Subscribe
Share
Share
Embed
Direct Current is a podcast about energy -- the kind that lights our homes, powers our lives and shapes our world. From the U.S. Department of Energy's digital team in Washington, D.C., Direct Current brings you fresh, insightful stories of how we generate and use electricity, what that means for the planet, and the cutting-edge science that's driving a global energy revolution.
(DIRECT CURRENT THEME PLAYS)
MATT DOZIER: Welcome to Direct Current. I’m Matt Dozier.
SARAH HARMAN: And I’m Sarah Harman.
(ELECTRIC MUSIC PLAYS)
MATT DOZIER: Lately, it feels like weather has been in the news … a lot. Aside from the usual seasonal changes in temperature, rain, snowfall, there seems to be a noticeable shift happening. I’m talking about bigger, more extreme weather events — stuff that was once considered so rare scientists called them “hundred-year storms” — now becoming a staple of our daily headlines.
SARAH HARMAN: But even though it feels like storms and weather are intensifying, are they really? It turns out that yes, scientists are noting an increase in the frequency of extreme weather events around the world.
MATT DOZIER: According to NOAA, the National Oceanic and Atmospheric Administration, weather that can wreak havoc — like heat waves and big storms — is, in fact, happening more frequently. Which makes it critically important for us to better understand and predict what’s coming, so that communities can be better prepared.
SARAH HARMAN: Good news is, there are people who study weather and these extreme events. They look at big storms both past and present to build computer models that simulate what the future might look like.
(MUSIC FADES OUT)
COLIN ZARZYCKI: I would argue I've been in the field since 2005, which is when I went and started undergrad, although if you asked my parents, they'd say I was the kind of oddball that liked watching the Weather Channel instead of watching cartoons at certain times when I was a kid.
MATT DOZIER: That’s Colin Zarzycki (“zar-zick-ee”). He studied meteorology and environmental engineering before working at the National Center for Atmospheric Research in Colorado. Now, he’s an Assistant Professor at Penn State University.
COLIN ZARZYCKI: I really, really liked hurricanes and, actually, the Weather Channel — 10 minutes until the top of the hour, they did what they called the tropical update — and I was really fascinated by the storms. Part of it was the maps and the satellite images, and it was kind of amazing to me that we were getting these images from space of these potential storms, and we were trying to forecast where they were going to go. And I kind of found myself looking up at the clock being like, oh, I need to stop doing what I'm doing now and go catch this update, see what's going on with the hurricanes. That's really kind of kicked off a passion of mine, to get into the field.
(INTRIGUING MUSIC PLAYS)
SARAH HARMAN: Since the early 2000s, the number of scientific studies that investigate the connection between extreme weather and climate change has gone up. We know a lot about weather. And we know a lot about the climate, but we still don’t fully understand how the two pieces, or models, fit together.
SARAH HARMAN: Speaking of, we wanted to take a minute to talk about something that’s pretty basic, but key to understanding this story — the difference between weather and climate. See, they’re closely related, but they aren’t the same thing.
MATT DOZIER: Weather is made up of individual events. We say, “What’s the weather like today?” … It’s different in different places and changes throughout the day.
MATT DOZIER: Climate, on the other hand, describes what the weather is like in one place over a longer stretch of time. When we talk about climate, we’re usually using things like average statistics. Things like temperatures, rainfall, wind — over a span of years, usually decades or hundreds of years.
SARAH HARMAN: Now, there’s no question here about whether global warming leads to greater extremes in weather — it does. That part is settled. But when a record-breaking heat wave scorches the Midwest, or unprecedented rainfall inundates Southern California, people often ask: “Did climate change cause this?”
MATT DOZIER: Some scientists say the answer to that question is more nuanced than “yes or no,” “black or white.” Colin is one of them. And instead of saying, “This hurricane happened because of climate change,” he prefers to think of it this way:
(MUSIC FADES OUT)
COLIN ZARZYCKI: Yeah, this hurricane existed, and we had 20 percent more rain than we would have had if this hurricane had appeared in 1850. And that component is due to anthropogenic climate change.
MATT DOZIER: That’s a really important thing for us to get a handle on, as we try to protect lives and property from these increasingly nasty weather events.
SARAH HARMAN: Because in Colin’s example, that 20% increase in precipitation could mean the difference between minor damage and, say, levees breaking.
