Constant's pations

If it's more than 30 minutes old, it's not news. It's a blog.

Monday, September 20, 2004

North Korea: The mystery continues


This note explain the problems with the North Korean's explanations for the recent explosions.

  • We test the merits of the North Korean's claim that there was 150 Tons used, and illustrate the maximum and minimum values are not consistent with the Richter scale readings. The sensor readings are inconsistent with the explosive power asserted to have been used.

  • We also identify the flaws with the North Korean explanation for why there is a difference between the detected explosion site and the asserted work. The asserted location is not consistent with the initial readings.

  • We identify a schema that would explain the required location of a detection system for the system to remain within tolerance. The conclusion is ridiculous.

    On all counts, the amount of gunpowder asserted to have been used, and the asserted 'actual location' of the explosion are without merit and fail. This note outlines the basis for asserting the North Koreans' explanations are without merit.


    There remains a viable alternative explanation for the 2.6 Richter reading. That a special purpose refueling rail car capable of carrying 1 million pounds of gasoline or kerosene exploded in a single blast, and caused a fire, during field-refueling incident near a NoDong Missile. Ref


    Still hoping to find, fix, or reconcile the various potential problems with this theory. Although the analysis appears to be sound, here is a list of possible problems with the analysis. Yet, at this point, none of these potential issues appear to undermine the conclusion that there is a reasonable basis to believe a refueling rail car exploded, and the North Koreans have spewed out non-sense.


    Well, we've certainly seen some interesting non-sense coming out of North Korea. But why expect an Administration that relies on non-sense to challenge it.

    North Korea claims that it used 150 Tons of gunpowder to create a dam. Small problem. The numbers do not add up. And the administration is silent.

    Let's run the numbers. The North Koreans say there were two explosions; and that the total amount of gunpowder used in both explosions was 150 Tons.

    One might think that it's "too complicated" to figure out "how many tons were in each explosions. Actually, it doesn't matter. As "no matter how you run the numbers, the numbers do not add up. Let's look at the numbers we have.

    First, a 2.6 Richter scale reading. This corresponds to 4.6 Tons of TNT.

    Second, the North Korean's train explosion in early 2004 which was estimated to be 800 Tons and a 3.6 Richter-scale rating. [Notional, but problematic numbers]

    Third, the North Koreans assert they have used a total of 150 Tons in two explosions; the total amount of gunpowder both explosions combined is 150 Tons.


    Using simple math, this means that at least one of the explosions had at least 75 Tons; and at most the highest number of tons is 150. Both these numbers form the upper and lower bounds for the "possible amount of tonnage" for a given explosion.

    The following graph illustrates one method to expose holes in the North Korean's explanation about "both blasts used a total of 150 Tons of Gunpowder."

    Graph: Upper and Lower Boundary Values for a possible
    Tonnage for at least one explosion

    A. Possible Tonnage For Two Explosions
    using 150 Tons of Gunpowder

    B. 2.6 Richter [R] and 3.6 Richter readings
    as reference

    C. o = notional placement of the explosion
    in terms of equivalent tons [T]

    150 I
    I \
    I \ 3.6R
    I \ 800T
    75 I\ \ O
    I \ \
    I \ \
    I \ \
    I O \ \
    I 4T \ \
    I 2.6R \ \
    I ______\___\___

    75 150

    TONS X1

    Discussion on the Graph Significance

    If the 150 Ton-number from the North Koreans is to be believed,

    and there were in fact two explosions, then one of the explosions had to have at least 75, and no greater than 150. In other words, if there was in fact 150 Tons of Gunpowder used, we know that one of the explosions had to exceed 75 Tons of Gunpowder.

    The "at least 75, but no greater than 150" represents the upper and lower value of at least one of the explosions in terms of tons.

    We define X1 and X2 as the values of each explosion, without consideration of the order. The two diagonal lines represent the upper and lower limits of the possible combinations for each of the explosions, where:

    X1 + X 2 = 150

    For purposes of simply identifying the possible values of the variables, there is no distinction between X1 and X 2; and we need not consider all combinations, as X1 and X 2 are interchangeable.

