Jordan: As a scientific concept, the butterfly effect can be traced back to an American mathematician who was making weather models in the 1960s. As a reference point for popular culture. You should know there have been three Butterfly Effect movies, for instance, that you’d probably trace its origins back to Jurassic Park in a brief, memorable scene.
Film Clip: It’s still not clear on chaos. Oh, it did it. It’s simply a building with predictability and complex system. The shorthand is the butterfly effect. The butterfly flaps its wings in Peking and in Central Park, you get rain instead of sunshine.
Jordan: The reason I mentioned the butterfly effect is because that scenario it showed up in an experiment is very likely to happen sometime soon in real time. In 1961 when this all started, Edward Lorenz was a mathematician who was running a numerical computer model trying to predict the weather. At one point, he entered the value of just one particular condition as 0.506. Instead of the longer but more accurate 0.506127 and when he did that, the entire weather prediction changed everywhere. And today we use a very similar and incredibly complex formula to predict the weather around the world over the next three days or five days or 10 days. It’s a system that is just as vulnerable to tiny changes in the information it uses as Lorenzo’s predictions from almost 60 years ago were. And that brings us to the wireless spectrum and to 5G technology in the United States specifically, and to a very real reason that our weather forecasts here in Canada and everywhere else in the world are about to become much less reliable. I’m Jordan Heath-Rawlings. This is The Big Story. Dan Vergano is a science reporter based in Washington, DC. He works for Buzzfeed News. Hey Dan.
Dan: Hey, how are you doing?
Jordan: I’m doing well, thank you. I’m going to get you to start today’s story with the absolute basics for people who don’t follow this and haven’t been paying attention. So what is 5g?
Dan: So 5G is the next generation of wireless phone service and sort of WiFi networks for everybody. It’s promising, you know, better, faster, stronger, brighter, happier, uh, wireless service to everybody.
Jordan: How is it different from 4G, 3G LTE, all that stuff.
Dan: 5g will basically be able to carry a broader stream of information to users. So you’ll be able to watch movies without buffering more easily. You’ll be able to basically have more information transmitted to your phone than previously, more easily. It’s basically just an improvement in our current service, but everyone is looking forward to it in the media business because they’re hoping people will start streaming movies like crazy and make a fortune for Hollywood and everyone else.
Jordan: Right. And how quickly is it being adopted?
Dan: It’s happening now. So South Korea has gone to 5g and it’s rolled out, uh, in North America and at least 40 cities. Is this going city by city location, by location all over the world. The perception is of the US and China are in a race to roll out 5G services and define the standards for 5G that will in turn, uh, enrich the technology companies making the equipment that supplies these five G networks.
Jordan: How much money is at stake for these companies? How big a deal is this?
Dan: It’s hundreds of billions of dollars. It’s the electronic economy of the future.
Jordan: So where did the first red flag come from concerning a weather forecast.
Dan: The weather community has been concerned about preserving its access to important wavelengths for decades. Uh, in the 5G case, the rumblings really started in the last year or two and sort of technical meetings, um, and burst into sort of public light, uh, around April, uh, February, April of last year, where FCC and Congress started getting into discussions about how are we going to auction off this one spectrum that’s particularly close to this important wave length for weather forecasting. So there was fighting back and forth, uh, when the auction of, uh, this particular wavelength, 24 gigahertz started, uh, around February, March of last year, Congress had hearings in April, uh, and in may, uh, where you sort of. The technical fight between two federal agencies came to light and, uh, it has sort of bubbled in verbal sense then. And, uh, there was a big international meeting in November where some standards were set that cause more consternation in the weather community and Congress again decided to intervene, demanding an investigation of the whole situation. And that’s sort of at the point where we said, okay, this is risen from the technical level to the political level that we really ought to bring it to our reader’s attention.
Jordan: Can you explain to me, uh, how weather forecast and weather satellites work and why there’s concern here?
