Hacker News Reader

Comment by xoa | Hacker News Reader

While I know Scientific American is a fairly mainstream targeted and softer science publication, this

>And the problem is now poised to get much worse because of the rise of satellite “mega constellations” requiring thousands of spacecraft, such as SpaceX’s Starlink, a broadband Internet network.

really could have used a LOT more qualification particularly since it's become a major recent talking point. There is one ultimate sure-fire way to reduce space junk: launch stuff to very low orbits. There is still some atmosphere (varying with heating and other factors) up a long ways in "space", and it's only above 600km or so that orbital drag becomes negligible enough that lifetimes really stretch out. The ISS for example requires regular reboosts or it would decay back into the atmosphere.

SpaceX focusing on economics has made it feasible to start planning constellations of comm sats that are low and very low earth orbit, with the understanding that inherently their lifetimes will be measured in single digit years. But that's ok since they can be replaced so cheaply. This is much better, not worse. Even if they go offline or got hit, the debris would have a very restricted lifetime. The media should do a better job of conveying how cheaper $/kg to LEO opens up a lot of new possibilities in what regulations are feasible and how we think about the basics of satellite design. A simple "Mega constellation bad!" is all wrong, and that's not necessarily going to be intuitive to everyone.

View on HN

Here's an awesome diagram of all tracked space debris from 1959 to 2021 showing the constant cleaning by drag happening at lower orbits. It's the higher ones we really need to care about.

https://www.reddit.com/r/space/comments/ld4vlq/gabbard_diagr...

Was waiting for that time in the '00s when China conducted an anti-satellite missile test. Did not disappoint.

Let's not do that again.

That's an amazing visual. Really helps to inform the debate. (Watch on full screen).

Can someone explain the reason for the mirror across the diagonal? Is there some altitude/velocity ratio that causes debris to spiral outward vs inward?

The mirror line would be a circular orbit, this plots each tracked orbiting debris with two dots, the blue dot being the lowest altitude of the orbit and red the highest.

I'm not an expert BUT...

I suspect the diagonal line between the colors represents the duration of a stable orbit at that altitude. Anything with a more elliptical orbit will drag when it's in the closer phase and decay. Anything with a lower orbit is already slowing to a tighter orbit (and more drag on the denser atmosphere).

So... this means we could theoretically handle the outer orbit space junk problem by putting more gas in those orbits!

This would increase the "drag" and deorbit debris in higher orbit much faster. My physics intuition says it would affect smaller things much more than bigger, which is what you'd want.

If it can ever be practical, I don't know, but conceptually that works, right?

Wow. Am I interpreting that correctly? Lower orbit stuff falls out of orbit pretty fast while the higher orbit stuff just kind of stays there.

Basically, yes. The atmosphere is a gradient, and the nearer you are, the higher the drag, and the faster you slow down.

I think you're interpreting it a little wrong.

While it does look like 500km is about where things fall out quickly, it also looks like there's a sweet spot of altitude to orbital period ratio where stuff just stays there. Outside of that sweet spot stuff still falls out fairly quickly.

I'm not 100% sure I'm interpreting your comment correctly, but higher orbits will always decay more slowly, all other things being equal. There's just far less atmosphere to bleed off the energy.

Edit: I think I now understand what you mean, but I suspect you're not realizing that every satellite appears twice on the chart: once in blue, for its periapsis (perigee, i.e. low point of its orbit), and once in red, for its apoapsis (apogee, i.e. high point of its orbit). You're probably noticing that some of the red dots, at high altitude, are decaying quickly, but that's because the corresponding blue dots are at low altitude.

Note also that orbital period is a monotonic function of altitude. More precisely, the square of orbital period goes with the cube of the semi-major axis (the long axis of the orbital ellipse).

I see what you're saying, and you're right, I am mistaken. Thanks!

Thank you for showing me the coolest graph I have ever seen.

>> launch stuff to very low orbits. >>Even if they go offline or got hit, the debris would have a very restricted lifetime.

That approach can help, but the problem persists. Lower orbits are much smaller than higher orbits. Focusing megaconstellations into a narrow 100-110km band is asking for trouble. The kessler syndrome could occur in a very narrow band, quickly rendering even short-lived satellites uneconomical. The narrower the band of orbits, the more likely and more aggressive kessler becomes. It is one thing to loose a few sats every year, very much another to have your entire constellation wiped out every year.

If you are low enough, Kessler is simply not possible because square-cube scaling means that the fragments deorbit in less than a single orbit.

> Lower orbits are much smaller than higher orbits.

Not "much".

A 1200 km circular orbit is ~10% longer than a 550 km one, so a max of ~20% more "room" in the larger orbital shell.

The quicker de-orbiting of debris from the lower shell more than makes up for this IMHO.

Not longer, three dimensions. Low orbits range from about 110 to 200km. The next range, from 200 to a few thousand km is vastly larger in three dimensions.

You can't have a stable orbit at 110km. The atmosphere is much too thick. You'll need at least twice that, and likely more to maintain vehicle attitude unless you have a very smooth drag profile, which is uncommon in orbital craft.

