Almost on the heels of the announcement of a test by North Korea, came confirmation of the test through seismic information reported by first South Korea, then the USGS.
It was clear that something had happened. It was undoubtedly an explosion -- the location was in an area that was not general active seismically, and the indications were that the depth was too shallow to be anything other than a blast. However, almost from the start, it was equally clear that something was just not right. There seemed to be a mismatch between what the North Korean government was reporting, and what instruments on the ground were showing.
After a day of reflection, it seems clear that what happened in North Korea was either a fake, or a failure. Unfortunately for the world, that doesn't matter.
Seismic Data
The Pacific Rim, including the waters off the Korean Peninsula, is among the most seismically active regions in the world. North Korea itself is remarkably stable (in a seismic sense, if in no other), but it's active neighborhood means that there are plenty of seismographs listening in nearby. A number of these are available for
public access, and what they have to say about the event.
|
Seismic events in Asia over the last week |
One way we can see the odd nature of the event in North Korea is to compare it to other events that happened in the region. In the image above, the event in North Korea is marked by the small orange square. The larger pair of orange squares to the south represents a 6.0 earthquake that occurred in the Philippines a few hours later. But before we look at either of this events, let's look at that little yellow square that stands alone near the bottom of our map and east of the Philippines.
That square marks an earthquake that happened near the tiny Island of Yap on October 5th. I've picked that event because, in this very active region, it happened to occur in a relatively quiet period, so it's easily spotted on its own, and because with a magnitude of 4.8, it's not too far from that assigned to the event in Korea.
|
Location of Earthquake off Yap |
Being in open ocean, there were fewer seismographs surrounding the Yap event, but there was a seismograph in Matsushiro, Japan.
|
Matsushiro, Japan -- Oct 6 |
At around 1252 UDT, the trace shows a small event. Compression waves from an earthquake, also known as P-waves, travel at around 6km a second through the crust. The Yap earthquake occurred at 1241 UDT, so the 11 minute lead difference is what you would expect in a seismograph located about 4,000km away -- which happens to be just the distance between Yap and Matsushiro.
|
Inchon, South Korea -- Oct 6 |
A couple of minutes later, the same event is visible on the seismograph at Inchon, South Korea.
|
Mudanjiang, China -- Oct 6 |
And finally, the small quake off Yap is registered by seismograph in China that's located just northwest of North Korea. The 4.8 quake is visible on seismographs at even greater distances, drawing a faint squiggle on a seismograph at Magadan, Russia more than 20 minutes after the event.
That's October 5th. Now, let's take a look at what happened in the middle of the morning, on October 9.
|
Location of 09 Oct 06 Event |
On this map, the orange star represents the location of the North Korean event. The other symbols on the map show historic seismic events, giving a good indication that this event was not part of the typical activity in the region. Now, let's look at the seismic data for that day.
|
Inchon, Korea -- 09 Oct 06 |
When looking at this trace, the first reaction tends to be "wow! There it is," but the large amount of activity showing up down around 1100 UDT isn't the event in North Korea. All that heavy duty activity is the trace generated by the 6.0 quake down in the Philippines. The red box shows the time period in which the event occurred, a time period marked by no visible movement.
|
Mudanjiang, China -- 09 Oct 06 |
The Mudanjiang site is actually marginally closer than Inchon. Once again, you can see the echoes of the Philippine quake, but it shows no obvious deflection during the time of the North Korea event.
So why would the Yap quake be so clearly visible more than 8,000 kilometers away, when the North Korea event is so obscure even for sites less than 400 km away?
There are several reasons why the North Korean event could be so hard to spot. First, it was rated between 3.5 and 4.2, which is smaller than the Yap quake. Second, the shallow locations mean that the rock surrounding it was less dense. The Yap quake occurred some 50 km below the surface, in a zone where the waves were easily transmitted by dense material. A "quake" centered less than a kilometer below ground is like trying to ring a bell made from Styrofoam -- even if you hit it hard, there's just not a lot of signal.
On it's own, the lack of significant deflection at nearby sites is not 100% proof that what happened in North Korea was not a nuclear event.
