How Common Are Satellite Collisions in Space?

[tribdog]

This guy I work with is under the impression that it must be difficult to keep satellites from running into each other. That they are constantly brushing by each other.
My argument was to say pretend the number of satellites was the total number of cars in the world. how often would you expect those cars to run into each other. they would be lucky to even see another car.
I'm just curious as to how many satellites are actually up there and what percentage are in a geosync. orbit. and finally is it possible to orbit in a plane that doesn't bissect the Earth evenly?

 

[LURCH]

Last question first; no, all orbits have the Earth's center of gravity at their center.

 

[tribdog]

Last question first; no, all orbits have the Earth's center of gravity at their center.

I knew that, I don't even know why I asked, but thanks

 

[russ_watters]

There are somewhere on the order of 10,000 satellites up there, the vast majority are are in leo, but I don't know the fraction (somewhere there is a map, but I can't find it. Enigma...?). And thousands of pieces of debris are being tracked.

You're pretty much right about collisions - its a concern, but not a huge one. For starters, most satellites orbit in the same direction (though at different inclinations). Mainly though, its just that 10,000 satellites spread over millions of square miles of space doesn't make the odds of collision very high.

That said, the space shuttle (Columbia?) once hit a paint chip on an opposing orbit and it went through something like 4 of 6 panes of glass on the windshield. At a closing speed of 30,000mph, a bolt would go through the shuttle or a satellite like a hot knife through butter.

edit:HERE's a map. The link says only about 4-500 of 8,000 trackable items are operational spacecraft . That seems low, but I guess its possible.

Another link on debris: http://www.corrosion-doctors.org/Space/Debris.htm

 

[LunchBox]

If you go to NORAD's website (http://www.norad.mil/) they are constantly tracking and updating orbital elements on everything that they are able to track.

The NORAD TLE (two line element) gives you all you need to create and propagate the orbit of any orbiting object.

Here are a few (link). Enjoy!

Cheers...

 

[600burger]

Here's another.

http://science.nasa.gov/RealTime/JTrack/3D/JTrack3D.html

Acctually found this one from i-am-bored.com! Doesn't have as many as the others, but its a java tool that's suposedlly in real time (like how am i going to check). You can also search spefic satellites by their name/abv. and get their info: flight path, alt., discription, mission, and origin. Its pretty fun to look at, and the 3D java map anit half bad.

-Burg

 

[enigma]

That.

Is.

So.

Cool!

So much to see there. You see the huge ring of geo-synch sats. You even see the bulges above central US and central russia where the gravitational perturbations stabilize the orbits.

You can see the various GPS satellites orbitting about halfway to Geo-synch. You can see the swarm of LEO satellites.

You can even switch on the groundtracks so you can see why they placed the GPS satellites where they did.

*bookmarked*

 

[BobG]

This guy I work with is under the impression that it must be difficult to keep satellites from running into each other. That they are constantly brushing by each other.
My argument was to say pretend the number of satellites was the total number of cars in the world. how often would you expect those cars to run into each other. they would be lucky to even see another car.
I'm just curious as to how many satellites are actually up there and what percentage are in a geosync. orbit. and finally is it possible to orbit in a plane that doesn't bissect the Earth evenly?

If you want the day by day boxscore for number of objects currently in space (plus the number that have been in space), try this link: http://celestrak.com/satcat/boxscore.asp.

There's about 3,000 'living' satellites up there (some don't work very well but are kept alive as an orbit test bed). There's around 9500 objects in orbit, counting dead satellites and larger pieces of satellite debris.

Even though your car analogy is pretty good for the most part, there are some orbits that are more popular than others. The geo belt is one of the more popular orbits. There's around 550 to 600 satellites in the geo belt, which gives you around 1 every 275 to 300 miles, if they were evenly spaced around the world. If you look at the JTrack linked by 600burger, you'll notice most people want to put their satellites over populated areas (most of the geo satellites are communications satellites, including satellite TV). Almost no one wants to station their satellites over the mid-Pacific.

You're right that there are almost no collisions between satellites - in fact, there's only been one confirmed collision: the French Cerise satellite and it's own rocket (that makes sense since the more similar two orbits are, the better the chances of intersecting each other).

That doesn't mean it isn't a worry. The International Telecommunications Union divies up slots in the geo belt based on the frequencies used on a satellite. Two satellites using the same frequencies have to be far enough apart that a satellite antenna can receive signals from one satellite without interference from the one next to it. If the two satellites use completely different frequencies, the satellites could be located on top of each other as far as the ITU is concerned. So there are some satellites that wind up working pretty close to each other.

Toss in dead satellites and you have some more problems. When you're done with your satellite, you're supposed to push it to a higher altitude so it doesn't clog up the geo belt. In the early days, they didn't worry about that - plus some satellites just die unexpectedly. Those satellites drift back and forth across the gravitational valleys Enigma mentioned (except he called them bulges). Fortunately, the orbit planes of dead satellites drift out of the geo belt due to the Moon and Sun, but they still pass through a couple times a day. Eventually (a little less than 30 years later), their orbit plane drifts right back through the geo belt. The number passing back through just keeps increasing.

If you're just talking about today, I'd say you're probably more right than your friend. A few people have problems because of their particular location, but most (especially all the satellites not in the geo belt) have very few close encounters with another satellite. But the frequency of close encounters in the geo belt is increasing fast enough that quite a few satellite operators are getting concerned about it.

It's not much fun to share space with an operational satellite from another country, knowing both of you have to make orbit adjustments to compensate for orbital perturbations. Neither of you know when the other plans to maneuver (although your smart operators can record a history of their neighbor's past maneuvers and make a pretty good guess) and there's enough political bureaucracy that neither can talk to each other. If you maneuver out of the other satellite's way without knowing what he's planning to do, there's a good chance you're just maneuvering into the path of his next maneuver. Without knowing, or at least being able to make a really good guess, the best option is still to rely on the fact that it's a small satellite and a big sky. In other words, just wait and see if the two satellites miss each other - they virtually always do (except that one time).