Why Does Our Sun Appear Yellow Instead of White?

[rogerk8]

Hi!

I wonder why our sun, being a star and all, is yellow and not white?

What is it in our atmosphere or jonosphere that makes it yellow?

Roger

 

[mathman]

It appears yellow mainly because the blue part of the spectrum gets scattered more than the rest by our atmosphere. That's why the sky is blue.

 

[Nugatory]

Google for "Why is the sun yellow?"
The first two or three hits have pretty decent answers.

 

[cjl]

Honestly, it looks pretty white to me.

 

[rogerk8]

Two things:

1) Scattering, what is that?
2) Reading that link suggests that the amount of yellow is due to the distance sunlight has to travel through the atmosphere.

Scattering sounds to me like something being reflected "away" from the observer. My thinking is that if scattering of blue light happens then the sky should be everything but blue. Please correct me while I'm wrong :)

Then take the only star I know, the Alpha Centauri (AC).

The light from AC certainly appears white to me.

Whiter than our sun at noon, even.

The only difference is that I/we cannot see it unless it is dark.

Correct me if I'm wrong, but isn't the distance for AC's rays of light through our atmosphere equally long as it is for our sun's rays of light?

Yet Alpha Centauri looks white while our sun looks yellow...

Roger

 

[Drakkith]

The Sun is not yellow, it is white. The only time it looks yellow is near sunrise and sunset.

 

[sophiecentaur]

Two things:

1) Scattering, what is that?
2) Reading that link suggests that the amount of yellow is due to the distance sunlight has to travel through the atmosphere.

Scattering sounds to me like something being reflected "away" from the observer. My thinking is that if scattering of blue light happens then the sky should be everything but blue. Please correct me while I'm wrong :)

Then take the only star I know, the Alpha Centauri (AC).

The light from AC certainly appears white to me.

Whiter than our sun at noon, even.

The only difference is that I/we cannot see it unless it is dark.

Correct me if I'm wrong, but isn't the distance for AC's rays of light through our atmosphere equally long as it is for our sun's rays of light?

Yet Alpha Centauri looks white while our sun looks yellow...

Roger

Sunlight can be altered either by partial absorption of some wavelengths or by some of the energy at some wavelengths being scattered from the direct path. Short wavelengths (the blue end of the spectrum) are scattered more than the longer wavelengths. We see the sky as blue(ish) because sunlight that would go direct to points on the Earth's surface remote from where we are standing is scattered in our direction. Meanwhile, the shorter wavelengths that would come directly to us, are scattered to other places - making the sky look blue to them.

Different stars have different temperatures so, inherently, they appear different colours. Viewing them through our atmosphere affects them all but still the bluish stars (hotter) look less yellow than the (less hot) stars like our Sun. The angle in the sky will also affect the colour - which is why the Sun looks much more red when it's near the horizon (sunset) - the path through the air is much longer, so more scattering. The short wavelengths the we 'lost' are seen by people to the west of us (it's still only afternoon for them).

If you actually measure the colour of AC when it's low on the horizon, it will also look more yellow than when it's high in the sky. But our subjection appreciation of the colour difference may not be good enough to spot this effect unless we are well trained observers.

 

[SteamKing]

Two things:

1) Scattering, what is that?
2) Reading that link suggests that the amount of yellow is due to the distance sunlight has to travel through the atmosphere.

Scattering sounds to me like something being reflected "away" from the observer. My thinking is that if scattering of blue light happens then the sky should be everything but blue. Please correct me while I'm wrong :)

Then take the only star I know, the Alpha Centauri (AC).

The light from AC certainly appears white to me.

Whiter than our sun at noon, even.

The only difference is that I/we cannot see it unless it is dark.

Correct me if I'm wrong, but isn't the distance for AC's rays of light through our atmosphere equally long as it is for our sun's rays of light?

Yet Alpha Centauri looks white while our sun looks yellow...

Roger

Is Alpha Centauri where you are from originally?

 

[D H]

Yet Alpha Centauri looks white while our sun looks yellow...

Alpha Centauri is fairly close in color to that of the Sun. It's a question of how our eyes work.

