Super(computer climate)model - with Dr. Travis O'Brien
Hi, folks. Welcome to Earth on the Rocks, a show where we get to know folks behind the science over drinks. I'm your host, Shelby Raider, and with me today is doctor Travis O'Brien. Travis, welcome.
Travis:Hello.
Shelby:So you are also part of the Earth and Atmospheric Sciences department here, but you are more on the atmospheric side. So you're gonna be the first person that we have on the podcast that's sort of representing the atmosphere. So do them proud.
Travis:In the clouds. Alright.
Shelby:How if someone were to ask you what sort of scientist are you, how would you answer that? How would you describe yourself?
Travis:I'm a climate scientist.
Shelby:And so what does that mean?
Travis:I do research on weather and climate, how that changes with time, what makes the weather around us work.
Shelby:And so we're gonna delve into that a little more today.
Travis:K.
Shelby:And we're gonna be doing this over drinks, and so I have to ask, what is your drink of choice, Travis?
Travis:Oh, drink of choice is often gonna be a type of whiskey. I I can't name a favorite. I like a lot of them. Rye whiskeys are usually the thing that I turn to if I have a lot of choice.
Shelby:Yeah. That is an excellent choice, and strong agreement there.
Travis:Mhmm.
Shelby:And this time of year, especially, I feel like as it starts to get cooler, those really hit the spot.
Travis:Yeah. Indeed.
Shelby:So you say you're a climate scientist. Mhmm. And, I think that, you know, folks that are listening may have very different ideas of of what that entails. And so in terms of how you specifically approach climate science, what does that look like? What is your approach for that?
Travis:What does that look like? Well, if you walk by my office at any given time, what it looks like is me just sitting there at a keyboard getting pasty. So I I sit inside. A lot of what I do is on the computer. I do I analyze weather data, existing datasets.
Travis:I run climate models. These are these computer models of how the atmosphere and ocean and all that work. Yeah.
Shelby:And and when you're doing this sort of work, what sorts of questions are you trying to answer? So what's your sort of specific aim whenever you're thinking about the climate?
Travis:So let me think actually, a good first way to answer that is, like, what what are some of the questions I've worked on really recently? And, for example, a lot of what I do focuses on extreme weather. So when a big weather event happens, like hurricane Helene or hurricane Milton, like just happened, the research group that I work with, we will often get together and go, okay. What caused this? Like, what were the weather conditions that made this happen?
Travis:What is it more intense than other events that have happened? Where does it fall on the record? And did climate change play a role?
Shelby:And so this could be any sort of, like, extreme weather event, or is it specifically sort of things like hurricanes like you mentioned, or can it can it be other sorts of weather events too?
Travis:It can be other sorts of weather events. We've the research group that I work with, it's this project called cascade that I've been working with for years out of Lawrence Berkeley Lab. A lot of the research has been on hurricanes. We've we did a paper in the last few years on the there's a big heat wave in the Pacific Northwest back in 2021, did some work on that.
Shelby:And so when you think of or or as you sort of posed it, you know, you ask, was this more intense than other storms? What were the weather conditions at the time, and did climate change have a role in that?
Travis:Mhmm.
Shelby:How do you sort of evaluate that? What do what do you mean by does climate change have a role in that?
Travis:Oh, man. Good question. Actually, I'm I'm teaching a class right now on this exact topic, which is one of the things I'm really excited about being at IU in doing is being being able to share my research with students who are really interested in learning about this stuff. So class is Current and Future Trends and Extreme Weather. And, you know, one of the things that we think about is reframing questions, because often people especially if you you hear media about big events, a question you'll ask is, did climate change cause this, which is not quite the right question to be asking.
Travis:A lot of these events could happen even without humans around, but have humans made these events more likely? Have they made them stronger? And there's a lot of different ways of doing it. Some of it is using climate models, which is sort of like digital twins is another term that people have used of earth that we can do experiments on. So turn things on and off, remove greenhouse gases, see what that does.
Travis:There's statistical or mathematical approaches that we can take to look at existing data that we have, the weather data that we've recorded over the last, you know, century or so and, compare that with trends in c o two, aerosols, which is stuff like pollution and things like that, and see what effects that has.
Shelby:And so you mentioned climate models, so I think that's something that, you know, you might hear these terms but are unfamiliar with
Travis:Yeah.
