Galactic Loops as Sources of Polarized Emission

Since I seem to have established myself as an arch-sceptic concerning the cosmological interpretation of the the BICEP2 measurement of the polarization of the cosmic microwave background (CMB), I couldn’t resist posting a link to an interesting paper by Liu et al. that has just appeared on the arXiv.

The abstract is:

We investigate possible imprints of galactic foreground structures such as the `radio loops’ in the derived maps of the cosmic microwave background. Surprisingly there is evidence for these not only at radio frequencies through their synchrotron radiation, but also at microwave frequencies where emission by dust dominates. This suggests the mechanism is magnetic dipole radiation from dust grains enriched by metallic iron, or ferrimagnetic molecules. This new foreground we have identified is present at high galactic latitudes, and potentially dominates over the expected B-mode polarisation signal due to primordial gravitational waves from inflation.

The authors argue that foreground emission from our own Galaxy has not been fully subtracted from maps of the cosmic microwave background. This emission could result in significant contamination of the CMB polarization if it is associated with dust grains aligned with the Galaxy’s magnetic field.

I’m grateful to one of the authors of the paper, Philip Mertsch, for sending me this map of the Galactic Loops with the BICEP2 region superimposed on it, demonstrating that there is potential for a contribution…




This paper is likely to provoke quite a discussion, so I thought I’d suggest one possible way of testing it, namely by updating the analysis presented by myself and Patrick Dineen in 2003 with new data. Here’s the abstract of our old paper:

We present a diagnostic test of possible Galactic contamination of cosmic microwave background sky maps designed to provide an independent check on the methods used to compile these maps. The method involves a non-parametric measurement of cross-correlation between the Faraday rotation measure (RM) of extragalactic sources and the measured microwave signal at the same angular position. We argue that statistical properties of the observed distribution of rotation measures are consistent with a Galactic origin, an argument reinforced by a direct measurement of cross-correlation between dust, free-free and synchrotron foreground maps and RM values with the strongest correlation being for dust and free-free. We do not find any statistically compelling evidence for correlations between the RM values and the COBE DMR maps at any frequency, so there is no evidence of residual contamination in these CMB maps. On the other hand, there is a statistically significant correlation of RM with the preliminary WMAP individual frequency maps which remains significant in the Tegmark et al. Wiener-filtered map but not in the Internal Linear Combination map produced by the WMAP team.

The idea is that cross-correlating the CMB pattern with Faraday rotation measures should provide an independent diagnostic of the effect of magnetic fields. Our analysis was based on old CMB data, so there’s an interesting project to be done updating it with, e.g., Planck CMB data and a larger set of rotation measures. See the comment below for a reference to more recent work along these lines, but still not including Planck.

Anyway, this all goes to show that there’s one question you can always ask about an astrophysics result: have you considered the possible role of magnetic fields?


24 Responses to “Galactic Loops as Sources of Polarized Emission”

  1. I may be getting this wrong, but as I understood it, they were examining the effects of “Galactic Loop I” on the CMB, and Loop I is not in the same place on the sky as the BICEP2 window. Is that right? So I’m a bit confused about what the direct connection between this and BICEP is …

    • telescoper Says:

      I think the point is that Loop I isn’t the only loop…

    • But the other loops (II-IV) appear to be even further away! On re-reading the paper they specifically mention that part of Loop I crosses the BICEP2 window, so maybe it is my spherical geometry that is letting me down, but I’m still confused …

      • Loop I (as defined by radio astronomers in the early 70ies) does in fact just touch the BICEP2 field, but note that there is another loop, that can be seen in 1.4 and 23 GHz polarised emission ( and that goes right through the middle of it! The reason we focussed on Loop I in our paper is that it is the most prominent one, at least in the Northern Galactic hemisphere.

      • Cheers Philipp!

      • Looks like I didn’t teach my own DPhil student to read papers carefully enough! What we actually wrote is “… the additional loop structure identified by Wolleben (2007), crosses the very region of the sky from which the BICEP 2 experiment has recently detected a B-mode polarisation signal”.