    Thus, a value of 76 and 74 [as two possible combinations for 150] occurs twice [76 and 74; or 74 and 76], but there is only one combination [76 and 74]. [Corrected the total; was incorrectly totaling 140, rather than 150]

    Outside this boundary layer we see something interesting: A Richter scale reading of 2.6, corresponding to 4.0 Tons of TNT, this falls below 75 Tons. Also, for sake of comparison, we show the 3.6 Richter reading of the Korean Railway explosion, corresponding to 800 Tons of TNT. [Problematic Richter number: 800 to 3.6]

    Depicting these two explosions-Richter readings helps "bound the possible values" of the actual explosions. In short, we can speculate as to "what would make sense" even though we have no actual data, nor can we confirm anything that we have been told [Richter scale-readings, or gunpowder tonnage used]. Specifically, we are not saying what did or didn't happen; we are simply saying, "using their numbers that they gave us, this is what is plausible." Overall, taking their numbers at face value, their story does not add up.

    Explosive power: Assumptions about gunpowder relative to TNT

    Putting aside the issue of "relative difference" between TNT and gunpowder, let's assume that TNT is the same as gunpowder for the moment.

    Let us suppose that the actual tonnage of TNT is the same as gunpowder; and that "one of the explosions" corresponds to 4.0 Tons, and 2.6.

    This is problematic in that we have 146 Tons of gunpowder unaccounted for; and the "remaining gunpowder" [150 less the 4 Tons for 2.6] "should" have a reading well above the 3.6 rating.

    If, in fact, there was a 2.6 Richter reading [corresponding to 4.0 tons of TNT], then we are to ask, "What happened to the other 146 Tons of Gunpowder?

    Yet, there was no report of a second measurable explosion at a reading over 2.6.

    Surely, if there were two explosions equaling 150 Tons, then by using "only" 4 Tons in one explosion, leaves us with 146 Tons "that never registered."

    Analysis and significance weighting

    This implies that the numbers the North Koreans are giving do not add up, nor support their version of events. It remains the North Korean's responsibility to clarify this inconsistency. It is implausible to think that "the error" lies with the Western sensors -- there are multiple sensors, apparently confirming the same Richter-scale reading of 2.6. Thus, we put greater weight and credibility on the 2.6 than 150 Tons.

    Comparing 150 Tons to a simple Richter Chart

    Let's now look at the possible explanations that would put more credibility on the 150 Tons than the 2.6 Richter reading. In short, we are hoping to bias the analysis by looking for anything that would tend to not only right the scales so that both the 2.6 and 150 were equally credible, but find a plausible method to suggest that the 150-number is more plausible than the 2.6 Richter reading.

    Let us assume that the North Korean number is valid and that there were two explosions [not proven]; this would require that at least one of the explosions had at least 75 tons, or 150,000 pounds of explosives [assuming 2,000 pounds per ton].

    Let us consider a chart. Not the Y-axis [up-down, not horizontal]: The units of measurement are log-scale, and the units are in pounds of explosives.

    Note the right scale entry of 120,000, corresponding to 60 Tons equivalent of explosives. This is less than the required explosive power of the 75-ton explosion. In other words, if the North Korean number is to be believed [that there were two explosives using 150 Tons] then this would mean that the required Richter reading would exceed the reported 2.6 and measure at least 4.0.

    Suppose this is true. Why is a 4.0-equivalent explosion only measuring as 2.6? The answer is: There was no second explosion of 4.0; the real single explosion was 2.6, corresponding to 4.0 Tons of TNT.

    If the North Korean numbers were true, then we'd have one explosion of at least 75 tons [greater than 4.0], and a second explosion corresponding to 2.6. Thus, if the North Koreans actually used 150 Tons of gunpowder, the only way we could have a reading of 2.6 is if the actual tonnage of another explosion exceeded 4.0. Yet we have no second reading, only one.

    That's the problem with the North Korean's explanation about "there being 150 Tons of gunpowder used." If that is true, then there had to have been one explosion with at least 75 tons, and that equals 150,000 pounds or an equivalent of greater than 4; followed by a second measurable detonation.

    To add up to 150, either one explosion was really large [which there wasn't, only 2.6 or 4 Tons]; or there was a simultaneous explosion at the same time. Which the North Koreans have not asserted, and is generally impossible given the wiring-challenges; plus, that wouldn't be "two explosions" but simply, "one big one."

