Please read the update at the bottom of this post; I correct a couple of errors I made.
In summer 1908, an explosion over a remote area in Siberia near the Podkamennaya Tunguska river devastated hundreds of square kilometers of land. Trees were flattened, the area scorched, and the pressure wave from the blast was felt around the world.
Decades later, it was determined that a small chunk of asteroid or comet came in at an angle and exploded high over the ground, and the fireball is what did the damage. It was assumed the rock was 50-100 meters in size, and the explosion yield was 10-20 megatons.
However, a new study just released by the folks at Sandia Lab indicates that the rock may have been smaller, and the yield smaller as well. The ever-increasing speed of computers has allowed them to make extremely detailed three-dimensional models of the explosion -- models that are incredibly difficult to make due to the fiendishly complex nature of how air moves under extremely high-pressure and temperature conditions.
Their results show that the blast created a tongue or finger of pressure which blasted downward from the explosion itself, as opposed to the explosion making a big spherical blast wave. Scientists have always assumed the blast wave expanded spherically, which means it takes a lot of energy to do the damage seen. However, a finger of destruction is more concentrated than a sphere, and can do more damage with less energy. So the rock and explosion were smaller than previously thought.
In my mind, that makes Tunguska even scarier than it was before: smaller rocks are more common in space. This means events like what occurred in 1908 may happen more frequently than we thought.
Unfortunately, the press release from Sandia is very scant with details. How big is their revised size for the meteoroid? What was the yield of the explosion? How does this change the frequency of impact?
They have some cool videos on the site too, but they are not labeled and have no supporting info.
The first couple seem to contradict each other as to how the rock came in, too. Maddening. If I get a chance, I'll try to call some of the people behind the study and get more info. Chapter 1 of my book is all about asteroid impacts, and I still have time to make some edits if needed!
UPDATE: I called Neil Singer at Sandia Labs about the press release and he cleared up some issues. First, I misread the release: the yield of the explosion was listed; it was something like 3 - 5 megatons (down from the earlier estimates of 10 - 20). That makes the focusing of the fireball very efficient! That's interesting. Second, the size of the meteoroid itself isn't given because it depends on too many other factors like speed, composition, and the like. This is actually in the press release itself and I somehow stupidly missed it. Now, the energy released by an incoming chunk of material depends pretty much linearly on the mass, so if the explosion itself was a factor of, say, 1/4 as big as previous estimates, then the mass of the object was 1/4 of the older estimate as well. That means its diameter was very roughly 60% of what was previously thought.
Also, as I said the videos aren't labeled (Mr. Singer said he was going to fix that), but I think I understand the first two now. The second one shows the blast from the side, with the bolide coming in at an angle of 35 degrees from horizontal and exploding well above the ground. The first one is most likely the same thing, but seen from a different angle, along the path of the incoming rock. That makes it look like it's headed straight down, but it's just a perspective effect. The explosion of the rock itself is dwarfed by the focused blast wave, so you don't see it clearly. The two videos are compatible, and don't contradict each other.
As far as changing the statistical frequency of such impacts, that's tough to nail down as well, because there is still too much fuzziness in the size estimate. The next time I talk to some of my asteroid expert friends I'll see what they think of this. No doubt David Morrison will be sending an email out about this result soon!
So my apologies to Sandia and Mr. Singer for what were essentially my own mistakes.