Hey folks, it's Quinn. Since we recorded this conversation, we've renamed the podcast. It used to be called important non-important, which is what you'll hear in the episode, and it's still the name of the company. But now the podcast is called, the most important question.
That is, what can I do? It's the question we get the most from all of you, and it turns out it's the question I've really been asking guests for almost 200 episodes now. So here we are, the most important question. Please enjoy this old conversation.
Thanks for listening, and thanks for giving us shit. Welcome to important non-important, my name is Quinn Emmett, and I'm Brian Colberg Kennedy, I'm back, he's back! And this is episode 16 with Dr. Kate Marvel.
We discuss the science and future of climate modeling. Dr. Marvel is a climate scientist and a writer. She's a theoretical physicist by training, like Brian.
She's now an associate research scientist at NASA's Goddard Institute for Space Studies in Columbia University's Department of Applied Physics and Applied Mathematics. Bam! Her research focuses on how human activities affect the climate and what we can expect in the future, using satellite observations, computer models, and basic physics to study the human impact on the experience from rainfall patterns to cloud cover. So yeah, she knows everything.
She knows every little thing you're doing. Every little thing you're doing. She puts into her computer and the answer comes out, we're fucked. That's your fault, Jeff.
Anyway, she was super rad, right? That was a lot of fun. Yeah, really awesome. Tiny preview, clouds, more controversial than you thought.
Not just confusing and beautiful, but as with most beautiful things, very controversial. Yeah, that was a very surprising part. Yeah, that was great. We'll get to it.
Quinn, welcome back from vacation. Thank you, I took a nice little trip last week, and I gotta say two things that are kind of related. I was in the British Virgin Islands, sounds super nice, it was really nice, very lucky to do it. You know, they got pretty fucked up last year.
Two category five hurricanes in six days. That's not great, Bob. Were you on land often? Often, yeah.
The US Virgin Islands and the British Virgin Islands, they both took it pretty bad. Obviously the stuff on the east side got it worse, but just unbelievable. Now it's eight months later, and just crazy. You can still see the beatings they took.
These islands, just natural resources just gone. You know, most of these places don't have their own water supplies, so water desalination plants if they're island could afford to have one shut down for a while. You know, obviously Puerto Rico is just still a nightmare. I think they just lost power again.
Again, yeah. But hey, you know what, the moral is, and what we try to take from in why we went is most of these islands and places like this, their economies are like 90% based on tourism. So go there. Go there.
If you've got somewhere you're going to go this year and you're going to spend the money, go there in some way. Every dinner you'd eat out in a stupid t-shirt you buy and snorkeling gear you get, it goes back and economy helps these people rebuild. So I really enjoyed it. The people were wonderful and they were thankful and still just an incredible place to go.
And the wildlife and the nature is amazing. So an also vacation. I mean, thank God, Jesus. Yeah, you're like the person in my life who the most deserves a vacation.
So I'm glad that you got to start. Kate Marvold's good use one. Oh yeah, she's good too. Also related on the flight back, I watched Geo Storm.
Oh, that's where you watch it online. It's amazing. Yeah, right. In the middle of terrible turbulence.
Yeah, okay. So good choice. I'm already terrified of flying. This movie, it's perfect.
It's perfect. I'm in the business of writing movies and I'm not going to make any comments on anything except to say it's perfect. It's in the cast. Andy Garcia, president, Ed Helms, Ed Harris.
Okay, okay. I'm very surprised. Ed Harris, Vice President of the United States. I mean, what a coup, right?
There's some twists. But I got to tell you, the best part of the whole thing was Zazzy beats from Atlanta. Have you seen Atlanta? Of course I have.
Donald, she's amazing. Zazzy beats. She might be from Atlanta. I mean, she's from the T.B.
Oh, yeah, yeah. She's amazing. Related, timing wise, it all comes together. I think she's domino in Deadpool 2, which is awesome.
Oh my God, I can't wait. Which comes out soon. And we talked to Dr. Marvel this week.
It's all... Unbelievable. Marvel, you're welcome. Oh, God, it's so great.
Oh, God, it's so great. Deadpool 2, Holy Cow, can't wait to throw that in there. Yeah, going to see Avengers next week. Don't know, I think it's like four hours.