You can't see Alpha Centauri in the daytime. Too much sunlight! You can only see it at night. Of course it looks white. Almost all of the stars look white. Use a telescope and the multitude of colors will jump right out. It's quite beautiful.

You don't see color at all (scotopic vision) or not very well (mesopic vision) at night. The cones that are responsible for our color vision don't work very well under low lighting conditions. The rods in our eyes take over when illumination decreases. There's only one kind of rod, so all we see under extremely low lighting conditions is black and white: No color at all. With slightly greater illumination we can see some color, but it is rather muted. You are looking at the stars with your cones working very poorly at best. That's why most stars appear to be white.

 

[Drakkith]

I just went outside and looked at the Sun for a moment. It looked pretty white to me.
Now excuse me while I go replace my retina...

 

[rogerk8]

I need to say thanks officially to all you guys but especially sophiecentaur and DH.

Take care!

Roger

 

[sophiecentaur]

I need to say thanks officially to all you guys but especially sophiecentaur and DH.

Take care!

Roger

You're welcome.
It's amazing just how counter intuitive a lot of this stuff is. Working it out on one's own can be really hard.

 

[sophiecentaur]

I just went outside and looked at the Sun for a moment. It looked pretty white to me.
Now excuse me while I go replace my retina...

Apparently, the navigators on old ships (pre-sextant days) all went blind in one eye at a very early age. At least they had the sense to use the same eye each time for sun sightings.

 

[rogerk8]

It sure is.

It thus seems like you are aware that you and DH have explained the phenomenon in a completelly different but yet scientific way?

Ain't that fascinating :)

Roger

 

[cjl]

Sunlight can be altered either by partial absorption of some wavelengths or by some of the energy at some wavelengths being scattered from the direct path. Short wavelengths (the blue end of the spectrum) are scattered more than the longer wavelengths. We see the sky as blue(ish) because sunlight that would go direct to points on the Earth's surface remote from where we are standing is scattered in our direction. Meanwhile, the shorter wavelengths that would come directly to us, are scattered to other places - making the sky look blue to them.

Despite all of this though, I would still say the sun itself is white, not yellow. People think of it as yellow in part because that is how it is depicted in paintings/artwork, and in part because it does actually look yellowish when it is near the horizon, which also attenuates the light enough that it isn't as uncomfortable to look at. Since it is normally painful (and potentially damaging) to look at the sun, we unconsciously assume that it is always yellow (as it is when it is near the horizon).

However, if you need further proof, look at a piece of white paper sitting in the sunlight. It is reflecting the same spectrum that is incident on it. Does it look like it is sitting in yellow light? If you think that the blue from the sky may be affecting this, do it inside, with the sunlight coming in through a smallish window to prevent much skylight from being able to change your results, and it should still look quite white.

 

[sophiecentaur]

Sunlight is certainly taken to be one of the 'white' illuminants. A surface (paint) that reflects all wavelengths equally will look White (that is what we call white) when it's outside. But of course, it is getting blue light from the sky around the Sun, too. Depending on the state of the clouds (red absorbers), elevation of the Sun and the presence of snow on the ground, we can get all sorts of 'whites', from warm white to cold white.

This is all subjective; it's not really a matter of "proof". You would find that your white paper will look very convincingly white when indoors, under all sorts of illumination (very yellow, dimmed filament lamps, for instance). That's because your colour vision is desperately trying to take out any perceived colouration due to to the illumination. Your brain tries to 'integrate to grey' - just like your digital camera also tries to do, and jiggles its colour channel gains accordingly. It can be very successful and it can also fail abysmally, when trying to match clothes colours in a shop, under artificial light. (Colour film is hopeless because it has only one setting for its colourimetry). That's why we take things out into the street to check.
The Sun is not particularly blue or particularly red/yellow, compared with all the other stars around there are much more extremes on other side of our 'white' Sun - intense blue stars that are burning up very quickly and are much hotter and some red giants that are cooling down at the end of their lives. Their particular colours of illuminant would be taken as normal (white?) by the local inhabitants.
Overall, coulorimetry is a fascinating subject, I think.

