Strange streak discussion: 2004 Dec 7 APOD

[Cloudbait]

I know little about cameras, and even less about digital ones. But it seems illogical to me that a flash would only be effective for a fraction of the time the shutter is open. 1/20 is a relatively short time. How long did the flash last for? I'm having difficulty accepting that it was effective for a fraction of 1/20..... (and therefore highlighting the bug for that instant only)...

Appreciate an explanation.

Depending on how bright the camera thought the flash should be, its duration was somewhere between 10 microseconds and 1 millisecond. Cameras are almost always set up to fire the flash immediately after the shutter opens (first curtain sync) or immediately before it closes (second curtain sync). So with a 1/20 second exposure, you would expect a very short flash right at the beginning or end.

[Guest]

Yeah, it's not obviously a bee. To me, it appears to be something greenish between the "wings", more like another "body segment" or "head", perhaps. Maybe a biologist could see whether it resonably resembles any insect sommon in the area.

Back on page 98, we got a report from the site - here's part of the message from back there.

I live an Darwin and have just visited this spot for the evening sunset. I noticed lots of flying ants at the rate of about twenty a minit. They are about half an inch long with their wings longer than the body. Their flight pattern viewed in the same direction seemed to match the angle of the track more often than not. they are slow flying but because of a light breeze tend appear to fly very straight. the wing beats are visible with the naked eye but only just. Body is the color of honey and the wings are transparent and rounded at the tips. The cliff is about 50ft with trees at the base so you are looking over the canopy of the trees. Other bugs of interest are dragon flies moving very fast and close to 4 inches long but their body is long and thin so highly unlikely. These white ants come out in the evening after rain in the previous days to mate and will last in plague preportions for at least another month until the wet season sets in properly. A photographer targeting these flying ants should have no trouble getting some in a photo to check for track based on shutter speeds similar to waynes photo.

Then could someone please post a picture of a Darwin white ant.

[Ricardo]

Just to add something to the discussion, perhaps. Of any of the responses I've read, concerning contrail shadows etc, no-one seems to have taken into acount the direction of the sunlight. which is in this picture from the upper right (note cloud shadows). To get a shadow that long and staight, of consistenet width and density, would require a contrail very high and dense. To be seen, the shaddow would also have to pass through a layer of cloud, and still have enough contrast to be seen against a dark background. And all this, only (for this picture) if the sun was in the opposite side of the sky. I would rule out the shaddow option.

I have seen an interesting contrail shadow before: when in a plane, I was looking at the ground from the window, and was trying to work out what a strange line was that crossed all types of terrain and infrastructure. Was is geological etc.? It couldn't be natural so long and straight to the horizon... surely I thought. Then, as the line became perpendicular to the horizon, appearing to go beneath me, we flew over a vapour trail of another plane by about 100m, heading perpendicular to our motion.

A dead straight long white bar, 10s of metres wide (near me, and tiny in the distance), stretching to the horizon in both directions, floating thousands of metres above the ground, is not something you often see.

Interesting disucssion, and yes not really that scientific anymore, but certainly getting us all thinking - and we all need to to that.

Ricardo - I am, therefore I think.

[victorengel]

I know little about cameras, and even less about digital ones. But it seems illogical to me that a flash would only be effective for a fraction of the time the shutter is open. 1/20 is a relatively short time. How long did the flash last for? I'm having difficulty accepting that it was effective for a fraction of 1/20..... (and therefore highlighting the bug for that instant only).

Having said all that, I did see the 'experimental scientist's' mock-up. In that image, the bee seems to cause the same sort of effect as the Pryde photo - but I don't understand it.

Appreciate an explanation.

Here is how the camera works. The shutter consists of two curtains. For the sake of this discussion, I will use the analogy of a theater stage with curtains that close from the sides. But whereas in a theater, the curtains close in the middle from each side, in a camera, both curtains go in the same direction, let's say left to right.

Before the exposure starts, both curtains are at the left. The left curtain is open, and the right curtain is closed. When the exposure starts, first the right curtain opens by moving to the right. Some time later, the exposure ends by the left curtain closing, also moving to the right.

