SpaceX has promised to redesign its upcoming megaconstellation of Internet satellites so that they cannot hurt anyone on Earth.
In a meeting with the FCC last week, SpaceX said that the vast majority of its Starlink satellites will no longer contain dense metallic components that could survive reentry and endanger people on the ground. “No components of…the satellite will survive atmospheric reentry, reducing casualty risk to zero,” SpaceX wrote in a letter to the FCC after that meeting.
In December, IEEE Spectrum reported that while each individual Starlink satellite had only a tiny chance of hitting someone after falling back to Earth, the aggregate risk of one of SpaceX’s nearly 12,000 satellites causing an injury or death was 45 percent every six years.
Casualty risks aside, having dead satellites fall back into the atmosphere is generally regarded as a good thing. Otherwise, they can stick around for a long time. For example, although Vanguard 1, America’s second satellite, stopped communicating with Earth six years after it was launched in 1958, it still orbits the Earth today as a piece of inert debris.
Vanguard 1 has since been joined by over 20,000 other pieces of space junk larger than 10 centimeters in size, as well as hundreds of thousands of smaller pieces of orbital debris. Any one of these could damage or destroy working satellites, crewed spacecraft, or even the International Space Station, potentially creating a dangerous cascade effect known as the Kessler syndrome.
Satellite operators are now required to ensure that dead satellites linger in their orbits no longer than 25 years, and that the likelihood of a human casualty at their end is no more than 1 in 10,000. Satellites in distant geostationary orbits are usually boosted further out into disposal orbits that are unlikely to pose navigation hazards to spacecraft, while those in low Earth orbits (LEO) are directed, or naturally decay, into the atmosphere.
The smallest CubeSats usually burn up completely on reentry, but fragments of larger satellites like SpaceX’s Starlinks, which are about the size of a Tesla Model 3, can survive. The FCC estimated that falling Starlink satellites would have resulted in as many as 500,000 separate objects reaching Earth’s surface every six years.
As originally designed, nine pieces of each Starlink satellite, including thruster parts, reaction wheels used for maneuvering, and silicon carbide communications components (probably mirrors for intersatellite laser links), would have reached the surface, SpaceX acknowledged. Some of these would have had enough energy to injure or kill, although the company contended that even the flimsiest of shelters would have offered some protection.
In early February, OneWeb, another company intending to build a megaconstellation of LEO Internet satellites, cited Spectrum’s casualty analysis in an FCC filing criticizing SpaceX’s plans. “The Commission should carefully consider the implications of approving an…orbital debris plan that fails to provide…an aggregate analysis or materially improve upon [SpaceX’s] already troubling casualty risk profile,” it wrote.
Late last month, the FCC itself asked SpaceX to clarify whether falling Starlinks could be directed to ocean areas, away from human populations. It also wanted Elon Musk’s company to provide additional, high-fidelity studies of casualty risks during reentry.
In its reply last week, SpaceX said that while its satellites could not be reliably directed to land in the ocean, that no longer mattered. “After extensive research and investment, SpaceX has now developed a system architecture that will be completely demisable,” wrote one of SpaceX’s lawyers.
SpaceX has always claimed that the design of its Starlink satellites would evolve. Now, it says that at the most, the first 75 Starlinks will include an iron thruster and steel reaction wheels likely to survive reentry. Any built subsequently will “use components that will demise fully in the atmosphere,” wrote the company. No satellites at all will be deployed with the silicon carbide components described in its initial filing.
SpaceX did not provide a reason why the first 75 Starlinks would use a semi-demisable design. It is possible that these satellites are already partially or fully built, especially considering rumors that the first operational satellites might be launched as soon as May. SpaceX launched two prototype Starlink satellites early in 2018.
Nor did the company say what would replace the silicon carbide mirrors in its satellites’ laser links. The space lasers are an innovation intended to allow thousands of Starlinks to share and network orbital Internet traffic in order to maximize bandwidth and minimize delays for users.
The FCC has yet to respond officially, but satellite experts consider SpaceX’s claim of eliminating all danger of human casualties to be very bold. “In principle, I don’t know if you can ever claim zero risk,” says Raymond Sedwick, an aerospace engineer at the University of Maryland's Center for Orbital Debris Education and Research. “The engineering challenge is making sure that as the satellite starts to reenter, the outer portion breaks up and exposes parts that would otherwise be protected, so they can experience high levels of heating.”
Such a redesign, combined with replacing iron, steel, and titanium components with less resilient materials like aluminum, could significantly reduce the chance of any fragments reaching the surface, according to John Crassidis, a professor of mechanical and aerospace engineering at the University at Buffalo. “But there’s nothing to say that aluminum won’t survive,” he says. “Anything can come through the atmosphere if you hit at just the right angle. If they’re guaranteeing it’s not going to cause an issue, then I’m going to have call BS on it.”
However, the simple act of SpaceX claiming that its satellites will burn up completely now puts pressure on rival megaconstellations to follow suit. Regulators are increasingly likely to want assurances that what goes up, no longer has to come (all the way) down.