Big Ideas Lab - NARAC

Episode Date: November 26, 2024

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?Enter NARAC, the National Atmospheric Release Advisory Center. Built on Livermore’s expertise in nuclear physics and supercomputing, NARAC was created to predict the spread of hazardous materials in the atmosphere during emergencies. From Chernobyl to Fukushima to Three Mile Island, NARAC’s scientists have worked on some of the hardest problems of our time—helping to assess risks, save lives, and inform critical decisions.This episode dives into the origins of NARAC, its groundbreaking mission, and how it continues to prepare for the unthinkable.-- Big Ideas Lab is a Mission.org original series. Executive Produced and Written by Lacey Peace. Sound Design, Music Edit and Mix by Daniel Brunelle. Story Editing by Daniel Brunelle. Audio Engineering and Editing by Matthew Powell. Narrated by Matthew Powell. Video Production by Levi Hanusch. Guests featured in this episode (in order of appearance): John Nasstrom, Chief Scientist for NARAC at LLNLLydia Tai, Health Physicist at LLNLLee Glascoe, Program Leader at LLNL for the Nuclear Emergency Support Team Katie Lundquist, Model Development Lead for NARAC at LLNLBrought to you in partnership with Lawrence Livermore National Laboratory.

It was April 1986, and scientists in Scandinavia were perplexed. The readings didn't make sense. Routine radiation screenings at nuclear facilities showed spikes in radiation levels. Spikes that seemed to be triggered by the scientists themselves. 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. 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. 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. In the days that followed, the lab's team at ARAC, the Atmospheric Release Advisory Capability, was able to trace the radiation back to a single origin point.

It was coming from the Soviet Union. 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 to have been released. And it was substantial. It's still the worst nuclear accident in history. On April 28, 1986, an announcement was made on Soviet television. It's now clear that the Soviet Union has suffered one of the worst disasters in the

history of nuclear power. Two days prior, on April 26, during a standard safety test at a nuclear power plant near Pripyat, 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. 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. 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 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. 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. Joe Knox, Marv Dickerson did something very bold. That's John Nastrom, the chief scientist for the National Atmospheric Release Advisory Center at Lawrence Livermore National Laboratory. 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.

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. 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. Lydia Tai is a health physicist at the lab. 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. 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 hoped 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 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. 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 dramatically improves the weather forecast. 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 Lawrence was working on weather prediction models and discovered that small changes in initial conditions could lead to vastly different outcomes. This concept led to Lawrence's unanswerable question. Does the flap of a butterfly's 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. 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.

Lee Glasgow is the program leader at Lawrence Livermore National Laboratory for their nuclear emergency support team. 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. Building on this groundbreaking work, others at Lawrence Livermore National Laboratory wanted to use atmospheric modeling for airborne hazard prediction, too. 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. 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. The government official said that a breakdown in an atomic power plant in Pennsylvania today is probably the worst nuclear reactor accident to date. In March 28, 1979, the Three Mile Island accident started unfolding. The accident occurred here at the Three Mile Island nuclear power plant a dozen miles south of Harrisburg. 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. At about four o'clock this morning, two water pumps that help cool reactor number two shut down. A malfunction in the secondary cooling circuit of the unit two reactor resulted in the failure of one of the main feed water 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. Officials say some 50 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 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. Three Mile Island was not a remote location. Situated in the London Berry 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.

How much radioactivity has already escaped from the plant? How high is the radiation level inside the plant? 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. 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? 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 held as an essential new tool. The people around Harrisburg, Pennsylvania, were told they can sleep in their own homes tonight.

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. Iraq 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.

DOE asked if we could take this operational prototype and apply it. Our staff worked around the clock on shifts for two weeks straight. We did time-evolving, three-dimensional atmospheric modeling forecasting that was basically real-time modeling. 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. 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. 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. 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,

so that's Laboratory Directed Research and Development Project, to develop a atmospheric model that will run on accelerated computer architectures. And it's really promising to revolutionize what we can do in our atmospheric models. As these advanced computational capabilities come into play, they promise significant enhancement in both speed and accuracy, transforming our approach to atmospheric modeling.

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. From there, we would then make refinements to that prediction. This means that when an emergency hits, within just 15 minutes, NARAC's automated system has a preliminary model and report ready for emergency responders to start using.

In a crisis, NARAC 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. But what NARAC has 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 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 to decision-makers the information that they need quickly, not just the atmospheric results. 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. 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. 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. On March 11, 2011, an earthquake hit off the coast of Japan. Quite simply the biggest, longest lasting earthquake I've ever experienced. The 9.0 magnitude quake created a massive tsunami. The wave reached heights of 45 feet as it barreled towards the coast.

When it struck land, it quickly overtook the seawalls of the Fukushima-Taiki nuclear power plant. Nuclear officials there are warning of a possible nuclear reactor meltdown. 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. Fuel rods are now exposed and if they stay that way, they could release radioactivity

and a disaster of unknown proportions. When the Fukushima accident happened in Japan, that was the biggest, longest-running, most intense effort that we supported. There was a huge concern about what's gonna 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. NARAC'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 U.S. government on protective measures needed for U.S. citizens in Japan and potential impacts in the U.S.

We set up a high resolution weather model over Japan, so we had the highest resolution U.S. 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.

It was grueling. 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 they're also trying to do several things at once. Not only analyzing what's already happened, how much radioactivity has been released from the plant, what are the contamination levels in the environment so that we can help the government in Japan, the U.S. government, assess what's already happened.

NARAC'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.

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. And once the NERAC assessments have been made, the Department of Energy is ready to provide that information to local decision makers. 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 an area or even relocating for the long term people who used to live in an area. In addition to these guidance documents, the NARAC process incorporates detailed health physics models to predict and assess radiation exposure. The health physics models are baked into the NARIC 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? 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, NARAC remains actively engaged. They respond to local emergencies such as toxic industrial chemical spills, fires, and potential nuclear facility accidents. NARAC 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. 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. That was a collaboration that NARAQ 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. 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. 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 witnessed extreme dedication during emergencies when people worked round the clock

to the point of extreme fatigue, but they people work around the clock to the point of extreme fatigue, but they did it because they wanted to do it and they were helping. They were dedicated to a mission. And like a true scientist, John has a specific formula he believes up and coming researchers can use to make themselves indispensable at NARAC. 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. Preparedness is the difference between catastrophe and control. Each member of NARAC stands as a guardian, 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, NERAC exemplifies the power of scientific innovation in addressing the challenges of today while preparing for the uncertainties of tomorrow. Lawrence Livermore National Laboratory invites you to join our diverse team of professionals where opportunities abound for engineers, scientists, IT experts, welders, administrative and business professionals, and more. At Lawrence Livermore National Laboratory, your contributions are not just jobs, they're a chance to make an impact from strengthening U.S. security to leading the charge in revolutionary energy solutions and expanding the boundaries of scientific knowledge.

Our culture at the lab values collaboration, innovation, and a relentless pursuit of excellence. We're committed to nurturing your professional journey within a supportive workspace and offering a comprehensive benefits package designed to ensure your well-being and secure your future. Seize the opportunity to help solve something monumental. Dive into Lawrence Livermore National Laboratory's wide variety of job openings at LLNL.gov forward slash careers where you can also learn more about our application process. This is your chance to join a team dedicated to a mission that matters. Make your mark. Visit LLNl.gov forward slash careers today

to discover the roles waiting for you. Remember, your expertise might just be the spotlight of our next podcast interview. Don't delay, uncover the myriad of opportunities available at Lawrence Livermore National Laboratory. history. 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.