The belch of a gassy galaxy

By Phil Plait | May 15, 2011 7:00 am

Spiral galaxies are inherently interesting. Something about their beauty is so enticing… but when you look at them more carefully, the science and physics behind them is terrifically compelling. And when you use different eyes — say, radio telescopes — then you see something different entirely:

This shows two views of the lovely face-on spiral galaxy NGC 6946. On the left is a visible light image, and on the right is the radio view, taken by the Westerbork Synthesis Radio Telescope (taken over the course of 192 hours). Amazingly, these two images are to the same scale!

Spiral galaxies emit light across the entire electromagnetic spectrum, including visible and radio light, but what emits that light is different. Stars and warm gas emit visible light, but cold hydrogen glows at radio wavelengths. At a wavelength of 21 centimeters (about 8.5 inches, much, much longer wavelength than visible light, by a factor of tens of millions!) cold hydrogen can actually be quite bright, making it a perfect target for big radio telescopes.

In this image on the right I superposed both shots so you can see just how much more there is to NGC 6946 than the eye sees. What this image immediately tells us is that cold hydrogen extends well beyond the region where hydrogen is warmer, toward the center of the galaxy. It also shows the gas still takes on a spiral shape well past the visible boundaries of the galaxy.

A more detailed analysis indicates there are over 100 holes in the cold hydrogen gas as well, and these correspond to areas where stars are actively forming. That’s hardly a surprise! Stars use up hydrogen gas when forming, and then heat up what remains around them in the neighborhood. Once warmed up, the gas doesn’t emit as much 21cm radio waves.

The astronomers also found a lot of this gas is moving at high speeds, up to 100 km/sec (60 miles/second, fast enough to go from the Earth to the Moon in a little over an hour!). This is probably gas that’s been blown up and out of the galaxy by stars and supernovae, only to fall back down due to the gravity of the galaxy. That’s not known for sure, but we do see such fountains in other galaxies, including our own.

I’ll be honest: I’m more of a medium-to-high energy guy than radio guy. That’s why I tend to talk more about X-rays and gamma rays from astronomical objects, but every part of the spectrum tells a story. Radio astronomy has been around for almost a century now, and it is still and always will provide insight into the mechanisms behind the Universe.

Image credit: Rense Boomsma/Digitized Sky Survey/WSRT; ASTRON/JIVE Archive

Related posts:

When beauty and science collide
IR M63. What RU?
Two nearby galaxies peek out through the dust

CATEGORIZED UNDER: Astronomy, Pretty pictures

Comments (27)

Links to this Post

  1. "Spiral Galaxy" | ArielsLight – Lighting Effects | May 16, 2011
  1. Messier Tidy Upper

    Tut, tut, BA such a crude way to describe such a glorious, splendid and surprisingly expansive galaxy that stands revealed for us! 😉 :-)

    Great images there – it is, indeed, hard to believe they are to the same scale. Not saying I don’t accept that they *are*, mind you! 😉


    PS. For those who might be interested : The countdown clock for Endeavour‘s final and the penultimate Space Shuttle launch (Source : NASA – space shuttle page) now reads 21 hours, 45 minutes and nine seconds.

  2. Joel

    It’s incredible, really. I’m reminded of the time when it first went click in my brain that the diffuse glow around optical light images of galaxies wasn’t just glare, but the extended halo, and the Galaxy as a whole was much bigger than what I thought I was looking at. It was pretty mindblowing at the time.

  3. chris j.

    do these cold hydrogen arms show the same rotation curves as the “inner” visible light arms do?

    is it possible to determine how big the milky way is, including “cold hydrogen” arms?

    do these cold hydrogen arms extend into or beyond the calculated dark matter halos for spiral galaxies?

  4. That dark spot roughly in the middle of the radio view — is that the black hole in the center of the galaxy? Or something else?

  5. Rick Johnson

    Part of the reason for the difference however is that this galaxy is severely attenuated in the visual part of the spectrum by galactic dust. To see it we have to look through a denser part of our galaxy. Deeper photos in visual light do show much of the same structure as seen in the radio image. Even my rather short visual light exposure of this galaxy (40 minutes of luminance data) with my 14″ telescope shows some of the brighter parts seen in the radio image.
    The arm running north by the red and blue “double” star is faintly visible as is the brighter part of the eastern brightening in the radio image. Going for several hours I likely would pick up most of what is seen in the radio image. I’ll give it a try later this year. Right now the sun limits me too much at my northern altitude so I’d need many nights to get enough data.

    The pink regions in my image are caused by hot hydrogen. They mark were active star formation is currently going on much like in our Orion Nebula. Though these regions are likely far grander than the Orion Nebula as I did not use a hydrogen alpha filter to bring out these regions.

