the back of a seismologist’s yard in Tokyo. And he later – Musha later worked for MacArthur’s people. He died, I think, around 1961. But he collected two or three of the – two of the – he collected  two of the written records of the 1700 tsunami in Japan. And then his successors came along and filled out those other blue dots. So, by the time we, over here in North America, were working out the North American side of this detective story in the middle ’90s, Japanese scientists paying attention to that could go back to their colleagues and say, what do you have from this time? Because, you know, 1680 to 1720 is not very long ago. And it falls in the time of these enormous volumes of written records. And they say, oh, that’s the one tsunami that we can’t find a home for. They’ve found home for tsunamis in Japan that came from Peru in 1586, Peru in 1687, Chile in 1730 and 1751, Chile in 1837. So, they knew about transoceanic tsunamis, especially from 1960 Chile, because that killed 130 people in Japan. But they didn’t – but they couldn’t find a home for the 1700 thing. So, they were really pleased. And they said, aha, this is the source. Well, we wanted – we wanted to  make sure of that because we just had – because a lot was going to hang on this. They said, okay, the flooding and damage was so great, you had to have tsunami heights between 2 and 5 meters. In Japan. From your tsunami. Okay? In 1700. And it’s hard to do that with a magnitude 8. They said, you got –you need a magnitude 9 earthquake. So, they voted for dinner sausage. Okay? [laughter] And they said, you know, your breakfast link model won’t do that. And so we needed to check that. And there are two sort of warring tree ring scientists who did this work – one at Lamont at Columbia University, and the other out on the West Coast, David Yamaguchi. And David did most of the work. Let’s turn off – well, let's see. Before you – Sam, I’m going to ask you to turn up the lights, but not quite yet. There is a tree called Witness that’s up on the top of the hill. And that’s one that lived through the earthquake. How’d it live through the earthquake? It was just high enough that the subsidence during the earthquake did not lower it into tidewater, okay? And then, it’s got a relative down there labeled Victim. Okay? And that used to be living dangerously just above high tides. Okay? And then, after the earthquake, it found itself covered by high tides. Hence, the salt marsh, and it’s sticking up through the salt marsh, so that’s the picture over there. So, we’ll just look real quickly at examples of those. So, if we could turn up the lights? Great. Okay, so the real sample this is  supposed to – that’s supposed to be here for this is somewhere in the baggage system of Southwest Airlines. [laughter] But, this works better because you can actually see the rings from wherever you’re sitting. [laughter] So, thank you to Southwest Airlines. And so, this stuff out here is supposed to be bark. And it’s that Witness tree, okay? Cut by Weyerhaeuser in the spring of 1987. And the real sample I was going to bring you goes from 1980 – a complete ring in 1986 all the way back to 1439. Okay? So it’s a really cool sample. And tree ring scientists went on – David – went out on the heels of these loggers. And he cut these radial pieces out of the stumps. And what was he doing? He was looking for a bar code, of good or bad years. So I tried to cartoon that here. Okay? There are some rings that are narrow and some that are wide, right? Okay, and the trees – if it’s a good  year for western red cedar in one place, it’ll probably be a good year for the same species in a nearby place. You know, just because of rainfall and fog and temperature – whatever makes the tree happy or not. So, you’ve got unhappy rings here and here and, you know, they give you this master barcode, okay? So David was able to construct a master barcode from 1986 back to, I think, about the year 900. Oh, sorry, those in the back need to see this better, okay. Okay. So anyways, just using this pattern of narrow and wide rings, so you can just think of it as a supermarket barcode, okay? And then – and then he went to a victim tree. And this is a slice from a victim tree. This was the outer edge of the victim tree. And those of you up close can see the rings terminate against the weather-beaten outside, okay? Because this has been 300 years of wind, and rain, and rot, and birds, and fire, and all sorts of things attacking the outside. And then inside, the ants have been busy eating away the wood. So, these are in kind of rough shape. But the tree ring scientist then measures these up and then slides this around to see if there’s a place where the patterns match. Okay? And then asks, is that the only place where the patterns match well? Okay? And then if that’s the unique match, then there’s some confidence that the dates that you know from the tree that Weyerhaeuser cut in 1987, okay, that those dates can be applied to this sample. Okay. So, they did that and then the geologists get in the act with their shovels, and they dig down to the roots –  to the bark here on the roots. Because that takes you out to the final year the tree was alive. And the remarkable thing was that, at the – Sam, you can turn the lights down if you like. That – at the four yellow dots shown up there – that four different estuaries, Copalis River, Grays Harbor, Willapa Bay, and Columbia River – that the victim trees at those four dots, the bark-bearing roots all showed the complete ring from the year 1699 as their final ring, okay? And so, the growing season runs through August and maybe resumes – a new growing season starts in May. So these trees lived through August of 1699. They’re dead by May of 1700. So, we’ve asked the trees, could you have died from an earthquake in the year 1700? [laughter] In January of 1700? And they said, yes, we could have. [laughter] So, that was the best – you know, the best test we could throw at it. We can’t really prove that that’s when they died – you know, from January 1700. We can’t prove that. But we sure had a good opportunity of showing that they died in 1697. You know? Or in 1703. And so, you know, it’s the failure to falsify, usually, that carries the day in this kind of hypothesis testing. So, there was this added