For extremely short exposures, the left and right curtains leave only a slit, which travels from left to right. For longer exposures, the right curtain opens completely before the left curtain starts closing. Such a long exposure is necessary for a flash exposure, except if the flash has a high speed sync feature, which I won't get into, except to say that in that case the strobe fires repeatedly as the slit crosses from left to right. That's not what we have here, so I'll drop discussing high speed sync.

Now, then, without high speed sync, the flash can only fire when both curtains are completely open. Now it does take some time for the curtain to travel from one side to the other. Then it takes additional time for the second curtain to also travel across. The fastest this can occur is the flash sync speed. For most cameras, this is 1/125 or 1/250 second. I don't know what the sync speed is for the G3, but for the sake of discussion, let's say that it's 1/250 second. Half of this time is spent moving one curtain. The other half is spent moving the other. That means it only takes 1/500 second to move a curtain from one side to the other.

The flash is timed to fire at the moment the first curtain moves out of the way, or the last possible moment prior to closing the second curtain. The feature of this picture we're discussing is in the center of the frame, so if the curtain takes 1/500 second to cross the frame, it takes 1/1000 second to get to the feature we're discussing. So there's possibly only 1/1000 second of blurring that could occur from ambient light after the flash.

Estimates of the flash duration itself are for about 1/1000 second at the longest, so the amount of additional trail after the flash should be comparable to the amount of blur in the flash-exposed image itself attributable to motion blur (not lens focus).

I hope I didn't lose you with all that.

Let me put it a different way. The camera manufacturers time the flash to occur at one end of the exposure or the other specifically so that when pictures are taken, it will look like a blur with a bright spot at the END of the blur. Since that is the design intent, I think it's unlikely we would see significant ambient light exposure beyond the flash.

[HawaiiArmo]

Hey Doc Bluto,
I'm still waiting for your fake images that you claimed to have created. You talk a good talk, but you have no facts to back up your statements. I want to see these images that you bragged about. Anything will do, even the most low-tech artificially created image would suffice, and it would give you some credibility on the issue. Right now, you sound as credible as Bush when he says he backs stem cell research. I don't want this to sound like a threat or a criticism, but the best way to change my opinion on the subject, as Hazei did, is to post an image that utterly refutes the other claims.
This thread is dead because everybody keeps beating a dead horse, and no proof has come yet to counteract Hazei's apparatus and image he created.

[stainless004]

This could be the aftermath of a lightening strike. A ground to cloud strike usually is more vertical, but is often straight as the charge leaves the ground. The shutter could have been snapped a split second after the strike, and the glow around the pole is plasma in the process of dissipating. The dark upward "shadow" could be rarified air produced by a hot bolt of electrons . Has anyone ever taken a picture milliseconds after a lightening strike?
If you don't like this explanation, I'll return your money.
—Eddie

[Can't use my Bad Buoy]

Just to add something to the discussion, perhaps. Of any of the responses I've read, concerning contrail shadows etc, no-one seems to have taken into acount the direction of the sunlight. which is in this picture from the upper right (note cloud shadows). To get a shadow that long and staight, of consistenet width and density, would require a contrail very high and dense. To be seen, the shaddow would also have to pass through a layer of cloud, and still have enough contrast to be seen against a dark background. And all this, only (for this picture) if the sun was in the opposite side of the sky. I would rule out the shaddow option.

And it IS sunset. The sun has left the surface but still shines directly on the clouds; so it's tangent to the Earth. Where does that contrail have to be then?

Yes, I think there are many reasons [this shadow geometry, shadow extends onto land below horizon, no appearance 15 sec before or after] to find the contrail theory invalid.

[hazeii3]

On the subject of bugs and flashes - do I have this right? If the flash went off at the end of the exposure, then the bug was travelling from top left to centre (conveniently placing it centre-stage). So shutter opens, captures fast-moving bug-trail, illuminates fixed position with flash, and then? Does the shutter close immedietly during or after the flash, or does it remain open for any length of time? If there was any time at all between flash and shutter-close, then the blurred streak should continue on for a while, no?

The reverse is true if the flash went off at the opening of the shutter (and bug travelling opposite direction) - any time between the opening and the flash, and the streak would start before the flash-point.