Dan: Uh, so yeah, easily! Uh, no, actually it’s complicated, but, uh, the weather satellites are, are in space, right? And they’re looking down so they get a beautiful view of the atmosphere. That’s a key thing, right? They’re not pinned into the atmosphere where, you know, we can only see. 10 miles, the atmosphere looking up from where they’re seeing across hundreds of miles. So there are beautiful devices for looking across broad swaths of the atmosphere. And the problem is like they have to figure out what’s going on down. So they do that by looking at the characteristics of certain molecules that are important in the weather, like water vapor and, uh, trying to, uh collect all that data and they in turn feed them into these weather models and what those are essentially are collections of what are called nonlinear partial differential equations. This is like the top end of your calculus class if you went on a did that. Systems of these that are solved continuously with super computers, and so all this data that they’re collecting, you know, the. The humidity, the temperature, the presence of water, vapor, ice, snow, uh, everything else is fed into these models. They’re essentially mathematical models and they’re crunched away by the supercomputers and they come out with predictions for what the weather should be like the next couple of days. It’s a global thing. It’s the entire globe being calculated this way and like 200 you know, cells across the globe. So it’s a gigantic circular math problem that’s being solved 24 hours a day continuously using data taken from these weather satellites that are taking into the atmosphere. It’s not a thing that’s just like there are, they’re looking down at Toronto and seeing, okay, it’s going to be windy tomorrow. It’s actually a gigantic math project that’s working all the time and requires a lot of data in order to be accurate. And the crazy thing is with partial differential equations, uh, they are, are susceptible to these nonlinear effects, the famous butterfly effect where you have, you know, a butterfly flaps its wings in the Amazon, and then there’s our hurricane over Lisbon. And so small errors in data in one location means that the solution that they get for six days later, or three days later, somewhere else is way out of whack. And that’s the kind of problem that they’re worried about, uh, with interference in the frequencies that they look at to determine the weather.
Jordan: That makes sense. But how does the 5g frequency mess with that?
Dan: So the, the 5g frequencies are the butterflies wings flapping over these cities and putting bad data into the numerical models of the, of the global weather forecast. So you basically inject a whole lot of noise over New York City, and that throws off the signal for this whole solution of the global, uh, weather forecast. And you might have screwed up weather forecast everywhere else days later, or, you know, uh, hours later, but somewhere else. So it’s not like, okay, there’s a lot of phone traffic over Toronto and it screws up the weather forecast for Toronto. It’s, there’s a lot of phone traffic all over Los Angeles, Chicago, Dubuque, and who knows where else. And all of a sudden, three days later, the weather forecast in Toronto was screwed up for mathematical reasons that we can’t predict very easily. So if you look up numerical weather prediction in Google, there’s a beautiful, uh, free picture. I know it doesn’t help you with a podcast now, but it might give you idea of the cell by cell sort of solution that they have to do for this kind of calculation. It really is a gigantic math problem that’s being done all the time by the various weather centers around the world.
Jordan: Why 5g specifically? What is it about that frequency?
Dan: So it’s not 5g in particular, it’s interference with the frequency. 23.8 gigahertz, which is the resonant frequency of a water vapor in the atmosphere, which is kind of important to know what water vapor is doing, right? You have to know your humidity. It’s basically a measure of humidity in the atmosphere. And what’s happened is the FCC in the United States is sold off, uh, rights to broadcast and frequency that’s really close to it. 24 gigahertz, right? So 23.8, 24 pretty close, right? And the problem is noise. It’s just like radio stations aren’t allowed to broadcast too close to each other. That’s why there’s numbers on your radio station right? On your radio channel. You can’t have them, uh, broadcasting over top of each other. And it’s worse here because the water molecules, it turns on, aren’t radio stations. So they’re not competing with these other signals that are bleeding into them. They’re just being bled onto. And so the fear is that unless you keep a really good buffer between the frequency at which this natural phenomenon happens, and the broadcast, which is going to be noisy as hell in a totally unpredictable way, because who knows when someone’s going to pick up the phone and make a phone call, that you’re going to have a totally unpredictable interference in the readings from the weather satellites and in turn that garbage is going to get fed into the weather models we rely on the predict the weather, and then in unpredictable ways our weather is forecast are going to be screwed up.
Jordan: So could the FCC just have sold off in different frequency and avoided this problem?
Dan: Uh, they could. The problem is that the frequencies are getting crowded, right? We’re becoming more and more reliant on the spectrum, and there’s only so much to go around and you know, you can only put so many buildings up in downtown Toronto. Right? Uh, and so the real estate is very pricey there compared to, you know, somewhere else. And so things are getting really tight. And this, uh, sort of part of the spectrum that has been set aside for, uh, emergency communications and weather and scientific aspirations is, you know, looking like free public land to the telecoms and they want it. Uh, and it’s, you know, they’re saying we’re going to give you a lot of money and build a better society, so there’s a lot of reasons to give it to them. The solution has been that they just require the broadcasters to use a wider buffer. And that’s essentially what the fight is, like how wide a buffer can we put around this, these frequencies and protect them? Where, you know, the telecoms are basically saying like, you don’t need much of a buffer at all, is the scientists are saying like, we need a gigantic buffer and you know, when NASA did this study and saying like, it’s quite a wide buffer we need. And the FCC is saying that it’s garbage. We don’t believe you. And you know, the problem is that it’s a kind of gigantic natural experiment. We’re going to see who’s right. But you know, in the past, NASA, you know, have been pretty damn good at predicting this kind of thing. So it makes the weather guys or, or you know, the people who bring you your phone service, you know, who do you trust, uh, you know, who has a better track record. That’s basically what we’re talking about here.
