Pod Insights Big Ideas Lab NARAC About this episode In the 1960s and 70s, researchers at Lawrence Livermore National Laboratory were pushing the boundaries of nuclear weapons technology. But these breakthroughs came with a daunting question: Now that nuclear technologies existed, how could the world manage their risks and keep people safe?
Transcript 0:03 Speaker 00 It was April 1986, and scientists in Scandinavia were perplexed. The readings didn't make sense.
0:13 Speaker 00 Routine radiation screenings at nuclear facilities showed spikes in radiation levels. Spikes that seemed to be triggered by the scientists themselves.
0:23 Speaker 01 They were coming to work and bringing very low levels of radioactivity, and that set off these detectors. So that was the first indication that there was radioactivity in the environment. And initially, no one knew where it was coming from. Then additional measurements came in.
0:38 Speaker 00 The researchers in Scandinavia reached out to other scientists across the continent, and the mystery deepened when they realized it wasn't just one facility that was registering spikes in radiation. Research labs across Europe were noticing the strange phenomenon, and they wanted answers. The United States quickly joined the investigation. Soon, the Department of Energy asked Lawrence Livermore National Laboratory for help.
1:05 Speaker 01 There was so much radioactivity released. It was initially measured in Europe, and our researchers were able to take new models that hadn't been applied to these kinds of problems before, that could simulate global transport, winds and transport of the radioactivity across the globe.
1:23 Speaker 00 In the days that followed, the lab's team at ARAK, the Atmospheric Release Advisory Capability, was able to trace the radiation back to a single origin point. It was coming from the Soviet Union.
1:36 Speaker 01 So using our weather models and our atmospheric transport models, we were able to estimate, based on the radiation levels measured in Scandinavia and parts of Europe, how much material would have been released. And it was substantial. It's still the worst nuclear accident in history.
1:52 Speaker 00 On April 28, 1986, an announcement was made on Soviet television.
1:58 Speaker 02 It's now clear that the Soviet Union has suffered one of the worst disasters in the history of nuclear power.
2:04 Speaker 00 Two days prior, on April 26, during a standard safety test at a nuclear power plant near Pripyat.
2:13 Speaker 00 A series of steam explosions and a subsequent graphite fire caused the reactor vessel to rupture and the reactor building to be heavily damaged. This resulted in the release of large quantities of radioactive particles into the atmosphere. The explosions were powerful enough to blow off the heavy steel and concrete lid of the reactor, which weighed 2,000 tons. Immense amounts of radiation shot into the air, surging upward in a plume that stretched across Europe.
2:41 Speaker 01 That's a really cool thing about atmospheric modeling, is you can use it to infer what happened, based on detailed simulations of the flow and transport, where the material would have come from.
2:55 Speaker 00 It wasn't until radiation was detected in Scandinavia that the Soviet Union began to admit that the unthinkable had occurred. Two days lapsed between the meltdown of Chernobyl's Number Four reactor and the admission from the Soviet Union. No one knows how long the mystery of the traces of radiation would have lingered, a silent danger, had the scientific community not come together to find answers. Only then could they start trying to assess the danger from the radioactivity released from Chernobyl. Because, as it turns out, the disaster was not unthinkable. Nuclear accidents have happened before, and they would happen again. But Lawrence Livermore National Laboratory is trying to make the world safer when they
3:40 Speaker 00 do. Welcome to the Big Ideas Lab, your weekly exploration inside Lawrence Livermore National Laboratory. Hear untold stories, meet boundary-pushing pioneers, and get unparalleled access inside the gates. From national security challenges to computing revolutions, discover the innovations that are shaping tomorrow, today.
4:17 Speaker 00 During the 1960s and 1970s, researchers at Lawrence Livermore National Laboratory revolutionized nuclear weapons technology and high-performance computing. But as these frontiers expanded, unprecedented questions and challenges arose along with them. There were a few visionary scientists who wanted to help address the potential consequences of nuclear technologies.
4:40 Speaker 01 Joe Knox, Marvin Dickerson, did something very bold.