piece of the detective story, anyways, that strengthened the link with the Japanese written records that had given us an exact date – the reality of an exact date. It’s deflating for a geologist because you struggle to nail things down with geological dating. You say, well,  you know, about 300 years ago. And the earthquake just doesn’t sound as credible as an earthquake that happened at about 9 p.m. on the 26th of January, 1700. [laughter] And so that’s what – you know, the Japanese written records gave Cascadia that kind of credibility, okay, of an exact – some sort of exact intersection with human history. And then also gave support to the dinner sausage model there. Okay, so repetition of these sorts of things. Bruce mentioned using the deposits as signs of repetition. There are sand – there are beautiful records of repeated tsunamis. The best of them published so far is in the brand-new issue if the Geological Society of America bulletin from a lake in the south central Oregon coast near Coos Bay called Bradley Lake. And it’s a record that goes back 4,600 years. And it’s a record of repeated tsunamis. I think all of them the authors believed to be from very big Cascadia earthquakes. Not all of them do they believe are full-length ruptures. Some of them they believe are partial – broke only part of the length of the fault. The graph I’ve shown you up here is for repeated earthquakes in south – on the subduction zone in southwest Washington. And the width of each yellow bar just says how poorly I’ve done my job. The wider the bar, the less  well I know the date. Okay? And so the 1700 event exactly dated over there at far right – the little skinny line. [laughter] And then – and you can see I did pretty good – I got pretty lucky with the  antepenultimate one, you know, the third to last – quite narrow. And you can see – you can see immediately that the earthquakes recur at regular intervals of 500 years. 300 years have passed from our most recent one. We have 200 years of bliss without any worry, right? No, I mean, to the contrary, the record is – the repetition looks quite irregular in time. And you can see precedent in this record for back-to-back earthquakes just 300 years or so apart. And so, here we stand 300 years from our previous earthquake. And we don’t know whether the next one’s, you know, going to happen tomorrow or two centuries or five centuries in the future. But there’s precedent for back-to-back for repetition, so it’s a good idea to prepare. A simple way to boil this down is to assume that the earthquake – that the subduction zone has no memory of when its previous earthquake happened. This is not a very satisfying assumption. It’s one of ignorance. But it may be the one best justified by the evidence that we have at this point. And with that assumption, a geologist working for the state of Washington has compared the odds of one of these happening with the odds of death from a car crash. And he has concluded that you’re eight times more likely to be on the  planet when one of these happens than you are to die of a car crash. So, you know, you wear a seat belt, and that’s why. But anyways. Whoops. So, here’s Santa Claus climbing a roof near a tsunami evacuation sign. These signs have sprouted up on the coast of – I think there’s some in Crescent City. Are they? And some of you have seen them. And then they’re along the Washington and Oregon coasts. They’re popular items to steal, because they look pretty good. The county emergency – well, the state people – they’re so costly, they’ve gone ahead and printed paper copies at full size and they hand them out and hope that people use the paper copies as wall decorations. [laughter] So, I mean, but it’s a serious problem for them. And there are a host of serious problems with these evacuation routes, when you get into the nitty gritty of it. Like you point a sign up a logging road, and then the timber company says, we don’t want the liability, and they throw a locked gate across it. Okay? And this is – you know,  these are real world issues that are really tough for the emergency managers to deal with. But nonetheless, you know, this – when the Sumatra-Andaman earthquake and its – and the horrible Indian Ocean tsunami that followed, when all that happened, a lot of this was already in place, and much more was in the works. And it was because of the detective story that I’ve just told you, that these precautions have been taken. Another precaution – a little hard to read here, but I think you can see. The green I used for high ground is not quite the same as the blue for the Pacific Ocean, okay? So those are those stripes coming down. Those are beach ridges. And that’s an area where Bruce is – near an area Bruce is doing quite a bit of innovative work on trying to link  tsunami models with the sediments. But you can see the challenges that the county and the city emergency people face here. You can see the roads where people live and, you know, how they’re going to get people to high ground, especially from nursing homes. But these sorts of maps – these evacuation maps that are based on the assumption of a magnitude 9 earthquake like Cascadia, okay – So this is that detective story coming kind of full circle and being used in a practical way to put lines on a map to guide people as to where to go. These sorts of maps – evacuation maps – are now available for nearly every sizable town and Indian reservation community, too, in the states of Washington and Oregon. Well, there’s one final element that touches on what Bruce did with the international – with the surveys he participated in. And that is that he and his colleagues opened up communications with scientists in Indonesia and Sri Lanka who were pretty much isolated from the kind of science that we’re able to do here. And that kind of opening is just essential for these countries to take their own precautions with inundation mapping, with the geology that can help to support that. And as a further effort, what we did in February was got some USGS funds to bring these four people to work in the south of Chile, in the region of that 1960 earthquake, to study the geological records of predecessors to the 1960 earthquake and its – and tsunami. And this is an effort that we hope also to continue in the future. Thank you.