I know little about cameras, and even less about digital ones. But it seems illogical to me that a flash would only be effective for a fraction of the time the shutter is open. 1/20 is a relatively short time. How long did the flash last for? I'm having difficulty accepting that it was effective for a fraction of 1/20..... (and therefore highlighting the bug for that instant only).

Having said all that, I did see the 'experimental scientist's' mock-up. In that image, the bee seems to cause the same sort of effect as the Pryde photo - but I don't understand it.

Appreciate an explanation.

I'll reference my explanation to my own pictures (since I know all the details in that case!).

Generally, the darker it is, the longer the camera shutter has to stay open for in order to collect enough light to get a good picture. The trouble is, if the shutter stays open for a long time, anything that moves appreciably during that time gets blurred along its direction of travel (if you try photographing a car race close up, you have to use shutter speeds of 1/1000th second or higher, and even then you have to track the car during the shot - that gets you a nice sharp car, but it blurs the background, no doubt you've seen the effect).

1/20th of a second is actually quite a low shutter speed - in fact, at that speed it's pretty hard to get a good handheld photograph because your hands shake a bit (at least, mine do) and then everything comes out blurred (that's one reason people use tripods to support their cameras).

So if we photograph anything moving with an exposure time of 1/20th second, we could reasonably expect it to be blurred along the line of travel. Now 1/20th second might not sound a long time, but even 1 mile per hour is about a foot and a half per second. So an insect flying (or being blown by) at 5mph works out to cover about 4 1/2 inches in 1/20th second - if it was close to the camera, that's several times its body length.

In the case of my photo, you can make out the 'trail' this caused; judging by eye, I'd say it's about 5 times the length of the body of the wasp that made it. As the wasp was about 1/2 inch long, the streak is about 2 1/2 inches.

So, while the shutter was open, the wasp moved across the field of view, blocking part of the sky in its travel and thus creating the dark streak.



But what about the bright blob at the end of the trail? Well, the reason for this is because I had set the camera up to use the flash as well. Now, normally a camera flash is used when it's dark to get around the problems of blurring; it adds a lot of light to the scene so letting you use a much shorter shutter speed (the flash itself lasts around 1/1000th of a second). In the case of the image above, in 1/1000th of a second the wasp would have travelled less than 1/10th of an inch - a small fraction of the body length.

So the picture above is effectively composed of the trail caused by the motion of the wasp during the slow exposure, plus a bright picture of the wasp caused by the short but intense pulse of light from the flash (though the wasp is also blurred because it's too close to the camera, which was meant be focused on the trees in the background; you can tell I'm an amateur!).

Effectively, the camera flash has acted like a single-shot stroboscope. What's interesting is we can tell that the flash actually occurred at the start of the exposure - I was spinning the wheel it was attached to so the wasp was flying head-first - see the picture below, if you imagine the camera being moved down to take a photo of this bug up against the sky background, you can probably relate the body shape to the 'blob' in the picture above.



Finally, I haven't explained why anyone in their right mind would use both a slow exposure with a flash, or why a supposedly idiot-proof automatic camera (like mine) would even allow such a thing. Well, I noticed my camera has a few different flash modes - red-eye reduction and so on - plus one called 'synchro'. I never knew what that was for, but I do now (thanks to this discussion!). Imagine you're out for a stroll at night with a friend, and they stop and lean against some railings; you think it'd make a great picture, but there's a problem. You want to catch the city lights in the background - so long exposure needed. But you want to use flash else your friend is only going to appear as a black silhouette. The answer? 'Synchro' flash! The long exposure captures the city lights perfectly while the flash catchs the smile on your friends face.

But when you get home and inspect the picture closely, you notice there's an odd streak on the picture, ending in a bright blob. It looks vaguely like an incoming meteor! For a moment you're puzzled; but then you remember the moth that fluttered by just as you pressed the button...luckily for you, it doesn't spoil the photo, and you quickly forget about it.

[Boldra]

For those who haven't seen today's APOD http://antwrp.gsfc.nasa.gov/apod/ap041221.html, there is a link noting that the best explanation http://www.cloudbait.com/science/darwin.html has been chosen. It is of course... a bug.

Various other explanations were NOT ruled out (fraud, rockets and strange lightning) but the bug explains most of the observed characteristics whilst not stretching probability very much.