4:45 Speaker 00 That's John Nastrom, the chief scientist for the National Atmospheric Release Advisory Center at Lawrence Livermore National Laboratory.
4:53 Speaker 01 They tried to take research on atmospheric dispersion models, weather models, nuclear materials and how they spread in the atmosphere, and also predictions of how the radioactivity produces radiation dose and what levels might be harmful so they can issue warnings and advisories on what areas people need to shelter in place and things like that.
5:16 Speaker 00 Joe Knox and Marvin Dickerson predicted that this new world of nuclear technologies would place new demands on safety, monitoring, and disaster management. The same advances in computing that were foundational to Livermore's nuclear weapons research and design could also be applied to help protect the public in the case of an emergency.
5:36 Speaker 03 The hardest thing to convince people of is that in a radiological emergency, it's not like the movies where everybody's face is melting off or something like that.
5:46 Speaker 00 Lydia Thai is a health physicist at the lab.
5:50 Speaker 03 For most people, the main risk is really an increased risk of cancer. It's something called stochastic risk, where cancer is really random. And being exposed to more radiation means you have a slightly increased chance of getting cancer, but it still doesn't necessarily mean that you definitely will get some kind of cancer. Explaining that is very difficult.
6:12 Speaker 00 Knox and Dickerson set out to calculate risks for the public. They put together a team of scientists to design a system that can predict health risks due to atmospheric releases of hazardous materials. The system used supercomputers to combine atmospheric science and nuclear physics. They hope this system could one day provide foresight in the critical first moments of a possible disaster. But to do that, you need to predict the weather, which is no small feat. Imagine you're playing a massive game of chess, but instead of a 64-square board, you're on an enormous field with millions of squares. Each piece moves according to its own set of rules, and the game is happening in real
6:54 Speaker 00 time with countless pieces making moves simultaneously. The weather patterns are like these chess pieces, each influenced by a unique set of conditions such as temperature, humidity, and wind speed.
7:07 Speaker 01 We start with an initial state of the atmosphere that's determined from weather stations on the ground, weather balloons that rise in the atmosphere and record temperature, winds, humidity, satellite observations that are used. So all these weather observations are used to basically initialize the model, a three-dimensional field of temperature and winds. And then the computational models progress forward to forecast into the future. As they're making those computations, they're assimilating new weather observations, and those are then fed into the model to constantly update it, which radically improves the weather forecast.
7:52 Speaker 00 Predicting the movement of these pieces—or the weather—is challenging, because every small change, like a slight shift in the wind direction or a temperature fluctuation, can drastically alter the course of the game. In fact, weather predictions are so complicated and independent that the phrase "the butterfly effect" actually originates from studying the weather. In the early 1960s, American mathematician and meteorologist Edward Lorenz was working on weather prediction models and discovered that small changes in initial conditions could lead to vastly different outcomes. This concept led to Lorenz's unanswerable question: Does the flap of a butterfly's
8:34 Speaker 00 wings in Brazil set off a tornado in Texas? With that in mind, let's return to our chessboard and add another layer. Say someone spills a bucket of paint on the board, representing a hazardous plume. Your task is to predict not just where the paint will land initially, but how it will spread across the board as pieces continue to move and interact. It's a near impossible challenge, at least for a human. And that's why Livermore's early scientists turned to supercomputers for help.
9:06 Speaker 04 The origins of atmospheric science at Lawrence Livermore. It goes back to 1959 when the laboratory purchased IBM 701 supercomputers to replace the Univac computers that had been used before.
9:28 Speaker 00 Lee Glasgow is the program leader at Lawrence Livermore National Laboratory for their nuclear emergency support team.
9:34 Speaker 04 And Chuck Leith, who was a physicist during the Manhattan Project, suggested to Director Edward Teller that he had a problem that could put that machine through its paces. His suggestion was modeling large weather patterns with a global circulation model. And so that was the first global circulation model ever developed and run on the supercomputer here at what was then Lawrence Radiation Laboratory.
