Photobombing satellites could ruin the night sky for space telescopes

0:23Welcome back to the Nature Podcast. This week, how satellite constellations could threaten space-based astronomy and the damaging consequences of video call glitches. I'm Nick Partridge-Hale and I'm Benjamin Thompson.

0:50The night sky is a glorious thing. But the night sky is changing. I remember when I was a kid, you go to the forest doing camping or stargazing, and sometimes you will see one tiny dot, not blinking, because blinking lights tend to be airplanes, but just continuously crossing the sky. Those are satellites. They were really hard to see, and now they are getting way, way, way more common. And that's a problem. This is Alejandro Borlaff from NASA's Ames Research Center in the US. He and his colleagues have been investigating how much of a problem the increase in communication satellites in low Earth orbit could pose to astronomy.

1:34And when he says their numbers are increasing, he's not kidding. There are a lot of them. Right now we have 15 ,000. Around four years ago we had 4 ,000. We have launched more satellites to space in the past four years than in the previous 70 years of space flight combined. And this is only the number of active satellites. There are also a huge number of broken ones, or bits of them, whizzing around above us in a low Earth orbit, the region of space less than 2 ,000 kilometres away from the surface of Earth. And these objects reflect light, which can manifest itself as thin streaks that cut across images taken by telescopes, causing all sorts of headaches for researchers.

2:19Sometimes we are interested in things that happen very quickly, like a supernovae. You are observing a galaxy and then suddenly a star far, far away explodes. And that looks like something that happens in only one image. If you happen to have a satellite crossing, you will lose that information forever. But if things can be difficult now, what might the situation look like in future, as giant constellations of telecommunication satellites are launched in the coming years? This is something that Alejandro and his colleagues have been looking at in a Nature paper this week. Specifically, they've been simulating how many images taken by space-based telescopes might be photobombed by the satellites they share low-Earth orbit with.

3:03In their paper, the team say that while there's been a lot of attention on the impact that future satellites could have on ground-based telescopes, there's been less on their off-planet counterparts. There is some evidence, though, that these satellites are already having an impact. A paper in Nature a few years back looked at the occurrence of satellite trails in images taken by the venerable Hubble Space Telescope, in orbital operation for well over 30 years. This work found that between 2018 and 2021, just over 4 % of images contained satellite trails. During this time, there were about 8 ,000 satellites orbiting Earth.

3:45It may sound really small. Like, OK, 4 % of the images, that's really a problem. The problem is how this scales over time. If you consult the number of satellites that are proposed to be launched, the number is staggering. We are predicting to have half a million satellites. To work out the potential impact 500 ,000 satellites might have, the team gathered information on things like the proposed altitudes and orbital periods for launches planned over the next 15 years, and used that information to build a computer simulation of the satellites speeding around the globe. The team then simulated the view that four space-based telescopes might have.

4:27NASA's Hubble and Sphear X instruments, both of which are already in operation, the European Space Agency's proposed Arrakis telescope, due to launch in 2030, and Shuntian, a telescope from the China National Space Administration, due to launch next year. The results of what the telescopes might see were stark. We predict that it's a little bit higher than one every three images from Hubble Space Telescope will show one satellite trail. That's a staggering number. That's really, really high compared to what we see right now. In the case of Spherix Telescope, Arrakis or Shintian, we would expect like 96 % of the images will be contaminated somehow.

5:08In terms of numbers, the authors predict that, on average, there will be just over two satellites observed in each Hubble image, about six for each Sphere X image, 69 for Arrakis, and 92 for Shun Tien. You can see what some of these simulations look like in Alejandro's paper. The sheer number of trails makes some look almost like really scratched vinyl records. Now, a record that damaged wouldn't play, the contents would be lost, and there's a real risk that the same thing could be the case for the astronomers looking for information in images like these. Despite this, Alejandro says that there's context to consider about the impacts of these trails.

5:53For example, some telescopes that stitch together multiple images of an object are less likely to be impaired. And where these telescopes are and what they're trying to observe is also important. Of the four telescopes they included in their study, Shuntian is due to be in the lowest orbit around the Earth. This means it would have the most satellites above it, but its camera resolution is very high, so these streaks will be very narrow in the images, potentially lessening their impact. Arrakis, on the other hand, although due to be in a high orbit, will have a lower resolution camera, meaning streaks will be wider, covering more of each image.

6:35But for Alejandro, seeing what the simulations kicked out and what the future could look like for low-orbit space telescopes was an eye-opener. I think I was very worried. It's hard to detach this from a personal perspective. I mean, I've been working with telescopes half of my life. Working in a way to make telescopes see better is something that we always push. At some point when you see that something is going backwards over time, then it's when you start having this sort of anxiety in a way. Michael Brown is an astronomer based at Monash University in Australia. He wasn't involved in this research, but shares some of that anxiety and says that the work provides a troubling snapshot of a potential future.

7:18Things are bad now with the satellite constellation. and the paper really drives home that they could get much, much worse in the coming decade. Just the scale of the constellations, how bright the individual satellites are, could get much, much worse. In the paper, the team actually investigated not just the impact of the half a million planned satellites, but what would happen if a million were launched. Spoiler, the results would be even worse. Michael thinks that the way the team went about their analysis was robust. I think they've gone about it in as reasonably a systematic way as possible, and it's a perfectly reasonable estimate of what would happen.

8:01Some of these satellite proposals do indeed talk about more than 100 ,000 satellites. So half a million satellites or even a million satellites isn't out of the question if these plans go ahead. my hope is they don't go ahead and they don't all come to fruition. Because of course the future isn't yet written. But it seems likely that a great many satellites will be launched and Alejandro wants the paper to highlight that many space-based telescopes face the same risks as their ground-based counterparts. He says that highlighting this issue will hopefully provide time for strategies to be put in place to mitigate the most serious impacts.

8:44Now, a number of recommendations already exist, but in their paper, Alejandro and the team make a few more to specifically aid space astronomy. These include things like having access to ultra-accurate open information on satellite orbits, and establishing upper limits for these orbits, so future telescopes can potentially orbit above them, minimising the trails they see. All these things will require dialogues between relevant parties, scientists, policymakers, and the companies that produce these satellites. Alejandro says their paper represents a worst-case scenario, and that doing something about it won't just benefit astronomers, but ultimately everyone.

9:28I think that astronomy is sort of like the current in the coal mine. They give you information on things that will happen far in the future, because the telescopes are so sensitive that there's you way before they start to be bright to our own eyes if we make space dark for spacecraft we will probably make it as well for people on the ground and you know when you go stargazing you will see bright stars you know 100 years in the future and they will be the same protecting the dark is something that michael echoes and he says this research emphasizes is that that is something that is at risk. I like to think of the sky as a bit of a shared wilderness.

10:08You know, it's something where if you go to a suitably dark place and look at the night sky, you can see the night sky as your ancestors saw it hundreds or thousands of years ago. And I've noticed it recently when using the telescope just for amateur observing that now I will see several satellites a night. And I think soon it's going to get to the point where the satellites become more and more common just to the unaided eye. So I think this is part of a growing chorus about the dangers to our shared night sky and that our night sky is changing. And I think this is part of that chorus and it should be raising alarm.

10:43That was Michael Brown from Monash University in Australia. You also heard from Alejandro Borlaff from NASA's Ames Research Centre in the US. To read Alejandro's paper, look for a link in the show notes. Coming up, how the glitches everyone experiences on video calls could have more of an impact than you might imagine. Right now, though, it's time for the Research Highlights with Dan Fox.

