Depends on execution, we could have billions of satellites without huge problems. But Kessler syndrome could happen even just with the amount of satellites already in orbit if people had been actively malicious. So, the “limits” are mostly just a question of what assumptions you’re making.
> Kessler syndrome could happen even just with the amount of satellite
Kessler as in a catastrophic loss of satellites. In LEO, it wouldn’t be a denial of space programme, nor even a systemic issue—there aren’t high enough orbital energies.
Kessler syndrome is both loss of satellites but also rendering orbits unavailable. My comment was presupposing malicious actors so the hypothetical orbits aren’t going to be identical to what we currently have. It doesn’t take that much material orbiting in the opposite direction to render geosynchronous orbit untenable for example.
There are 1.5 billion cars in the world, and they all drive on a single plane on a very small percentage of the surface. Now increase the circumference by a lot and add a whole new dimension. I think we'll be fine.
One critical thing you're missing is that when cars smash into each other, the debris stops moving soon, and a spec doesn't completely destroy other cars a week later.
Also car interactions are seldom between competing space programs or warring nationalities with their own interests and limits on maneuvering capability and energy expense quotas.
Collision avoidance at spaceX is largely automated but for most operators there is a wide variety of response capability
Not disagreeing with the conclusion, but cars are almost always stationary (parked) and when they do move they are many orders of magnitude slower than satellites. Which means it's not an obvious 1:1 comparison.
And cars are very maneuverable -- can stop, start, and turn in seconds; and can safely pass within inches of each other. (Most) cars also have one human each paying full attention to where it is going and controlling its path.
Satellites don't have so many degrees of freedom.
A bad way to extend the cars analogy is to compare with people. There are 8 billion of us and we walk around all the time but almost never get damaged by running into each other.
Satellites follow a extremely well known path and are tracked by the operators to a ridiculous level of precision. I can't imagine us needing more than 10 providers of 100K sats each at an upper limit. You could fit them all in a few (10 say) large football stadiums if you piled them up (a sat is probably 5m³ at most, a stadium is 100x100x50m so 500000m³)
There have already been six "Unintentional high-speed collisions between active satellites and orbital debris" [0], the most recent of which was in 2021.
There is more to worry about than just tracking satellites following "extremely well known paths". You could have a situation like Intelsat 29e [1] where a satellite is torn to pieces, creating a debris field with parts too small to track. You can't avoid what you can't see.
It’s a great analogy but to nitpick a little I wouldn’t say the circumference is expanded ‘a lot’ - they’re generally in LEO at 5-600km so an additional 1000 km which is essentially 3,141.5 ish kilometres on top of 40,000ish - under 8%?
Of course, each additional orbital plane is basically another set of roads but for all the internet sat constellations being planned extra distance = lag, hence we’re going to see the most growth in that LEO category
Two cars sitting still in a car park won't collide with each other, but there's no way to sit still in orbit.
Any two circular orbits at the same altitude will intersect at two points. You have to be very careful to make sure objects in such orbits will not collide with each other. It's far easier to make the orbits slightly different sizes or shapes, so that they don't intersect at all.
In any case, the need to be in an orbit to stay in orbit severely restricts your free choices for where to put your objects, so there is not as much free space as you might expect.
Given that low earth orbit satellites supposedly travel around 17,000 miles per hour, that is a frighteningly small amount of space. If we're voting, I say let's not get that crowded.
That means the satellites should be aware of each other positions, just like aircrafts. It’d take care of collision. The majority of satellites will be commercial, not military. There will be space control for them.
Also may be not 500 billions, but how about 100 millions? So 5000 times fewer.
When two satellites cross each others plane it is highly predictable and easy to maneouver an avoidance. OTOH on a freeway lane changes happen constantly are sometimes very unexpected.
We could not do that inadvertently. To block even 1% of light using Starlink sized satellites (~30 m^2 with solar panels deployed) would require tens of billions of satellites. We could do it on purpose with huge rotating solar reflectors, and it should honestly be considered as a real option, but we couldn't and wouldn't do so just by launching communication satellites.
Starlink satellites mass 300 kg each. There's about 5000 of them in orbit, so that's 1,500,000 kg.
The Earth has about 5.15x10^18 kg of atmosphere. 5,130,000,000,000,000,000 kg of gas. If each satellite aerosolized completely (which it wouldn't, metals would rapidly precipitate out of the air) then they would make up a couple parts per trillion of the atmosphere. Way lower concentrations than cloud seeding.
In contrast, routine operations of Air Force One emit around 41,000,000 kg of CO2 per year. (They fly not just the president, but his security detail, and the presidential limo to every foreign visit)