Oh, good. And everybody gets one scene. I just hope some people find the guy. That's how they can afford to pay all those actors I imagine.
Yeah. Hey, what's going on with our website? Yes, pretty great. Oh, man, we checked some stuff off of this finally.
So it feels like we're at important, not important 3.0, right? This started off as a little newsletter. No website or anything. Then we got a website, but it didn't do anything.
There was basically nothing on it. So sorry. And then we started the podcast in January. So we got anyway to ramp up for a little bit.
Now what do we got now? We're almost two years into the newsletter. Which isn't saying. Crazy.
Coming up on 100 issues. And now every one of those is archived full text on the website. You can find them by month or by week. They've even got tags, Brian.
Tags? So you can search by particular issue or a thing you're into. Everything is becoming extremely searchable. You click on one tag.
You're welcome. It took a lot of hours. Click on one of them and I'll show you all the other interrelated things through our short history. It makes it very easy.
And now how many episodes do we have to? I think we just made 16 years. Is it 16? Talk tomorrow.
This is 16 today. So we got 16 now there's a page for each of those. So right instead, just a stupid link. We own the page with a full text transcript.
Usually up three to four or five days afterwards. That is really cool. Oh, it's the accurate. Sure.
They do a pretty good job. 90%. They're all the things we talk about and some jokes. And now everything is searchable.
There's a search box too. It's crazy. You don't have to do tags or dig through one by one. You go in there and type in Dr.
Kate Marvel. You're going to find everything we've ever done about her. Which just again sounds creepy. So creepy.
It's not meant to be that way. Anyways, if you're new to all this or you just want to take a deeper dive because you like it, I'm sorry. It's all there for you. Right?
When we have time to update it because... One and a half man show baby. One and a half man show. We're trying.
But it's up there. And finally what happened last night Brian? Our store went live. That's right.
We're gonna do it. We'll do another one. But check it out now if you haven't already. Important.unimportant.com.
Slash store. What do we got Brian? For our listeners? We got t-shirts.
Not just t-shirts but t-shirts. T-shirts. And there's, how soft are they? I have, I think I own like eight now.
They're so soft. What would you compare them to? I only want to wear them. What about a lamb?
Like maybe the skin of a newborn baby or newborn baby lamb? Yeah, yeah. I want a lamb involved. But like the chosen one.
Right. Not just any lamb. Everyone's like lams are soft and they're like, but this fucking guy. That's not true.
That's how it's coming baby. Alright. Anyways, lots of awesome stuff. We're excited.
Thank you for everything Quinn. Let's do, thank you Brian. Let's do episode 16. Dr.
Marvell. Dr. Marvell. Alright.
Our guest today is Dr. Kate Marvell. And together we're going to discuss modeling climate change in 2018 and beyond. What's it all mean?
Dr. Marvell, welcome. Hi. Thank you so much for having me.
So Dr. Tell us real quick who you are and what you do. Great. So I am Kate Marvell.
I am a associate research scientist at the NASA Goddard Institute for Space Studies and Columbia University. I trained as a physicist and I actually did astrophysics before I realized that really this is the only good planet. And I want to know more about this one. So I am a climate scientist right now.
I work with climate models and sort of simulation data. I also work with a lot of satellite observations. Wow. Super cool.
That's a lot. So let's set up our conversation for today. We are big believers in action oriented questions. We want to get to the bottom of today's topic.
So you really get it. And then because these times call for action, form like some specific steps that everyone here can take to make a world in our universe. Sounds good. Yeah.
Okay. So we're going to start with one important question to really get to the heart of why you are here today. And instead of it being, you know, tell us your life story or something, we like to ask, why are you vital to the survival of a species? Oh, that very low stakes there.
We're like, you use it into it. I mean, look, I am a tiny little cog and what I think is the greatest thing that humans have ever done, which is science. I think it's just incredible that we know how the planet works. And you know, that means that we know what we're doing to the planet and we have an idea for what to expect in the future.
So I can't believe they let me do this job. It's so fun. And it's also so incredible. It's so interesting.