  Altitude  Lifetime
  200 km    1 day
  300 km    1 month
  400 km    1 year
  500 km   10 years
  700 km  100 years
  900 km 1000 years

Ha - and the altitude of the space station is 408 km so most of the junk should clear out in a year? That sounds too easy.

The space station has to add energy almost weekly to overcome drag. If it didn't it would likely come down within a couple years. When they want to throw trash away, they literally just drop it off the side and it burns up eventually: https://gizmodo.com/iss-ditches-2-9-ton-pallet-of-batteries-...

Would that stuff cause a huge fireball in the night sky? 2.9 tons is a hell of a lot of mass... amazing to think it will just burn up in the atmosphere.

I'd say it would burn up pretty nicely. Its not gonna be one homogenous lump of stuff that holds together on re-entry, its a bunch of smaller "things" (battery cells) bundled up and held together with an external shell.

As soon as that shell burns through its going to break apart and each individual part will start burning up.

Yeah probably most of the mass is aluminium sheet which melts quite easily hence Starship and the MiG-25 switching to steel, and also can burn quite well - it was the fuel in the shuttle's boosters.

I don't think it is to worry about, 1 cubic meter of iron weighs 7.8 tons for example... it is not a huge mass in the end.

Because it is really that easy. Space-junk in low-earth orbit is not really a concern, and alarmist articles about space-junk in connection with projects like Starlink are just click-bait that rely on the general public's ignorance of orbital mechanics and space.

Medium and high orbits are a different story. Space junk is absolutely a real concern here.

Also, take out SpaceX from the name and add all the competitors (potentially lower quality) that are starting or about to get into the space and you can understand the concern.

edit: my concern is the long chain of the curve of low-orbit satellite companies as price continues to have downward pressure.

>edit: my concern is the long chain of the curve of low-orbit satellite companies as price continues to have downward pressure.

While this isn't entirely unreasonable, consider a few things. First, this is one case where even natural incentives line up pretty well. For the foreseeable future, no one is going to have more economic incentive to prevent the loss of usable LEO than companies that build themselves around usage of LEO. There are also virtuous spirals in the very technology that makes low cost possible, ie., the only way to reach SpaceX's targeted launch costs with Starship are to have a full reusability, and that itself cuts a ton of orbital debris (spent stages).

Second, much cheaper mass and higher cadence changes everything in space engineering, and that includes giving regulators significantly more leeway in what they can reasonably require. Things like more redundant controlled deorbit systems, requirements for lower orbits by default with higher orbits reserved, requirements for materials, and so on all ultimately boil down to how much it costs to get a kg to orbit and how regularly it can be done. More leeway there makes a lot of things easier without destroying utility, which in turn raises the chances we can make solid systematic changes.

Not saying there isn't plenty of room for error overall, or that regulators shouldn't be thinking about it too. But I do think the "mega constellation commercial" focus is mostly misguided. If anything the biggest risks seem to be from government actors, in terms of things like a-sat weapons and Old Space big companies not feeling the need to care.

> For the foreseeable future, no one is going to have more economic incentive to prevent the loss of usable LEO than companies that build themselves around usage of LEO.

This line of reasoning never seems to work out the way you are suggesting. It's more likely to wind up in a giant Mexican standoff once there a multiple entities who all have the individual ability to ruin it for everyone.

Which is when it's time to sit down and institute a system of property rights for what has now become a scarce resource.

> and that itself cuts a ton of orbital debris (spent stages).

Almost all upper stages for LEO missions get deorbited (or at least they're meant to be deorbited). It's typically only the high energy missions like GEO/GTO where upper stages get left in a "graveyard orbit".

Full reusability makes sense and limits space junk - though the counter to that is they are launching more satellites into space - so probably negate each other to a certain point.

I guess my concern would be who actually has regulatory authority over space and how can they ensure all actors play fairly. I would have to imagine this is a situation that is similar to the tragedy of the commons type scenario.

To be clear - I really don't fall into one side of the camp or the other but I can see how a lot of the waste management side of things can fall by the wayside given our track record on waste cleanup (mines, municipal solid waste, nuclear)

Why not incentivize these companies with grant such that they will get the grant if their TO orbit launches _include_ the retrieval of space junk as well on the RETURN back to earth.

In order to go up, you have to bring something back.

Rendezvousing with, capturing and then manoeuvring with a captured object is an extremely hard and as yet unsolved problem, although there are some experiment. Also it is only feasible to rendezvous with a target on an almost identical orbit, otherwise the fuel and thrust requirements to reach it very quickly balloon to hugely prohibitive proportions.

> Rendezvoing with, capturing and then manoeuvring with another object is an extremely hard and as yet unsolved problem, although there are some experiment

Umm...this is utterly wrong, and has been since Gemini 10 in 1966, not to mention Skylab, Mir, ISS

> Umm...this is utterly wrong, and has been since Gemini 10 in 1966, not to mention Skylab, Mir, ISS

There's a big difference between docking with a spacecraft that is designed to be docked with and capturing a satellite that was not designed to be captured.

People are making good progress on this - Northrup Grumman just had a successful life extension satellite mission very recently. But it's far from a mature technology.

