 Bingo! We're back to Think Tech. It's the one o'clock rock. I'm Jay Fidel. Welcome to Research in Manoa. Our show today is called Earthquake Early Warning and Tsunami Prediction. Our guest for the show is James Foster of HIGP, the Hawaii Institute of Geophysics and Planetology. Welcome to the show, James. Thank you, James. Pleasure to be here. So we're going to take a look at the low-cost networks for Earthquakes and early warning and tsunami prediction. The mass production of smartphones can be integrated with GPS and accelerometers in revolutionizing fields that can take advantage of these low-cost sensors. So James is collaborating with USGS to build an early Earthquake Warning network in Chile. Let's hear more about that. This is a revolutionary ability that has developed, as you said, through the development of cell phones and the fact that these are being built now in millions and millions of millions. Everybody in the world has them. Everybody wants increasing little gadgets that can be integrated into the cell phones and that can allow them to do new and useful things. It turns out that some of these gadgets that are being now built into cell phones can be used to do science and hazard mitigation. Two of the key ones are the accelerometers. Everybody knows that you turn your cell phone and the screen turns with you. So the cell phone has in it an accelerometer to be able to recognize what you're doing with the cell phone at any given time. Those accelerometers are actually sensitive enough to detect Earthquakes. Just because my phone feels an acceleration and movement is that enough? Or are you talking about collecting signals from lots of cell phones? Well, initially it would be our concept is using a fairly dense network, but it's not everybody's network. Not everybody's cell phone. It could be. And there's movements afoot to see if we can develop an app that anybody who wants to can download, install on their phone and become an active part of a mobile, evolving network for Earthquake 31. Brilliant. So if I have a city, for example, and I have say a million people living in the city, then I can enlist say 100,000 of them to be part of my network and they will opt in and their cell phones will have an app that feels the vibration and sends vibration to a central place and the central place, maybe with a geophysicist present, will say, uh oh, we're having an Earthquake. So you're not talking about sensors in the ocean, you're not talking about satellite sensors or all that new kind of thing, new fangirl technology that's all over the Pacific now or that hopefully is being installed all over the you're talking about the cell phones right there in the city which is being affected by the Earthquake. That's the concept. We'll go before that can be put into an operational setting and the first step is the network that we're developing right now in Chile where we're actually using a commercial smartphone, we're taking advantage of its onboard accelerometer and we're taking advantage of GPS capability. Right now we actually have attached an external GPS unit but there's a new generation of cell phones and firmware, the Android operating system that would give us the capability to run all of that internally on the cell phone. And what's particularly cute about this whole concept is that normally when we do a field deployment, the biggest problem is communications. How do we get our field data back out of the field? Cell phone already has it. And not only that but the cell phone has computing capability onboard so we can even do onboard computing, we can filter the data, we can do some preliminary processing to improve the robustness of the signal, to make sure that we've got nice, clean, compact data that we send back to our central server and the facility. So once you've downloaded the app, this happens in the background. You don't necessarily see it happening. So yes, for this crowd sourced concept that would happen in the background right now, the ones we're installing, they're in a box. They're put onto buildings in Chile. We've been working with some of the military installations because they've got security. You mean it's not deployed to an individual? It's not deployed to an individual. Exactly. So we've actually fixed these on roofs, on posts anywhere that's relatively fixed that's got a good sky view so that the GPS is operating well and so that the cell phone has good communications. But what about me? I'm walking around the street, having my pocket. I use it all day. I make telephone calls. This app would help you through me also, wouldn't it? Absolutely. So as I said, we're not there yet because the problems of dealing with 100,000 people who are doing things that are going to cause shaking. They're getting on and off of buses. They're going up and down elevators. They're doing all sorts of things. So you'd need some pretty sophisticated cloud-based computing behind the crowd-sourced version of this to recognize what's just a person doing day-to-day things and what's 100,000 people all suddenly starting to shake in the same sort of way because an earthquake is... So what I hear you saying is that there would be kind of a filter on this phone and it would say, oh, this is the sound of a bus or he's driving in a car. It's not the same thing. We're looking for a footprint of an earthquake and it's different and we can filter out all the other sounds and vibrations