Jeffrey LewisHow Tall Was the Box

The debate continues over Yousry Abushady’s remarkable appearance on Egyptian television.

The interview is notable, as one observer notes, because Abushady “is speaking in his personal capacity, which is soon-to-be retired.” You just don’t see IAEA officials doing television interviews like this.

Although I still have lots of questions about the type and size of the reactor, I continue to think the evidence, on balance, supports the reactor hypothesis.

James noted his skepticism regarding Abushady’s comparsion of the heights of the probable reactor at Al Kibar (Dair Alzour) and the reactor at Yongbyon.

Frank Pabian sends along a careful comparison of photos that makes that skepticism explicit. To read his complete rebuttal to Dr. Abushday, just click on the image, which more accurately shows the relative heights of the building.


  1. Yossi (History)

    Abushady chose this weak argument against the nuclear theory because he appeared on TV and needed something simple that the Egyptian masses can easily understand. There are many other arguments as Abushady himself says and Jeffrey’s gayness is premature. USIC honor wouldn’t be saved by exploiting a mistake made by a soon to be retired official.

  2. Rwendland (History)

    It doesn’t make any real difference to Frank Pabian’s overall conclusion that BoE is about 20m tall from the USIC evidence, which I broadly agree with; but the apparent distortion in the Yongbyon photo Frank Pabian identifies on page 2 of his rebuttal is mostly, if not entirely, due to Abushady using the original photo from Sig Hecker’s website, but Frank Pabian using a version that has been image corrected by a Wikipedia editor.

    The original photo, taken by a Senate staffer with a handheld camera, was a fair bit off-vertical. So a Wikipedia editor decided to correct that, and Frank Pabian has used this corrected version. This vertical-axis correction makes the left side of the Yongbyon look lower compared to the original, which Frank Pabian has mis-identified as a distortion in the (original) photo that Abushady used.

    My part in this is that I put the original photo onto Wikipedia commons. Another editor subsequently decided to do the vertical correction, as is the way on Wikipedia, and I decided to leave the corrected version as the main Wikipedia version. Perhaps I should have reverted that change back to the original to avoid this confusion!

  3. FSB

    did anyone ever see any other inside pictures besides that one of the “core”? if so we can get dimension from those also to compare.

  4. James Acton


    I don’t normally refrain from commenting on comments but you are being disingenuous here.

    If USIC had said they had 30-40 reasons for something, but only made one public and that turned to be very dubious you would not be defending them—and quite right too.

    You should apply the same standard to Abushady.

  5. M. Calloway (History)

    Alternatively, Yossi, the height argument was chosen by Abushady as being [supeficially] convincing or perhaps his strongest (what would that say about the other 30-40 faults he claims to have found?).

    I’m not saying this to stand up for the USIC’s hono[u]r, belive me, I’d be the last to do this – but as Blowback said in a previous post:

    “it is unhelpful to have such analysis coming from an IAEA official in his “personal capacity”. If the IAEA has concerns about this they should be in the official reports”.

    If there are convincing arguments (and I’ve not seen them here or elsewhere) to refute the claims made in the ‘CIA briefing’ that the BoE was indeed a reactor then they should be brought into the open, not just aluded to.

  6. Hmm... (History)

    Is Abushady only soon-to-be-retired or maybe also soon-to-enter-politics? Time will tell I guess.

  7. mike

    just curious, but with the talk of the BOE having a substantial underground – what is the deal with Yongbyon? How much of that is below ground and to what purpose and necessity?

  8. Yossi

    James Acton, thanks a lot!

    I gladly accept your critic. I may well be biased because of my strong support for the United Nations and its institutions. This is just an ideological tie, there is no formal or personal relation. I also have instinctive dislike for situations where a great power seems to push around a small backwards country but of course this doesn’t mean the great power is wrong.

    If the USIC said it has 30-40 reasons and published only the worst one we would have a moral duty to ask for more info but it would look strange and people would probably mention ancient cases not supportive of good reputation. However the mighty USIC with its tens of thousands of employees and huge resources can’t be easily compared with one retiring official of a relatively poor agency who appears in his personal capacity on TV.

    It may be incorrect in principle to compare the IAEA with the USIC as the former must consider the benefit of all UN members while the latter must abide by the interests of a single country. It’s natural to assume the IAEA has purer intentions than any national intelligence agency and I think history supports this view.

    God help me but I start to sound like Ms. Fleming.

  9. Andy (History)

    While this discussion of the BOE is very interesting, I think we all need to be aware of the threat imposed by the world’s newest nuclear power.

    I will say I did another BOE height calculation on two images using shadow length and sun angle and I came up with 21m and 20.3m.

  10. Andrew Foland (History)

    By calculating a perspective transform on one of the images, I found in April that the height was 53+-5% of the side length. Assuming a side length of 45m gives a height of 24+-2 m.

  11. Yale Simkin (History)

    Back on Oct 29 and Oct 30 last year I posted what I think was the first open literature estimate of the height of the Big Box.
    (Yeah, ok… someone disillusion me…)

    My initial crude estimate was ~24 meters, which I feel is between an eighth to a third too high of actual.

    This is the post which used the same methodology later used by Frank Pabian:

    These are estimates (in increasing size) using a variety of techniques from ACW:

    blowback: Less then 10m

    Yousry Abushady: 10m

    AYA: 10 to 12m

    Geoff Forden: 17.5m

    Rwendland: about 18m

    Frank Pabian: 19.6m

    Andy: 21m and 20.3m

    Andrew Foland: 24m +/-2

    Yale Simkin: 24m

    hairs: 26m +/-5 most likely towards the lower range

    Here are some various useful images, actual and virtual:

  12. Jeffrey Lewis (History)

    Thanks, Yale.