Makes no sense. Satellite retrieval is something that you definitely want to run a separate mission for - you don't want to bolt a visit to a completely separate orbit (and then transfer back to reentry) on top of a regular mission. The amount of fuel you need to launch something grows exponentially with the amount of fuel you need at the final stage.

Not to mention, if you're thinking of making LEO launches do this, then they'd either have to capture other LEO junk (makes little sense, most of it will deorbit itself), or you'd have to bolt on additional stage, which may very well introduce more junk itself (are people still using explosive bolts for upper stage separation?).

So make 'em separate missions, of course. You have to run the retrieval mission and have it succeed before you're granted a launch license for the original mission of putting something up.

Mission-for-mission makes no sense, who decides what mass or orbit of junk counts for your mission? You really have to wait years and set up a whole different mission architecture before you can launch your own project? It's bananas. However a cleanup tax on satellite launches used to fund recovery efforts might work.

The whole article is about how no one knows a good way to do that yet.

When these satellites burn up in the atmosphere, what happens to all the particles? Do we have lots of likely toxic material just constantly being burned and circulating around?

We estimate that the amount of dust and meteorites entering earth's atmosphere is about 100 metric tons per day. Combine that with emissions from volcanoes, many of which are highly toxic, and vaporised satellites are really insignificant.

The proportions remind me of another somewhat common space-related misconception, which is that rocket launches are big polluters.

With each launch emitting roughly as much as a single commercial flight and typically around 100 launches happening globally per year, even a tiny change in the tens of millions of commercial flights per year has a far larger impact on emissions than launches will for the foreseeable future.

> With each launch emitting roughly as much as a single commercial flight

It's one of those "connective" facts that puts so many distinct things into perspective. When I first learned it a few years ago, I became simultaneously very relieved that a rocket launch doesn't really have that big of a carbon footprint, and horrified by how much emissions a single passenger plane can make. It's a lesson about how visuals can be misleading: a plane looks tiny and doesn't really seem to be doing anything, while a rocket is big and rises to heavens on a pillar of flame, propelled by the anger of hell itself - and yet it turns out they're roughly the same, emissions-wise.

An easy way to compare the amount of CO2 they release is looking at the size. They only can burn the fuel they carry, and the density of the fuel is similar. The best images I got are

https://twitter.com/rocketrundown/status/956079854511972352

https://twitter.com/erdayastronaut/status/123959552711483801...

To make the calculation more difficult, like the 80% of the volume of a Boeing 747 is room for humans, and like 50% of the volume of a Falcon 9 is room for Oxygen and thrusters.

Some rockets use toxic fuel though, so it's not entirely apples to apples.

Some of the propellants (especially the hypergolics) are pretty toxic. However, Soyuz, Long March 5 (and later), Falcon 9, and Delta III (first stage) all use high-grade kerosene (RP-1 or its Russian equivalent) and liquid oxygen. Jet-A isn't too far from high-grade kerosene. Ariane 5 uses liquid hydrogen and liquid oxygen in its first stage. Many of the smaller players use RP-1/LOX, and some of the next-generation rockets will use methane/LOX propellants.

Though, the Delta III and Ariane 5 boosters do contain ammonium perchlorate, which is worse for the environment than jet fuel.

No, we have very small amounts of likely toxic material circulating around.

Earth's atmosphere is around 5 x 10^18 kg. All the satellites put together are not going to increase the amount of toxic material enough to matter whatsoever.

If you are concerned about metal oxide dust, then you would want to shut down every blast furnace in the world, and also pave over the Sahara, since it throws millions of tons of iron oxide dust into the air every year: https://en.wikipedia.org/wiki/Saharan_dust

> Also pave over the Sahara, since it throws millions of tons of iron oxide dust into the air every year

And at the same time fuck up our planet in a major way. There is a lot of life dependent on the plant on that dust. Everything from micro organisms in the sea, the Amazon gets its phosphorus from it, etc (it is a long list of things).

Sigh.

Strictly the low orbits of StarLink actually make it a sort of non-issue: there's PLENTY of atmospheric friction at the orbits used so re-entry is only a matter of time.

The bigger issue is higher orbits where orbital decay has far small perturbations to rely on. Those can be 10,000 year orbits which makes for a real problem.

I really wonder tho what this does to isolated tribes. To look up at an unchanging night sky, and on a regular basis new movement, new pinpoints of light. What does this do to the conception of The World ?

They will likely view it - correctly - as an ominous omen of change and upheaval.

Wait, how does launching stuff to low orbits reduce space junk?

Edit: I think I now understand your comment. I mean sure it maybe adds less space junk than launching things into higher orbits, but it doesn't really reduce it, right?

It's like building a structure in the middle of the plains where it'll sit untouched for decades, versus building a structure on the beach where the incessant waves will grind it to bits in short order.

Objects in low orbit don't stay there. They're slowed down by the atmosphere, and re-enter in a few months or a few years. Only if they're healthy enough to orient themselves, and have the fuel to do it, can they perform the re-boost maneuvers necessary to overcome atmospheric drag and stay up for longer.