Jordan: Do we have any idea how badly the forecasts might be impacted? Do we know where and when or is it just random?
Dan: It’s both random and, um, it’ll increase over time as the five G networks build out, but it’s a question of how much are the telecom sort of listening to stories like this one in yours right now, do these complaints and sort of being prudent and not, and not going after this part of the spectrum so much themselves. So it might be that they take special care, you know, they realize this is a problem and they take special care and it has no effect. It might be that they ignore it completely or they just say, FCC let do it so off we go, and they’re going to cause a lot of interference. We don’t know yet. This has yet to be built. But the prediction has been that, you know, the weather forecast go back to the essentially 1980 level by the head of NOWA, where in 1980 and if you’ve got three days of weather prediction, you’re in good shape. And you know, uh, they were just much less reliable. They say there’s a 70% chance of rain. Doesn’t mean it’s going to rain, right? That means that there’s a seven in 10 chance it’s going to rain, but you could be pretty sure that there’s a seven and 10 to 10 chance of rain, and in the future it might be, maybe that’s really not such a good guest, you know? So it causes a lot of problem, a lot of uncertainty for people. You know, everyone’s got a pack, an umbrella time.
Jordan: You mentioned that it could make. Three-day forecast, less reliable, or kind of take it back from an 8 out of 10 to a 7 out of 10 but are there bigger implications in terms of severe weather events, which we’re now seeing more frequently? Like is this going to make it harder to figure out where hurricanes are coming or massive snow storms? So that kind of stuff.
Dan: It could potentially make the predictions for things like massive hurricanes, massive snow storms, the weather track, harder to predict. You know, it’s nicer to know it’s aimed in Miami, seven days ahead of time rather than three days ahead of time though. The hurricane is a good example where you see these tracks, like we know that in six days there’ll be here where we might have the point that’d be at the point where like in six days we have no idea where it’s going to be. That sorta thing costs a fortune, right? The companies start packing up and moving for no reason, six days ahead of time, and then they have to turn around and send the trucks back to, you know, with the goods. And so you’ve just caused a whole lot of costs to everyone, which whole society will feel for no reason.
Jordan: And just to be clear, this is a fight that’s going on in the United States right now. This is the FCC that sold it, but the potential implications are global.
Dan: It’s a global fight. Uh, it’s focused in the U S right now because of Congress asking for an investigation. Uh, and the FCC auctioning the spectrum to U S telecoms, which were hugely influential, but it is a global fight. It’ll affect the weather globally, and it is the International Telegraph union that had the meeting in November in Switzerland. It’s set aside a buffer levels to operate on this part of the spectrum.
Jordan: So 5g across the world will run on the same frequency everywhere. It’s not like the buffer could be larger in Canada than it is in the U S or as it is in Italy or whatever.
Dan: The 5G will run on various spectrums that the ITU, uh, the International Telegraph Union says it is allowed to run on. One of them, uh, will be this 24 gigahertz spectrum that the FCC is licensed to telecoms. So it doesn’t matter if Italy says, no, no, you can’t do it. Or the Europeans who have been, you know, more concerned about this than the FCC say, no, no, you can’t do this. They, they fought for and got a stricter, but not. Perfect offer, uh, at the meeting in November. It doesn’t matter because it still might be the weather prediction over, you know, Rome that’s screwed up because of bad data that’s been injected into the weather models from the United States.
Jordan: So what have other countries, uh, said to the United States as, have they been pushing back against that frequency in particular? Well, the, the fight was in November at the ITU meeting, at the International Telegraph Union meeting, and basically the European, uh, spectrum authority fought back against FCC, which wanted a very loose buffer. And so they got a stronger one. But it’s a compromise. It’s a consensus. Organization, and so they got a stronger one, but it’s still at least eight times more powerful than the Europeans wanted. And then then NOWA suggested would be safe too. So there was a little bit of a pushback, but it’s not fair to call it a compromise because it’s still one that poses a threat of interference to the weather forecasts.
