76 comments

[ 2.9 ms ] story [ 143 ms ] thread
It definitely doesn’t help that astronomy students are taught that the sun and sun-like stars are called yellow dwarfs, and almost all visualizations of stellar types associate spectral class G with yellow as well…
Sapi-Whorf strikes again!

Seems like a name change is in order.

"White is unambiguously the correct answer."

Hmm, isn't sun wavelength actually closer to very very bright green?

My understanding is that the spectrum follows the black body radiation curve with a temperature of approximately 6000K. Peak color is green. White is closer to the truth IMO because it is a combination of multiple colors, but still misses the mark because how does "white" contain UV, radio waves, infrared of very different magnitudes?
The sun is not a monochromatic source. It's an incoherent, broadband source. The power spectrum distribution of those wavelengths pretty-closely follows the black body radiation curve. If you had a monochromatic source at that peak it would look green.
The sun doesn't emit a single electromagnetic wavelength; the light it emits instead broadly falls on a, uh, spectrum, including wavelengths that are either too large or too small to be perceived by our eyes.

It makes the most sense to call that white. Although, the wavelength of peak intensity is a green.

The spectral peak is blue-green but the emission is relatively even along the visible light range, so if any color is perceptible it's only a very slight tint. The atmosphere scatters blue light more than other colors so direct sunlight viewed from Earth at noon is even more balanced and pure white.

https://commons.wikimedia.org/wiki/File:Solar_spectrum_en.sv...

The color of something is not defined by its peak radiative frequency. I can easily create a spectrum that has a peak in blue but looks red, by having a lot more red overall but not at one frequency.
The way I think of it is that our vision system has evolved to work in sunlight.

Therefore the sun looks "neutral" to us. Which is "white".

Or at least sunlight does. The sun itself looks like "pain" :)

Thank you! This is the response I came to find/make. The sun's color, when observed by our eyeballs, is "pain" or "ow" or "I can't tell" or something like that, exactly. The sun's color is a recursively unknowable concept which demonstrates the permanent disconnect between us the observers and reality.

This is a Carnap-level kind of interpretation, but it's so very important if we're going to be scientific to be understanding of the limitations of our builtin sensors and the resulting limitations of our modelling.

Or perhaps the best way of answering questions from five year olds isn’t causing these people who know about spectra and wavelength-response to go into pedantic and excessive detail?

If you answer “white” then the next question is “why does it look yellow”. This article seems to presuppose that there is a “correct” way to answer a five-year-old that will prevent further questions.

It would be a great time to get a little glass prism and show him that it really has a lot of colors in it!
(comment deleted)
Back in the age, seeing the color spectrum rolled out by the prism made me conclude that it's the prism that colored it...

As odd, as it may seem, the presence of the object stole the attention away from the subject. Only later, learning about wave refraction/reflection on the media border has finally elucidated, so to speak, the priciple and placed the origin to the property of the sunlight.

It could've been summarized as "I believe in Prizm" :)

A bit embarrassed to ask this, but could someone ELI5 - why does it look yellow? Or rather, why is this a common perception? Why is this the colour kids reach for to draw the sun?
Sunlight shifts reddish as the sun sets and its blue light is Raleigh-scattered away by a thicker slice of atmosphere. This also makes it dim enough to look at directly. At its zenith, when the sun is whitest, it can't be naked eye observed, so the balance of subjective impressions is formed by late-day observations when its color is shifted toward yellow/orange/red.
Very fancy way of saying “because people only look at it at sunset”.
(comment deleted)
Green Blue and violet wavelengths scattered by the atmosphere
Well all the other children and past children have drawn it yellow, so it is a part of the culture of children's drawings. The culture really does matter, for example young children in different cultures draw human figures quite differently. As someone else noted in these comments, you can't draw white on white paper well, and yellow is the closest thing really in a box of crayons, so that probably has a large effect as well.
My guess is that since the sun is too bright to look directly at, the yellowness is some effect from overloading the eyes and/or looking at it peripherally.
Good guess. Near as I can tell, the sun's yellower at sunrise / sunset, and I can almost look at it during that time, so I imagine that selection bias is involved.
> then the next question is “why does it look yellow”

Then you say it doesn't look yellow at all, outside of sunsets.