Shelby:With what those are. And so I think it it might be nice to talk a little bit about what is a climate model, how do they function, are you running those here through IU, what does that look like in that sense? So maybe you gave a little bit of a description, but what exactly is a climate model? What do you mean when you say that?
Travis:So if you imagine taking the whole of the atmosphere and dividing it up into cubes, cubes that are stacked next to each other, joined on the side, What these models do is, you know, if if you take ever take a physics course, one of the first things that you learn is that mass is conserved, energy is conserved, momentum is conserved. There and that's the basis of all, like, physics based fields of study. So the climate models take these divide the atmosphere and the ocean up into boxes, and they there's equations that describe how mass, like humidity, moves through the grid boxes, how winds flow, that's momentum, or how temperature moves, that's energy. And they basically they convert these horrendously looking calculus equations for, you know, students who ever take calculus into algebra, just really big algebra problems that you need a computer to solve. I mean, actually, it's it's kinda cool.
Travis:Some of the earliest weather models, which are the same type of model, involved people sitting in a room, like, in a big stadium room and doing these calculations, like each person sort of representing a grid box and, you know, passing momentum to 1 person mathematically and so on. So, yeah, computers do this, and we we run these here at IU. Yeah. And, actually, I I've managed to get, I think, 12 undergraduate students. Maybe it's 13 or 14 at this point since coming to IU to learn how to actually run these models at IU, which has been really exciting to actually see something that people traditionally think is, like, so complex that you you'd need to spend years learning you know, getting over the learning curve.
Travis:But, partly because IU's got free supercomputing and partly because the particular model that I'm using, we've got it set up in a way that there's just a handful of skills that the students need to learn in advance, and they actually get to run it.
Shelby:So you mentioned a supercomputer. That's probably also a term that folks are not familiar with. So IU has a supercomputer, which is which is sort of impressive. And so what is a supercomputer, and how do you utilize it?
Travis:Well, off, it's not a computer with a cape. If only. If only. Yeah. I know.
Shelby:It's really what
Travis:it is, it's it's a bunch of computers acting together as one gigantic computer. So imagine that, you know, you take a laptop, you take 10 laptops, you connect them together, and you give part of a computational problem to each of them. So maybe you give each of the 10 laptops, like, a tenth of the planet to do the math on, and then they work on that part of it and share information between them. A supercomputer is usually gonna be, like, thousands of these computers. You don't necessarily use all 1,000 at the same time, but you can.
Shelby:Have you so I work at IU, obviously, and I've never seen the supercomputing facility. Have you seen it?
Travis:Here? No. I have seen the one at, the National Energy Research Supercomputing Center NERSC in Berkeley, which is where I worked prior to coming here. They are cool.
Shelby:What are they can you walk us through what they look like or what it is to to experience a supercomputer?
Travis:Yeah. So, so you you you walk into this room and what you won't see actually immediately recognizes a computer. What you will see is what looks like big fancy looking cabinets that people usually paint, like, fancy imagery on. You know, like, I I so the current one at NERSC is Perlmutter, named after Saul Perlmutter. He's a recent Nobel Prize winner.
Travis:So it's got a picture of him and some imagery related to the work that he did on astrophysics that earned him the Nobel Prize, and it's like rows and rows of these racks of computers. If you open up the racks, it's these metal racks. They they they're called blade servers, so it's like a computer that's, you know, maybe 2 inches tall, couple feet wide, and a couple feet deep, and there's just stacks of these computers, and then very nicely organized cables everywhere. I mean, for a person who's really, like, you know, likes organizing things, it's very pleasing to see. Yeah.
Travis:But what's really cool, though the computers themselves are kinda neat. What's really cool is that the places where the data is stored, because these you know, when you run simulations like this in any field, it's gonna produce a lot of data. And believe it or not, even now still, the cheapest way most efficient like, space efficient way to store data is magnetic tapes, like tapes from the 19 eighties, you know, sharing mixed tapes. And so what's really cool is these tapes are in racks. Like, they just sit there.
Travis:They're not being used most of the time, and these robotic arms go and grab them. And so you've got hundreds of people using these super super supercomputers, and a person can request data on a tape. A robot will literally move an arm and grab it, or it might be multiple tapes. And so these robotic arms are just going back and forth, grabbing tapes, loading it in, and all that.