And then at times it's depressing because the science is really super clear on what carbon dioxide is doing. We know that carbon dioxide is greenhouse gas. We know we're emitting a lot of it and we know the planet is warming up. And so, you know, I think if I am important, which I don't believe I am, that is my importance as being allowed to be, you know, a little part of this giant, wonderful enterprise that we call science.
That's awesome. I totally dig that. It is incredible. We could be dear or possums and not really have any idea what's going on.
But instead we do. And like you said, it comes with just a bit of a burden these days, which is knowing the mess we've made and what it's going to require to try to dial it back. Totally. So, all right.
So as we kind of get into the main topic, what we want to do is establish some context for folks for today's question. So this is your friendly reminder. Context 101 with Professor Brian is terribly oversimplified, often completely off-winded. Of course.
And sometimes they're never really intentionally a bit wrong. That's why we've got a resident expert here, Dr. Marvel, to correct us. So Brian, tell us about the history climate model.
Absolutely, no problem. Apparently, apparently is not a good start. We're talking about science. You're like human to 2002 Wikipedia.
Super reliable, Brian. I've read some stuff. Okay. Apparently it goes back to 1922, okay, an English mathematician and meteorologist started climate modeling with what it is today, like building equations using various inputs to predict web conditions over a period of time using numerical methods for the first time, building on some earlier ideas from Norwegian meteorologist and a Swedish chemist.
And then in 1938, this guy named Guy Callender, pretty big deal. Stop. His name was Guy Callender? Yeah, Guy Callender.
Okay. Spelled a little bit differently. Okay. I love the first name, Guy.
So Guy used the model to show that rising CO2 levels were warming the atmosphere, although that paper was pretty much ignored until the 60s. Perfect. In 1950, some meteorologists used the electronic numerical integrator and computer, like those giant member hidden figures and a huge, massive, enormous. What are we going to get to get to get in figures cinematic universe?
Like I'm done with the Avengers, you know? Oh, yeah, just a whole, just be cool. Spinoffs. So, so they use that in a 52 to forecast the weather like for the first time.
But it like took way too long. It was not very accurate. So baby steps. So then you know, time goes on bigger and more powerful computers are built.
Forecasts are more on point. All these groups are formed around the world to better predict everything. And you know, yeah, yeah, yeah, here we are right now. Wow.
That was exceptional. That might be your best one yet. Thank you. Okay.
And then finally, like who's paying for all this, right? And here's where it gets real good. I read an article about Guy named Steve, who's a professor of computer science at Sirius Institute. Steve.
Steve. Yeah, just Steve. He's a professor. He's a scientist at climate modeling like a climate model can cost like hundreds of millions of dollars, like if not billions.
Wow. All right. Well, I'm sure Dr. Marvel will enlighten us on the actual numbers.
Dr. Marvadino, Steve. Actually, why not Steve? Maybe I know Steve.
I don't know. As you like scrolls through a rifle and Steve Steve. Where's Steve? Right.
Is he like Prince? Did he just go by one name? I think he just goes by. Yeah, it's probably just a symbol.
No. Professor from a claim. It's just a symbol for a client. It's just a ring flag.
Okay. So now we're here. We're at today. So, this administration's first two budgets have proposed big cuts to various science and planetary first though by Congress.
Right, even when raises sometimes. Yeah, sort of. Awesome. So we'll talk more about that later, but you know, it's complicated.
Okay, so with that incredible history lesson for some context, let's focus on our topic of the week, climate modeling in 2018. Where are we, where are we headed and how accurate are those predictions? We want to start with one basic question, which might seem incredibly basic actually, but I'm only one coffee deep in, well, that's where we like to start. Dr.
M, what's the difference between climate and weather? Oh, that's a really good question. I get that a lot. So there's a guy at, I believe the University of Georgia, Marshall Shepherd, and he says, weather is your mood and climate is your personality.
Oh, that's good. Isn't that good? Like I wish I'd come up with that. Wait, that's got to go on a t-shirt for sure.
Right. And so, you know, a lot of times people ask, you know, how is it that you can predict climate in 100 years if you can't predict weather, you know, next month or next week? And that's because we're talking about two totally different things here. You know, I cannot tell you what the weather is going to be like next January in New York.