  6. Pete Jackson

    I haven’t harped on this for a while, but the optical picture is way underprocessed. It’s nice and pretty to make the sky between the stars look black, but much information is lost. By raising the background level until it is gray (instead of black), you can see much fainter features showing weak optical emission corresponding to the neutral hydrogen arms, about half way out to the edge of the frame. I suggest that you try this whenever the agency issuing the press (ASTRON in this case) hasn’t done so.

    A long-exposure optical photo taken to the sky background limit might show much more faint optical emission. Of course, it would totally white out the central regions – not pretty.

  7. Ronan

    Hrm. Magnificent photo, but I’m puzzled by your statement that once the hydrogen is warmed up, it doesn’t emit as strongly at 21 cm. With an increase in temperature, I would have thought that although the peak emissions would shift to a shorter wavelength, the intensity of the 21 cm emissions would still increase. Is the decrease in radio emissions due to the hydrogen being heated up so much that it becomes ionized, so there are no electrons left to undergo transitions the characteristic transitions seen at colder temperatures?

    It seems like the only way you could get such low-energy electron transitions would be if the electrons were already at quite a high energy level, and were only passing from that level to one just above or below them–which would imply ridiculously high temperatures for such “cold” gas, but cold and hot can be awfully relative terms, so I guess that works. And they wouldn’t be likely to be disturbed by much, out there in the void, so maybe such excited states could be relatively stable even at really low temperatures.

  8. Casper

    Something i don’t quite undestand here? how can the larger disk display a spiral structure at all? If the arms of a spiral galaxy is simply the visual result of heavy stellar formation, which create a lot of class O stars, the large disk should not have a spiral structure. The few “holes” of star formation in the cold gas surely cannot explain a spiral on such a gigantic scale. Not only that, but the spiral in the cold gas disc seems to be an extension of the arms of the inner spiral, which dosen’t make any sense at all.

  9. I had not seen such a scale comparison before; I am amazed by the scale of the cold hydrogen extension of the galaxy’s arms…!

    Any idea of percentage/amount of mass they contain…?

  10. Jamey

    So what does this galaxy look like in gamma, UV, IR, microwave, etc?

  11. George Martin

    Ronam @6

    Hrm. Magnificent photo, but I’m puzzled by your statement that once the hydrogen is warmed up, it doesn’t emit as strongly at 21 cm. With an increase in temperature, I would have thought that although the peak emissions would shift to a shorter wavelength, the intensity of the 21 cm emissions would still increase.

    The 21 cm line of hydrogen arises from a “spin flip” of the electron in the ground state of the hydrogen atom. The intensity of the radiation is due only to the number of hydrogen atoms in the line of sight. The velocity of the hydrogen atoms relative to the line of sight shifts the observed frequency.

    If the hydrogen is warm enough to say emit the visible alpha line, electrons moving from the second exited state to the first exited state, then there will be very many fewer hydrogen atoms in the ground state so the 21 cm line will be much weaker if detected at all; few or no electrons in the ground state to flip spin.


  12. George Martin


    Thinking about what I quoted from Ronam @6, I think that post the post @6 showed some confusion between spectral lines and the Planck black body curve.


  13. Ronan

    Ah, thanks for the explanation; hadn’t realized that there were emissions associated with electron spin transitions, but that makes sense.

    And it’s quite possible I am confused there; my understanding is that the black body curve is what you’d get for a material capable of absorbing and emitting light across infinitely many different wavelengths (effectively what you’d get if you looked at the light emitted by, say, a plasma that was capable of absorbing and emitting light translationally, yes?), and that for something like electronic transitions you’d still get…well, a spectrum made up of distinct emission peaks, but with the intensity of each of those peaks following what you’d expect for that frequency were you dealing with a continuous blackbody spectrum. Is that incorrect?

  14. dave cortesi

    The physics is “compelling” — yeah. In particular, it must be compelling, maybe bouncing off the ceiling whooping level compelling, to _somebody_ that “the gas still takes on a spiral shape well past the visible boundaries of the galaxy.”

    In fact, in your superposition, the hydrogen arms seem to fit very well with and continue, the visible arms. This has to be fascinating to somebody that (unlike me) actually understands some of this. Obvious possibilities: One, the spiral structure is composed entirely of hydrogen, and the visible stars merely outline it (because the stars formed from gas that was already moving in the spiral dance and retained that motion); Two, whatever fizzicks create the spiral structure operates equally on stars and on diffuse cold gas; Three, the stars AND the gas are merely passengers on a spiral structure that is formed in something else, e.g. dark matter.

  15. George Martin

    Ronan @11

    …well, a spectrum made up of distinct emission peaks, but with the intensity of each of those peaks following what you’d expect for that frequency were you dealing with a continuous blackbody spectrum. Is that incorrect?