    Analysis and significance

    Again, despite giving more deference to the 150 number, we are again forced to conclude that the 150-number is less likely than the 2.6 Richter reading. Bluntly, because the "chain of events" required to support the 150-number are not supported by later readings, we are left to put more weight on a single reading of 2.6 Richter scale, rather than the self-reported 150 Tons of gunpowder from North Korea.

    Implications of "at least 75 Tons used in one explosion corresponds to 2.6"

    Let's further bias the analysis, and presuppose that there is a possible "to be determined" explanation for this inconsistency. In short, what we are doing now is essentially ignoring the above two analysis, and seeking to find a plausible explanation, and then assess the merits of that argument.

    Let us assume for the sake of argument that the 2.6 Richter reading [equivalent to 4 Tons of TNT] corresponds to the larger of the readings, above 75; and let us assume that the 75 Ton-minimum for one of the explosions corresponds to the 2.6.

    What could we say about the "effectiveness of the gunpowder used" if at least 75 Tons of TNT created "only" 2.6 on the Richter scale? That's some pretty weak gunpowder, indicating the North Koreans really aren't in a position to be making a dam: They've spent about $5M dollars on some pretty low grade stuff.

    If I was in North Korea and we were having to pay the "western capitalist dogs" that kind of money for such weak stuff, I'd be pretty upset. But they continue to buy it or make it themselves. However, the "they're too stupid to make good stuff" misses the point: They've created a rocket that flew over Japan, and they do have a fleet of many rockets. So I'm not buying the "they spent alot for bad stuff".

    I'm more inclined to think that the 150 pounds of gunpowder is a non-sense number; they hoped to "muddy the waters" by saying "150 pounds" corresponds to both explosions, not thinking that "at least one of the explosions had to equal 75 tons; but 75 Tons doesn't correspond to 2.6, so the 150 Ton-explanation doesn't make sense either.

    Analysis and weighting

    Again, for the third time, the weighting and deference to the 150-ton number doesn't stand up relative to the 2.6 Richter reading. In fact, we can only speculate that the North Koreans "to make their 150 number stick" would have to come up with far more creative explanations to justify this number.

    It is assumed that the United States doesn't want to let the North Koreans lose face; rather than publicly challenge them over the discrepancy and possibly make the North Koreans feels as though they are in a corner, it appears as though Washington is simply saying nothing definitive.

    Burden of Proof

    In the idea world, we would go to the North Koreans and say, "hay, this story you've given us doesn't make sense. Here's the data to prove you're wrong." Rather, what the world is doing is playing a game of walking on eggshells--we suspect that North Korea may actually be preparing for a nuclear test.

    However, let's suppose we do live in a universe where those who lie are held to account. We might have a very difference discussion.

    Bluntly, the burden of proof rests with the North Koreans. Whether this "at least 75 Tons of gunpowder" really corresponds to the 2.6 remains in question; it is not to be presumed, but proved.

    Therein lies the faulty analysis. North Korea is "presumed" to be telling the truth for sake of 150 Tons; yet the numbers do not add up; rather than get a "credible set of numbers" the West has quickly joined in the "it must be something wrong with the sensors." Surely, this is not a credible explanation as there's been much testing and development of these testers; bluntly, it is amazing how quickly we assume the 150 number to be true, and that "everything else" is easily explained away.

    What about the second explosion>

    The other issue to consider, if the North Koreans story is true, is "Why didn't the residual (150 less the [amount above 75]) show up as a reading on a second Richter scale?" Therein lies the problem.

    The North Koreans, by committing to 150 Tons, have also committed to 75 tons or greater; and at the same time the technology to support the 2.6 finding is very good.


    To be clear, the sites in Japan are only 3 of many which are also located in the United States, Russia, China, and elsewhere.

    For purposes of confirming nuclear detonations and the 2.6 Richter reading, there are at least three monitoring stations in Japan; and the North Koreans would like us to believe that the "error of 30 miles [67 km] relates to a measurement problem.