10:10 Speaker 00 Building on this groundbreaking work, others at Lawrence Livermore National Laboratory wanted to use atmospheric modeling for airborne hazard prediction, too.
10:19 Speaker 04 They came up with a concept that it would be great to have a hazard prediction capability based on the weather and transport, all in one system that we could easily use for a response and for informing teams.
10:37 Speaker 00 It would be years before this small research project would have its time to shine. Not when disaster struck on the other side of the globe, but much closer to home.
10:46 Speaker 05 A government official said that a breakdown in an atomic power plant in Pennsylvania today is probably the worst nuclear reactor accident to date.
10:55 Speaker 04 In March 28, 1979, the Three Mile Island accident started unfolding.
11:02 Speaker 05 The accident occurred here at the Three Mile Island Nuclear Power Plant a dozen miles south of Harrisburg.
11:08 Speaker 00 In the early morning hours of March 28, 1979, scientists at the nuclear facility known as Three Mile Island knew something was wrong. Sensors were showing dangerously high temperatures in reactor number two.
11:22 Speaker 06 At about 4 o'clock this morning, two water pumps that help cool reactor number two shut down.
11:27 Speaker 00 A malfunction in the secondary cooling circuit of the unit two reactor resulted in the failure of one of the main feedwater pumps. In the control room, the operators tried to solve the problem by turning off the emergency cooling system, which was a devastating mistake. Heat pooled in the reactor and the core overheated. Temperatures exceeded a staggering 4,000 degrees Fahrenheit, leading to a partial meltdown in reactor number two.
11:54 Speaker 06 Officials say some 50,000 to 60,000 gallons of radioactive water escaped into the reactor building, and that the radioactivity penetrated the plant's walls. Steam escaped into the atmosphere, and radiation was detected as far as a mile away.
12:10 Speaker 00 Three Mile Island was not a remote location. Situated in the Londonberry Township in Pennsylvania, the nuclear facility was nestled into the heart of a thriving community. Parents could see the towering nuclear reactors from their kitchen windows as they watched their children play in the front yard. They had no idea if, or how much danger they could be in.
12:32 Speaker 06 How much radioactivity has already escaped from the plant? How high is the radiation level inside the plant?
12:38 Speaker 00 The damage done to the reactor had already resulted in radiation leaking out into the atmosphere. It was at that moment that the research happening at Livermore became pivotal.
12:52 Speaker 04 The next day, the Secretary of Energy contacted the Director of Lawrence Livermore and asked, "Hey, I know you've been working on this thing, can you put it into action?"
13:05 Speaker 00 In early April 1979, the Atmospheric Release Advisory Capability, or ARAC, was officially stood up and quickly went to work. They teamed up with the scientists already at Three Mile Island, where their research project was hailed as an essential new tool.
13:22 Speaker 02 The people around Harrisburg, Pennsylvania were told they can sleep in their own homes tonight.
13:26 Speaker 07 After an extensive technical review of the troublesome situation at the crippled nuclear power plant, the governor of Pennsylvania said he'll issue no evacuation order at this time.
13:36 Speaker 00 ARAC was instrumental in providing model predictions needed to assess the situation, complementing the measurements already done by other teams. The technology that was once only theoretical became vital.
13:50 Speaker 01 D W asked if we could take this operational prototype and apply it. Our staff worked round the clock on shifts for two weeks straight.
14:00 Speaker 04 We did time-evolving, three-dimensional atmospheric modeling, forecasting. That was basically real-time modeling.
14:10 Speaker 00 This type of real-time weather and plume modeling had never been done before and was essential in determining the threat levels, as well as helping officials decide how best to respond to the crisis. ARAC proved itself during the Three Mile Island incident. Not long after, it became the National Atmospheric Release Advisory Center, or NARAC. This transition broadened its scope of responsibilities to provide national-level emergency response for atmospheric releases of hazardous materials.
14:41 Speaker 06 NARAC is a team of about 40 staff members. We have researchers, we have model developers, as well as software developers who are building the system, and computer IT people who are keeping our computers running, the assessors who are running the models and communicating with the local and government agencies.
