APOD: Journey into the Cosmic Reef (2020 May 18)

[Chris Peterson]

I can't even get them out of data, though. Like you'd have to either be using a specific set of filters to make them look green. Don't think you'd get green stars as long as you're using wideband visible filters.

No... because wideband visible filters, by design, are intended to approximate the same colorspace as the human eye/brain system does. To fully understand the "color" (using the term loosely) of something you need its spectrum. An infinite number of images made through an infinite number of infinitely narrow filters.

[geckzilla]

I can't even get them out of data, though. Like you'd have to either be using a specific set of filters to make them look green. Don't think you'd get green stars as long as you're using wideband visible filters.

No... because wideband visible filters, by design, are intended to approximate the same colorspace as the human eye/brain system does. To fully understand the "color" (using the term loosely) of something you need its spectrum. An infinite number of images made through an infinite number of infinitely narrow filters.

do I need also to divide by zero?

[Chris Peterson]

I can't even get them out of data, though. Like you'd have to either be using a specific set of filters to make them look green. Don't think you'd get green stars as long as you're using wideband visible filters.

No... because wideband visible filters, by design, are intended to approximate the same colorspace as the human eye/brain system does. To fully understand the "color" (using the term loosely) of something you need its spectrum. An infinite number of images made through an infinite number of infinitely narrow filters.

do I need also to divide by zero? :lol2:

It will work as long as you take forever to get there.

[Ann]

---

The only thing I find even remotely misleading about space images is that nebulas are probably a lot greener than we let it be... Note, however, there aren't any green stars. Green stars? weird. Green nebulas? probably all over the place.

Thermal black-body radiation drops off exponentially at short wavelengths so we can discern both red & orange stars. However, once the peak gets to yellow/green near the center of our vision it fires up all our vision cones and appears white (by definition?).

I can't even get them out of data, though. Like you'd have to either be using a specific set of filters to make them look green. Don't think you'd get green stars as long as you're using wideband visible filters.

Blackbody curves, reasonably similar to light curves of stars.

---

A planetary nebula spectrum.

---

The difference between a green (probably blue-green) nebula and a star is that the blue-green nebula emits its greenish light as spikes, whereas the star emits it as a blackbody light curve.

Human color vision has evolved to see the brightest light available to us, which is sunlight, as white. More specifically, we see daylight as white. If the Earth had orbited a K-type dwarf like the components of 61 Cygni, we would have seen our K-type daylight as white.

When the sky is blue, sunlit areas are yellower than areas in the shadow.

---

When the Sun peeks through clouds in a gray sky, the Sun looks white.

---

I'm sorry if I'm insulting your intelligence by pointing this out, Geck, but it is one of my pet peeves that the Sun is not yellow. It is white because we have to see it as white, because we have to see daylight as white.

On a day when the sky is blue, sunlit areas are yellower and areas that are in shadow are bluer, because much of the blue light of the Sun has been scattered in the atmosphere to make the sky look blue. This in turn makes direct sunlight yellower. The yellower hue of direct sunlight makes us think that the Sun is yellow. But on a semi-overcast day, when the Sun can still peek through the cloud cover even though the sky is gray, the Sun is seen as white.

Everything that looks green on the Earth has to have a "spiky" spectrum, with a sharp spike in the green part of the spectrum. Like grass, or like planetary nebulas.

Ann

[geckzilla]

But our eyes can see green colors on a computer screen, and stellar color data can be captured in a way that isn't saturated to the point that it is white. Thing is, they never turn up green, ever. I don't really care how nitpicky you get about it. The sum of its parts just don't end up green.

[Ann]

But our eyes can see green colors on a computer screen, and stellar color data can be captured in a way that isn't saturated to the point that it is white. Thing is, they never turn up green, ever. I don't really care how nitpicky you get about it. The sum of its parts just don't end up green.

That is partly my point. I mean that the white color of the Sun is not just a matter of over-saturation. I mean that if we could keep the light curve of the Sun unchanged and just turn down the intensity of its light, the color of the Sun would still be white, or neutral.

I don't think it is possible to make the light curve corresponding to an actual stellar spectrum look green, because our eyes don't know how to interpret a blackbody curve as green. Of course, this is my very very amateur opinion.

Ann

[geckzilla]

Well, if we interpret color with our eyes, and our eyes never see green in any stars, I don't know what to say other than stars aren't green. And if that makes me stupid then I'll take stupid.