    That doesn't wash. The instruments to "measure earthquakes" are not new; in fact, each generation of instruments created to monitor earthquakes are an improvement over the earlier. Earthquakes were fist studied in the 1600s; today, using 3 detectors in Japan [ignoring the detectors in China, Russia and the US], we should have the ability to narrow the area down to within a few miles, not "maybe plus or minus 100," which the North Koreans want us to believe.

    Implications of the "horizontal error"

    The North Koreans assert that the true source of the blast is at a dam, some 37 miles east of the Richter-reading-source. Note, the "direction of error" [Slide 30 miles, 67km to the East] was horizontal, not vertical [North and South] into China.

    Let's consider the implications of the shift, and what it says or implies about the accuracy of the detection equipment. If we are to believe the North Koreans, and also allow for the design limits of the detection equipment, we'll notice something very unusual.

      Diagram: Showing relative position of a hypothetical
    Detector-location to fall within the bounds
    of the detector-limits

    I North

    Actual [k]: Position, actual site of explosion
    Asserted [s]: Position, asserted site of explosion

    k s
    --> [Delta A]
    [ ] N. Korea
    I \
    Distance I [ ] Japan \ <-- Delta B
    ["D"] I "J" \
    I \------>
    I [ ] Hypothetical-required
    ["H" or "Hawaii"]

    If the amount of the error was 'truly of this magnitude' [off by 30 miles, 67 km; from K to S], if we consider the "design limits of the instruments" [At "J"]... we would have to conclude something about the 'distance of the instruments from the source" that is at odds with reality.

    Specificity, the way these seismograph-detectors work, is that they triangulate on a point; and they can do so with +/- error rate.

    Supposing the "error of 30 miles was valid" [off by 67km], then this would mean that "for the instruments to be operating within their design limits", the distance from the source of the earthquake to the detector, would have to be extended well beyond Japan, to well within the Pacific Ocean, near Hawaii.

    In fact, the range of errors is not in dispute. The instruments have a known error rate, part of the performance rating, and the contract award.

    Referring to the diagram, Delta B would have to be small, not by shifting the location of the blast, but by assuming that the "hypothetical detector" had a much larger distance [D] from the location, in order to fall within the design limit angle-error, Theta.

    What the world is asking us to believe is that "despite these contract specifications" and "all the history of detecting earthquakes" [since 1600s], that the ability to detect the location of earthquakes has suddenly degraded, and the "massive amount of money spent on pinpoiting sources of nuclear explosions has been for naught."

    Oh, come now. Surely, there are not enough buffoons in the world that would have us believe this non-sense. But there are.

    The only way to believe that the instruments were all wrong in Japan, but still comply with the contract specification requirements on errors, would mean that the detector was somewhere else much further from North Korea, than Japan...somewhere in the middle of the Pacific Ocean. Yet, the detectors are actually inside Japan.

    There is nothing the North Koreans can do about this, nor can they explain away the historical error-rate by asserting "there were 150 Tons" and that's it. In other words, by committing to 150 Tons, the North Koreans have also committed to a long list of other data-assumptions that do not add, but detract from, the merits of the North Korean's story.

    So why is the US so reluctant to "come forward" to "point out the non-sense"?

    Couple of explanations.

    Failure to understand

    First, the US doesn't realize the implications. Indeed, the same engineers could be arm-twisted not to go to the NYT over the Vice President's abuses through the Office of Special Plans, so the White House can be sure the intelligence community's credibility is shot.

    Analysts silenced

    Second, those who have figured out the implications are being asked to be quiet; perhaps the better word is "threatened with loss of employment" if they speak their minds. Those mortgage monsters need to be fed.

    No need for a diversion

    Third, the US has decided it doesn't want to argue with North Korea over this issue; North Korea has no oil; and there's no need to emphasize this diversion given the Kerry campaign's inability to organize.


    Fourth, why bother to fight/argue with the North Koreans publicly over an issue that the US really can't do anything about; the US really can't do anything to the North Koreans, regardless the facts. There's no reason to admit publicly that "nothing can be done" by the US; so why not simply pretend there is no problem.

