DPRK: N-Test, Take 2

It was early October, 2006. Overhead, satellites skimmed the sky above the Democratic People’s Republic of Korea (DPRK) and took careful note of the goings-on below.

On the ground, as well as far beneath it, technicians busily prepared for North Korea’s highly-anticipated, first nuclear weapons test – an event that was confirmed in dramatic fashion on October 9th, when it appeared on one of the seismographs I was monitoring:

Though the event appeared on dozens of other seismographs, this one, from seismic station QIZ in Guangdong, China, seemed particularly interesting.

In the lower-right corner of this QIZ seismograph, you can see the “boom” of a large, explosive detonation. In comparison to normal seismic activity (seen in the middle section of the graph) the explosion created a much less jagged seismic ‘signature’ than a standard earthquake when it rolled into Guangdong. On other seismographs, the North Korean “bang” looked similar to a normal earthquake (except for its abrupt onset), but QIZ sits in a sort of sweet-spot when it comes to catching ‘vibes’ from North Korea.

Estimates of the test’s yield varied from 0.5 to 15 Kt (kilotons); the former figure coming from a US arms expert and the latter having been issued by the Foreign Minister of the Russian Federation. A consensus developed, at least in Western circles, around the idea that the DPRK’s nuclear test had released energy equivalent to just over 800 tons of TNT (0.817 kT) – a relatively small bang in nuclear terms, but a concern nevertheless.

The US Pentagon officially considered the test to be “fizzle”, which is to say that it produced a yield substantially less than intended, but there was a lot of discussion in the media (and in “situation rooms” worldwide) about the actual meaning of the event.

Here’s the signature of the test, statistically smoothed:

States of Alarm
On October 11th, 2006, shockwaves of another variety rocked the debate when an earthquake prompted many to believe that another (larger) test had just taken place. (A quick review of the seismic data took the tension down a few notches.)

A couple of days later, the US confirmed that, according to air samples collected on October 11th, North Korea’s Oct. 9th detonation had indeed been a nuclear one. (Tension climbs back up – one notch.)

On October 16, 2006, the governments of the United States, Japan and South Korea simultaneously indicated that preparations for a second nuclear test were underway and that the test was likely imminent. (Two notches!)

And then… nothing. The days ticked by…

On October 18, 2006, the technicians on the ground seemed to be packing up and moving on to other duties, but no one seemed to know exactly why the seemingly imminent second test had been cancelled or postponed. There were, undoubtedly, some who took the opportunity to congratulate themselves for having bent the will of North Korea’s intransigent leader (and master prevaricator) Kim Jong-Il.

Meanwhile, North Korea quietly informed China that it was intending to conduct as many as three more nuclear tests.

But what happened between October 16th and October 18th?

Why, October 17th, of course!

And this:

– a strong quake (R6.6); Papua, New Guinea region; as measured at QIZ

Most of the activity seen in the graph above is due to a strong earthquake (R6.6) that took place on October 17, 2006, near Papua, New Guinea (P.N.G.). However, in the midst of the recorded tremor there are some anomalies.

None of the other available seismometers showed the same sort of alternating signal cancellation and augmentation that can be seen in this graph from the QIZ seismo – the same seismometer that rendered the most unique “echo” of the October 9th nuclear test.

So, I had an idea:
What if I took the signature from the Oct. 9th test and compared it against the anomalies in the Oct. 17th seismograph?

A strange thing happened when I scaled the earlier signature by 245% (about six times the energy displacement of the Oct. 9th test):

The first peak of the previous signature (in Red) aligns perfectly with the maximum signal cancellation in the P.N.G. quake at Point A.

At Point B, the P.N.G. quake (but only when measured at QIZ) undergoes another transition.

The climb to the second peak (Point C) of the first signature coincides with a strong augmentation of the P.N.G. data.

At the next peak, Point D, the P.N.G. seismographic data suddenly changes once again.

Statistically speaking, the strong correlation between the two signals dramatically increases the likelihood that a second North Korean nuclear test did take place – eight days (almost to the minute) after the first … on the very day when the United States, South Korea and Japan were bracing for another blast.

Fizzle 2?
If the first test was more “fizzle” than “fissile”, then what are we to make of this possible second test?

Should we assume that the first test was supposed to yield 4kT (the information confided by North Korea to China) but could manage only 0.817kT?

Should we assume that the second yield (at ~5kT) was “better than expected”?

Or, dare we speculate that each test actually achieved its desired result?

If we logically travel down that path, we’ll find a number of devices with ‘dial-a-yield’. Some large. Some small. Among the lighter weight designs, there are tactical nuclear land-mines, nuclear demolition devices, so-called “back-pack nukes” and nuclear artillery ordnance that offer variable yield capability. Many of these offer a choice of three or more yields.

If the design tested by North Korean was based on one of those models, then it’s possible that they tested the same device twice (at two different yield settings) – and only got caught once. This (along with the geophysical facts) may help to explain why the signature of the first test matches up so well to the anomalies in the Oct. 17th readings.

Unfortunately, the device might also have a third setting of about 31kT.

(Note: This assumes a weapon with symmetrical scalability, but the range of selectable yields would ultimately depend on the geometry and energetics of the ‘undisclosed’ design.)

Variable yield devices, on average, tend to be smaller and lighter than the average nuclear warhead and are usually Plutonium-based, all of which makes them ideal for North Korea’s nuclear program, which is predicated upon its Plutonium-breeding reactor complex at Yongbyon. The smaller size and lighter mass of these types of devices also makes them suitable for reconfiguring as missile-deliverable payloads.

It was reported just last summer that the A.Q. Khan proliferation network had been flogging designs for advanced nuclear mechanisms for some time. To me, anyway, it looks like this could be one of them.

Take (n)
So, what has North Korea been up to for the past 2.5 years – besides shooting ballistic missiles over Japan and alternately dragging and stomping its feet?

Has the DPRK conducted any further nuclear testing?

Tough questions.

Many live seismographs from Russia and China have been off-line for a while now, making it more difficult to easily collect data from specific ‘listening posts’. Some have recently reappeared, but I have a hunch that more time will pass before we see QIZ – or any of the other Sino-Korean seismic sweet-spots – openly published again.

If the North Korean leadership figured out that they could muffle their nuclear tests in the noise produced by the frequent earthquakes of the Ring of Fire, then it’s possible that they may already have completed a number of limited yield tests since October, 2006, but larger tests (of 15, 20, 30kT or more) would definitely be more difficult to hide.

Echoes & Variations
QIZ responded to the Oct. 9th test with more gusto than many of the seismos located closer to the action, in South Korea. Even the one at Inchon, just over the border and across the peninsula, barely burped during the event window!

– from the INCN seismometer at Inchon, South Korea

This would seem to indicate that the nuclear test suites prepared by North Korea are “shaped” (the North Koreans are very good at digging) in order to avoid detection. The mountainous terrain and soil conditions around the site (near Kilchu) may also help to dampen vibrations.

One reason that such good measurements could be collected to the southwest is due to the shape of Korea Bay, on the Yellow Sea, which acts like a type of seismic echo chamber for the upper Korean peninsula – similar to the way that your voice is projected when you speak into the small, open end of a hollow cone.

Okay, so now you know.

But, what can we do about it?

To start, maybe we should take a penetratingly good look at every country that is known to have dealt with A.Q. Khan’s network – and pay special attention to the ones who are working aggressively on long-range missile programs. That second group (a very ‘exclusive’ club) appears to have only two members: North Korea and Iran.

Of course, there’s also Pakistan, but that’s a whole other predicament.

—

(Update: For info on the May 25, 2009 test window click here.)