This is going to be fun. Let's get three to five companies to launch several thousand satellites each all at once into the same general orbital paths. What could go wrong?
This is going to be fun. Let's get three to five companies to launch several thousand satellites each all at once into the same general orbital paths. What could go wrong?
It's unlikely. These satellites will decay on their own if unmaintained, and they've been positioned very carefully. Plus, space is huge. Consider the size of the Earth and realize their orbital...
It's unlikely. These satellites will decay on their own if unmaintained, and they've been positioned very carefully. Plus, space is huge. Consider the size of the Earth and realize their orbital paths are significantly wider than that.
It would be the most hysterical answer to the Fermi Paradox if every other alien race shot up too many satellites to get anything else outside of orbit.
It would be the most hysterical answer to the Fermi Paradox if every other alien race shot up too many satellites to get anything else outside of orbit.
There is a lot space up there, and unlike the surface of the earth there is a third dimension to play with. AFAIK the biggest issues are the fragments from explosions that Russia, China, India,...
There is a lot space up there, and unlike the surface of the earth there is a third dimension to play with. AFAIK the biggest issues are the fragments from explosions that Russia, China, India, and the USA have caused by testing their capabilities for sat destruction. However, those fragments do have much more to hit with these new constellations.
Project Kuiper will consist of satellites at three different altitudes: there will be 784 satellites at 367 miles, 1,296 satellites at 379 miles, and 1,156 satellites at 391 miles. The Verge
In November 2018, SpaceX received US regulatory approval to deploy 7,518 broadband satellites, in addition to the 4,425 approved earlier. SpaceX's initial 4,425 satellites had been requested in the 2016 regulatory filings to orbit at altitudes of 1,110-kilometer (690 mi) to 1,325-kilometer (823 mi), well above the ISS. The new approval was for the addition of a very-low Earth orbit NGSO [non-geostationary satellite orbit] constellation, consisting of 7,518 satellites operating at altitudes from 335-kilometer (208 mi) to 346-kilometer (215 mi), below the ISS Wikipedia
Let's go crazy and assume a total of 100,000 satellites from all companies combined. Would you be really concerned about 100,000 cars on the 2D surface of the earth, each of which are tracked by radar, and only have Newtonian physics to worry about?
I work in the space industry...it's true that there's a lot of empty space in orbit, but Kessler syndrome is also a very real risk that shouldn't be underplayed. Biggest factor missing from your...
Exemplary
I work in the space industry...it's true that there's a lot of empty space in orbit, but Kessler syndrome is also a very real risk that shouldn't be underplayed.
Biggest factor missing from your calculation is that satellites are not uniformly distributed in their orbits. A single Falcon 9 can carry several dozen spacecraft, and they all get deployed by the launch vehicle into more or less the same orbital trajectory (this is why SpaceX's Starlink satellites formed a "train" right after launch). They should maneuver to separate themselves apart, but those maneuvers are not zero-risk and carry their own chance of collision.
There's also only so many feasible orbits. A lot of satellites do "polar" orbits, for example, where they pass over the north and south pole once per orbit, creating a higher risk of "traffic jam" at the poles. These satellites from Amazon and SpaceX targeted at providing internet are more likely to be in inclined orbits, for example only going as far north & south as 45 degrees because that's where the bulk of their possible customers are.
100,000 satellites is a huge overestimate. There's currently around 5,000 man-made objects on orbit, and about 10,000 total launched since the beginning of the space age. Just between SpaceX and Amazon, they're talking about doubling, tripling, or quadrupling the number of spacecraft on orbit.
Instead of cars on the Earth's surface, think of it more like this - at any given time, there's X planes in the air over the United States. A year or two from now, that's going to increase to 3X or 4X. Do you think air traffic controllers would be concerned?