You know, January 22nd, 2019, I have no idea what the temperature is going to be. But I do know that it's likely to be cold. It's likely to be colder than, you know, New York and June of the same year. And that's because I understand the climate of New York.
I understand, you know, what shapes the sort of long term climate conditions here. And so, you know, that analogy, climate is your personality of whether it's your mood. You know, you can feel angry, you can feel hangry, you can feel, you know, upset. But that doesn't mean you're not fundamentally a very positive, nice person.
So when we talk about climate, we're talking about kind of long term statistics. Got it. So talk to me a little bit about, I guess, again, just a further clarify for folks. And again, this is so they can arm themselves in their own conversations with crazy people who make that argument.
Talk to me about the difference. What are the typical inputs that go into weather modeling versus climate modeling? I guess what really sets the science apart? Well, I mean, weather modeling is necessarily short term.
So you're basically tracking individual, like masses of air and water around in the atmosphere. And, you know, we've got amazing capabilities to do that. You know, we can give you forecasts, you know, with some accuracy about five days out. If you try to do a weather forecast for, you know, 20 days from now, that's going to be fairly useless.
Because the conditions that are affecting the weather now aren't necessarily going to be the conditions that are affecting the weather in 20 days. When we're talking about the climate, we're basically talking about, you know, these really large scale inputs and outputs. So we know that the Earth receives all of its energy from the Sun. And we know that energy is conserved.
And so basically the Earth heats up and re-radiates that energy into space as we call infrared radiation. So basically thermal imaging. And the Earth's climate on these really large scales is basically the balance between that incoming solar radiation and that outgoing infrared radiation. So, you know, that's fairly useless.
If you want to know, you know, should I pack an umbrella in 20 days? But it's really, really useful if you want to know are humans having an impact on the Earth's climate as a whole? Got it. OK.
That's good. That was good for me also. Yeah. Like, you know, that's all right.
So just for more clarification, I guess for listeners, because I feel like this is like really a topic where all folks see headlines and never the context. There are all kinds of different climate models. Yes, like not just one God model that everyone points to. They're about 25 to 30 modeling groups around the world.
Oh, so talk to us a little bit about is there redundancy there? Is it just difference in geography or politics or funding or do they all have different aims? I imagine it's pretty complicated. But if you could just again talk us through it all, it's just going to help people really understand, you know, what is driving these headlines that are saying?
Actually, it's worse than we thought it was, et cetera, et cetera, to really understand the machine behind it. I mean, that's that's a fascinating question. And, you know, the answer is kind of all of the above. So a climate model is just physics on a computer, right?
So I have a lot of people ask me, like, can I see your climate model? And I think it's like a diorama or something. Right? God, that'd be awesome.
Right? I kind of want one, though. But, you know, basically, what a climate model is. We just we take everything we know about how air and water work and land.
And then we put those equations on a computer. So you can come up with a really simple climate model. You can just write down on a piece of paper where you have energy coming in from the sun and then be being re-radiated from the earth. That is a climate model.
And you can actually write that down and you can solve those equations. To be clear, you can. I mean, I believe that anybody can solve those equations. This is right.
I can figure it out. OK. I believe in you guys. But there's sort of limited usefulness to that climate model.
You can learn a lot about the energy balance of the atmosphere. But if you want to actually say, like, how is that going to change precipitation patterns, you need to go a little bit more sophisticated. And so you can write down equations that describe how air currents and oceanic currents interact with each other and how the land surface interacts. And there are a lot of things that we can sort of explicitly solve.
But then we run into a problem of resolution. So climate models have to simulate basically the entire globe because everything is interconnected. What happens in the Arctic doesn't necessarily stay in the Arctic because that can have ramifications for the jet stream, which have ramifications for our weather, which may have ramifications for our weather all over the planet. So because everything is so connected, generally, in order to understand the climate, you have to simulate the entire Earth.
But what we're interested in is a lot of processes that affect the Earth's climate that happen on really small scales. So an example is cloud formation. Clouds cover. You look at those views of the Earth from space.
And you see lots of clouds. Clouds cover a significant proportion of the Earth's surface. But if you think about what cloud formation actually is, it's basically you can put like a tiny little sand grain or dust grain in the atmosphere. And then water droplets and ice crystals kind of coagulate around that tiny grain.