I suggest that the people who still support alternative theories (UFOs, contrails, meteorites or Tina Turner) need to focus on disproving the bug theory to make headway.

Boldra

[DC]

I posted the other day that I had seen some "distortion" in looking at a diff of before and after. It turns out that the distortion I could see was apparently movment in the tree between the lamposts and not in the water. However, while investigating that, I found that this thing, whatever it is, has more structure to it which becomes apparent in different wavelengths when doing a diff after doing a filter. It is hard to point out, but definitely there. I'll try to come up with some images in the next few days that clearly point out what I am referring to.

Ideally I would like to see from the bug theorists a 3D model with computer generated reflections that could show how the bug's anatomy could produce that image. What I've seen so far, unless I've missed something, is that they produce an image with a bluish blur next to a yellowish blur, then think they have a proof. I voiced my approach to a disproof in the last few pages, but in those pages I didn't see a response.

And I definitely would like to see a picture of the flying ants being discussed. I am particularly interested in their leg and wing positions in flight, as well as coloring.

[Deckham]

Thanks for the brief lessons on photomechanics (?) - I appreciate it. The curtain analogy was particularly stimulating.

Here's another one for you, if you would:

Mr Pryde has set up his camera to photograph what is essentially a landscape scene timelapse.

So he either decides to use the flash option to illuminate passing insects (couldn't resist, sorry), or he neglects to 'switch' it off. Can't think of any other reason it would be left on. There is no designed foreground focus to illuminate.

Given it was in error, would someone have an approximate time a digital camera of that ilk erroneosly configured to 'flash' every 15sec would be able to continue to do so - with a standard battery?

If this was not an error (I now understand why flash would be used to light up your friend's smile) what was the purpose of it?

Perhaps this could be a question posed to the photographer.

By the way - I totally realise that this has become more of a psychological discussion more than anything else. I enjoy it even if it is only hypothetical. I admit that should the pic not be a hoax, the bug theory is for me.

However - I am not even nearly convinced at this point that it is, in fact, a naturally occuring phenomenon....

[hazeii3]

Thanks for the brief lessons on photomechanics (?) - I appreciate it. The curtain analogy was particularly stimulating.

Here's another one for you, if you would:

Mr Pryde has set up his camera to photograph what is essentially a landscape scene timelapse.

So he either decides to use the flash option to illuminate passing insects (couldn't resist, sorry), or he neglects to 'switch' it off. Can't think of any other reason it would be left on. There is no designed foreground focus to illuminate.

Given it was in error, would someone have an approximate time a digital camera of that ilk erroneosly configured to 'flash' every 15sec would be able to continue to do so - with a standard battery?

If this was not an error (I now understand why flash would be used to light up your friend's smile) what was the purpose of it?

Perhaps this could be a question posed to the photographer.

By the way - I totally realise that this has become more of a psychological discussion more than anything else. I enjoy it even if it is only hypothetical. I admit that should the pic not be a hoax, the bug theory is for me.

However - I am not even nearly convinced at this point that it is, in fact, a naturally occuring phenomenon....

I believe he posted an explanation earlier, which was something like the flash was on because it was necessary to get the exposure he wanted.

Ref. the battery life, in my experiments I took probably about 100 shots, all with flash, without having to recharge.

However, I'd certainly like to see the rest of the photographs he took, to see if there are similiar trails in any of them.

[Guest]

I suggest that the people who still support alternative theories (UFOs, contrails, meteorites or Tina Turner) need to focus on disproving the bug theory to make headway.
Boldra

Quite Right.

My problem is the 90 degree bank while still in straight flight. That's not a bee. While a bee would be making the aforementioned proverbial beeline home [Dusk is about and the bats are just coming out]. Contrary to testimony here, the bee is not usually wind deterred. Just as an aircraft does, a bee has the power to yaw into the wind and maintain the flight path he intends. Otherwise he doesn't fly.

And a fly, common around Darwin, especially in Summer, doesn't usually exhibit straight flight. It could be another insect but not the then prevelent flying ant which is extremely suseptible to wind.

But it's the illumination on two counts which bothers me most with this theory.
1] Flash time 10usec-1msec. I go with other posts in that the flash probably maxed out. Being set for end curtain the camera would have been attempting a balanced exposure on a non-existant foreground.

