Red spirals are not dead

If you have ever taken an introductory astronomy course, one of the first things you learned about galaxies is that elliptical galaxies are red, and spiral galaxies are blue.  Two major classifications or morphologies (plus irregulars and lenticulars), two distinct colors – done.  Close textbook.  We have an easy way to distinguish between galaxy types without looking too closely at them, and we can move on with our lives.  However, it appears that the real picture may not be so simple: the discovery of a population of red spiral galaxies in the local universe (as well as some blue, elliptical-like spheriods, but we won’t get into that here) has led astronomers to begin to doubt this traditional paradigm.  The author of this research note sets off to investigate this odd population of spiral galaxies and uncover what is, or is not, responsible for their rosy appearance.

Before we can puzzle out the reason this seemingly abnormal group exists, we first need to revisit the physical explanation for the optical colors of “normal” galaxies.  In slightly more technical terms than “blue” and “red”, astronomers say that spirals live in the “blue cloud” of a color-magnitude diagram, ellipticals live in the “red sequence”, and a few interlopers fall into the “green valley”.  (See this astrobite for a diagram and a similar discussion).  The galaxies in the blue cloud are actively forming stars; their colors are dominated by emission from the youngest stars, which are the brightest and hottest, and therefore the bluest à la blackbody radiation (like the bottom of a flame).  In contrast, galaxies in the red sequence are mostly devoid of gas and cannot form many new stars.  Because massive, blue stars evolve and explode as supernovae relatively quickly, only the long-lived, low mass stars – which are the dimmest and coolest, and therefore the most red – are thought to remain in these types of galaxies; thus, we say they are “red and dead”.

Figure 1: Red spirals (top) and normal, blue spirals (bottom) (Masters et al. 2010).

And now, time for the strange galaxies.  Citizen scientists contributing to Galaxy Zoo have helped astronomers uncover about 300 red spiral galaxies in the Sloan Digital Sky Survey (SDSS).  (Note that individually classifying all of the galaxies in the SDSS would have been a near-impossible task for professional astronomers without assistance – see Galaxy Zoo co-founder Kevin Schawinski’s guest post about how to get involved in citizen science yourself!)  To find out if these red galaxies are also dead, the author of this paper adds ultraviolet (UV) data from the Galaxy Evolution Explorer (GALEX) along with infrared (IR) data from the Wide-field Infrared Survey Explorer (WISE) to the optical information from SDSS for 255 of the original galaxies.  Ultraviolet emission can be used to infer the star formation rate, knowing that the vast majority of the UV again comes from the youngest, hottest, most massive stars, and assuming some underlying distribution of masses for the stars that aren’t emitting in the UV (the Initial Mass Function, or IMF – e.g. these astrobites).  Simply measuring the UV emission isn’t enough, as some of this radiation is absorbed by dust grains along the way.  Luckily, the UV photons heat the dust grains, causing them to emit thermally in the infrared (IR); thus, if we have data in both the UV and the IR parts of the spectrum, we can combine them to infer how much UV emission tried to leave the galaxy in the first place.

Figure 2b: Edge-on galaxy (SDSS)

Figure 2a: Face-on galaxy (SDSS)

Of course, there are two perfectly adequate yet mundane reasons why spiral galaxies might be red.  Happily, the authors of the original paper in which these data were published (Masters et al. 2010) have already pared down the sample to avoid such confusion.  The first explanation is reddening of by dust, which scatters blue light more effectively than red light, moving blue light out of our line-of-sight.  This could cause intrinsically blue galaxies to appear red.  The same physics (Rayleigh scattering) explains why the sky is blue while sunsets are red; in the first case, we are looking at indirect, scattered light, and in the latter case, we are looking directly at the Sun through a thick, light-scattering atmosphere.  To cut down on dusty galaxies, the authors select galaxies that are nearly face-on (Figure 2a), as opposed to edge-on (Figure 2b), based on the ratio of their projected major versus minor axes.  An image of our own Milky Way in the infrared (e.g. from WISE, as discussed in this astrobite) shows the logic in this approach – all of the dust lies in the plane of the Galaxy, so an edge-on galaxy will be much more affected by reddening than a face-on galaxy.

The second factor that could naturally redden a spiral galaxy has to do with the size of its central bulge relative to the star-forming disk.  Almost all spirals have this spherical, quiescent (and red) component, and in the case where the bulge dominates the luminosity, the galaxy could thus easily appear red.

For each galaxy, the authors model the surface brightness profile – which describes how much light the galaxy emits as a function of position, here defined solely by the distance from the galaxy center – by treating the disk and the bulge as two distinct components, using an exponential disk for the former and a de Vaucoulers law for the latter.  They then compare the contributions from the two profiles, selecting only galaxies where the bulge contributes less that 50% of the light, or fDeV < 0.5 in Figure 3 below.

Figure 3: Increasing bulge fraction from left to right (Masters et al. 2010). fDeV is the fraction of the galaxy's luminosity contributed by the bulge; a higher fDeV leads to a much redder looking galaxy.

Figure 4: NUV - r color versus stellar mass. Contours show a distribution of local galaxies, and red points are the red spirals in this sample.

Now that we have ruled out the uninteresting options, we can ask the question we are really after: are the red spiral galaxies forming stars?  The answer, surprisingly, is yes!  Not only are these red spirals still active, but they seem to be forming stars at the same rate as other spirals in the local universe.  The contours of Figure 4 show the underlying bimodal distribution of local galaxies as categorized by their near-UV (NUV) to optical (r-band) color (lower NUV – values indicate bluer colors).  From here, it is apparent that the majority of the spirals from this sample, which are plotted as red points in the figure, fall in line with the typical blue spirals.  So, why do they appear red in the optical?  The author suggests that the mass of the galaxy is the key parameter.  All of these red spirals are massive (> 10^10 solar masses), and although they are forming stars at a normal rate, a large, underlying population of older, red stars dominates the stellar mass content in these galaxies, shifting their optical color to the red.  Thus, we are left with a cautionary tale: at high masses, optical color is no longer a good proxy for morphology, and a red spiral galaxy may be nothing more than a red herring!


About Alice Olmstead

I am a fourth-year graduate student at the University of Maryland, College Park. I currently do astronomy education research with Chandra Turpen, Ed Prather, Joe Redish and Colin Wallace, focusing on how professional development workshops help faculty to grow as educators. Prior to that, I studied distant, highly magnified, gravitationally lensed galaxies to investigate where they formed their stars and why. Outside of academics, I love travel, hiking, music, and vegan food. 🙂

Discover more from astrobites

Subscribe to get the latest posts to your email.


  1. Astrobites@Anchorage: Tuesday Morning Sessions | astrobites - [...] formation history predicted the complicated distribution of galaxy types we see today, from the “large, red, and dead” ellipticals…
  2. Not Dead Yet: Signs of Cool Gas Around Elliptical Galaxies | astrobites - [...] learned that elliptical galaxies appear red in color and that spiral galaxies appear blue (but see this astrobite). The…

Leave a Reply