The yellow Sun paradox Hereis a brand new observational paradox I stumbled across today. The Sun looks overwhelmingly yellow to my eyes (and reddish at dawn and dusk).
But science disagrees, and claims instead the Sun is “…the definition of whiteness.” On earth’s surface, apparently not enough blue light is scattered out to justify the Sun looking yellow. This is a first-class paradox that unfortunately has not received the publicity it deserves.
A cite from the source:
Assuming the source is factual, can sciencey folks here explain in layperson terms what’s going on?
“the amount of blue light scattered out is far too small to have a noticeable effect on the sun’s color.”
Evidently wrong. Source is not factual.
“This is a first-class paradox”
[Inigo Montoya]You keep using that word.[/Inigo Montoya]
Trouble is, that the definitions are all over the place.
The CIE definition of a 6500K illuminant is not for “pure white” they are for “things illuminated in daylight” white. There are other standard illuminants.
If you put an object out in the sun, it is illuminated by both the direct sunlight, and by the blue sky. The 6500 illuminant takes this into account. If you use some sort of masking device to only illuminate with the direct sun, and shadow out the blue sky, the apparent colour of the light changes, and becomes noticeably more yellow.
None of these standard illuminants are what you would consider mathematically pure white - which would require equal intensity of light at each wavelength across the visible spectrum. That would be - to our eyes - much bluer.
Where it all gets difficult is that our eye/brain compensates as much as it can anyway. We don’t notice how seriously yellow incandescent light bulbs are unless we compare them with a different source.
The sun, and incandescent light sources are black body radiators, and so have a spectrum that is defined by their temperature. The Sun is more complex as we must add in the effects of the atmosphere (both ours and the sun’s). This shifts the balance a bit, and also results in a bit more roughness. The CIE defines a range of illuminants that take into account the sun’s angle in the sky, so you have an average blue sky, midday sun, and others. What science does not do is define sunlight as pure white.
Francis Vaughan is correct, the daylight whiteness is the sum of the Sun and the sky colours, yellow plus blue adds up to white…ish.
As a matter of fact that’s what most LEDs for lighting do, well, same same but different; they use a blue LED (instead of a yellow Sun) to illuminate a yellow phosphor and the result produces “natural” light tones.
Edit to add, you can clearly see the yellow phosphor inside a white LED.
Why would “pure white” be the case where all wavelengths are of equal intensity? Thatr’s an absurd definition.
What we define as white is approximately the light from the sun. You can say it’s “all over the place”, but it isn’t, really. If you look at the CIE chromaticity coordinates of most common definitions of white – mainly averaged skylight (with sun) or noonday sun, they’re all in pretty much the same place, and their spectra are very similar. As for atmospheric absorption, itr’s certainly present, but doesn’t much alter the overall spectrum or the CIE coordinates.
Here’s the CIE diagram:
Here’s a listing of the Standard Illuminants. You can see how they cluster in the same space:
I devote a chapter in my book to The Yellow Sun Paradox, which I believe the OP is quoting. The explanatioin is given there.
Because that is the conventional definition of white noise. If you talk to an engineer about spectrally white energy in any context other than the visual spectrum this is the definition they would assume.
When I said “all over the place” I meant that the definitions the OP was quoting were all over the place. Not that the CIE definitions were - they make perfect sense. If you look very closely you can see some adsorption effects that take them away from a pure black body. The difference between midday sun and afternoon sun is entirely due to atmosphere.
However the CIE definitions only apply to humans using sunlight on the Earth. White is different on Mars. There might be a CIE standard for it, maybe one day there will be.
A “pure” definition of white could be equal intensity in every wavelength band, or it could be equal intensity in every frequency band. The two are not equivalent.
And for what it’s worth, astronomers don’t define “white” in terms of the Sun (a G2V star), but in terms of Vega (A0V). Color is defined in terms of difference of magnitude through different filters, and magnitude through any given filter is calibrated such that Vega is magnitude 0 in every filter. Relative to Vega, Sol is in fact a bit yellowish.