Airplanes are also easy, because if two of them collide, the debris falls back to Earth. That may cause damage on the ground, but at least the debris doesn't just hang there, posing a navigation hazard to every other plane. That's the real risk of Kessler syndrome - a collision is unlikely, but every single collision raises the odds of a future collision. That can get into a positive-feedback loop very quickly.
Sort of...normal actuarial tables track likelihood of death, and lots of people die, so there's pretty good data to base them on. Satellites shouldn't ever collide, so you can't really do the same...
Sort of...normal actuarial tables track likelihood of death, and lots of people die, so there's pretty good data to base them on. Satellites shouldn't ever collide, so you can't really do the same sort of analysis.
The USAF does track just about everything in Earth orbit, originally to detect Russian missiles or other space-based weapons. They run simulations based on that to project if anything is about to collide, and will notify affected satellite operators if they think that's likely to happen.
For example, my company got a notification several months ago, that one of our satellites had a 0.04% chance of colliding with another satellite. Their projection was that in 3 days, we'd pass about 1 kilometer from each other. We've had closer approaches than that in the past, including warnings like "don't maneuver without coordinating with us first, because the other satellite wants to maneuver to avoid collision and you shouldn't both move in the same direction".
The biggest challenge with projections like that is that is that orbits determined by ground-based radar have uncertainty built in to them, and that uncertainty grows the farther out you try to project into the future. Orbits also decay very slowly and gradually, so even without new satellites being launched, they've got to keep continually scanning the sky and updating orbital data on everything.
I'm not terribly familiar with the launch-planning side of things, but I'm sure there's a similar process done before launch to verify that a target orbit won't have an abnormally high probability of collision with any existing orbits. That'll only get more complicated as the sky gets more crowded.
If I may ask, are you guys worried at all about these new constellations and thousands of new objects in LEO? I mean if Kessler Syndrome was a true concern at this point, the new constellations...
If I may ask, are you guys worried at all about these new constellations and thousands of new objects in LEO? I mean if Kessler Syndrome was a true concern at this point, the new constellations wouldn’t be allowed, correct? Do you have faith in this regulatory regime?
I have faith in the old-school, NASA/FAA/FCC/NOAA/etc regulators, and a lot of skepticism of the VC-funded startups coming in to "disrupt" the space industry. You know how AirBNB "disrupted" the...
Exemplary
Do you have faith in this regulatory regime?
I have faith in the old-school, NASA/FAA/FCC/NOAA/etc regulators, and a lot of skepticism of the VC-funded startups coming in to "disrupt" the space industry.
You know how AirBNB "disrupted" the hotel industry by ignoring / finding loopholes in existing local hotel laws? Picture that same mentality but applied to launching rockets and operating satellites on orbit.
Silicon Valley is "move fast and break things". Aerospace is the opposite. If you move too fast, the wings fall off and you die. But also, if you move too slow, you stall and you (maybe) die. So you try to find a happy, stable, sustainable middle ground.
When Amazon says they want to launch 3,000+ satellites, you know that means there's a spreadsheet somewhere at Amazon to figure out they want X satellites to be operational, and Y% of them are going to fail over time, so they actually need to launch Z of them. Then they build them with the assumption that Y% are going to fail anyway, so does such-and-such component really need to be that reliable? That's a fine assumption to make when you're building datacenters on Earth, but not when you're launching potential space debris.
For another example of where the tech industry and the aerospace industry don't mesh well, AWS has "groundstation as a service" where the idea is that they'll slap satellite antennas on top of each of their datacenters and rent them out, same as they rent out the more general-purpose hardware inside.
The problem they've run into is that regulatory agencies (like the FCC and their international counterparts) want the location of any big radio transmitters to be public knowledge, and AWS considers its datacenter locations (and even the number of discrete locations that make up a region) to be trade secrets.
If Amazon could get away with not asking permission to launch these satellites, they would. If they can cut corners to save money along the way, they will.
While this is an utter flight of fancy... Until I thought about AWS in space, I didn't think about the possible legal implications of a server in orbit. I know this constellation is not about...