Shelby:That is pretty wild. Yeah. So how long are those tapes stored? Are they stored indefinitely?
Travis:Yes. You're you're probably getting a question that I don't really know the answer to. I can guess that. The these tapes so there there's redundancy built in. Actually, usually, it's not just one tape center, but they're, you know, there's the tape center at the supercomputing facility, and then they exist somewhere physically different, just completely copied.
Travis:These tapes degrade over time. For example, something that people don't usually appreciate, but maybe if you listen to an old, Radiolab podcast, you might hear that, you know, cosmic rays. They're bombarding us all the time. They degrade these high energy particles from outer space. They hit these tapes and degrade data.
Travis:I'm I'm guessing that the the tapes are occasionally checked and probably copied and replaced, but, yeah, they just sit there, and they archive petabytes of data, petabytes. So typical laptops these days has, like, a terabyte of data. A 1,000 terabytes is a petabyte. I think NERSC has, like, 10 petabytes of storage. So Wow.
Travis:You know, 10,000 laptop drives worth of storage.
Shelby:That's no small feat. No. So when you are running these models, so you're you're reaching out to these supercomputers that are basically thousands of regular computers linked together and working as a unit. So that's a lot of computation power.
Travis:Mhmm.
Shelby:How long are these how long does it take these simulations to run? So is this something that when you say, I wanna run this climate model, you hit let's run, and then, you know, 30 seconds later, you get a result? Is it an hour later? Is it a day? What do these look like?
Travis:Oh, that's that's an interesting one, and that really depends on the type of experiment that you're doing. One of the things that really controls how long it takes is how big your cubes are. So you divide the atmosphere up into cubes that are 100 kilometers, like the the roughly the width of the state of Indiana on a side. If you wanna run 30 years of climate simulation, that might take a few hours. That's not so bad.
Travis:If you wanna instead divide the Earth up into boxes that are 10 kilometers on the side, you're gonna have to every time you cut that grid spacing in half, you have to multiply the amount of time by 8. So, yeah, a 100 and a half is 50, 50 and a half is 25, 25 and a half is close enough to 10. So that's 8 times 8 times 8, which is a lot lot longer. I'm not gonna try and do that math off the top of my head right now. But yeah.
Travis:So so if you wanna do a single climate length experiment, that takes a long time. Some of the work that I do is looking at scenarios of, like, given the, you know, given the weather conditions we have right now, where might that where might the atmosphere go? Like, what sorts of really where extreme weather events are possible? In that case, you're not running one long simulation, you're running a bunch of independent simulations, and you can actually you can give each, you know, subcomputer of the nodes is the term that we use of the supercomputer. You can give each of those a climate model, and they can all run-in parallel.
Travis:So you I could actually take over, like, all 1,000 nodes of of Big Red 200 is our supercomputer here. They all run simultaneously. That might actually take an hour, and I could get, like, weeks of simulation from each, but over, you know, 1,000 simulations.
Shelby:Right. And so splitting up these grid cells into different sizes, is sort of changing the resolution of these models. And so I'm assuming for some things, you would want really small cells that you get really high resolution so you can sort of make out very small changes and things. And for other aspects, it might not be that important to have very fine resolution. Yep.
Shelby:And so for for what you look at, so looking at things sort of in this atmospheric level, are you more interested in these fine scale changes or in these big scale changes?
Travis:Yeah. More on the fine scale, and that that actually gets back to the question he asked me. Climate model is probably the general term, but regional climate modeler, if I'm in the more specific science community. So regional climate models are climate models that just focus on a spur certain part of the globe. They don't worry about weather moving around in other parts, and that allows you to reduce the size of the problem.
Travis:And so you can afford to have smaller grid boxes, find a resolution, and that allows you to, you know, focus on things like mesoscale convective events, these big thunderstorms that often come through in the summer, things that cause the ratios and tornadoes and stuff like that.
Shelby:So for either of those examples, either looking at sort of tornadoes in this region, which if anybody that's listening is from the Midwest, you're gonna be very familiar with those storms that Travis is describing or for these large hurricane events that just happened. How can you sort of walk us through how you approach that in terms of your simulation? So what data do you feed these models? What are you changing, and what are you trying to figure out sort of on the end?
Travis:Yeah. So let me give an example of a hurricane. We haven't done this particular experiment for hurricanes Helene and Milton, but we did it for some earlier hurricanes. So you might take so in terms of data, there are these datasets called reanalysis. They're kind of observations, but it's actually one of the big problems in atmospheric science is it's hard to observe the atmosphere.
