Saturday, May 09, 2009

What's wrong with this picture?

14 comments:

Len said...

Okay...

The secondary rainbow has the order of colours reversed.

The centre of the arcs is above the horizon, which would meant that the Sun would have to be below the horizon -- no rainbow!

It's hard to tell from the picture, but the direction of illumination would have to be head on, the buildings look as if they are lit from the side.

Dean Baird said...

Nicely played, Len! You nailed the biggies square-on. There's a wee detail you didn't mention.

Nevertheless, you advance to Round 2: Why is the image a fake? (You've already established that the image is a fake.) What possessed the artist to do this?

Len said...

Is the wee detail the spacing of the arcs? The secondary bow looks too close to the primary.

There wasn't much effort put into the "buildings" at the base of the arcs, either...

As to the motivation of the artist, you got me there -- I don't recognise the skyline. Could it be that we're looking at the city from the North, and thus would never see a real rainbow (since the Sun could never be at my back)?

Dean Baird said...

Correct again!

The path to solving the motivation lies in finding the original image. That will reveal the identity of the city as well.

Len said...

Okay, it's Baltimore... Still thinking...

This is so much more fun than marking!

Len said...

...Aaand we're looking North. Okay, so the Sun could be behind me, but it would have to be at high noon, putting any potential rainbows much further down in the sky.

Baltimore is ~39°N, so the best case scenario would be a noon on December 21 or so, when the Sun is about 25° up from the horizon, putting the centre of the rainbow arcs 25° below the horizon. The primary arc is 42° from the centre, so any rainbow would just be skimming over the tops of the buildings.

Dean Baird said...

The owner/artist/originator reveals his purpose on his blog. It's innocent enough, but the image reveals a lack of understanding of rainbows that I doubt makes him unique.

Len said...

This is feeding my OCD.

After reading the original blogger's post here, I tried simulating the view with Starry Night -- he would have had to be looking approx. NW, and the low angle of the Sun would have produced pretty nice arcs, just not as big as he made them.

Funny that he states that the rendition is "fairly accurate"! Great blog, though -- I spent a lot of time going through old posts, admiring the pictures. He has a great eye.

Dean Baird said...

He remembers the double rainbow, but is not burdened (as we are) with knowledge of the technical details. He made no attempt to deceive. He just didn't get things quite right.

More importantly, where's *your* recreation?

Len said...

You mean the Starry Night one? It was nothing special, I just wanted to see the nature of the arcs. If you are familiar with the program, I created two FOV circles, one for each type of rainbow, centred at an altitude of -5° (assuming the Sun wasn't too high in the sky).

As expected, the arcs are spaced farther apart and make more of an angle with the horizon. I'll send you a picture under separate cover (check your mail!).

Stevie Ray said...

Not yet mentioned - the color bands are too uniform in thickness - ignoring the greater scattering of the violets and blues (and Indigos- sorry Deano) which gives the glorious wider yellow and red bands seen in real rainbows.
Also, I would think the distance from the rainbow would have to be a lot farther away and at an altitude much greater than sea level to yield such protractor-shaped arcs.

Alp said...

What is wrong: this spectrum would only be visible under circumstances not yet recorded (e.g., if a hostile alien power projected a death beam onto a city)

Dean Baird said...

@Alp: You got me there.

Converter said...

i think the size of the rain bow is unusual.