So, a satellite which loses control, or gets smashed into bits which can't individually control themselves, just becomes fodder for the drag. It gets swept out of orbit by the wisps of atmosphere out there. Low orbit is very "clean" in terms of junk, because junk simply can't linger there. Just like there aren't a lot of ancient ruins on the beach.

The exact degree of drag is unpredictable because the exosphere is subject to a lot of variables, hence the "months to years" ambiguity. But they don't stay out there for decades or millennia, like junk in higher orbits.

> It's like building a structure in the middle of the plains where it'll sit untouched for decades, versus building a structure on the beach where the incessant waves will grind it to bits in short order.

That is a fantastic analogy.

Does that mean the material can still fully burn when falling down from under 2000km altitude? Or would we see pollution for example from battery materials?

The original altitude is not relevant to how the item burns up. Its not the vertical velocity when it hits the atmosphere, its the horizontal velocity.

Anything re-entering the atmosphere is subjected to temperatures around 8000 deg. Most things are just totally incinerated, unless they are well-enough insulated to withstand that temperature until they slow down from orbital velocity.

Atmospheric drag causes a relatively speedy deorbit if not actively resisted. The starlink satellites use Hall effect thrusters to maintain their orbit. If a unit fails, it should deorbit in under 5 years. If the thrusters work, they could also force it to deorbit much faster. The risk comes from debris from a possible collision being ejected into a higher orbit.

https://en.wikipedia.org/wiki/Orbital_decay#Atmospheric_drag

https://en.wikipedia.org/wiki/Starlink#Space_debris

> The risk comes from debris from a possible collision being ejected into a higher orbit.

Is that a thing that can happen?

My intuition is it would be possible if a rocket currently boosting to a higher orbit were to collide with something on its way there, but two objects in the same orbit colliding couldn't get enough delta-v to actually get to a higher orbit. They could maybe get "higher", but not at orbital speeds and so would rapidly decay.

I know next to nothing about orbital mechanics, so maybe someone who does can provide some better insight here.

The way to think about orbits is not one height but two: apoapsis and periapsis.

Debris from an explosion/collision can have an unlimited apoapsis, but the periapsis is at most the altitude of the incident.

So if it started in low orbit, it will still drop down regularly, and drag will still remove it over time.

The intuition is: all the debris from a collision now has an orbit passing through the point of collision (at that point of time). Collision at 300km means all the debris must have an orbit that passes through 300km at that moment.

The orbits will be elliptical: exactly how elliptical depends on the directions of the two objects hitting each other and the dispersion of the fragments (how much the fragments have sped up or slowed down compared with original objects, and change of direction).

The fragments the stay in orbit the longest should be those that remain closest to a circular orbit?

> Ejected into a higher orbit. Is that a thing that can happen?

Apparently not as such. See sibling comment here:https://news.ycombinator.com/item?id=26809297

I am no expert either, but a paper-napkin calculation tells me yes, this can happen.

Basically, just based on impulse and energy conservation, and some assumptions on the size and number of fragments, a head-on collision of two satellites going opposite directions at the same orbit should send some pieces of debris into much higher orbit.

How likely this is to happen in reality, and thus how big a problem this could be is a subject of modeling and much more complex calculations.

It turns out that intuitive reasoning for orbital mechanics is not easy. Two satellites having a head on collision become space junk because now you have many fragments that themselves can't control their altitude or deorbit. Fragments that are ejected from the collision cannot have more than energy than they had originally, nor can they change direction very much. If pieces are ejected vertically, they are always going to have less lateral velocity than they had going into the collision, therefore will end up in an elliptic orbit with a high probability of colliding with the atmosphere at perigree.

See "can you throw a baseball from the ISS and hit the earth" (No, you can't.)

The key insight there doesn't have that much to do with orbital mechanics, it's more that in a collision, small fragments can come out with vastly more momentum than they went in with, which in orbits translates to a higher apogee. If you have ever bounced a small ball off a large one, you know the effect. See this video for an intuitive example https://youtu.be/2UHS883_P60

Yes, it translates into a higher apogee, but not a higher perigee. So they will still come down to the altitudes with higher drag.

Strictly speaking this is only true if you start with a perfectly circular orbit, right?

Nope, any orbit. Think of it this way: an orbiting particle (or piece of debris) with no independent control/thrust will stay along whatever orbital path it's following, round and round (modulo any further orbital decay). After a collision, the particle is on a new elliptical trajectory, but the new trajectory still passes through the location of the collision. So the lowest point of the new orbit cannot be any higher than the altitude of the collision.

Likewise, if the particle is slowed down, the highest point of the new orbit cannot be any lower than than the altitude of the collision.

An interesting feature of orbital mechanics is that a change in velocity at a given point in orbit will change the altitude of the orbit at every point except where the change took place.

I think you're claiming that it is impossible for two colliding objects to release fragments that have more energy than either of the original objects. That's wrong.

If it helps, start by imagining two satellite-like machines colliding in interstellar space (i.e. not in orbit), where they were initially moving at 1m/s in opposite directions. Even ignoring the possibility of an explosion (unspent fuel, pressurized areas, I dunno), it's still very easy for interactions to violently fling individual parts outward at speeds higher than 1m/s.