Travis:Like, we don't know what the wind is everywhere. We don't know what the temperature is everywhere. We can measure it in certain locations. We have to guess somehow at what wind and temperature and humidity is locations in between. So and we there are a lot of sources of information.
Travis:I mean, we have you know, people measure specific stations. We send up balloons twice a day at 1,000 ish locations across the globe. We have satellites viewing the Earth, but each of those gives a sort of incomplete picture. A reanalysis actually takes a weather model, which is the atmospheric part of a climate model, and basically combines it with those observations to come up with a better guess of what the atmosphere is everywhere. So that's sort of our starting dataset is these things called reanalysis.
Travis:We feed those into a regional climate model that that allows you to not worry about the weather everywhere else, but maybe zoom in on the region of interest. So, modern day reanalysis, its resolution might be 25 kilometers. With a regional climate model, we might be able to afford, like, 1 kilometer grid boxes, which allows you to actually resolve convection, these these thunderstorms and stuff that are important for generating tornadoes. We can't resolve tornadoes though. That's that's a big catch.
Travis:That's, like, that's one of the big golden questions, but it's computationally just can't do that. Yeah. So, you know, these reanalysis feed in, ex an experiment that we might do to ask what effect did climate change have. So we do 2 versions of the event, the event that happened. So we simulate that event at high resolution with a regional model, and then we do a version where you there's increased greenhouse gases, warmer temperatures, higher humidity, a a version of that event, sort of like asking the question, if this same event happened, you know, 50 years from now in a warmer climate, what would it look like?
Travis:Would it be worse? And the answer turns out to almost always be, yes, it would be worse. Typically, more rainfall, is it for things like, hurricanes and stuff.
Shelby:And so is that generally the approach that your group and that you take is taking events that have already happened and then simulating how they may change in the future given variations in certain climate or weather conditions?
Travis:That's one of a lot of our approaches that we take. We we part of the thing that I love about working with this cascade group that I'd I'd mentioned is we have a lot of different expertise from different areas, statistics, computer science, physics, sort of like myself, physicist. And we have a lot of different approaches to these problems. So some of them involve climate models, so that approach that I just described, you can and there are multiple approaches with climate models. So another thing that we might do with the climate model is take these really long simulations, these scenarios of, you know, what might weather look like in the future, how how so if I wanted to ask the question, how often might we get really bad thunderstorms in a future climate?
Travis:I might take a really long running climate simulation where greenhouse gas ramps up in a way that we think it might over the next century and actually search for mesoscale convective systems in those, count them, look at the weather environment around them, look at how much rainfall falls in them and how that changes as c o two increases. So that's another really common approach. We also use just, like, purely statistics based approaches. So some of my colleagues are statisticians. They got their PhD in statistics departments, and they've developed new methods that allow us to actually just look at observations alone.
Travis:And so the reason this is really important is because none of these approaches on their own is perfect. They've all got problems that might be lead a person to go, well, this this, you know, these climate models aren't perfect. They have errors in them. Maybe they're getting maybe they're saying, you know, rainfall will be stronger, but, you know, convection is not often really well resolved, and these these models are well represented. But when we combine it with multiple approaches from different angles that are totally independent, keep coming to the same answer, that really boosts our confidence that we're that we know what we're talking about, that we're making good guesses about the future.
Shelby:Yeah. I mean, it sounds like the way you describe it, not only is is the team that's working on this very interdisciplinary, but the approach ends up being interdisciplinary to try to increase your your confidence in the the results that you all are getting, which sounds like that would be a pretty exciting place to sort of work in. Right? So you're always being stimulated by a lot of different fields at the same time.
Travis:Yeah. It is. It really is. And I've learned a lot along the way. I mean, I've I've learned a lot about I've I never actually took a statistics class in college, but I actually know a lot about it now.
Shelby:And so you had mentioned that that you've had several undergrads who have worked on this, and it just takes, you know, a small number of skills that they develop. What are some of those those skills that that you feel like students could benefit from having if they're interested in this sort of of work? Because I also think you might have thrown around some words that may be a little scary to people like statistics, physics, that sort of thing. And so so what does this look like as a student who may be coming into this?
