Jeffrey LewisFact-Checking From Low-Earth Orbit

Fact-checking political rhetoric has lately become something of a cottage industry in America. Getting into that line of work seems to require a cutesy visual rating system involving wooden noses, flaming trousers, or some such. But with such a sweet target of opportunity, who can resist? After all, this may be the first time that space imagery has debunked a political claim.

Hey, maybe VERTIC could set up a program.

Our story begins in South Korea, where President Lee Myung-bak of the Saenuri party is reaching the end of his single term. Best known in the West for his resemblance to Pee-wee Herman and for having been the object of a North Korean campaign of vilification that was extreme even by Pyongyang standards, Lee is thinking of his legacy and pulling for Saenuri candidate Park Geun-hye, daughter of late dictator Park Chung-hee.

Just last month, he visited Yeonpyeong Island. That’s where, to the shock of South Koreans two years ago, a North Korean artillery raid killed two ROK marines and two civilians. According to an article last week in the Joongang Ilbo, Lee actually took the occasion of his visit to declare the incident a victory. Really?

Before going further, just so there’s no confusion, let it be known that I don’t have a preferred candidate in this race. (I was about to write “a dog in this fight,” but, well. Yeah. I’m more of a bulgogi type anyway.) Also, in no way do I wish to make light of the Yeonpyeong-do incident.

Let’s continue.

Joongang’s correspondents wrote:

According to the [ROK Defense] ministry, on Nov. 23, 2010, the South Korean military had conducted its regular firing exercise and was performing maintenance on its K-9 self-propelled artillery.

It was then that North Korea attacked the island by firing 170 artillery shells and rockets. In response, the South Korean military returned artillery fire using 80 K-9 self-propelled artillery shells against Mu Island, where the North Korean artillery unit was located, 13 minutes after the North attacked.

According to unconfirmed reports, 10 North Korean soldiers were killed and 30 were injured.

Industry insiders note that one of the reasons why the Defense Ministry decided to define the battle a success after almost two years came after President Lee Myung-bak’s recent visit to Yeonpyeong Island.

After being briefed about the skirmish, President Lee is known to have told military officers that “Being here in person, I come to think that we [South Korea] counteracted well [against the North].”

He also said, “It’s our victory, [and the military should review and redefine] the Marine Corps’ reaction [to the attack].”

In other words, first impressions can mislead:

Unconfirmed reports? Surely someone can confirm them.

Turns out, someone has.  DigitalGlobe released an imagery product showing the aftermath of the shelling. It’s at the NKEconWatch website. (Thanks to a very good friend for the tip.)

The presentation is gob-smacking, showing three images of three batteries of North Korean MLRS launchers in position and camouflaged prior to the attack.  (I am not sure about the dates shown in the presentation).  Here is one such image:

It also includes this stunning view, taken later, of craters from the ROK Marines’ counter-battery fire at 37° 47′ 14″ N, 125° 35′ 25″ E, all in the general vicinity of now vacant firing positions. (You can find the same view using the historical imagery feature at Google Earth.)

Ahem. Judging by where the craters are, all of the shots missed.  And now, to deliver the verdict, Walter Sobchak:


  1. blowback (History)

    “on its K-9 self-propelled artillery”

    Does this artillery cocks its leg and piss on you?

    • Magpie (History)

      Just for a second there, my brain parsed the sentence funny, and I thought – “what the hell is a ‘self propelled artillery shell’? A dog with a bomb?”

      You have to admit, that’d fit the K-9 designation pretty well.

    • joshua (History)

      Perhaps the K-9 subtly expresses a hidden love of gaegogi?

    • John F. Opie (History)

      The K-9 is a self-propelled artillery system. All it takes is two seconds on google to find that out. Y’all should know that it means the artillery system is self-propelled, rather than being towed…

    • Magpie (History)

      Yeah, we waz just jokin’.

  2. Eve (History)

    OSGEOINT often had flaming trousers. Pity there is no OS now.

  3. Juuso (History)

    I have always read that SK counter fire could silence NK artillery positions in a few minutes, but after seeing those pics i’m a bit more sceptical towards those claims.

  4. Peter Hayes (History)

    This is a two way street. Reportedly, half of the DPRK shells went into the ocean around the island; and half those that hit land were duds. It’s clear that the DPRK attack killed some South Koreans. It’s not clear that they hit anything that they targeted, except randomly.

    This duel should give both sides pause by reducing confidence in precision strikes with conventional artillery.

  5. rwendland (History)

    Does anyone have a reliable public source on where the aim points of the initial K9 live firing exercise into the sea was? Specifically, was any point within 12 nm of the North Korean coast, but further than 12 nm from the Yeonpyeong Island coast? ie a place North Korea would clearly regard as in thier territorial waters?

    The two best newspaper level reports I have found suggest it probably was at such a point. Hankyoreh of Dec.21, 2010 shows a map which suggests a large part of the firing range, to the west-south-west of Yeonpyeong, was within 12 nm of the North Korean coast:

    The JoonAng Daily of Nov 25,2010 states “the South Korean military prepares “with caution” and keeps their firing “4 to 5 kilometers [2.5 to 3.1 miles] away from the Northern Limit Line.”” As the NLL is about 3 nm from the North Korean coast, this suggests the aim point was around 6 nm off the North Korean coast.

    So it looks likely this terrible and unfortunate episode was caused by the running sore of the NLL, and an excessive reaction. South Korea trying to assert the NLL’s continuing military existence, and North Korea trying to assert its territorial waters and the non-existence of the NLL in international law.

