So by now, everybody has read about the meteor which broke up over the Chelyabinsk Oblast late last week. The reportage on it was pretty interesting in the beginning -- a lot of between-the-lines skepticism was being put out there by American news outlets. I was a little wary myself, too, as a lot of the initial reports from Russia were pretty sketchy, buffeted primarily by Russian dashboard cameras, which, in our Photoshop and AfterEffects age, are probably not at the top of our list of "reliable sources." For people who care about Cold War science and technology, of course, there's the additional fact that ChelyabinskOblast is a major site for secret Russian military-industrial developments. It'd be like reports of suspicious explosions around the Nevada Test Site, or Los Alamos, or Pantex. Chelyabinsk Oblast is the home of Chelyabinsk-70, the Soviet Livermore, and just north of it is the city of Sverdlovsk (now Yekaterinburg), the site of a 1979 anthrax leak that the Soviets tried to cover up by claiming it was something more "natural" in origins. Add in the legacy of the Soviet response to Chernobyl, the relative rarity of this sort of meteor strike -- once a century is the frequency that's been cited -- and the extreme rarity of something like this happening over inhabited land -- most of the planet is devoid of human occupation -- and some skepticism in the absence of solid evidence was, I think, not unwarranted. Eyebrows raised, including mine, but apparently it all checks out.
How powerful was the explosion? NASA currently is saying it is the equivalent of a 500 kiloton blast, which is a lot. 500 kilotons is (as you can see) half a megaton, is about the upper-limit of a pure-fission nuclear weapon, and is, as journalists love to breathlessly relate, some 20-30 times the power of the bombs that hit Hiroshima and Nagasaki. That the only result was a lot of injuries caused by windows blowing inward -- something that occurs with a shock wave of one pound per square inch or above -- is attributed to the fact that the meteor exploded many miles above the ground, away from the city.
Personally, I cast a dubious eyeball towards the comparisons of natural phenomena with nuclear weapon energy releases. It's an incredibly common trope, though. Wikipedia's coverage of the 2004 Indian Ocean earthquake is actually quite reflective of how this gets talked about, even if it is somewhat dorkier in its citation of units than the average journalistic account:
The energy released on the Earth's surface only (ME, which is the seismic potential for damage) by the 2004 Indian Ocean earthquake and tsunami was estimated at 1.1×1017 joules, or 26 megatons of TNT. This energy is equivalent to over 1500 times that of the Hiroshima atomic bomb, but less than that of Tsar Bomba, the largest nuclear weapon ever detonated. However, the total work done MW (and thus energy) by this quake was 4.0×1022 joules (4.0×1029 ergs), the vast majority underground. This is over 360,000 times more than its ME, equivalent to 9,600 gigatons of TNT equivalent (550 million times that of Hiroshima) or about 370 years of energy use in the United States at 2005 levels of 1.08×1020 J.
Lots of numbers thrown around, lots of energy involved, yes, but what does it mean? I have two major objections to this form of analysis, where nuclear weapons are used as some kind of barometer for general energy release. The first is about the character of energy release is important -- because it affects how these things are felt at the human scale. The second is about whether these sorts of comparisons are actually clarifying to the general public.
On the character of nuclear and non-nuclear blasts
The key thing about nuclear weapons is that they discharge most of their energy as heat and blast. Most of the energy release occurs over a very small amount of space and time. You can essentially regard the physics of a nuke as being a the creation of a tiny point in space that suddenly is heated to tens of millions of degrees, and this results in all of the effects that we are pretty well familiar with. The results are extremely localized: even the massive Tsar Bomba had a fireball only five miles in diameter, which is huge by human standards but minute by geological or geographical standards. The vast majority of the energy is discharged within a few milliseconds, as well. It's a bang that matters on a human level because a huge amount of energy is released very quickly in an area of space that corresponds fairly well to the sizes of human habitation centers. The fact that a huge amount of that explosive energy (around 50%) is translated specifically as a blast wave -- the thing which destroys most of the houses and people and all that -- is perhaps the most salient thing about nuclear explosions from a human standpoint.
