Today, Explained - Why we still can't predict tornadoes

Episode Date: December 13, 2021

Most people get about eight minutes' advance warning of a tornado. This episode of Vox’s Unexplainable podcast explores how scientists need to confront more of these storms, head on. Transcript at vox.com/todayexplained Support Today, Explained by making a financial contribution to Vox! bit.ly/givepodcasts Learn more about your ad choices. Visit podcastchoices.com/adchoices

Get groceries delivered across the GTA from Real Canadian Superstore with PC Express. Shop online for super prices and super savings. Try it today and get up to $75 in PC Optimum Points. Visit Superstore.ca to get started. A series of deadly tornadoes ripped through the United States on Friday. One of them might have been the longest tornado in the history of the country, and most of that one ended up in Kentucky. The damage is unlike anything the state's ever seen.

Entire neighborhoods were destroyed. Thousands of homes are just gone. Some 60 people are dead at this point, and that number is expected to rise, and it'll take years to rebuild what was lost. It goes without saying that there wasn't enough advance warning for people, and with tornadoes, there never is. The average warning time for a tornado is about eight minutes. This past summer, our friends at the Unexplainable podcast tried to figure out why that is, why we can't predict tornadoes better than we do. It turns out, if we want to get better at it, we're going to need to chase a whole lot of tornadoes instead of letting them chase us.

Today, we're going to run that episode of theirs so you can better understand what we know and don't know about tornadoes. Here's Brian Resnick and Noam Hassenfeld from Unexplainable. Forecasters are just really bad at this. And when these warnings do come, they often don't precede a tornado. 70, 80% of warnings don't actually come before a tornado. They're false alarms. What? So you're saying people usually only have like seven minutes to get to safety. warnings don't actually come before a tornado. They're false alarms.

What? So you're saying people usually only have like seven minutes to get to safety, and then like three quarters of the time there's no tornado, even when we do get the warning? Yeah, this is where we are. How is that possible? How do we not understand this better? The heart of it is actually kind of simple. Okay. Tornadoes come from thunderstorms, but meteorologists, they can look at two thunderstorms. These thunderstorms will look identical, but one will produce a tornado and the other one won't,

and they really don't know what makes the difference. So what do scientists know about how tornadoes form? Yeah, I was wondering that too. I talked to Robin Tanamachi. She's a meteorologist at Purdue University in Indiana, and she knows everything there is to know about tornadoes. I teach mainly atmospheric science, radar meteorology, and severe storms meteorology. She's been fascinated by them her whole life. She told me this story of when she was a little girl, there was this tornado captured

on the news by the news helicopter. We're about three quarters of a mile from the actual touchdown at this time. This is really spectacular. And what did I see on the screen? But it was this gorgeous helical tornado, and it was live. I'm going to have to leave this area here. The debris is drifting my way here. Please do, Max. And I believe at this point the helicopter pilot is realizing, whoa, we've got a live tornado on our hands, and so they're maneuvering around it,

trying to keep a safe distance from it. Right now the tornado is located over a wooded area, and it is ripping entire pine trees into the air, thousands of feet. She remembers just, like, thinking, I can't get enough of that. She couldn't look away. My thought was just, what is it that's making this happen?

You know, what is it in the environment that's necessitating that this thing exists? And what makes it so powerful? I really wanted answers to those questions. I know you would ask me some questions here. Unfortunately, I've been watching the tornado instead of where I'm going. Yeah, lots of questions. What's Robin learned so far? I know you would ask me some questions here. Unfortunately, I've been watching the tornado instead of where I'm going. Yeah, lots of questions. What's Robin learned so far?

Well, obviously, she hasn't gotten to that holy grail of being able to predict and know everything about the formation of a single tornado. But on the bigger picture, she does know where they come from. Mainly, they come from here. They come from the U.S. There are definitely hot spots all over the world, but the United States by far gets the most of them. Why are there so many tornadoes in the U.S.? So like I said, tornadoes come from thunderstorms.

A thunderstorm you can think of as an organism that ingests and eats warm, moist air. And warm, moist air? We got that in spades. That comes from the Gulf of Mexico. And that's moving north from the Gulf. This tide of moist air that just keeps sloshing

northward from the Gulf of Mexico over the southern tier of states. And then it hits two dry streams. So one, we got dry air from the Rockies. And two, we've got dry air coming off the Mexican plateau. So that warm, moist air from the Rockies. And two, we've got dry air coming off the Mexican plateau. So that warm, moist air from the Gulf, it meets those other two streams and gets pulled up into them. But those two streams are moving in different directions.

So all that air begins to twist and condense and form clouds. It will start to rotate just on its own spontaneously. That air will be torqued and twisted. In my head, I'm envisioning, you know, like Looney Tunes, when the three cartoon characters all, like, run into a big brawl and it kind of, like, spirals into a cycling. Yeah, something like that. I'm imagining something more of, like, a wizard summoning these airstreams that converge into a cauldron and create these vortexes.

But we have different brains. How does Robin think about it? It's like a big barber pole in the middle where the air is rising, but it's also twisting as it goes up. This is the heart of a supercell. Supercell? Yeah, supercells are these mega thunderstorms

that spawn tornadoes. They rotate, so they're kind of like these mini hurricanes. And the big question is, how do you take this big kind of lumbering rotation in the sky from these supercells, these big, twisting thunderstorms, and those sort of give birth to tornadoes. So do scientists have any right ideas here? Like, what sets this process off? Well, before they even get to there, it's more like they don't see what sets it off. That process is probably happening on timescales of just a few seconds or less. All this happens so fast.

Weather radar, like our number one tool for observing storms, it just can't see it. Why can't radar see tornadoes? Well, one part of the problem is it basically picks up on moisture, rain. And you can infer things from rain, like wind speed, like you determine how fast that rain is moving. But it's like this incomplete picture of a storm. The other problem with radar is physically radar is this dish that spins. And so every time it spins, it takes an observation of the thing that it's observing, like a storm. But that takes a few moments for it to

spin. You can have a big lazy vortex in one scan, and by the time the radar swings around again, you've got a tornado. The effect is kind of like if you were watching a dancer under a strobe light. The dancer being the storm and the strobe light being the radar that can only capture it in certain moments. So the dancer starts dancing, she's spinning, and you get a slice of it. And then the dish rotates a bit, and it sees a bit. And then it misses a bit, and then it sees a bit, and then it misses a bit.

We're scanning it too slowly to see this very rapid process that's happening. Whatever causes a tornado to spawn, it often happens when that dancer is in the dark. You're saying that whatever happens to jumpstart this tornado, it can happen so fast that it could happen in the dark of a strobe light, like in that moment when the radar dish has its back turned? Yeah. Researchers are so in the dark, like they're not always sure if like the tornado forms from the storm cloud down or from the surface of the earth up, or if both things happen at the same time and then they connect

in this explosive way. It's just the basic idea here is what we can't see, we can't predict. A lot of times this process is either invisible or opaque. It's like a magician's sleight of hand trick. Again, imagine two thunderstorms that look identical and one produces a tornado. Well, the one that's producing a tornado is performing a trick that we just can't see. And that's really why tornado warnings are only sent out when a tornado is seemingly detected. Our current paradigm for warning for tornadoes is something called worn-on detection, meaning either it's indicated by radar or somebody in the field has phoned in and reported it. So the tornado has been detected, you issue a warning.

Wait, but I thought you said there were a lot of false alarms, like the majority of the time it's sent out, it's a false alarm. So how could it be a false alarm if they detect the tornado and then alert people? Oh, it's because often when a tornado is indicated by radar, it's not a real tornado. So it's a false detection. And it's a false detection for all the reasons why I said radar isn't that great. It just gets it wrong. It just can't see everything that's happening. And these false alarms, it's not just a meteorological problem. It's a psychological problem too. You know, like the boy who cried wolf syndrome. If you issue too many tornado warnings, people start to not trust them as much. We need to bring that false alarm rate down so that when

we do issue a tornado warning, we're extremely confident people are actually going to take action. So if scientists are already struggling to predict tornadoes, and then they're sort of training people to react to tornado alerts in exactly the wrong way with this boy-who-cried-wolf syndrome, I mean, how do you solve the tornado problem? So for one, Robin's really pushing for just better, faster radar that can scan areas more quickly, and maybe then it can pick up on more clues to just better see

what's happening in these storms and to see that moment of tornado formation. Like a faster strobe light. Yeah, a faster strobe light. But really what meteorologists want is better prediction, better artificial intelligence that can just look at a storm and model it in a computer and then press fast forward and see what will happen. If they can do that, then they can warn on forecast and say like 30 minutes from now, we've run the computer models and this storm is going to create a tornado and you should get out of the way.