Here's a picture of a star that ended up green.
https://flic.kr/p/2iFw1Hk

[neufer]

Well, if we interpret color with our eyes, and our eyes never see green in any stars, I don't know what to say other than stars aren't green.

oh darn! You're telling me that the light from 5,772 K yellow/green peaking star like the Sun won't even make my white sheets drying out on the clothes line look yellow/green ... and therefore can't be referred to as yellow/green? Way to kill my dreams.

[TheZuke!]

Well, if we interpret color with our eyes, and our eyes never see green in any stars, I don't know what to say other than stars aren't green. And if that makes me stupid then I'll take stupid.

Here's a picture of a star that ended up green.
https://flic.kr/p/2iFw1Hk

Here's another star that was green.

http://www.movieactors.com/actors/margarethamilton.htm

[TheZuke!]

Everything that looks green on the Earth has to have a "spiky" spectrum, with a sharp spike in the green part of the spectrum. Like grass, or like planetary nebulas.

Ann

Such as the "rare" green sunset.

[geckzilla]

Well, if we interpret color with our eyes, and our eyes never see green in any stars, I don't know what to say other than stars aren't green.

oh darn! You're telling me that the light from 5,772 K yellow/green peaking star like the Sun won't even make my white sheets drying out on the clothes line look yellow/green ... and therefore can't be referred to as yellow/green? Way to kill my dreams.

it's like yellow and green are now the same color!

[Chris Peterson]

I'm sorry if I'm insulting your intelligence by pointing this out, Geck, but it is one of my pet peeves that the Sun is not yellow. It is white because we have to see it as white, because we have to see daylight as white.

"Color" has many different meanings. Astronomically, the color of a star is unrelated to its visual appearance, and refers to the part of the visible spectrum where its peak output lies. That's determined by its temperature, and is typically measured by looking at the intensity of the star through two filters, most commonly a green and a blue one, and then taking the difference as an index of color. (This is error prone if the star has a gaseous envelope or there is intervening dust, in which case the astronomical color is better determined spectroscopically.) The Sun is properly considered a "yellow" star in the astronomical sense of the word.

What we call "color" when we're talking about human vision is a combination of physical and physiological effects. Light stimulates three different receptors in our eyes, with broad sensitivities at short, medium, and long wavelengths (which we tend to think of as blue, green, and red, although that's not very accurate in the case of our eyes). In dim light, our rods are involved, too, so there are actually four sensors... and indeed, we see colors quite differently in dim light than in bright. So our brain receives signals corresponding to intensity and the intensity ratios of the different sensors, and turns that into perceptual color. Perceptual color doesn't just include hue, but also intensity. So if you take a single wavelength light source (which you might think of as a "pure" color- for instance the OIII channel of an astronomical image) and vary just its intensity, you will interpret that as a variation in color. And, of course, your brain further messes with reality by trying to force what you're seeing into an intensity range of black to white, and by making the illumination source "white". Put on pink glasses, and after a few minutes your brain will be telling you the light is neutral.

[Ann]

I'm sorry if I'm insulting your intelligence by pointing this out, Geck, but it is one of my pet peeves that the Sun is not yellow. It is white because we have to see it as white, because we have to see daylight as white.

"Color" has many different meanings. Astronomically, the color of a star is unrelated to its visual appearance, and refers to the part of the visible spectrum where its peak output lies. That's determined by its temperature, and is typically measured by looking at the intensity of the star through two filters, most commonly a green and a blue one, and then taking the difference as an index of color. (This is error prone if the star has a gaseous envelope or there is intervening dust, in which case the astronomical color is better determined spectroscopically.) The Sun is properly considered a "yellow" star in the astronomical sense of the word.

Okay. So the Sun, whose spectrum peaks in the green (if not blue-green) part of the spectrum, is considered, astronomically, to be a yellow star. And Vega is the "white standard" of all stars, as it is considered to be the perfect example of a perfectly white star.

Vega, the perfectly white blue star. Photo: Stephen Rahn.

---

Blue white dwarf in old globular cluster M4. Hubble image: NASA and H. Richer (University of British Columbia); Ground-based image: NOAO/AURA/NSF

---

Never mind that Vega, if it were to replace the Sun at the center of our solar system, would not only kill us by hitting us with very harsh ultraviolet light, and, if we managed to stay alive, blind us, because Vega is some 50 times brighter than the Sun (image having 50 Suns!! in the sky), and also quickly make the Earth's ocean's boil, but also, if we managed to somehow not only survive but keep our color vision intact, bathe us in blue light.

Similarly, white dwarfs, which are in most cases blue, since they are blisteringly hot and have not had time to cool down to non-blue temperatures, are called "white" because... Well, because astronomers don't seem to like calling blue things blue.