    Hide reality

    Fifth, the US has actually a very good detection capability and doesn't want to reveal to the North Koreans we know exactly what is going on. This is possible, but raises the question: Why would we deny that which is known in North Korea, but go to such lengths to fabricate which doesn't exist in Iraq; plus, there is the credibility issue: Why publicly state "there is no problem" despite the surprises in India/Pakistan; and it is absurd to believe that the US would lie about a "known problem" unless the US doesn't want to force the hand and have people say, "There's a real problem in North Korea, but we have no troops: Looks like we need a draft."


    Sixth, the US would rather spend time harassing its own population than publicly challenging the North Koreans over this "nonsense": What that "non-sense" is remains to be seen; and it is noteworthy that "despite all that we don't know," all the physical evidence is explained away with explanations that well, don't make sense.

    Program funding

    Seventh, there are program managers who are pushing for an upgrade and/or modernization of nuclear-weapons detection equipment; this "inability to detect" is the perfect excuse to justify continued funding and research of something that isn't broken, but merely a program used to "keep people employed" or a "make work program" for engineer who are otherwise unemployable at WalMart.

    Saving face

    Eight, is the reality that North Korea could very well have in place a nuclear program that is about to deploy. The last thing the US wants to do is back North Korea into a corner. Even though the story doesn't add up, there's really nothing constructive to come from publicly ridiculing the North Koreans for a non-sense story.

    At best, the North Koreans will simply create another story that will have to be "reproven wrong" to the satisfaction of policy makers. Chances are, they've already privately agreed that there was an actual explosion of a rail car carrying kerosene or gasoline at the site of the 2.6 Richter reading.

    At worst, the North Koreans will simply stop communicating. The US would rather have bad information and the appearance of diplomacy, rather than have silence. At this point, anything from North Korea is a good sign, even if it is non-sense.

  • The other analysis

    One other area to consider is the "size of the hole created" and compare it to other holes that had at least 75 Tons of gunpowder, and compare the results.

    This isn't all that difficult. We know from history that Guy Fawkes would've leveled Westminster Abbey with ___ X ____ tons of gunpowder. [Detailed discussion and analysis]

    If we compare the size of the explosive force that would've occurred in this explosion [See map], and compare it to the size of the "newly created hole" in North Korea, the comparison will yield some interesting results.

    In fact, let's not stop there. Although the North Koreans think they've pulled the wool over everyone's eyes, we can simply look at the size of the machinery directly adjacent to the hole in the "site visited" and compare that equipment to other sites.

    In short, the "size of the hole" can easily be calculated", and then compare "what they're doing at this dam, and see what is going on.

    Again: The issue is: This "dam-site that was visited" does not match the source of the Richter-scale-reading. Indeed, a "Detailed analysis of the dam-site" would yield information about "what is going on at the site," [maybe] but does little to answer: What caused the 2.6 Richter reading; and why was the site visited not consistent with the triangulated results?

    The time spent "messing with the video at the dam-site" delays attention, energy that would otherwise be put to the actual origin of the 2.6 Richter; and also permits the North Koreans to grade-over the site.

    It remains to be seen whether the Western Satellites continue to monitor the real source of the explosion to monitor night-time operations. The North Koreans know the satellite times; it remains to be seen whether they would cover the area with tarps or other shielding to prevent Satellite imagery; or whether they will use humans [vice machines] so as to not attract attention to the actual explosion site.


    There is no basis to assert the actual location was different than the original readings. The inspectors visited the wrong site, while the real explosion site is being sanitized.

    The North Korean's explanation doesn't make sense. The World has embraced the theory that "there must be something wrong with the sensors." North Korea has not been held to account for this non-sense as there is little anyone can actually do.

    Nothing the North Koreans have stated has satisfactorily explained away the possible alternative explanation: That a special-purpose rail car carrying 1 Million pounds of kerosene or gasoline exploded during a NoDong Missile refueling.

    The west is reluctant to state what is known as this would reveal knowledge of the scope of the missile deployment activity; there is no urgency in stating the truth; and the west has chosen to remain silent rather than publicly debate with the North Koreans over the missile mishap.

    There is much non-sense spewing forth from both DC and North Korea. Bad enemies make great friends.


    If you are having trouble sleeping and would like to read more details of the site visit, and the problems with the message traffic, read here for a blow-by-blow breakdown of the information spewing out of the Koreas.