 

[Solon]

Here is an image of the Sun taken with a neutral density filter.
Sun Spots. 30 Sept 2011. 1203hrs 200mm | ISO 50 | f/13 | 1/8000sec | 10 stop ND filter
http://mcalisterium.wordpress.com/2011/09/30/sun-spots/
So surely if we used exactly the same set-up from outside Earths atmosphere, we would know how much effect the atmosphere has on what we see from Earth? A simple experiment, but has never been done AFAIK.

 

[sophiecentaur]

A simple experiment, but has never been done AFAIK.

What, no happy snaps taken on the Moon with standard film? - Perhaps before your time!
But photographs tell us nothing about perceived colour. Colour is subjective and even the best camera can only make a stab at it. There is no doubt, though, that there is atmospheric absorption and scattering so the colour would have to be different out in space. Is this really an issue worth discussing?

 

[cjl]

Sunlight is certainly taken to be one of the 'white' illuminants. A surface (paint) that reflects all wavelengths equally will look White (that is what we call white) when it's outside. But of course, it is getting blue light from the sky around the Sun, too. Depending on the state of the clouds (red absorbers), elevation of the Sun and the presence of snow on the ground, we can get all sorts of 'whites', from warm white to cold white.

The "blue light from the sky around the sun" is the reason I suggested doing it through a smallish window, with the sky mostly blocked. That having been said, you're right that white isn't terribly well defined.

This is all subjective; it's not really a matter of "proof". You would find that your white paper will look very convincingly white when indoors, under all sorts of illumination (very yellow, dimmed filament lamps, for instance). That's because your colour vision is desperately trying to take out any perceived colouration due to to the illumination. Your brain tries to 'integrate to grey' - just like your digital camera also tries to do, and jiggles its colour channel gains accordingly. It can be very successful and it can also fail abysmally, when trying to match clothes colours in a shop, under artificial light. (Colour film is hopeless because it has only one setting for its colourimetry). That's why we take things out into the street to check.
The Sun is not particularly blue or particularly red/yellow, compared with all the other stars around there are much more extremes on other side of our 'white' Sun - intense blue stars that are burning up very quickly and are much hotter and some red giants that are cooling down at the end of their lives. Their particular colours of illuminant would be taken as normal (white?) by the local inhabitants.
Overall, coulorimetry is a fascinating subject, I think.

Sure. I debated going into a more lengthy discussion about various color temperatures, and how an incandescent light looks white despite a significant red/infrared bias in its spectrum, but I didn't feel like going into that much detail. That having been said, I don't think anyone following normal conventions would call a 5800K blackbody "yellow", and looking at solar spectra, I cannot see any evidence of significant blue attenuation (compared to the rest of the spectrum) when the sun is at a high angle in the sky. Sure, blue is scattered more than other colors (and thus the sky is blue), but the vast majority of blue light from the sun is still making it through the atmosphere unimpeded. As a result, I don't see how the sun could be considered anything other than white to most people, at least when it is high in the sky. (Admittedly, this is all somewhat academic, since the intensity of the sun is far too high to look at directly without a filter in any case, and once you add a filter, the apparent color will be strongly affected by the filter's exact properties).

 

[sophiecentaur]

There are many sources that are hotter than the Sun - electric arcs, for instance - and many that are not so hot. It is strictly a matter of taste, of course, but it would be reasonable to call the different 'whites' by familiar names - namely blue, yellow and orangy-red. But that's nothing to do with Physics, is it? You'd need to ask your granny or someone else who is strictly non-technical. (Apologies to all the Grannies with PhD's in Physics - there must be quite a few on PF - or perhaps they would have more sense than to get into the testosterone laden maelstrom that is Physics Discussion)

 

[Solon]

@sophiecentaur

There is no doubt, though, that there is atmospheric absorption and scattering so the colour would have to be different out in space.

Well I'd like to see the difference.

Is this really an issue worth discussing?

The curse of an inquiring mind. Sorry, won't do it again.

@cjl

and once you add a filter, the apparent colour will be strongly affected by the filter's exact properties

That's why the ND filter is needed, it attenuates all wavelengths equally, so should show a true colour.

 

[cjl]

That's why the ND filter is needed, it attenuates all wavelengths equally, so should show a true colour.

