[mindwok]

How feasible is it though once the network reaches a huge size? Starlink satellites are tiny, and they've been deploying thousands of them over the last few years. I imagine it would take enormous resources to shoot them down, especially if the US does treat them like a strategic resource and adds more for redundancy.

[lucubratory]

The huge size of the network is itself a risk. Kessler syndrome is something everyone is currently trying to avoid, but if you wanted to intentionally induce it you could just start launching giant payloads of tiny ball bearings into their orbits, or take down enough of them that the shrapnel becomes equivalent to that anyway. Starlink is low enough that the debris from even a full Kessler syndrome cascade will deorbit very rapidly, but we're still talking a 3-5 year timeframe, not months, and trying to rebuild capacity in that period will just worsen + extend the problem.

[mlindner]

This is something commonly misunderstood. Kessler syndrome is a statistical process that happens over many years. It is not a sudden cascade like is seen in movies like Gravity. Statistical processes are not what militaries are interested in.

It's actually thought that Kessler syndrome is kind of already happening right now, which is why there's a lot of push right now to try to de-orbit the very large pieces of debris, so they can't act to form further debris.

[lucubratory]

It happens as fast as it happens. Any actual projection would depend on the specific orbits, masses, volumes, materials, and numbers of satellites - Starlink's orbits have a lot of satellites now. There's a very big difference between "everyone trying to avoid it" and "one of the world's largest space programs trying to cause it" in terms of how much we should be worried about it happening for any given orbit in the near term.

The reason it's a scary outcome is because it's an exponential. It can look like an isolated incident or incidents, then the next day be not practically stoppable.

[mlindner]

There's already significant amounts of debris that transits through Starlink's orbit.

And I'm telling you, your "image" of what this looks like is just incorrect. The kessler syndrome is likely already occurring. Yes creating more debris will make it happen more, but it's not like lighting a match to a pile of tinder.

And it's not in fact exponential in the sense that people commonly imagine when they hear that. It's an exponential that's very close to flat, i.e. an exponential with an exponent barely above one. Given enough time, yes it can destroy all satellites in Starlink's orbit, but it's not on time scales that's relevant to a war.

[robbiep]

Why can’t every satellite have a small rocket/firework like thing on the back pointing out to the expanse and if the power goes out or it doesn’t receive a signal from the dead man’s switch for long enough then it ignites? Even with a big mass you don’t need to give it much of a shove downwards in a zero friction environment to speed the de-orbit period up.

I’m sure I’m missing something but it just seems like a no brainer to make the deorbit process speed up with something relatively failsafe, as opposed to hopefully/maybe saving a bit of fuel to push it that way eventually

[FeepingCreature]

Satellite failures often involve uncontrolled spinning. So you've turned a piece of debris in a known stable path that will eventually deorbit into a piece of debris on an unknown but potentially energetic orbit.

Satellites do have deorbit thrusters, but they're a lot more deliberate. I think Starlink have a whole separate remote controllable system just for deorbit control.

[mlindner]

1. Most significant satellites already have propulsion, only the smallest do not (which is a side problem). The problem is that when satellites fail, the guidance will often fail too.

2. Engines and fuel are heavy. Including one on the smallest satellites may take up the entire mass availability that would go to the instruments, leaving the satellite with nothing to do. There are people working on this, one idea is including a small air canister and a balloon. At end of mission the balloon can be inflated which greatly increases the drag of the satellite causing it to de-orbit relatively soon.

3. As a side note, you don't want to fire "out to the expanse" as that won't de-orbit your satellite. It'll just "twist" the orbit, lowering the perigee and raising the apogee. Primarily it'd just waste fuel. To de-orbit you want to slow down, so you need aim "backwards" along your orbit's path.

4. With a big mass you need an equally large amount of fuel as what determines your ability to de-orbit is the satellite mass, your engine's propulsion efficiency, and the amount of fuel you have.

5. The problem isn't existing satellites. The problem is very old defunct satellites and rocket bodies and existing small debris. Many rockets used to (and still do to some extent) leave large pieces of themselves in orbit.

[wongarsu]

Pretty feasible for anyone who has enough ballistic missiles to target about 5000 targets, or is willing to invest a couple billion into stocking 5000 overpowered fighter-launched missiles. Starlink isn't that high up, and in military terms 5000 targets isn't that much.

The effort of getting a ballistic trajectory that peaks at 500km is a lot smaller than reaching a stable orbit of that height. And just like WWII aircraft you don't need to hit them, just produce enough shrapnel in their vague vicinity.

The biggest hurdle is the universal international condemnation you would receive for such an act

[supertrope]

Even if anti-satellite missiles are too expensive to be used to shoot down thousands of targets, the ground stations could be bombed instead. Hacking the control plane and sending de-orbit commands could be even cheaper.

[mlindner]

Starlink satellites use inter-satellite lasers and can send those signals arbitrary distances via multiple satellites. Taking out a ground station will just require routing changes and the constellation will continue to perform.

And you can't just wave around "hacking the control plane". Russia's been trying to interfere with Starlink for a while and they haven't had any long term success. And finally, even if the did somehow get access to the control systems at SpaceX, the satellites can't de-orbit quickly. It takes weeks to de-orbit, over which time they could be commanded to reverse course.

[jessriedel]

The new generation of Starlink satellites have laser connections between them (which is what the article is discussing). They can send data to the other side of the globe to a friendly country for the ground link. (That’s a less efficient use of the inter-satellite bandwidth, of course, but worth doing for war.)