PARIS European space scientists were aware of the potential for a close encounter between Russian and U.S. satellites before they crashed.
But the difficulty of predicting orbits and "noise" from thousands of pieces of debris made a definitive prediction of a collision impossible, European officials said on Thursday.
"The 'catalogue' of objects and debris showed a possible approach between the paths of the two satellites but an approach doesn't necessarily mean a collision, and you would need more information to be certain," said Philippe Goudy, deputy director of the French space operations control center at Toulouse.
"It is not a case of two satellites coming together out of nowhere; they had been followed. The U.S. catalogues can give an alert but these are not necessarily completely exact."
Washington reported the first collision of its kind, between a privately owned U.S. communications satellite and a defunct Russian military satellite, on Wednesday.
Each day, officials at the Toulouse center, operated by France's CNES space agency, examine catalogues of radar tracking data supplied to the world's space-faring nations by U.S. Space Command, the only steady source of orbital mapping information.
"What we do for the 15 satellites we control is to watch them every day in orbit and we verify whether there is an alert. If so, we look for more radar information," Goudy told Reuters.
The crash occurred on Tuesday 485 miles above the Arctic in a crowded polar orbit used by satellites that monitor weather, relay communications and perform scientific surveys.
"If you want a global coverage with high resolution, this is where you go," said Bo Andersen, director general of the Norwegian Space Agency, referring to the 800-km altitude level.
"If you go much further down, you get more drag from the atmosphere which shortens the life of the satellite; if you go higher up, the resolution gets worse," he said.
The collision happened not far from the orbit of a defunct weather satellite blown to pieces by a ground-based missile in a Chinese weapons test in 2007. European and U.S. officials argue the resulting debris made it harder to identify crash risks.
The United States used a naval missile to destroy a tank of toxic fuel on a defective U.S. spy satellite last February.
Scientists say predicting collisions is difficult because of the unpredictable behavior of other objects, solar radiation and the gravitational effect of the moon and earth, while at 800 km molecular wisps of atmosphere can gently skew orbits.
"You can calculate potential collisions some time in advance but orbits change, especially when you have a defunct satellite which you cannot correct," Andersen said.
It was not immediately clear whether the U.S. government was aware the satellites were likely to pass close to each other.
One senior U.S. official, who asked not to be named, declined comment "for several reasons."
The Russian satellite lacked a propulsion system and could not have been moved to safety. It was not clear whether Iridium, which owned the U.S. craft, tried to avoid the collision or had warnings about what space officials call a "close conjunction."
An Iridium spokeswoman did not say whether it had been warned but said the firm always monitors its satellites closely.
Even when an alarm is raised, officials said it may be decided the risk of collision is too small to justify moving a satellite, something that takes it temporarily out of service.
The chances of a space collision are so low that only three damaging crashes of any significance have been recorded involving a satellite and debris, France's Goudy said. In 1996, a French spy satellite called Cerise was severely damaged by a piece of debris from the rocket that launched it.
The United States tracks debris or micro-meteorites down to 10 cm wide, but objects as small as a scrap of peeled-off paint can pose a threat once they start hurtling through space.
Andersen said the incident would reinforce calls for a new European system to keep track of dangerous shrapnel in space.
(Additional reporting by Andrea Shalal-Esa, Jim Wolf; Editing by Robert Woodward)