SpaceX's Starlink satellite constellation has been worrying world astronomers for years. Now, a new study published in The Astrophysical Journal Letters by astronomers at the Zwicky Transient Facility (ZTF) at Caltech's Palomar Observatory shows how bad the problem is - and how much worse it is likely to get.
Since 2019, SpaceX has launched batches of low-Earth orbit (LEO) satellites to build a constellation network that will provide Internet access from any point on the planet. There are already close to 1,800 satellites in orbit, with plans for tens of thousands to eventually be deployed.
The problem is that these satellites reflect light from both the Sun and the Earth over the horizon, making them visible in the night sky even to the naked eye. While this may be a spectacular sight, or even a nuisance, for terrestrial observers looking up at the stars, it is a serious problem for terrestrial astronomical observatories.
When the human eye stares at the night sky and sees a Starlink satellite, it looks like a particularly bright star moving across the sky. It does so only because our brain is constantly updating the visual data it receives from our eye to create that movement.
However, this is not how telescopic observatories work. In order for these observatories to take pictures of distant objects, they have to collect a lot of light over a long period of time and then create a single image out of all that light. This is how they can capture such vivid images of objects that our eyes cannot detect on their own.
However, when a Starlink satellite passes through the view of these observatories, it does not appear to be moving. It appears as a light line across the image, blocking what is behind it. This is especially true for images taken during the twilight hours at dawn and dusk, which are essential for identifying objects that may be between the Earth and the Sun, such as. terrestrial asteroids.
"In 2019, 0.5 percent of twilight images were affected, and now nearly 20 percent are affected," Przemek Mróz, lead author of the study and a former Caltech postdoctoral fellow, now at the University of Warsaw in Poland, said in a statement.
"We do not expect Starlink satellites to affect non-twilight images, but if the satellite constellation from other companies enters higher orbits, this could cause problems for non-twilight observations," Mróz added.
Twilight observations are especially important for planetary defense
The reason why twilight observations are so important is that most of the objects in the solar system orbit the Sun on one level. Some objects have far more eccentric orbits - comets, in particular - but all asteroids are remnants of the Sun's original growth disk during the formation of the solar system.
So asteroid orbits are not very eccentric and can sometimes cross our own and hit the planet. Usually this is not catastrophic as the meteors (any space object that enters our atmosphere is defined as a meteor) burns up in the atmosphere and simply creates a light show in the form of shooting stars or more spectacular fireballs.
These are the major meteors we need to worry about, and it is important that we track as many asteroids as we can that could threaten life on Earth. This is where twilight observations come in.
If the Earth is between an asteroid and the Sun, there is no problem tracking it. It will be visible deep at night and will even reflect sunlight, making it easier to spot. But that's only half the sky. The other half is completely overwhelmed by the sun's light, making it impossible to observe anything during the day.
This makes it almost impossible to spot an object orbiting the Earth and the Sun, as many asteroids do. It also means that we lose track of any object that can orbit further away from the Sun than we do if the relative position of the object is only visible during the day.
So how do you spot these objects? Fortunately, you can point a telescope at the position of the Sun on the horizon just after sunset or just before sunrise and get a pretty clear view of these objects.
This allows us to keep track of the position of terrestrial asteroids and project their orbits into the future, thereby identifying possible threats far enough in advance to do something about them.
But they are also small points of light in the sky, so telescopes have to take very long exposure pictures over several days in order for them to appear and for us to determine their speed and orbits. And now about 20% of these images are compromised by SpaceX's Starlink satellites orbiting them.
Tom Prince, Ira S. Bowen Professor of Physics, Emeritus, at Caltech and co-author of the ZTF study, notes that the problem is not that bad yet.
"There's a small chance we'd miss an asteroid or other event hidden behind a satellite strip," Prince said, "but compared to the impact of the weather, such as a cloudy sky, these are pretty small effects for ZTF."