15:06 Speaker 00 That's Katie Lundquist, who leads model development at NARAC. Today, NARAC isn't only responsible for responding to emergency situations when asked. The team is also responsible for improving, maintaining, and testing the computing, modeling, and reporting systems that they rely on so heavily.
15:25 Speaker 06 The weather models that we currently use, and this is going on in all sorts of scientific computing, they were architected to work on CPUs, so on central processing units in the high-performance computing systems. But right now graphical processing units are coming online and they're enabling what we call accelerated computing. So accelerated computing would use a mix of CPUs and GPUs. And models that run on this type of computer architecture can run maybe up to 100 times faster than our models could run on the CPUs alone. So right now I have an LDRD,
16:08 Speaker 06 so that's Laboratory Directed Research and Development Project, to develop an atmospheric model that will run on accelerated computer architectures. And it's really promising to revolutionize what we can do in our atmospheric models.
16:25 Speaker 00 As these advanced computational capabilities come into play, they promise significant enhancement in both speed and accuracy, transforming our approach to atmospheric modeling.
16:35 Speaker 06 We do this research. We develop our models. We develop improvements for our models. Over time, those pieces of research and the model developments that go along with them, they become more robust. And then they can go through an operationalization process where we transfer that technology into the system and begin to use it once we feel confident in its use under almost all atmospheric conditions. From the time that we start the prediction to when we make our products, the way that we want the automated portion of our system to run is that it can run within 15 minutes. So once an analyst begins working on the problem, that would be the initial prediction.
17:19 Speaker 06 From there, we would then make refinements to that prediction.
17:23 Speaker 00 This means that when an emergency hits, within just 15 minutes, NERAC's automated system has a preliminary model and report ready for emergency responders to start using. In a crisis, NERAC provides actionable information to responders and policymakers, ensuring that complex scientific data, like concentrations of radioactive material, translates into practical guidance that can save lives.
17:50 Speaker 06 But what NARAC is really focused on, and I think is a unique strength of ours, is turning that concentration into impacts on humans. So a decision maker may not understand what to do with that concentration because they don't have enough expertise to understand what the impacts are. And so, most of our products, they aren't necessarily showing the concentration on the ground. They focus more on turning that concentration into the impacts on people. These are the health impacts that this population would incur. That's NARAC's strength, is developing products that show impacts to humans and can communicate
18:32 Speaker 06 to decision makers the information that they need quickly, not just the atmospheric results.
18:38 Speaker 00 But what happens if an emergency impacts the NARAC facility itself? In order to be ready to respond no matter what, the team needs to ensure their data processing and communication systems can withstand any potential disaster.
18:52 Speaker 04 We drill or exercise on a weekly basis. We're always keeping our system up to date, checking the systems, making sure that they work. From an infrastructure perspective, we are in a building that ensures uninterruptible power to our computers. We have generators, we've got batteries. If the rest of the site goes down, our building will not. We carry pagers. When you're on call, you gotta have your pager going. We do have landlines, we do have fax lines. If we need to, we can get on the landline and talk to somebody or fax, just like what they did for Three Mile Island.
19:37 Speaker 00 In addition to this enduring backup infrastructure, NARAC is also equipped with the latest high-speed networks, web-based software tools, and massive global databases of weather, terrain, land use, and map data. This kind of vigilance has always been at the heart of what NARAC does, and it would be needed again in 2011, off the coast of Japan.
20:12 Speaker 00 Kyung-la, our reporter there, was saying that she thought it was four to five minutes. Was that your sense as well? On March 11, 2011...
20:20 Speaker 07 Well, it was absolutely unlike anything I've ever experienced.
20:23 Speaker 00 An earthquake hit off the coast of Japan.
20:28 Speaker 07 Quite simply, the biggest, longest-lasting earthquake I've ever experienced.
20:33 Speaker 00 The 9.0 magnitude quake created a massive tsunami.
20:39 Speaker 00 The wave reached heights of 45 feet as it barreled towards the coast.