11:11Researchers at CERN in Switzerland have trapped more antimatter more quickly than ever before. Antimatter particles are mirror versions of matter, with opposite values of properties like electric charge. When these antiparticles encounter their matter equivalents, they annihilate, releasing energy, making them tricky to study. Scientists at the Anti-Hydrogen Laser Physics Apparatus, or Alpha Collaboration, used an indirect laser cooling technique to bring positrons, the antimatter equivalent of electrons, to a near standstill and bind them with antiprotons forming atoms of anti-hydrogen. They accumulated more than 15 ,000 anti-hydrogen atoms in under seven hours, a rate eight times higher than the previous record.

12:04This enhanced trapping technique should allow more detailed studies of any differences between matter and antimatter, perhaps helping to answer one of the biggest questions of all, why matter dominates antimatter in our universe. Read that research in Nature Communications.

12:26Hydrogen is often touted as a green fuel. But a new analysis indicates that the benefits vary considerably from sector to sector. Researchers evaluated hydrogen's real-world environmental impact by looking at the full life cycle of emissions from 2 ,000 existing or planned low-emission hydrogen projects. They conclude that when it comes to powering road vehicles and domestic heating, direct electrification would provide greater emissions reduction than hydrogen. But turning hydrogen into liquid biofuels for aviation and shipping, or using it in steel production, are effective ways of cutting overall emissions.

13:07All this assumes the hydrogen itself came from a low-emission source. So-called green hydrogen extracted from water might also replace the hydrogen currently used in crucial industries such as fertilizer production. Currently, the majority of this hydrogen is produced from fossil fuels without capturing the carbon, releasing something like 1 billion tonnes of CO2 every year. Find that paper in Nature Energy.

13:41I'm willing to bet at some point since the last podcast, you've been on a video call with someone and something has gone a bit wrong with the picture. Maybe the person you were speaking to froze for a few seconds or they pixelated or the audio cut out. These little interruptions or glitches are an annoying feature of modern life but according to a new study in Nature these glitches can potentially have real negative consequences. To find out more I gave one of the authors Jacqueline Rifkin a call on a stable connection. She told me how this study started five years ago, right about the time something else was going around.

14:22So that was the height of the pandemic and what we were observing is that pretty much everything you might want to do as a person was being moved over to video call and it was just becoming more and more prevalent and a more and more kind of acceptable substitution for in-person interaction. And so, you know, these glitches are sort of becoming embedded into the DNA of our lives every day. And we really wanted to know what might be going on when they're happening in all of these important calls. So to understand this better, you actually did a series of studies and experiments, and I wanted to break it down a little bit here.

15:02So you started off by looking at some archival footage of thousands of strangers getting to know each other via video call. And interestingly enough, for this one, you had a measure of how glitchy these calls were. What did you find when you looked through these? What we were able to do was we were able to look at the calls that people reported glitches in and the calls that didn't. And we were able to look at survey data of how connected those people felt to each other after the fact. So did they feel like they liked each other? Did they feel like they kind of found common ground, maybe even wanted to connect in the future?

15:41And what we found was a correlation where the calls that had glitches in them, where the participants said, yep, there was an issue with my call. We were able to go into the footage and verify that they had had a glitch. Those people felt less connected to their conversation partners after the fact, compared to the people that had kind of seamless, no-glitch calls. And you also looked at a job interview situation and this is a common way where people meet for the first time and they would also worry about glitches. What did you find was the effect of glitches here on whether or not people got hired?

16:17Yeah so we had sort of a simulated job interview. We had recorded videos of someone who was making their statement for why they thought they were a great fit for a job but only for half of participants. We edited in glitches So they saw some freezes and some other sort of disruptions in the call. And those glitches, again, caused a significant decrease in participants' interest in hiring this person, supposing that they were interviewing them for their own firm. Speaking of impactful life circumstances, you also looked at parole hearings. And I should say at this point that this wasn't an experiment you ran.

