Old 30th Mar 2017, 23:37
  #1110 (permalink)  
Uncle Fred
 
Join Date: Aug 2008
Location: Vendee
Posts: 137
True Onetrack. I was not endorsing the Sun but rather commenting more on the headline that they had written for it.

For a more serious delve into how we know some things about NK, I pulled this discussion from the SlateStarCodex. Open Thread 72.25 | Slate Star Codex Unfortunately they is not a Permalink to this.

Chap by the name of John Schilling. Sounds as if he is working for a think-tank in the U.S.

John Schilling says:
March 29, 2017 at 9:47 pm ~new~
Weíve talked about North Koreaís nuclear weapons, and the missiles they have to carry them. Before we move on to the final act, which is to say what they plan to do with all of those wonderful toys, thereís an obvious question that some of you might be asking: Just how do we know so much about the strategic weapons of one of the most secretive and isolated regimes on the planet? Itís not like they are going to just tell us about them, right?

And actually, they sort of do. As no less an authority than Dr. Strangelove pointed out, the whole point of a deterrent weapon is lost if you keep it a secret. The North Koreans go out of their way to tell us, show us, and demonstrate for us their missiles and even occasionally their nuclear warheads. And they usually arenít lying. They arenít being entirely honest about it either, exaggerating some parts and concealing others. So weíre not going to take their word for it. But their parades and photo ops and public demonstrations are often a good starting point.

Other good starting points: seismic data from nuclear tests. Thereís a pretty unambiguous seismic signature from an underground nuclear test, giving us the time and location and approximate yield. It is in theory possible to spoof by setting off a nuclear device in a very large artificial cavern, but we keep a close enough watch on their usual test sites to be confident they arenít doing that sort of excavation. Thereís a network of seismometers being monitored 24/7 for this sort of thing. And while itís theoretically possible to detect missile launches from seismic data (the reverberation from rocket exhaust hitting the pad can be detected across a few hundred kilometers), thatís a much trickier signature to interpret. So we typically just let the US or South Korean military tell us when the North launches a missile. Their satellites and radar will detect the launch in minutes, they will send out a warning to their own key people almost as fast, and then follow up with a press release in a few hours because if they donít it will leak out anyway. North Koreaís press releases usually take longer.

Whatever the initial source, once the news breaks we have to figure out what happened and integrate that knowledge into our broader understanding. Melissa Hanham of the Middlebury Institute (Monterey, CA), offers one account of how this proceeds. I particularly like her comment about tweets travelling faster than seismic waves; it does convey the almost literal wave front of tweets, texts, and emails that brings the news to the geographically-dispersed community of wonks that deals with this sort of thing. Iíve worked with the Monterey team in the past, but am more closely associated with the 38 North project run by Johns Hopkins University. Itís a friendly and cooperative rivalry.

If we do start with North Korean propaganda, the next step is to look for what they are hiding. As Melissa notes in her account, the Monterey team has the tools and technical expertise to detect most any sort of image manipulation, and while it is rare for North Korea to fake photos of things that donít exist they quite frequently edit them to hide things that do Ė e.g. cutting away from one missile just as it begins to explode, and going to a long shot of a different missile ascending majestically into the heavens. Monterey is equally skilled at geolocation; if an interesting picture comes to their attention, in an hour or two there will be a tweet saying when and where it was taken and what other interesting stuff happened there.

Weíd rather have our own pictures, and for that we have satellites. Google Earth doesnít have the resolution or the refresh rate we really need; fortunately the 38 North team has ties to Digital Globe which can provide 30cm imagery on a daily basis if we tell them what we need. If what we need are pictures of an event that happened yesterday afternoon, probably the best we can do are before-and-after shots, but even those are helpful. And for some propaganda events, e.g. parades, there will be foreign journalists present Ė usually Chinese, but some of them will share their own unedited photos if we agree not to publish them,

Pictures can tell us the size and shape of e.g. a missileís exterior, and sometimes telling details like the location of propellant fill/drain ports or extra riveting for structural reinforcement. We want to know whatís on the inside, and more importantly how it will perform.

If thereís an actual flight test, performance can be straightforward Ė radar tracking will give us the range and apogee, or for a satellite launch the orbital parameters and usually the first-stage impact point. Weíd like more detailed trajectory data, like the acceleration profile, but the people with the tracking radars only rarely release that level of detail. For ground tests, we can sometimes read bits of telemetry data off the screens Kim Jong-Un insists on posing in front of, but mostly we are limited to studying rocket plumes. Plume size can give approximate thrust, color and intensity the propellants used, and detailed geometry can give hints to e.g. operating cycles (is there a separate turbopump exhaust?) or control mechanisms (separate verniers, or jet vanes?).