and just send you the geophysicist in the central location will send you... In fact, the central location could be anywhere, couldn't it? It could be here. There could be many of them. So yes, ThinkTech could be a central location. But no, that's exactly right. There has to be some serious computing that recognizes that not just one phone is showing kind of characteristic shaking but all of the phones in that vicinity are showing that same shaking. So that can't operate within the app. That has to be controlled through some central facility that recognizes that all of these phones are sending data in at the same time that show the same. And the GPS thing is going to give you essentially a map. You have a map, you know where it's coming from, so you can see it you can see all the lights lighting up on the map. Now, would you be able to triangulate and determine the epicenter of the earthquake? Exactly. So that's the whole point of partly the earthquake early warning but then taking the earthquake early warning and mapping that over to tsunami prediction. So the earthquake early warning is really it's just about detecting shaking and saying, hey, there's some serious shaking, damaging shaking coming your way. Here's a few seconds of warning ahead of that and reaching you that you've got time then to do something. You can crawl under the table. You know, you can if you're a surgeon in a hospital operating room, you know, you can withdraw your instrument from close up from the patient that's lying on the table in front of you. If you're a power station, you can spin down your generator. All of these things that, you know, a few seconds of warning before that damaging shaking hits could be critical to maintaining critical infrastructure or lives in the event of heavy shaking. That's the other part of this. It's a two-way street. One is you send your vibration information data to whoever is analyzing it. But the other part is somebody at the other end say, hmm, we have a problem here. You know, Houston? We have a problem. So we're going to advise everybody in the network, okay, maybe every in fact, the people who opt in to send new data is only a subset. When you go back to say, whoops, you better get under the table. You want to go wider than just your special agents. You want to tell everybody, right? Oh, absolutely. You need to be plugged right in, you know, for the general public, you're going to have to go through some warning system like Hawaii's. How do you do that? That's not necessarily the same app. Because that would go to people who have not opted in, not necessarily opted in. No, that's going to go to a broad public warning system. Those are operated by every county, has its own emergency response system. But then very specialized things like the hospitals, like the power companies, they may have a very specific warning that they need, that plugs straight into some emergency response software that starts things happening that you don't want it to go to a designated person who then has to run to hit that red button. There's no time for that. You have to have stuff that starts to happen automatically. So you have to be robust. You don't want to be shutting down the city's power grid on a false alarm. It's a voice message though, it comes back. It could be almost anything. It could be something that automatically gets interpreted by some governing software that says, hey, we've got to shut things down right now. There's no time for somebody to be sitting in the middle of this and hearing the message and responding because the warning could be really one, two seconds. There's no time for some of these for human intermediary for the messaging system. So the messaging system is not necessarily part of what you're working on? No. I'm a geophysicist. I'm a researcher. My problem is not how best to get these messages into the public into those stakeholders that need to be able to respond. That's the role of people whose jobs are in emergency management. But you have to communicate to say the city warning system. Is it just digital? Is there a problem with the reasoner problem? Or you talk about the intensity of the problem or talk about what kind of advice should be given by that city warning system? It's a very subtle problem actually about what the warning should be so that people respond in the right way to that warning. You have to be very, very careful about the language you use so that people really understand in that second that they have to ingest the data exactly what their response should be. So it's a many-fold problem. We as a geophysicist as a researcher within the institute, we're working to be able to provide the emergency management people with the best information we have and we'll work with them so that they understand what it is we're giving them and they understand then what they want to broadcast out to the public. But yeah, you have to be very, very careful. You can't just say, you know, there's damage and shaking coming. You're going to panic in all this. And so there has to be an educational program that educates the public so that they know in advance, you know, they don't just get this warning and it's like, what am I going to do? They have to know through school programs, through radio programs, you know, through any sort of media that gets, reaches out to the public that this sort of thing is happening. So the mainland has these shake alert programs that