[Note (July 2012): I've made some minor edits to this blog entry since it was originally posted, cleaning up the text a bit and updating a dead link.]
The UK newspaper "the Sun" is not exactly a bastion of journalistic integrity -- it's a bit like the Weekly World News with a Monty Python accent. So I'm not too surprised that it would have some stupid stories in it, but one they had on their site today really angered me. It's short, and here it is in its entirety:
AN [sic] asteroid big enough to wipe out mankind is overdue and could strike at any time, scientists fear.
It was thought an impact of the size that wiped out the dinosaurs happened only once in 1/2 million years.
But new evidence studied in Australia suggests a hit equal to 500 Hiroshimas occurs every 1,000 years -- and the last was 4,800 years ago.
First off, which scientists? No names, nothing. That's some journalism!
Next, "wipe out" is way too strong a phrase. An asteroid big enough to kill every human on Earth would have to be roughly 10 kilometers across or bigger, and we have most of those charted (half the near-Earth asteroids bigger than 1 km have already been charted, and 90% is the goal by the end of 2008). Yeah, we're scared of them -- you'd have to be nuts not to be -- but we also know that none that big is headed our way in the immediate future.
The really scary ones are the smaller ones -- a few hundred meters across -- that can come in undetected and kill millions of people if they impact over a city or near a coastline. That would be a disaster of unprecedented proportions for humanity-- but it wouldn't cause an extinction event.
So strike one for The Sun.
Second, it's not thought that an impact like the dinosaur killer happens every 500,000 years. If that were true, we wouldn't be here! The last big impact of that size was... when the dinosaurs were wiped out. Duh. The average time between impacts like this is about 200 million years (see page 10 of David Morrison's paper "Earth Sterilizing Impact"). It's true that an impact that could destabilize civilization happens every million years or so, but this is different than an extinction-level event.
The last sentence in the article is accurate. I already discussed this before. There is some evidence to suggest we get hit more often than previously thought, but the evidence is scanty. A lot more data are needed.
But what really irritates me is the idea that we are "overdue" for an impact. That's garbage. Impacts are a stochastic process; they're random. We could get hit tonight by a big one, or it might be a million years from now. They don't happen like clockwork, so the impact frequency is a statistical one, like saying how often you get heads when you flip a coin. You have a chance of getting heads every time you flip it, but you cannot say with certainty that you will the next time. And if you flipped tails last time, it doesn't mean it'll be heads the next.
The same with asteroids. The idea that we get a big impact every millions years or so doesn't make asteroids rain down on a schedule.
Asteroid impacts are scary, and there is absolutely no need to make them any scarier by fear mongering like this. Worse, it adds to the "cry wolf" problem that is already plaguing impact scientists -- whenever a potentially hazardous asteroid is announced, and sails past us, the public get that much more inured to the real threat from these things (even when scientists are saying the right things). It doesn't help to have gas thrown on this fire by irresponsible journalism.