    But it would be useful to know for certain that this was the case. Of course North Korea’s over-reaction was terrible; if they felt a military response was essential a proportionate response would have been to fire into Yeonpyeong Island territorial waters (3 nautical miles are asserted by the United Nations Command), not at the island.

    • joshua (History)

      It appears that the NKs may have made shelled both north and south of the NLL in August 2010. Different media accounts from that time, all based on anonymous sources, do not agree on where the shells (or maybe rockets?) fell. There was a similar around-the-NLL shelling event in Aug. 2011, according to the ROKs.

      I seem to recall that shortly after the YP-do incident, the NKs threatened a repeat performance in response to further ROK artillery drills, but did not follow through.

      In any case, there is a long and growing record of tit-for-tat actions, including deadly incidents at sea, in this area. YP-do seems to have been the first case involving civilian casualties.

    • rwendland (History)

      Josh, did you mean NK in that reply? You seem to have answered a different question to the one I was asking.

      I was wondering if we, in the public-domain, knew for certain where exactly the initial SK live firing exercise aim point was – before NK fired on the island, prompting NK’s wrong-headed response. The two newspaper articles I highlight suggest it was at a point within 12 nm of the NK coast, which NK regards as its territorial waters – but are those newspaper reports correct?

      I’m worried that the NLL will cause an even more serious flare-up, especially if SK insists on asserting the NLL by firing at points within 12 nm of the NK coast. Basically I support Joel Wit and Jenny Town’s recent suggestion in Foreign Policy and 38North that replacing the armistice that with a permanent peace arrangement, including sorting out the NLL, should be an Obama 2nd term priority.

      NB In the comment above I forgot to give the url for JoonAng Daily article of Nov 25, 2010. It is:

  6. George William Herbert (History)

    Uh. I am excessively confused.

    Your cache / link to the original doc is not working for me. I also can’t find it on their site…

    Google Earth shows a now-former NK rocket artillery firing position at/near 37°46’44.38″N 125°35’1.04″E but it’s not the one in the picture with those coordinates, unless I’m horribly confused. The second set of coordinates are for the SK artillery battery, on the Island, I assume, but are not for the photo they’re tagged on (which appears to be another NK rocket artillery battery, at yet another location).

    Shelled battery appears to be at 37°47’13.44″N 125°35’24.72″E.

    The battery is about 100 meters long (center of position of two end vehicle revetments). That puts two rounds into about 25 meters from launchers in that photo, and several more within 50 meters. That is not going to blow up the vehicle, but fragments are likely to be bouncing around (moreso downrange, which in this case is away from the rocket launchers, but in general). 10 fatalities from those strikes would be a very high number, the crews should not have been anywhere near the center of the shelling area about 100m north of the firing points and should not have been closer in to the north (that’s where the rocket exhausts will be going). So in vehicles or to the south; perhaps east or west. 10 casualties at that point would be unlikely, but more than zero.

    I count 14 to perhaps 20 shell impacts (they may have missed some in the images) in that area; so if they fired 80 rounds that’s 4-5 target areas. Scanning around…

    37°47’41.55″N 125°35’59.11″E has another battery to the same specs, less evidence of impacts there but there are 5-8 circular possible craters nearby / just south of the battery.

    37°46’25.15″N 125°35’48.69″E is near the center of 3 4-revetment firing posts, but no evident shelling.

    There seems to have been a recent grassfire fire among bunkers at the base of 37°46’10.56″N 125°35’56.97″E, a hill with a characteristic defensive hillfort. Not clearly artillery impacts though.

    Another 4 revetments near 37°46’14.62″N 125°36’27.44″E no sign of incoming fire.

    Scaling back…. Nothing more stands out as impact areas in the DigiGlobe hi-res area there. But that’s only a small fraction of the target area within 40 km of Keunyeongpong Island; the K9s could fire as far as nearly or to the outskirts of Ongjin and Haeju to the north and northeast respectively. So there are about 60 artillery shell impacts out there to find in the wider areas.

    • Magpie (History)

      (I also couldn’t find it on the NKEconwatch site, but didn’t want to say anything for fear of looking like an idiot).

  7. joshua (History)
  8. rwendland (History)

    On the K-9 reliability/accuracy, there is an interesting claim in the JoonAng Daily of Dec 29, 2010:

    “among the six K9s that were stationed on Yeonpyeong Island, more than half failed to operate on the day North Korea attacked.

    At a National Assembly hearing to examine the attack, the K9 units were found to have problems with its engine being damaged by an antifreeze liquid.”

    Two ways of looking at this. Positive spin is that only 2 or 3 K-9s were needed for an effective response. Negative of course is that preparedness was poor.

  9. Bob (History)

    It is now well known that the SK response to the attack was a farce, they were simply caught of guard. Their equipment was poorly maintained, the counter-battery radars did not seem to be working properly and it took far too long to respond. That said, conventional artillery is not pinpoint accurate- with just 80 shells fired one would not expect them to have destroyed the NK battery unless they were very lucky. Especially against SP artillery which quite clearly did a “shoot and scoot” from a pre-prepared position. That said, had the NK artillery still been in position and firing at least some of the SK shells (based on their impact point) would have had a suppressive effect from their splinters and and blast not to mention the psychological effect of their impacts. On balance though, NK won this one and left SK licking its wounds but at the same time gave them lots to learn that they can easily fix.