Ann

[Chris Peterson]

I'm sorry if I'm insulting your intelligence by pointing this out, Geck, but it is one of my pet peeves that the Sun is not yellow. It is white because we have to see it as white, because we have to see daylight as white.

"Color" has many different meanings. Astronomically, the color of a star is unrelated to its visual appearance, and refers to the part of the visible spectrum where its peak output lies. That's determined by its temperature, and is typically measured by looking at the intensity of the star through two filters, most commonly a green and a blue one, and then taking the difference as an index of color. (This is error prone if the star has a gaseous envelope or there is intervening dust, in which case the astronomical color is better determined spectroscopically.) The Sun is properly considered a "yellow" star in the astronomical sense of the word.

Okay. So the Sun, whose spectrum peaks in the green (if not blue-green) part of the spectrum, is considered, astronomically, to be a yellow star. And Vega is the "white standard" of all stars, as it is considered to be the perfect example of a perfectly white star.

Vega is "white" only because the standard filters are normalized to give it a B-V of zero. Again, astronomical color is a quantitative value that isn't directly related to physiological color. Astronomically, the color of a star is basically just a measure of its temperature.

[neufer]

Planckian locus in CIE 1931 x,y space

---

Okay. So the Sun, whose spectrum peaks in the green (if not blue-green) part of the spectrum, is considered, astronomically, to be a yellow star. And Vega is the "white standard" of all stars, as it is considered to be the perfect example of a perfectly white star.

Sol is 5,777 K [Yellow/Green]

Rapidly rotating Vega is:
10,060 K [Cyan] at the poles &
8,152 K [Green] at the equator

The color (chromaticity) of black-body radiation depends on reverse the temperature of the black body; the locus of such colors, shown here in CIE 1931 x,y space, is known as the Planckian locus.

[Ann]

Planckian locus in CIE 1931 x,y space

---

Okay. So the Sun, whose spectrum peaks in the green (if not blue-green) part of the spectrum, is considered, astronomically, to be a yellow star. And Vega is the "white standard" of all stars, as it is considered to be the perfect example of a perfectly white star.

Sol is 5,777 K [Yellow/Green]

Rapidly rotating Vega is:
10,060 K [Cyan] at the poles &
8,152 K [Green] at the equator

The color (chromaticity) of black-body radiation depends on reverse the temperature of the black body; the locus of such colors, shown here in CIE 1931 x,y space, is known as the Planckian locus.

You wouldn't quote mumbo-jumbo, Art, and I have seen this graph thingy before, so I know that there is some serious science between this Planckian locus.

For all of that, this graph assumes, as far as I can see, that colors exist independently of human vision. But yellow-green is simply the response of our brains to certain stimuli from our retinas. Therefore, "yellow-green" only exists inside our heads (and possibly in some animals' heads), not independently in the outside world.

But if "yellow-green" is a sensation that only exists inside our heads, and we can never see the Suns as yellow-green, how can anyone claim that the "true color" of the Sun is yellow-green?

Without our brains, there are no colors and no sounds, just electromagnetic waves of different wavelengths. So even though this Planckian thing obviously has a sound scientific grounding, by saying that the blackbody curve of our Sun is yellow-green in about the same way as the speed of light is ~300,000 kilometers per second, I choose to completely ignore it.

Thank you.

Ann

[Chris Peterson]

For all of that, this graph assumes, as far as I can see, that colors exist independently of human vision. Otherwise, how could the Sun be called yellow-green, when we certainly never see it as the least bit greenish?

The Sun is not called "yellow-green".

Astronomically, the Sun is called "yellow".
Physiologically, the Sun is called "white".
As a blackbody, the peak intensity of the Sun is in the part of the spectrum that most people perceive as yellow-green.

[Ann]

For all of that, this graph assumes, as far as I can see, that colors exist independently of human vision. Otherwise, how could the Sun be called yellow-green, when we certainly never see it as the least bit greenish?

The Sun is not called "yellow-green".

Astronomically, the Sun is called "yellow".
Physiologically, the Sun is called "white".
As a blackbody, the peak intensity of the Sun is in the part of the spectrum that most people perceive as yellow-green.

That's not what I have heard.

Animation showing an approximation of the color of the Planckian Locus through the visible spectrum. It doesn't look as if a temperature of 5,777 Kelvin doesn't seem to correspond to a yellow-green color. Animation by Lucas(CA) at Wikipedia.

