Why the Sun appears yellow

[George]

The clouds are now thicker. I'd settle for a silver lining at this point.

The SAD (Sunlight Attenuator Device) is now fully operational. I've called Houston and they have given me a "go". (ok - it was really "go home" but let's think positive).

The initial test was on a 60 watt clear light bulb. The inner markings and filament reflection off the glass were easily seen. Then I turned it on....(just kidding).

I still think the Sun will look white because of the irradiance curve is so flat down here. Maybe NASA will take it and give it to an astronaut to see what he sees. Blue-green still has a fightin chance considering the irradiance curve the Sun radiates.

[George]

I am now in Orlando with full sunshine but no S.A.D.! [I''l forego the pun] I did not bring it as I heard the weather was bad. [Not to mention the concern I have for it getting out cf calibration during the flight.]

SAD2 will be complete by tomorrow. The SAD2 strobe disk is an aluminum pie pan and the rotating shaft is a tube from a paper towel roll. The S.A.P. (SAD Activating Person) is prepared to operate.

I hope this does not distract you from your interest in other great events such as our landing on Mars.

[George]

Another update.

Today the sun peeked through and I conducted the normal preliminary tests in my lab (bedroom).

FAILURE. The main drive shaft (and only shaft) to the strobe plate de-coupled! Apparently, Elmer's glue is not suitable for aluminum. I had feard this but did not want you t o worry too much.

I am repairing. Should be ready just after sundown, probably.

[Eroica]

The tension is unbearable! ](*,)

[George]

The tension is unbearable! ](*,)

Hey....your name's not George!

Oh, sorry, Eroica. The solitude of a lonely inventor can be stressful.



Update....di-dt, di-dit, di-dit...

Rain.

However. I did test SAD2 (semi-repaired) on the hotel lamp. I saw darkened bands. After climbing high enough to see the bulb.....sure enough....2 little flourscent bulbs. You sawie it coming too, right?

[George]

S.A.D. 3 will be electric powered. A hand fan or battery powered drill will drive the strobe plate.

I wanted everyone to know in advance to thwart all the anticipated media questioning of future SAD development. Now that they beat Pres Bush down to spilling everything about Moon and Mars trips, I figured they'd be here next.

[George]

Left SAD 2 behind as the aluminum pie pan would probably look like a land mine to the airport radar security. Hope the Smithsonian will forgive me.

Back in San Antone..

Yeah...Sunlight! And it isn't even Spring.

I was the S.A.P. and the Sun's image was.......WHITE. No surprise.

The hand spinning was non-uniform. I went ahead and upgraded SAD into SAD 3. An old variable speed drill held the pencil shaft nicely. The image was reasonably steady at the higher rpm (up to 1200 rpm).

The slot was about .5 mm wide (1 cm long). It was 7 cm from center of strobe plate. This reduced the net sunlight by almost 99.9%! Still the sunlight was very intense but I could see the disk, barely. (Solar angle above horizion was about 45 deg.)

I did not test it closer to the horizion. But will see how the color compares with the color changes near the horizion tomorrow, hopefully.

Try to get some rest, all. You deserve it.

[George]

Update....di-dit, di-dit, di-dit

Finally got a sunset to watch.

No solar hue variation with SAD3. The sun's setting and changing colors seem to match what was seen through SAD3.

This seems to confirm the conventional wisdom that the color changes of the Sun near the horizon are atmospheric related only and not a due to the eye's color cones thresholds possibly being exceeded [when it looks white].

It also may confirm the Sun's color, as seen on Earth, to be white.

Now we need a SAD4 to go up with an astronaut where the Sun's irradiance is nearly twice as intense in the blue and green range as opposed to down here. Reducing the Sun's intensity might make all the difference.

Any advice? Any astronauts want to be a S.A.P.?

[Eroica]

=D> Well done, George! A positive contribution to the science of astronomy. Give that man an award!

So, if the sunlight is reduced to about 0.1% of its normal intensity, it looks white - is that your finding?

I don't suppose you could reduce it to, say, 0.01%, or even 0.001%? Baader AstroSolar Safety Film reduces the intensity of sunlight by 99.999%. But it does not reduce it equally across the spectrum - blue wavelengths are attenuated much more than red - so it's not much use to us here.

[Swift]

This is a serious comment, not a joke.... you should publish this. No thoughts as to where.

[George]

Well done, George! A positive contribution to the science of astronomy. Give that man an award! So, if the sunlight is reduced to about 0.1% of its normal intensity, it looks white - is that your finding?

Yes. White. Essentially the hue that you see with a quick and analytical glance of the sun is the same hue of color you would see when the intensity is lowered to 99.9%. It is white during most the day. As the sun looks yellowish near the horizon, so also does the view through SAD3.

I don't suppose you could reduce it to, say, 0.01%, or even 0.001%? Baader AstroSolar Safety Film reduces the intensity of sunlight by 99.999%. But it does not reduce it equally across the spectrum - blue wavelengths are attenuated much more than red - so it's not much use to us here.

I can clearly understand why they have to cut out 99% more than I did...it is still very, very bright even through SAD. So, if you add a mirror for light gathering, it would be imperative to cut it down.

I assumed any kind of filter or even glass might alter "true color" viewing, so, the strobe idea makes more sense as it is "pure" sunlight (albeit atmosphericly bleached).

It was interesting to note that the image was becoming quite weak as the sun neared the horizon due to the amount of reduction through SAD3. I tried SAD on Sirius, as it was high and bright, but it was not visible. The moon was visible through SAD.

The quality of the solar viewing was not clean enough to note detail such as sunspots, although, a future SAD might improve enough to see them. A filter works easier in this case, however.

This is a serious comment, not a joke.... you should publish this. No thoughts as to where.

Thanks much! That comment makes it all worth the "colorful" effort.

The real value of SAD will be whether or not it makes sense to send it up with an astronaut who will take the few minutes to give it a try. This, I presume, would mean convincing someone with pull that the hypothesis makes sense. Specifically, the hypothesis being that the true color of the sun can not be determined until the observer is outside the atmosphere and with the ability to attenuate the solar visible light intensity to a more reasonable level for the human eye.

Since the irradiance of the sun in the blue and green range above the atmosphere is almost twice that as the irradiance below the atmosphere, there is a fair chance that we might have a blueish or greenish or blue-greenish or something other than white star. I think humanity is ready for this super advanced information. It would also be cool to see the BA add it to his future, hopefully, new book.

If you can reword this to make it "more better", let me know.

Any way I can show you the S.A.D. 3? I can easily get a jpg on it but don't know what to do after that as I do not have a working website at this time. Of course, you know what to expect as it is fairly "sad".

[George]

The SAD3 in space would be nice. However, what if Hubble took a look at the "Sun's Twin" - 18 Scorpii. This star is very close in temperature to the Sun. If it's absorption lines match then it may be another way to get a handle on the Sun's color.

Another possibility, but likely weak in true color accuracy, would be to heat an object close to a blackbody to the Sun's 5700 deg. C temperature. The Sun is not too far off from a blackbody radiation curve in the visible portion. I can't imagine what we would heat-up that would be accurate. I still don't understand the hole in the furnace trick for that matter. It is also possible the Sun's absorption lines would make a difference in the true color afterall.

Therefore, a S.A.D. and a S.A.P. might be the best method afterall.

[Swift]

Another possibility, but likely weak in true color accuracy, would be to heat an object close to a blackbody to the Sun's 5700 deg. C temperature.

As someone who regularly heats things to 1200 to 1400C, heating something to 5700C would be no easy thing to do.

[George]

Another possibility, but likely weak in true color accuracy, would be to heat an object close to a blackbody to the Sun's 5700 deg. C temperature.

As someone who regularly heats things to 1200 to 1400C, heating something to 5700C would be no easy thing to do.

We've got a plasma unit that gets close but I do not think it would be worth the effort due to the amount of effort and absorption within the spectrum problems.

It would be nice to hear..."Oh,yeah, we do this all the time and whatever we heat to this looks blue". Not likely, heh?

[George]

I found an interesting site that shows the Sun's color to be blueish.

This site displays color rendering based on blackbody curves.

I do not know why the "peachy pink" site varys with this result.

Naturally, I favor the blue sun site over the yellow or peachy pink sites. It also uses Fortran which I always liked (except for the careless "do loops").

Since the Sun's spectrum is so strong in blue and green, this analysis makes more sense to me.

It's possible other sites are post atmosphere "bleaching" in their color rendering.

So the quest continues!! 8) \/

Has Hubble, Cassini or others taken any "natural" or "true color" shots of G class stars?


[The report that 18 Scorpii (our supposed solar twin) was yellow-orange did not thrill me. I wonder what evidence they use to make that claim.]

If I listed other G2 stars, would anyone check them out in their scopes???