- We’ll take questions, now. There are microphones on both sides. We’ll get them turned on. And we’ll just kind of alternate between those microphones taking questions. If for some reason you can’t get to it, raise your hand and I’ll bring you this one. We’ll start with the microphones, since I said we would

- When a tsunami is passing over land, is it a vertical wall of water? Is it noisy? What does it sound like? And the second question. A plate being subducted is said to be pulled under. Is that more accurate than pushed? If so, why?

- For the first question, part A, it depends on how big the tsunami is. It comes in like a very fast-moving tide. So, it’s somewhat gradual. If it breaks, it forms this vertical wall of turbulent water. And as Bruce described, in some places it might be doubled up or tripled up like that. There was a very fascinating Japanese study who looked at the sounds tsunamis make. And they make a variety of sounds. As they hit cliffs, they make one kind of booming sound. If there’s a gravel beach, the backwash of the tsunami will make another very loud sound. So, these have been reported in Japan and kind of been classified. And it’s extremely variable depending on what the geology is like there. And then the second part of your question, again, was – oh, the subduction zone. It’s a constant controversy – well, I don’t think it’s that much of a controversy anymore. I kind of err on the side of slab pull. But there’s ridge push, slab pull. From what I know about it, slab pull seems to be favorable. But, there’s probably both forces going on.

- And I can comment from eyewitness reports. In some of the tsunamis, there are reports of a noise preceding the tsunami going on land or even while it’s going on land. And there are also descriptions sometimes of it being nearly a vertical wall of water. But at other times, it is more like a fast-rising tide. So, it’s variable.

- Is there any kind of connection between the Indonesia earthquake and the bunch that we’ve been getting in California?

- We don’t have the right person here to answer that question. [laughs] But probably not is the right answer, I think. There is – it’s an active issue in seismology. Long-range triggering. I think they first saw it with the Landers earthquake a few  years ago seeing these earthquakes – I do say, in this issue of Science,  it describes a lot about those Sumatra earthquakes. There’s an article about triggering in the Alaska region. They think that’s triggered from this earthquake. So, it could exist. Whether these earthquakes that we just had, I kind of doubt it, but you never know.

- [inaudible]

- Sure.

- Okay. Part of that puzzle is that cluster of magnitude 9 earthquakes. That’s a genuine cluster in the middle of the 20th century. And then the dearth of them, all the way through the rest, until 2004. And that – you know, this is Kamchatka, Alaska, southern Chile. So, you know, they’re not very close in space. But somehow, they came together. Nearly all of the seismic energy radiated in the 20th century was done at that time.

- What was the final magnitude of the December 26th earthquake?

- From a well-respected seismologist here at the USGS, I don’t think we’ll ever know. [laughs] It really depends on what you call the final magnitude, whether – and it’s depending on whether you’re looking at the kind of earthquakes we’re normally used to feeling – the short-period waves, the very long-period waves going to this vibration of the entire Earth, or static deformation – just is permanent deformation. So right now, it ranges between 9.0 – I think that’s still the official USGS magnitude – to 9.3. I think most people are kind of focusing on 9.15 to 9.2.

- But – [inaudible] that microphone, but the … Let’s see. If you make a comparison between the Sumatra-Andaman event and the 1964 Alaska event, using the same criteria, ’64 Alaska is still bigger.

- Hello. Yeah. I was wondering – it appears that you all had information that these big tsunamis were coming from subduction regions. So my question is, apparently this Sumatra area was overlooked or something. In other words, there was no detection equipment, apparently, in that area. Which I remember from the news reports. So, is that quite true, that you’ve known about these subduction zone correlations for a long time? Or was this – you know, cemented the evidence, in a sense? And why weren’t there – why weren't  there things in the Indian Ocean there?

- It’s an excellent question. They were overlooked, and I’ll show you why, hopefully, in a slide. First …

- We’re going to have to  get out of the way. First looking … The picture of the Sumatra subduction zone. This is what we call a very oblique subduction zone. So unlike Cascadia, where the oceanic plate is being pulled or pushed directly at the continent, here it’s moving at a very shallow angle. So, right at this epicenter here, it’s moving enough to cause a thrust motion, but this is almost no subduction at all. We still have the plate underneath the overriding plate. But there’s very little of it going down. It’s actually moving side to side. And the current thought is that, if you didn’t have the momentum coming from the south, it wouldn’t have generated an earthquake on its own. So, this is a very oblique subduction zone, and they were, I think, overlooked. I think most people didn’t realize these were strongly coupled. And there’s two other regions in the world that we now are looking at. There’s the far western Aleutians, a very oblique subduction zone where we have the Pacific Plate going underneath the North American Plate, and the Puerto Rico subduction zone, which is of extreme concern because of the population density. Again, a very oblique subduction zone. So certainly, after this earthquake, we’re seeing, oh, magnitude 9 earthquakes can occur here. We’re looking at the regions – similar regions around the world.