        Wolleben, M. 2007, Astrophys.J., 664, 349 [arXiv:0704.0276]

        I don’t know how to embed a picture here, otherwise would have shown this structure going right across the BICEP2 window.

        In any case as you say Peter there are many, many other ‘loops’ in the Galaxy which are not resolved against the diffuse galactic radio background. We know they are there, otherwise one cannot explain the observed angular power spectrum of the 408 MHz all-sky survey (Mertsch & Sarkar, JCAP 1306 (2013) 041, arXiv:1304.1078). I wrote a paper long ago (MNRAS 199 (1983) 97) pointing out that such old SNRs may account for most of the ‘diffuse’ galactic radio background.

        The problem with discussing science through blogs, facebook et cetera is that it is easier to make an ignorant remark before making the effort to understand the necessary background – the discipline of an academic seminar is simply not there!

        I am told that over on Facebook, someone called Tom Crawford has also made the momentous discovery that Loop I does not actually cross the BICEP2 region … and 13 people “Liked” it. I suppose that it must be hard work to actually read what we wrote and look up the reference – so old-fashioned!

      • Thanks Subir.

        In my (partial) defence, I had read the sentence regarding the Wolleben loop structure, but I hadn’t realised that you had not included this structure in the loops you showed in your figures. Philipp has now made the additional figure you were referring to available on Facebook and this clears up the confusion!

      • Scott Dodelson Says:


        I’ll take the hit for the facebook page and whatever confusion that
        caused. Tom Crawford is one of the top CMB experimentalists, having played a major role in SPT for a decade; also a great guy. Just saying, in case anyone misconstrues what you wrote.

        In either case, exciting times; it will be interesting to see how things shake out.


      • Dear Scott, sorry if I sounded a little aggrieved … but this is a serious issue and people should think carefully before making comments if these social media discussions are not to become rumour and gossip mills. I accept that it can be a valuable way for the community to stay abreast of a rapidly moving topic – but only when there is a sensible moderator (like yourself)!

        What really irked me was the suggestion that by pointing out this important possible systematic we are somehow running down the BICEP2 result. We are doing science here – not selling used cars! Please let’s all focus on the job rather than take cheap pot shots at each other. As you say these are exciting times and there’s room for everyone to make a contribution.

  2. Interesting stuff Peter. Thanks.

    I’m sceptical too.

  3. Regarding your idea, Peter, of using correlations between Faraday Rotation and CMB maps to look for foreground contamination, as you know this was recently (but pre-Planck) followed up with WMAP data. The problem, especially in the “Southern Hole”, is however the scarcity of rotation measure data.

  4. James Gallagher Says:

    Oh great, so you guys have known about this for years and you didn’t think to tell the guys at the South Pole that they’re pointing their detectors at a dirty part of the sky.

    (Kinda jokin’ here – but you get my point)

    • stringph Says:

      Yes, people kept the issue a secret from the BICEP people by publishing about it in scientific journals.

      • James Gallagher Says:

        How are they supposed to read journals at the south pole – it’s bloody cold you know.

  5. stringph Says:

    Magnetic fields? has a few.

  6. […] a research paper that suggests the signals could be galactic in origin rather than cosmological (here). The observations merit close scrutiny from those qualified to do so – but the hype […]

  7. […] a research paper that suggests the signals could be galactic in origin rather than cosmological (here). The observations merit close scrutiny from those qualified to do so – but the hype […]

  8. […] Collaboration have used extragactic Faraday Rotation measures to probe the Galactic Magnetic field as I suggested on this blog not long ago. The article that first advocated doing this with CMB maps can be found […]

  9. Have you seen our new all-singing all-dancing all-sky RM analysis? Is this useful for CMB?

  10. […] Coles, “Galactic Loops as Sources of Polarized Emission“, In The Dark, April 8, […]

  11. […] famous Auditorium A near the end of the first talk, about the Imprint of Radio Loops on the CMB (a subject I’ve blogged about), which is a shame because (a) its interesting and (b) some of my own work was apparently […]

  12. […] ja palaan asiaan tilanteen tarkentuessa. Aiheesta enemmän Physics Worldissa ja Peter Colesin blogissa (myös kommentit ovat […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: