Please Don’t Try to Shoot Diaper Gel Into a Tornado
A tornado expert explains the science behind Twisters, from diaper gel to anchor drills and why the film didn’t mention climate change.
Twisters is a more solemn movie than its predecessor, which you can tell because there are no flying cows. Its heroes’ goals reflect this: They don’t just want to study the formation of tornadoes but potentially destroy them, using a combination of hyperabsorbent polymers and silver iodide.
Is anything like this even remotely plausible? To find out, I caught up with Dr. Jana Houser of Ohio State University, who studies tornadoes and how they form. Houser is also a storm chaser, and she played a tangential role in the production of Twisters, tagging along with cinematographer Sean Casey as he shot footage of real storms for the film’s green-screen effects and B-roll. She was patient enough to answer all my questions about the real science underlining the movie’s plot, including whether or not it’s safe to shoot chemical compounds into a tornado. Please don’t email her with your ideas!
What was your review of Twisters, both as a scientist and as a moviegoer?
It was an exciting movie. I thought the plot was pretty good, there were a lot of good action sequences. The thing I was kind of disappointed about — and I’m pretty sure they tried, but it didn’t work out — was I was really hoping that Kate’s mom was going to be Helen Hunt. Which she wasn’t, but I knew that going in.
From a scientist, there were ups and downs. I have to give them props because I know they did their research. They had consultants from the National Severe Storms Lab, the Storm Prediction Center, and from the National Weather Service, so that made a big difference. They made a lot of references to certain parameters and environmental conditions. “Outflow” and “storm interactions” and “the cap” — those are all science-y things that we look for.
There’s some legitimacy to the technology, particularly with Javi’s phased-array radar, the triple-radar network. That is something that scientists have been trying to accomplish for years, with very minimal success. And if we’ve been trying to do this for 30 years, you can understand that the success rate is low for a reason: Logistically, it’s very challenging to do. The fantasy part is there’s no way we would ever be getting within 300 meters of a tornado. You’d lose an instrument every single time.
How do you really do it?
You set up farther away, within ten kilometers or so of the tornado. And most of the time, we’re only operating with two radars because getting the third one in there usually implies that you’re in the back. It’s challenging to set up from the backside of the storm as a storm is moving away from you. You can’t really get yourself in position without the storm basically mowing you over first.
One aspect of tornadoes that Twister and Twisters both highlight is that there’s a lot of uncertainty around how exactly they form.
When we’re out in the field, we don’t really have a sense of, Is that storm actually going to produce a tornado in five minutes, in ten minutes, in 30 minutes? We understand what we need, ingredients-wise, for a tornado to form. We need a strong rotation at the ground that basically gets sucked up by a strong updraft immediately above that. But the questions then are, how does that strong rotation at the ground get there? Where does it come from? How do we know when the storm updraft above is going to be strong enough to basically suck that rotation upwards and stretch it out into a tornado?
Similarly, you can have two storms that are separated by 20 miles or so that appear to be in similar environments, and one of those storms produces a tornado and another doesn’t. What was special about that one particular storm? Those are kind of the details that are still unknown. While we’re definitely further ahead now than we were even ten years ago, the details still elude us. And part of the reason for that is because we don’t have sufficient observations of the process.
In the movie, it’s like, We’ll get up close to a tornado and shoot something into it, and in real life, it’s more, Let’s stay a safe distance and shoot radar at it?
Yeah. The radar is basically like an X-ray for the tornado. With multiple radars looking from different angles, you can triangulate the three-dimensional wind field. But information that we don’t get from radar includes temperature information and pressure information and even, to a degree, moisture. It is important to know the details of that temperature profile. Where is the air cold? What’s the boundary between the cold, storm-generated air versus the environment? We need three-dimensional thermodynamic data to help put the nuances of the storm into context. As far as I know, that technology doesn’t exist.
Speaking of technology that may or may not exist, I’m curious about the idea of using absorbent polymers from diapers to suffocate a tornado. I imagine that people don’t do that in real life. But is it based on anything?
Hypothetically speaking, it might be sort of possible. But the amount of material … I mean, when you think about a tornado, there are kilotons of moisture. It is heavy, heavy-duty. We’re not talking about a baby’s diaper being good enough. So you would need tons and tons of this material to be somehow ingested into the storm in the right spot at the right time. And honestly, the environmental implications of basically covering the earth with this, coupled with the implications of potentially using silver iodide — those would cause far more pressing issues than the tornado itself.
It might be a bad idea to take tons of gel, shoot it up in the sky, and have it come down wherever?
Probably not a good idea. And that was one of the elements of the story line where there was sort of a breakdown between the science and the fiction for me. I appreciate what the writers were trying to do. They wanted something with a big bang, literally and figuratively, and that was the direction they went. The cringey part for me is I get emails all the time from people asking bizarre things, and there is going to be somebody who’s going to try to do something like this, and they could put themselves and others at risk. So that, to me, is the scariest part of the whole story line. Because this is going to seed ideas in people who really have no idea what they’re doing, and it could potentially be problematic.
Can you share some examples of bizarre emails you get?
I’ve had people talk about shooting missiles into tornadoes to blow them up, very seriously. I’ve had people say that they can predict tornadoes knowing the phase of the moon coupled with the magnetic phase of the Earth. Or let’s build this humongous wall across the central plains so that tornadoes run into it and die.
Going back to the movie’s premise, let’s say there was a big tornado headed for Oklahoma City, and we did it: We destroyed the tornado. The problem becomes, Okay, you destroyed it at that specific point, but the environment outside of that point is still the same. You disrupt the storm for five, ten minutes, but that doesn’t mean that in 20 or 30 minutes, it’s not going to be capable of doing the same thing again. If we disrupt or destroy it, what are the side effects of that? What are we going to be inducing downstream that we aren’t anticipating?
Is it harmful to shoot fireworks into tornadoes, or is that a victimless crime?
Trying to get something into a tornado like that, I think you could maybe do it for a really weak EF0 or EF1, where the winds aren’t too bad. Clearly if you’re getting it into the tornado in the first place, you’ve got to be in the tornado. And for some of these bigger tornadoes, oftentimes there’s so much debris and cloud material and rain in the way that it would never create the beautiful, luminescent spectacle that we observed in the movie.
If your truck had a little corkscrew thing on the bottom, would that be sufficient to protect you from a tornado?
I love that question, because actually, this is a real thing. There are individuals that have vehicles equipped with spikes like that. In fact, when I went out last spring with the film crew to get the background footage, I was with Sean Casey. We didn’t have these big corkscrew augers, but his vehicle is equipped with spikes. Sean and I go back like 20 years. I was with him on one of the first chases where he actually implemented this idea. The whole vehicle is on a hydraulic system that lowers it to the ground, and then you drill in spikes that go down ten feet and anchor the vehicle. They’d chop your leg off if you were in the wrong spot. I hesitate to say they’re “safe,” but your vehicle probably won’t roll in some of these lower-end tornadoes. The biggest problem is not so much the winds, but what’s inside the tornado. That tornado is picking up trees and houses and farm material. If something like that whacks you at 200 miles an hour, your vehicle’s rolling no matter what.
There’s been some back-and-forth over whether the movie should have mentioned climate change. The director, Lee Isaac Chung, has said that was partly because the science is unclear on how much climate change affects tornadoes. Can you fill me in on the current scientific consensus?
He’s absolutely right. The science just isn’t there. And the reason why the science isn’t there is because there is such a spatial and temporal gap between climate and tornadoes. Climate is long-term, hundreds, thousands of years, and regional climate conditions are hundreds of miles wide. Tornadoes last seconds, maybe an hour if you’re lucky, with spatial scales of max 2.6 miles, the widest tornado ever recorded. So they’re much smaller, and much shorter-lived.
Tornado generation is linked to small-scale processes within the storms themselves that are not necessarily climate-related. When we look at long-term trends in our tornado record, there is no evidence to suggest that we are getting more tornadoes or more intense tornadoes. I just pulled all the data myself and generated a whole bunch of plots, and if you look at the EF2s or greater, those actually might be decreasing in number over the past 30 years. The number of weak tornadoes is definitely increasing, but is that increase physical or is it driven by our ability to better identify the damage and the tornadoes themselves? Most likely the latter.
What we can determine from longer-term trends is that we are seeing a change in the distribution of tornadoes in terms of geography. Places like Alabama, Tennessee, Mississippi, over to Georgia a little bit, up into Midwest states like Illinois and Indiana, even over into Iowa. Over the past ten years, these states are seeing an increase in environments that are favorable for tornadoes. What that means is that we’re seeing this shift away from the central plains. I hesitate to say this, because the central plains are still maximized, but we’re coming to a little bit more of an even playing field between the traditional Tornado Alley area and the southeastern United States up into New England.
Is that driven by climate change? Possibly, because those larger-scale environments are climatologically linked. So when the surface conditions are warmer, when they have more humidity, coupled with cold air above and strong, larger-scale low-pressure systems that can then be made more robust because of stronger temperature contrast, that might have some play in the climatological sense.
We’re also seeing more tornadoes earlier in the year. Wintertime tornado outbreaks appear to be on the rise while summertime tornadoes appear to be on the decline. So when you look at the U.S. as a whole, or look at the year as a whole, the numbers are staying the same. But when you drill down into specific geographic regions, specific seasons within the year, we are seeing some indication that things are changing. And places like New England and the metroplex across the eastern part of the U.S., it is entirely possible that they might end up seeing more tornadoes than would historically be expected.
We had a tornado in Brooklyn a while back. I remember being like, Huh, didn’t know that was possible here.
When we look at the tornado events in that area, these are the sort of lower-end tornadoes: relatively short-lived, relatively weak. But still damaging. There’s more out there to damage than there is across the central plains, which is mostly farmland with minimal structures. So people’s lives are still impacted. Trees still come down. People’s houses are still damaged.
This interview has been condensed and edited.