16:55Rather, you looked at the transcripts of parole hearings to then understand how glitches were having an effect. What did you find when you looked at parole? So we were able to read all of these transcripts and come up with a dictionary, essentially, of words and phrases that we thought indicated that there had been a glitch happening. Words like freezing or, you know, staticky noise, things like this. And what we found was a really kind of interesting, but perhaps concerning correlation was that the calls that contained those indicators of glitches in the language were associated with a lower likelihood of that particular inmate being granted parole.

17:35So we can't necessarily make a causal claim because this data is correlational. We're just observing it here. But it did strike us that there may be this sort of even broader potential concern. They might be having these sort of very subtle changes to our psychology and the way we feel about the person on the other line that could even have implications for someone's freedom. And were you able to build up an understanding of why it was that these glitches were having these effects? You could probably come up with a number of different reasons. First of all, they're annoying. They're frustrating, right?

18:11Everybody notices them. So maybe it's the frustration issue. There's also the kind of understanding and comprehension issue. So if someone's cutting in and out all the time, you can't totally tell what they're saying. And if you can't totally tell what somebody's saying, you just might not like what they have to say as much. we thought that there might be something else a little bit more nuanced going on. And so what we chose to think about was the fact that video calls are so good at mimicking a real life interaction. So compared to any other technology we have had in the span of human history, video conferencing specifically kind of feels like you're sitting across the table from the other person.

18:57You can kind of suspend disbelief for a moment and simulate that you're face-to-face sitting across the table. The issue though is that glitches kind of break that simulation and what that does, that sort of shattering of the illusion, is it creates a sensation called uncanniness. And you looked at the alternatives that you suggested such as like oh maybe you can't hear someone properly or something like that and you still decided that uncanniness is the most plausible explanation. Yeah, exactly. So we did our best to sort of say, you know, well, could it still be comprehension issues? So in the job interview context, we said, do you remember what the person said their major was?

19:40Do you remember what the person said their greatest professional strength and weakness was? Things like that. Every time the glitches had no impact on comprehension. So being exposed to these glitches did not make it any harder to understand or sort of get what the person was saying. So it wouldn't be that. We also found that when we were able to statistically control for things like disruptiveness or just sort of the mere kind of interruptiveness of the call, that did not have as big of an impact. And so we were able over the course of our studies to sort of chip away at these other explanations and saying, sure, they may have some impact in the real world, absolutely, but uncanniness, this strange and eerie sensation, is still exerting this negative impact on these evaluations and judgments.

20:34And when it came to the implications section of your paper, when you discussed, you know, what all this means, you wrote that glitches have the potential to fundamentally alter the trajectory of individuals lives. And this may sound like quite a bold claim. So how was it that you came to this conclusion? So I think this really just comes from the scope and the frequency of these interactions that we've been studying in this paper. Of course, in any of our studies, it's, you know, a single simulated job interview with a single rating of, you know, if this person we're trying to join your firm, how interested would you be in hiring them?

21:15So we're not trying to overstate the size of any given result that we have in our studies. But where that kind of sentiment came from was the idea that in a given day, someone might have 5, 10, 15 calls on video conferencing. And in a given year, they might not just do video conferencing for work, they might do it for fun, for their health, for their business, for their family. So there's these many, many exposures to video calls. And with those come many, many exposures to glitches to the point where, you know, if you're, you know, someone with access to less stable internet, if you're the type of person that your calls are constantly being damaged by these glitches, it's hard to say whether there might be this compounding effect where you might be just systematically less favored in a job setting or systematically less likely to take on advice given to you from professionals on the other end of the call or systematically less likely to get social support and feelings of connectedness from a given social call.

22:28So these things might just sort of accrue over time because of the scope and the frequency of what we've been looking at. That was Jacqueline Rifkin from Cornell University in the US. For more on that story, check out the show notes for some links. And that's all for this week. As always, you can keep in touch with us on X or Blue Sky. We're at Nature Podcast. Or you can send an email to podcast at nature.com. I'm Benjamin Thompson. And I'm Nick Petrichow. Thanks for listening.

23:05Thank you.