Jordan: So what happens next? Uh, both politically and I guess in terms of, we just have to wait and see how badly, uh, the forecasts are impacted as this thing ramps up?
Dan: So on the politics side, uh, what happens is that the government accountability office will likely pursue an investigation of how NOWA, NASA, and the FCC came to a logger heads on this and sort of what went wrong. Why were these people talking to each other? And that might lead, uh, to Congress to take more steps to legislate stronger limits on that in the United States on, on what frequencies can be used. Well, what size buffer could be used. It’s possible. That would, that would be one place where the action would be. It might be that, uh, there is a reaction in Europe to this going on and, you know, they have a say in it or, and push limits that, you know, in effect, uh, telecom interests so strongly that they steer away from this. But yeah, we really have to see, I mean, the suspicion I’m hearing from people, I spoke to more people, in talking to the story is that, we’re just going to have to see how these 5G networks, these base stations are built out. And there’ll be a certain point where it will be apparent like, Hey, the weather predictions are less accurate than they used to be. You know, we should have a data, you know, like you’ll take a six month swath at one point say, yeah, their predictions were more often now than they were five years ago. And you know, the most likely candidate is in fact this 5g interference. The problem of course, is that it won’t be interference that happens in a systematic way that can be removed like a lot of the, uh, telecomm folks seem to think, you know, it’ll be random. And so that’ll be a nugget, a little harder to detect. But at a certain point, you know, they’re expecting the next three years or so, you know, the weather forecast will start to degrade and that’ll be detectable. And then we’ll have to turn around and have an emergency ITU meeting and you know, further restrict the buffers or whatever. As it is a plan now is that the limits are on the buffers until 2027, uh, in order to encourage the industry to build out. And then they get a little tighter after 2027 and it was only supposed to, it’s only scheduled to take a look at this again in 2023. So, you know, worst case, we’re looking at 2023 for another reconsideration. All this.
Jordan: In the bigger picture. I mean, you said something fascinating a little while ago about how you can only build so many buildings downtown. Um, I’m presuming that 5g won’t be the end of this and that there will be 6G or whatever else, uh, comes next
Dan: Yeah, they’re already thinking about it.
Jordan: So yeah. Where, so where are the limits? Like what do you do at that point, when it’s just too crowded?
Dan: So the, the weather forecasters are really terrified about this because there’s a lot of other valuable parts of the spectrum too they need for, for, you know, other phenomena, for hurricanes and tornadoes and things like that. And what if those get bought up and they’re really having a hard time telling the weather to people? So this is a question we’re going to have to solve, uh, as a sort of global society. If we’re going to going to, you know, operate the airwaves in such a crowded manner, there has to be a better system to bring more players to the table than just, uh, the telecoms in order to just decide what parts of the radio spectrum are actually preserved for things like weather forecasting or, you know, scientific endeavors. There’s a similar fight, you know, going on right now over satellites and astronomy. You know, where there’s so many satellites going up the astronomers are complaining, is getting in the way, their telescopes and they can’t see the sky. You know, this is same, similar kind of crown he was going on. As we become sort of a more space-based society that is very valuable, real estate is going to become more and more fought over and fraught.
Jordan: And there’s nobody, uh, there’s no global body to adjudicate that really.
Dan: Well, I mean, there is the eye, the ITU, and you know, we could do it through the UN.
Jordan: America would love that.
Dan: Yeah. Well, it’s a hell of a country. Um, I, you can speculate how likely that would be to, uh, go over. On the other hand, uh, you know, we have, uh, our telecoms also want to collect money from the Chinese and the Indians and the Canadians and the Europeans. And so if they raise enough of a ruckus, then, um, you know, things could change. It’s just a huge fight. This is a very lucrative industry. This is a very well resource industry, the telecom industry, right? And these are the people we’re talking on a cell phone right now. The concern of course, is that, you know, anything sort of vaguely scientific or that doesn’t, you know, enrich those powerful industries will, you know, get a shove.
Jordan: That’s Dan Vergano science reporter for Buzzfeed News. This was The Big Story, for more from us, head to thebigstorypodcast.ca. Find us on Twitter @thebigstoryfpn or in your favorite podcast application, whatever one you might prefer. If it lets you leave us a review, give us five stars. Tell us what you think. Thanks for listening. I’m Jordan Heath-Rawlings, we’ll talk tomorrow.
Back to top of page