---

Lou Mayo of NASA wrote:

So, the sun actually emits energy at all wavelengths from radio to gamma ray. But, as can be seen in the image above, 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! This maximum radiation frequency is governed by the sun’s surface temperature, around 5,800K. A higher surface temperature would result in a shorter maximum wavelength and our sun might peak in the blue or violet part of the spectrum (or even the ultra violet!). A lower surface temperature, and our sun’s spectrum might peak in the yellow or orange or even red part of the spectrum.

Ann

[Chris Peterson]

For all of that, this graph assumes, as far as I can see, that colors exist independently of human vision. Otherwise, how could the Sun be called yellow-green, when we certainly never see it as the least bit greenish?

The Sun is not called "yellow-green".

Astronomically, the Sun is called "yellow".
Physiologically, the Sun is called "white".
As a blackbody, the peak intensity of the Sun is in the part of the spectrum that most people perceive as yellow-green.

That's not what I have heard.

Lou Mayo of NASA wrote:

So, the sun actually emits energy at all wavelengths from radio to gamma ray. But, as can be seen in the image above, 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! This maximum radiation frequency is governed by the sun’s surface temperature, around 5,800K. A higher surface temperature would result in a shorter maximum wavelength and our sun might peak in the blue or violet part of the spectrum (or even the ultra violet!). A lower surface temperature, and our sun’s spectrum might peak in the yellow or orange or even red part of the spectrum.

Ann

Again, nobody calls the Sun "yellow-green". Ignore the sloppy non-technical language in that quote (which if you read it carefully, is just staying what I said, which is that the peak intensity of the Sun's blackbody curve is in the part of the spectrum we see as yellow-green. That most certainly does not mean we see the Sun's output as yellow-green, and the quote doesn't suggest otherwise.)

[neufer]

Lou Mayo of NASA wrote:

So, the sun actually emits energy at all wavelengths from radio to gamma ray. But, as can be seen in the image above, 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! This maximum radiation frequency is governed by the sun’s surface temperature, around 5,800K. A higher surface temperature would result in a shorter maximum wavelength and our sun might peak in the blue or violet part of the spectrum (or even the ultra violet!). A lower surface temperature, and our sun’s spectrum might peak in the yellow or orange or even red part of the spectrum.

Lou Mayo of NASA also wrote:

For our sun, this black body curve or “Plank [sic] Function” is a smooth almost bell shaped curve involving electromagnetic (EM) radiation at many different wavelengths from very long infrared to very short ultraviolet wavelengths.

“Plank [sic] Function” is NOT an almost bell shaped curve;
it is a quite asymmetric and depends upon the choice of spectral variable:

<<The location of the peak of the spectral distribution for Planck's law depends on the choice of spectral variable.

---

For example, using T = 6000 K and parameterization by wavelength, the wavelength for maximal spectral radiance is λ = 482.962 nm with corresponding frequency ν = 620.737 THz.

For the same temperature, but parameterizing by frequency, the frequency for maximal spectral radiance is ν = 352.735 THz with corresponding wavelength λ = 849.907 nm.>>

Perhaps one should talk of the median value rather than the ambiguous peak value.
But the Planckian locus in CIE 1931 x,y space above is the best way to go IMO.

Again, nobody calls the Sun "yellow-green". Ignore the sloppy non-technical language in that quote (which if you read it carefully, is just staying what I said, which is that the peak intensity of the Sun's blackbody curve is in the part of the spectrum we see as yellow-green. That most certainly does not mean we see the Sun's output as yellow-green, and the quote doesn't suggest otherwise.)

[Ann]

Again, nobody calls the Sun "yellow-green". Ignore the sloppy non-technical language in that quote (which if you read it carefully, is just staying what I said, which is that the peak intensity of the Sun's blackbody curve is in the part of the spectrum we see as yellow-green. That most certainly does not mean we see the Sun's output as yellow-green, and the quote doesn't suggest otherwise.)

Like you and Art pointed out, Lou Mayo probably isn't a highly reliable and knowledgeable science popularizer.

That said, I still don't understand where you have found the information that the blackbody curve that best corresponds to the temperature of the Sun peaks in the yellow-green part of the spectrum. I have always heard that it peaks in the green part of the spectrum, not specifically in the yellow-green. Where have you found that information?

Christopher S. Baird wrote:

The color of the sun is white.
...
It may be tempting to examine the color content of sunlight and identify the brightest color (the peak frequency) as the actual color of the sun. The problem with this approach is that peak frequency does not have a concrete meaning. The peak frequency is different depending on whether you are in frequency space or in wavelength space, as shown in the images below. In wavelength space, sunlight peaks in the violet. In frequency space, sunlight peaks in the infrared. Which is right? They are both right.
...
Furthermore, astronomers like to model the sun as a perfect blackbody, which it is not. According to the wavelength-space blackbody model, the sun peaks in the green! When astronomers say the sun is green, they mean that their inexact model peaks in wavelength in the green.

