I want to return to the question of why the Witch Head Nebula looks so relatively red in today's APOD. My answer is that the photographer probably relied rather heavily on an Hα filter.
Let's look at Stanislav Volskiy's amazing portrait of Orion again. I haven't searched very diligently at all, but I haven't managed to find out what filters he used for his amazing image. So I'm going to have a guess: I think it was either LRGB filters (for luminosity, red, green and blue filters) or, more likely (I think) HαRGB filters.
RGB filters are typically broadband filters, which is to say that they detect a wide range of wavelengths. The image above shows you the sensitivity of a particular set of RGB filters: The B filter detects wavelengths from 350 to 550 nm, the G filter detects wavelengths from 440 to 640 nm, and the R filter detects wavelengths from 550 to 750 nm.
Approximate blackbody curves for the Sun (at 6000 K), Betelgeuse
(at 3000 K) and Rigel (at 12,000 K). New Jersey Science and Technology University
Starlight is broadband light, because stars emit a huge number of wavelengths. In the picture at left you can see an illustration of all the wavelengths emitted by the Sun between 400 nm and 700 nm. In the picture at right, you can see the blackbody curves of stars of different temperatures, which show us where their emission peaks. The emission of a star like the Sun peaks at wavelengths a little shorter (or bluer) than 500 nm, a star like Betelgeuse peaks in the invisible infrared part of the spectrum (and Betelgeuse emits very little blue light), whereas the blackbody curve of a star like Rigel peaks in the invisible ultraviolet. Note that Rigel emits more visible blue than visible red light, but Rigel still emits a lot of red light, too.
In the picture at right, I have tried to show that the wavelengths scattered by dust from a star like Rigel is not just blue light, but green, violet and ultraviolet as well. (Yes, red and yellow light is also scattered by dust, but a lot less efficiently than shortwave light.)
My point is that starlight is broadband light, and reflection nebulas are also lit up by (scattered) broadband light. Therefore the best way of photographing the light from stars and reflection nebulas is, in my opinion, to use broadband filters.
Yes, but the light from emission nebulas is something else entirely! Because emission nebulas (except planetary nebulas) are typically extremely strongly dominated by a very narrow red wavelength, hydrogen alpha, at 656.281 nm. The spectrum of an emission nebula typically looks like this:
The hydrogen alpha wavelength is almost exactly the same as the NII emission wavelength, and these two (almost coincident) wavelengths typically dominate emission nebulas completely. The green OIII wavelength is sometimes strong, but, except in planetary nebulas, it is almost never dominant.
What does it mean that "a single red wavelength" typically dominates the light from emission nebulas so completely? Let's look at two versions of the Lagoon Nebula, one that shows hydrogen alpha light only, and one broadband RGB image where Hα has been used to enhance the luminosity and the red hues of the image:
The Lagoon Nebula in RGB and Hα. Image: Ignacio Diaz Bobillo
You can see that the Hα image is much "flatter" in hue and "all red". The RGB+Hα image, by contrast, shows various shades of pink, because the very red Hα is being diluted by green OIII, bluish cyan Hβ and scattered blue starlight. Note however the dull color of the "wing" to the left of the "bright Lagoon Nebula proper". The light here is probably all Hα. The RGB+Hα image is nicer-looking, if you ask me, but the "Hα only" image has captured almost as much nebular light as the RGB+Hα image.
The point I'm trying to make is that Hα is almost always very bright in an emission nebula. By using an Hα filter, an astrophotographer can bring out huge amounts of Hα and lots and lots of details in an emission nebula, because it doesn't take so long to get a good Hα exposure. A good broadband B filter exposure takes much longer.
Another point of mine is that it is sometimes "too easy" to bring out a lot of narrowband Hα at the expense of broadband filters. Then again, you can't lie with an Hα filter: Where there is no Hα, no Hα filter will detect any Hα.
That point has been dramatically demonstrated by Alistair Symon in his portraits of the Milky Way:
The Milky Way from Cygnus to Scorpius. Image: Alistair Symon.
In the image above, Cygnus is at upper right and the tiny little Antares and Rho Ophiuchi nebula complex is at lower left. So why is the area around Cygnus so extremely red? It's because there is so much Hα light in Cygnus. Yes, but look at the long stretch of Milky Way to the left of Cygnus in this image - Sagitta, Vulpecula, Aquila, Scutum - which seems to be almost completely lacking in red. Why is there no red there? D'uh. Because there is so little Hα light there that is not hidden by dust in the Milky Way. An Hα filter there would not bring out what is not there.
You do not have the required permissions to view the files attached to this post.