Geoff FordenChinese BMD Eyewitness: Real or Not?

click on the image for a larger version

I’ve been working on a rather long piece about the recent Chinese Ballistic Missile Defense test but persistent reports of an eyewitness (complete with photos) have sidetracked me. These reports purport to be from a Chinese citizen who appears to have witnessed multiple flashes/explosions. (The original English translation seems to have disappeared, luckily I printed it out to pdf, which can be viewed here.) The question is: are these credible reports/photos?

For the moment, let us assume the photograph is associated with the interception. What could it be? My guess is that it is not the initial interception. The eyewitness seems to have watched a number of phenomena in the sky before taking out his cell phone and taking a picture. (That is certainly believable. In fact, it would be too incredible a coincidence for him to capture the interception.) Also, the first things he witnessed do not appear to have been the plume from the interceptor rocket. He certainly would have reported an initial streak of light if that had been the case rather than “moons” appearing.

Instead, the image above could be a large fragment from the target burning up in the atmosphere as it reenters. Using a typical camera phone field of view of 50 degrees implies that the streak is about 1 arc second long. If it originates at about 50 km altitude—somewhere around the altitude where the atmosphere starts to get fairly dense—then that corresponds to about 0.8 km long. Of course, it has been foreshortened by some unknown amount.

For the moment, and for the sake of continuing to speculate, let us assume there is no foreshortening. We might expect a target velocity (depending on the unknown range of the target rocket) to be somewhere between 3 and 6 km/s. With no foreshortening, that implies a “shutter” time of between 0.15 to 0.3 seconds. (Shorter range target rockets would imply longer shutter times.) I’m not an expert on cell phone cameras, but that seems to be somewhat longer than I would expect possible. (Readers?) The inevitable foreshortening would lengthen that shutter time still further and assuming a higher altitude would imply an even longer shutter time. These same arguments rule out this being an image of the initial interception. So the credibility question comes down to: how long does a cell phone camera integrate over a scene at night?

There is still some wiggle room here. I need to try to calculate where in its trajectory (ie what altitude) a piece of debris would become visible but my initial reaction— subject to a lot of further work —is that this is not directly associated with the interception.
It is still possible that it is a piece of debris burning up.


  1. bph (History)

    I know nothing of exposure times on cell phone cameras, however….

    That cannot be a 1 second of arc. A cell phone camera cannot resolve that small an angle. Further, if the long axis is 1 second of arc, than the short axis is 0.2 seconds of arc. The rule of thumb is that the best ground based telescopes take images of 0.4 – 0.6 seconds of arc in size. The atmosphere blurs images to at least that size, and generally much worse at sea level (see wikipedia’s discussion of astronomical seeing for why).

    The basic equation for resolution is wavelength divided by aperture diameter (yielding an angle in radians). So a 1cm camera delivers images of ~12 seconds of arc resolution (the human eyeball is like 20-30 seconds for comparison)

  2. Geoff Forden (History)


    Quite correct! I really meant to write 1 degree, I have no idea why I put 1 arc second. (It is about 1/50 of the field of view, which is assumed to be 50 degrees. I seem to have had a “brain cloud.”) I have corrected the diagram.

    After sleeping on it for a little while, (where I realized the above mistake!) I also realized that it is possible the user had an unknown camera zoom. That could also reduce the field of view and make the required shutter time that much shorter. (Think about that for second!) However, a zoomed image just brings up another issue: there is no indication of jitter that I always get when I take a digital photo in poor lighting conditions.

  3. Jochen Schischka (History)

    I think it may be interesting to check out if something in the lower range of speeds perhaps results in more consistent findings (the ‘shutter time’ can be longer in this case); Do we know anything reliable yet about the launch location and the ‘aim point’ of the target (aka the range/delta-v of that missile)?

    Additionally, i also wouldn’t exclude the possibility that the picture was taken at a somewhat lower altitude than 50km (i’d personally assume something more in the range of 10-30km for such a light-show, especially at the lower end of possible reentry speeds).

  4. Geoff Forden (History)


    Very true, and the lower the altitude, the shorter the required integration time and the more likely the image is associated with reentering debris caused by the interception.

  5. Rupert (History)

    Camera shake could smear out the image too.

  6. Tom (History)

    I think that it pretty much has to be post-intercept debris. If the intercept occurred at 100km, and we believe the story that it was 45 degrees off the horizon at Jiuquan, then in Urumqi (where the photo was taken) the intercept would have been 2.2 degrees off the horizon, and at about 900km slant range (if it happened near to right between the two cities, otherwise things get worse).

    Since the ground isn’t in this photo, and since at that slant range a cell phone exposure wouldn’t show appreciable streaking for 6km/s, it has to be some of the debris.

  7. kme

    0.15 to 0.3 seconds certainly sounds like a reasonable shutter speed figure, particularly at night.

  8. Anon.

    Cell phone cameras (possibly with very rare exceptions) don’t feature any sort of mechanical zoom devices. Instead, zooming is achieved by taking an area at the center of the CCD cell and blowing it up in dimensions to the size the user wanted.

    This scaling up of the image introduces significant blurring to the image, since it cannot contain any more information than the (small) area at the center of the cell, yet that information has to fill the entire picture.

  9. Lee (History)


    I’m not sure that’s correct any more. On both my aged n95 and new htc hero make audible lens adjustments while taking pictures. (You can hear the optical zoom adjusting.)


  10. RAJ47

    We must first calculate the distance of horizon visible at say 45 degrees. Then calculate the time taken for two missiles at supersonic speeds traveling in opposite direction and colliding. The phenomenon if at all visible to naked eye (like the usual satellites in orbit)will be visible for not more than 5-6seconds. It will be extremely difficult even for a professional photographer to take aim and shoot.

  11. Anon.


    Nokia’s website states the N95 has 20x digital zoom, while HTC doesn’t make any mention of an optical zoom feature. Which they probably would if there was one as that could become a major selling point. The sounds you hear are from the autofocus system.

    But we’re getting quite off topic here, and since the consensus appears to be that is is a bit of debris rather than the intercept itself, there’s little to be gained from overanalysing this single photo.

Pin It on Pinterest