And so in order to be really accurate, you would need to simulate every single dust grain in the entire atmosphere. And we don't have a computer that's powerful enough to do that. And so what that means is that we have to approximate those processes. We have to kind of make choices.
We have to simulate them by empirical relationships as opposed to explicitly simulating those really small scale processes. And those empirical relations are, you know, you can make different physically defensible choices for how you model these small scale processes on a really large scale. And so different climate modeling groups make different choices for those small scale processes. I mean, this is as dumb down as it gets.
But essentially, everyone is considering everything. They're just, it comes down to the choices they make on degrees, almost, of what they include and how much to include. Is that? Yeah.
And choices for how we approximate these processes that we know that we can't resolve in a climate model, because it's two cores. You know, so, you know, an example to think about is like, let's say you have a country. It's too laborious. It's too hard to go through and count the votes of everybody in the entire country.
Because let's say there's like 300 million people. And so what you do is you subdivide the country into, say, 50 units. And you make approximations. So you say, OK, this unit to the West, I'm going to color that blue.
And I'm going to approximate the votes of everybody who lives in that unit by coloring that blue. And then, you know, this place in the South, I'm going to approximate the votes of everybody who lives in that place by coloring that red. And so instead of 300 million things to keep track of, now you've got, you know, basically 50 things to keep track of. More often than not, that sort of corsing process that, you know, looking at 50 instead of 300 million leads you to accurate results.
But sometimes you get a disconnect where, you know, the fact that you haven't counted all 300 million or whatever. And you use this 50 instead. It gives you something that that looks a little weird, if that makes sense. Right.
But seems pretty necessary. Because like you said, what's the other option? Count 300 million. Not going to happen.
Exactly. Yeah. That makes sense. Talk to me a little bit about how far we've come.
You know, in one of the most important things that we've learned, not necessarily about climate, but more specifically about modeling in the past 20 years, you know, what's changed since you came on to the scene with your ability to do your job. That's a good question because we've learned a lot. But by some metrics, we haven't learned very much. So I'm really interested in this quantity called the equilibrium climate sensitivity, which is kind of a fancy word for how hot is it going to get?
Right. And you know, the number one reason that we don't know how hot it's going to get is because we don't know what people are going to do. You know, we don't know. Are we going to get serious about cutting emissions?
Right. We're just going to be business as usual. Are we just going to be like whatever and like set fire to oil barrels because why not? No, it's a bit.
Right. Totally. Like who knows what we're going to do. But even if you remove all that uncertainty associated with human behavior, we still aren't 100% sure how warm it's going to get.
And that's because there's a lot of physical processes that we still don't understand. So if you looked at, you know, kind of three model generations ago, the predicted range, you know, what models we're saying. Basically, they're saying in response to a doubling of atmospheric CO2, it'll get warmer. And that warming will range from about a degree and a half Celsius to four degrees Celsius.
And if you look at the most recent IPCC report, it says, we think that it'll get it'll get hotter and that'll be between a degree and a half and four degrees Celsius. So that range kind of hasn't narrowed, even as the models have gotten more sophisticated and better. And that's been really frustrating for a really long time. But now I think we're kind of starting to realize exactly why it is that models disagree on how warm it's going to get.
And it turns out that models disagree on how warm it's going to get because models disagree on what clouds are going to do. Interesting. Fucking clouds. So talk to explain that.
Tell me why clouds are the problem. So clouds are what we call a feedback process. So for me, the most intuitive way to understand a climate feedback is to understand that when the Earth gets warmer, stuff changes. And that stuff that changes can sort of feedback onto that warming process and either speed it up or slow it down.
So a really good example is ice melting. Ice is super reflective. And so the fact that both poles are largely covered in ice means that they're reflecting a lot of sunlight back to the air or back into space. So when you park your car on a really hot day and you put those reflective screens on your car on the windshield.
Mine's the inside of the Millennium Falcon. Awesome. Continue. The ice is the inside of the Millennium Falcon for the climate.