Now that 1/1000sec flash is 1/50 of the 1/20sec exposure. I would expect the flash to expose the insect for 1/50 of its flight in the frame.

I'm failing to see that. Plus the photo is much less distinguishable [focal plane:infinity/lens:f5.6 vs ?] than those posted by hazeii which were taken five times slower @ 1/4 sec. Additionally there is no trail remotely resembling straight nor long enough. At sunset, away from the hive you might catch a bee on a straight, 30 mph dive for home. But still, that is only 2.2 feet in the 1/20 sec exposure. Will that explain the trail? If it's to be an insect, I would say it has to be other than a bee.

2] The abdominal exposure. Having read all the posts, the most likely explaination of the flare in the 'insect scenario' were that it was a flash exposure of either a full pollen load or the surface of the insect's abdomen. If so, I don't see how that flare can be patially obscured by an object in the background [lamppost]. I understand the lamppost would appear in its entirety but it would normally be washed where overlayed by the insect's flash exposure; not be obscuring that exposure occurring much closer to the camera.

Those are my objections to the insect theory. With each of those allayed I too would have to join based upon 'bicycle boy' and his outstanding demonstration.

Why aren't more of the proponent's here doing !/20sec f5.6 exposures focused to infinity around their own yards? Yeah, so winter throws a wrench into the old insect watching evenings.

I think it is premature to declare it an insect without tighter demonstration and a fast flying insect prospect which can roll [granted, a curtain cocking, flash charging, etc may have startled the insect into the roll. But without wobble in its flight path?] during straight flight with nary a wobble. Though it may constantly perform defensive aerobatics while on the day's final flight to refuge. That leads to the slight wobble seen in the trail?

I for one would like to see a few more demonstrations before declaring this a bug infested event. The major ground work was done by the guy with the disassembled bicycle who can't be trusted warming KFC

And I thank you, sir, for a very good demonstration. 8)

[MrMoon]

I've ony seen the picture now and I didn't read all the replys but I did read the conclusions and I think that there's one thing that wasn't mentioned.

When looking at the picture 1:1 I noticed this: the trail starts "from the sky" and it goes down intercepting a white cloud at about the same level (in the photo) as the tree in the left. If you look at it careful you will see that the trail doesn't pass in front of it but it loops around it.

I don't have an explanation to the trail but with my experience with cameras and pictures I know that the flash shouldn't be on because there are no bright spots in the nearer plants.
Also I know that bugs are just visible in this kind of picture at a small gap of possible distances. If they're too close they're too out of focus (like when you picture something through a wire fence. It does not appear in the picture. And the diametre of the wire is usually bigger than the diametre of an insect). If they're too far they just too small to be seen.

The size of the burst and trail and the straight line that it makes in the picture just don't fit with a bug passing by.

[Can't use my Bad Buoy]

In my last post a half hour ago I posted my concerns with the insect theory but neglected to post to which photo of hazeii's extensive body of work I was referring.

[gibsong]

Is it possible that is was a model rocket being lauched from the light post? The rocket engine ignition and power build up to lift off would cause the bright light and apparent smoke. The contrail is the rocket pasing through the view of the photo from right to left.

[MrMoon]

in 0.05seconds ?

[Sombrero4594]

I would comment upon the insect “consensus” about the streak. As a lifelong student of entomology and the natural sciences, I would offer two problems, both fatal, with the insect idea.

1) As others have pointed out, the shutter speed rules out any insect leaving a trail that long (or any other creature from Earth) on a picture taken at that speed.

2)The trail is perfectly (within the limits of my digital copy) straight. No creature on Earth can do that—not even a peregrine falcon in full stoop at 198mph.

Also, any rocket-like device would have left the cloud associated with liftofff fully visible (unless it was the landing end of trajectory). Since the photographer said there was no trace of it in the next shot, that sort of rules out the landing end of a trajectory.

[Guest]

Then could someone please post a picture of a Darwin white ant.

Not being from Darwin or Australia for that matter, this was what a couple of google searches on "White Ants" Australia came up with..so take this for what it is...a SWAG.