20:45 Speaker 00 When it struck land, it quickly overtook the seawalls of the Fukushima Daiaki Nuclear Power Plant.
20:52 Speaker 06 Nuclear officials there are warning of a possible nuclear reactor meltdown.
20:57 Speaker 00 The loss of primary electric power and the water flooding the emergency diesel generators caused a blackout. Without power, the cooling systems failed, leading to an uncontrolled increase in reactor temperatures and a partial meltdown.
21:11 Speaker 06 Fuel rods are now exposed, and if they stay that way, they could release radioactivity and a disaster of unknown proportions.
21:20 Speaker 01 When the Fukushima accident happened in Japan, that was the biggest, longest-running, most intense ever that we supported. There was a huge concern about what's going to happen next. These plants were unstable. The flooding due to the tsunami knocked out their backup power generators. So they were bringing in generators to try and keep the water pumping, to keep the reactors cool, and that was not working completely. So there was a huge change in the conditions over time that we had to deal with.
21:55 Speaker 00 NERAC's involvement included real-time simulations of radioactive release and spread, leveraging meteorological data and complex atmospheric models to forecast the potential impact areas. These predictions were essential for informing the US government on protective measures needed for US citizens in Japan and potential impacts in the US.
22:16 Speaker 01 We set up a high-resolution weather model over Japan. So we had the highest-resolution US government model running for that area. We had experts in weather forecasting and using meteorological observations to improve the forecast. So we were doing very high-resolution weather forecasts for the region and constantly updating those as new weather data comes in. So in that case, we worked round the clock on shifts, seven days a week, for 22 days straight. And then after that, we were still analyzing the problem as the emergency was less severe, but people still needed analyses of what the areas were potentially impacted from releases from the Fukushima nuclear power plant in Japan.
23:00 Speaker 00 It was grueling.
23:02 Speaker 01 Wake up, drink some coffee, go back in. It became a routine schedule for several weeks like that. So this is going on over weeks and weeks. Reactor conditions are changing, weather conditions are changing, and we're also trying to do several things at once. Not only analyzing what's already happened, how much radioactivity's been released from the plant, what are the contamination levels in the environment so that we can help the government in Japan, the US government, assess what's already happened.
23:32 Speaker 00 NERAC's models help determine the potential radiation doses over time, allowing authorities to make informed decisions about public safety and resource allocation. Their expertise ensured that both Japanese and international agencies could respond effectively to the evolving situation. While some of the damage caused by the tsunami was obvious, others, such as radiation exposure, are not so immediately obvious.
23:58 Speaker 03 We use physics-based computer modeling to predict where radioactive material might go through the atmosphere and then eventually land on the ground. So then we take that calculation of material on the ground and predict how that would affect people who are living in that area. Depends on a lot of factors. It depends on how long people stay there.
24:20 Speaker 00 And once the NERAC assessments have been made, the Department of Energy is ready to provide that information to local decision-makers.
24:27 Speaker 03 What we try to do in advance of an incident is to make sure that our products and our reports are designed and written in a way that helps tell what we're trying to tell, helps the responders explain to the local decision makers what those effects are and what they can do about it. The federal government has a lot of already existing guidance documents because we don't want to be making this up in the event of an emergency. We have it all prepared in advance. For example, the Environmental Protection Agency has a guide about what kind of radiation dose levels would warrant taking protective actions, like sheltering in place, or evacuating
25:10 Speaker 03 an area, or even relocating for the long term, people who used to live in an area.
25:16 Speaker 00 In addition to these guidance documents, the NERAC process incorporates detailed health physics models to predict and assess radiation exposure.
25:24 Speaker 03 The health physics models are baked into the NERAC process. And so after the physics models are done and they have predicted material on the ground, like grams per square meter of soil, that's how much material is on the ground, then the health physics models take that amount of material that's on the ground, and you put in certain assumptions, like we assume that the person is going to be standing there outside for four entire days after this material lands on the ground. How much radiation dose will they be exposed to then?