So it's reflecting a lot of sunlight back to space that would otherwise reach the surface and warm it up. But as it gets warmer, you've got less ice. So the ice goes away. And so instead of having this really effective, reflective shield, you've got dark land or dark water underneath.
And that absorbs that radiation from the sun, as opposed to reflecting it. And so that is a feedback process that accelerates the warming. It makes it faster. It makes it worse.
And scientists call this a positive feedback. And I think that is the worst term ever. Because normal people hear positive feedback and they're like, oh, I'm doing a great job. Sure, of course.
And in this context, positive feedback means a thing that will probably kill us all. So it's the opposite. It's going to be a better word for it. So I like destabilizing feedback.
The nightmare scenario. Bad thing. So here's my question again. Total neophyte below that, whatever the word is for it.
I mean, we were so able to understand the process of the cloud feedback loop. But you said our single biggest hindrance seems to be we don't know what clouds are going to do. Where's the disconnect there? Right.
So we know ice is an example of a positive feedback that we understand really well. Because we know that when it gets warm, ice melts. Sure. You know, that's something that science has figured out.
Thank you. Right? Good job, science. But when you look at all of the different processes that we think could constitute these feedback loops in the future.
And we understand all of them pretty well, except for clouds. So ice is an example of a feedback process that we understand really well. Clouds are an example of a feedback process where we're like, oh, I don't know. And that is because clouds are incredibly complicated.
They play this really dual role in the climate system. So low clouds, you know, the cloud, what we think of as a cloudy day, right? That's characterized by the fact that we don't really see the sun on a cloudy day. Because those clouds are blocking sunlight that would otherwise reach our surface.
So those clouds are acting to make the planet cooler than it would be otherwise. But you know, there's like, you know what a serious cloud is? It's kind of like thin, wispy clouds that you get on sunny days. I did not think there was going to be a test today.
I just gave the answer. By the way, you're my fighter a lot of times like, what cloud is that? I'm like, oh, man, there's cumulus. There's serious, serious.
And then there's some other great ones. Right. Like, I actually don't really like clouds because I'm from California. And I hate weather.
But I like super important to my job. So I like to study them. I mean, we're in California. And it turns out we haven't had clouds for a little while now.
And everything is on fire. So more clouds would be better. Totally. But anyway, those high clouds that you see on a sunny day, those are made of water vapor.
And water vapor is a greenhouse gas. So those clouds are warming it up. So clouds kind of play these dual roles. Like they block sunlight, but they also trap heat from the planet.
And like you pointed out, there's like a bajillion different kinds of clouds. And all of these different kinds of clouds have different effects on the planet's climate. And so how are all of these different clouds going to change in the future? We don't really know because cloud formation is really hard to simulate in climate models.
Because that's something that happens on a really small scale. So it's this like vicious cycle of this thing that's really, really important, that could go either way, and is really, really difficult to simulate in a climate model. Is there any hope of cracking clouds? Like, so is it a technical?
All right. So let me look at it this way. So not to get super nerdy, but GPUs, right? Graphical processing units.
It was always about CPUs, right? That's what drove Moore's law for the past 25 years and made you have an iPhone. How did we get a computer in our pocket? Right?
But now GPUs have come along almost accidentally from companies like NVIDIA. And they were for some people who do Photoshop and the computers and play games. But suddenly, it's made other visionary and graphical things like self-driving cars and virtual reality. A reality day.
Shut up, Brian. So is it a technical challenge or is it we don't have enough data? You know, we've talked a lot on here about big data and algorithms and when it comes to infectious disease or something like that. Is it something that's missing?
Kind of all of the above. Great. So we are, you know, that's going to be my answer for everything. It's like, we need everything.
But you know, I think we're getting better. You know, we have more power computers now than we did in the last sort of generation of climate models. And so we can actually, you know, simulate cloud formation in a lot of climate models on some scales. So we're getting better at modeling clouds.
We're also getting better at observing clouds for a really long time in order to figure out what clouds we're doing on kind of long time scales. We had to stitch together a bunch of different weather observations. And so as a result, like these long term cloud data sets are kind of like patched together from like one satellite here, one satellite there, then that satellite by a lot of the sky. And that was nothing for a little bit.