    That’s extremely helpful.

  13. Rwendland (History)

    We need to be wary of methods based on direct measurements of the photos from the USIC video, as one image clearly has about +35% y-axis (height) distortion. If the photos also had a similar height distortion then a 20m apparent height would mean a real height of 14.8m, and 18m apparent height 13.3m real height

    Below are two half size images for easy comparison, clearly showing the y-axis distortion.

    First the distorted satellite image from the USIC video (from GlobalSecurity – full size image link ):

    Compare to a satellite image from ( full size image link ):

    I carefully measured the image roofs on a 17” 1280×1024 LCD. The y-axis / x-axis ratios of the roof from 3 images are:

    USIC-GlobalSecurity: 6.3cm / 4.3cm = 1.465 image above: 6.0cm / 5.6cm = 1.07
    Google Earth: 17.3cm / 15.4cm = 1.12

    So comparing the roof ratios, the USIC-GlobalSecurity image has +37% y-axis ratio distortion compared to the image (1.465 / 1.07), and +31% y-axis ratio distortion compared to the current Google Earth view (1.465 / 1.12).

    I’ve watched the USIC video, and the distortion is in there, not introduced by GlobalSecurity.

    This could easily be innocent distortion from image to video conversions, or a bit of carelessness. It is more likely this distortion only affects this single image, but we cannot disregard the possibility this also affects other images.

  14. Yale Simkin (History)


    Pabian did not use ODNI images for his “Methodology #1 (Comparative Length of Shadows)” determination of Box height.

    He used an off-the-shelf picture:

    – not that it matters, since measuring relative shadows works just as well with a skewed image.

    The height he extrapolated (19-20 meters)was right in the ballpark with the central estimates generated by others here using differing methods.

    His 2nd method, using a side photo of the Box, was a reality check against the first.

    If the “ground truth” photos by the Syrian mole were seriously compressed in the X-axis in the video (deliberately or otherwise), then the 2 methods would disagree in results.

  15. AYA

    Dear all, I see you are still guessing the Syria building height with simple assessments. As a professional in the field of satellite imaging for many years, I would like, for once, provide you with precise results based on calculations not assessment. To calculate the height of any building based on a satellite image, we need the following: Latitude (35 42’ 28.20”), longitude (39 50’ 00.05”), date when the image was taken (e.g. 18 August 2007), exact time of the image (unknown) or Altitude (Angle of the Sun rays with the ground level|) (also unknown) or the Azimuth Angle (horizontal shadow angle measured eastwards from north) [≈ 150 degree). Measuring the total shadow length and noting the camera off-angle (Nadir angle or deviation from exact verticality), the horizontal shadow can be calculated. Using all above parameters in a sophisticated software, the result of the building height would be 10.5m with +-1m uncertainty. That is why the figure for the height between 10 to 12m is a realistic scientific result.
    Now let me respond to some of repeated errors in other calculations, mainly:
    The effect of the Nadir (off) angel is actually ignored. The total shadow was taken as the horizontal shadow ignoring that part of this shadow is the vertical side of the building itself.
    The use of the mixer cement truck as reference for assessment was also not precise, as a correction should be introduced as the two object were not differently oriented towards the Sun. The height of the truck was taken as 4m causing the shadow ratio to be 1 : 2.25, while the maximum physical height of similar truck is 3 – 3.5m which should give a ratio of 1 : 1.5 . With these adjustments, with no precise calculations one may still get the height from 10 to 14m.
    Although Abushady referred to the lowest value (10m), one could understand that he was talking to public which needs much simple presentation to understand it.

  16. AYA

    Just one final quotation:

    “Even when the experts all agree, they may well be mistaken.”
    Russell, Bertrand

  17. Rwendland (History)


    Yes, I think the shadow analysis techniques have much to commend them. I rather like Geoff Forden’s technique that uses the satellite timestamp and his calculateHeightFromShadow program, that comes up with a 17.5m estimate.

    Comparing shadow length with that of a unknown concrete truck must have a large error margin. I noticed that some of the shadows at the nearby Halabiye Castle historical site are about as large as that for the BoE, on Google Earth. These appear to be from the same satellite pass. If anyone had detailed measurements of the objects and terrain of that historical site, that could make for a very useful new estimate.

  18. Andy (History)


    It’s been a long time since I’ve done these kind of calculations, but based on my figures, 10m seems almost impossible. I plug your numbers into this model (150 degree azimuth angle and 10m height) and it calculates the shadow length at 4.7m, which is quite a bit less than what I measure it as. The off-nadir angle for this image is small according the digitalglobe’s metadata (which is probably why it was picked for GE). I don’t think the small off-nadir can account for the discrepancy, but it’s been a long time since I’ve done these kinds of measurements, so I may be wrong.

    For everyone’s benefit, here’s the metadata for the image currently in Google Earth:

    DOI: 18 August 2007
    Avg Off-nadir angle: 2 deg
    total max off-nadir angle: 2.77 deg
    area max off-nadir angle: 0.82 deg
    area min sun elevation: 64.63
    avg target azimuth: 27 deg

  19. Yossi

    Yale and Rwendland has kindly provided the following two views of a satellite image in which BoE’s electric cable seems to be bent under its own weight:

    This seems incompatible with an image shown in the famous USIC presentation:

    What do you make of it? Is it a reason for impure thoughts?