3 ms·
Seems like FUD to me. Consider that 11,000 satellite-sized objects are spread over the surface of the earth-- given an entire lifetime to search, how many do y
by aerovistae 8y ago
Seems like FUD to me.
Consider that 11,000 satellite-sized objects are spread over the surface of the earth-- given an entire lifetime to search, how many do you think you could find?
It helps a little to imagine it as people instead of objects. Imagine there are only 11,000 people on earth instead of 7 billion-- just the population of one small town, spread over all the oceans and all the continents. Do you think you would ever find even one of them?
Then consider that orbital height expands the "surface area" of this satellite spread by an enormous amount....the surface area of a sphere at LEO is substantially larger than the surface area of the earth. I don't know the exact numbers, but I know it's a huge difference.
So the odds of ever finding any would be substantially lower.
I don't think this is something we have to worry about, even if it was 100,000 satellites.
But...I'm not an expert on orbital mechanics. I could be wrong, for sure. Just seems suspect to me.
- jstanley 8y agoI think you're pretty much right, with the proviso that most satellites orbit in roughly the same plane.
- bad_alloc 8y agoThe problem is that these 11000 objects are moving among the existing ones. Some are nearly stationary on a geostationary orbit, while the LEO objects tend to zip around on wildly different orbits. So they are not sitting there, waiting to be found, but each object has a certain chance each orbit to hit some other object. These probabilities go up the more we put in orbit.
- AllegedAlec 8y agoBad explanation; the LEO and the Geostationary satellites will never ever bump into each other; they are in completely different orbits that never cross. Furthermore, we don't just randomly chuck these things into orbit without any thought or consideration. They do not 'have a certain chance each orbit' to hit something.
- Michaelanjello 8y agoIt's collisions that trigger a runaway Kessler syndrome. Collisions have happened before and they are a serious concern. Just ask the inhabitants of the space station, for example. Probability dictates that sooner or later, something or someone will run into you. I don't think Starlink is doing all that they can to safeguard their satellites, especially considering that managers were fired for not prioritizing quality over speed.
- saagarjha 8y ago> Imagine there are only 11,000 people on earth instead of 7 billion-- just the population of one small town, spread over all the oceans and all the continents. But not evenly. There are certain orbits that most satellites try to stay in.
- dmurray 8y ago> the surface area of a sphere at LEO is substantially larger than the surface area of the earth. I don't know the exact numbers, but I know it's a huge difference. The surface area of the sphere is not much bigger, LEO starts a few hundred km up compared to the Earth's radius of thousands of km. Maybe 1.5x the surface area. But it expands in 3 dimensions: if the interesting part of LEO is 500km thick and a satellite takes up 10m of it from bottom to top, that's 50,000 concentric Earth-surfaces.
- Jabbles 8y agoThen consider that orbital height expands the "surface area" of this satellite spread by an enormous amount....the surface area of a sphere at LEO is substantially larger than the surface area of the earth. I don't know the exact numbers, but I know it's a huge difference. Surface area grows with the radius squared. Starlink are meant to be at (a max of) 1200km above the surface. Earth radius is 6371. ((6371 + 1200) / 6371)^2 = 1.41 There is 41% more surface area at their orbit than at sea level.
- emiliobumachar 8y agoConsider that each person is as fast as Flash in the comics. And if two ever collide, each blood droplet just keeps going.