Bob Atkins shot a video on a beach while walking his dog. Later he noticed a flying object on it. There has been some media coverage such as
GeneralThere are numerous possible explanations. This kind of ufo cases has become common when digital cameras have become an everyday item available for everyone. JPEG/MPEG compression tends to make random bugs, drips, or dirt appear smoother than they really are. The unexplainable nature of the Cumbria ufo has been justified with the "immense" speed of it. I've tried to approximate the speed here. Absolute speed would require information about the object's size, but calculating the speed relative to its diameter is rather straightforward to measure. Shape
There isn't much we can say about the shape of the object, which is typical of this kind of "camera sightings". The shape doesn't appear to be exactly regular. It's difficult to say how much the JPEG/MPEG compression has affected it. It's notable that there is apparently no motion blur which suggests a limited speed relative to the exposure time. Duration of movementFrames 79..85
Difference between frames
The object appears only in two different locations. I approximated the time it took to move between these locations. For this purpose I used the video file uploaded to Facebook by Bob Atkins because it seems to be the highest quality version available. On first sight of the video we can see that it's rather jerky, the frame rate is irregular. The nominal frame rate is 25 frames per second, but the real rate is lower at times, especially during zoom. During these times the last frame of the video stream is duplicated because the camera hasn't shot frames quickly enough to fill in the 25 fps stream. The ufo appeared while Mr. Atkins was zooming in. At this time the real frame rate appears to be between 5 and 10 frames per second. The ufo appears on five frames, which of four last are identical. Frame #80 shows the ufo in its first position (on the right-hand side). Frames #81, #82, #83, and #84 show it in the second position (left and further from the camera). The duration of the movement calculated directly from five frames would be 5 * 1/25 s = 0.2 seconds. However, as we can see the timing of the camera is irregular and not exact. Rather than using this directly calculated value we should consider some error margins. The second position frame appears during four 1/25s frames in the result video stream so the timing error must be at least 4 * 1/25 s = 0.16 seconds. For simplicity I've used the approximation 0.3 seconds for the time of movement, but we have to keep in mind that if we append a 0.16s error margin in both ends of the time slice, it will grow by over 50%. One way to think about this is, that the object may have appeared already during frames 76 to 79 while the camera didn't actually produce any new frames. In the similar way, its moment of disappearance may be a few frames off in either direction. In any case, this "error consideration" doesn't have much impact on the result below, it is just to emphasize the degree of uncertainty. DistanceIt's impossible to measure the absolute distance the object travelled because we don't have any idea of its size. However approximating the distance relative to diameter is trivial. I've used the two still photos uploaded by Bob Atkins to Facebook as they seem to be the highest quality available (720x540 pixels). The width of the object is about 23 pixels in the first frame, 17 pixels in the second one. The distance travelled, measured between the centerpoints, is about 404 pixels. So during the 0.3 seconds the object travelled about 20 times its width. I've only considered the movement component perpendicular to the screen. The speed component along the depth axis appears to be rather small in comparison. SpeedIf we presume different kinds of sizes for the object, we can reach the following absolute speeds for it from the above measurements (speed = 20*width/0.3s). Because of the timing inaccuracy these should be treated as rather rough approximations with potential error of at least 50%.
These approximations seem to fit several natural explanations. I will take a closer look in one of them. The dogThe actual target for the video was Bob Atkin's dog. Mr. Atkins and another person (unseen in the video) were playing with the dog just prior the "ufo scene". This is shown in the original video clip released by Mr. Atkins on YouTube. It's revealed that the dog has been jumping and running deep in a pool of water. Just before the "ufo zoom" the dog appears in front of the camera, a bit on the left side, shaking water and dirt off itself. Then it runs out of the picture to the right side of the camera. After the ufo scene, the camera turns right, back at the dog, who is sitting next to Mr. Atkins and wagging its (seemingly dirt covered) tail. One explanation for the object might be a splash of dirt, dirty water, sand, or similar, which flies when the dog shakes itself or just wags its tail next to the camera. The trajectory of the object - from right to left, and away from the camera - fits this explanation, as well as the color and irregular shape. If the object is e.g. a splash of dirt about 1 centimeter wide, the approximated speed would be around one meter per second, which would seem reasonable. SoundThe mono audio track doesn't reveal much, but we can see if there is any evidence to support the above explanation on it. At 02:28 on the original clip we can see the wet and dirty dog shaking itself. This creates a series of sharpish noises. Some of these noises are "two-parted" which might originate e.g. from its flapping ears or the collar. These two "parts" are separated by 25 to 30 milliseconds. Sonogram of this shaking noise is here:
Right before appearance of the unknown object, after the dog has ran next to the camera on its right side, we can hear two somewhat similar sounds. These are "two parted" sharpish noises. Separations between the "parts" are about 25 and 35 milliseconds. These noises can be seen in this sonogram at about 2:36.8 and 2:37.4. The ufo appears at about 2:38.7. The timestamps refer to the original, longer video clip.
About 1.3 seconds passes between the latter, possibly "shaking" noise, and appearance of the ufo. If we presume the object was a dirt splash about 1 or 2 centimeters wide and started its flight when we hear the sound, it would have advanced about 1 to 2 meters before appearing on the video. This would seem consistent with the dog's apparent position. At that moment the dog is also very close to the pool of water, so merely running or jumping could be enough for a splash to be thrown. Conclusion
There are usually numerous possible explanations for this kind of videos. The one discussed above, a dirt splash thrown by the dog,
seems quite reasonable. The dog has just been in the water, is dirty, has been shaking itself just seconds before the "ufo".
It's position next to the camera fits with the trajectory. The speed of the object fits this explanation and some support
for it may be found in the audio.
2010-10-17 pouko@touko.cjb.net |