While this is an utter flight of fancy... Until I thought about AWS in space, I didn't think about the possible legal implications of a server in orbit. I know this constellation is not about servers, just routers, but eventually there will be some, right?
From a founder of Protonmail about their legal responsibilities:
Essentially, unless you are located on a ship 100 km offshore, you will have to fall under the jurisdiction of some country and must follow the laws of that country. Almost all countries require companies to assist in some manner in criminal investigations, and Switzerland is no exception. Protonmail AMA
I know that the FCC and FAA regulate most satellite launches, but I wonder, will there eventually be some opening here?
Putting a server in space would have all sorts of disadvantages. Power supply, heat dissipation, maintenance and replacement, radiation if it is high up enough. However, it is probably a lot more...
Putting a server in space would have all sorts of disadvantages. Power supply, heat dissipation, maintenance and replacement, radiation if it is high up enough. However, it is probably a lot more feasible to put a server 100 km offshore with a communication constellation.
Interesting how there's not a single mention of SpaceX's similar endeavor.
This is going to be fun. Let's get three to five companies to launch several thousand satellites each all at once into the same general orbital paths. What could go wrong?
It's unlikely. These satellites will decay on their own if unmaintained, and they've been positioned very carefully. Plus, space is huge. Consider the size of the Earth and realize their orbital paths are significantly wider than that.
It would be the most hysterical answer to the Fermi Paradox if every other alien race shot up too many satellites to get anything else outside of orbit.
There is a lot space up there, and unlike the surface of the earth there is a third dimension to play with. AFAIK the biggest issues are the fragments from explosions that Russia, China, India, and the USA have caused by testing their capabilities for sat destruction. However, those fragments do have much more to hit with these new constellations.
Let's go crazy and assume a total of 100,000 satellites from all companies combined. Would you be really concerned about 100,000 cars on the 2D surface of the earth, each of which are tracked by radar, and only have Newtonian physics to worry about?
Now, I don't mean to accuse anyone here of 2 dimensional thinking. 🖖
EDIT: listen to the child comment by @spit-evil-olive-tips, not me, but here is a Kurzgesagt video on topic: https://www.youtube.com/watch?v=yS1ibDImAYU
Related thread on HN: https://news.ycombinator.com/item?id=20364928
I work in the space industry...it's true that there's a lot of empty space in orbit, but Kessler syndrome is also a very real risk that shouldn't be underplayed.
Biggest factor missing from your calculation is that satellites are not uniformly distributed in their orbits. A single Falcon 9 can carry several dozen spacecraft, and they all get deployed by the launch vehicle into more or less the same orbital trajectory (this is why SpaceX's Starlink satellites formed a "train" right after launch). They should maneuver to separate themselves apart, but those maneuvers are not zero-risk and carry their own chance of collision.
There's also only so many feasible orbits. A lot of satellites do "polar" orbits, for example, where they pass over the north and south pole once per orbit, creating a higher risk of "traffic jam" at the poles. These satellites from Amazon and SpaceX targeted at providing internet are more likely to be in inclined orbits, for example only going as far north & south as 45 degrees because that's where the bulk of their possible customers are.
100,000 satellites is a huge overestimate. There's currently around 5,000 man-made objects on orbit, and about 10,000 total launched since the beginning of the space age. Just between SpaceX and Amazon, they're talking about doubling, tripling, or quadrupling the number of spacecraft on orbit.
Instead of cars on the Earth's surface, think of it more like this - at any given time, there's X planes in the air over the United States. A year or two from now, that's going to increase to 3X or 4X. Do you think air traffic controllers would be concerned?
Airplanes are also easy, because if two of them collide, the debris falls back to Earth. That may cause damage on the ground, but at least the debris doesn't just hang there, posing a navigation hazard to every other plane. That's the real risk of Kessler syndrome - a collision is unlikely, but every single collision raises the odds of a future collision. That can get into a positive-feedback loop very quickly.