Well, sure, but 10 seconds with Google should show that you can find images taken through a neutral density filter that range from a bit bluish to fairly yellow, and everything in between. In addition, the filter isn't the only thing affecting the color of the picture - the camera's white balance settings can make a significant difference too. Basically, there's not a whole lot you can tell about the sun's "true color" from a picture taken through a neutral density filter, since there's too much variability in the results.

 

[Solon]

Well, sure, but 10 seconds with Google should show that you can find images taken through a neutral density filter that range from a bit bluish to fairly yellow, and everything in between. In addition, the filter isn't the only thing affecting the color of the picture - the camera's white balance settings can make a significant difference too. Basically, there's not a whole lot you can tell about the sun's "true color" from a picture taken through a neutral density filter, since there's too much variability in the results.

Well that's why I suggested the exact same set-up, same camera and settings, same (overhead) shot, that should be as close a comparison as we can get.

 

[voko]

Well that's why I suggested the exact same set-up, same camera and settings, same (overhead) shot, that should be as close a comparison as we can get.

The white balance settings are not specified at all. Neither are the postprocessing steps.

 

[Solon]

The white balance settings are not specified at all. Neither are the postprocessing steps.

I have posted the question to the photographer, maybe he can recall those details.

 

[cabraham]

I remember reading that solar spectra peaks in the 510 nm region. So the sun is definitely a cool white. The peak emission is bluish green. It's definitely not yellow, as yellow is around 580 nm. Maybe the perception of yellow has to do with the surrounding sky color of cyan (greenish blue). But it really looks white to me. If I'm under tungsten lamps, I have trouble seeing the difference between a black sock and one that's navy blue. But when I hold up the socks to the window during daylight hours, the difference is easy to see.

Under sunlight, blue, even very dark blue, is clearly distinguishable from black. But tungsten lamps are very yellowish, and blues just don't stand out at all. The color spectrum of sunlight is clearly towards the cool end, vs. tungsten which is very warm.

Just what I have read and observed. Like previous posters have stated, color is subjective.

Claude

 

[Solon]

A pink Sun?
http://casa.colorado.edu/~ajsh/colour/Tspectrum.html

This begs the question then about how we can base the model of distances in the cosmos on the colours of the distant Suns when the colour of our own Sun is stll so contentious?

 

[Drakkith]

A pink Sun?
http://casa.colorado.edu/~ajsh/colour/Tspectrum.html

This begs the question then about how we can base the model of distances in the cosmos on the colours of the distant Suns when the colour of our own Sun is stll so contentious?

It's not contentious. The Sun's COLOR is white. But since color is purely perceptive, we don't use it for accurate measurements. Instead, we measure the spectrum and plot the output per wavelength using a spectrograph.

Also, we don't base distances off of color. We have a cosmic distance ladder that involves several different techniques, none of which are based on color. http://en.wikipedia.org/wiki/Cosmic_distance_ladder

 

[chemisttree]

Of course the sun is white.

 

[sophiecentaur]

I have posted the question to the photographer, maybe he can recall those details.

If you really want the answer to this question, you need a graph of relative amounts of atmospheric absorption.
There is a graph in here that shows the difference in the transmittance over a wide range of wavelengths. Look at the 400-900nm region, which shows that red light is reduced half as much as blue light. There are graphs in several other articles, too. Just do some googling on atmospheric optical absorption. That amount of slope will have a significant effect on the perceived colours of the Sun, above and below the atmosphere. A photograph would be no more conclusive than that graph.

 

[sophiecentaur]

Of course the sun is white.

But is it Persil White?

 

[davenn]

OK ... regardless of how it looks visually or photographically via the Earth's atmosphere,
The Sun is deemed a yellow class star...

cheers
Dave

 

[chemisttree]

But is it Persil White?

Up close it is.

 

[Solon]

@Drakkith.

It's not contentious. The Sun's COLOR is white. But since color is purely perceptive, we don't use it for accurate measurements. Instead, we measure the spectrum and plot the output per wavelength using a spectrograph.