    I think that’s incorrect but there are other people who read this blog better able to comment than I can. But the intensity of a spectral line is not due to temperature but due to the number of atoms (or molecules) making the transition. Raising the temperature, i.e. energy, will promote more atoms or molecules to a higher energy level and therefore different spectral lines, but that won’t increase the energy of a given spectral line. If there are fewer atoms at a given energy level to decay, because an increase temperature caused electrons to move to higher energy levels, then the intensity of that line will decrease with temperature.

    Increasing temperature will increase the width of a spectral line but won’t increase the intensity.
    In fact astronomers use the ratio various lines of the same atom to determine the temperature of a body.

    By the way electron transitions are not the only source of spectral lines used in Astronomy. Molecules give rise to spectral lines. Molecules rotate. The angular momentum of the rotation is quantized. That is how radio astronomers detect inter stellar molecules.

    Also the electrons of the molecules vibrate back in forth like they were connected to the molecule with springs. That vibration is quantized. The vibrational transitions are primarily of use in infra red astronomy but are also important in radio astronomy.


  16. @5. Pete Jackson : I think that’s where negative images with the stars or in this case galaxies in black on a white background come in very useful and handy for revealing faint detail too.

  17. Joseph G

    Wow, I never realized… So when we say something like “the Milky Way is approximately 100,000 light-years wide,” does this mean we need an asterisk in there denoting areas with stars? That’s a big ol’ fat error bar! :)
    I’d always wondered how so many stars could continue to form even in old, established galaxies – I guess that hydrogen “fuel tank” is a lot bigger then I’d been picturing…

  18. Joseph G

    Also, the spin-flip that causes the 21cm hydrogen line – can it be reversed? Or does each hydrogen atom only make the transition once? It sounds like this spin-flip transition is different from a regular electron energy transition, but perhaps I don’t understand what we’re talking about (this is highly probable).

  19. How long have we known about this substantial extension? Does this shift our understanding of the balance of baryonic and dark matter at all, or is this old news with a pretty new picture?

  20. danezia

    @Bipedal Tetrapod, looked at the photos and thought at the same thing…

  21. Joseph G: Yes, it can be reversed, if radio waves of the same frequency were to impinge on the gas. Any emission line is also an absorption line, but whether you can see it depends on whether there’s much to absorb.

  22. Messier Tidy Upper

    I think that’s where negative images with the stars or in this case galaxies in black on a white background come in very useful and handy for revealing faint detail too.

    For instance, see the discovery image for the tiny irregular moonlet Trinculo :

    orbiting Ouranos as one example.*


    for a colour negative and see :

    FIgure [sic] 6. Deep negative B-band image of NGC 5128 shows the system of optical shells at the edges of the galaxy. This system of shells is one of the clearest indications that NGC 5128 has undergone mergers in its past.

    for a third.

    Aussie astrophotographer David Malin has taken some great examples but alas not too many that I can easily find online.


    * Trinculo was discovered in 2001 making this year (August 13th to be precise!) the tenth anniversary of its discovery – and also the discovery of no less than eleven small eccentric orbiting moons of Jupiter : Autonoe, Thyone, Hermippe, Eurydome, Sponde, Pasithee, Euanthe, Kale, Orthosie, Euporie & Aitne. Grand and poetic names for these tiny and obscure shards of rock or ice in tilted and tumbling paths around their gas giant primaries. Anyone seeking a magical sounding incantation (for whatever reason) could do worse than chanting the names of some of these lesser knowns moons or moonlets! Oh & while the Jovian moonlets are named from Greek mythology, Trinculo was named for a drunken jester from a Shakespearean play. 😉

  23. Joseph G

    Uranus has a moon called Trinculo?
    That is just awesome to know 😀

  24. abadidea

    Tomas Sedovic: I would assume (as a layperson) that the dark dot in the middle is caused by all 21cm hydrogen having long since been sucked into hot stars at the core. Black holes in the center of spiral galaxies cannot be directly imaged (insofar that there’s anything to see?) since they are always surrounded by a thick sphere of these stars.

  25. Yoruichi

    Radio technology shows what I have been thinking; That thing is a epicly monstrous!… though it really is a pretty picture 😛

  26. Rense Boomsma

    @Tomas Sedovic: the hole in the middle is an artifact produced by how the radio image has been created. There, the neutral hydrogen is seen in absorption against the bright nucleus instead of in emission. Probably there is a black hole in the center, but this image will not show that.

    Images at other wavelengths suggest that the holes seen in neutral hydrogen have in fact a real deficit of gas.

    Most optical images generally show only the tip of the iceberg. Take a look at:
    and notice that the stellar distribution is also much more extended and can be traced in the “fat” northern hydrogen spiral arm.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar