The Kessler Syndrome: Space junk in orbit
onezero.medium.com
onezero.medium.com
In the past it was not practical to put satellites in low orbit for several reasons. One, because the same atmospheric drag that cleans up debris would pull the satellites out of orbit too, severely limiting their useful life. Two, because a satellite in low orbit can only see a small part of the Earth's surface at once and passes over it very quickly. You can't use stationary satellite dishes to communicate with such satellites, and establishing continuous coverage over any part of the Earth's surface would require hundreds or thousands of satellites.
Two things have happened that changed the game. One, phased array antennas are now cheap enough for consumer applications and they are able to communicate with multiple fast-moving satellites at once without physically pointing at them. Two, SpaceX has decreased launch costs to the point where it is now feasible to launch thousands of cheap satellites instead of dozens of expensive satellites, and replace them all within a few years instead of expecting them to last for decades. Starlink satellites orbit at 550km where debris lasts only a few years before automatically being cleaned up by the atmosphere. Future generations may orbit even lower where debris is cleaned up even quicker.
The non-SpaceX proposals for large constellations mostly chose higher orbits. I think it's time to prohibit this. Starlink competitor OneWeb just had a satellite failure at 1200km and that dead satellite will pollute orbit for centuries if it is not retrieved (and there is no realistic plan to do so, nor proper incentives to ensure that it happens). Heaven forbid that it collide with something and produce an ever-expanding debris cloud that can never be fully cleaned up in our lifetimes. Meanwhile, Starlink has had tens of failed satellites and they are all deorbited already or well on their way.
It just doesn't make sense to do things the "old space" way with a small number of expensive high orbit satellites. Low orbit cheap satellites are the future.
For the lower orbits I think there are some points to be made about the additional flights to maintain enough satellites, as well as the questions of what happens to all this junk as it de-orbits? If we're talking about sustainable infrastructure, what kind of new weird material/chemical build-ups are we going to be dealing with after 50 years of tens/hundreds of thousands of private satellites partially burning up and partially raining down on random spots on the planet?
Perhaps this is less of an issue than I'm imagining it will be, but I tend to think we'd be better off figuring out some kind of autonomous robot LEO satellite aggregation/recapture solution for cleaning up satellites from higher stable orbits that can operate farther and live longer. We should be allocating our launches to other projects.
Even without considering the emissions of launches, there is a significant resource energy/cost for each one and the noise pollution also takes a toll on surrounding ecosystem. If we're going to be ramping up launches to scale the mass to orbit, we're also going to be scaling those negative externalities. I think we really need to make sure the mass we're taking up is worth it and not just governed by whoever happens to have the most money to burn.
I can imagine the very near future where people are paying to launch crypto-mining satellites for memes, or some other such wasteful nonsense.
Space dust contains plenty of things that are more hazardous than anything in these satellites. According to [1] the abundances on this page multiplied by [2] 40000 tons per year, the stuff naturally hitting the atmosphere each year includes 0.3 kg uranium, 1.6 kg thorium, 3 kg radioactive isotopes of potassium, 10 kg mercury, 18 kg cadmium, 50 kg lead, 72 kg arsenic, and 120 tons of chromium, among other things. This stuff burns up in the atmosphere all day long, and has for billions of years.
As far as emissions, one Falcon 9 launch emits 300 tons of CO2 and they've been launching one or two times per month to build out the Starlink constellation. Meanwhile commercial aviation emits 2 million tons per day and is still a small fraction of the world total.
[1] https://periodictable.com/Properties/A/MeteoriteAbundance.v....
[2] https://link.springer.com/chapter/10.1007%2F978-1-4419-8694-...
A rocket launch powered by fossil fuels is comparable to an oceanic jet trip, surprisingly. So quantities aren't a big deal, yet. Elon Musk has promised to switch to carbon neutral fuel for his next generation of rockets, synthetic methane for Starship.
Firstly reusable rockets cut the amount of discarded hardware massively. With expendable launchers almost all of the hardware dropped into the ocean or burned up is from the launcher. The satellite is a tiny fraction. This is a huge win.
Secondly the total mass of satellite hardware burning up is dwarfed by natural meteorites, of which approximately 100 tons of silicates, metals and hydrocarbons enters our atmosphere every day. That’s an entire ISS every 4 days. That’s 600 times the mass of satellites the US launches per year.
The thing is Starship may well increase mass to orbit by that much. Even so, that would “only” double the natural rate of material entering the atmosphere.
Another strategy that might help, for high altitude satellites, would be to move to fewer bigger satellites, with racks for different companies instruments, and with self de-orbit capability. Treat them like orbiting data centres. Starship can certainly handle big satellites, and if launches are cheap then it could be affordable to fit such satellites with de-orbit capability. Or maybe even send a Starship to swallow it up and bring it back.
"Heaven forbid that it collide with something and produce a debris cloud that can never be fully cleaned up in our lifetimes."
