On Galaxies

Every now and then I find something that captures my interest for one reason or another, and the search to know more sets off a cascade of questions, firmly booting me down the rabbit hole of research and discovery. This time, what kicked it all off was a little object by the name of SDSS J012320.31+332050.0:

This little object was shared on one of the astronomy pages I frequent, labeled as a suspected planetary nebula. But having spent many hours pursuing the SDSS imagery and working on various projects in the SkyServer, I thought it looked familiar. I’ve seen a bunch of objects with similar appearance- little, mottled, low surface brightness blue wisps. But I’d always understood these little blue wisps to be irregular galaxies, so I became curious- is this an irregular galaxy, or could it really be a planetary nebula?

The object in UV light.

The object in UV light.

First, I checked out the object in Aladin Lite to see if there was any imagery at other wavelengths. Besides visual wavelengths, the only other imagery high resolution enough to show individual galaxies in that area of the sky was GALEX, showing ultraviolet wavelengths. GALEX shows the object as being very bright in UV, especially compared to the surrounding galaxies-this little SDSS object is even brighter than NGC 507, the brightest galaxy in the cluster!

This make sense when you consider the morphology of the galaxies in the cluster. Most of the galaxies present are ellipticals with a high percentage of old, red stars, and not many blue stars. However, you will also notice that there are a few spirals in the cluster, and they too appear reddish! This was quite confusing to me, since the cluster is not at a particularly high redshift. Averaging the redshifts given by Simbad for some of the brightest galaxies in the cluster (NGC 503, NGC 507, NGC 494, and NGC 499) yields a redshift for the cluster of z=0.017. Compare that to clusters such as Shakbazian 376 (z=0.066), Hickson 35 (z= 0.054) and galaxies such as UGC 4709 and UGC 2690 (z=0.028 and z=0.021, respectively). Despite the higher redshifts of all of those clusters and galaxies, the spirals present in that sample still appear bluer than the very red spirals of the low-redshift NGC 507 cluster!

NGC 495 on SDSS imagery.

NGC 495 in particular shows extreme reddening compared to what would be expected of a barred spiral with such a low redshift, with a g-r of 0.88. It is important to note that the g-r given for NGC 495 uses magnitudes from the core, and not the spiral arms. The SDSS did not create measurements for any areas of the spiral arms, and only caught the outer parts in two locations. The error on the magnitudes in those two locations was too great to get a meaningful measurement of g-r.

By “eyeballing” it you can see that the arms are almost exactly the same color as the core, but I wanted to be sure of that, so I took color samples of the image in Photoshop. I measured according to the hue scale, meaning that 0=pure red and 120=pure green. The core of NGC 495 has a hue of 43. The spiral arms have a hue of about 36 to 38, with the hue lowering along with the brightness. This makes sense when you consider that the background sky has an average hue of about 0: as the brightness lowers, the percentage of the light which comes from the background sky becomes more significant. Each measurement of the galaxy is the average of a sample of 51x51px, to negate the effects of noise. The measurement of the background sky is the average of a sample of 101x101px.

It should be noted that the SDSS imagery uses the g filter for the blue channel, the r filter for the green channel, and the i filter for the red channel, so the image that you see is false color shifted toward infrared wavelengths. In simple terms: green light=blue pixel, red light=green pixel, and infrared light=red pixel. This causes objects to appear a bit redder than they usually do.

Checking the field with the NASA/IPAC Infrared Science Archive Galactic Dust Reddening and Extinction tool (phew, that’s a mouthful) shows that while NGC 495 (along with the rest of the far northern side of the cluster) has locally higher amounts of reddening, the magnitude of this effect is too small to account for the observed lack of blueness. The reddening from galactic dust is estimated at only 0.06 magnitudes B-V.

Ok, so our view of the cluster isn’t being reddened by extinction. Another interesting possibility then arises: is it possible that these spirals are just intrinsically red? According to this fascinating paper by K. L. Masters et al., the answer is yes! I’d never thought about the fact that spirals can be red, though I’d run across a few of them while browsing the SDSS imagery. But according to that paper, about 6% of face-on spirals are intrinsically as red as elliptical galaxies. These red spirals occur in greatest concentrations in the outer regions of galaxy clusters, and a disproportionately large number of the red spirals have bars. NGC 495, the extremely reddened spiral I examined the color of above, is a prominently barred spiral.

So, perhaps the answer to “why is this cluster so red?” is this:

NGC 495 is intrinsically red, while since most other spirals in the cluster are viewed at oblique angles, they appear reddened from their own dust.

This hypothesis is supported by the fact that NGC 496, at a slightly higher redshift than the rest of the cluster but a less oblique angle, does have visible blueness in its spiral arms (making it apparently the only galaxy in the cluster to exhibit blue arms).


Well, I did get a little sidetracked there, so to get to the original point: The object!

I compared the SDSS and GALEX imagery for the object with some irregular galaxies I found that had similar appearances on the SDSS. You can see the comparisons on the left. The object in question is on top. You’ll notice that the objects on top and bottom appear much brighter in UV; however the noise is also very significantly brighter, so I think the greater apparent brightness is a consequence of the data being stretched differently.

While the objects all seem to increase in brightness in UV by about the same amount, that’s not immediately conclusive- a quick search reveals that planetary nebulae also have some emissions in UV. What’s difficult to find out is how much!

So, I used SkyTools to generate a list of planetary nebula for comparison on the imagery of six different sky surveys. GALEX provides images in UV wavelengths, DSS2 provides images in visible light, the SDSS and DECaLS provide images in a composite of visible and infrared, and 2MASS and WISE cover different portions of the infrared spectrum. In total, I examined 16 planetary nebulae (and 4 galaxies misclassified as planetary nebulae; they are not included in the final sample).

Identifier Appearance on DSS Appearance on GALEX Appearance on 2MASS

No comments yet.

Leave a Reply