PHOENIX, Ariz. — In June of 2025, we were greeted with a set of space images so special that one scientist even deemed them worthy of the title “astro-cinematography.” Indeed, they were unbelievable, dotted with TV-static-like dots representing millions of galaxies, printed with nebulas resembling watercolor canvases, and bursting with data about some of the farthest cliffs in our observable universe.
These were the first portraits to come out of the Vera C. Rubin Observatory, humanity’s magical new cosmic watchkeeper fitted with the world’s largest digital camera and a telescope with an enormous field of view. Rubin has the ability to thoroughly image the night sky over and over again from its vantage point atop Cerro Pachón in Chile, and with unprecedented efficiency at that. This is an instrument anticipated to revolutionize astronomy and reveal things about the universe we won’t be able to fathom until we find them.
It all sounds like a dream — but a wakeup call may be looming. An Earth-based telescope approaching the limits of modern technological power is unfortunately forced to contend with another kind of scientific advancement happening in space: the exponential rise of satellites in Earth orbit.
As of writing this article, there are about 14,000 satellites orbiting our planet — nearly 10,000 of which belong to SpaceX — and the number is going to increase aggressively as commercial interests in this realm continue to grow. Blue Origin and Eutelsat’s OneWeb, for instance, are following SpaceX’s example, as are several Chinese companies, and smaller startups are readying their own endeavors. SpaceX has actually recently floated the idea of a data center in our planet’s orbit, which would involve putting something like a million more satellites up there.
Priceless Rubin images could therefore be tainted by commercial satellite interference, or “streaks,” as astronomers say.
Why satellite streaks are a big problem
You may have heard about this issue before, as satellites are already well-documented to interfere with astronomy imaging. You can even see a streak right here in this relatively inconsequential image from the Gemini North Telescope, and there are quite a few studies written about why we need to be worried about megaconstellations populating Earth orbit in general.
“Astrophotography is a valuable educational tool for raising awareness and interest in the natural world,” Federica Bianco, a scientist with the University of Delaware, said during the conference. “The night sky environment is often culturally significant, and dark sky tourism has been recognized as an important factor in sustainable development of rural and remote communities.”
Just this month, physicians and scientists from Northwestern University announced they’re worried about satellites in Earth orbit disrupting our sleep patterns.
“They change the night sky,” Rawls said. “Turns out, telescopes are not the only things that look up.”
But even though satellite interference with science isn’t a new conundrum, the existence of Rubin brings the consequences into clearer focus.
“The same features that made Rubin really amazing for discovery are the same features that also make it vulnerable,” Rawls said. “Its wide field of view; its very sensitive camera; the fact that it can scan the whole southern sky every three nights — it’s going to see a bunch of satellites.”
“If all you’re going for is a pretty picture, fine — you can Photoshop out the streaks,” she said. “But trying to get the science out of that is a little bit trickier, because you’re introducing systematic errors in a way that’s actually very, very tricky to account for.”
The depth of the issue
At this year’s AAS conference in Phoenix, a cadre of scientists affiliated with Rubin spoke about different ways satellite streaks could affect the telescope’s discoveries. The goal? Find a solution before things get out of hand.
The room was absolutely filled, speaking to the urgency of the topic.
One effort, perhaps the most obvious one, revolves around scientists working with satellite operators and builders to help them create spacecraft that mitigate interference.
For instance, the International Astronomical Union (IAU) recommends keeping satellite brightness at magnitude seven or lower. If you’re unfamiliar, in this centuries-old system, smaller numbers indicate a higher brightness than larger numbers. So, magnitude zero is 100 times brighter than magnitude five, and the full moon has a magnitude of around -12.6, while the sun lies around -27.
“Above this [magnitude 7] limit means that the satellite will most likely saturate the pixels, making it impossible to recover the data underneath the satellite,” Jeremy Tregloan-Reed, a member of the IAU Centre for the Protection of the Dark and Quiet Sky from Satellite Constellation Interference, previously told Space.com.
Rawls said that, when a satellite’s brightness is around four or five, it can cause considerable interference — especially when many spacecraft with this brightness level are present in an image. “You start to get this crosstalk effect,” she said, “where you get these extra streaks kind of parallel to the main streak.”
“The whole object, really, for [Rubin’s mission], is to discover something really new, something that’s going to blow our minds, not something that’s been known for decades before — a new class of object, a new kind of thing — and discovering the unexpected, the needle in that haystack is made much more difficult by having a whole lot of foreground systematic errors,” said Anthony Tyson of the University of California, Davis.
The responsibility of satellite companies
“There is a possibility that we may be able to coordinate with some of the operations’ brightest satellites to actually have them reorient their hardware or adjust their attitude, as they say in satellite speak, so they would not glint as brightly where Rubin happens to be pointing at night,” Rawls suggested.
Some headway has been made on that front, but historically, there has been a bit of friction. All companies haven’t been adhering to the magnitude guidelines, for instance. Most notably, the BlueWalker 3 satellite, operated by the Texas company AST SpaceMobile, was over 400 times brighter than the recommended magnitude seven.
More recently, Connie Walker of the National Science Foundation’s NOIRLab said the Rubin team has been in contact with about 16 different satellite companies. “Some have been more forthcoming in talking with us than others,” she told Space.com. She also said, however, that she’d just been speaking at the conference with the company Reflect Orbital, which appears to be receptive.