Thank you for the very informed response! Is there something analogous to actuary tables for satellites which predict the chances of collisions?
Sort of...normal actuarial tables track likelihood of death, and lots of people die, so there's pretty good data to base them on. Satellites shouldn't ever collide, so you can't really do the same sort of analysis.
The USAF does track just about everything in Earth orbit, originally to detect Russian missiles or other space-based weapons. They run simulations based on that to project if anything is about to collide, and will notify affected satellite operators if they think that's likely to happen.
For example, my company got a notification several months ago, that one of our satellites had a 0.04% chance of colliding with another satellite. Their projection was that in 3 days, we'd pass about 1 kilometer from each other. We've had closer approaches than that in the past, including warnings like "don't maneuver without coordinating with us first, because the other satellite wants to maneuver to avoid collision and you shouldn't both move in the same direction".
The biggest challenge with projections like that is that is that orbits determined by ground-based radar have uncertainty built in to them, and that uncertainty grows the farther out you try to project into the future. Orbits also decay very slowly and gradually, so even without new satellites being launched, they've got to keep continually scanning the sky and updating orbital data on everything.
I'm not terribly familiar with the launch-planning side of things, but I'm sure there's a similar process done before launch to verify that a target orbit won't have an abnormally high probability of collision with any existing orbits. That'll only get more complicated as the sky gets more crowded.
If I may ask, are you guys worried at all about these new constellations and thousands of new objects in LEO? I mean if Kessler Syndrome was a true concern at this point, the new constellations wouldn’t be allowed, correct? Do you have faith in this regulatory regime?
I have faith in the old-school, NASA/FAA/FCC/NOAA/etc regulators, and a lot of skepticism of the VC-funded startups coming in to "disrupt" the space industry.
You know how AirBNB "disrupted" the hotel industry by ignoring / finding loopholes in existing local hotel laws? Picture that same mentality but applied to launching rockets and operating satellites on orbit.
For example: FCC fines Swarm $900,000 for unauthorized satellite launch
Silicon Valley is "move fast and break things". Aerospace is the opposite. If you move too fast, the wings fall off and you die. But also, if you move too slow, you stall and you (maybe) die. So you try to find a happy, stable, sustainable middle ground.
When Amazon says they want to launch 3,000+ satellites, you know that means there's a spreadsheet somewhere at Amazon to figure out they want X satellites to be operational, and Y% of them are going to fail over time, so they actually need to launch Z of them. Then they build them with the assumption that Y% are going to fail anyway, so does such-and-such component really need to be that reliable? That's a fine assumption to make when you're building datacenters on Earth, but not when you're launching potential space debris.
For another example of where the tech industry and the aerospace industry don't mesh well, AWS has "groundstation as a service" where the idea is that they'll slap satellite antennas on top of each of their datacenters and rent them out, same as they rent out the more general-purpose hardware inside.
The problem they've run into is that regulatory agencies (like the FCC and their international counterparts) want the location of any big radio transmitters to be public knowledge, and AWS considers its datacenter locations (and even the number of discrete locations that make up a region) to be trade secrets.
If Amazon could get away with not asking permission to launch these satellites, they would. If they can cut corners to save money along the way, they will.
While this is an utter flight of fancy... Until I thought about AWS in space, I didn't think about the possible legal implications of a server in orbit. I know this constellation is not about servers, just routers, but eventually there will be some, right?
From a founder of Protonmail about their legal responsibilities:
I know that the FCC and FAA regulate most satellite launches, but I wonder, will there eventually be some opening here?
Putting a server in space would have all sorts of disadvantages. Power supply, heat dissipation, maintenance and replacement, radiation if it is high up enough. However, it is probably a lot more feasible to put a server 100 km offshore with a communication constellation.
It's frustrating to see the night sky being polluted by a small collection of conceited, smug billionaires.