they involve schools and they involve the public and they go through these dry runs of how do you respond if the sirens go off? And what do you say? And what exactly? Do you tell them to get under the desk? Do you tell them to take their instruments? What do you tell them? Well, and you tell them in voice, no, because you have the ability to do that on the cell phone. You do but you don't have the time to give a complicated message. So you really just have to give them a very short cue that, you know, this is time for you to respond in the way that you have been trained to respond. So, you know, there's no time for complicated instructions. Pre-training is what... Pre-training is absolutely crucial. So something you mentioned in both of our of the show a moment ago that concerns me is that you don't have a lot of time here. We're talking about real time. We're talking about, you know, vibrations happen in a number of phones. The process to analyze those vibrations is almost immediate. And the message goes out, whoops, you got an earthquake here. And you said it's only a matter of seconds. How much time is it really? I mean, how much time is it really? It's interesting. Our understanding of exactly how much lead time you might have is sort of evolving as, you know, the science evolves. A lot of the work on the mainland is based on the assumption that the damaging shaking comes with surface waves from the seismic, from the earthquake. And they propagate relatively slowly. And so some of the design of the system is based on the fact that you might have 20 seconds, 30 seconds. Wow. You can't do much in that period. You can't do very much. It's looking like, you know, depending on how you design your system, you know, the practical systems we can design right now, you may not even have more than one or two seconds. Can I save myself in one or two seconds? Absolutely. There are crucial things that you can do as an individual to reduce the risk of you being hurt or killed in damaging shaking. What are they? Getting under the table. You're looking to be in a place where, if things are starting to fall, you're protected from at least the worst of that. So, you know, you're getting into a structural safe. You know, if you've got time, you can get out of the building into open, clear space. Get out into the street. Otherwise, if you don't, you know, just get into the table. You know, get into something that's going to provide you some protection from things falling off of the roof. Because those are the things that really... Well, speaking of time, James, we're going to take a short break now. Aloha. I'm Bill Sharp, your host of Asia Review. Watch us every week, every Monday afternoon for exciting, up-to-date information and analysis about contemporary affairs in Asia. Aloha, Kako. I'm Marcia Joyner, and I'm inviting you to navigate the journey. We are discussing the end of life options, and we would really love to have you every Wednesday morning at 11 a.m. right here. Hi, I'm Tim Appichella. I'm the host for Moving Hawaii Forward, and the show is dedicated to transportation and traffic issues in Oahu. We are all frustrated by seeing in our cars in bumper-to-bumper traffic, and this show is dedicated to talking with folks that not only we can define the problem, but we hopefully can come to the table with some solutions. So I invite you to join me every Tuesday at 12 noon, and let's move Hawaii forward. You know, nothing untoward has happened in this last minute either, but it couldn't. And we would have had, you know, the luxury of a minute to save ourselves, you know, given the time frame you've been talking about. That's James Foster of HIGP. He's been working on a system for Chile, you know, which will identify vibrations that suggest an earthquake and then allow municipal authorities to warn people to get safe. So interesting. But I'm going to talk about, you know, you're a geophysicist. You know, we don't usually see a geophysicist getting involved in this kind of electronic system that would give warnings to people and save a community. How did you get involved in this? It's been a slow evolution. I, as you say, I did a degree in geophysics, and most of the opportunities for employment are within the oil or mining fields for geophysicists generally. I was interested in volcanoes. I came to Hawaii to work with the volcanoes. But, you know, initially I was interested more from a scientific intellectual point of view. You know, how do volcanoes work and you know that the application is so that we can understand them better and so that we can warn people in advance of things that we understand now might be about to happen. But I did a PhD here in Hawaii. I've stayed in Hawaii. My interests have broadened beyond volcanoes. I look at things going on in the atmosphere, tsunamis on the ocean. And, you know, you're looking for a role. Okay, what is it? I do geophysics. And, you know, you want it to be more than just a dry intellectual exercise, which is very fulfilling. But, you know, it's sort of isolating from the broader community that we work within. And so, you know, as my career has evolved, and I've realized the various ways in which the things that I do can be put to very direct, concrete purpose to, you know, provide additional information, help them to particularly, you know, working with hazard mitigation for communities. And Hawaii's exposed to a lot of hazards. We've got the