Of course, my guess is that the possible amounts of additional delta-v are pretty low, like 100 m/s, and as such the resulting orbit would not be much higher than the original orbit. But that's just wild conjecture.

If you were to collide two LEO satellites and turn them into fragments, even if you give those fragments much higher velocity than the initial satellites, you _cannot_ launch them into a higher orbit.

Fragments that are launched down obviously hit the Earth. But an orbit is a closed ellipse, so the fragments that are launched up will _also_ hit the Earth -- they'll just go up steeply, turn around, and come down. The only fragments that won't initially hit the Earth are the ones that are ejected tangentially, parallel to the Earth's surface. Those will go into an elliptical orbit with a high apogee and perigee at the altitude of the collision. Which means they will _still _come down to that altitude and gradually lose energy.

I agree that the best case (uh, best for people who like Kessler cascades?) is that the new fragment has a perigee equal to the location of the collision, and a higher apogee. That means, yes, that it will suffer whatever degree of drag was present at the altitude of the collision.

However.

The grandparent post claimed that "Fragments that are ejected from the collision cannot have more than energy than they had originally, nor can they change direction very much", which I think is clearly what I addressed in my comment, and I stand by my correction of that, and your reply doesn't address this at all.

Also, with respect to "you _cannot_ launch them into a higher orbit", you just agreed that you can have a higher apogee than the collision. If the input satellites had perigees lower than the altitude of collision, you can also have a lower perigee. More velocity at this point on the orbit, higher apogee, higher perigee.... So how do you figure you cannot have a higher orbit?

Parts could be squeezed between high pressure zones and get shot out of the collision, but even those parts won't gain a higher orbit because the perigee will be below the original orbit.

First of all, no, the perigee need not be "below the original orbit", where did you get that? The correct thing to say is that the new orbit's perigee cannot be any lower than the location of the collision.

But more generally, how does this justify your claim that "Fragments that are ejected from the collision cannot have more than energy than they had originally"? Or were you conceding that point and making a different point?

See also my more-detailed cousin of this comment.

Set a baseball on top of a basket ball and drop them both from waist high, the baseball will shoot really high into the air by stealing energy from the basketball. The same principle can occur when you start breaking pieces off of spacecraft with collisions. Also the two colliding objects might not be in "the same orbit" one could be in a polar orbit and one could be in a more equitorial orbit. One could be circular and the other highly eliptical, etc. This could boost parts of one of them into a higher orbit (though honestly it would be just as likely to slow things down).

> I mean sure it maybe adds less space junk than launching things into higher orbits, but it doesn't really reduce it, right?

Sure, but that's not really the point. Using biodegradable plastic is much better than traditional plastic that stays in landfills forever. "Just don't generate trash" isn't an option.

There is literally no problem with plastics that stay in landfills forever. Landfills are underground, like the oil that the plastic is derived from.

Landfills are generally very close to the surface, filling holes created by activities like strip mining, so they do occupy surface area. Unless you're wiling to dig down under existing landfill to dig out voids to pack with more landfill.

Landfills are covered when they are full, effectively isolating them from surface ecosystems.

The point is you can't use that area for getting rid of landfill anymore. From that perspective that area is 'used up'.

If the cost to LEO is a fraction of the cost to HEO or geostationary, it becomes economical to send cheap satellites to LEO and let them burn up in a few years than to send expensive satellites to HEO or geostationary and let them become multi-decade space junk.

Eg if it is $10m to LEO and $15m to HEO you will pay the extra to get to HEO and put a satellite designed for long life - but if it is $1m to LEO you can build a cheaper satellite and launch it 5 times over the following decades.

This is only true if the cost of the launch dominates the total mission price. That just isn't true any more. Engineering labor is the dominant cost until you have very high volumes. Even then, component costs are still far higher than launch costs.

But they aren’t building 1 satellite and launching it 5 times over the next decade. They’re maintaining a constant fleet of 42000 satellites, which means putting up new junk to replace the old junk while its still flying junk and not de-orbited junk.

I like the idea of having internet on my sailboat as much as the next guy, but it takes some serious cognitive dissonance to convince oneself that 42,000 pieces of nearby junk is better than a handful of pieces of far away junk.

>They’re maintaining a constant fleet of 42000 satellites, which means putting up new junk to replace the old junk while its still flying junk and not de-orbited junk.

No. First, The whole reason they need to maintain a fleet, rather then just having it sit there for decades, is that they're so low. The vast majority of planned Starlink satellites from that number are from Phase 2, and are V-band VLEO sats with orbits around 340 km, which is really low. At that altitude, natural decay time is measured in weeks at best. To work they'll need both active thrusters providing regular boost and aerodynamic low drag design (and maintaining orientation for that will itself require fuel). Should they actually lose all control through malfunction or a collision, they and resulting debris will deorbit very rapidly (they'll both have no boost and be less aerodynamic).

And second, "through malfunction or collision", because it's not as if SpaceX (and other LEO satellite operators) doesn't have, and indeed are required to have, plans for controlled deorbit at EOL. Most of the satellites can be expected to get deorbited in a controlled way as planned. Making sure everything burns up has been one of the things slowing Starlink development, it took some work to ensure the optical links would properly go for example.