And so, you know, it's it's janky and we know that it's janky. But you know, there's there's now kind of more dedicated cloud data sets that are giving us better data. And so we have better observations. We have sort of more powerful models.
And the hope is that that's going to allow us to really get a handle on what clouds are doing now and what they're going to do in the future. I will say that we've been looking really hard for evidence that kind of clouds are going to save us because that would be awesome. Oh, okay. You know, somehow like the cloud feedback was stabilizing.
You know, clouds we're going to change in such a way to kind of slow down the warming and like maybe buy us a little bit more time. And we have not found any evidence that that is a thing that that's going to happen. That kind of sucks because it's like nature is not going to save us from ourselves. And we don't deserve that really.
You know, I mean, and like maybe, you know, if that did happen, it would just be like cloudy all day and like this is this is what global warming was. It's just going to make us all like have seasonal effective disorder all the time. But it looks like that's not going to happen. And so, you know, kind of it is probably up to us.
She says with a question mark the end of the sense. Is there sorry? So that probably won't happen. Is there something else or other things that you that you are looking for to that seem like they, you know, might be coming down the road the next two years or five years or 10 years?
In terms of like the climate saving us or in terms of like. Oh, God, no, that's not going to happen. No, no, no, no, no. I would say more.
Yeah. Your ability to particularly end game. Basically, no, what's going to change in your job and your profession? What is and this is always a stupid question, you know, what is predictably going to change besides the advance of, I guess, the technical quick capabilities?
You know, what are you guys looking forward to that is probably most evidently going to come to pass. Yeah. So like I'm going to get super nerdy about this. You can tell me about it.
No, let's do it. Never. Never. We call it Earth System Models.
So I feel like every time you talk about climate change, like some, some dude and like, I'm sorry, it's usually a dude. It's always, of course it is. We're the worst. Yeah.
We'll come along and be like, well, like plants love carbon dioxide, you know, you put more carbon dioxide in there and like plants are going to thrive. And so like this is a great thing. Do people say that to you? Yeah.
All the time. No, you didn't. And it's like, it's, the things, it's true. Like plants do thrive under increased CO2.
Like you put more CO2 in the atmosphere and all of the things being equal, you get more plants. Something else that's really good is you put more CO2 in the atmosphere and plants get much more efficient at using water because they don't need to open their stomata quite as much in order to sort of transfer it. And so you put more CO2 in the atmosphere, you get more plants and those plants are more efficient at how they use water. And so that sounds great.
But on the other hand, it's really hard to open or close your stomata when you're on fire. And so it's a really fascinating question, you know, and it has implications for things like drought. What's going to happen to soil moisture? What do we expect in terms of drought in the future?
And climate models are getting much more sophisticated actually, including these effects. So what happens to plants in a future climate? How do we simulate that? The answers that we're starting to get from these much more complex models are like, it's a little bit more complicated than just like plants love CO2 and everything's trying to find.
It's so weird. Yeah, right. So I think that's amazing. I'm really excited about our enhanced ability to kind of understand how the biological systems that live on the earth are responding to increased carbon dioxide in the atmosphere.
And then to revisit the topic of feedback, like how those changes feed back on the warming. Groovy, got it. All right, so you're very exciting. Earth systems modeling.
You're doing a climate model. You get an output. What's the most frequent use case? What's the most frequent use?
So you're not just doing it. So when you run it, you run a model and you get an output. What is a typical practical application of the data you're getting out of that? Who are the models customers per se?
How's it being put to use? So you run a climate model and you get, you don't just get temperature. Like you get an incredible amount of data when you run one of these models. So you get temperature all over the planet.
You get temperature in the troposphere, temperature in the stratosphere. You get humidity. You get precipitation. You get cloud cover.
You get things like leaf area index and snow cover. You get an insane amount of data when you run one of these models. And all of the modeling groups in the world kind of come together and do these coordinated experiments with their models where everybody agrees to do an experiment where you abruptly quadruple carbon dioxide in the atmosphere and then you see what happens. And then every single modeling group agrees to do an experiment where you increase carbon dioxide at 1% per year and you see what happens.
And because there's this coordinated project to sort of make all of the models run under the same conditions, then you can really kind of get a handle on the uncertainty. You can get a handle on, well, why do different models give us different projections? Or why are all the models saying the same thing? This must be really robust.
