by Ann » Mon Apr 29, 2019 1:27 pm
Lasse H wrote: ↑Mon Apr 29, 2019 10:09 am
The image as processed by Josh Lake, linked to by "featured image" in the Explanation, is much more beautiful than the APOD – in my opinion. Most of the pink background is gone and replaced by blueish areas, while at the same time the major stars that appeared bluish are more pink. What are the principles for the colouring? How much of the colours are original, and how much is enhanced, or even replaced, in these two images?
Interesting, Lasse H. As Bruce Daniel Mayfield pointed out, there may have been a mixup, and the picture that the APOD linked to may have been the intended APOD.
Nevertheless, why are the colors so different? The short answer is that today's APOD is a broadband image, and the "featured image link" one is a narrowband image.
Look at the picture at left. Broadband filters react to optical wavelengths in much the same way as the human eye does. RGB (for Red, Green, Blue) imagery shows objects in space the way the human eye would see them, if our eyes were many times more sensitive to faint light than they are. In the picture, the RGB filters correspond to the faint shaded areas in blue, green and red.
But today, many pictures of nebulas are narrowband images instead. In the picture at left, you can see the principal wavelengths that are picked up by the typical narrowband filters: Hα at 656 nm, SII at 658 nm, and OIII at 501 nm. There is usually no filter for Hβ at 486 nm.
The "featured image" in today's APOD is probably an OIII-Hα-SII image. If so, OIII, which to the eye is relatively green, will be mapped as blue. All the blue stuff that you could see in the "featured image" would be OIII. The stars, by contrast, look orange for reasons that I can't explain, but I do know that blue stars usually don't look blue in narrowband images, and that the SII filter, usually mapped as red, is particularly good at picking up starlight.
The Orion Nebula in probable RGB. Photo: Francesco Battistella
So what is the "true" color of nebulas?
The stunning image at right shows you the colors of a nebula that can often be seen in a superb RGB image. Note the dominant red color of the Orion Nebula. What we are seeing is Hα.
Yes, but note that the color close to the central engine of the Orion Nebula, the Trapezium cluster, isn't red. It's more yellow. What the filters react to here is the presence of both red Hα and green OIII, because the area next to blisteringly hot stars is usually relatively rich in OIII.
So what will you see if you look at the Trapezium region through a telescope?
The Trapezium Region. Photo: Clarkvision.
The picture at left is supposedly an RGB image, which gives you a relatively good idea of what the eye can see if you look at the Trapezium region of the Orion Nebula through a telescope. I agree that the Trapezium region looked green to me what I looked at it through a telescope, although the color was less saturated.
Why is it that we can see the green color of OIII near the Trapezium Nebula, but not the red color of Hα?
Night and day sensitivity of the human eye.
Source: Deutsches Kupferinstitut.
The answer is that the human eye is quite insensitive to faint red light, whereas, by contrast, our eyes are fairly good at seeing faint green light. And all astronomical objects except the Moon and the brightest planets are faint. As you can see from the picture, our eyes are relatively sensitive to faint green light, but quite insensitive to the deep red color of Hα.
Okay, but why would you use narrowband imagery in the first place when you photograph nebulas, but not when you photograph, say, star clusters and galaxies?
That is because stars and galaxies emit light at all wavelengths, but nebulas typically only emot light at specific wavelengths. It has to do with the fact that nebulas, which are made of gas, only emit light when they are hit by photons of certain energies. One such case is how ultraviolet light from hot stars affects hydrogen atoms in the vicinity of the hot stars. In the picture at left, you can see that the proton of the hydrogen atom is surrounded by five electron shells. The hydrogen atom only has one electron, and usually it is located in the second electron shell.
Yes, but when an energetic photon of ultraviolet light hits the hydrogen atom, it can knock the electron into a higher electron shell. Usually it just knocks it "one shell up", to the third electron shell. But the electron soon falls down again. As it does so, it emits light at the exact wavelength of 656 nm, the wavelength of Hα light.
If the photon contained even more energy, it might knock the electron "two shells up". When such an electron "falls down", it will emit a photon of 486 nm, the wavelength of Hβ light. But in rare cases, the electron of the hydrogen atom might be sent to the fourth or fifth shell and emit Hγ or Hδ emission.