This is a great sign, seeing as Reflect Orbital’s ultimate hope is to put over 50,000 satellites in Earth orbit by the year 2035. Why? Well, to place mirrors around our planet that can reflect sunlight, so that this sunlight can be sold as a service to light up different parts of Earth on demand. “That’s going to be challenging for astronomy,” Walker said.
“It’s essential for satellite operators to publicly share where they are and, ideally, how they’re oriented, so that we can understand when it’s going to be a problem for us,” Rawls said.
Walker also explained that SpaceX has given the team permission to test satellite reflectiveness with the materials the company uses to make its satellite constellations. “They give us an old one, and we can do some modeling on that.”
But at the end of the day, despite satellite companies being open to conversation at times, “it’s not a full-scale charge on that,” Walker said. “Their priorities are as a business, but where they can, they help out.”
Ironically, there appears to be a way for scientists to help satellite operators by giving them the satellite-streaked images that aren’t great for astronomical purposes. “We have been providing data to a dozen different companies or so, including SpaceX and Planet Labs,” Walker said.
“I think it is a nice ancillary thing that we can offer,” Rawls told Space.com, “because we’re not redacting any of the pixels, ultimately.”
Maybe the burden falls on scientists
This brings us to option two: Find a way to optimize Rubin data analysis so satellite streaks don’t impede the incredible levels of data it’ll be revealing. And, according to scientists speaking at the conference, it looks like those impediments could be pretty sizable.
Consider how satellites reflect sunlight while also circling our planet very quickly.
“They orbit fast — once every 90 minutes in low Earth orbit — and so you get a bright streak. Sometimes, it’s a dashed line, sometimes it fades in and out, sometimes it’s really bright, sometimes it’s skinny, sometimes it’s short,” Rawls said. “It’s actually necessary to be intrinsically fainter as you go to higher altitudes for the impact on Rubin’s camera to be about the same. This is counterintuitive, because further-away things are, in fact, dimmer, but further out, things in orbit move slower, so therefore they linger.”
To illustrate the potential negatives of this, Sarah Greenstreet, lead of the Rubin Observatory’s Solar System Science Collaboration’s Near-Earth Objects and Interstellar Objects working group, explained what could happen to solar system science conducted by Rubin amid a satellite frenzy.
For context, the observatory is expected to discover five million new solar system objects by the end of its 10-year-long survey. It’s also going to be conducting what’s called a “near-sun twilight, near-Earth object micro survey, which will look at objects very close to the sun.
“We’re going to have potentially hundreds of these moving object detections, and then we take many, many images, so you have to be able to tell which dots connect to which dots moving in which direction between all of these hundreds of objects in each of these images,” Greenstreet said during the conference, pointing to a visualization of this picture. “If we add artificial satellites into this [picture], you now have a whole bunch of other moving objects.”
As Greenstreet explained, in order to detect a moving object with high certainty, Rubin will need to capture four pairs of images on four nights within 15 days. But what if even one of those eight images were tainted with a satellite streak?
“We fail to discover that moving object,” she said.
She also pointed out a study from 2022 in which the authors simulated what kind of impact 40,000 satellites around Earth at less than seven magnitude — again, the recommended brightness level — would have on Rubin’s mission, known as the Legacy Survey of Space and Time (LSST).
The conclusion was that between 10% and 30% of the main LSST fields would contain streaks. And if you put those satellites at lower altitudes, up to 50% of the exposures could have streaks present.
Study authors also found that, during twilight hours, every exposure taken will likely have at least one streak in it with these circumstances. About 15% of near-Earth objects expected to be found with Rubin could go missed because of satellite interference. “It’s not huge, but I would be happy if it was much lower than that,” she said.
“One caveat to this is, of course, that these numbers rely on satellites being in this range less than seven magnitude,” she added. “If they’re brighter than that, then all of these numbers are quite a bit worse.”
To Rawls’ point about working with satellite operators, Greenstreet also explained that knowing where the satellites are before pointing Rubin at a certain section of night sky — sacrificing about 10% of LSST time — could cut the number of main survey fields with streaks in half. There is also hope for the twilight survey specifically, because more images are taken. That means there’s a higher chance the necessary four pairs are achieved with backup images.
“We’re really worried about phenomena that would happen in a single exposure but may not appear in the second exposure,” Bianco said.
“There was, in fact, already a claim for the discovery of a [gamma-ray burst] at z = 11,” she said. (“Z” is a measurement of redshift, how much light has been stretched on its journey to our telescopes. A redshift of 11 indicates an incredibly old light source — one that was emitting just 400 million years or so after the Big Bang.) “It was revealed to be a streak from a satellite.”
Rawls says her team is working on a database that labels where streaks and glints could fall due to satellite interference when looking through Rubin data.
The idea is to let scientists know when they’re viewing objects in a zone known to have satellite streak occurrences, “so you have a better understanding of maybe where you would want to be cautious about being like, ‘Oh, we found an amazing new supernova, never-before-seen.’ And then you’re like, ‘Well, it was in that streak area … maybe that’s actually just a glinting satellite,” she said.
“There’s this idea from Chris Stubbs of using different ground-based small telescopes at different places, right along the ridge, where all these telescopes are in Chile, and to look at the effects of these satellite trails moving across. But I don’t think that it’s reasonable or cost effective to have multiple LSSTs, for example, looking at different angles,” Tyson said.
Though the conversation didn’t quite reach any actionable solutions beyond what has been discussed as already in progress, like Rawls’ work, there was one clear message potently felt across the room — one of urgency.
“This is now a working observatory,” said Bob Blum, acting director for Rubin Observatory Operations, said.
“Every night — every minute — is precious.”