Travis:So skill wise, one of the big skills is scientific computing is the broad term that I would give it. So most most of the times nowadays when people interact with a computer, it's with their touchpad, or maybe if it's an iPad, they're physically touching the screen. That's one way that you can interact with a computer. Historically, the first way that when computers were originally built, one of the original ways that people interacted with them is typing commands to a computer. Basically, all modern computers have an ability to talk to them through this thing called the command line.
Travis:That's these supercomputing systems, you just kinda have to interact with them in that way. And there's a language to talking to the computer in that way that needs to be learned. And so learning how to interact with a computer in a command line, that's one of the skills that students build. And there's a lot of really good free resources out there on the Internet that that that's sort of the the first couple weeks of when I start interacting with a student. Almost nobody has these skills, because they're it's kinda esoteric.
Travis:It's like, I don't know. You know, most people don't need to use them, and they don't learn them unless they need to. Yeah. So they pick them up quickly enough, and so they start to be able to type commands that tell the computer, you know, run this first step in the climate model. Open this text file, and let me edit it.
Travis:Combine these output files from the climate model, and do this. And then they learn a little bit of programming too along the way. And that that goes into scientific computing. Then it's yeah. I mean, that's that's one of the big things.
Travis:I think if a student really wanted to get really into any branch of science, like, programming is a really key skill.
Shelby:Yeah. I would agree that it's really useful. Yeah. I'll I will certainly date myself when I say this. So my sort of extent of computer programming is learning HTML code so that I could personalize my Myspace page when I was younger.
Shelby:And even that, you know, you feel so powerful when you can do that. I feel like for students that are learning how to to talk in this language to computers, that that would be a lot of fun and and something that would become pretty useful regardless of where they ended up.
Travis:Yeah. It it is it's it's really empowering because you can once you learn that skill, you can get computers to do things that nobody made them do. Like, computers can do a lot of really cool I I don't know. For me, like, I love programming. I'm I'm just a professional nerd through through and through here.
Travis:But, for me, I I get really bored with, you know, tedious, repetitive tasks, and that is a source of joy for me is to write a little computer program to get the computer to do the tedious task that I don't wanna do.
Shelby:Yeah. So. So so how did you end up in this this sort of field? Was this something when you were young, you thought, oh, man, I wanna I wanna go into sort of atmospheric modeling, or did you have an a fascination with the weather? What No.
Shelby:What spurred you to come here?
Travis:We and it's actually somewhat unusual. A lot of people who go into atmospheric science are, like, they knew from when they were kids. Like, talk to anyone, any of the atmospheric science students, like, they they knew forever. No. I never I mean, I enjoyed being outdoors and stuff, but when I was a little kid, I wanted to be a chemist and a professional baseball player.
Travis:Perfect combination. Yeah. You know, I got into college. I didn't entirely know what I wanted to do. I was between whether I wanted to do study music, whether I wanted to do physics, whether I wanted to do math.
Travis:I loved all of these things. I ended up going doing physics. As I said, I love programming, and I I've loved it since I was a kid. I I was fortunate that my my dad worked in the tech field, like, you know, his his whole career. I learned how to program when I was 13.
Travis:He helped me build my first computer when I was 13. So, like, I I've loved this stuff for a long time. And what was cool is that, you know, fast forward to the field that I'm in, I actually found a field that combines all these things. Not quite the music that I do on the side, but, like, computer programming, math, physics, this is this is what I do. I got my undergrad degree in physics.
Travis:I, you know, I did research in material science for my undergrad thesis there, and it was interesting, but it wasn't something that really drove me. And I I definitely had the thought early on, you know, I don't know if I could do this for 6 more months, let alone, like, my whole life. Maybe this isn't for me. What got me so I was living in Santa Cruz, California. It's a coastal town, beautiful campus.
Travis:It's in the redwoods. If you ever get a chance to go, it's really unique. What's funny is you can't even see that the campus is there. Like, you go on a tour, and it's just it's hidden in the redwoods. And, like, there's such and such building, and, like, where?
Travis:Yeah. But what's unique about it is it's in the fog. Like, California is known for its coastal fog. For those of you who've never visited California, you have the image in your mind of sunny beaches. That's true on some days, but also bring a sweater because that fog, when it comes out, it's really cold.