You just think it does because that's what you have been taught to crayon at school. In the normal sky it's white.

In Japan they get taught to color it red, so kids there think the sun is always red.

This seems like an ambiguous question deriving from the fact that "color" is a bad representation of light outside of a small window of power and frequency.

What color are x-rays?

> So why do astronomy graduate students not know what color the Sun is?

Without any proof, I will offer a guess: Because it is the wrong question!

If you ask "What is the wavelengths of the light than the sun emits?", you are setting an expectation of an accurate answer. "What color is the sun?" seems a casual question that expects a casual answer.

> White is unambiguously the correct answer.

Obviously false without knowing at what relative speed and in which direction the Sun moves relative to the observer. There are always different degrees of accuracy and expectations need to be set to get the correct answer.

White is often defined to be the color of sunlight that reaches the earth’s surface (go look at lightbulb specs to see what I mean).

That accounts for any relativistic and atmospheric issues.

But diffusion through the atmosphere can change that color. For example, at sunset or sunrise.
Lightbulb specs often describe their white color in terms of color temperature, in Kelvin.

This refers to perceptually matching the color of an ideal black body glowing at the specified temperature.

The documented color temperatures of different lights vary considerably. Rarely do they match sunlight. Daytime, overhead sunlight at the Earth's surface has a bluer spectrum than most lightbulbs, but sunlight doesn't look bluish-white because color perception adjusts rapidly for ambient light, and sunlight is so bright it dominates even when it's seen indoors through a window.

Color is an ambiguous term, but most people interpret it as it is used in human visual perception. Under that definition, the author is wrong. Outside the atmosphere, it will be perceived as white, but here on Earth, it will vary from yellow to red. The environment that the person receiving the question can assume is the environment they are asked the question in. Therefore, the most popular answer of yellow is correct.
> here on Earth, it will vary from yellow to red

If that were true, how would we ever see blue things?

By looking at things our visual cortex perceives as blue, typically by activating the blue photoreceptors more than others relative to objects that are not blue but also via negative after images and a number of other methods. Look at these blue hearts, for example: https://cdn.cpnscdn.com/static/blog/2015/02/greenblueheartss...

You seem to be making the mistake I specifically warned against of defining color by the distribution of wavelengths coming from the object. Once again, the definition of color is ambiguous, and most people will use the human perception definition. Purple things don't have to be composed of pure violet light.

What activates blue photoreceptors?
Electromagnetic radiation with wavelengths between 380 NM and 550 nm. What point are you getting at? It is as yet unclear whether you understood my point.
I'm not making a point; I'm asking a question. If the light from the Sun were "from yellow to red", what would trigger our blue photoreceptors?

And no, I don't understand your point. I don't see what it has to do with yellow-red light.

Agreed - this is the all-too-common willing ignorance of colloquial semantics to pedantically demonstrate a non-point.
Trying to correct your answer to the question without reading the article.

If the Sun would only emit one wavelength, you'd need an artificial light source to get white (R+G+B) light.

The answer white is correct in regards of the visible light spectrum. But you need to add infrared and UV to it.

Red and blue shift is relevant for objects that move relatively to us observers. E.g. other stars within our galaxy, or galaxy relative to other galaxies.

Since we orbit the sun relatively slowly you won't notice significant red or blue shift here.

This subject isn't really related to physics or astronomy except in a kind of shallow sense.

It has way more to do with the biology and psychology of how we perceive and interpret color (which is an incredibly in-depth and complicated subject!)

Overall the post gives off huge crank vibes, and browsing around the rest of their site does nothing to disabuse me of this impression.

What a bad article.

Color is hard because of broad spectrum sources, atmospheric scattering effects, biology effects (2-4 wavelength receptors in human eyes), and human psychology ("auto white-balance"). This leads to ambiguous language.

The article does nothing to shed light on those issues.

(comment deleted)
I just asked my 6 year-old this question.

— What color is the sun?

— White. (pauses...) But some people color it yellow.

— Why is that?

— I don't know, because it's a bright color. (shrugs and runs off to play)

It dawned on my that when drawing on a white piece of paper, it's not exactly easy to draw a white object without resorting to inverse video (coloring the blue sky in such a way as to leave a spot that represents the sun).

In the blue sky, the sun will be perceived as yellow, not white. Painting it as white would be just as wrong as painting grass hot pink. https://eclipse2017.nasa.gov/what-color-sun
From the nasa link:

> * it emits most of its energy around 500 nm, which is close to blue-green light. So one might say that the sun is blue-green! *

Fantastic!

https://www.lrc.rpi.edu/programs/NLPIP/tutorials/photosensor...

So human eye is more sensitive to the same wavelengths the Sun emits more! So it all evens out and the Sun only appears white! That's why we define white as white as this is "lit by the Sun / reflects all sunlight" - our only source of light for many years. So if e.g. dog's eye is sensitive to different wavelengths it should see the Sun in different color? But then again, a dog should call it "white" as this is what you would call "lit by sun" as opposed as e.g. "lit by fire".

Fascinating!

More likely the only time that people look at the sun is at sunset because it’s so bright. But at midday it’s white.
It emits white light, but that's only one definition of color. Normally when we talk about color, we talk about reflection and absorption. If we shined a super sun on our sun and wore super sunglasses that block out all of out sun's relatively dim light, what would we see reflected back by the hydrogen-helium ball?
Not much. The sun is a blackbody. (But if you discount blackbody radiation, instead of just trying to out-shine it, that's an interesting question.)
Because it is a stupid question.

Somebody asks what colour is the sun, the follow up should be why? Are they going to draw the sun on a piece of paper? Then expect a traditional red sun. Or yellow. Nobody draws the the sun as white.

Note that the sun is red, at dusk.

Only when there is a lot of smoke or dust in the air.

Any time you can actually look at it, otherwise, it is very distinctly yellow. When you can't look at it, it doesn't have a color, because duh.

Video game designers often draw the sun as white (except when the scene is intended to be near sunrise or sunset). I just doublechecked all the skybox textures I have.
The original question was posed by a 5 year old. How many 5 year old videos game designers do you know?
The word everybody here is looking for is, "pedantry".
> White is unambiguously the correct answer.

OMFG. It most certainly is not. "White" is a very good explanation for the sum of all visible light that reaches the environment on the ground.

But if you look at (or really towards) the actual bright thing in the sky, you see it as yellow. Preschool kids and amateurs aren't wrong on this. It's yellow. Go look.

I mean, even that's not correct. The disk of the sun itself is too bright for your retina to actually sense, saturates the cells, and leaves weird artifacts (and no small amount of permanent damage, of course) that look like all sorts of colors.

But the glare around the sun that you see before you look away is, for sure, yellow. The shorter wavelengths have been preferentially scattered away on the direct path to your eye. They are the "blue" that someone else sees when they look to that path in the "sky". What's left... is yellow (relative to the general color balance of the non-direct light, that is).

Sure sure, "solar irradiance" is as good a definition for "white" as you'll find. But "the sun" is yellow.

Note the question in the article is not the question in the HN title. It's « I'm told the Sun is a big hot ball! Awesome! I love balls! What color is the ball? ».

Under « Methods - Question design » in the article:

« The "ball" helps clarify intent - 'object, viewed from space', not 'eyeball observation made through atmosphere'. »

The title is from the article, not the HN submitter. Clearly "What color is the Sun?" reflects the question they want the reader to have in mind, not the specifics of the question they bury down at the end.

So... I don't get it. Your point is that the article's interpretation is "right" because they purposefully designed the question to be confusing? Would you react the same way if you were reading an opinion poll that played the same trickery?

I mean, if they wanted a "correct" answer, why didn't they just ask "what color is sunlight?"

I guess I'm that guy. My answer would be "principally Ultraviolet C." Only if you asked "What color is the sun after filtering through our atmosphere?" am I going to say "white."

Similarly the correct answer to "Why is the sky blue?" is: "It isn't - small particles reflect ultraviolet even better than blue. We just can't (consciously) see those colors."

ipRGCs do detect UV, and our brains do get this message, and we can report it under at least some conditions.

> Similarly the correct answer to "Why is the sky blue?" is: "It isn't - small particles reflect ultraviolet even better than blue. We just can't (consciously) see those colors."

Cliff Stoll gives a great anecdote about this. He's asked "Why is the sky blue", and to every answer he gives the interviewer asks "Why?" Turns out it's a bit complicated.

https://www.youtube.com/watch?v=yfm3MFj6LLU

> I guess I'm that guy. My answer would be "principally Ultraviolet C."

> "Why is the sky blue?" is: "It isn't - small particles reflect ultraviolet even better than blue. We just can't (consciously) see those colors."

Aren't "colors" defined as the sensations our eyes give us depending on the wavelengths entering them ? Hence "ultraviolet" isn't a color, since our eyes can't detect them.

There is "light" and "visible light" but afaik there is no such thing as "colors" and "visible colors", as "colors" = how our brain interpret visible light.

The only pathology I see here is the claim that a vaguely specified question has a single unambiguous correct answer.
I looked directly at the Sun for 20 seconds without blinking. My brain could not settle on any particular color - it kept shifting hues.
I love this line: "Ah, deadlines. So this is online, but still needs work. Hopefully before any viral attention."
Well the other thing is that color is kind of a human concept, so we are also looking at the Sun with a specific set of equipment that is different from that of an insect, for example. Even questions like "what color is that flower" look very different in different wavelengths.

Even the lumen is calibrated to the receptivity of the human eye over various wavelengths.

https://en.wikipedia.org/wiki/Luminosity_function

I think a much better example of this idea is shown in one of my favorite 80s videos, "A Private Universe" which asks Stanford alumni (in their robes no less) why the moon has phases. Then they show how their misconceptions go all the way back to primary school and were never challenged.

https://vimeo.com/113349804

If you’re a film student, the answer is obviously “blue.”
Go outside look at the non-sunset sun, it's white.

Look at photo's on the internet, it's clearly white.

Being clever about scattering blue is wrong. If it did change the color of the sun to yellow during the day, the clouds would be yellow around it like sunset clouds.

The only 'clever' answer why it's not white might be, that small part of the day when people see it without pain, it is yellow.

If you had never seen a sunset or sunrise or a NASA false color sun image or a crayon drawing as a kid, you'd be adamant it is white.

This is what the article is about, people here while trying to be smart still don't seem to get it is actually white most of they day. There is no trick, no optical illusion, it's just a simple question about a thing we are told not to look at.

http://solar-center.stanford.edu/SID/activities/GreenSun.htm...

The color of the Sun can be white, yellow or orange. Out in space, it will always be white. Down on Earth, Sun's light is scattered in the atmosphere. The short waves (blue) more so than the long waves [1]. The end result is that the sky looks blue and the Sun looks like the complement of the sky. At midday, the light from the Sun comes in the most direct way, so there is little scattering, and we perceive the sky as less blue and the Sun a very bright yellow. At sunset, the rays go through a much thicker layer of atmosphere, there's a lot of scattering going on, and you end up with a dark color of the sky (blue-purple) and an orange tending to red Sun.

Whoever says the color of the Sun is always white has no idea what they are talking about.

[1] https://en.wikipedia.org/wiki/Rayleigh_scattering

What a strange article. I’m a physics grad student and I entirely disagree with his rather smug answer. From astronomy grad students, he seems to expect a plot of the black body spectrum when asking a casual question over lunch, yet he asks the same question to five year olds? If he wanted an “unambiguously correct” answer then he should ask an unambiguous question. The only pathology demonstrated here is his own.