4 ms·
This is a bit of a misunderstanding of the problem. Imagine you took a car and you drove in a straight line in any direction in the US where you'd loop back at
by rjf72 8y ago
This is a bit of a misunderstanding of the problem. Imagine you took a car and you drove in a straight line in any direction in the US where you'd loop back at the end. And now imagine somebody else did the same thing. How likely would you be to collide? You'd have either 0 or 1 collision point in your paths. Usually it'd be 0. And even in the case where you had 1, the chances of you both hitting that point at the exact same time would be infinitesimally low. Now have thousands of cars do the same thing. And all the above still holds true over one 'iteration', or two or three or... However, any two cars that have at least one intersection point will eventually collide (excepting a perfectly synchronized orbit timing) so long as they keep going. It might take a long time, as in potentially thousands of years, but if you get enough cars doing this and you run them long enough - eventually you'll get a pair that will have a collision within some foreseeable time. This is analogous to how things collide in space, except space is much bigger.
In other words Kesseler syndrome does not create some sort of impassable barrier -- in the above analog imagine your risk of getting hit by a car if you stand in one place for 30 seconds - again, basically 0. It just makes collisions, over the longrun, more likely to occur between orbiting bodies. There's absolutely no risk of impairing our ability to get off Earth -- just our ability to circle around it for years in fixed orbits. And in reality, I think it's almost certain we'll eventually see a Kessler syndrome type event. The first thing two capable nations involved in an armed conflict would do is to destroy each others' satellites, resulting in a severe mutual crippling of both offensive and defensive capabilities (as well as civil chaos - imagine if GPS/comms were all taken out tomorrow!) at very low cost. So the only way we do not see a Kessler Syndrome type event is if there's literally never a significant armed conflict between any of US/Russia/China/India, and whatever other future nations develop this technology. The odds of that seem to be pretty much 0.
- mokus 8y agoWhat makes the Kessler scenario worse is that every collision creates dozens or hundreds more cars, increasing the probability of a subsequent collision. Through positive feedback, large classes of orbits become unusable for meaningful mission lifetimes, even if the odds were still low for an individual spacecraft to have a collision in some short time window.