Unfortunately, the Vera C. Rubin observatory, which is under construction in Chile, is far more sensitive than the ZTF and will be far more susceptible to satellite streaks during its observations. With tens of thousands of satellites planned in this decade, it will be disastrous for the work of the new observatory.
"There is nowhere you can go [the Vera Rubin] and point it out without having one of them in sight, ”Tony Tyson, an astronomer and physicist from the University of California Davis, who works at the observatory, told Vox in December 2019.
What's the big thing about streaks?
So what if there is a white line across an image? Is not the rest of the image still perfectly usable?
Yes, the rest of the image is usable, but everything that was behind that streak will be missed. It may appear on a subsequent image, but without knowing its position on the first image, you can not measure its speed or trajectory.
If one out of every five images can be compromised, then can we not just take more images to compensate? Yes, and we will have to do that, but one in five images will be compromised by Starlink satellites when there are only about 1,800 satellites in LEO. SpaceX already has approval to put 12,000 Starlink satellites into orbit and plans to put another 18,000 satellites on top of it.
In addition, SpaceX is not the only company that wants to create satellite constellations. Companies like Amazon and OneWeb are planning their own constellations.
What happens when there are 100,000 satellites in LEO, all of which sweep across the sky in the twilight and destroy these vital observations? So many are likely to be deployed within the decade between the various satellite companies. With so many satellites in LEO, every single observation taken in the twilight will be affected. All. And it will not be a single line, it can be dozens.
Can technology help?
It is possible that machine learning algorithms can help compensate by removing these dashes, but it has the potential to effectively erase the data and create objects that are not there, or delete objects that are.
It is also possible to coordinate observations with satellite operators so that no observations are made when satellites pass overhead. It's easy enough for now, but will it still be easy when there are hundreds of thousands of satellites in orbit?
Could Starlink satellites be made of a non-reflective material? Possibly, and SpaceX has already put a sunshade on the satellites and placed their solar panels to reduce the amount of light reflected back to Earth. This has reduced their brightness significantly, but not enough to avoid streaking across observatory images.
Although SpaceX took all satellites out of orbit and scrapped its Starlink plans, there is nothing to stop a Russian or Chinese company from doing something similar. It is not the case that a constellation of Chinese Internet satellites is getting better.
So what is the solution? That is the big question and no one has come up with a satisfactory answer yet. Astronomers have approached the UN for help in regulating these satellites to protect terrestrial observation. However, this will not prevent these companies from implementing their satellites. SpaceX alone puts one hundred or so satellites into orbit each month. After all, there is money to be made.
"The mechanisms of [international] law that could have allowed us to avoid this, the wheels are turning so slowly that when we reach any consensus on a political solution, it will all be over, "said John Barentine, an astronomer and director of public policy. at the International Dark-Sky Association. "I just think it's a numbers game that astronomy probably can't win."
What it means to have the night sky taken from us
Despite all the alleged benefits to humanity - and do not get me wrong, it's undoubtedly a good thing to bring true broadband internet speeds to developing countries and rural areas - no one should forget that SpaceX, Amazon and others do not. of altruism.
They do not bring free internet service to the world. They will pay for it and probably make money on their investment in their satellite constellations. Making money is the whole point of the business, twilight sightings of objects near Earth or just simple stargazing cursed.
"Satellite constellations ... hit the core of humanity's scientific and cultural relationship to the night sky, affecting millennia-old celestial traditions and cultural practices across all nations around celestial cycles and the night sky," Aparna Venkatesan, James Lowenthal, Parvathy Prem, and Monica Vidaurri wrote in a study published in Nature Astronomy in November 2020.
They also advocate an international consensus on protecting the night sky for future generations: "This approach calls for a radical shift in the policies of international regulatory bodies towards the view of space as a global common space of an ancestor, containing the heritage and future of humanity's scientific and cultural practices. "
With little regulation and no international rules for rolling out satellite constellations, it feels like screaming into a void, a void that will soon charge you money for the privilege of doing so. No matter what happens, we should never forget what is being taken from us and why.