Travis:So I was fascinated by this, by, like, how variable the fog is. I I remember one day being in my house and looking out one side, and it's completely sunny. The other side, it's just gray and dark because the fog bank was off, it was it was just coming on shore. That fascinated me. I was really interested in it, and I started reading about it after I graduated.
Travis:I was like, what what are clouds even? I actually didn't know that. I'd I completed a whole physics screen. I was like, I actually don't know what a cloud is. Like, literally don't know what a cloud is.
Travis:And so I started doing reading. It's like, oh, this is really cool. And so I went back to talk to one of my professors who taught a climate class. It was I think the class was called Earth's history. I took it my sophomore year.
Travis:Well, I I didn't actually go to talk to her. What what was funny is, like, serendipity is, like, a lot of my career has just been, you know, random luck happening. I I went down to the coffee cart. I was, you know, doing that material science research just after my senior year. I was thinking, you know, I should go talk to her.
Travis:I think she's in this building. She happens to walk out right then. I hadn't seen her since I'd taken the class. She looks at me. She goes, oh, you were in my class.
Travis:I remember you. I was like, that's really funny. I was just thinking about you and, you know, that started the conversation. I met with her. We organized a meeting later on.
Travis:It turned out she was looking for a grad student. And she remembered me from class, apparently made a good impression, applied for grad school, and that was it.
Shelby:That was it. Yeah. And and so what field was grad school in?
Travis:So my degree is in earth and planetary science, but my research was in climate science. And so I got a actually, a broad training in geology as part of part of my background, although I don't do much geology or any geology, really.
Shelby:I think that that for both the Earth and atmospheric signs aside, I think that that's one thing that I've have come to appreciate both as a graduate student and then afterwards is that, you know, you don't have to go through an undergrad degree in an Earth or atmospheric science program to end up being very qualified and have very useful skills for a graduate program. So, like, I knew folks who came in who, you know, were interested in things like the climate modeling side, who had math degrees, but had no geology exposure whatsoever, or folks who may come into a geochemistry program with a chemistry degree, but very little geoscience exposure. And so as you sort of mentioned before, this sort of interdisciplinary nature really gives a lot of flexibility to how you sort of enter the field and the things that you can do with it, in my opinion, which is part of why I think it's so exciting.
Travis:I agree with that. And I think I mean, it talks about how a lot of anxiety that I felt when I was younger was around what do I do? What am I gonna do? And it it it's, you know, it at least in this field, and I've seen this actually in other fields too, you you can change your mind at almost any point. And I I knew somebody who got actually a degree in psychology and ended up being admitted into a graduate program in physics.
Travis:Like, that that big of a change. Yeah. It was it it was a challenge, of course. She had to learn a lot of background starting off. But, I mean and and it is really common in our field because there's not many undergraduate programs in climate science.
Travis:There are some in atmospheric science. So a lot of the grad students who come in, like you say, are from other fields.
Shelby:Yeah. And so then, you know, you sort of mentioned this changing a little bit. So after graduate school, you spent some time at a national lab.
Travis:Yeah.
Shelby:So first of all, what what is a national lab? What does that mean, and what what goes on there?
Travis:Yeah. So national lab, you might be thinking of Hawkins National Lab in Indiana. National National Labs, they they have their origin, you know, around World War 2. This need to do big team science on these big problems of, like, well, at the time, bombs. So some of the National Labs do really high security research, like and they have some of these National Labs, you go to them, you can't visit.
Travis:They they've got guards with machine guns at the gates because they do research on nuclear materials and research that's classified in some cases. The lab that I was at was Lawrence Berkeley Lab. It's actually one of the few that you're not allowed to bring confidential information or, onto that campus. It's a completely open campus. It's right next to UC Berkeley, Very, like, college like atmosphere, so really pretty open.
Travis:You can wander in and out. I mean, there are they they do have guards because it's not a it's not meant to be a public facility, but it it's not like a high security facility. You don't see guns on the guards, so pretty chill environment. So they I mean, the National Labs exist to do these big team science problems that bring together big groups. That's sort of their their legacy.
Travis:They do research in all of the scientific fields, biology, astrophysics, computer science, life science, climate science, earth science, and all that. Some national labs specialize in different things. So I was I was in what ended up becoming the, Earth and Environmental Sciences area of Lawrence Berkeley National Lab from, for my postdoctoral position. So my first position after I finished my PhD, onto being a research scientist there.