All that said, though, the primary reason most people think of the Sun as yellow is that it’s usually seen against a blue background, and by comparison with the background, it’s more yellow.
Another reason I think is that people don’t usually look at the Sun when it’s high in the sky because it’s too bright to look directly at. Most people would recall seeing the Sun in late afternoon or in the morning, when it’s lower in the sky and more blue light has been scattered out due to the longer angled trip through the atmosphere.
So, is this a situation of people meaning two different things by the same word and not seeing the distinction?
CIE is useful for color-related issues on earth. It’s a “good enough for 99.9% of problems” definition but it isn’t quite true, just like accounts of Newtonian physics taught outside physics departments.
Hence, it’s not really a paradox, just an instance of the term “white light” being used in two different ways and those ways being mixed up .This gives rise to apparent contradictions which are just semantic issues.
If one started using quantum mechanics-related definitions in a Newtonian physics context, it would also produce apparent contradictions, wouldn’t it?
I think it’s a combination of only actually looking at the Sun when it’s near the horizon and surrounded by an reddish-yellowish sky, and subconsciously associating it with other hot things like fire, which does appear yellow.
Another instance where common experience and science are in apparent contradiction which it would be inaccurate to call a paradox. We associate hot things with being red but things that are even hotter will either move toward blue/violet or white.
Actually, since this is a good place to ask the question:
Why does a red flame* that becomes hotter glow blue but a piece of iron that becomes hotter glow white?
I get that blue is higher frequency/energy than red but why is it blue rather than white-ish? I’d expect the frequencies of a flame to combine to form a white-ish color.
And why is it different for a piece of iron that is heated?
And what happens to violet? We hardly ever see a violet flame.
*A typical one seen on earth like a fire or a welding gun.
Iron is close to being a blackbody. A flame, at least one produced by burning most familiar fuels, is not. You’ll get spectral lines in there, some of which are apparently bluish.
From my physics-ignorance POV, it seems that fluids tend to differ significantly from the ideal blackbody whereas solids tend to be closer. Then again the sun is close to being a blackbody.
So, what influences how close something is to being a blackbody?
And do you know about the missing violet light on flames? It’s present in rainbows so I wouldn’t expect it’s a question of amplitude.
I would make the argument that the sun is, by definition, pure white because color is subjective based upon the observer. As humans have evolved to see visible light, specifically the wavelengths our sun gives out then, from a human perspective, how can it be anything else?
Now, obviously the Rayleigh scattering has some effect, as the blue scattered out isn’t in the sun, but I’m unsure how much of that affects it. I think more of it is that it appears yellow because it’s against a background of blue. Consider if you take two circles of the exact same shade of gray and put one on a white background and one on a black background. The former will look darker relative to the white background than the latter. Our brains are great at perceiving relative color, but not so much at seeing absolute color. This is why when you wear sunglasses, after a short period, we start to see colors as they are relative to the tinting of the lens. So if we’re wearing blue tinted lenses, something that is absolutely green will appear yellow to our brains.
In this sense, I think that’s probably the dominating reason that the sun appears yellow. White on a blue background will appear yellowish. And both the Rayleigh scattering and the relative coloring effects disappear when observing the sun from space.
Any individual gas will be a poor blackbody, because its spectrum will consist entirely of lines. If you have a very large, complicated molecule, you can end up with a very large number of lines, enough that they start to blur together… but large, complicated molecules tend to have high boiling points, so they’re usually not encountered as gases. If you have a very large number of different gases mixed together, you can again get enough lines to blend together, but there aren’t enough different gases in a flame. And if you have a high enough density and/or thickness of gas, eventually even the off-line portions of the spectrum add up, but flames are much too thin for that.
The CIE space is generated based on the RGB response values of the human retina for applied use in human-factors based color analysis. I could think of many experiments or applications where one may want a “pure white” standard that is independent of human physiology.
e.g., a standard for measuring the vision of animals, or machine vision of robots or of sensors. Or of anything with a “visual” spectrum that sees ranges beyond human vision boundaries.
Otherwise, when artificially intelligent robots arise, it may be discriminatory to not have an 800nm Red crayon for them to color with
