Strange streak discussion: 2004 Dec 7 APOD

[victorengel]

Forget the trail for a second and thing about the smoke. They said the light bulb didn't work after it happened. Light bulbs have a tendance to get real bright just before they go out. Some type of power spike cased the wires to heat causing the smoke and just took time before the bulb popped.

Not of they're sodium vapor lights. Those tend to start cycling on and off until someone notices this and replaces them.

[victorengel]

You've got the insect hypothesis wrong. In the insect hypothesis, there is NO FLASHING of the lamp. It is only the camera that flashes. It is no coincidence that the camera's flash goes off when the shutter goes off, by the way.

Flash is not used for landscape photography (it is disabled when "infinity focus" is selected on my camear). You cannot use flash photography with such extreme depth of field unless you have flash energies in the megajoules, and a flash picture would not have the lighting and deepening shadows which are so blatantly obvious in the series.

The EXIF data in the pictures say the flash was fired. I trust the EXIF data. Note I never said the flash would be useful to light the scene. It's useful to light the bug, which was close enough to be illuminated by the flash. Everything else was far enough away that the flash did nothing.

[Axle]

Has the photographer stated if there was or was not a flash unit attached to the camera?

Sorry if I missed this in a previous posting.

[Prometheus Bound]

I went back to every page looking for pictures or links and didn't find any ethereal bugs. I did learn a bit about flying poles. Cool pictures but they look nothing like the image in question. Can someone point me to a page number?

I think the bug theory carries a lot of weight. But... If this is what was captured, I can't believe that it's the first time, there have to be more pictures like this.

It would be nice to get some info as to why the experts discount meteors (I never thought lightning would be that straight, so that this has been discounted does not surprise me). I'd like to give some weight to the experts, but without their reasons, I can't accept their conclusion.

[Friend]

What is most humorous is the great lengths folks will go to try to explain something while tossing out the easiest and simplist solutions. While the wildest and craziest ideas are fun, they're bunk and consist of absolutely ZERO critical thinking. Just a hint here... PhotoShop!!! Think PHOTOSHOP!!!!

[Nat]

Looks to me like a trail of smoke from a powered flight. It's moving pretty fast--the trail is nearly straight, and it looks to me like it's narrower close to the water as if a small rocket rapidly came in from the upper left, its smoke trail spreading behind it. Could the "flash" actually be a splash?

[Guest]

You've got the insect hypothesis wrong. In the insect hypothesis, there is NO FLASHING of the lamp. It is only the camera that flashes. It is no coincidence that the camera's flash goes off when the shutter goes off, by the way.

Flash is not used for landscape photography (it is disabled when "infinity focus" is selected on my camear). You cannot use flash photography with such extreme depth of field unless you have flash energies in the megajoules, and a flash picture would not have the lighting and deepening shadows which are so blatantly obvious in the series.

While I would completely agree with you that one should not use flash for landscape photography, the EXIF image data clearly shows that a flash was fired in "red-eye reduction mode". Load the JPEG up in Photoshop if you don't believe me.

While the flash would have been useless (except maybe for catching out-of-focus bugs in the foreground), it was fired. I don't know about your camera, but mine supports a nighttime flash assist mode that doesn't change the exposure settings when you turn the flash on. This is so that you can catch both background objects that wouldn't be lit by the flash and foreground objects that would be. This is a perfect match with the behavior seen in the image, as nothing in the landscape seems to be affected by the flash.

[gbusch]

Not likely a coincidence that the flashpoint occurs conveniently at the top of a light post. Given that the light is not working It is highly conceivable that the light blew. The source of the 'streak' perhaps is a shadow of some internal structure between the camera and the filament. Looking at the photo, the shaddow progressively fades as it extends to the brighter upper left quadrant of the sky. As it 'passes' in front of the bright cloud it disappears altogether where light overpowers the effect of the shaddow....continuing on...the shaddow reappers only to fade further on in the lighter sky.

[kralassoc@aol.com]

My father witnessed similar grey streek behind bright object which was moving rapidly into chimney of our farm house . He ran into house to witness what he called in Czech language "ball lightning". The bright ball was slowly moving around warm stove. He turned back in panic and that think exploded moment later. He was not hurt, there was no heat to start the fire, but windows were blown and room was quite rearranged. This object was about football size and there was no sound associated with movement. Object in this picture appears much larger if it is in vicinity or behind the light pole.

[Guest]

I went back to every page looking for pictures or links and didn't find any ethereal bugs. I did learn a bit about flying poles. Cool pictures but they look nothing like the image in question. Can someone point me to a page number?

The link was http://asterisk.apod.com/vie ... &start=138

[Axle]

Regarding the Photoshop "hint". I saved the image, inverted, desaturated, and increased contrast. Then zoom in...

It is true that the image appears more clear than ever to be a flying insect. But, I guess I am still a little skeptical.

[James H]

Ok: Lets say it is a bug ... and lets say the "flash" froze the bug midlight (Note the EXIF data says that the flash was used in the shot)

Now if the length of the wingspan of a typical fly is say 15mm, then we can use this to calculate the approximate speed of the bug.