Shelby:And how how did sort of your day to day vary from your time at a national lab to where you are now in in, faculty position?
Travis:So well, it it varied even in in my time there. I mean, I started off as a post doc. And what I what really stands out to me from that time was both this just feeling of excitement, like of like, wow. I'm I'm an expert in something, and I'm actually getting to just lead my like, research in in something that I'm really excited about. So that was really cool.
Travis:All I did every day was research. Like, I, you know, analyzed data, ran simulations, wrote papers and stuff like that for the first couple years. Once they opened a research scientist position that I applied to and was accepted into, it started to change a bit. I started to, for example, supervise people for the first time. So which was a really strange experience.
Travis:I I I hired a post doc, and I was only 2 years out from my PhD. And this guy's, like, basically the same age as me. And I I my first conversation with him was, I don't know what I'm doing. Like, I guess we're working together. Like, let's figure this out.
Travis:Yeah. Yeah. And then as I went on, I got more into, like, science administration. So I, the Cascade project that I mentioned, that started in 2013. I helped write that proposal.
Travis:I was on the leadership team for that from the beginning, and got to the point where I co led the project at one point while I was there. And I'm actually again co leading it now. It's so it's in its 10th year now. But yeah. And now my day to day is just like all sorts of things all over the map, interacting with students.
Travis:Some of it's my own research, but that seems to be getting less and less in time as time time goes on. More advising people and guiding people in research. A lot of it's teaching and interacting with students. Increasingly now, it's, you know, faculty help run the university and increasingly getting into roles that actually help make that happen too.
Shelby:Yeah. And every now and then being kind enough to pop onto a podcast.
Travis:Yeah.
Shelby:So so during your sort of career as a student, as a researcher, and now as a faculty, were there any sort of moments that stood out to you as either very formative or things that felt very memorable for different reasons?
Travis:That's an interesting one. You know, one of the this is super random. I and I don't even know why this bubbled in into my mind, but that that summer where I was transitioning from my undergrad research into material science and then finally realizing that, you know, I I might be able to do grad school. I was living in on an alpaca ranch in the Santa Cruz mountains. I managed to get this find this really great deal where I got to live rent free in exchange for cleaning up alpaca poop every morning.
Travis:I had alpacas right outside my window. Like like, literally, like, my kitchen sink overlooked one of the one of the pens and, you know, the big floofy hair looking at the Yeah. Me in the morning as I'm making coffee. Like, that time was really exciting. And what what was really exciting about it too was this transition from feeling very uncertain and also kinda bad about myself, like, not knowing what what I was gonna do in the future.
Travis:I had just learned I'd been rejected from my first round of applying to grad school. I'd applied to, grad school in physics. I had no idea, by the way, at that point what was involved in that. I mean, my brother and I were the first in our family to go to grad school. So, fortunately, I got advice from him after the fact.
Travis:Should have actually thought to ask him Yeah. Because he he'd, gone into a master's program. The the application process was totally different. It was all about who you know and making connections. And I fortunately just made that really fortuitous connection with the person who ended up being my PhD adviser.
Travis:So it was this really cool time of, like, oh, I can do this. And and then when I started doing research in climate, it just clicking and going, this I can do for another 6 months. This I can imagine doing indefinitely. And that that feeling has never changed. Like, that's that's really cool to have found the thing that I just love to do day in and day out.
Shelby:Yeah. I mean, that's why I think we all stick with it because because we love it. I had some, I'd say, unique or interesting living situations as a student. Mhmm. Nothing came close to living on an alpaca farm.
Shelby:That sounds amazing. I was all in until you mentioned the alpaca poop, and then I maybe pulled back a little bit, but it sounds like all in all, that's a pretty fair trade.
Travis:It was a fair trade. It was a really cool experience.
Shelby:And so so sort of thinking about folks who may be listening to this, would you have any advice or words of wisdom for people who are, you know, maybe they hear this and think, oh, it sounds like a really cool field, something I'd be interested in checking out or pursuing or learning a little more about?
Travis:Yes. Absolutely. And this is something that I always repeat to students because I don't think it's widely known. If you want to go and get a graduate degree, like a master's or a PhD in a science related field in the United States, you can do it for free and actually get paid doing it. I didn't know that.