- I’d like to comment on that also. Just from a practical standpoint. In Banda Aceh, the tsunami was there 15 to 20 minutes after the earthquake. You would have to have very, very good infrastructure to be able to get a warning out in that time and to have evacuation plans in place to get the people away from the area of danger. There’s a refreshing story from the island of Simeulue where there were large tsunami waves there in northern Simeulue. I visited a village where the waves were on the order of 10 meters high, and nobody died. And the reason why is, in 1907, there was an earthquake and tsunami – large tsunami – that killed many people in Simeulue. And so they had a recent enough memory and a knowledge of tsunamis – they call them smongs there – where they – as soon as the shaking stopped, they all went up to the hills. And so, there were no deaths. Many houses destroyed, but no deaths.

- A story related by Gegar? Did he tell you this one?

- I’m not sure.

- About – I think it’s that same island. I’m not sure.

- Oh yeah, yeah, yeah.

- With the uplift?

- Yeah.

- Okay, so, also using the recession of the sea as a sign that a tsunami is coming, okay? So, was it the March 28th event that raised?

- Yes.

- Okay, so, it raised the shoreline, and people up in the hills saw that the sea was still receded. So they declined to go back to their village, because they figured a tsunami was coming. It was a permanent deformation. [laughter] A permanent uplift. And so, this Indonesian scientist spent some days trying to explain that, uh-huh, okay, this is the way it’s going to be from now on. [laughter] The shoreline’s at a new place.

- They thought it was a very large tsunami coming. [laughter] Actually, they were in the hills for one to two weeks before Gegar got there. Yeah.

- Okay. Me? Several years ago, I saw a TV program about a wildlife preserve in the very northern tip of Sumatra. And the Sumatran tiger there and other rare animals were there. And I wondered how this wildlife preserve fared in the tsunami.

- I’m not sure of the area you’re talking about. There are some intriguing stories about how wildlife in places where the tsunami hit were not killed. I know in Sri Lanka, there are stories about elephants who got away.

- Hm.

- And so, I don’t know about that specific preserve and what happened there, but it’s possible that there is some instinct in animals that we don’t have, to get away from...

- In Sri Lanka, the elephants detected the presence of the approaching tsunami, somehow?

- Somehow, somehow. And there’s theories on that. Actually, it was a problem because they evacuated the elephants, evacuated to high places. And it’s dangerous to be around elephants. [laughter] So, people couldn’t evacuate to the same places. [laughter] But – I actually – one of the interesting things about going on these international field teams, and you have websites, and people see that and you get some interesting emails. And one email was about – you know, with different theories about sound that was created by flexure of the plate that the elephants somehow detected.

- Yeah. I know elephants can hear a lower frequency sound than we can hear.

- Yeah

- Yeah.

- First of all, I’d like to say, I think you’ve done some really good work, here. [laughter] My first question is, what sort of wave heights and inundation you’re expecting from a dinner sausage- type repetition, Cascadia subduction zone type, along the coast and then down into California? And the second question is, what is the state of research on what could be the big gorilla in tsunami generation in the Pacific, is a flank or a sector collapse in the Hawaiian islands? The last modeling that I saw on that was a 30-meter wave along the West Coast.

- Do you want to ...

- Do you want to do the – it’s simulated, so ...

- Yeah, yeah, we just completed a study – well, it’s still in the works –  working with a group at NOAA simulating local tsunamis for, again, a wide variety of scenarios. But we’re getting runup heights and – runup heights, now vertical ones – very similar to Sumatra – 30 meters or so. We’re getting inundation distances, and this is where Bruce comes in, by looking at the tsunami deposits for 1700 and for 1964, even. That can help constrain these tsunamis. So, it’s, in many ways similar to Sumatra.

- It’s a difficult question to answer. There’s – in terms of – and correct me if I’m wrong, Eric. In terms of the modeling from the deformation from the earthquake, there’s a lot of unknowns, and so it’s difficult to do it forward. And, in terms of modeling from the tsunami deposits, we still have to learn more about the relation between the tsunami and the flow. I will say, in that study, before the Sumatra earthquake occurred, we found some deposits that had what we thought was a landslide. You know, it was hill material that was very different than what looked like was brought in from the beach in the regular tsunami deposits. And, after the Sumatra earthquake, we’re reevaluating, thinking maybe the wave got up into the hills and stripped it, like we saw in Sumatra.