What about OIII? The answer is that in nebulas surrounding clusters of hot stars, like in today's APOD, the green color of OIII is usually not prominent in RGB images. That is because red Hα is typically the dominant wavelength of emission nebulas. But there is a special case where OIII emission becomes quite visible and dominant, and that is in some planetary nebulas.
OIII-rich planetary nebula Abell 39.
Adam Block / Mount Lemmon SkyCenter / University of Arizona / CC BY-SA 3.0
Planetary nebulas are the fluorescing cast-off envelopes of dead stars, whose tiny naked inert cores are still blisteringly hot. Many planetary nebulas are poor in hydrogen, since the star's hydrogen has been used up or cast off before the star entered its planetary nebula evolutionary stage. Many planetary nebulas contain appreciable amounts of oxygen, however, and the oxygen is ionized by the extremely energetic photons emitted by the core. Therefore, many planetary nebulas look green or bluish to the eye when viewed through a telescope.
To summarize: Nebulas, unlike stars and galaxies, emit their light at certain narrow well-defined wavelengths. They can therefore be photographed through narrowband filters, which isolate the wavelengths that the nebulas emit. This also makes it easier for amateurs to photograph nebulas, because narrowband photography is typically easier than broadband photography..
But since the dominant wavelengths are green (OIII), red (Hα) and red (SII), it is necessary to map red Hα and red SII as different colors, otherwise you can't tell the difference between them. For that reason, Hα is often mapped as green in narrowband images, while SII is still red. OIII, then, is mapped as blue.
Ann
[quote="Lasse H" post_id=291824 time=1556532557 user_id=123973]
The image as processed by Josh Lake, linked to by "featured image" in the Explanation, is much more beautiful than the APOD – in my opinion. Most of the pink background is gone and replaced by blueish areas, while at the same time the major stars that appeared bluish are more pink. What are the principles for the colouring? How much of the colours are original, and how much is enhanced, or even replaced, in these two images?
[/quote]
Interesting, Lasse H. As Bruce Daniel Mayfield pointed out, there may have been a mixup, and the picture that the APOD linked to may have been the intended APOD.
[float=left][img2]https://static1.squarespace.com/static/5a18a65ba803bbe2355d8de2/t/5b322d37575d1ffdcb6fdd9c/1530015038870/jeart2.jpeg[/img2][c][size=85]Broadband and narrowband bandpasses.
Source: https://www.itelescope.net/new-blog/2018/6/26/narrowband-imaging-the-dark-magic[/size][/c][/float]Nevertheless, why are the colors so different? The short answer is that today's APOD is a broadband image, and the "featured image link" one is a narrowband image.
Look at the picture at left. Broadband filters react to optical wavelengths in much the same way as the human eye does. RGB (for Red, Green, Blue) imagery shows objects in space the way the human eye would see them, if our eyes were many times more sensitive to faint light than they are. In the picture, the RGB filters correspond to the faint shaded areas in blue, green and red.
But today, many pictures of nebulas are narrowband images instead. In the picture at left, you can see the principal wavelengths that are picked up by the typical narrowband filters: Hα at 656 nm, SII at 658 nm, and OIII at 501 nm. There is usually no filter for Hβ at 486 nm.
The "featured image" in today's APOD is probably an OIII-Hα-SII image. If so, OIII, which to the eye is relatively green, will be mapped as blue. All the blue stuff that you could see in the "featured image" would be OIII. The stars, by contrast, look orange for reasons that I can't explain, but I do know that blue stars usually don't look blue in narrowband images, and that the SII filter, usually mapped as red, is particularly good at picking up starlight.
[float=right][img2]https://smd-prod.s3.amazonaws.com/science-red/s3fs-public/styles/image_gallery_scale_960w/public/atoms/OrionDust_Battistella_960.jpg?itok=7U1EC-5K[/img2]
[c][size=85]The Orion Nebula in probable RGB. Photo: Francesco Battistella[/size][/c][/float]So what is the "true" color of nebulas?
The stunning image at right shows you the colors of a nebula that can often be seen in a superb RGB image. Note the dominant red color of the Orion Nebula. What we are seeing is Hα.
Yes, but note that the color close to the central engine of the Orion Nebula, the Trapezium cluster, isn't red. It's more yellow. What the filters react to here is the presence of both red Hα and green OIII, because the area next to blisteringly hot stars is usually relatively rich in OIII.
So what will you see if you look at the Trapezium region through a telescope?