Unfortunately, "The sun is Green!" makes for more exciting headlines than, "The sun is white and would peak in the green if it were a perfect blackbody and if you measure in wavelength space." Although not as exciting, the ultimate truth is: the sun is white; its spectrum peaks in the violet in wavelength space, in the infrared in frequency space, and in the green according to the wavelength-space blackbody approximation.

I don't see anything in Christopher S. Baird's description of the color of the Sun that suggests that the Sun is yellow-green.

Ann

[geckzilla]

Another fun color astronomical puzzle is Neptune. Very, very few people have ever seen it directly, resolved through a telescope, but at least one person I know, Dr. Heidi Hammel, described it as more blue than cyan, and leaning more toward saturated and deep than pale. Another person with his own kind of color expertise is Björn Jónsson, who analyzed the color spectra of Uranus and Neptune to illustrate them in this article:
https://www.planetary.org/blogs/guest-b ... rence.html

While he does show Neptune as being more blue and less pale then Uranus, Heidi's own visual approximation based on her experience viewing it through a 24-inch telescope on Mauna Kea looks like this:
https://pbs.twimg.com/media/DRG-f6rUQAA ... =4096x4096

Certainly, I think she disagrees with Björn's conclusion. Who's right though? The person who analyzed the data, or the person who saw it with her own eyes? It's hard to say for sure, and human perception is fallible, but is Björn's analysis necessarily correct? I find myself wanting to lean toward Heidi's impression.

[Ann]

Another fun color astronomical puzzle is Neptune. Very, very few people have ever seen it directly, resolved through a telescope, but at least one person I know, Dr. Heidi Hammel, described it as more blue than cyan, and leaning more toward saturated and deep than pale. Another person with his own kind of color expertise is Björn Jónsson, who analyzed the color spectra of Uranus and Neptune to illustrate them in this article:
https://www.planetary.org/blogs/guest-b ... rence.html

While he does show Neptune as being more blue and less pale then Uranus, Heidi's own visual approximation based on her experience viewing it through a 24-inch telescope on Mauna Kea looks like this:
https://pbs.twimg.com/media/DRG-f6rUQAA ... =4096x4096

Certainly, I think she disagrees with Björn's conclusion. Who's right though? The person who analyzed the data, or the person who saw it with her own eyes? It's hard to say for sure, and human perception is fallible, but is Björn's analysis necessarily correct? I find myself wanting to lean toward Heidi's impression.

Thanks, Geck, very interesting!

I have long suspected that the "almost-too-blue-to-be-true" color of Neptune as seen in the Voyager 2 data really is too blue to be true. Of course, I have no means of knowing what the color of Neptune "really" is.

I very much enjoyed Björn Jónsson analysis of the color difference between Uranus and Neptune! Hugely interesting! Of course, Heidi Hammel may still be correct, and Neptune could be just as "cornflower blue" as the the Voyager 2 images showed it to be.

Thanks again! I so enjoyed this!

Ann

[neufer]

50 Million Schoolchildren Can't Be Wrong

---

I have always heard that it peaks in the green part of the spectrum, not specifically in the yellow-green. Where have you found that information?

From the Planckian locus in CIE 1931 x,y space at 5777 K.
(Also scattering by the atmosphere makes it redder.)

Besides: "50 Million Schoolchildren Can't Be Wrong "

• "The Last of the Red Hot Mamas:"

Click to play embedded YouTube video.

[Chris Peterson]

Again, nobody calls the Sun "yellow-green". Ignore the sloppy non-technical language in that quote (which if you read it carefully, is just staying what I said, which is that the peak intensity of the Sun's blackbody curve is in the part of the spectrum we see as yellow-green. That most certainly does not mean we see the Sun's output as yellow-green, and the quote doesn't suggest otherwise.)

Like you and Art pointed out, Lou Mayo probably isn't a highly reliable and knowledgeable science popularizer.

That said, I still don't understand where you have found the information that the blackbody curve that best corresponds to the temperature of the Sun peaks in the yellow-green part of the spectrum. I have always heard that it peaks in the green part of the spectrum, not specifically in the yellow-green.

Keep in mind that the location of the peak- a single wavelength- does not correspond to visual color. Our eyes are stimulated by all of the light on both sides of that peak, as well. If you were to take out a slice of the energy around the peak that was a few hundred nanometers wide, few would likely describe the color as a pure green (like they would with a narrow slice right at the peak). And, of course, color is subjective. People describe the visual experience of color very differently.