COLIN ZARZYCKI: Generally, when we think of extremes, we think of something that's pretty rare and pretty consequential for society. So, if it's something that happens, once a week, then society is likely pretty prepared for that. But if it's something that's kind of big, and that we're talking about happening, you know, once or twice a year, or even some of the extremes that I'm really interested in, which are more of these, like once every 10, or once every 100 year type events, those are the things that really have kind of that big generational impact.
(TECH-Y MUSIC PLAYS)
MATT DOZIER: So, in order to get ahead of these “generational” storms and droughts and all the other weather-related disasters waiting to happen, you need a computer model — something that can crunch up all the information we know about how the atmosphere works and spit out a prediction — or rather, lots and lots of predictions — of what’s to come.
SARAH HARMAN: The better our climate models, the clearer picture we’ll have of what to expect, weather-wise. That can help us identify the biggest risks and opportunities to mitigate damage ahead of time.
MATT DOZIER: When it comes to better preparing in advance for these types of events, Colin says we have to take a dual-faceted approach … We have to look backward, at the historical data about weather events in the past, and we have to project into the future, with modeling fueled by supercomputers. So, let’s start off by talking about the side of the coin that investigates the past.
COLIN ZARZYCKI: We kind of start to think back, maybe, to pre-industrial times, so thinking back to like, our great, great, great grandparents or something like that. But even looking since maybe like the 1950s or 1960s.
SARAH HARMAN: The quality of historical weather data that’s available for Colin and his team to analyze varies quite a bit … As you can imagine, getting reliable, quality data from a hundred years ago is pretty challenging. There are records, but they’re sparse.
MATT DOZIER: As technology became more sophisticated, local airports began tracking things like temperature and rainfall, which are useful data points. Then about 40 years ago, satellites started collecting all sorts of weather-related data. So, the team brings together all of these pieces of information and basically, they search through the archives to compare extreme weather events from way back when, to today.
(MUSIC FADES OUT)
COLIN ZARZYCKI: And what you can actually do is you can look at the ratio, how many record highs you have, how many record lows do you have? If our climate wasn't changing, that ratio would actually stay constant. And what we've seen historically, especially over the last couple decades, is record highs are rapidly outpacing record lows, which also gives us some confidence that we're seeing this trend line in both the temperature, but also the extremes associated with that temperature.
SARAH HARMAN: The other side of this research coin in Colin’s two-pronged approach is focused on what’s yet to unfold. By building models based on our physical understanding of Earth system, Colin and his team are working towards being able to make prognostications, or projections of what we can expect to happen in the future.
COLIN ZARZYCKI: At the end of the day, it's just boiled down to a bunch of computer code. What we do is we kind of put all these pieces of code together, you know, I like to tell my students, it's somewhat like putting a bunch of Lego blocks together and kind of creating this big, what we call a model, and then we essentially run the model.
MATT DOZIER: Running the model, as Colin says, isn’t as straightforward as pressing a button once or flipping a switch. Instead, it’s like rolling dice hundreds or thousands of times and logging each outcome. But the information that they’re using is so much more complex, Colin’s team runs it on a supercomputer.
SARAH HARMAN: You can think of a supercomputer like hundreds or thousands of laptops collectively working on the same problem at the same time. Each laptop is given a little piece of the problem, like what’s going on in the Great Lakes region, and then when brought together, the collective results simulate what the surface of the earth would be like at a future point in time.
COLIN ZARZYCKI: And what the model is really doing is it's simulating the weather over and over and over and over again, in many cases for many years, or many decades, or even many centuries. And we can run different iterations of the model to kind of look at different sets of weather patterns — different rolls of the dice, you might say. And then what we essentially try to do is we try to use those results to say something about how we would expect extreme weather events to change.
SARAH HARMAN: If the idea of a bunch of numbers on the computer swirling together to form a picture of future weather seems a bit… abstract… well, you’re not wrong. But these calculations can reveal patterns.
MATT DOZIER: So, Colin and his team track down historical weather data, write code, build the models, and then ship it all off to a supercomputer to crunch the numbers. Of course, even that is kind of oversimplifying — there’s a lot of back and forth to make sure everything looks right, and then the output that they get back from the supercomputer is, frankly, super overwhelming!