Travis:And when I was thinking about grad school, I was like, oh, man. I'm gonna have to take out some big loans, but I really wanna do this. And then come to find out that our federal government pays researchers to do research and pays for training of the next generation of researchers, covers your tuition, and you get paid. It's not a lot, but you get paid to do it. So that's incredible.
Travis:Like, that's a game changer. And I've I've in talking with students, that realization is often a big game changer. It's like, oh, I can do this. Like, financially, I can do this.
Shelby:Yeah. And even like you said, it's not a lot, but it is it is more than having to pay your own way. It makes such a big difference and gives you a lot more flexibility in how you sort of spend those years in graduate school. I was the same way. I had no idea that that was the case.
Shelby:Yeah. I was a first generation college student, and that was just such a pleasant surprise whenever that came to be known.
Travis:Yeah. Actually, since since we're on the record here and we're mentioning, 1st generation, my so my brother and I did we're we were the 1st to go to grad school. He's my older brother. I got my PhD first.
Shelby:Hey. Alright. We were we're gonna lead with that. Whenever this comes out, we'll be sure to advertise that widely and make sure that he listens. Thank you.
Shelby:Well, Travis, thank you for coming on. Yeah. We will sort of wrap up today with our yes, please segment.
Travis:Okay.
Shelby:This is an opportunity for both of us to spend a minute on a soapbox talking passionately about something that we're invested in at the moment. I can go first. You can go first.
Travis:I would love to hear what yeah.
Shelby:Okay. Please go. Okay. So what I'll ask is for you to time me with a verbal cue when we've got 30 seconds, 15, and 5 to go.
Travis:Okay. Will do. You ready? I'm ready. Set.
Travis:Go. Yes.
Shelby:Please, let's keep the focus on women's basketball. Mhmm. The WNBA season just recently ended at the time of this recording at least. Shout out to the New York Liberty for their championship run. Obviously, Caitlin Clark brought a lot of viewers to the field as she should have, but there are so many amazing players who have been there for years establishing the league.
Shelby:DeWanna Bonner, Aja Wilson, Stewie from New York. You name it. They've been there. Now I want that to carry over. Women's college basketball season is about to start.
Shelby:IU women have historically been a very, very strong team the whole time that I've been here, and it is so much fun to go to the games. And so I want to keep this energy alive and have more eyes on women's sports, especially women's basketball
Travis:Yes, please.
Shelby:As we get into this new season. Don't let that energy from the Caitlin Clark effect as great as it was, wane. I know she's not in college anymore, but there are so many great players that are in college, including at at Iowa. So please, folks, tune in and watch the next generation of women's basketball.
Travis:Wow. Perfectly timed. It's like you've done this a few times.
Shelby:I know. Right? Who would've thought? Okay. Are you ready?
Travis:Yes.
Shelby:So this is Travis O'Brien and his yes, please.
Travis:Yes, Please. Brandon Sanderson. Keep him coming. I have been invested in the entire Cosmere series from Brandon Sanderson, fantasy writer. This guy is prolific.
Travis:I love it. The it's intimidating. If you see one of these books, these books are like 3 inches thick. You're like, okay, I am committed. The series that I've been really focused on, the Stormlight Archive, his new book is coming out in December.
Travis:I'm so excited about it. It's just what's really cool about his writing and his fantasy world is 30 seconds? He's so creative. So this book, the star Stormlight Archive, starts on a world where basically everything's a crustacean of some sort. And and actually the premise for the book is, you know, the heroes of the world, they've been holding up the world, you know, sacrificing themselves Interesting.
Travis:And they decide to walk away and not tell humanity. They tell humanity, no, we've won we've won the war, but they didn't. And so that's the starting point for the book. Huge book. Really excited.
Travis:I recommend you give it a read.
Shelby:That sounds incredible. Are you gonna gift yourself the next book for Christmas?
Travis:Oh, yeah.
Shelby:Yeah. I think that's a perfect time for that to come out. Well, Travis, thank you again for joining us. This has been great to have you on. I think that you did the atmospheric science side very proud. You've got big shoes for the rest of them to fill. And for folks listening, hope we'll have you back next week when there'll be a new person that we'll get to learn about over drinks. We'll see you then.
Shelby:Earth on the Rocks is produced by Cari Metz with artwork provided by Connor Leimgruber with technical recording managed by Kate Crum and Betsy Leija. Funding for this podcast was provided by the National Science Foundation Grant, EAR dash 2422824.