Yeah, we'd love it if people could have us up to 30 minutes to an hour. I mean, that's where you make a completely different set of decisions. And I assume to get a better AI system, you just got to input tons and tons of tornadoes all over the country, right? Yeah. And a big challenge here is that tornadoes are so varied. They come in all these shapes and sizes. From very skinny funnels that are only, you know, a couple hundred feet wide all the way up to two and a half miles wide, I believe was the

widest one on record. They seem to be very sensitive to local environmental conditions. So what forms a tornado in one area might not form it in another. Maybe just like the presence of trees changes the equation. And tornadoes behave differently in different parts of the country. That huge spectrum makes studying tornadoes really cool, but it also makes studying them really hard. Yeah. And if you need to get a bunch more tornadoes to input into a new system to figure out how to predict tornadoes, how do you find those tornadoes if we don't know how to predict tornadoes? Yeah, yeah. That's a funny conundrum here. We need more data from the tornadoes themselves,

but we don't know when they're going to show up. We need to be prepared to show up when they do, because to get that data, we solve this in the most badass way possible. We have to go storm chasing. Storm chasing? Storm chasing. After the break. Support for Today Explained comes from Aura. Aura believes that sharing pictures is a great way to keep up with family. And Aura says it's never been easier thanks to their digital picture frames. They were named the number one digital photo frame by Wirecutter. Aura frames make it easy to share unlimited photos and videos directly from your phone to the frame. When you give an Aura frame as a gift,

you can personalize it, you can preload it with a thoughtful message, maybe your favorite photos. Our colleague Andrew tried an AuraFrame for himself. So setup was super simple. In my case, we were celebrating my grandmother's birthday. And she's very fortunate. She's got 10 grandkids. And so we wanted to surprise her with the AuraFrame. And because she's a little bit older,

it was just easier for us to source all the images together and have them uploaded to the frame itself. And because we're all connected over text message, it was just so easy to send a link to everybody. You can save on the perfect gift by visiting AuraFrames.com to get $35 off Aura's best-selling Carvermat frames with promo code EXPLAINED at checkout. That's A-U-R-A-Frames.com promo code EX at checkout that's a ura frames.com promo code explained this deal is exclusive to listeners and available just in time for the holidays terms and conditions do apply bet mgm authorized gaming partner of the nba has your back all season long from tip off to the final buzzer you're always taken care of with a sports book born in Vegas.

That's a feeling you can only get with bed MGM. And no matter your team, your favorite player or your style, there's something every NBA fan will love about bet MGM. Download the app today and discover why bed MGM is your basketball home for the season. Raise your game to the next level this year with BetMGM, a sportsbook worth a slam dunk,

an authorized gaming partner of the NBA. BetMGM.com for terms and conditions. Must be 19 years of age or older to wager. Ontario only. Please play responsibly. If you have any questions or concerns about your gambling or someone close to you, please contact Connex Ontario at 1-866-531-2600 to speak

to an advisor free of charge. BetMGM operates pursuant to an operating agreement with iGaming Ontario. It's Today Explained, and today we're running an episode of the Unexplainable podcast, also from Vox. It ran in their feed back in July, and it's one about the mysteries surrounding tornadoes. And one way to solve those mysteries is to storm chase. And you're about to hear from a guy who's very excited about chasing tornadoes. So excited he even took his kid chasing a tornado. I know that might be hard to believe considering this weekend's events, but this is a professional who knows what he's doing. And just for the record, he is talking about chasing storms back in July of this year. All right. With that being said, here we go back to Unexplainable.

You know, it's always interesting when the sirens come out, the normal public goes down to the basement and the scientists go to the roofs. So this is Jeff Weber. He's an atmospheric scientist. He stood on a few rooftops. He's chased a few storms. He actually chased one recently with his 12-year-old son. Admittedly, I planted the seed, but he is really into it. And after seeing his first tornado last week, he can't get enough. He wants to just keep doing it.

So what happened on this day? Well, we're in the house. Tornadoes were not forecast to develop anywhere near us for this day. But a tornado warning went off on my phone. As an atmospheric guy, I have about five weather apps, and my phone started blowing up with this tornado warning. I'm like, a tornado warning? That doesn't seem right.