This appears to be an illustration rather than a photograph..
http://www.kendall-bioresearch.co.uk/ISOP3.GIF

I also found some photographs of the "workers" which may be indicative of the overall color..
http://www.termite.com/images/termites.jpg

[Can't use my......]

1] How long is it?
Can you imagine it, a very small gnat close to your eye and out of focus which travels but a couple inches?

2] Somewhere between the meteorific distance you first imagined and the least distance just outlined lies the true path of the event. Aren't there insects which could cover the shortest of these?

[Bob Peterson]

Ha Ha---we've all been had---this is like an Alfred Hitchcock movie (a bad one) where we are all unwitting subjects of an evil scientist---thius is now obviously a psychological forum, not an astronomy based forum, which was the evil intent to begin with---as we degenerate into our caveman (any cavewomen out there?), proving that "the Lord of the Flies" was an accurate portrayal of the human condition, it is time to search out and find this Pryde fellow and string him up---let's hire a dectective and get to it!!
PS, I still don't think it's a bug but just a co-incidental record of a bulb exploding and some unrelated event on the land mass--pppssfffttthhhttt!!!

Right Hawaiianmike.
There's Gay Pride Day
so I'm proposing we have
Get Pryde Day
And that day is today

P.S. I'm not gay

[Guest]

1) As others have pointed out, the shutter speed rules out any insect leaving a trail that long (or any other creature from Earth) on a picture taken at that speed.

This has been discussed in depth, assuming the bug was reasonably close to the camera,( and would have to be in order for it to have been lit up by the flash ) it doesn't have to be flying at anything very fast at all

EXAMPLE:

The streak is about 1175 pixels long and about 14 pixels wide. The relative darkening of the sky is roughly 2%, which could be caused by a dark object 2% the length of the total streak, or about 24 pixels. Using those numbers, this streak could have been made by a small bug about 0.1" wide by at least 0.15" long, making an 8" streak during the 1/20th second by flying a mere 9 MPH. Or, it could be a bee 0.3" wide by about 0.5" long making a 24" streak by flying 27 MPH, which bees can easily do. Those are just examples -- it could be an even smaller bug flying slower or a larger bug flying faster -- but the point is, this streak is very much within the range expected for a motion-blurred bug shot at 1/20th second exposure.

2)The trail is perfectly (within the limits of my digital copy) straight. No creature on Earth can do that—not even a peregrine falcon in full stoop at 198mph.

Can we PLEASE kill the idea that the line is straight.. It's NOT.. there must be at least a couple of dozen posts showing the line does deviate. So if we say well it's PRETTY straight, OK, but 1/20th of a second is NOT a long time for anything to fly/move/descend/take off/blow in the wind/whatever in a "PRETTY" straight fashion..especially if it's only an inch away from the camera.


TO THE FORUM ADMINS: Would it be possible to post a link on the top of each page to the excellent discussion: http://www.cloudbait.com/science/darwin.html that was referenced in the APOD solution link?

[Guest]

I would have to say that Chris Peterson's insect theory seems pretty weak.

While his discredit of the other theories seems logical and well surmised, he does not apply the same rules to the insect theory.
He claims that it is not a missile because "it seems very unlikely that military air-to-surface weapons would be in use in an area like this". He embraces "likelihood" as an discredit to this theory, but ignores likelihood when considering the straightness in flight path for the insect.

One point I might add is, if the image captures the insect silhouette, and a light reflection from the insects wings, this would tell us that the insect is still in the photo at the time of exposure. Where is the insect? If Mr Peterson's analysis is detailed enough to dislocate the flash from the light pole by 8 to 10 pixels, Say that the trail is 2% less bright than the surrounding sky, surely the there would be some evidence of the bug itself.

I agree with his conclusions on what the image does NOT depict, but he contradicts himself when he so readily accepts the insect explanation on the same grounds that he uses to dimiss the others. I am of the opinion that none of the theories that Chris Peterson sites - including the insect - are the root cause for the image anaomaly.

[Can't use my Bad Buoy]

TO THE FORUM ADMINS: Would it be possible to post a link on the top of each page to the excellent discussion: http://www.cloudbait.com/science/darwin.html that was referenced in the APOD solution link?

I would not favor that.