COLIN ZARZYCKI: And then essentially, what the supercomputer does, is it prints out petabytes of information. And this is a tremendously massive number, like, you and I are used to thinking about megabytes in terms of what we can send through email, or maybe gigabytes for our hard drives. And this is orders of magnitude more. This would require thousands of external hard drives to be able to store it if I wanted to store it here in my office.
(INSIGHTFUL MUSIC PLAYS)
SARAH HARMAN: For reference, a petabyte is one million gigabytes. The models Colin works with basically report the weather for every single hour over the next hundred years. From there, the team has to figure out how to mine that data for the important stuff.
COLIN ZARZYCKI: We write software code that goes in, like the claw machine, like something you'd play within an arcade, and goes in and extracts those extreme weather events. Because like I said, they don't happen that often, but that's what we're really interested in.
MATT DOZIER: So how do you go about doing that? Last time I checked, most research labs don’t have the space or the budget for thousands of laptops running at once. Instead, Colin taps into the supercomputing strength of the Department of Energy. DOE and the National Labs fund and operate supercomputers across the country, which researchers like Colin can apply to use. Once a proposal is approved, the researchers can then build and run their models from afar.
(MUSIC FADES OUT)
COLIN ZARZYCKI: A lot of our research really relies heavily on supercomputing capabilities offered by the Department of Energy. We have a supercomputer here on campus and we use it a lot for testing, and maybe doing a little bit of data analysis. But it's orders of magnitude smaller than the kind of computing capacity at some of these DOE centers. And we really, really want to make the most efficient use of computing, we want to really simulate these future climates with as much realism as possible, we want to be able to kind of see as many of these extreme weather events, we really need to leverage that state of the art cutting edge, supercomputing capacity, which is offered by the Department of Energy.
MATT DOZIER: Colin’s supercomputer of choice is called "Perlmutter". It’s in California and a part of National Energy Research Scientific Computing Center. Also in California is Alan Rhoades. He’s a research scientist with Lawrence Berkeley National Laboratory, and he’s one of those climate modelers who spend a lot of time investigating some of the extreme weather scenarios that Colin mentioned.
ALAN RHOADES: A lot of my work is sitting in front of a computer and accessing the NERSC supercomputing system, which we have here locally at Berkeley Lab. And basically, that's racks and racks of blinking lights that crunch all of the mathematical equations of atmospheric motion and thermodynamics and everything that are in our Earth system models, in a really fast and consistent way.
SARAH HARMAN: Alan has been working in climate modeling for a decade. Like Colin, his fascination with atmospheric science started early on in his life. He grew up near a large mountain range in California, called the Sierra Nevada, and he remembers a big storm in the late 1990s that dumped a lot of water on the West Coast. For Alan, the effects of the storm were so memorable, it helped to spark a fascination, and eventually, an academic path in weather and the environment.
ALAN RHOADES: I think I had a lot of interest in doing observations and going on expeditions — for example, in my undergrad — to far off places. But for some reason, the allure of the “crystal ball” of Earth system models or climate models to look at the future and what it might hold for us, I think that really got me excited about being an earth systems modeler and working here at Berkeley Lab.
(SERENE MUSIC PLAYS)
MATT DOZIER: Alan focuses on studying precipitation-related weather events, specifically “atmospheric rivers.” These are features in the atmosphere that move large amounts of water vapor from the tropics to the poles — like rivers, but in the sky. And when they make landfall, they can unload a ton of precipitation either as rain or snow and are largely responsible for most of the major flood events on the West Coast.
ALAN RHOADES: These features are about three kilometers deep, about 850 kilometers wide and 1000s of kilometers long. So, they're even larger than some of the biggest rivers in the world. And depending on how extreme they are, some of them have been considered to be on the order of about 7 to 15 Mississippi Rivers worth of water transport.
(MUSIC FADES OUT)
MATT DOZIER: To do this type of research, Alan spends a lot of his time writing computer code to fuel the models that run through all sorts of hypothetical weather and climate patterns. He gets data from field observations gathered by people who go out into the mountains to measure snow and rainfall, and more recently, from things like satellites and airplanes.