I did this, and calculated from the start of the dark streak to the bug must have been about 320mm.

Therefore for something to travel 320mm in 1/20 s it must be travelling at 6.4m/s or 23km/h ...

Quite reasonable for a bugs speed! Note the black streak does NOT begin at the edge of the frame, but rather a short distance in, this means that the objects speed could be no greater than that if the dimensions of the bug are correct.

If, however, it is actually an explosion, then by dead reackoning, I'd say the flash is about 2-3m in length ... again using the same process this means the speed of the object must have been about 2-3000 kph ... Far to fast if no sonic boom was heard ...

So therefore the bug theory is the most believable!

[/img]

[DC]

From the diff images, in spite of what I said above, it does appear the trail extends past the sky and over the water.

I am still focused on the shape of the reflection in the water. See my image several pages ago with lines drawn over the image. The angular part of the reflection does not go through the circular part of the reflection. Therefore, the water at that point must have a concave depression, causing the circular part of the reflection. I expect this could be caused by a compression wave preceeding a meteor which exploded just before entering the water.

I can think of no other satisfactory explanation for the shape of the reflection. This sounds good to me.

[Guest]

Forget the bug option. No chance. From previous post, I agree and believe the light is on a mast. The high poles look like street lights I am use to and the two smaller ones do not.

The smoke from below is probably from a boat starting its engine and a light on the mast may in fact blink on occasionally (e.g. during startup).

The timing of the light burst and the trail are far too coincidental. This is clearly an optical affect probably due to the light itself. The halo affect around the light again is due to the light itself and possibly the smoke.

While a contrail may look like this, again, it is far too coincidental it shows up exactly when the burst of light shows.

This should clear up everything.

It might, except for the fact that everyone from Darwin who has talked about this has discussed how the utility workers inspected the light pole's bulb. One would think that they might have noticed that what they were inspecting was a mast and not a streetlight pole. And one would think that they wouldn't be calling it a streetlight pole even after that. Don't you think?

[Kagato]

Assuming it's a digital camera, I'd wager the "smoke" might be nothing more than a camera artifact produced by the CCD.

Most digital cameras, when pointed at an intensely bright light source, show streaking across the picture (commonly a vertical band in one or both directions) through the light source.

The streak in this picture appears to be about 30 degrees off horizontal. Some cameras have CCDs with the red, green and blue sensors arranged in a triangle:

Code: Select all

        R G
         B

So, if the street light had flared up brightly for some reason, it would have mostly affected the red sensors... which in my diagram is at the top-left corner -- at 30 degrees off horizontal.

Of course, I don't really know a lot about how CCDs work, so I could be way off base...

[Crimulus]

The most difficult part of the bug theory is that the bug doesn't look the same through the whole picture. You have to realize the shutter was open for the entire 1/20th of a second. The bug should look the same no matter where he is in his trajectory. The flash could have changed his appearance but according to this the flash lasted way less than 1/20th of a second and occurred only at the very last moment the shutter was open. This is not very efficient if you are a camera maker.

It simply cannot be a bug. Looking at all these pictures the line looks most like a contrail, but it shouldn't be there. The circle near the flash at the light could be a lens flare, but what would make that much light.

It does appear that if the straight line in the "smoke" is actually a reflection on the water, it is perpendicular to whatever the black line coming down to it is.

I think that whatever may be certain it is that the correct interpretation of this picture has not been proposed. (Save the photoshop theory, but I'd prefer to think they used the Gimp -- but why would someone waste their time on a picture that really has no weight as a hoax)

There is always at least one aspect in contradiction to all these theories. Perhaps a meteor came in in two pieces and ones explosion reflected in the others ripples? but what is that line going across the ripple?

Why do we even care !!! Blast this cursed intrigue

[John Poling]

I agree with Paul Doherty from the Exploratorium. I have seen shadows of jet contrails a number of times and they look just like this. They can be seen only when the viewing angle is just right, and they usually can't be seen for very long.

John Poling
Physics Department
California Polytechnic State University

[Guest]

What is most humorous is the great lengths folks will go to try to explain something while tossing out the easiest and simplist solutions. While the wildest and craziest ideas are fun, they're bunk and consist of absolutely ZERO critical thinking. Just a hint here... PhotoShop!!! Think PHOTOSHOP!!!!

Digital camera have enough funky artifacts of their own -- you don't need Photoshop to get weird shots like this. People are speculating about this particular image only because it's not immediately obvious what the artifact actually is.

[Quinkin]

I feel it is worth pointing out to the anti-meteor proponents that there have been a number of loud, bright phenomena in the skies of Australia, presumably a precursor to the Geminid shower.

http://ta.harrisgroup.com.au/editopinion.cgi?id=154577
http://taree.yourguide.com.au/detail.as ... =2004&m=12
http://www.smh.com.au/news/National/Met ... click=true

Although not neccesarily related geographically to this particular instance, it does raise one possibilty.

Possibly a larger meteorite was explosively fractured with the resultant turbulent shockwave imparting large lateral velocity upon some fragments of the original meteor. This would explain much of the high speed (the original meteor had a much higher terminal velocity), shallow angle of incidence and possibly even the perceived downward bend of the "trail" (as friction finally catches up with it). None of this even starts to look at the effects of non-spherical objects and rotational dynamics.