I've never seen a colour calibration card used out in orbit, or a white balance reference card. Maybe it's because colour being perceptive they don't bother?. SOLSPEC has 1 nm sampling, but calibration is needed regularly, and I don't see any charts showing how the spectrum changes, if it does, over time. Is ours a variable star?

Also, we don't base distances off of color. We have a cosmic distance ladder that involves several different techniques, none of which are based on color.

I though they used light and colour curves when examining the Cephid variables? Maybe not.

 

[russ_watters]

I've never seen a colour calibration card used out in orbit, or a white balance reference card.

Not in orbit, but...
http://marsrover.nasa.gov/mission/ spacecraft _instru_calibr.html

 

[Drakkith]

OK ... regardless of how it looks visually or photographically via the Earth's atmosphere,
The Sun is deemed a yellow class star...

Yes, which is a misnomer.

The term yellow dwarf is a misnomer, as G stars actually range in color from white, for more luminous types like the Sun, to only slightly yellow for the less luminous GV stars.[8] The Sun is in fact white, but appears yellow through the Earth's atmosphere due to Rayleigh scattering. In addition, although the term "dwarf" is used to contrast yellow main-sequence stars from giant stars, yellow dwarfs like the Sun outshine 90% of the stars in the Galaxy (which are largely orange dwarfs, red dwarfs, and white dwarfs, the latter being a post-main-sequence star).

http://en.wikipedia.org/wiki/Yellow_dwarf

@Drakkith.I've never seen a colour calibration card used out in orbit, or a white balance reference card. Maybe it's because colour being perceptive they don't bother?. SOLSPEC has 1 nm sampling, but calibration is needed regularly, and I don't see any charts showing how the spectrum changes, if it does, over time. Is ours a variable star?

No, the Sun is not a variable star.
While the Sun does go through changes, none of these have any effect on the average temperature of the Sun's photosphere, which means that it continually outputs light with a spectrum close to a perfect blackbody at about 5,800 kelvin.

I though they used light and colour curves when examining the Cephid variables? Maybe not.

To my knowledge they only measure the magnitude changes.

 

[cjl]

OK ... regardless of how it looks visually or photographically via the Earth's atmosphere,
The Sun is deemed a yellow class star...

cheers
Dave

That picture is clearly misleading though - it shows a 6000K blackbody as being on the orangey side of yellow, which it clearly isn't (as anyone who has seen a 6000K blackbody will attest). Even 3000K, which that picture shows as being distinctly red, will look fairly white to most people (a lot of incandescent light bulbs operate at around that temperature), though it will probably look somewhat yellowish (especially if you have a whiter reference light to compare it to).

 

[sophiecentaur]

Have you never come across false colour before? It's used all over the place to 'model' variables with different colours. Why do you take such a picture so literally? You would never make the same mistake with a thermograph.

This thread has really got out of hand. Stars have a range of surface temperatures and that produces different 'whites', which can be described in terms of yellow, blue, white etc. Those names are only used 'colloquially'. Anyone who wants to be precise will talk in terms of temperature.

The atmosphere changes the colour of anything viewed through it because it is not a neutral density filter. Is there any doubt about that? Any 'experiment' to demonstrate this needs to be quantitative and properly carried out. And it has been done many times. Random photographs do not constitute evidence of anything.

 

[CubicFlunky77]

I have never thought the sun was yellow. It is too bright for me to observe for more than 5 seconds (even with very dark sunglasses). Color doesn't seem to be an objective measure since I cannot tell if my perspective of a certain color (in this case, yellow) has the same shade/physical characteristics as your 'yellow'.

 

[davenn]

Have you never come across false colour before? It's used all over the place to 'model' variables with different colours. Why do you take such a picture so literally? You would never make the same mistake with a thermograph.

This thread has really got out of hand. Stars have a range of surface temperatures and that produces different 'whites', which can be described in terms of yellow, blue, white etc. Those names are only used 'colloquially'. Anyone who wants to be precise will talk in terms of temperature.

The atmosphere changes the colour of anything viewed through it because it is not a neutral density filter. Is there any doubt about that? Any 'experiment' to demonstrate this needs to be quantitative and properly carried out. And it has been done many times. Random photographs do not constitute evidence of anything.

Totally agree

that diagram, which is the standard model for stars
of course there's going to variations within the ranges displayed. It just gives a good indication of where our sun/star is positioned in the great scheme of things

Dave

 

[cjl]

Have you never come across false colour before? It's used all over the place to 'model' variables with different colours. Why do you take such a picture so literally? You would never make the same mistake with a thermograph.

Of course I have, but it seems like a lot of people (at least among those who haven't studied physics/astronomy very much) do take those charts fairly literally as representing the apparent color. In some ways, I'd prefer if it didn't bother with the color at all, and rather just plotted luminosity and photospheric temperature. The color serves no real additional purpose on that chart, and as I said, it's not really even remotely representative of what blackbodies at those temperatures actually look like (since that would require a slightly yellowish white in the lower right, moving to a slightly bluish white in the upper left, none of which would be terribly saturated colors).

This thread has really got out of hand. Stars have a range of surface temperatures and that produces different 'whites', which can be described in terms of yellow, blue, white etc. Those names are only used 'colloquially'. Anyone who wants to be precise will talk in terms of temperature.

The atmosphere changes the colour of anything viewed through it because it is not a neutral density filter. Is there any doubt about that? Any 'experiment' to demonstrate this needs to be quantitative and properly carried out. And it has been done many times. Random photographs do not constitute evidence of anything.

I think you might be taking this a little too seriously...

 

[sophiecentaur]

I think you might be taking this a little too seriously...

'Appropriately' seriously, I think.

 

[D H]

When the Sun is far above the horizon it is indeed "white" as our eyes would perceive it. We only have three color sensors, and how sunlight from a high Sun excites those three kinds of sensors is pretty close to white.

We can't see that white Sun because we can't take anything more than a fleeting glance at the Sun when it is far above the horizon. Even that fleeting glance hurts, and anything longer would cause permanent eye damage. We can only look at the Sun long enough to deduce it's color when it's close to the horizon and the sunlight has passed through a long stretch of the atmosphere. The scattering that results from this long stretch of atmosphere is what makes us perceive the Sun as yellow (or orange or even red). The perceived color changes as the Sun gets closer to the horizon due to this scattering. When the Sun is high overhead, it's white. Just about as perfect a white as can be.

 

[sophiecentaur]

Which 'white' did you have in mind? Look at the controls on your computer monitor and you'll find a lot of options even on that. Are you certain that, if someone altered the 'white' on your screen whilst you were away, you'd necessarily notice?

 

[Drakkith]

Which 'white' did you have in mind? Look at the controls on your computer monitor and you'll find a lot of options even on that. Are you certain that, if someone altered the 'white' on your screen whilst you were away, you'd necessarily notice?

The kind of white the Sun looks like near zenith?

 

[sophiecentaur]

The kind of white the Sun looks like near zenith?

What sort of a day and where in the World? This is just not accurate enough - even for colour printing and certainly not for Science.

 

[Drakkith]

What sort of a day and where in the World? This is just not accurate enough - even for colour printing and certainly not for Science.

I don't know what you're getting at Sophie.

 

[D H]

Which 'white' did you have in mind?

CIE D65. That is as white as white can get, and it's the color of the Sun at noon. Note that correlated color temperature is not quite the same blackbody temperature. CIE D65 is *the* standard definition of white. http://en.wikipedia.org/wiki/Illuminant_D65

 

[sophiecentaur]

I'd go along with that - as long as you acknowledge that D65 is an Illuminant and not a 'colour'. It's defined in terms of Daylight / Sunlight, which is not only direct light from the Sun. The rest of the sky is there too.

 

[sophiecentaur]

I don't know what you're getting at Sophie.

I'm getting at the fact that the Sun is not something that we ever look at, in fact. So, on its own, the Sun is no use as a reference, and the state of the sky and the surrounding scene is going to contribute to what we see.
When Sunlight (as a luminant) is said to be 'white', the light comes from all over the sky and not down a narrow tube. The colour of a sheet of white paper, at midday, in the shade, is not the same as out in the open. All that indirect light that gets into the shade is also getting to the paper when it's not in the shade. If we're trying to be scientific about this, we have to be precise and have a possible measurement system - not just to say what it 'looks like'.