SARAH HARMAN: But despite the recent advances that remote and automated technology can offer, ultimately, Alan says the amount of data points available isn’t enough to run the models. So, they use more coding to fill in the blanks. And once they have these more-complete data sets, they use the computer models to answer questions like:
ALAN RHOADES: What if we were to model atmospheric rivers in present climate versus in a pre-industrial or if greenhouse gases hadn't occurred? Or what if we warmed the planet by one and a half degrees, or two, or three, which are in line with some of the IPCC — Intergovernmental Panel on Climate Change — warming levels of interest from a policy perspective? And so, a lot of my work is kind of working in the virtual world. How atmospheric rivers and virtual mountain snowpack might respond to these different levels of warming.
MATT DOZIER: As of Spring 2023, when this podcast was recorded, atmospheric rivers have been dominating headlines on the West Coast of the US. In January alone, a whopping *nine* atmospheric rivers inundated the area. So, Alan is interested in how well the modeling systems can represent these kinds of events, even as they’re happening. He’s also learning about the downstream impacts for water managers in drought-prone places like California and Colorado.
ALAN RHOADES: Water management in the West is very complicated, we get a lot of booms and busts. We're constantly in between flooding and drought. And so, a lot of the work that I do is trying to interrogate why that's occurring, and how those occurrences of floods and droughts might become amplified, or more persistent in different warming worlds. And I think the tool that I use to do that and rely upon are these earth system models that I keep mentioning.
SARAH HARMAN: Alan is part of a DOE-funded project called “Hyperfacets.” It’s a collaboration between a dozen research institutions that aims to get cutting-edge modeling data into the hands of people who need it most. People like water resource managers in Southern California, for example.
ALAN RHOADES: I think it's important because it allows us to be proactive about changing our management of our current infrastructure, or think about how do we scale up other water management strategies that might be more responsive or resilient in a warmer world where you get bigger, badder atmospheric rivers, maybe less snowpack, or flashier snow melt, or more potential for flood events, or maybe you have persistent, low-to-no snow conditions where you can't rely on mountain snowpack anymore as a natural reservoir? And so, I think our models allow us to really be proactive about thinking about those things, and doing it in a way that's not too simplified, or can maybe provide us with physics informed estimates of what those plausible futures might be.
(POSITIVE MUSIC PLAYS)
SARAH HARMAN: We may not be able to pull out a crystal ball and predict future extreme weather events, like Alan dreamed about when he was younger, but with this type of collaborative research, we may be one step closer. And for Colin, all of the looking backward and projecting into the future is worth it, to help minimize the destruction caused by extreme weather.
COLIN ZARZYCKI: What I'm really interested in is being able to provide some degree of information to society, about how we want to prepare and harden ourselves against these extreme weather phenomena So that we can preserve life and property. Essentially, at the end of the day, t hat's really our goal is to make sure that people are staying safe, that we're keeping property as buffered and hardened as possible against these impacts. But we can't plan and we can't think about how we're going to do that, unless we have some understanding of what extreme weather is going to look like in the future.
(MUSIC FADES OUT)
(OUTRO MUSIC PLAYS)
MATT DOZIER: That’s it for another episode of Direct Current! Thank you to our guests, Alan Rhoades and Colin Zarzycki, for lending their expertise.
SARAH HARMAN: If you want to learn more about the science of understanding and predicting extreme weather, check out our show notes. You can find those, and all of our other episodes, at energy.gov/podcast.
MATT DOZIER: We also wanted to give a big shoutout to Ashleigh Papp and our Office of Science. Ashleigh was a huge help in reporting and writing this episode.
MATT DOZIER: Direct Current is produced by me, Matt Dozier. Music and sound editing assistance by Michael Stewart.
SARAH HARMAN: And our episode artwork is by me, Sarah Harman.
MATT DOZIER: If you’ve listened this far, you should know that this is my last episode as host and producer of the podcast. Making this show has been one of the most fun and rewarding parts of my job since we launched it in 2016, and I’m really proud of the stories we’ve been able to tell and the important work we’ve highlighted. It means a lot that so many people have shared the show and gotten in touch with us. Your feedback has been greatly appreciated — and I hope you’ll keep listening! Because there’s lots more to come, in Season 4 and beyond.
SARAH HARMAN: This is a production of the U.S. Department of Energy and published from our nation’s capital in Washington, D.C.
MATT DOZIER: And as always, thank you so, so much for listening.
(MUSIC FADES OUT)