Let's understand why this is happening. So there is no one output of a climate model. The new generation of models that's coming online pretty soon. That coordination project is going to give us about 50 petabytes of data, which is just like an incredible amount of stuff to play with.
Damn. So I guess then who is utilizing that data that's going to come out? 50 petabytes. Is it most of the governments?
Is it industry? How is that being used? You can use it if you want. Again, maybe not.
I mean, it's all freely available. So you can go to something called the Earth System Grid and download the output of all of these models. Brines MacBook Air just went up and forth. You know, the data is there for anybody to use.
Practically, it's mostly researchers using it. So it's people like me who are interested in what's a drought going to look like in the future? How hot is it going to get? But it's also people who might work in city planning and want to get an understanding of the uncertainties that they should take into account when they're building seawalls or designing insurance programs for coastal or low-lying areas.
So the fact that this data is publicly available means that it's not just for basic science. People are using it to try to influence decision makers and try to use this data to try to make the most informed decisions possible. Interesting. So to follow up on that, I guess, and this is moving towards our listeners becoming participants here.
How do they become more invested in these models and this publicly available data? You said it's out there. Brian is, I guess, that's just who he's going to be doing now. Yeah, I'm pretty busy with this.
So is there room for innovation here? Is there something that could come out and surprise us? Or I guess, let's say some young nerd goes on to the website and starts digging through this stuff. What does someone need to do to become your mentee?
The next Dr. Marvel? Would you even advise them to go into climate modeling? If not, then where else?
Where can nerds have the biggest effect? I mean, I have a lot of fun in my job. I would recommend being a climate scientist to anyone. But if somebody wanted to go into climate science, I would say you need a background in physics.
So you need a solid understanding of why stuff does what it does. And then you also need fairly solid programming skills just to sift through this insane amount of data. Something that people can do is help us shape the questions that we answer. So I'm really interested in drought because that is something that's very relevant for California.
That's a place that I love very much. And I think this has to be a dialogue. It has to be kind of two ways where we don't just pass on scientific wisdom from above. But we actually listen and we say, what are the questions that we might be missing?
What do different people care about? And really make sure that we're directing our research to answering questions that actually matter to people. I love that. It's back and forth and a group effort.
Yeah, no, that makes a lot of sense. So, all right. I can imagine, though, stepping back a little bit, a lot of this might seem pretty almost detached from our listeners' day-to-day life. And that's kind of what we've always tried to do is cut through all the bullshit every week and give people what's most important.
But at times, all anybody can get on their communities' headlines. We just try to give them the most important ones, the existential ones, the positive or the negative. So, how is this modeling affecting our listeners on a day-to-day basis, even if they don't know it? Like I'm Brian.
I'm just a guy doing some acting, really bad acting, by the way. I don't know if you've seen me. Little podcasting, some motorcycling. Terrible, stupid motorcycling.
Well, we agree to disagree. But the point is, in between all of these activities, I might think, I wonder what Dr. Marl discovered about the future of our climate today and how that affects me or my kids in the future. So, what is the day-to-day impact of what you're doing?
I mean, we have to use models because we don't have any observations of the future. No. Yeah, that's super convenient. So, this is really our best hope of understanding what it's going to look like and what we should prepare for.
And climate models allow us to explore counterfactual scenarios. Like, what would the Earth look like if nobody lived here? What would it look like if there were volcanoes where the only thing affecting the climate? What would it look like if we decided to do something insane, put up giant solar shields that blocked the sun?
Into it? Yeah. Let's not do that. Like, I'm not endorsing that.
Got it. Agreed at this point. Right. But I would say kind of two things.
Climate models allow us to sort of see what's coming and help us prepare for it. And some of these worst case scenarios that are popping up in the projections, we still have time to avoid those. A world that is four degrees warmer than pre-industrial is a really scary world with really terrible impacts. And we still have time to avoid that.
And I think models are kind of really telling us they're making it really explicit. This is what it's going to look like. These are what the consequences are. And I think your point that this doesn't feel really relevant is really well taken because we don't, basically humans don't think about climate change because we've never really had to think about climate.