25:57 Speaker 00 Everyone at NERAC mobilized to assist with the Fukushima disaster, playing a crucial role in mitigating damage and guiding the international response. This collaboration with the Department of Energy, working with agencies and organizations around the world, is another one of the team's key differentiators. This ensures that best practices and technological advancements are shared globally. Such partnerships are vital for managing cross-border environmental impacts and ensuring a coordinated global response. Even in the absence of major global disasters, NERAC remains actively engaged. They respond to local emergencies, such as toxic industrial chemical spills, fires, and potential nuclear facility accidents.
26:41 Speaker 00 NERAC continues to innovate, developing new technologies and refining predictive techniques to better prepare for future incidents. In a world that is rapidly changing, comes new threats. And NARAC stands ready to look out for people around the globe, from Europe to Pennsylvania to Japan. Not just to respond to disasters we know, but also to make sure we're ready for ones we haven't faced yet.
27:07 Speaker 06 For research, we definitely use the high-performance computing systems, including the new hybrid architectures that have GPUs in them. For example, like one of the projects that I worked on was looking at nuclear winter. This is the idea that in a nuclear war, there would be fires that could arise after a detonation and that smoke could rise into the stratosphere where it would have a long residence time and it could block sunlight to the Earth's surface and it could affect, for example, agriculture and our ability to grow food at the Earth's surface. So we did a project looking at that recently.
27:49 Speaker 06 That was a collaboration that NARAC did with the Climate Program. We simulated the fires, whereas the Climate Program then took the smoke from our fires and ingested those into the climate model to run the longer-term impacts.
28:05 Speaker 00 Nuclear technology is part of our modern landscape, and the dangers associated with it are irrefutable. But we can all rest a little easier knowing that NARAC is not only watching, waiting, and ready for the call, they are actively working to improve the world.
28:22 Speaker 01 I do get super excited about the future, mainly because we have some of the brightest young scientists in the world working here that are really dedicated to the science, but also the mission of making the world safer in the small but important ways that we can. We're a mission-focused lab. We do great science, which is valuable, but we apply it to a mission. So the dedication of our people to that mission and making the world safer in the ways that we can is extraordinary. I've witnessed extreme dedication during emergencies when people work around the clock to the point of extreme fatigue.
29:06 Speaker 01 But they did it because they wanted to do it and they were helping, they were dedicated to a mission.
29:12 Speaker 00 And like a true scientist, John has a specific formula he believes up-and-coming researchers can use to make themselves indispensable at NARAC.
29:21 Speaker 01 Follow your passion and learn as much as you can. Those go together. You have to have a passion for what you're doing. And the more you can learn, the more you can educate yourself to fulfill that passion. Those are the keys. That's what people at our lab do. That's what makes an exciting place to work.
29:45 Speaker 00 Preparedness is the difference between catastrophe and control. Each member of NARAC stands as a guard, always watching and always ready. As we look to the future, Lawrence Livermore National Laboratory's commitment to leveraging cutting-edge technologies like quantum computing and advanced sensor networks is poised to help the team continue to revolutionize environmental modeling. With a mission to make the world safer, NARAC exemplifies the power of scientific innovation in addressing the challenges of today while preparing for the uncertainties of tomorrow.
30:30 Speaker 00 Lawrence Livermore National Laboratory is hiring. If you're passionate about tackling real-world challenges in science, engineering, business, or skilled trades, there's a place for you at the lab. Right now, positions are open for a Senior Guidance, Navigation, and Controls Engineer and a Senior Business Operations Specialist. These are just a few of the more than 100 exciting roles available. At Lawrence Livermore, you'll work on projects that matter, from national security to cutting-edge scientific advancements. Join a team that values innovation, collaboration, and professional growth. Explore opportunities at llnl.gov/careers, where your next career move could make history.
31:26 Speaker 00 Thank you for tuning in to Big Ideas Lab. If you loved what you heard, please let us know by leaving a rating and review. And if you haven't already, don't forget to hit the follow or subscribe button in your podcast app to keep up with our latest episode. Thanks for listening.
Previous El Capitan Next Site 300 Page updated November 6, 2026 · transcript and summary are machine-generated