The Color Of The Sun: Revelation
    By George Cooper | July 4th 2008 04:18 PM | 36 comments | Print | E-mail | Track Comments
    About George

    Amateur astronomer, B.S. M.E., self-acclaimed heliochromologist....

    View George's Profile

    Continuing from the prior article... “The Sun Ain’t Yeller”, cries the heliochromologist.  Undaunted by tradition, dogma, or tens of thousands of erroneous magazine and textbook images of our Sun, heliochromology, a colorful heterodox, is winning the day because the Sun is what it is, color and all, regardless of other’s puerile incognizance. 

    The answer is not a vague, subjective one, but an objective one, as sure as red apples are red.  Heliochromology is our path to enlightenment that will bring resolve to this color conundrum -- a subtle polemic that has been dormant to all of astronomy for hundreds, nay thousands, nay nay, tens of thousands of years, perhaps since mankind first discovered the Sun and its daily color metamorphosis. 

    We know not who the sagacious sapien may have been who indubitably made the first astronomical discovery of all time – the Sun.  Yet, dare none to ask its color?  Who knows what color lurks hidden behind the blaze of Sun? The heliochromologist knows!


    The Sun is certainly a worthy topic of study, one with a dynamic and powerful past. Our star has no equal as an observable object and it is even an object of great worship.  Consider Ra the Egyptian god.  “Ra Ra shish cum bah” – where do you think that mantra comes from?  Even today our host star is worshipped, especially during the spring equinox when thousands stretch along miles of beach and soak in excesses of radiation, possibly proportionate to their beverage consumption.  Yet, ask them the color of their god and they only offer wild guesses, more often than not yellow is their response.


    So, as other worthy scientists have been marching ahead including astronomers, solar physicists, helioseismologists, nuclear physicists, and others, let us now join our color paladins, heliochromologists, as they enter the tail of this triumphant science parade and learn of their brooming efforts that has helped make a cleaner and more colorful path for all. [/corn]


    Why ain’t the Sun yeller?  There are two strong elements that come together to reject any yellow argument:  the Sun’s color we observe here, under appropriate attenuation, and the colors that were scattered away by our atmosphere (extinctions).  This simple finding was not expected initially but was realized after much stumbling along my amateur way.


    A number of attempts were made in trial and error fashion to help get a handle on what might and might not advance this color quest.  The following are most of the examples:


    1)      The SAD -  Solar Attenuation Device consisting of a black painted paper plate with a pencil glued to its center.  A 0.5 mm slot was made in the plate that allowed me to observe the Sun with a 99.9% attenuation while the plate was spun by a cordless drill.  [It costs about 15 cents to make as all parts were used.  Perhaps it is the worlds cheapest scientific instrument.]  The Sun was still too bright to determine its color, as seen terrestrially.


    2)      Random orientations of quality 3 inch convex lenses and prisms were tried but produced little toward any progress.  Worse, it did manage to catch my blue jeans on fire.  [When a restaurant hostess now asks me if I smoke, I say “No, except when I’m on fire.”]


    3)      Invention of the asterochromograph.  This device uses a simple, calibrated mask that alters the Sun’s spectrum that has been dispersed by a prism -- one with five place accuracy of refractive index – to obtain the matching energy for each wavelength of that of the AM0 spectral irradiance data found from space observations. 



     This adjusted spectrum is then scrambled and homogenized into a small dime-sized spot to allow us to simply see the color of any object with known spectral irradiance data.  This spot’s color accuracy can be easily verified with a simple spectrometer.  The original prototype was made and mounted onto my 8” SCT to provide Solar tracking.  

    4)      Taking advantage of Kitt Peak’s great AOP (Advanced Observing Program) for the public, I did an all-nighter with their technician Roy Lorenz.  [What a fun time, an event I recommend to everyone.]  Roy was very helpful and patient as he helped me obtain images of solar twins.  The following image is of 18 Sco and uses a astrophoto technique known as progressive defocusing. 


    18 Sco

    Taking the image progressively out of focus while altering the tracking rate allows the color of a celestial object to be revealed.  Stefan Seip is a photographer who has produced beautiful images using this technique. Here is one of his images.


    5)      Using various computer color programs we can determine color given the known spectral irradiance data.  Of the few that I found, none seemed to work.  One model can be found at this site, but claims the Sun is peachy pink!  Is our Sun a girl star???  [Actually, peach pinkish is the author's color descripton, but I’m in this for the fun, as well as, truth.]


    6)      Observing the simple objects such as clouds, the Moon, and other objects.  If the Sun were, say, green, wouldn’t the Moon look green?   Wouldn't clouds look green, too?  They all look white, don’t they?  Look at the color of this astronaut: 


    Ed White


    He not only looks white, he is white….Ed White!![He was the first US astronaut to walk in space.]  [I had the pleasure of meeting him and hearing his speech.  Every time he referred to his cockpit companion by saying “Jim and I”, people laughed, but I was too young to catch the pun (Gemini).]


    7)      Finally, the best of all evidence:  The Solar image of the Sun from Kitt Peak taken by Roy Lorenz and Dr. Drew Potter.


    Suns color




    This last one came days after the all-nighter with Roy Lorenz, and while my wife slept in the Spartan astronomer’s dorm – that was also the night of our 25th wedding anniversary (thanks again Honey!).  The next day, after I had no sleep, I took her on the late morning tour through the McMath Pierce observatory, world’s largest Solar telescope.   Their main projection table had a small image about the size of a dime or nickle cast upon it.  This intrigued me and I asked what it was.  The person who was adjusting the tracking system said it was Venus.  Venus in broad daylight was impressive enough, but here was a reasonable nice disk of it, too.  He said you can do a lot when you have a f50 telescope.  Their other and smaller projection table had a nice, bright solar image upon it, though I saw no sunspots.  I was intrigued with the observatory more than I was cognitive of my original goal. On the way out of Tucson, and after some sleep, it finally hit me what I missed.  The color of the Sun was flooding down on me and I had overlooked every drop!  The Solar projection table had the proof that refuted any claim whatsoever that the Sun is yellow.   This, however, was  predicated on their telescope not using any filters.  I later confirmed with them that they were not using any filters, which could have easily altered their Solar image. 