However, it does indicate a few things about the scale of the event. A nuclear weapon, even one on the scale of the US "Trinity" test, or the first bombs dropped over Japan, produced a large concussive force that would typically be expected to inject more energy into deep strata and produce a more clear-cut "signature" at seismographs across the region. These signatures can be wrong (as in 1997 when some scientists worried that Russia had violated the test ban treaty), but they're hard to erase completely. Those first US weapons produced yields equivalent to between 5,000 and 20,000 tons of TNT. That's typical for a "standard" nuclear warhead. In fact, it's easier to produce a blast this size than it is to go much larger or smaller. But the numbers for the North Korea event suggest what would be expected from an explosive yield of about 100-500 tons -- something about 1/50th the size of the Hiroshima bomb.
Now, it is possible that the North Koreans attempted to build a small weapon. They have a limited amount of plutonium available, and they may have been trying to divide it into as many devices as possible. But building a small bomb is difficult, requiring much more sophistication than a device in that 5K to 20K window. And building a device that yields 100-500 tons? No. You just can't easily get a successful nuclear blast that size.
There are only two real possibilities for what happened in North Korea. Either this was not a nuclear device at all, but a blast generated through conventional explosives, or it was a nuclear test gone wrong -- "a fizzle."
It was a fake, or a failure.
How could it have failed?
If the device was a fake -- just a lump of TNT or C4 playing a role -- we should know when we get the results of planes "sniffing" the air for released isotopes. But if it was a failure... well, there are several consequences.
|
Nuclear Implosion Device (courtesy Wikipedia) |
Because of the nature of their nuclear plants, the material North Korea has available for bomb-building is plutonium. That's unfortunate (for them), because enriched uranium is really much easier to work with, and the designs for uranium weapons (the so-called "gun type" device) are so simple that the US uranium bomb was not even tested before it was dropped on Hiroshima.
A plutonium bomb is a different animal. Basically, it contains a sphere of plutonium that must be brought to critical mass through the compressive effects of an explosion. To do so, the explosion has to be exceptionally well-timed, giving nearly equal pressure from all sides. During the Manhattan Project, extreme care was taken to make the triggers scattered around the plutonium exactly the same. Every wire was measured to a thousandth of an inch. Every initiator made precisely alike. Also, the device is equipped with an "initiator" composed of some materials capable of throwing out a big, sudden burst of neutrons to help get the weapon started (I'm intentionally leaving out a lot of information on initiators).
So what went wrong in North Korea? If they failed to take extreme care in compressing the plutonium, the bomb could have become uneven, blowing some of the material off to the side before it could all be driven to criticality. Or the initiator could have been poorly designed, causing a slower start to the explosion and leading the device to fall apart physically before getting the full effect.
In either case, the result is called a "fizzle." In that case, you can get a lot of radioactive isotopes, a good deal of heat, and a much smaller explosion than the design would suggest. In essence, only a tiny part of the material is converted into energy. Even in a well-design nuclear bomb, the amount of matter converted into energy is about the size of a stick of gum. In a fizzle, it can be microscopic.
My initial inclination is to go with the idea that this was faked. After all, North Korea has made similar threats of tests before, and coming only a short time after the US had (foolishly) moved to cut off their access to banks, this was clearly intended as an effort to gain the isolated North Koreans some means of leverage. They have incentive to make us believe they can do this.
However, most experts believe that North Korea has enough plutonium on hand to create half a dozen or so devices, so a failure is just as likely. If that's true, we'll know soon even if the data from the sniffer planes is never made public. If this was a real test and it failed, North Korea will be keen to do it again. Soon.
Fake, or failure, we've little choice but to treat North Korea as if they've joined the "nuclear club."
Update [2006-10-10 17:11:15 by Devilstower]:
As linked below by Magnifico, CNN is now reporting that US officials believe this test "went wrong."
The official added that the unexpectedly small blast, of a half kiloton or less, indicated "something went wrong."
Nuclear tests in 1998 by India and Pakistan were about 24 to 50 times as powerful - or in the 15 kiloton range, according to the Federation of American Scientists.
The United States still cannot say for sure the underground detonation was a nuclear blast; the working assumption is that it was, but not very successful, the official said.
I promise that I am not an unnamed US official.