- Now, for the Hawaii question, correct, the ...

- Yeah.

- This is what we call the volcanic sector collapse, or flank collapse.

- Sector collapse, yeah. -There’s actually two scenarios. One for Hawaii – the south flank of Hawaii, and one for a volcano in the Canary Islands. These are – they’re controversial in many aspects. The hydrodynamics, there’s two different sides to it. The likelihood of that type of sector collapse is pretty low, in comparison to the earthquake-generated tsunamis. But I think all that said, it could happen. Now, the important thing for a landslide-generated tsunami, or one of these volcanogenic tsunamis, is the speed at which it goes. And it can – the speed, we know, can range. If you go up 280, you can see these goofy slides that take forever to go down. Or they can go real fast. And if they’re going fast, they can generate tsunamis. But it’s a huge uncertainty. We don’t know what landslides are doing underwater, basically. We have these instrumental records – seismograms all over the world telling us these detailed pictures of what the earthquake’s doing, but we don’t have the equivalent data for submarine landslides.

- Well, those aren’t submarine landslides. They start above sea level.

- That’s true, and …

- And if I read the record correctly, most of them have been characterized, at least in the Hawaiian Islands, as avalanche-type collapses.

- That’s probably true, too, yeah. But again, I don’t think one has ever occurred in human history. They probably could occur at some point. But it’s one of these likelihood – one thing is, when we talk about hazards of all types, there’s four components to keep in mind. The severity – so this scenario is high severity. Location – if it’s just going to affect Antarctica, we’re not so concerned about it. Extent, is like magnitude 9 earthquake versus a magnitude 7.2 earthquake. And then, likelihood. So, we try to keep all four of those in mind. So, I think, if somebody said, you know, an asteroid tsunami. That’s maybe a high hazard but very low risk, because it’s on the 10,000- to 100,000-year timescale. So, a lot of things to keep in mind in terms of that.

- But that would be a terrible disaster if it does let go.

- Yeah, and it certainly could. Yeah.

- My question is related to the last one. It’s not about an earthquake-triggered tsunami. But a Scientific American article a couple of years ago described one of the biggest dangers that we could have for the East Coast, is a landslide from a volcano on one of the western islands of the Canary group. And can you comment on that? Because that one could be far worse than anything we could ever see on the East Coast – on the West Coast.

- It, well certainly, the problem with the East Coast is they’re not at all prepared for a tsunami, like, we – I mean, there’s ways we can go on the West Coast, but – again, this is a controversy that we’re talking about, whether or not – I can see why the scenarios in places with one of the steepest bathymetric regions in the world, an active volcanic region. There’s some controversy about the hydrodynamics. Are they including all of the necessary things you need to include. So, the wave heights 5 to 10 meters from one hydrodynamic theory, and other hydrodynamics say 1 to 2 meters. Even so, Bruce showed that a 1- to 2-meter high tsunami is worth worrying about. But I guess I always go back to the likelihood. Is this likely? Should we be worrying about Brian’s Cascadia event and the probabilities he’s talked about, or should we be worrying about the Canary – I think these questions really come down to, how do we deal with uncertainty? You know? And that’s the big problem. These are very uncertain phenomena in comparison to earthquakes. So, I don’t have a good answer.

- But, it’s partly the job of the scientists to reduce that uncertainty. It’s tough for engineers to design for something very uncertain. Because then, to be cautious about it, they’ll spend – they’ll have to spend a lot of money for something close to a worst case. So, if you can define the hazard more exactly, people can take action against it. In a way, the Japanese written records did that for us at Cascadia. They defined it as a worst case, but at least they gave us a definition. And one that tsunami modelers could use without being ridiculed. In the case of the Canaries, my gut reaction is that somebody went to graduate school in Delaware, and spent some time mucking around salt marshes in Delaware, punching holes in them, and looking at thousands of years of salt marsh history and then thinking about what we see out here in Cascadia. You know, that one way to approach this is to consult those natural records that go back on the East Coast probably 5,000, 7,000 years. You can – you can look back in salt  marsh history and see whether some great catastrophe swept onto the Atlantic seaboard during that time. And from that, come up with the probability side of that equation. I think some people in risk management talk about risk as the aggregate phenomenon. And they say that it’s the product of these two things – probability, consequences. Okay? And so yes, you know, Canary Islands disaster, consequence is way up there. But the probability is vanishingly small. And so if I were living, still – I’m from Connecticut – if I were living on the East Coast, I’d be worried much more about the kind of thing that happened with the Grand Banks in 1929, where an offshore earthquake set off a submarine slide. The slide turned into a turbidity current that famously broke a dozen transatlantic cables, one after another. But the slide also set off a tsunami that killed 29 people in Newfoundland. And there’s a lot of potentially unstable continental slope off the Atlantic coast. of the U.S. that could be destabilized a bit further by a landslide. And that doesn’t give you much warning between the time of the event and the arrival, right? And then also if I lived on the Atlantic coast and vacationed in the Caribbean, I’d think a little bit about the tsunami history there. With the numerous examples of fatal tsunamis. Large numbers – thousands of people in total – have been killed by historical tsunamis in the Caribbean subduction zone there.