SpaceX is not interested in leaving them up there. I mean, FFS people, who would be hurt more than SpaceX by Kessler Syndrome!?

>I like the idea of having internet on my sailboat as much as the next guy, but it takes some serious cognitive dissonance to convince oneself that 42,000 pieces of nearby junk is better than a handful of pieces of far away junk.

Your dismissiveness towards hundreds of millions of underserved people and challenging use cases and ignorance of orbital dynamics does you no favors here.

> and maintaining orientation for that will itself require fuel

With some gyroscopes, I think they only need electricity for it, not actual reaction mass.

> Your dismissiveness towards hundreds of millions of underserved people and challenging use cases and ignorance of orbital dynamics does you no favors here.

Imagine thinking that a product that costs $500 up-front and $99/mo thereafter and uses ~100 W constantly is designed for "hundreds of millions of underserved people".

> Your dismissiveness towards hundreds of millions of underserved people and challenging use cases

This is pompous and elitist nonsense of the highest order. People have other more pressing concerns than fast internet. You know...like water, food and cheap power.

Communications infrastructure is essential for any developing/developed nation. It's easy to take for granted while living in an urban center of a 1st world nation just how game-changing communications is. A truly global, easily accessible network has the potential to lift millions out of poverty through facilitating economic growth of remote, unconnected areas. Dismissing satellite constellation networks as merely "faster internet" misses the bigger picture.

Also it's worth noting that communications networks is but one application for LEO satellite constellation technology.

It's not a matter of 'more' or 'less' rather than 'easier to deal with'.

LEO sats will probably produce more space junk by mass but junk in LEO will be gone in a few years, no matter how much or little of it their is.

On the other hand Geostationary satellites produce much less junk in terms of mass, but once it's up there it's not coming down for centuries.

Low Earth orbit satellites drag in the fringes of the upper atmosphere a lot more, so their lifetime in space, unboosted, is typically only a few years. If your satellite dies, it's "self-cleaning."

That said, an impact at this level can still spew debris into higher orbits, as the chaos of the collision can impart enough energy to some collision ejecta to change their orbit significantly. So you still don't want things bumping into each other.

It's not less so much as zero. Space junk isn't really a problem to worry about below a certain point.

There's atmosphere up there, so it creates drag which slows the stuff down and causes it to re-enter and burn up or crash onto the planet in only a few years. Compare that to higher orbit, where there's much less drag and space junk could last centuries or more before it deorbits.

It’s subject to drag from the whispy atmosphere and relatively quickly de orbits and burns up (thus no junk is left behind)

Can you give a rough estimate of what fraction of space waste is due to differing orbits of satelites? Like very low vs low vs geostationary? I don't know the gradations between those either.

Wikipedia to the rescue (note the log-log scale) https://en.wikipedia.org/wiki/File:Spacedebris_small.png

The biggest issue is by far the 600km to 2000km altitude band. It is popular for both LEO and SunSync sats and is high enough above the atmosphere that minimal propellant is needed for orbit boosting.

The actual amount/distribution of stuff can be misleading. What really matters is the risk of collision, which is the highest for Sun-synchronous orbits. The fact that lower SSOs are also the most convenient for remote sensing work doesn't help.

GEO is more or less safe because it's unstable (dead stuff drifts outwards), well kept (slots are limited by the resolution of customers' terminals) and typically doesn't have high relative velocities.

You seem to know a lot about this subject! Years ago, an acquaintance of mine told me of their idea to use railguns to remove space junk. Is this a remotely feasible solution?

From space, maybe. A railgun on the surface of Earth that was trying to slow down a piece of space junk would, by necessity, be aimed at a low angle. Getting a projectile a couple hundred kilometers up while launching at such angle would make this railgun be something between a window-shattering nuisance and a weapon of mass destruction, depending on muzzle velocity and the size of the projectile.

The problem is, of course, the atmosphere. That thick soup of gas that's at its densest near the surface, and has the annoying tendency of engulfing hypersonic projectiles in a thick ball of screaming-hot plasma.

> There is one ultimate sure-fire way to reduce space junk: launch stuff to very low orbits

Really? If there's a collision in LEO, doesn't debris spray in all directions, including some in the direction of higher, more stable orbits?

Just to add to sibling comments who already answered this, that isn't how orbital mechanics work (or else a lot of things would be much easier!). A good place to start reading is on Hohmann transfer orbits [1, 2]. They're commonly covered in the context of interplanetary travel, but the exact same principle applies to changing orbits around a single gravity well. The key take home as applied to here is that a single instantaneous addition of energy (delta-v) to an object in stable circular orbit just changes it to an elliptical orbit where the high point is determined by the added delta-v but the low point is the same as where it started. It takes a properly timed second instantaneous application of energy to stabilize at a different orbit.

So any ejecta from a single collision will still orbit through whatever altitude they started at, and in turn will be affected by atmospheric drag at that point regardless of how much farther out they get at the high point. In principle it's of course not impossible that they could collide with something else already in a higher stable orbit, but the odds of that get very, very low particularly in the short time frame they have before decay assuming they're starting in VLEO.

