« What are the benefits of being full professor? | Main | The best sentence I read today, circa 6:36 a.m. »
Department of Uh-Oh
Until recently, nearly all the thinking about the risks of space-rock strikes has focused on counting craters. But what if most impacts don't leave craters? This is the prospect that troubles Boslough. Exploding in the air, the Tunguska rock did plenty of damage...
That is Gregg Easterbrook in the latest Atlantic Monthly, June issue, "The Sky is Falling," not yet on-line. Here are previous MR posts on the asteroid problem.
Posted by Tyler Cowen on May 10, 2008 at 06:10 AM in Science | Permalink
Comments
I don't doubt the brilliant scientific achievements of NASA, but won't they all go to waste if an impact kills us? If I was president, I would devote 90% of NASA to creating a space rock detection system. The last statistic I read said currently, we are capable of viewing 1% of space. I do not know if that is at one time, or a scanning technique, but asteroids move. If we had 99% of space covered, we could still miss one or two million, space is very large. This may be why we just leave it to fate, and work on trying to send people to the moon. But seriously, with the computing power we have today, can't we find 40,000 NASA employees(the number that worked on the shuttle(airplane) project that were just fired) to create a system that at least gives us a year notice with 100% certainty?
Posted by: brainwarped at May 10, 2008 8:50:21 AM
To put your mind somewhat at ease, let me point out that the sizes of Asteroids are believed to follow a power law distribution. This means that the kind of "dinosaur-killer" asteroids which are large enough threaten human civilization, should they hit Earth, are just not that common. This means that the mean times between impacts of asteroids of this kind is of the order of millions of years. These asteroids of > 1km in diameter are believed to number ~1100, of which 689 had been found by 10/1/2006 putting us on track to have effectively finished the job by early next decade. We are now looking to start taking on the more challenging job of locating at least 90% of the far more numerous "regional-class" asteroids ( > 140m in diameter). These latter group of asteroids are incapable of destroying human civilization (except, perhaps, by triggering a nuclear conflict, were one to land in the wrong time and place) but could cause wide spread devastation. Here is a link to my source material for this (look on p 26 of the document = page 13 of the pdf): http://www.b612foundation.org/papers/NASA-finalrpt.pdf
My take on it is as follows: I think we are putting an effort commensurate with the risk involved (i.e. there is an extremely low probability that an undiscovered dinosaur killer will impact the Earth before it is found). For regional events, the MTBE is ~5000 years, so that the risk of taking a few decades to reach the 90% level is not extraordinarily high. My understanding is that the limiting factor is not computer power but telescopes with which to survey broad swaths of sky. In short, you don't want to ignore the killer asteroid question, but neither should you panic and do a "chicken-little".
Posted by: Greg at May 12, 2008 8:36:25 PM