    Remember me saying that if we knew the color of the Sun seen terrestrially that we could add the colors back into this image in proportion to those scattered by our atmosphere.  This would be rather tricky except for the fact that the Sun’s image on the table was white -  all white, from limb to center.  This makes the non-yellow claim extremely strong.  Why is that?  Well, recall that our atmosphere scatters blue light far greater than any of the longer wavelength colors.  [Rayleigh Scattering describes this and shows that this scattering is inversely proportional to the fourth power of wavelength.  Blue is not quite half the wavelength of red, and it scatters blue photons about 8 or 9 times more than red photons, depending on which wavelengths of those colors you want to compare.  This scattering effect is why the sky appears blue.]  So, if we add primarily blues to a white object, could it ever cause it to become yellow?  Nope, not no way, not no how.  The Sun ain’t yeller, indeed.


    But could it be a blushish-white star since we are adding blues to this white object?  I doubt it since the amount of blues, along with all the other colors in diminishing amounts, are not enough to cause us to see anything but white.  Part of this reasoning is supported by another eye phenomena known as color constancy.  Our brain likes to make light sources appear to be white.  Turn your car’s headlight on during a sunny day and notice how it appears yellow, but turn it on at night and our eyes see it as white. 


    Since the Sun is alone in space no other light source will trick us in seeing anything other than a white star. Yet there is more evidence for a white Sun.

    The Sun is a ball of gas and the layer that we see is the photosphere.  As we look into the central portion of the Sun’s disk we are able to see deeper into its atmosphere.  As we peer into deeper regions of the Sun, we are seeing hotter and hotter gases.  The Sun’s surface temperature is about 6,390K in the central zone.  As we look to the limb of the Sun, we see only the very top of the photosphere where the Sun is only 5,000K.


    This temperature gradient is part of what is known as the center-to-limb variation (CLV).  It is important to us because if the Kitt Peak image is not yellow at the limb, at 5000K, then there is really no hope, once again, of the Sun ever appearing yellow.  [In 4 or 5 billion years, that will be a different story.]


    But, what happens when we add those atmospheric blues into the hotter, central disk region.  Will this portion of the disk appear bluish white?  I don’t think so, but there is that possibility.  Color models do indicate that objects at this temperature should look blue.  Until someone goes up into space and uses some simple attenuator (a SAD, perhaps) we may never know.


    Nevertheless, until that day, our Sun should be regarded as a white star as it illuminates the entire solar system.    Hopefully, artists will take it from here and will occasionally portray it in its true color.  Hopefully, also, they don’t always do this since I also like seeing a colorful Sun, just not all the time, please.


    Allow me to leave you with a somewhat true color representation of our Solar system, for even the planets (especially Venus) are portrayed poorly by most others.]



    Venus here courtesy of Ricardo Nunes; and the other planets courtesy of NASA/JPL


    BTW, the fun doesn't end here. Since the Sun is still classified as a yellow dwarf, yet it is really white, now what? Pluto was an easier correction. :)

    [edit for grammar corrections]


    double post oops
    Science advances as much by mistakes as by plans.
    Interesting, very interesting, however there are some serious problems with your reasoning and conclusion.   First of all you never precisely defined what the color of a star was.  Since most objects do not emit blackbody radiation like the sun does.  In fact they are illuminated by the black body radiation of other objects.  So for most objects it can be said that the color of the object is defined by it's emission spectrum... the set of wavelengths that are emitted from it's surface under illumination.  Or equivalently the color of an object is white minus the frequencies it absorbs. Since stars emit so much visible radiation along almost all wavelength's  their emitted light will appear to be white.    Observe the dark lines in the absorbtion spectrum of our sun.  These are the wavelengths that are absorbed by the elements it is made of. It can be said then that the star's color is white minus those dark lines. Looking at this spectral data the sun emits the most consistently in the red and yellow.  This, not superstition, is why so many people when drawing the sun draw it red, orange, or yellow even when they depict high noon when more of the higher wavelength's your greens blues and violets will get through the atmosphere.  

    Basically while 90% right you are 10% wrong.

    However I like your simple approach to the problem.   Dont tase me bro
    Science advances as much by mistakes as by plans.
    Helio George
    [Hmmmm, I lost my original reply to you.] Hfarmer, you are correct that the color of a star is defined, though loosely, by its emission spectrum. However, it is not correct to claim a star’s color is white less its absorption wavelengths, no more than its color would be white plus its emission band colors. The color of a light source, if seen at a proper intensity level, is determined by the combination of all the photons of all the wavelengths within the visible band of light, as well as, the spectral sensitivity our eye – brain (retinex) has. To take any given spectral energy distribution and compute a net color seems beyond our abilities when it comes to certain stars, especially ones like the Sun which has a peak within the visible spectrum. Looking at your illustrations, these are not actual spectrums of the Sun. The absorption lines are much too clean and the color distribution is wrong; yellow is a very narrow banded color. Here is a very good, perhaps best, Solar spectrum. [Also taken from the McMath-Pierce telescope at Kitt Peak.] Notice the slight fuzz to the absorption bands and how narrow is the yellow band. This narrow band was mentioned in Part II of this article. [Go to the top of this article for the link.] Thanks for your interest.
    Lighten Up! You're made of stardust!
    Thanks for the discussion. I think Hfarmer missed the point though. Absorption lines aren't visible to the human eye anyway so how is it relevant.

    As you say the sun is certainly white as white is defined. Of course we're really talking about a black-body curve so it's not flat or uniform in wavelength.

    This gets us to the definition of "white" itself. In our human case, "white" is biologically (and species-specifically) defined by what we've evolved to. For example, other species see farther below and above our "visible" spectrum and presumable if you asked a bumble-bee what "white" is you'd get a slightly different answer in a strictly physics wavelength sense. Evolution "found" it convenient to make our biological sense of every day illumination a flat, uniform response to the extent necessary for survival without being wasteful.

    In our case we generally have our white defined by the aggregate of three (3) biological dye response curves (roughly Red, Green and Blue) in our eye. If you have a certain gene mutation there's an extra dye curve between Red and Green. Alone this doesn't fit the curve of the black body of the sun. So there must be an additional curve formed by neural nets, probably in the retina, that effectively combine the whole 3-dye curves responses to result in a "1 - blackbody" shape to a first order, ultimately resulting in the "flat response" that allows individual wavelengths to suddenly "have color" rather than the "all color" of white. Then again, that may be our brains fooling us again.

    Helio George
    Jes, it seems logical that a flat distribution of colors would represent a white sensation in our brains. I favor this hypothesis, but evolution doesn't require it since it is only a passive process. Other animal have evolved with such different visual senses, though ours is superior to many of theirs, no doubt. [This is one reason I mentioned the white-tail deer.] It might be more logical to have an evolved eye that sees an ultimate white as the actual photon flux distribution we see during midday. The photon flux I showed in the prior article (part II) is one seen from space, thus, an even steeper drop-off will be found for the blues and some greens as seen terrestrially due to the greater scattering effect for these colors. This might mean the "most white" is one that has a much greater reduction in blues, perhaps even half that of the other colors. Considering that only about 2% of our color cones are blue [receptors], that last idea is even more remarkable to suggest. This all goes to show that it is no simple task to determine the Sun's color. If the Kitt Peak image had been yellow at all, then yellow would still have a chance of being its true color. Since no yellow was found, then the Sun can not be yellow (again, as seen in space at an appropriate attenuation level). A while back, one astronomer rightly asked about the brightness of the KP projection since our eyes will tend to make bright light sources white, which I also mentioned in the article. Roy and Drew, once again, stepped in and, after I sent them a light meter, measured the surface brightness and found it to be well within our photopic comfortable zone.
    Lighten Up! You're made of stardust!
    I read somewhere that the Garnet Star, so called from its appearance in binoculars or a small telescope, appears yellow in a large, 12 inch (?) instrument. Another thing - what is brown? I've been playing with the colour picker on my favourite graphics application, and I get a respectable brown from Hue 40 Saturation 100% Value 60%. When I change the value to 100% I get orange. Yellow has a hue of 60, and when I reduce the value to 60%, the colour is more a dull green than anything else. Robert H. Olley Physics Department University of Reading England
    Robert H. Olley / Quondam Physics Department / University of Reading / England
    Helio George
    A yellow Mu Cephi through binoculars may have more to do with the binoculars than the color of Mu Cephi. Mu Cephi, one of the largest known stars, is an M-class star and and should look orange or red. These stars are cool enough to have more complex molecules. TiO is one example that is common to many class M stars. TiO, to a much lesser extent, is also found in cooler regions of the Sun (ie sunspots). Perhaps the complex molecules give it a yellow look, but I doubt it. Also, if Mu Cephi is near the horizon, you will get a redening effect caused by our atmosphere, which is why the white Sun looks yellow or orange. [Thus, even more reason for a non-yellow result. I seem to be anti-yellow these days. :)] No blackbody distribution, as typically found in stars, will give you a brown color. However, the newer class of stars known as the T-class are about the coolest you can get and still be a star. At these lower temperatures weird things can happen. One T-class star, as I mentenioned in the article, is storong in red and somewhat strong in blue. That is not unusual except that it has almost no mid-range colors. Combining blue and red makes the star appear maroon or crimson, depending on which university you might be rooting for.
    Lighten Up! You're made of stardust!
    Well the by the definition you are using all stars would be white then. Their would be no red or blue giants. Which by observation there are. My definition is all about precise, quantitative and empirically defineable things. While on the certain level that you worked at you are right, on a more fundamental level you are wrong. Dont tase me bro
    Science advances as much by mistakes as by plans.
    Helio George
    The Sun is blue! But let me 'splain... First, Hfarmer, my view does not suggest all stars are white. It all depends on the relative strengths of all those photons entering the eye. Stars, like the Sun, that have a flat distribution, will appear as white. But, not all stars have flat distributions. [I tried to keep the article short, which limited the explanation needed. Sorry for the confusion. The blue thing is coming.] For instance, consider the M-class "red" stars. Their surface temperatures are under 3500K. At 3500K the peak energy comes at a wavelength just beyond red, which means all the M-class stars will have energy peaks redder than red, if you will. Since stars are nearly blackbody radiators, then the strong "reds" must be added to the slighly weaker "oranges", as well as, the weaker "yellows", etc. It is this combination of colorful flux that gives our brain the signal that, in this case, will yield a red or orangish-red color determination. This same arguement will apply to the blue end, but the blue end is somewhat different. Recall that photon flux is the best model to use for light when our eyes and cameras are the instruments. Since "blue" photons have nearly twice the energy of the far "red" photons, it will take a larger blue-ended hump in the Planck distribution to give us as rich a blue appearance as we enjoy the richness of red in the M-class stars. As a result, blue stars are more blue-white, since the other colors are influencing us more so. It is intersting that the spectral energy distribution (SED) that we see in a rich blue sky will approximately be matched with a 15 million deg. temperature; the SED slope in the visible spectrum is very steeply inclined to blue and beyond. [The peak is near 2 Angstroms.] This condition happens to exist in the core of the Sun. Therefore, if we could safely see a properly attenuated view of the Sun's core, it would appear as a rich blue object. So, the Sun is blue, but only inside. :) George
    Lighten Up! You're made of stardust!
    Now, what colour is the Moon? Look at these PICTURES. Robert H. Olley Physics Department University of Reading England
    Robert H. Olley / Quondam Physics Department / University of Reading / England
    Helio George
    That is a great shot from EPOXI (its new name following the Deep Impact mission). I don't trust, however, the color accuracy in this image, but I do trust the relative brightness between Earth and the Moon. The Earth's albedo is 0.37 (avg.) and the Moon's albedo is 0.12, or 1/3 the surface brightness of the Earth (given the same light source). The Moon rocks seem to be gray, but there are varying color regions on the Moon. On one Apollo mission, they found on a crater's rim, IIRC, a very bright orange region. The image of your link, nonetheless, is very remarkable. Have you seen the animated version? Animated Lunar Transit
    Lighten Up! You're made of stardust!
    On page 112 of Sky & Telescope Observer's Guide Series "City Astronomy" (1994) by Robin Scagell are the words: "The Sun is actually pure white in color, with a slightly yellow limb. ... If the Sun appeared to us as yellow, we would never see an object by day as a white, for it would be the color of Sunlight -- yellow. ..."

    The author has some more words about this but I don't see that they are needed for this point.

    The argument by the author appears to remove the point about how very bright lights overwhelm our eyes and cause us to see white when the light isn't white. Another point is that I have attenuated the Suns light with a foil-like solar filter when looking directly at the sun with either my binoculars or telescope or reflected off the moon with a gray or polarising lunar filter with my apochromatic 4" refractor. In every case the Sun was always white. (The highly reflective portions of the moon would be white and other parts would be gray or even brown, but nothing yellow.)

    I believe their are many errors in astonomy and cosmology. For some other big ones visit my homepage listed above.

    I didn't see this science blog showing my website and since I referenced it here it is in the text

    Helio George
    Hi Vince,

    Yes, there have been many assumptions made within astronomy that have been shown to be incorrect.  The Sun's color is an odd error by many since it is taken more as a subjective issue and not highly relevant to advancing science.  They have accurate data on the Sun's spectrum, mass, size, composition, temperatures, etc., and color accuracy just seems to have fallen through the little cracks.   Other stellar colors also could use some revision since red giants are more orange than red, and blue stars are never deep blue but bluish white instead.

    Thanks for the 1994 S&T article.  It is rare to find any claims for a white Sun from back then.   The arguments stated that the Sun must be white because white paper looks white is not that great an argument, however.   If the Sun were a yellow star and sunlight were yellow, we would still see any white paper look white because the Sun would still be emitting all the other colors, too.   So the paper could be, say, a bluish-white paper as seen in white light, but would appear white in yellow sunlight.

    When I glance at the Sun, I see a yellow rim around the Sun, though it is unclear why most people see this.  Regardless, the white Sun claim I make is on the basis of us seeing the Sun from space and appropriately attenuated to avoid the blinding affect you mention.  In this case, the limb region will still be white because there is essentially no yellow in the limb seen terrestrially  (see the Kitt Peak image).  Since our atmosphere removes more of the blue end of the spectrum, then if the Sun limb were indeed yellow, it would appear to us more prominently yellow (or yellow-orange) in imaging taken terrestrially.

    You have an interesting site, btw.  I would recommend you give the forum a try as you will find many to discuss your views.

    Lighten Up! You're made of stardust!
    Yellow rim around a white object in a blue field?

    Could it be that your tired retinas are after-shadowing yellow at the line between "OW! Bright!" and "blue?"

    The sun-ward parts of the sky are often tinged with warm colors on the horizon, but even in daytime the sky can bleach a sort of platinum-blonde around the sun. Apparent contrast, maybe.
    The atmosphere's diffusion of blue light away from the sun (which is responsible for the color of the sky) also leaves an area along the direct line of light that's slightly richer in colors-other-than-blue.

    Maybe it's not the sun being yellow that's the misconception:
    Maybe it's imagining the sky is blue. The sky is only blue against dark space or dark hills, at a wide angle to the sun. Against the sun (like a "sky eclipse"), or on shallow angles from sunward, the 'sky' color is red-orange-yellow-green, or brassy-white.

    See Goethe's theory of color, or Aerogel, depending on which century you prefer.
    Watered-down skim milk with a light shining behind & then from the side will also give you a pretty good analogy.

    Before you get all excited to tell me the color of the sky, can I distract you with another 'white' object - the moon?

    I want to see the differences between solar and lunar light spectra, and my budget is about what you spent on choice #1.
    I hear that there are some additional colors absorbed by the moon, which have to be factored in when using moonlight as a proxy for sunlight in astronomical calibrations.
    Is that just theory? or does someone have a nifty chart like the one of the Sun, showing the differences between direct sunlight and moon-reflected light?
    It's the differences that interest me, not the aggregate "white" effect.


    Helio George
    Hi Erica,

    The best comparison between the Sun and Moon can be seen in comparing their spectral irradiance.  The best place I know to get sp. irr. data for the Sun is:  

    No doubt, there is some sp. irr. out there for the Moon, but I don't have a hand link for you.

