Another physics question: sunsets and colors

[Alex Franke]

As has been clearly demonstrated in the past, this is my go-to forum for all things woodworking and physics. So here's what's puzzling me now... Can anyone shed some light on this? (har har)

I get that the sky is blue because the purples and blues bounce around a lot more and blue is the sweet spot. I also get that the sun often goes from white to yellow to orange to red as it sets because the atmosphere gets more time to filter colors out. So what happened to green?

I understand that green flashes are more about refraction than scattering, so that's not really what I'm talking about. But if purples and blues are already scattered, and the oranges and yellows are scattered more as the sun sets, then shouldn't we at some point either see a blue-green sky or a yellow-green sun?

[Kristofor]

I get that the sky is blue because the purples and blues bounce around a lot more and blue is the sweet spot. I also get that the sun often goes from white to yellow to orange to red as it sets because the atmosphere gets more time to filter colors out. So what happened to green?

The sky appears blue during the day because short wavelength light is being absorbed and re-emitted in random directions. That is light that wouldn't have been visible to you if it had kept on it's original course. The actual color of the sun if you look up when it's overhead isn't blue, instead it's yellowish (with some portion of the blue wavelengths scattered away).

When the sun is near the horizon it looks more red because now in addition to the the blue component being scattered out more of the intermediate wavelengths (like green) are also being scattered so the light passing through to you looks orange/red. If you look to the side away from the sunset you can probably find a light-bluish area (mix of green and blue) scattered light.

It's easy to see on any app with an RGB color selector... Start filtering out the blue channel and if that's all that's blocked you'd have a nice yellow. Then start filtering out the green channel and you shift into the oranges and then red.

<note: edited for clarity>

[LCHIEN]

along the same lines, when Japan melts down, will we see a glow in the western sky, the eastern sky, or will the entire sky glow???

[Bruce Cohen]

along the same lines, when Japan melts down, will we see a glow in the western sky, the eastern sky, or will the entire sky glow???

Hey Loring,

I not sure if that's funny?

Now, if Viet Nam blows up, would anyone care???

Bruce

[MilDoc]

A green sky does occasionally occur before severe thunderstorms and occasionally in winter during a snow storm. Only occurs AFAIK if the sky overhead is cloudy for 20 miles or so around the observer. That again has to do with light scattering. In fact, the light from 20 miles away that reaches you, even in clear sky, is more greenish, but you don't perceive it because the predominant light is still blue.

[Alex Franke]

It's easy to see on any app with an RGB color selector... Start filtering out the blue channel and if that's all that's blocked you'd have a nice yellow. Then start filtering out the green channel and you shift into the oranges and then red.

Ah -- Exactly!!! We'd only see green if the oranges and reds were filtered out first because it's mixed color and not separated color. Thanks for this and the perfect way to demonstrate it!

[cabinetman]

Now, if Viet Nam blows up, would anyone care???

Bruce

I might. I could have some children there.

.

[Bruce Cohen]

I might. I could have some children there.

.

Hey Cab,

Does your wife know of this???

And I'm not going anywhere near that comment. I enjoy living.

Bruce

[MilDoc]

Ah -- Exactly!!! We'd only see green if the oranges and reds were filtered out first because it's mixed color and not separated color. Thanks for this and the perfect way to demonstrate it!

Uh, not exactly. Green is indeed a separate spectral color, as seen in a rainbow. Remember the "Roy G Biv" device for recalling a rainbow's colors?

Had to look it up, but the perception of green is evoked by light having a spectrum dominated by energy with a wavelength of roughly 520–570 nanometers.

So the occasional green colored sky seen from a distance that I mentioned before is because all other wavelengths are filtered out, not just the ones you mentioned. In the real world, green is a real color, not mixed.

In the case of paints, etc, green is a mixture in the subtractive color system - a mixture of yellow and blue. But in the additive color system green is indeed a primary color, along with red and blue.

You can't compare these real and mixed systems to a computer monitor, which uses an "optical mixing" system. It's not additive because the colors do not overlap. Each of the red, blue, and green pixels are separate. Because the pixels are so small and close together our eyes blend them together, having a similar, but definitely not the same, effect as additive light.

[Alex Franke]

Uh, not exactly. Green is indeed a separate spectral color, as seen in a rainbow. Remember the "Roy G Biv" device for recalling a rainbow's colors?

What I mean is that if you take full spectrum sunlight (what I called "mixed color") and start to attenuate from the upper end of the spectrum (as these colors scatter more), the resulting color goes from white to yellow, then straight into orange and then reds. And, the only way you'd see green is if the oranges and reds were filtered out before the greens (but also after the blues, which I didn't make clear). I imagine this is what's happening in the green sky phenomenon that you describe.

I think we're on the same page, but thanks for clarifying.

This forum rocks.