[JohnOwens]

The idea that the Sun is white because it's intensity is too great for us to see it's color, can be tested. What if we make a strobe (a slit in a rotating disk) to attenuate the total light received, then, maybe it will look yellow. Naturally, the sky is 100% overcast.

My bet is it will still look white as the spectrum (post atmosphere) suggests.

Or, you could make a different device (you could even use the same materials!), by putting a pinhole in the pie tin (or a piece of paper, cardboard, etc.) and projecting the Sun onto a white surface (but you'd better hope it's an honest white!) like you do for looking at sunspots or eclipses. That would give you variable attenuation too, by changing the distance between the two; you could make it as bright as you can stand to (hopefully) get the best cone response.

-----------------------

Going back a bit, during the sky & cloud color discussion, I was wondering: what about the green color of the clouds you see just before hail and/or tornados? What brings that about? I've found that far odder than the blue sky or a perceived yellow Sun.

[AstroSmurf]

I had an idea recently. While the sky's blue colour is due to Rayleigh scattering, shorter-wavelength light is also scattered, as can be seen near sunset and sunrise. However, this very same process is still happening during the day, but since the length of atmosphere traversed is less, the "scattering angle" is similarly less.

The end result of this is that the sky will look yellow, but only very close to the sun, giving us the impression that the sun itself is yellow.

Comments?

[George]

The idea that the Sun is white because it's intensity is too great for us to see it's color, can be tested. What if we make a strobe (a slit in a rotating disk) to attenuate the total light received, then, maybe it will look yellow. Naturally, the sky is 100% overcast.

My bet is it will still look white as the spectrum (post atmosphere) suggests.

Or, you could make a different device (you could even use the same materials!), by putting a pinhole in the pie tin (or a piece of paper, cardboard, etc.) and projecting the Sun onto a white surface (but you'd better hope it's an honest white!) like you do for looking at sunspots or eclipses. That would give you variable attenuation too, by changing the distance between the two; you could make it as bright as you can stand to (hopefully) get the best cone response.

This is actually a good idea. I understand that Kodak makes "gray" paper, or some other name, that reflects visible light evenly.

However, this will cost more than the 19 cents allocated for the project. :wink:

Going back a bit, during the sky & cloud color discussion, I was wondering: what about the green color of the clouds you see just before hail and/or tornados? What brings that about? I've found that far odder than the blue sky or a perceived yellow Sun.

I do not know. I have always assumed it was the presence of hail stone which added some sort of refractive factor to give green the advantage. However, I have been in these storms and no sign of hail.

My guess would be that this green cloud condition ocurrs near sunset so much of the blue has already scattered away which makes green the winner when enhanced by the cloud properties. It would be easier to make this claim if the rest of the sky was green.

I am still reading up on scattering and have a long way to go. My 19 cent SAD may be worth more than what I actually know. 8-[

[George]

I had an idea recently. While the sky's blue colour is due to Rayleigh scattering, shorter-wavelength light is also scattered, as can be seen near sunset and sunrise. However, this very same process is still happening during the day, but since the length of atmosphere traversed is less, the "scattering angle" is similarly less.

The end result of this is that the sky will look yellow, but only very close to the sun, giving us the impression that the sun itself is yellow.

Comments?

I am curious if you actually see the sun as yellow when it is near noonday. I see it as white.

The main factor to keep in mind is that Rayleigh scattering varies as the 4th power of the light's frequency. Blue light is about 10 times more effective at Rayleigh scattering than red. [Violet even more so but the Sun does not radiate much violet and, I think, our color cones are not so receptive to violet (but I once heard otherwise).]

Particle size is another major factor. Nitrogen and Oxygen, apparently, do a great job of scattering by themselves. Particles close to or greater than the wavelength of the light will not experience Rayleigh scattering.

Scattering angle is also important. Rayleigh scattering is found at all angles. Mie scattering is strong at narrower angles which tends to whiten the sky. This is why the sky is so blue overhead since the angle is around 90 deg.

At sunset, the blue sunlight is so scattered that you don't see it, rather, you see what is left.

Oddly, Mars exhibits nice blue sunset's near the sun. This might be due to the larger particles which do not allow much blue to scatter but might cause the other colors to skedaddle. [see "Martian Sky" thread for more]

[darkdev]

I didn't read all four pages, so I apologize if this has been suggested...

Can we just video tape the Sun with a device with equal RGB absorbance, then watch it on a properly calibrated monitor?

If it was actually more yellow, the screen would show it. Otherwise it is only an effect for us.

[George]

I didn't read all four pages, so I apologize if this has been suggested...