- Thank you.

- Are tsunamis the same as storm surges?

- I’m going to deflect that. Ihaa conflict of interest because that’s my son. [laughter]

- Is this a plant? [laughter]

- No, storm surges don’t have the high velocity that a tsunami does. And they also, the storm surges have the waves that Eric was talking about. The waves that are very short-period, short-crested. So, if you asked me, would I rather be in a storm surge that’s a meter high or a tsunami that’s a meter high, I’d say, give me that storm surge.

- [laughs]

- Because the tsunami that’s a meter high will be moving fast.

- I’ve got a question. I’ve been doing some studying. You know, the Monterey Bay shape? What’s the consequence of a tsunami in Monterey Bay? Because you got to look at the shape of the bay, plus the canyon. It’s perpendicular, got deep water, shallow water. That wave hits.

- Well, we did have a tsunami in Monterey Bay from the Loma Prieta earthquake. It was a little landslide let go in Monterey Canyon. It was observed, and damaging, I think – if I remember correctly – at Moss Landing. The bathymetry – bays are tricky, because ideally, tsunamis are focused on headlands and points. So we know that pretty well. Bays, theoretically, they should be deflective. They should unfocus a tsunami wave. But you also have to take account the reverberation within the bay. So it really depends on the shape of the bay. Obviously, something –  well, the inundation maps, most have taken this into account because they did full-blown hydrodynamics of the bay. So that takes that into account. But it’s difficult to say with bays. You can pick out a place like Crescent City, the headland. That’s  going to focus the tsunami waves. Bays are a little ...

- Because – and I’m asking that, it’s because you have deep water in the ocean. You’ve got shallow water on the other side. The canyon is deep.  So the water goes this way, hits the canyon. You know, it goes upward. [inaudible] climbs upward.

- Well, the tsunamis, unless they’re coming straight – and they could, if they’re coming straight up the canyon – but if they’re coming at any kind of oblique angle, they’re actually going to be deflected. Kind of like that effect I saw – we saw for Bangladesh.

- Right.

- So any shallow angle, they’re going to be deflected, actually, along the edges of the canyon. If they’re going straight across the canyon, they’re not going to see it. They’re going straight up the canyon, they’re not going to see it. So there’s – the canyon itself is what we call caustic. And it might actually even amplify the waves along the edges of the canyon.

- And …

- Go ahead.

- To answer your question, I don’t know anything about this, I’m just thinking if you wanted to look into what 1964 Alaska did. What 1960 Chile did. It’s hard to think of a tsunami much bigger from Alaska than 1964. And from the diagram Eric showed, hard to think of one that’s better directed at this part of California. Okay? So, ask yourself what happened there, and then you could use that as something of a worst case for an Alaskan tsunami.

- Okay, another question.

- I’d like to comment on that, since I live in Santa Cruz. [laughter]

- USGS has an office-specific science center in Santa Cruz that’s primarily coastal and marine geology. And there’s about 50 of us there. 1946, there was somebody who died in Santa Cruz from a tsunami generated off Alaska. For those of you who know Santa Cruz, near the wharf, the Dream Inn,  the big hotel, there was somebody walking there and the tsunami reflected off the cliff there. And the person got taken out to sea. So, I think Brian’s answer is very appropriate. You look at the historical to see what there’s been there. But it’s possible that, at least for tsunamis coming from the northern part, that they bend around and it’s the northern part of the bay that gets hit. Just like Crescent City gets hit. And I saw the same effect in east Java when I went there,  that the tsunami actually refracted strongly around the headlands and turned around a bit. And the parts of the bay that were kind of closer to where the generation area was where they got hit.

- Okay, another question. Okay. Think of the weight of the water in this world. Does that have any effect on the deformation of the crust? Does it have to be – sort of weight pushing down on it?

- That’s a very good question. There was actually a presentation at the last Seismological Society of America meeting for the Indian Ocean tsunami. And there was a sense that they could detect the pressure of the water on these GPS stations in Sri Lanka and India. So, that’s an active – that’s an idea and that seems to be observed. I think it would have to take a very  large tsunami, but it does seem to be – I think it has to go through all the peer review process and stuff like that. But, I think that’s a very likely explanation, yeah.

- Okay, thank you.

- Hi. Thank you all for an excellent presentation tonight. It was really very informative. Two quick questions. First of all,  on the long runout landslides on the southern end of Hawaii, you were talking – NOAA said 30 meters, possibly, for a tsunami? And if so, what kind of a timeframe would we be looking at here on the West Coast for that? And secondly, for you, how do you perceive the recovery of the people of Sumatra and the neighboring areas affected, in terms of time for a complete recovery? Now, are we talking years for these people?

