12/30/2004: Green flash at sunset

[Undertoad]

Credit and copyright Tony Cook and used with his permission. This was the Earth Sci pic of the Day a week ago. Taken on Tenerife in the Canary Islands, it's a great photo and also shows an interesting phenomenon - at the top of the sun, a green flash appears. This green is not on the sun itself, but a product of how the atmosphere affects the light. From the ESPoD explanation:

Green flashes are created by variations in refraction near the horizon. The refractive layer causes sunlight to be weakly dispersed into the constituent colours of red, yellow, green, blue and violet. Violet and blue light are normally scattered in the Earth's atmosphere, with the result that the last portion of the dispersed light to be observed as the Sun sets is green.

[Guess]

very cool

[glatt]

pretty

[BigV]

More pics of green flash.

http://www.polarimage.fi/sun2/ataulu2b.jpg

I must say that this fella's website has the most incredible collection of sky / weather / water pictures. It was a BIG time sink for me when I came across it.

Enjoy!

[xoxoxoBruce]

On Golden Pond.

[capnhowdy]

I understand the color bit, but I wonder why the edges are so rough. Almost looks like an overexposure. Very beautiful to say the least.

[blase]

Didn't we see a similar image a while back, I seem to remember something similar...

[linknoid]

I actually liked the last one that was used for IotD a bit better:

http://www.cellar.org/showthread.php?t=2395

This one is still pretty cool.

[cweekly]

capnhowdy: I understand the color bit, but I wonder why the edges are so rough.

that roughness -- specifically the striations -- are caused by different levels of light absorption/refraction at different heights in the atmosphere. by the time you see those, the sun is actually below the horizon(!) but you see the light from it bent/redirected through the atmosphere at slightly different levels according to altitude. the density of matter in the upper levels of the atmosphere is less than in lower zones: also, rather than being a steady gradient, the atmospheric zones are delimited at fairly distinct altitudes. thus the stripes in the setting sun are a distorted representation of these atmospheric pressure levels.

another note bene: this is also related to the reason sunsets are so pretty: the red and orange hues are caused by the lengthening of ths sun's rays as the sunlight travels further, it's in effect being stretched, and longer wavelengths are more red, while shorter ones are more blue. (this "redshifting" is an example of the doppler effect, which also explains why car engines sound higher pitched as they race towards you then suddenly lower as they recede)

that's the gist anyway (from memory),someone else may post a more technically precise answer

happy new year
-cweekly

ps I almost never post but I've lurked here for years. hi!

[xoxoxoBruce]

someone else may post a more technically precise answer

I hope not. Your explanation was clear enough for even me to understand. Thank you and welcome!

[linknoid]

Originally Posted by cweekly

another note bene: this is also related to the reason sunsets are so pretty: the red and orange hues are caused by the lengthening of ths sun's rays as the sunlight travels further, it's in effect being stretched, and longer wavelengths are more red, while shorter ones are more blue. (this "redshifting" is an example of the doppler effect, which also explains why car engines sound higher pitched as they race towards you then suddenly lower as they recede)

that's the gist anyway (from memory),someone else may post a more technically precise answer

Sorry xoxoxoBruce, but his explanation of the colors at sunset is completely off. The reds and oranges you see in the sunset have nothing whatsoever to do with red shifting or stretching of wavelengths. For that to happen we'd have to be moving a significant percentage of the speed of light away from the sun.

But the real answer to why you get the reds and oranges at sunset is the same reason the sky is blue. Go back and read the quote that Undertoad posted with the original picture again:

Green flashes are created by variations in refraction near the horizon. The refractive layer causes sunlight to be weakly dispersed into the constituent colours of red, yellow, green, blue and violet. Violet and blue light are normally scattered in the Earth's atmosphere, with the result that the last portion of the dispersed light to be observed as the Sun sets is green.

See, when it's scattered, all the blue end of the spectrum gets scattered, and the red end isn't. Near sunset the light has to go through so much atmosphere that most of the blue end of the spectrum has scattered out (which makes the sky look blue), and all that's left is the reds and oranges (and some green, which is what the IotD is about).

(I was going to explain why blue scatters and red doesn't, but I figured the more I write, the less people are going to take the time to read it)

[capnhowdy]

Thanks for the lessons, Guys. It does make sense, even to me. It seems like the sunrise & sunset would look almost alike, but sunrises are always loaded with yellows, golds, and oranges while sunsets are filled with reds and purples. A great gift nature has given us. I hope to enjoy thousands more at each end of the day. Imagine trying to explain how a beautiful sunset looks to a blind person who has never seen one. Now that would be a challenge.....

[Happy Monkey]

I wonder if the temperature of the air causes the sunrise/sunset color difference. In the morning, the sunlight is advancing through cold night air, and in the evening it is retreating through warmer daytime air.

[xoxoxoBruce]

Originally Posted by linknoid

See, when it's scattered, all the blue end of the spectrum gets scattered, and the red end isn't. Near sunset the light has to go through so much atmosphere that most of the blue end of the spectrum has scattered out (which makes the sky look blue), and all that's left is the reds and oranges (and some green, which is what the IotD is about).

If I remember art classes correctly green is yellow & blue and violet is blue & red.
So how can we get green if the blue component in already dispersed?
And how can the violet get dispersed if half of it is red yet the red doesn't?

[linknoid]

Originally Posted by xoxoxoBruce

If I remember art classes correctly green is yellow & blue and violet is blue & red.
So how can we get green if the blue component in already dispersed?
And how can the violet get dispersed if half of it is red yet the red doesn't?

The short answer is that there are two ways of mixing colors. One is by mixing what is absorbed, subtractive mixing, which happens when you mix paints, and the other when you mix colors of light, additive mixing. The mix of colors coming from the sun is the dividing up of white light into the colors of the rainbow, not of mixing pigments.

Our eyes only see 3 colors: red, green, and blue. Using those 3 colors you can simulate most any color, which is why TVs and monitors only use red, green, and blue (RGB). There are 3 types of cone cells in your eyes, one for each of those colors, so every every color you see is based on the proportions that each of those cells are activated.

When you see yellow, it means that both red and and green receptors in your eyes have been activated. Yellow light activates both red and green receptors (since its wavelength falls in between red and green), but if you mix red and green light, it has the same effect of activating red and green receptors. And when you see either yellow light or red and green lights mixed, you can't tell the difference, your brain just interprets it as yellow.


On the absorbtion side of things, the primary colors are cyan (anti-red), magenta (anti-green), and yellow (anti-blue). That's why color printing uses Cyan, Magenta, Yellow and blacK (CMYK) instead of red, green, and blue to reproduce all the colors. Cyan absorbs Red light (and reflects green and blue), magenta absorbs green, and yellow absorbs blue, so they're the exact opposite of the 3 primary colors of light. If you take a picture on film, the negative will turn reds into cyans, magentas into greens, and blues into yellows.

But when you mix paints, you aren't mixing pure red, pure yellow, and pure blue. It's hard to tell what light frequencies are being absorbed without using a spectroscope, so the exact colors you see when you mix two colors of paint all depends on which frequencies (colors) of light are being reflected and which cells in your eyes are being triggered by each frequency, and how strongly.

It's not a simple subject at all, and I'm not going to attempt to explain any more here. Hopefully everything I've written here is understandable. If you want to know more, there's an excellent (and much more technical) explanation here:

http://hypertextbook.com/physics/waves/color/