One can see the point in distinguishing about the amount of energy over time and space by considering the Sun. The amount of energy from the Sun that reaches the Earth's surface every moment is tremendous -- equivalent to billions of tons of TNT -- but it is spread out over a huge area, so instead of totally obliterating us when we go outside, it pleasantly warms us and maybe, at its worst, gives us a painful, peeling burn after several hours of intense exposure. So that is a lot of energy released over a short unit of time, but it is diffused over a very large area. The converse situation can also be considered: a given city absorbs an immense about of energy from the Sun over the course of a year, but because it is spread out in time, it isn't anything like a nuclear explosive's yield.
What about meteors? Yes, there's a lot of kinetic energy in those rocks falling from the sky. But they don't translate most of that energy into shock and heat. Even the famed 1908 Tunguska event reached temperatures "only" in the tens of thousands of degrees, as opposed to the tens of millions. You can regard the kinetic energy of such a thing as 20 megatons of yield, but the actual blast effects were more than five times less than that because the energy didn't transfer very efficiently. (Still quite a blast, though!) The Chelyabinsk meteor was much smaller than that and it exploded in the atmosphere -- a reaction more like a chemical explosive than a nuclear one. So in some sense, comparing a meteor explosion to a nuke is better than comparing an earthquake or a tsunami to a nuke, but it's still not very exact.1
On the public understanding of nuclear explosions
My other issue, though, is about public understanding. The Chelyabinsk meteor exploded with an energy release of 500 kilotons. Is being told that going to mean anything to the average person, except to say, if it had hit the city, it would have been equivalent to a nuclear explosion? Does saying it is 20-30 times more powerful than Hiroshima mean anything to the average person, except the conjure up potentially incorrect misconceptions of what those effects would be for their cities? The truth is, as we've seen again and again, the average person has almost no intuitive point of reference for making sense of nuclear explosions. Heck, I barely have any point of reference and I'm constantly searching for them! The average person cannot distinguish between the results of a megaton-range explosion and a kiloton-range one unless you translate it into terms that are meaningful to them. That was the whole point of the NUKEMAP: to take these numbers and try to come up with geographical representations that make intuitive sense.
And so here's the problem: since the physics aren't the same, any intuitive generalization made from a nuclear analogy will be necessarily highly flawed. The effects of the Chelyabinsk meteor were not really equivalent to the 1952 Ivy King nuclear detonation, which was a nuclear explosion of 500 kilotons in yield. Even the Tunguska event was not really equivalent to a five megaton nuclear explosion in its phenomenological effects, even though it was a pretty big boom.
"Hey," you object, "we're just trying to communicate to people it was a big explosion!" Yeah, I know, but it's misleading. If you want to communicate the size of things, don't talk about the energy release in terms of nukes -- the effects aren't the same. If you want to convey the effects... talk about the effects. A better way to talk about the Chelyabinsk event is to not talk about the energy output but instead to talk about the radius and nature of effects -- exactly how many square miles of the city had their windows blown out? Even just saying that thousands were injured by broken glass does a lot more work to convey what this was -- and how scary it was -- than anything else. If you want to say, "if it had directly hit the city before blowing up" -- a big counter-factual but whatever -- "so-many square miles would have been destroyed," that too would make a lot more sense.
Using nukes as a genericized way to talk about energy output is highly misleading both from the point of view of the expert, but even more so from the point of view of the layman. I really don't see the advantage to it either way. I fear in talking about asteroids as nuke equivalents people may be trying to emphasize their threat -- which is totally legitimate -- but at the same time may end up inadvertently down-playing nukes. After all, if a 500 kiloton airburst only knocked in a few windows, what's all the fuss? Yes, we can explain why they are different -- but we wouldn't have to do that if we just described the effects better in the first place, rather than taking a lazy recourse in how-many-joules-equals-how-many-megatons equations. Rather than using nuclear terminology, and then down-scaling to explain how the effects are actually not quite the same... just tell us the actual effects and forget the nukes! If one must do things in response to nukes, do it the other way around: find out the actual effects of the meteor (or whatever), then tell us what yield nuke gives you those effects. It's less sensational, sure, but it'll help people understand both meteors and nukes better.
- For helping me think through the physical comparisons, and providing some interesting references, I was aided by e-mail conversations with my AIP colleagues Charles Day, Paul Guinnessy, and Ben Stein, as well as my old Harvard colleague Alex Boxer. Any interpretive errors are of course my own! [↩]