For internal arrangement, it helps that so many of North Koreaís missiles are based on old Soviet technology, because the Russians have published a fair amount of detail on their older models and we can sometimes talk to Russian academics with inside knowledge from the old days. Now that the North Koreans are designing their own engines and other systems, thatís going to be a bit harder Ė but in partial compensation, they are now showing us pictures of their engine ground tests. So we can maintain a sort of shopping list of e.g. rocket engines available to North Korea, and see which of them are a good match for any new missile we might see.

Also particularly helpful, in 2012 they finally managed to launch a satellite successfully, and the first stage of the launch vehicle necessarily landed far enough from home that South Korea, with a proper oceangoing navy, was able to recover it before their Northern cousins. That was a space launch vehicle, not a missile, but there are clear indications that they are using some of the same technology. So that gives us a pretty good idea of the North Korean state of the art in missile structures and systems in 2012. In their next satellite launch, the first stage disintegrated into several hundred radar-trackable bits just after separation. This probably wasnít an accident.

Finally, we look for anyone else who might have insight or information we lack. As I mentioned, thereís a friendly rivalry between several teams in this field, and we all talk. Beyond that, the most obvious sources are the national intelligence services of nations with a particular interest in North Korea, namely the United States, South Korea, Japan, and China. They are doing basically the same things we are and with about the same tools, but some unique edges. The United States Government has better satellites than Digital Globe. South Korea and Japan have radar sites positioned to track North Korean launches, and the South Koreans have a bunch of North Korean defectors they can talk to (alas not usually technical specialists). They all typically issue terse press releases with whatever they publicly want people to know, and we arrange occasional closed-door meetings to give them an opportunity for greater candor. But these are all basically spies, and so not to be entirely trusted. They rarely outright lie, but every truth is carefully selected to serve an agenda, and if we canít verify it directly we at least want to see different governments with different agendas saying the same thing.

Other knowledgeable sources include the Iranians and Pakistanis, who have purchased missile technology from North Korea and in Pakistanís case sold them nuclear technology in return. These arenít exactly open and transparent governments, but they have different ideas about what needs to be kept secret and what they ought to brag about. Thereís also the important resource that is random nerds on the internet. When North Korea first paraded their mobile ICBM, most of us focused on the missile. Within a day, a bunch of heavy-equipment otaku identified the exact make and model of the transporter chassis; a Chinese product that the Chinese government promptly threw down the memory hole because they hadnít meant it to be seen carrying North Korean missiles. Nice try Ė scrub it from the internet, bar the gates of the Great Firewall of China, Iíve already got a copy of the advertising brochure complete with a dimensioned three-view drawing.

For nuclear warheads thereís usually less visible evidence, as the tests are conducted underground. But the seismic signatures of those tests can be precisely measured, and their implications are well understood. We also have satellite imagery of their reactors, whose waste heat really canít be hidden, though the uranium enrichment centrifuges unfortunately can be. To bridge the gap between the fissile materials we infer them to be producing and the earth-shattering kabooms we hear them produce, we make comparisons to the extensively documented history of the early US nuclear weapons program. Also the programs of nations like Sweden, South Africa, and Iraq Ė less well documented, but more on point for what an emerging nuclear power in an established nuclear age might do.

But all of these sources combined only add up to fragmentary pieces of a puzzle we have to assemble. For the rest, thereís engineering judgement, analysis, and the presumption of competence. The people who have demonstrated the ability to launch satellites into orbit, we assume will design their other rockets in a manner that at least plausibly could work. The regime that has survived three generations with most of the world turned against it, we assume will adopt military strategies that plausibly could defend that regime against its enemies rather than assure its destruction.

So, given the fragments of hard data, we make educated guesses as to what might make up the rest, apply various modeling tools (and Iíve written a couple of custom codes to handle the missile side) to estimate the performance of the postulated system, and see if the result makes any sense from the North Korean standpoint. Given the number of missing pieces of the puzzle, the problem isnít finding an answer that makes sense, itís narrowing it down to one. And establishing confidence bounds, given that some of those puzzle pieces come from very sketchy sources. My initial assessment of the KN-08 ICBM listed six possible configurations plus ďmaybe itís just a hoaxĒ, and it took another three years (and four more North Korean photo ops) to get to a single reasonably firm conclusion.

Thatís for assessing the technology, where at least there are specific, knowable answers to questions of the form, ďIf I put together a missile thusly, will it work and how far will it deliver a payloadĒ? Things are a bit fuzzier when the question is, ďWhat is Kim Jong Un going to do with all those missiles, and what can we do about itĒ? But we havenít been ignoring those questions, and Iíll deal with them in my (probably) final segment next time.
Uncle Fred is offline