[float=left][img2]https://encrypted-tbn0.gstatic.com/images?q=tbn:ANd9GcR4dLNXMCUvNJuK0xPCSLI6pB_1_GSvnDYwpolLvF1hLIAT8AEZ[/img2][c][size=85]The Trapezium Region. Photo: Clarkvision.[/size][/c][/float]
The picture at left is supposedly an RGB image, which gives you a relatively good idea of what the eye can see if you look at the Trapezium region of the Orion Nebula through a telescope. I agree that the Trapezium region looked green to me what I looked at it through a telescope, although the color was less saturated.
Why is it that we can see the green color of OIII near the Trapezium Nebula, but not the red color of Hα?
[float=right][img2]https://www.kupferinstitut.de/fileadmin/user_upload/kupferinstitut.de/de/Images/Werkstoffe/Anwendung/Licht/LightingFigure9.png[/img2][c][size=85]Night and day sensitivity of the human eye.
Source: Deutsches Kupferinstitut.[/size][/c][/float]
The answer is that the human eye is quite insensitive to faint red light, whereas, by contrast, our eyes are fairly good at seeing faint green light. And all astronomical objects except the Moon and the brightest planets are faint. As you can see from the picture, our eyes are relatively sensitive to faint green light, but quite insensitive to the deep red color of Hα.
Okay, but why would you use narrowband imagery in the first place when you photograph nebulas, but not when you photograph, say, star clusters and galaxies?
[float=left][img2]http://montessorimuddle.org/wp-content/uploads/2012/02/Emission_spectrum_H_annotated.png[/img2][c][size=85]How hydrogen emits light.
Source: http://montessorimuddle.org/2012/02/01/emission-spectra-how-atoms-emit-and-absorb-light/[/size][/c][/float]
That is because stars and galaxies emit light at all wavelengths, but nebulas typically only emot light at specific wavelengths. It has to do with the fact that nebulas, which are made of gas, only emit light when they are hit by photons of certain energies. One such case is how ultraviolet light from hot stars affects hydrogen atoms in the vicinity of the hot stars. In the picture at left, you can see that the proton of the hydrogen atom is surrounded by five electron shells. The hydrogen atom only has one electron, and usually it is located in the second electron shell.
Yes, but when an energetic photon of ultraviolet light hits the hydrogen atom, it can knock the electron into a higher electron shell. Usually it just knocks it "one shell up", to the third electron shell. But the electron soon falls down again. As it does so, it emits light at the exact wavelength of 656 nm, the wavelength of Hα light.
If the photon contained even more energy, it might knock the electron "two shells up". When such an electron "falls down", it will emit a photon of 486 nm, the wavelength of Hβ light. But in rare cases, the electron of the hydrogen atom might be sent to the fourth or fifth shell and emit Hγ or Hδ emission.
What about OIII? The answer is that in nebulas surrounding clusters of hot stars, like in today's APOD, the green color of OIII is usually not prominent in RGB images. That is because red Hα is typically the dominant wavelength of emission nebulas. But there is a special case where OIII emission becomes quite visible and dominant, and that is in some planetary nebulas.
[float=right][img2]https://s22380.pcdn.co/wp-content/uploads/Abell_39.jpg[/img2][c][size=85]OIII-rich planetary nebula Abell 39.
Adam Block / Mount Lemmon SkyCenter / University of Arizona / CC BY-SA 3.0[/size][/c][/float] Planetary nebulas are the fluorescing cast-off envelopes of dead stars, whose tiny naked inert cores are still blisteringly hot. Many planetary nebulas are poor in hydrogen, since the star's hydrogen has been used up or cast off before the star entered its planetary nebula evolutionary stage. Many planetary nebulas contain appreciable amounts of oxygen, however, and the oxygen is ionized by the extremely energetic photons emitted by the core. Therefore, many planetary nebulas look green or bluish to the eye when viewed through a telescope.
To summarize: Nebulas, unlike stars and galaxies, emit their light at certain narrow well-defined wavelengths. They can therefore be photographed through narrowband filters, which isolate the wavelengths that the nebulas emit. This also makes it easier for amateurs to photograph nebulas, because narrowband photography is typically easier than broadband photography..
But since the dominant wavelengths are green (OIII), red (Hα) and red (SII), it is necessary to map red Hα and red SII as different colors, otherwise you can't tell the difference between them. For that reason, Hα is often mapped as green in narrowband images, while SII is still red. OIII, then, is mapped as blue.
Ann