----

1: https://ocw.mit.edu/courses/aeronautics-and-astronautics/16-...

2: https://solarsystem.nasa.gov/basics/chapter4-1/

> high point is determined by the added delta-v but the low point is the same as where it started

Is it the low point that is that same? I would have thought it was the current point (at time of delta-v addition), wherever that is, that must be part of the new orbit. Not that this is relevant for the topic at hand, just trying to check my understanding.

This is correct. The simplification being made is that the original orbit was circular, so if you increase the debris' velocity, the current point is now the low point of the new orbit.

Even for circular orbits, is that the case? The new low point would be lower if impulse is not parallel to the velocity.

Correct, so basically the lowest point(periapsis) of the new orbit cannot be any higher than the point where the collision happened.

Another way to think about it is, an orbit around a single body is entirely defined[0] by your current position and your instantaneous velocity. Six numbers. Where you are, which direction you're going, and how fast. Such orbits are closed, which means wherever you are right now, you'll return to that exact point later.

So if you're at point P in your orbit and suddenly kinetic energy was added, your velocity changes in some arbitrary way. While you aren't sure where you'll go and when you'll get there, as long as the new (position, velocity) pair still defines a closed orbit around the body, you can be damn sure you'll return to the exact point where you are now.

From this follows that the lowest point of your new orbit cannot be higher than where you are now, and the highest point of your orbit cannot be lower.

[0] - Ignoring gravitational effects of other bodies, residual drag, magnetic fields, solar pressure, etc. None of these matter on the scale of days to months.

Yes, if you start from roughly circular orbit, and the impulse is in the opposite direction to the direction of travel, it will slow you down and then the location of impulse is now the new highest point on orbit, and the lowest point after the impulse is even lower than before it.

Thank you, this is really informative.

That's a comprehensive answer, thank you. TIL.

This would just create an oblong orbit, which would still cause re-entry into thicker atmosphere for a portion of the orbit.

In the direction, yes, but it would still cross the original orbit. You need a second collision to actually get stuff into higher orbits.

I thought the debris that gets lifted upward tends to go into elliptical orbits. Apogee would still create a significant amount of drag, wouldn't it?

Yes, I'm aware that orbital mechanics is not that simple, that's why I specifically didn't talk about "upward" as a higher orbit isn't necessarily caused by a push in that direction.

Sibling comment ( https://news.ycombinator.com/item?id=26809297 ) is better at outlining that I was wrong: apparently a single impact cannot produce an entirely higher orbit, for debris sent in _any_ initial direction. At best, it produces elliptical orbits with a higher apogee, and the same drag at perigee.

With my limited knowledge of orbital mechanics I believe this wouldn't be a problem. It takes a lot of energy to move to a higher orbit. I would think the probability of enough energy to fling something to a higher orbit where it would be stable for a noticeably longer time would be very low.

Think of exploding something 2 miles up the side of a mountain, what are the chances a piece goes another mile up the mountain. Like that but even harder.

Not a physicist, but I doubt the debris would have the energy to reach geostationary orbits.

Not relevant? There are many orbits that are below geostationary and are also high enough to avoid atmospheric drag for a long time.

This is important correction. But a high-speed collision at a very low orbit can trow debris to not-so-low orbit.

How big of a problem this can be is something to model and calculate, it's not something one figures out just from the general principles.

A high speed collision in low orbit can change a circular low orbit into an elliptical eccentric orbit that intersects a higher circular orbit, but unless there is an additional accelerating event at that higher altitude, it cannot recircularize its orbit at that higher altitude.

There are thus two takeaways:

1. By definition, this means that part of the orbit will always be at low altitude, regardless of the collision dynamics. So this means that it will continue to decay over time, albeit perhaps at a slower rate (decay being proportional to the time spent at lower altitude).

2. While that eccentric orbit will intersect with a higher circular orbital plane, it does so in a predictable fashion that can be routed around. The higher orbits are also much sparser, so the chance of this intersecting with a satellite that is already present is very, very small.

Thank you, others pointed this out too. Proves that "celestial mechanics" is trickier than I thought.

I guess if a collision is messy enough, there would be secondary collisions between debris pieces, and it sounds like these in principle can push some junk into higher orbits. But I think the probability is really low; this should not be a concern.

> a high-speed collision at a very low orbit can trow debris to not-so-low orbit

How so? According to my understanding of orbital mechanics, if an object begins its orbit at a certain altitude, it will return to that altitude exactly one orbit later. While a collision can change the shape of the object's orbit, it can't change the spot of collision - which it will return to every cycle of its orbit.

Some of the debris will be thrown into eccentric orbits reaching significantly higher altitudes as part of the orbit. If these fragments hit objects in these higher orbits, that will create further debris clouds in more stable high orbits, etc, etc. That way you can get a runaway chain reaction that could escalate to arbitrarily high orbits and debris cloud lifetimes.

I don't see how that's a major issue, It'd take a pretty odd angle for a high speed collision to push something into a higher orbit. That seems like a pretty rare event.

