Video: Debris from an unprecedented collision of two satellites on Tuesday is expected to spread out over time. An animation from the firm Analytical Graphics simulates what might happen (Courtesy of Analytical Graphics, Inc; www.agi.com)

Two satellites smashed into each other on Tuesday, creating a mess of space debris that is still being counted.
The collision, which involved an Iridium communication satellite and a defunct Russian satellite, is the first to occur between two intact spacecraft 鈥 but experts say it is unlikely to be the last.
How did this happen? New 杏吧原创 takes a closer look at whether the collision could have been avoided.
Advertisement
Is anyone keeping an eye on everything that鈥檚 up there?
The most comprehensive catalogue of space objects is maintained by the Pentagon, which uses optical telescopes and radar to track more than 18,000 objects measuring 10 centimetres across or more. The US Air Force compiles data on the orbits of these objects. A large fraction of the information is made publicly available on the website .
Does the US government calculate the risk of collision for every spacecraft?
No. The military regularly calculates the risk of collision for priority spacecraft, including military satellites, the International Space Station, and the space shuttle, when it is in orbit, warning operators if there is a risk of a close encounter.
But not all satellites are given that treatment. 鈥淭here just isn鈥檛 the manpower or computer capabilities to do that at this point,鈥 says Andy Roake, a spokesperson for Air Force Space Command.
Why is it so hard to calculate whether two objects will collide?
If every object orbiting the Earth stayed on a simple path, multiple observations of an object would yield precise estimates of its orbit and position. But orbits change.
Unexpected glitches 鈥 such as accidental releases of fuel 鈥 can shift a satellite鈥檚 orbit.
But so can the drag from Earth鈥檚 atmosphere, as well as gravitational tugs from the Sun, Moon, and Earth, which is somewhat squashed in shape. The military uses models of these environmental perturbations to help refine its estimates of satellite orbits.
How well can the military estimate satellites鈥 orbits?
That鈥檚 not clear, as the Pentagon may not release its best estimates of satellite orbits.
But space analysts say the data the Pentagon does make publicly available is not precise enough to accurately estimate the probability that two objects might collide.
Tuesday鈥檚 smashup is a case in point. Using a collision prediction program to perform a retroactive analysis of the satellites鈥 orbits, aerospace analyst found that the Pentagon鈥檚 public data showed that the two satellites would have missed each other by 584 metres.
But he says the uncertainty in that distance could be several kilometres. 鈥淭here鈥檚 no reason looking at the data that was available [to think the Russian satellite] was an immediate threat,鈥 says Kelso, who has been working with Iridium to assess the risk that the debris could collide with other satellites in the firm鈥檚 fleet (see image above right).
Could Iridium have predicted a collision?
Since Iridium likely communicated directly with its now-destroyed spacecraft, chances are the firm had a more precise measure of its position and orbit than could be determined by radar or telescopes.
Still, uncertainties in the orbit of the Russian satellite might have made it difficult for the firm to assess the probability of a collision. 鈥淚t鈥檚 like driving a car with a GPS unit and the windows blacked out. You know where you are, but you have no idea where the other cars are,鈥 says , a technical consultant for the Secure World Foundation in Superior, Colorado.
Indeed, according to , John Campbell, Iridium鈥檚 executive vice president for government programmes, said at a forum in June 2007: 鈥淓ven if we had a report of an impending direct collision, the error would be such that we might manoeuvre into a collision as well as move away from one.鈥
At the time, Campbell said that Iridium was receiving an average of 400 reports per week of objects coming within 5 km of one of their satellites. The reports were issued by the US Strategic Command鈥檚 Joint Space Operations Center. 鈥淭he ability actually to do anything with all the information is pretty limited,鈥 he said at the time, putting the risk of a collision from the close approaches at 鈥渁bout 1 in 50 million.鈥
But an Iridium spokesperson told Reuters that the company did not get a warning before Tuesday鈥檚 collision: 鈥淚f the organisations that monitor space had that information available, we are confident they would have shared it with us.鈥