At that altitude the time for natural orbital decay is not within our lifetimes by tens of thousands of years. If mammoths and saber tooth tigers littered up there it would still be a hazard to us today.
https://www.google.com/search?q=phased+array+antenna
Basically antennas that can programmatically steered without physically moving.
> In antenna theory, a phased array usually means an electronically scanned array, a computer-controlled array of antennas which creates a beam of radio waves that can be electronically steered to point in different directions without moving the antennas.
Also with the cost of sending a satellite to space about to potentially plummet (starship?), wouldn’t it just make sense to make a lot of cheap satellites that you throw up in a low enough orbit that a collision would only require a few years til the shattered pieces de orbit and burn up? We may be on the cusp of a revolution on satellite technology as the engineering work going into each individual satellite drops precipitously and the turnaround time becomes weeks not years. I could see a similar Cambrian explosion in satellite tech from cheap space access similar to when we went from mainframes to pcs. Thoughts?
International space law, like most other international regulations, is only for the small fries. Any country capable of inserting into geosynchronous orbit with a domestic manufactured launch vehicle is usually also nuclear armed, hold a permanent seat on the Security Council or allied with a holder, and is immune to most low level sanctions. No real space power can be truly bound by treaty if things heat up because they are usually the enforcers.
That's essentially Starlink, though for other reasons. But lower orbits see much less of the earth. Earth has a radius of about 6000km, at a low-earth orbit of around 600km you are way too close to have a good view, so you need a lot of satellites to cover the globe. There are also useful special orbits like geostationary orbits (where the satellite is always in the same spot relative to earth's surface) or sun-synchronous orbit (the shadows always look the same because you are in the same spot relative to the sun) that are much higher up.
However with cheaper satellites, cleaning up dead satellites or larger chunks of debris also becomes cheaper, we just need to figure out who has to pay for it. Right now we are kind of stuck with everyone being affected but no one feeling responsible.
So, it seems a bit disingenuous for the article to make a big deal of LEO satellites.
The assumption I think you're making is all debris remains in a fairly localized decaying orbit, which is not the case when a high energy collision occurs. The debris will fan out in two radial smearing patterns that loop back around at the collision point. Some pieces will hit more atmo for longer periods while others will achieve higher altitudes for a time and those can debris fields last longer. Should a LEO cascading collision event happen we would see lots of debris fields reaching higher orbits and a non-zero risk of those satellites being hit too.
* we do not have any kind of technology to get stuff back from orbit. Anything that is there and anything that we put into orbit will stay there, passively, until the laws of physics take care, the only exception being satellites that are made to plunge back into the atmosphere with their left over propellant. We cannot even hope to get meaningful amounts of plastics back from the surface of the oceans which is a similar problem only many, many orders of magnitude more feasible.
* When you put satellites into lower orbits you'll need more of them to obtain the same coverage of ground. At the same time, the volume of space is smaller in lower than in higher orbits, so more crowding. You can use smaller, less powerful satellites, but because of atmospheric drag, their lifetimes will presumably also more limited in low orbits, meaning considerably more additional traffic due to the increased need of de-orbiting and replenishing activities.
* Should anything go wrong (and something will go wrong sooner or later, as indicated by the fact that we already have to close keep tabs on the course of the ISS), a collision will spread parts into all directions. One would imagine that in LEO roughly half of the sky is occupied by nearby Earth, but the other half is pointing sideways or up from the Earth, so if two 100kg satellites smash into each other you still have like ~100kg of new debris that will fly into directions not prone to natural fast decay. Of course this is just a rough estimate on which I'd like to be proven wrong.
* Low orbit and increased number of objects also mean more interference with astronomy. I cannot believe when folks claim that now that we 'have AI' we can just 'use AI to clean up astronomical imagery'. Our entire ability to put stuff (and people) into space at all is predicated on many, many millennia of gazing up to the skies and trying to figure things out. No astronomy, no going to space. When you take an image and there's an object blocking the view, or worse adding a very considerable amount of reflective light that saturates your sensors, no amount of 'AI' will get back that info to you because the data is, essentially, just not there.
The question isn't "Will there eventually be some sort of incentive to clean up space junk?" (which would lead to some sort of free-market amelioration if not solution), it's "Can this be prevented?" Free market solutions to cleaning up commons often come too late and are imperfect.
The one saving grace here is that other living things won't be harmed; so there will potentially be less ripple effects.
In this case, the commons, if Kessler's predictions are correct, will become unusable much more quickly than has been the case with other commons in the past. That creates a much stronger free market incentive to clean the commons up.
This is just like overfishing - every fisherman is motivated to catch more fish himself, rather than to leave the stock to grow, so it gets fished until the population collapses. And, they keep fishing right into the collapse, they don't even stop when their haul declines, just invest more in becoming more efficient at depleting the commons faster.
The likely result is that everyone will keep putting up satellites, making minimal safety measures for Cover-Your-Ass publicity, and then it'll cascade collapse one day soon.
