Image Credit: NASA
The graphic above shows the space debris that NASA follows around the Earth. Most of this is just junk, pieces of spacecraft and satellites that have disintegrated (or that have been purposefully destroyed, including by the United States). The stuff closest to the Earth is gradually slowed by the tenuous outer reaches of our atmosphere, and it falls back to Earth and burns up. A lot of the meteors you see at night are actually human junk following back to Earth. But the higher stuff can last for decades or hundreds of years, because the atmosphere gets even thinner the higher you go.
Space debris is dangerous stuff. All the debris in the picture above is large enough that we can detect it with radar. But there is a lot more debris that is quite small -- flakes of metal or chips of paint or pieces of bolts and nuts and rivets -- and we can't detect it from the Earth. This junk can be moving at very high speeds relative to other satellites (hundreds or thousands of miles per hour), so if it hits a satellite, it can cause a lot of damage. And if that other satellite happens to be a space shuttle or space station, a single, undetected screw could puncture a pressure hull, exposing the astronauts to the vacuum of space, likely killing them.
For every space shuttle flight, NASA calculates the risk that the shuttle will be catastrophically lost due to space debris. As long as that risk is less than 1-in-200, then the launch can go forward. If the chance is higher than 1-in-200, then the flight is reviewed.
A news story about the upcoming Hubble Telescope repair mission announced that the chance that the shuttle Atlantis could be lost to space debris is 1-in-185, or too high for NASA regulations. So, NASA is going to have to review the risk and see if it is too dangerous for our astronauts.
Now, something in that risk doesn't quite add up to me. The mission lasts 11 days, and it has a 0.5% chance of being destroyed by space debris. If those odds are constant for any satellite in that orbit, then the Hubble Space Telescope (launched in 1990) has had a 95% chance of being destroyed by space debris over its lifetime. But obviously it hasn't been destroyed (knock on wood). So, it seems likely that the odds of space debris destruction are too high.
There are problems with my analysis. Hubble is smaller than the space shuttle, so that reduces its likelihood of being hit. The Hubble may be capable of surviving hits that would puncture the space shuttle. And the likelihood of debris impact may be going up, as both the U.S. and China have recently practiced shooting down satellites, which created a lot of new debris, raising the chances of a hit.
Space debris is an important concern for any satellite, whether manned or not. We need to work to ameliorate the problem, by means such as building satellites that can be de-orbited and by not destroying satellites for military purposes. However, the space shuttle has larger safety concerns than space debris. It is good and right for NASA to be concerned and alert about space debris, but it would be a little overkill to cancel a mission over a 1-in-185 worry, when the shuttles have roughly a 1-in-70 chance of other fatal mishaps. Let's go fix Hubble!