Not really. High speed collisions are high entropy events, so debris fills state space, and you get SOME debris at high altitudes (usually in highly eccentric orbits). That being said, low altitude is much safer since you don’t get as many long lived dead satellites

From super-low Earth orbits, more of the orbits that go really high also impact Earth or its atmosphere on the other side, so they only are up there once, rather than over and over and over.

Yep!

> But a high-speed collision at a very low orbit can trow debris to not-so-low orbit.

This seems wrong, from what I understand of orbital mechanics.

Yeah, the terminology is wrong. It's not a high orbit, it's a high ballistic trajectory, which then is terminated on the low side. But if that object improbably happens to collide with a high-orbit object then you could get a spreading of the consequences to higher orbit.

Curious how big of an issue this actually ends up being since they can't be put in a stable higher orbit without a second burn/collision to circularize. Without a change to the orbit perigee it would still decay, just more slowly?

> inherently their lifetimes will be measured in single digit years. But that's ok since they can be replaced so cheaply

Oh cool, let's shoot even more soon-to-obsolete pollution into the atmosphere. That's working so well on Earth. /s

1000s of satellites are irrelevant to atmospheric pollution. The numbers are so small compared to the vast size of the atmosphere it isn’t really a concern to worry about.

Atmospheric pollution from earth based sources is different because it’s something that is being done in the scale of billions of people, not 1000s, and CO2 doesn’t naturally disappear from the atmosphere over time and is growing decade after decade.

You're so confident. How adorably naive.

Do you math?

Over one hundred metric tons of space dust falls on the earth daily. Do you really think SpaceX de-orbiting a 200 Kg satellite once a day or so is going to change that significantly?

Comments like this aren't constructive.

Aren't their lifetimes so short because they will fall out of orbit? That's not really polluting orbit.

SpaceX is planning to launch up to 42 000 satellites (there are about 3000 now). Cheaper $/kg leads to more space junk and higher risk of Kessler syndrome, that doesn't need any qualifications, I think. If something goes wrong the fact that the cloud of debris will be there for 10 years and not 100 years is not a huge consolation

Timing is absolutely a huge qualification to Kessler syndrome.

If the cloud of debris is gone in 5 years, that's a hurdle to some careers and a (hopefully) temporary shutdown for some businesses. Covid has taught us that the modern economy has traded a lot of robustness for profit and efficiency, and those would certainly be significant losses of billions or trillions of dollars, but it would be a small fraction of GDP, we would hopefully learn from the mistake, and we would still be a spacefaring species.

If Kessler syndrome occurred among geostationary satellites, that could potentially be a problem for millions of years. That's six orders of magnitude different and definitely needs qualification.

But there is a lot more space in geostationary orbits and lot less stuff so the risk of chain of collisions is lower

There is far less space in geostationary orbit, because the locus of GEO is a one-dimensional ring above the equator.

It's true that there is less GEO real estate in this sense. But to kick off Kessler syndrome, you need two objects in different orbital planes so that they collide at very high speed. It wouldn't be possible for two defunct GEO satellites to collide, because two nearby GEO satellites are moving in the same direction at the same speed.

but aren't the satellites in GEO stationary relative to one another? And if by chance they all get destroyed in collision the debris would mostly also stay in that ring so we could still launch rockets and have satellites in other orbits that don't cross it.

And this is exactly the sort of comment that grandparent addressed with their post.

More stuff in orbit doesn't necessarily lead to higher risk of a catastrophic kessler syndrome event, if it's low-earth orbit. The risk is not the same as geosynchronous where it can stay there forever. The satellites themselves (and even the debris if there is a collision) will decay on a very short-term basis (single-digit number of years).

By making launch cost-per-kg cheaper, it is now plausible to put things in LEO instead of geo or other orbits. And doing that probably actually reduces the risk of a catastrophic event, even if there is more stuff.

Possibly higher risk of a smaller event, sure, but that is true of anything that puts more stuff in space. That's a generic argument that we shouldn't be doing anything in space at all, which I think is not compelling.

> If something goes wrong the fact that the cloud of debris will be there for 10 years and not 100 years is not a huge consolation

Seems like a pretty big consolation to me, no?

No kidding. Order of magnitude

How does the cloud of debris stick around? For context, I've played Kerbal so I know a bit about orbits but not much.

My intuition is that the velocities involved in being in orbit in the first place are so big (7 kilometers per second according to a quick search) that the orbit of the debris won't be substantially different, so it will still decay. Even if you give enormous kicks to the debris, most of it will end up with orbits intersecting the earth. The ones that don't will still intersect the original collision location, and so still experience decay during that part of their orbit.

Basically, to change your orbit you need to boost once to push the other side of the orbit out and then again at the other side to push the original side out. A collision can only give one of the boosts, and so is sort of inherently bad at moving the orbit.

You are totally correct (and yay KSP!)

But the danger is the debris that is ejected from a collision at a lower orbit would then collide with debris that is at a higher orbit. The debris from the 2nd collision would have its original periapsis, which may be high enough to stick around for decades.

Reply on news.ycombinator.com