Can we just video tape the Sun with a device with equal RGB absorbance, then watch it on a properly calibrated monitor?

If it was actually more yellow, the screen would show it. Otherwise it is only an effect for us.

A video or camera observing the sun on the ground should show it as white if they truly simulate eye response. These devices were not designed, however, to observe something so bright.

This problem is even worse when trying to observe the sun from space as it is even brighter.

Filters would help but might corrupt the observation. However, I am not a filter pro. A simple strobe would seem to be the best alternative and a good camera might do the trick (attached to the strobe) but might not as cameras are designed to give reasonable true color based on conditions down under the atmosphere with less UV, etc. I just don't know.

[Sorry, I just noticed your post.]

[John Dlugosz]

It must be remembered the "yellow" of the sun really isn't that "yellow." i've projected the disk onto paper and its not /that/ apparent. It would seem most likely at this time that it is a physiological effect.

In a book on lighting for digital scenes, it teaches that sunlight is yellowish, compared to shade which is bluish. Setting virtual lights of the right colors from proper directions does indeed make a rendered scene look more real. I'm sure the advice is based on studio lighting skills; for example making something look like it's outside when it's really in a soundstage.

A video or camera observing the sun on the ground should show it as white if they truly simulate eye response. These devices were not designed, however, to observe something so bright.

I could shoot a photographer's Grey Card which is designed to be pure in color, and compare the R,G,B values recorded (with the camera set identically, auto-whiteballance turned off) for the card shot in sunlight and in shade.

Point is, I disagree with the proposed idea that the disk of the sun looks yellow but actually produces white light, because artists always portray sunlight as yellow. Exagurated yellow cast gives the impression of sunny.

My conjecture:

If you measure the spectrum of direct sunlight at noon, factor that curve by each of the three response curves of the human R,G, and B sensors, then plot the resulting tristimulus value on the CIE diagrams, you'll find it is right at the spot labeled "white". By definition! The experiment was a tautalogy.

But, the visual perception doesn't work with absolute values. If you take a film-based photograph inside using artificial light, the slide (or uncorrected print) will look orange. A spectrograph would tell you that it is indeed an accurate photo. But we didn't perceve the room as orange when in it!

The vision is calebrated based on overall ambiant light. So, what is the overall lighting when outside? The diffuse ambiant light is the scattered blue light from the sky. We adjust our vision so the white paper looks white to us, under this slightly blue light. Now, with our internal white-ballance set up that way, pure sunlight looks yellow, and pure diffuse light looks blue. (R+G makes yellow). That is, R,G,B has been factored into (R+G),(B) from two sources.

That doesn't explain why a sunset goes from a yellow disk to a red disk without passing through green. update: what was I thinking? I shouldn't post anything requiring thought while skimming boards during a quick 'fresher at work. I see now that after taking the B out and starting to remove some G, it will be more and more R.

[tjm220]

...That doesn't explain why a sunset goes from a yellow disk to a red disk without passing through green.

Because green isn't between yellow and red in the EM spectrum? :-k

[George]

In a book on lighting for digital scenes, it teaches that sunlight is yellowish, compared to shade which is bluish. Setting virtual lights of the right colors from proper directions does indeed make a rendered scene look more real. I'm sure the advice is based on studio lighting skills; for example making something look like it's outside when it's really in a soundstage.

Bluish light from shade makes sense as it would be reflecting blue sky lighting (assuming a non-cloudy day). Yellow light from the sun, IMO, is limited to mornings and evenings. Earlier in the thread I made a S.A.D. (really cheap strobe) and we observed the Sun. Not very high from either horizion, it truly looks white with no hint of any color.

A video or camera observing the sun on the ground should show it as white if they truly simulate eye response. These devices were not designed, however, to observe something so bright.

I could shoot a photographer's Grey Card which is designed to be pure in color, and compare the R,G,B values recorded (with the camera set identically, auto-whiteballance turned off) for the card shot in sunlight and in shade.

I elected to just use my 19 cent strobe. I hope you will try the Grey Card and let us know your findings.

Point is, I disagree with the proposed idea that the disk of the sun looks yellow but actually produces white light, because artists always portray sunlight as yellow. Exagurated yellow cast gives the impression of sunny.

My conjecture:

If you measure the spectrum of direct sunlight at noon, factor that curve by each of the three response curves of the human R,G, and B sensors, then plot the resulting tristimulus value on the CIE diagrams, you'll find it is right at the spot labeled "white". By definition! The experiment was a tautalogy.