    Since color is the product of the luminous source, spectral reflectance of the object, and our eye-brain's spectral sensitivity, the Moon will be a little different than the Sun since it's spectral reflectance properties come into play.  I do expect you will find a difference even though the Moon is considered generally gray in color.

    "Why the Sky is Blue" (Hoeppe, 2007) is an excelent book on the sky's color and Goethe's color theory is addressed.   The best model we have today is found in Rayleigh Scattering, which explains your milk in the water experiment nicely.  [I think, however, the color fringes we have mentioned might be handled fairly nicely by Goethe's theory.]

    I happened to be looking at Betelguese early yesterday morning with an expensive pair of binoculars (not mine).  I defocused to get a better idea of its color and noticed some off-color outer rings, which I attribute to the lens coatings.  Betelguese was yellow-orange after defocusing, btw, which did not surprise me even though it is a "red" giant.

    Thanks for your comments. 
    Lighten Up! You're made of stardust!
    Helio George


    You raise an important aspect of science... definitions.  We simply assign the visual "green" sensation, for instance, to wavelengths that cause that "green" effect.  This can me a monochromatic green laser, or, in this case, it can be a sensation of green caused by several combined colors (i.e. metamers).

    Once we establish an agreeable definition, then the sky is either "blue" or its not, with some exception for certain people that have different color sensitivities, though we all likely have some minor variations, but within our color definitions.

    These colors can be measured objectively by spectrometers, which gives us an objective base to work with.  Thus science can easily become involved with colors. 

    On the other hand, if we are talking about "pink elephants" or "good art", where spectrometers are of no effect, then this is a subjective-base view and typically outside the purview of science.  

    The true color of the Sun is very much within the purview of science because of our color definitions and ability to measure the Solar spectrum, as well as, reproduce the Solar spectrum for our eyes to see and our brain to decide.  If the result of such a reproduction creates color variations for many observers, then the Sun's color would remain in question, yet this article attempts to show why such variation is highly unlikely. 


    Lighten Up! You're made of stardust!
    The reason 5) is pink is because it uses the 6500K, blue sky, web standard, D65 white point. Which is noticeably blueish. A white object rendered with a blueish white point looks redish.

    Just as a white object rendered with a redish white point looks blueish. As when a 5000K, incandescent light, print standard, D50 white point is used.

    Which is why I encourage people doing sky map software, and astronomy outreach, to use a non-standard white point of 5800K, instead of D65. Because the user compares star colors to the screen's white, not to their room. To a white window on their laptop, not to a blue window in their wall. So a G2 star is best shown as white (D58), instead of as pale pink (D65).

    An extreme case of white point deflection is using Vega. The Sun is yellow relative to Vega. And to the Cookie Monster. One is a blue spectroscopic reference star. The other a blue muppet. Neither makes a great white point. The Sun is also blue relative to yellow Big Bird, purple relative to green Kermit, green relative to a purple dinosaur, and cyan relative to red Grover.

    But yes, the Sun is white, and it would be nice if astronomy content stopped using the Cookie Monster as a white point.

    Helio George

    Thanks for the explanation.

    My no. 5 point is partially meant for colorful hyperbole as I suspected the peachy pink color determination has more merit when not in the narrow conditions of simply observing the Sun from space, attenuating its intensity, and seeing what its true color is -- and assume that 100 others doing the same thing would agree with the color result.

    Your reference to Vega is an important one, I think.  I suspect the "yellow Sun" popularization, at least for the scientific community, came as a result of the spectral color comparison of the Sun relative to the hotter important reference star Vega.  Perhaps Crayonic adolescence is responsible for the rest. *wink*

    I was reading George Gamow's book today on the Sun and noticed that in his chapter on star colors he gave a couple of examples of stars to illustrate the range of star colors: reddish to bluish.  He mentions yellow as an intermediary color yet never mentioned any star of this color. Considering his book is specifically about the Sun, the omission is striking and may be support for the more spectroscopic "yellow" assignment for the Sun.  [I haven't finished reading his 1940 book, admittedly.]

    I think somewhere in this old, 3 part blog, I mentioned that my interest piqued when I read an article by and East coast astronomer claiming the Sun to be green (off-color green). Perhaps he's a Kermit fan. :)

    Lighten Up! You're made of stardust!


    Part of the problem is one both you and Mitchell Charity have alluded to, or walked around, depending upon how one actually interprets what you two have been saying.  Professional visual artists know that the "color calibration" of their computer monitors are critical for getting the fine results they shoot for.  Additionally, they also have recognized since even before the use of computers, that our color perceptions are greatly influenced by what colors we see nearby.

    As a result of the former, a disk or field of color, calculated to represent the "true" color of our sun, as seen from space, can easily look "pinkish" on an uncalibrated computer monitor.  As a result of the former, one can look at pink lighting (given by white and read floodlights, for instance) and see white (while seeing the shadows cast by the read floodlight as green).

    So, even a slight yellow sun will appear "white" when that is all one is viewing.

    However, to help in showing some of the variety of star colors, check out the Kepler Transiting Planet Candidates.  The linked version has "The colours of the stars are meant to represent how the eye would see the star outside of the Earths atmosphere."  Of course, if your monitor is not properly calibrated, and/or you have too much "background" color near your monitor, what you see/perceive will not be a correct representation.

    Anyway, just trying to help calibrate things, here.


    Helio George
    Hi David,

    Thanks for your calibrated input.

    Background colors should indeed be taken into consideration.  Many people suggest the Sun has a bit of a  yellow limb when it is relatively high in the sky, compared to a yellow setting Sun.  Others don't see much of a yellow edge.  Some argue that the blue sky background creates a yellowish effect adjacent to the edge (limb) of the Sun.  This seems to be in line with your argument.  [Warning for the casual reader: Don't look at the Sun directly.  Retina damage can easily be permanent.]

    Seen from space,  and with the naked eye, the background color is pitch black, so color contrasting effects are, I assume, not a problem.

    Your point that even a slight yellow sun would appear white is important because our brain has a color constancy function that tends to make the brightest light source a white source even if it is not.  The example of a car's headlight, mentioned in one of my three blogs, demonstrates this effect.  But the key point isn't whether it is slightly yellow and looks white, it is that it looks white, period.  If a light source always appears white, then by definition, it is white.  Now if we grab a hot 0 class star, which is very bluish white, and place it next to th eSun, then, as you mentioned, the background light, which is now the brighter source of light, may cause the white Sun to turn a bit yellow, maybe.  One star is sometimes seen as green for this very reason, yet there are strong arguments why no star can look green. 

    If you are familiar with color calibration, consider the industry color estimation for light sources that are almost 6000 kelvins.  Often bluish white is the color estimation, but this too is wrong, and partly for the reason of color constancy, perhaps.

    The Kepler transit chart in color is, in my opinion, flat wrong, and I mentioned this to them when it first came out.  I believe their values came from a source known for a peachy pink calculated result out of Colorado. 

    F and G class stars will be white if seen with the naked eye, from space, and at a reduced intensity level.

    In case my arguments in this column were not enough, you might enjoy hearing that astronaut Don Petit has heard my plea and has planned an experiment aboard the ISS, which he has recently boarded.  His plan is to do a pinhole projection, and he will support his color finding of what he sees with shots from a new standard-issue Nikon camera that are aboard the ISS.  I believe this will be the first time ever that mankind has looked at the Sun at a proper intensity level without the color distortion caused by our atmosphere.  The WYSIWYG can't be better than this.
    Lighten Up! You're made of stardust!


    It would have been preferable if you would have actually set your reply to my post as a relply.  Then I would have known about your reply days ago, rather than just seeing it because I thought to take a look here.

    You say

    ... But the key point isn't whether it is slightly yellow and looks white, it is that it looks white, period.  If a light source always appears white, then by definition, it is white. ...

    If what you want is to define "the Color of the Sun" in terms of human perception, including all the well known (and, potentially, yet to be known) inaccuracies, and "foibles" of this easily fooled system, then so be it, you have done a fine job.

    However, if one want's to be scientific, and independent of the easily fooled human perception apparatus, then there is much yet to be desired, here.

    Later, you indicate:

    In case my arguments in this column were not enough, you might enjoy hearing that astronaut Don Petit has heard my plea and has planned an experiment aboard the ISS, which he has recently boarded.  His plan is to do a pinhole projection, and he will support his color finding of what he sees with shots from a new standard-issue Nikon camera that are aboard the ISS.  I believe this will be the first time ever that mankind has looked at the Sun at a proper intensity level without the color distortion caused by our atmosphere.  The WYSIWYG can't be better than this.

    Once again, is the thrust of this "Quest for the Color of the Sun" seeking the human perception answer (in which case, anything from pink to greenish can appear white, under very simple and common circumstances), or a more scientific answer?

    You see, looking at "a pinhole projection" of the Sun, whether in space or on the surface of the Earth, is guaranteed to appear white (unless one has a "white standard". there, for the human observer), due to the nature of the human perception system we have already discussed.

    As for taking a picture with "a new standard-issue Nikon camera that [is] aboard the ISS":  Is the camera carefully color calibrated?  That includes the picture files produced therefrom (including the final version shown to the public).  Then, there's the question of the calibration of one's computer monitor, and the nature of one's surroundings when viewing such (including the color of the lighting in one's home or office [which is often taken into account when choosing the nature of the calibration of one's monitor, though, for commercial uses, there is often the additional consideration of the color of the lighting the consumer will be viewing within]).

    So, once again, is the "quest" for scientific "color", or human perception?


    Helio George

    Thanks for the "reply" tip.    Let me know if this doesn't notify you.

    ...However, if one want's to be scientific, and independent of the easily fooled human perception apparatus, then there is much yet to be desired, here
    If we restrict the definition of the Sun's true color to be what the vast majority of observers would claim the Sun's color would be when they were viewing it at both a comfortable intensity and from dark space, thus without distortion by color altering media, such as our atmosphere, then we have a reasonable determination of the Sun's color. 

    It is unclear to me by what you mean by a more scientific view of the Sun's color.  If a computer program, or other scientific approach, takes the spectral irradiance data provided by any of the space instruments, and calculates a color result that is distinctly different than that constantly seen by a vast majority of  people -- assuming the object has only a black background and is the only luminous source of light, thus limiting the ambiguity that comes from other effects with the exception of color constancy  -- then which system is in error: human or non-human?  The color constancy effect only boosts the white determination, but this is fair given the context in which I frame the conditions to establish the Sun's true color.  Is this not reasonable?

    The peach pinkish claim from U. of Colorado [], which is likely the source of information used for the excellent, but chromatically flawed, exoplanet/star size montage, seems to be based on the color the Sun relative to a D65 light source.  There is, however, no reason the Sun would appear to the human eye to be peach pinkish since there is no D65 light source near the Sun to allow the peach pinkish color view.  The color temp. of D65, ~ 6504K, is a temperature that exceeds all surface temperatures found on the Sun.  The limb is around 5000K and only the center region comes close at 6390K.  It may be very useful for industry to use the D65 standard, but I fail to see its application in our quest for the Sun's true color.

    If, however, we want to put the Sun's color on a monitor that uses the D65 standard, and place it in this D65 "white" background, then perhaps the Sun will look a bit peach pinkish.  [If we were to use a 6500K Planck distribution as the standard background color temp., then I think this could be erroneous given the specific deviation of the Sun's spectrum from a blackbody, especially as found in the blue end of the spectrum, which would make it appear more bluish than it would pinkish.  I simply know too little of the industry's techniques to know if this is the logic used for the color temp. standards.]

    What should be interesting to scientists, and some have commented on it, is just how flat the photon flux distribution is for sunlight (as seen from space).  Perhaps a perfectly flat photon flux distribution across the visible spectrum could someday be the best definition for "white", maybe?

    The montage of the stars with their respective transiting exoplanets is a clear example of how color fallacy is still propagated, even if unintentionally.

    Seeing the Sun from space in its natural dark habitat, if you will, where no atmospheric extinctions alter the color of sunlight, and at a greatly reduced, and spectrally unaltered, intensity should be the most fair method in determining its true color.  Is this not reasonable given the specific colors system (retinex) we have is a human perceptional phenomena?

    Lighten Up! You're made of stardust!
    Helio George

    I meant to comment on this as well because it may help show the argument you may be making...
    As for taking a picture with "a new standard-issue Nikon camera that [is] aboard the ISS": Is the camera carefully color calibrated?

    My recommendation was that he set the color temperature to around 5850K, which is close to the Planck "overlay" temperature for the Sun's color, and more accurate than the one derived from the Solar Constant conversion.  If the color temperature is much lower, then the camera image will give the Solar disk a bluish tint.  If the color temperature is higher, then the Solar disk could look yellowish, especially near the limb.

    If, however, we tell the camera what the color temperature is for the light source, and the light source is the object we are shooting, then we should, I think, get a very close color representation of the color observed by Don Petit, right?  [I'm not a professional at this at all.]

    Lighten Up! You're made of stardust!


    Thanks for the replies.  So, according to your responses "what you want is to define 'the Color of the Sun' in terms of human perception, including all the well known (and, potentially, yet to be known) inaccuracies, and 'foibles' of this easily fooled system, then so be it, you have done a fine job."

    So, rather than something more scientific and observer independent, you are, essentially, guaranteeing a "white" observation, as I have already pointed out.

    That's OK, but it is of less interest to me, because it simply plays on well known inaccuracies and "foibles" of the easily fooled human observational system:  One that is actually "designed" to see "white" in any light from pinkish to greenish (which is actually an advantage, when trying to hunt and/or survive in the variety of lighting conditions on this planet).

    I was, however, hoping for something more scientific, but, whatever...


    Helio George
    The best science view of the Sun's color, or any object, can be found by analyzing the SED (spectral energy distribution) of that object.  This presents the specific energy level we observe for each wavelength.  [Spectral irradiance is used for the Sun since it is an extended object, and not a point source.] 
    Knowing an object's SED and it's brightness, it's color can be determined from the various models science has developed.  You may see, for instance, a comparison of the D65 SED with an actual SED of the Sun.  it's important to note that the SED of the Sun varies hourly when measured here below our atmosphere.   Its SED in space, fortunately, does not vary over time, at least in the visible portion of the spectrum.

    Since, however, other surrounding light sources can easily affect the color observed, some consider the  color of the Sun, or any color, to be rather subjective.  For this reason I qualify the Sun's color to be as we would see it from space, and at a comfortable intensity.  This not only eliminates the atmospheric color distortions but also places it within a black background, eliminating other color distorting effects.  This allows for an objective color determination, one that a multitude of actual observers would agree upon.

    As you know, such conditions will favor a white result do to color constancy.  So, perhaps, a more scientific approach might be to determine its color if we could eliminate the color constancy effect.  I think this more to your question.  I would assume we could simply compare, as mentioned before, the Sun's SED, as seen from space (AM0), with the various color models known.  The intriguing thing I see is just how flat the SED is when we convert to a photon flux distribution.
    Lighten Up! You're made of stardust!


    You end with:

    As you know, such conditions [the Sun's color to be as we would see it from space, and at a comfortable intensity] will favor a white result do to color constancy.  So, perhaps, a more scientific approach might be to determine its color if we could eliminate the color constancy effect.  I think this more to your question.  I would assume we could simply compare, as mentioned before, the Sun's SED [spectral energy distribution], as seen from space (AM0), with the various color models known.  The intriguing thing I see is just how flat the SED is when we convert to a photon flux distribution.

    Yes.  This is "more to my question."

    I will look into the correctness of your AM0, when I have sufficient time.  I've seen many other spectra for the Sun, out in space, that don't show quite such a rapid falloff in the high energy end of the spectrum.

    Oh, one quibble.  You said "Its[the Sun's] SED in space, fortunately, does not vary over time, at least in the visible portion of the spectrum."  However, even within the visible portion of the spectrum, the SED of the Sun does vary with the Sunspot activity, even if filtering out all the actual Sunspots (IIRC).


    Helio George
    Your interest has spurred me to consider doing another piece on this topic that is more scientific. The three separate pieces I did were done in a corny style because I have had a lot of fun playing with this topic. So, look for another blog from me when I get time, which may be in a couple of weeks. I'll try to notify you when done. The prior blog ( ) has a graph that illustrates the significant difference between the Sun's spectral irradiance based on wattage vs. its spectral irradiance as a function of photon flux. Since E=hf and blue light is nearly twice the frequency (half the wavelength) of red light, then the blue end of the spectrum will have almost half the number of photons compared to the number of red photons for any given energy. The impact this modeling has on the solar spectrum is illustrated in the linked blog graph. What looked to be a very blue-favored spectrum is actually a mildly weak blue-ended spectrum when using the photon flux density model. Since the photon flux model better represents what the retina is responsive to, then it becomes more clear why white is the likely result since the spectral distribution of photons is so flat. I am doing a "heliochromology" presentation soon, so I should be able to produce a much less corny blog soon.
    Lighten Up! You're made of stardust!


    Yes, I'm familiar with the Solar Spectral Irradiance plot in your previous article.  It's that plot to which I was referring.

    I'm also quite familiar with the transformation to photon flux, and quite comfortable with how that transforms such a plot.

    It's the Spectral Irradiance plot (before transforming to photon flux) that looks possibly questionable, compared to other spectral plots I have seen (before the "filtering" of our atmosphere).

    By the way, I did recognize that these articles "were done in a corny style because I have had a lot of fun playing with this topic."  Unfortunately, this is why I felt I needed to ask about your "definition of the color of the sun".  It was just somewhat unclear due to the "corny style".

    Anyway, I'll look forward to your new article.


    Helio George
    Here's another "Reply to this" attempt after logging-in...

    [Another quirk is that, after using Word, I am unable to create paragraphs. Ug.]

    You are correct in noting the less than normal looking spectral irradiance curve in the prior blog.  I used the Thuilier 2002 data set at the time I first did that graph.  Their data set is in ~0.3 nm increments, so I interpolated that data to convert to 5 nm increments to make for a nicer graph.  Also, I superimposed it onto a graphics softwared package to get the nice coloring presentation.  I was more interested in the accurate placement of each color for the wavelengths than I was in the more eratic energy values.   [I have seen a number of scientists mistate solar spectral peaks.  One will say the peak is in the green -- this is only true if you assume a perfect Planck distribution for a solar temperature like 5777K, though it is borderline cyan at this temp. -- and others have said its peak is in the yellow.  The graph shows the peak(s) are in the blue, which is correct but the plot has clipped them off a bit.   Regardless, these tweaks to the data causes the slope to look much more linear than it normally would.  So, your scrutiny of the plot has revealed a glitch that I ignored.  I'll make sure the plot is more accurate next time.  Thanks for the heads-up.

    I've found that SORCE [ href="" ] has almost daily sp. irr. data sets since about Oct. of 2003.  When doing a few quick comparisions, there is very little or no variation in energy changes during the Solar cycle, at least since 2004, within the visible spectrum.  The variation is often stated to be about 0.1%.  The far UV portion does have a greater variation, but it is not as much as one might expect.  Of course, terrestrially, the UV sp. irr. does vary, probably due to the very thin ozone layer changes from time to time.

    For me, the story I enjoy telling is more about the adventure than the results.  The color result is certainly no discovery but it is something that seems to be stuck in a dark well, and I find this fact to be both ironic and entertaining, hence the corn.  I'm an amateur, and my hope is that others will be encourage to jump into things of interest to them whenever they see something they can enjoy, as well as, use as a way to engage the amazing world of science.
    Lighten Up! You're made of stardust!


    Every time your reply has actually shown up as a nested reply to one of my messages I have been properly notified.  Every time your reply has shown up at the top level (leftmost side) of the postings, the way a post that simply uses the "ADD A COMMENT" area at the bottom of the page will appear, I have received no indication that there was any reply whatsoever.

    It mattered not one iota whether you were logged in (like the message I am replying to), or whether you were not logged in, as below.  In both cases, I was notified, and you can see that they are nested under my messages.




    By the way, why didn't you use the "Reply to This >>" link for your latest reply?  I only saw it this evening because I had the "wild idea" to check here.


    Hi David,

    Thanks for clarifying your thoughts on the sp. irr. info.

    I suspect there may be a system problem because I have been using the "reply to this" button to respond. Perhaps my loging-in or my host status is altering the notices to you.

    This response is via the ananomous route to test this idea. My next post will be using my normal logged-in approach.

    Please let me know which, if any, notice you receive.


    Hallo ! I think that this page is very interesting and scientific . So the Sun is really a very bright ("blinding") cool white to our eyes viewed from space .


    I think perhaps you misunderstood the effect of Rayleigh scattering in the atmosphere. If the Sun were pure white light (i.e. approximately flat spectrum in "photon flux", as your plot showed, and as your Kitt peak photo confirms), then you would expect the stronger Rayleigh scattering of blue light in the atmosphere to add a *yellow* tint, not a blue tint. This is because the scattering *removes* more blue light from our line-of-sight to the Sun than non-blue. The blue sky is caused by scattering of blue light *in* to our line-of-sight from the atmosphere, but in the direction of the Sun this is negligible in comparison to the effect of removing a (blue) fraction of the much stronger sunlight due to scattering *out*. It's basically the same effect which produces a sunset.

    "Remember me saying that if we knew the color of the Sun seen terrestrially that we could add the colors back into this image in proportion to those scattered by our atmosphere. ... So, if we add primarily blues to a white object, could it ever cause it to become yellow? Nope, not no way, not no how. "

    This makes it sound like you think the scattering would add blue light to the image. But this is not true, it would actually add yellow to the image, as I explained above.

    So there is no problem! The color of the Sun seen terrestrially is yellow-ish, and your Kitt-peak plate is white, thus the yellow light was added entirely by the effect of atmosphere scattering, which is what we expect Rayleigh scattering to do. Your results are pretty strong evidence that the apparent yellow color of the Sun in the sky is almost entirely due to atmospheric scattering.

    How astute and keenly visioned is the audience?  4.5 years after this was published, a reader wrote on the site contact form (comments were closed, obviously, and I opened them manually to leave this one) that the graphic with the Natural Colors of the Suns and Planets is wrong - the Earth is backwards.  To see it (can you see it?), you have to open the image and then zoom in. I thought NASA/JPL created that but if they are getting the planet backwards, we are all in big trouble.