Also, has anyone actually inspected the top-plate of the street light? I can't be sure on the make but the lamp seems remarkably similar to many street lights that have a quite large and heavy galvanised steel top plate. It is more than conceivable that a micro-meteor could impart enough energy to completely anhilate itself, while not having enough enrgy (and to oblique an angle of impact) to damage the street light such that the damage could be observed from the ground.

The insect seems to fulfill Occam's Razor, but personally I find the micro-meteor fragment concept far more satisfying.

Q.

[Guest]

So, let me get this straight... the odds that this is some weird straightline lightning caught perfectly on camera are a bazillion to 1; the odds that this is a meteorite striking the water or a lamp or a mast or a bug and it was caught on video is a trazillion to 1; the odds that this is bug or a lamp blowing up is a million to 1; and the odds that somebody with 2 minutes on their hands and a decent computer with a graphics program and the will to edit a photo are probably 2 to 1... hmmm lets see... what would I suspect first of all? Golly, that's a tough one... no no what was I thinking?? It's gotta be the bug from Jupiter hitting the exploding light bulb with an atomic dust trail from space trailing behind it... yeah, that's it.. gotta be!!!! hahahahaha!!!

[Gustopher]

Point #1: This is certainly not a meteorite impact.

The length of the path "trace" in the photo is about 205 meters based on the assumption that the white sedan near the lamp post has a height = 1.45 meters, and a further assumption that the path trace is in the same plane of view as the sedan (plane perpendicular to camera line of sight). The analysis is not sensitive to these assumptions.

Given a shutter exposure speed of 1/20 sec, this means the meteorite must have had a minimum flight speed of over 4 km/sec. This is a plausible atmospheric entry velocity for a meteor (10-70 km/sec according to http://csep10.phys.utk.edu/astr161/lect ... pacts.html).

However our hypothetical meteorite must have been quite small to avoid any damage, as evidenced by the "after" photo taken 15 seconds later (sequence determined by EXIF timestamp not by filenames; exposure speed also verified from EXIF data). Smaller meteorites can be slowed to terminal atmospheric free-fall velocity, which is more like 0.05 km/sec. So either you have a monster meteorite, relatively unhindered by air resistance and moving fast but with no evidence of final impact, or you have a tiny meteorite moving impossibly fast. The simple conclusion is that this is not a meteorite trace & fireball event.

Somebody with more time could probably estimate a meteorite size given the size of the fireball in the image and assumed impact with the lamp post - try playing with the calculator at
http://liftoff.msfc.nasa.gov/academy/sp ... azard.html.

Point #2: This event, whatever it was, had something to do with the lamp post. The illumination from the fireball bears this out.

Check out these three analyses of the original images:
http://ccouch.home.insightbb.com/meteorite.htm

Image 1 shows the before, during, and after shots.
Image 2 shows a threshold setting applied equally to all 3 images. Clearly the lamp post indicated by the pink arrow is brighter in the middle frame.
Image 3 shows another lamp post, the one nearest the fireball, also clearly brighter.

This implies the fireball is somewhere near the lamp post in image 3 and is providing illumination for the lamp posts in question. Therefore the fireball is NOT an insect near the camera lens being lit up by the camera flash.

So, this leaves us with hypotheses about the lamp post. Exploding bulb? Other? Direct inspection on-site should yield clues.

Hope this helps - write me for clarification - ccouch@insightbb.com

[dave potter]

I think it is the shadow of an airplane contrail from a cloud layer above the one we are looking at.

[Guest]

Regarding Point #2 - of course another possibility is that the image is faked, but I'm inclined to give the benefit of the doubt in light of the lamp post illumination demonstrated on my web page. This would be a pretty subtle thing to fake, but by no means impossible. Heck, I thought to check illumination using threshold settings, so I guess a good faker might think of it too.

[Guest]

The most difficult part of the bug theory is that the bug doesn't look the same through the whole picture. You have to realize the shutter was open for the entire 1/20th of a second. The bug should look the same no matter where he is in his trajectory. The flash could have changed his appearance but according to this the flash lasted way less than 1/20th of a second and occurred only at the very last moment the shutter was open. This is not very efficient if you are a camera maker.

There is a very commonly-used camera mode (for both digital and film cameras) called nighttime assist. It takes a long normal exposure to catch ambient-lit objects and then uses a quick low-level flash to light up objects in the foreground. This is how you take a flash picture at night and get city lights in the background. This is not a poor choice on part of the camera manufacturers, it is a very useful mode feature.

Now I don't know why that mode was turned on for this particular shot, but the fact that we know that the flash was fired without underexposing the background would suggest that the camera was in fact in this mode.

[knalty@swbell.net]

Greetings,

I believe this to be a saturation effect on a progressive scan LCD camera.

This camera scans each image from the bottom right up. However, the location in memory of the scanned cell does not make a vertical line due to refresh blanking intervals. I believe the camera caught the flash, and the dazzled pixels were seen in the darkened streaks leftward in the image.

I would be very interested in learning which model camera caught this image.