- I’ll do the Hawaii one. I don’t know – I can’t remember the exact numbers for West Coast tsunamis. Under this catastrophic – cataclysmic scenario, you would get very large tsunami runoffs, particularly in Hawaii. Also possibly on the U.S. West Coast. We would have the same amount of time as if an earthquake-generated tsunami occurred. Which they did. In 1975, we had observed tsunamis from that earthquake in Hawaii on the U.S. West Coast. And, again, I’d have to check, but it’s on the order of maybe five, seven,  eight hours. Something like that. Enough time for a warning. And hopefully soon there’ll be these DART stations in there that’ll help confirm or deny whether a tsunami has been generated by that kind of – so one of the regional warning centers is at Ewa Beach outside of Honolulu. One of their missions is Hawaii tsunamis, and then the rest of the Pacific. We’re going to get our warning – they’re all linked – but we’re going to get out warning from Palmer, Alaska, at the West Coast Alaska Tsunami Warning Center. But they  have seismometers all around Hawaii in case this would occur. So the warning centers hopefully – you know, this is a pretty cataclysmic scenario, I guess is the best way to describe it. So, you know.

- Was the concern that a tsunami generated in Hawaii would hit us? Or, concern for local …

- Yes, how it would affect the West Coast and in what timeframe.

- Steve Ward did a simulation for a volcanic flank failure off of Kilauea. I think that’s what this is coming from.

- Okay. And then, in answering your question on the time scale for recovery, I think it’s going to be quite long. I was there in April, in Sumatra, and in the large cities – well, certainly Banda Aceh, there were portions that weren’t affected by the tsunami, and it looks like business as usual. And, even in the tsunami zone there, there’s starting to be some rebuilding. And there’s another city maybe 100 miles south, Calang, where there was starting to be rebuilding. But everywhere else, the smaller villages, there just weren’t very many people around. Some of these places, it was, you know, greater than 95% of everyone who lived there was killed by the tsunami. It was frankly very difficult to go to these areas, because of the tragic event that had occurred there.

- Not too long ago, when the tsunami warning flashed across our TV screens, here, and I found out about the one in Northern California, I flashed back to what I thought, at the time, was a fairly large earthquake down in Chile. Off the coast of Chile. And I know you guys haven’t talked about it, so it probably wasn’t significant in terms of tsunami. But, am I just blowing steam here, or was there a fairly large quake ...

- Just recently?

- Just not too long ago.

- Yes, there was. I’m trying to remember. Because it was right near that June 14th event in northern California. I don’t remember a warning going out from the Hawaii Warning Center going for that. It must have been under the 6-1/2 to 7 threshold for that. I remember it was a big earthquake, though. And, incidentally, Chile has their own local tsunami warning system, just to throw that in.

- Yeah, I have a couple questions. One is, since you’ve done this correlation relative to earthquakes in the subduction zones and so forth, in the areas where you have that occurring and there’s volcanic, is there any correlation in the timing of these as it relates to volcanic activity? Or vice versa?

- I’ll give you just one particular for Cascadia. You know it already. When did Mount St. Helens last have a big eruption, okay? And you know, May 18th, 1980. And then the very big tsunami and earthquake in January of 1700. Before the January 1700 earthquake, St. Helens had, in the late 15th century, a pair of eruptions. One of which was about – has been estimated to have produced as much ash as did the 1980 eruption. The other about five times that volume. The same tree ring scientist who did the Red Cedar work, David Yamaguchi, dated both of those events. One of them to the off-season for the trees, between 1479 and 1480. And the other, the same thing, 1482-83. And subsequently, glass – volcanic glass with the St. Helens fingerprint – chemical fingerprint – was found in the autumn 1479 ice layer in Greenland. In the Greenland ice core. But in any case, I’m giving you those little details just to say, okay, these events are pretty well located in time. Or we could go out to the banks of the Columbia River and see the geology of the 1700 earthquake written in the banks. And then we could go find that 1479 ash that got washed down the river and deposited. And you can see that they’re separated in time. So, at least for St. Helens, there’s no obvious correlation. Now, there are good examples of volcanos going off at or about the time of very big subduction earthquakes. But there are many counter examples, as well.

- Okay. One last question. Can you comment about the risk to the Bay Area communities relative to any Cascadia subduction event, or a strike-slip event here more locally, relative to tsunamis?

- I can go through a tour of the North Pacific. The strike-slip ones – there’s very odd event in 1898 between the Hayward Rodgers Creek we’re still trying to sort out. There’s some reports of a 3-meter tsunami inside the bay.

- I’m not sure that’s the question.

- That’s not the – well, the strike-slip events.

- Is that your question, or ...

- Either one.