No, they aren't; that's just the point. When the commons becomes unusable, nobody can exploit it any more. And if that happens quickly enough, the short term benefit of exploiting it is not sufficient to outweigh the long term cost of it becoming unusable. That seems to me to be happening in the case under discussion. And once it happens, there are only two options open: (1) figure out how to not exploit the commons, i.e., all the parties who want to use it have to come to agreement on how to use it sustainably; or (2) nobody gets to use it at all. If there are significant benefits to be gained by using the commons, as there certainly are in this case, there is a strong incentive for the parties to figure out how to do option #1 instead of just giving up and taking option #2.
> they keep fishing right into the collapse, they don't even stop when their haul declines, just invest more in becoming more efficient at depleting the commons faster.
The reason this works for overfishing (where it has indeed happened) is that fishing is not the only source of food. Once overfishing has completely depleted a fishery, people just switch to other food sources.
In this case, there is only one Earth orbital space, and it doesn't seem like there is any viable alternative to providing the kinds of services that are being provided by satellites. But if that turns out not to be the case, if some of the players are able to find alternatives, then yes, that would increase the likelihood that Earth orbital space would not be cleaned up by the individual market participants.
I'd really need to see an example where this actually happened, because your pitch in the first paragraph is utterly unconvincing.
The incentives for individual players/nations are still to use earth orbital spaces more - the more each can put up there, the more they reach their goals, and prevent competitors from reaching their goals.
Of course there is the global motivation to use it better in the long term, but this is in conflict with the individual near-term incentives - the entire point of the Tragedy Of The Commons is that everyone DOES see that it is in their interest to preserve it. Yet, before the tipping point, it is ALSO and in the short term MORE advantageous for each player to exploit it for their own advantage.
Russia just demonstrated as much last week. It is totally in their long-term interest to preserve it, and even in their immediate interest to not endanger their cosmonauts or their investment in the ISS. Yet, they went and scattered thousands of projectiles all over the space.
So, the assertion that this is somehow different needs more evidence. AFICT, everyone sees the need, yet everyone is still tossing up more satellites without a plan or agreement to fix it.
Where is the immediate (not long-term) motivation to not use it, and where is evidence that it is being fixed? It is already quite late.
That's one factor, yes. Another is the time horizon for gaining the benefits of using the commons. This is another respect in which the Earth orbital space scenario is very different from fisheries.
When fisheries are overfished, the benefit to the overfishers occurs over a very short time horizon. They have to either eat or sell the fish fairly quickly, before the fish spoil. So as long as they think they can go out and catch enough fish on their next trip, which typically lasts on the order of days, to compensate them for the effort and make their expected profit on top of that, they have a short term incentive to continue to overfish. In short, the time horizon for the short term incentive is on the order of days for overfishing.
A company deciding whether to launch additional satellites has a very different short term time horizon. A satellite launched now will take years to pay back the investment made in it. So the company has to be able to forecast that that satellite will remain in service, and not get impacted by extraneous factors like orbital debris, for a period of years after it is launched. Plus, the company has to invest resources in planning for years before the satellite is even launched. So if the company projects that in, say, ten years, the risk of satellites becoming impacted by orbital debris is unacceptably high, they have an incentive to do something about it because otherwise they will lose the investments they are making now in satellites that will have to deliver them profits in that time frame.
> Russia just demonstrated as much last week
What Russia did last week had nothing to do with short term incentives for private exploitation of Earth orbital space using satellites. Indeed, the main point of their demonstration appears to have been that they can deny the use of Earth orbital space to anyone with anti-satellite missiles if they so desire. If they were to carry out such a threat, it would prevent private companies from exploiting this commons.
E.g., fisherman have massive capital investments in their boats, and also will not make money unless it is long-term sustainable.
There are already the Space Debris Mitigation Guidelines," which was put together by the Inter-Agency Space Debris Coordination Committee (IADC), and unanimously adopted by the member states of the United Nations Committee on the Peaceful Uses of Outer Space (COPUOS).
Russia straight up violated them. Yes, this may prevent private companies from using space, but it would also prevent them from doing so, and as a nation-state, they have an even longer term need & incentive.
With all these incentives you outline, major steps should have already been taken by private companies restraining themselves. Can you point to any?
I'd love you to be right, but I don't see how it'll happen.
Cool essay by the way. Learned something new.
Or are you saying you think companies will invent/build more resilient satellites in response to junk? I am not an expert but I wonder how feasible that would be given the weight constraints and extreme speeds involved.
Perhaps that would take thousands of years....
Humans are organisms. Humans go to space, and will go there more with it getting cheaper. Orbital space is an environment. Debris collisions are an ecological concern for spacefaring organisms.
Less snark: I think definition (or more accurately, your expansive interpretation of it) misses the spirit of the way that word is generally used.
Not to mention all the communications systems we have in orbit that we rely on for our economy which would result in significant loss of logistics / economic efficiencies and straining growing social political issues. All of which means higher likelihood of damage to local and regional ecologies
Highly recommended!