They have made a determination based upon very incomplete investigation and simulation.
As far as they went they're fine. I just don't yet agree that the discussion is closed; hence I feel their words would tend to prejudice the outcome of any discussion entered here.

[Ernst Lippe]

My problem is the 90 degree bank while still in straight flight.

There are several possible explanations:
First of all it is likely that the insect is not flying completely parallel to
the camera, so that it will have its head or its abdomen oriented somewhat
towards the camera. One explanation is that insect was experiencing
a side wind, and that it rolled to counter the wind. Another
explanation is that the insect was captured with its wings in a strange
looking position (after all we normally can't see how insect wings normally
move because they are to fast for our eyes). Yet another explanation
is that the wings will only reflect the flash when they are in one
specific position, because they act like mirrors.

I agree that the position of the wings looks somewhat odd, but
on the other hand, most of us don't really know how a flying insect
looks in a flash photo.

That's not a bee. While a bee would be making the aforementioned proverbial beeline home [Dusk is about and the bats are just coming out]. Contrary to testimony here, the bee is not usually wind deterred. Just as an aircraft does, a bee has the power to yaw into the wind and maintain the flight path he intends. Otherwise he doesn't fly.

But even a fast flying bee will be influenced by the wind.

And a fly, common around Darwin, especially in Summer, doesn't usually exhibit straight flight.

Virtually all insects can and do fly in straight lines. But most people will
only spot insects when they stay long enough in the same place and they will
probably miss insects that are flying longer distances. Actually, the only
time when there are not flying in straight lines is when they are searching
(e.g. for food), and there is probably nothing interesting for an insect near
the camera because it is located at the coast.

Anyhow the actual length of the streak is at most a few meters (and probably
much less).

It could be another insect but not the then prevelent flying ant which is extremely suseptible to wind.

But it was quite windy when the picture was taken, just look at the trees.
The wind speed during the day was 9-29 km/hr which means that
in 1/20 s the wind movement alone was 12.5-40 cm.

But it's the illumination on two counts which bothers me most with this theory.
1] Flash time 10usec-1msec. I go with other posts in that the flash probably maxed out. Being set for end curtain the camera would have been attempting a balanced exposure on a non-existant foreground.

We don't know if the flash occured at the beginning or at the end,
this camera can do both.

Now that 1/1000sec flash is 1/50 of the 1/20sec exposure. I would expect the flash to expose the insect for 1/50 of its flight in the frame.

We still don't know how fast the flash was. If it was indeed 1/1000 s
the movement blur should have been greater than it is in the picture
(assuming that the insect moves at a uniform speed). But even a
flash of 1/2000 s is short enough to be consistent with the observed
motion blur. Now the minimum exposure time of this camera is
1/2000 s, and it seems reasonable to assume that the flash should
also be able to handle this.

I'm failing to see that. Plus the photo is much less distinguishable [focal plane:infinity/lens:f5.6 vs ?] than those posted by hazeii which were taken five times slower @ 1/4 sec. Additionally there is no trail remotely resembling straight nor long enough. At sunset, away from the hive you might catch a bee on a straight, 30 mph dive for home. But still, that is only 2.2 feet in the 1/20 sec exposure. Will that explain the trail? If it's to be an insect, I would say it has to be other than a bee.

We don't know the size of the insect, and it is also very difficult to
say how far removed it was from the camera. But even a very short
path (say 0.1 m) would cause the streak if the insect is close enough.

2] The abdominal exposure. Having read all the posts, the most likely explaination of the flare in the 'insect scenario' were that it was a flash exposure of either a full pollen load or the surface of the insect's abdomen. If so, I don't see how that flare can be patially obscured by an object in the background [lamppost]. I understand the lamppost would appear in its entirety but it would normally be washed where overlayed by the insect's flash exposure; not be obscuring that exposure occurring much closer to the camera.

We don't know the orientation of the insect, because that depends
on the synchonization of the flash. Actually, I think that it is
most likely that we are looking at the reflection of the head.
Why do you believe that the lamp post obscures the flash?
If you look at the difference images one of the striking features
is that you can't see any static features from the background.
In particular the lamppost is completely absent. This is a
very strong argument against all explosion theories, if it
was a bright light flash in the distance, the difference image
should show shadows of nearby objects, but there are none.