Here is a computer rendition that is colorful..... Peachy Pink !! (Have mercy )

Here is an irradiance plot of sunlight above and below the atmosphere... here

Notice how flat the spectrum is from blue to red on the ground. My interest is what color would the eye see above the atmosphere with the much greater blue component. The intensity would first need to be reduced with a strobe or, possibly, use the "Grey card" reflection, or both.
We just might have a bluish star afterall. 8)

But, the visual perception doesn't work with absolute values. If you take a film-based photograph inside using artificial light, the slide (or uncorrected print) will look orange. A spectrograph would tell you that it is indeed an accurate photo. But we didn't perceve the room as orange when in it!

The vision is calebrated based on overall ambiant light. So, what is the overall lighting when outside? The diffuse ambiant light is the scattered blue light from the sky. We adjust our vision so the white paper looks white to us, under this slightly blue light. Now, with our internal white-ballance set up that way, pure sunlight looks yellow, and pure diffuse light looks blue. (R+G makes yellow). That is, R,G,B has been factored into (R+G),(B) from two sources.

That doesn't explain why a sunset goes from a yellow disk to a red disk without passing through green.

Near as I understand, Rayleigh scattering is pulling blue light out first, then green, etc. By the time the narrower green becomes dominant over blue, the yellow, orange and red become more of a factor. Also, I suspect the green is somewhat added to the red, as you noted above, to give yellow the edge to our visiual interpretation. It's probably not quite that simple as other scattering effects are also part of the equation, too.

[edit: "Peachy Pink" link is now correct - (Thanks John)]

[centsworth_II]

Hi all,

I found your discussion very interesting, especially George's contribution. Experiment trumps theory, don't you think? Anyway, I was involved in a similar discussion in a painting forum and suggested that they take a look here. I thought some of you might like to see it from the artists' perspective, so here's the link http://www.wetcanvas.com/forums/show...ge=1&pp=15. By the way, I'm DuhVinci over there.

[George]

Hi all,

I found your discussion very interesting, especially George's contribution. Experiment trumps theory, don't you think? Anyway, I was involved in a similar discussion in a painting forum and suggested that they take a look here. I thought some of you might like to see it from the artists' perspective, so here's the link http://www.wetcanvas.com/forums/show...ge=1&pp=15. By the way, I'm DuhVinci over there.

Welcome to the board, centsworth_II. =D>

The BA's book addresses the misconception of the Sun's color. Your link is interesting. A couple of posts brought up the idea of seeing a color as a result of another color's influence. This can be quite dramatic. It may play a role when it comes to certain shadows, etc. However, I suspect the color of the sun would not suffer from much contrasting colors since it is up in the sky. The strobe I used should be a true representation of it's apparent color. I painted the strobe plat flat black to help avoid any psychological coloring effects.

[centsworth_II]

Thanks for the welcome, George.

Color is a fascinating topic since it involves the physics of light as well as the biology of the human eye and the psychology of human perception.
I'll remember your experiment when I think of how we see the sun. Once the sunlight hits the objects around us, things really start to get complicated!

[George]

If you find any detail specifications on the eye's performance, let me know. I would be curious to find out how many photons at different wavelengths the color cones can handle till they are "maxed-out". My hope is to answer a question I have, namely, if all three cones each receive enough photons each, does this cause "white" to be the color regardless of actual ratios? It should be an easy - yes.

Hope all goes well for you and that your studies, well, are colorful. :wink:

[StormSeeker]

You know, something flashed in my mind early this morning about why we don't see the sun as green, but it kind of slipped away before I could grab hold of it. Something about why chlorophyll is green, and the absorption of red light, and something about if we could see what color the inside of our eyes are. It made sense at the time.

[George]

Ya'll have gotten soft, dag nab it! This polemic problem is still unsolved.

From Ask the Space Scientist #35

Sunlight is not 'white' at all, but is only described that way as an approximation. The sun is an honest to god yellow star by DIRECT observation (watch your eyes), and any combination of colors that matches its color will appear yellow...but we call this particular combination 'white' for reasons that seem as much based on tradition as on poor observation.

"All answers are provided by Dr. Sten Odenwald (Raytheon STX) for the
NASA IMAGE/POETRY Education and Public Outreach program."

He didn't capitalize "god" so I knew he wasn't absolutely, positevely right on this.

I'd say "yellow" is the tradition, not "white". I'd send him a S.A.D. (if he'd take it) but the postage is more expensive than the instrument.

Note: This is probably an old answer. He may be reading the BA's book as we speak, whereupon, he will edit his answer, right?