So I tell my son to run out into the backyard and see if he can see it. We did take some video video and five seconds later, I hear him shrieking at the top of his voice. I see it! I see it! I do too! So we all run out, take pictures from the backyard and I'm like, hey buddy, you know, this is pretty cool. He goes, we got to go. We got to chase it. Let's go. And so we ran to the truck. And as we were driving over the top of the hill, we could see the whole layout of the storm. We could see the tornado from the top of the cloud deck all the way to the surface of the ground. We could see the hail shaft.

We could see the heavy rain. We could see the shelf cloud rotating above the tornado. Oh, my God. By the time we got up close to it, it was spectacular. We could see the rotation. We could see the dirt being dug up from the ground. It was just an incredible event. My heart's pumping just listening to this.

It is awe-inducing. It is truly awesome. My mouth is open as I'm staring at this thing spinning. It's amazing. It's unreal. What was it like to share that with your son? It was one of the highlights of my life. To be chasing a tornado and to see my son so excited about it.

I think I have a future atmospheric scientist on my hands. So to start simply here, why is chasing tornadoes so important? Why can't we study them more remotely? Well, the most important things for a tornado to form is temperature, humidity, wind direction, and speed. But we need them at the surface, we need them at the upper atmosphere, and we need them at levels throughout the entire atmosphere. And so to get that complete dataset all around the tornado at all the different levels is very difficult. Why don't we have this now? Is it just like our typical weather radar doesn't take on all this information? It's difficult to get this complete dataset because tornadoes don't stay on the ground for a long period of time, and we don't know in advance where they're going to show up.

So if you want to do a contrast with hurricanes, for example, they take days, if not weeks, to cross the Atlantic. And we have the luxury of that time to fly planes through the eyewall of that hurricane and get those measurements. With a tornado, we don't have that luxury of time. They might be on the ground for as little as a few minutes. And so to be able to marshal all of those instrumentations and planes or sensors, it's very difficult. So if we did have all the right equipment in place and at the right time, and we caught a perfect tornado, would that be enough? Heavens no. We need dozens, if not hundreds.

They're kind of like snowflakes. No two tornadoes are exactly alike. For example, of 100 tornadoes, you get 98 of them that spin clockwise, but you're going to get a couple of them that even spin counterclockwise. What? And we have no idea why. I think as we get greater technology, in particular with drones, either on the ground or in the air, it will give us a greater capacity to collect these datasets without putting humans in harm's way. And so I'm hopeful that over the next decade or two that we'll have more complete datasets so that not only can we look at one dataset of the tornado, why it formed, but we can look at nuances and subtleties and differences on why one tornado formed different than another one.

And that's where we can unravel the mystery. Why are you confident that it's just a problem of more data? Could it also be the case that these storms are just so chaotic, so sensitive to very slight variations in geography or place or conditions where they're just very hard to predict and we might not ever get much better at this problem? Yeah, I appreciate that perspective. I'm a firm believer, though, that science can solve a lot of these problems if they have the proper data sets to analyze it.

We know what we're dealing with. And if we can close the box, so to speak, around the tornado, we should be able to explain what's going on. We have a very good understanding of thermodynamics and instability, but we just don't know quite how they're interacting with each other at these finite scales that a tornado is operating in. You're saying that our atmosphere is knowable, like things that we can measure, we just need finer and finer grained resolution. Exactly. There's no new physics being brought about by a tornado.

The atmosphere is very well understood. The laws of physics are very well understood. We just need to see at these smaller scales how they interact to create the tornado. I love how a lot of this boils down to, well, to better understand tornadoes, we need to keep chasing them, right? Is that what I'm hearing here? Most definitely. Just photographs from the public can be incredibly helpful because we're still at that point

where photographs can gain insight into the structure of the tornado. I don't know if we should be telling people on a podcast to take photos of tornadoes, right? Only if you're trained and know what you're doing. They are very dangerous, but if you approach them from the backside, that's the smart approach. They should be going away from you, I imagine. You should be chasing the tornado, not having the tornado chase you. This episode was produced and edited by Brian Resnick, Noam Hassenfeld, and me, Meredith Hodnot, with extra help from Bird Pinkerton. Mandy Nguyen checked the facts, Christian Nayala sound designed the tornadoes, and Noam scored the episode with some help from 3,000 Rivers.

Lauren Katz heads up our newsletter, and Liz Kelly Nelson is the VP of Vox Audio. You can sign up for our newsletter at vox.com slash unexplainable and email any thoughts you might have about the show to unexplainable at vox.com. Unexplainable is part of the Vox Media Podcast Network, and we'll see you next Wednesday. Thank you.