- What Cascadia would do to …

- That’s one of them.

- I was answering the second one first, I guess.

- I’m just interested [inaudible]. [laughter]

- Well, this is from the West Coast and Alaska Tsunami Warning Center, so it’s got some … This is really designed to predict what the tsunami amplitudes are on these DART buoys right here. But you can see – like Brian really showed in a lot of his slides, that the tsunami energy is heading off towards the Pacific, eventually towards Japan. The question of how big these tsunami amplitudes are going to be down here in San Francisco Bay is an obvious question that we need to address. It’s not a simple question to answer. Because instead of these direct waves coming straight off a earthquake, we get what’s called these coastal trapped waves – these edge waves. The Cape Mendecino earthquake –  I forget the date of it. They were quite visible.

- ’92?

- ’92, ’92. They were quite visible at the Crescent City tide gauge. So we know these edge waves are excited by subduction zone events. So I think that’s going to be a key piece to understand. To do that, you really need good bathymetry all the way from Cascadia all the way down to San Francisco, and I’m not sure we have that, to really do an accurate job of it. But, I’m sure we would see some effect. I think we’re looking at about one to two hours, depending on where the earthquake rupture starts, if it’s a breakfast link or a dinner sausage. [laughs]

- But, keep in mind also, 1964 Alaska. Look back in the papers and check on this, but a tsunami historian, Jim Lander from NOAA, has compiled a book of historical tsunamis along the U.S. Pacific coast. And he mentions in there that, in 1964 dollars, the losses in San Francisco Bay from the the 1964 Alaska tsunami are about 1 million. Okay? And it was from strong currents. And it’s well known also that Los Angeles Long Beach Harbor suffered an equivalent loss in 1960 dollars from the 1960 Chile tsunami. And again, strong currents.

- That’s a good point, yeah.

- Okay. Thank you very much.

- I work at the Office of Emergency Services in San Mateo County. And I work on tsunami planning. I get a lot of questions about what the bay effect is going to be. And the only answer I have is a preparedness answer – to watch for flooding and be alert for a warning. But, primarily our target audience for warning is the coastline. Do you have a simple answer for the effect in the bay from a 42-foot wave on the coast side? Which is my planning number. [laughter]

- Yeah I was wondering how you came up with 42.

- It’s my worst case scenario. [laughter]

- We did a simulation of tsunamis in San Francisco Bay, from really looking at the 1906 and this kind of odd 1898 earthquake. What seems to happen – there’s a really good connection through the Golden Gate. Of the tsunami energy that gets through – so there’s a limited amount that gets through – there’s a real strong connection. The Bay – you probably know this already, but it’s very deep in some sections. Most of the Bay is very shallow. So, you see – if you compare tide gauges, like Brian was mentioning, the 1964 and 1960, if you compare from Fort Point to Alameda, you could place one on top of each other. There’s very efficient transfer of energy from Fort Point all the way to Alameda. But then you get into these shipping channels. And once you get into the really shallow water, it really starts to lose energy. I think maybe from bottom friction more than anything else. Plus it’s spreading out, so that little bit of energy that’s going through the Golden Gate has to all the sudden spread out into the vast reaches of the Bay. So generally – you know, and it doesn’t look like there’s very much reverberation within in the Bay. There does seem to be reverberation in the Gulf of the Farallones, interestingly enough. But not within the Bay, it seems like.

- Okay. Second part to that, the seiche effect. I don’t find much study on the seiche. Are you experienced with that?

- We have a senior scientist who was here at noon. It was Ralph Chang. And I was in with a nice – a very active seminar. And one of our hydrodynamic models we built was a seiche model. And-

- What is a seiche?

- A seiche, thank you [laughs] is – it’s a sloshing of water. And it can be triggered by a number of phenomena. They’re somewhat common in the Great Lakes from atmospheric disturbances. For earthquakes, we got seiches from the 1964 all the way down in the Gulf of Mexico. That’s from the surface waves – the seismic surface waves – traveling around the Earth. They can also be triggered by the tsunami itself. So the tsunami can all of a sudden send the – get this sloshing going. It doesn’t look like the bay itself is the right geometry for seiching. There was a study of Lake Tahoe that does seem to be – look like it’s the right geometry for seiching. And they’ve got faults bounding it. Briefly, it’s a very irregular shape, unlike Lake Tahoe, for example.

- Great. Thank you.

- Is there anyone that I need bring the microphone to that we missed so far?

- What’s the DART stand for?

- Deep-ocean Assessment and Reporting of Tsunamis.

- That’s good.

- NOAA loves acronyms. [laughs]

- They’ve changed the name now. They’re calling it tsunamimeters?

- Tsunamimeter, or something like that.

- Tsunamimeters. Because no one knows what a DART is. [laughter]

- Okay, well, I’d like to thank everyone. You’re welcome to come up and ask individual questions.


And please join us again next month for California Rock and Roll.