volcanoes. We've got earthquakes. We've got hurricanes. We've got tsunamis, you know. We're sort of a bullseye in the middle of it. My procosm of anything you'd ever want to know about geophysics. So, it's a fascinating place to work intellectually. But it's also a place where you realize that the work I'm doing has concrete direct application to helping, you know, our society. And so, I've evolved. Is it true that, as a geophysicist, when you walk down the street, you have a different perception of the world around you. In other words, as your feet touch the pavement, they're looking for vibrations, you feel that you're on top of an Earth which is always changing and moving? No, it's true. You know, as a geophysicist, as a geologist, you know, we look at the Earth, you know, from this very long-term, large-scale process, sort of objective, as well as from earthquakes, which is very short time scales, but, you know, driven by these big, sort of global scale processes. And I work with sea level change. I work with the tsunamis. And so, yeah, you look at the ocean, you look at the way it changes, and you think about what's happening if the land here might be subsiding a little bit. You know, that's making sea level worse. And you look around for clues about that might be happening. Yeah, you, it's... And you're thinking, you know, what's going to happen in the next few years and a year of bites, except what we're talking about today is like 10-second bites. Yeah, so we span this huge range of time scales and this large range of spatial scales, as well, of, you know, the various different processes that are involved in the things that we study. So, okay, so it occurs to you, as a geophysicist, who, you know, appreciates, you know, earthquakes, that and I'm really talking about the connection now, the outreach from HIGP and science and your work there, through the community, not only in here, of course that's interesting, and I do want to cover that. But in Chile, how does that happen? Did you wake up one morning and say, hmm, Chile, that's what I want to do? You know, things evolve in science, particularly in a research angle, you know, often very organically. You know, we don't start with a concrete plan that we need to address this very specific problem in this very specific area. So, as a research group, I work within the GPS group, the Pacific GPS facility within HIGP and that group and the various different leaders of the group whose mantle I know hold, have a long history since the early 90s of working in Chile. So we've got a long, you know, we've got connections with the researchers in Chile, the institutions in Chile that deal with this problem within Chile. So, you know, as we develop new technologies and we realize, wow, this is a really exciting technology that has real application in this very specific field, we look around, it's like, okay, so where could we deploy this? It's like, well, we have these connections in Chile where they've got a very serious problem, you know, many of the tsunamis that have impacted Hawaii have come from Chile and they have very, very large earthquakes very regularly and so it's a real, real problem that they're struggling to find the best solutions to. And so, Chile was partly because they have a very concrete problem that they haven't yet managed to fully address and we have the connection. Concrete problem? That's kind of a double entendre. It is. They actually, you know, they've addressed the construction part of the problem and that's one of the biggest, perhaps the biggest factors that, you know, impacts the impacts of earthquakes in regions is making sure that you build things properly so that they can sustain these things and Chile's done a great job of that and so part of what they're looking for now is to get better local tsunami warnings. They want to know whether that big earthquake they had has generated a, you know, potentially dangerous tsunami and they need to know that very quickly because it's a near field problem. And then, of course, the shaking, you know, can they improve the ability of their population to get into safe places quickly enough to save lives? They're a perfect laboratory for this because they have so many earthquakes and tsunamis. Exactly. You know, they're right on this subduction zone that runs down pretty much their entire coastline and they've got a very, very long coastline. And so they have big magnitude eight and nine, you know, eights and occasionally magnitude nine earthquakes. So it's a real problem for that country so it was organic in the sense that somebody knew somebody in Chile in the scientific establishment in Chile and said, hmm, why don't we try this in Chile because they need it and we here in Hawaii understand it and we can bring the elements together to actually deliver a system to Chile. Well, yeah, I mean, and I should make clear that actually the USGS has led this particular portion of this effort. They acquired the grant that provided the money to support this. And so the leader of that effort, Dr. Brent Brooks, was actually the director of the GPS facility here in HIGP before I took over. So this, you know, this again, it's just this organic element, you know, he the seeds of this particular idea were sewn here in Hawaii between a conversation that he had with me and with Gerard Freier who's until recently at the Pacific Tsunami Warning Center and Gerard was asking about the capability of low-cost Tsunami, low-cost GPS chips to help with Tsunami Warning, and we weren't sure that we could conceive of that particular element, but we thought, well, you know actually for an earthquake problem, you know, there's a potential for a GPS. It's not only low-cost, it might be better. It might actually be better. And this is, this gets to one of the the elements that is sort of a bit of a sideways step for people in the research community, which is that quantity is not necessarily better than quantity. You know, we like to think in research, you know, we just want to get the most accurate measurements, we possibly can, and that's going to tell us everything or at least the most we can get out of our observations. But for a very applied problem like earthquake early warning, what's better is to have a lot of sensors covering the entire area that you're concerned about, and you can compromise a little bit on the actual quality of those measurements, and this is where, you know, this is the realization that Ben had and that led to this whole concept for this network in Chile is that, wow, you know, we don't have to spend $20,000 on putting in these high-quality research sensors, which mean we can only put in a few. We can spend $350 on a cell phone and a little solar panel to power it, and we can put out hundreds. And that's actually going to be a better solution to this very specific problem. It won't tell us some of the answers to research questions, but it's going to practically save lives and all that. So it's an electronic play, though. Of course, HIGP is well skilled in building the database and making the geographical analysis over a map. But to get that signal, to get that data, you have to have the map and you have to have a system to accumulate the information through a server, I guess, somewhere to pull all that in. Is HIGP able to? Is it involved in doing the electronics on this? The fundamental electronics no, but that software element we're actually working to have in-house capability. Right now, that warning portion of the system is handled elsewhere, but that's the capability so as you say, we have the capability to do that. We've not yet installed and got it running, but there's a robustness in having several places doing that simultaneously. So we're going to be doing that and we need to have that because the concept is that this isn't just restricted to Chile, there are other countries that could benefit from this sort of network and we want to be able to reach out to those and offer a full comprehensive solution to the problem that they struggle with. A footnote to what you said, it's worth mentioning I think that my experience, my observation, scientists sometimes find that they need to do something in another discipline in order to get their job done. And of course, they can reach out and collaborate with other elements in their network, but they can also, and I've seen it happen many times at UH in Southwest in HIGP where they can learn that stuff and do it on their own and good for them. That's the kind of grit that you need to do science and get the job done, yeah? No, you're absolutely right. As a researcher, you're constantly reinventing yourself. You're finding that the skills, the tools you have to hand are actually great for this other problem that you've never even thought about and you're right, you can address that by just bringing in a specialist in that field or you bootstrap your own expertise and you just go ahead and do it. You still need the help of other people. You rarely do things on your own. You need lots of different skill sets that you can't possibly hope to develop all by yourself, but yeah. Must be a lot of gratification in that, James. The invention of gratification. That's absolutely right. So one more minute before we have to go, but my last question to you was what's going to happen now? What's the timeline like to get this in place in Chile? Within the networks running in a, you know, we haven't got the four where the plan is to put 200 stations out. Right now I think we have almost 60 in the ground and running, and so we're doing detections of earthquakes. We've detected four earthquakes to date with the network. None of them have been damaging, but they've been relatively small, and so, you know, they're demonstrating that we can actually detect earthquakes that are smaller than the ones we were concerned about, which is a great demonstration of the network's capability. So we'll hope to have the full network rolled out by the end of this year. We'll be building our in-house capability to bring all the data in and do the analysis, and our in-house goal is to be able to translate not just the earthquake early warning, but to connect that to tsunami predictions. So we'll be looking to see whether the earthquake has actually moved the seafloor enough that there needs to be a warning of a tsunami threat to the local communities. It's a new world, isn't it? It changes everything. Actually, I believe it will. It opens up all sorts of avenues for countries that couldn't afford these really expensive networks to be able to implement these life-saving infrastructure, saving networks, and I think it's going to transform things. Good for you, James. Really, you're saving lives, doing the right thing, taking science to humanity. That's great. Thanks for coming down. It's been a pleasure. Thanks for having me on the show.