Batman's gas is all wonky

You don't need really high resolution observations to get a lot of interesting information about a galaxy's gas. An unresolved spectrum can still tell you quite a bit about the total content and rotation, just not anything about which bit of a galaxy is doing what. A typical example (taken from today's paper) looks like this :

This is just brightness on the vertical axis as a function of velocity. Due to the rotation of the galaxy, one half of the of the gas is at a lower velocity (as it comes towards us) and the other is at a higher velocity (as it moves away from us). The double-horn "batman" shape is because most gas tends to be moving at a constant rotational velocity. But why is it all wonky ? Did Batman order dodgy costumes again ?

I've often wondered if we would extract more information from the spectra than just a crude estimate of the rotation speed. If we could measure the wonkiness, could we at least get an idea of whether the gas is being disturbed by something ?

The answer from this paper seems to be basically "no". They note that as others have found, measuring the asymmetry - the ratio of the flux in both halves of the spectrum - is a surprisingly subtle process. There are so many uncertainties in the measurement (finding the exact centre, defining where the gas ends and the noise begins) that you can only get a reliable measurement of asymmetry for the brightest galaxies, with signal to noise above 30, and even then they've got to have really quite large ratios before you say they're at all interesting.

This paper uses the HIPASS survey to quantify asymmetry across the whole southern sky. HIPASS was the first ever all-sky survey, but it has some serious limitations in sensitivity and resolution. Not only could it only detect bright, relatively nearby galaxies, but its field of view is so large that there can often be multiple galaxies seen in the telescope beam. So they restrict their analysis to those galaxies which have only one plausible optical counterpart, so they know the asymmetry isn't just the result of measuring multiple galaxies at once. They also quantify how much gas the galaxy has lost (if any) compared to typical galaxies of similar size and morphology.

What they find is.... well, not much, to be honest.

They divide their sample into four "distance"* bins, and plot both their parameters (deficiency and asymmetry) as both maps and as functions of overall galaxy density. They only see significant gas loss within the Virgo cluster, which is present in the first and second bins (and a little bit in the third as well).

* To be more accurate they use velocity, which is a rough proxy for distance on very large scales.

In their closest bin, there's a clear trend with gas loss as a function of density. Which makes sense : the more galaxies smashing around, the more the gas is going to get torn out. And more importantly, high-density environments like Virgo experience other effects (like ram pressure stripping from hot external gas) that's far more effective at stripping gas than galaxy-galaxy encounters. But in the second bin the trend is MUCH weaker and far more scattered, while in the third and fourth bins it's gone completely. This is despite probing even more dense environments in the more distant bins (which groups or clusters these are they don't say).

Deficiency, like asymmetry, is a difficult parameter to measure because galaxies have strong intrinsic variations anyway. You can really only use it to quantify things in very broad terms : galaxies are either gas rich, normal, somewhat deficient, or strongly deficient. And almost all the strongly deficient galaxies are present in the first bin, with very few in any other bins. In part (as they say) this is a sensitivity effect of HIPASS, which will only be able to detect the most deficient galaxies at the lowest distances.

There's another problem which they comment on : measuring galaxy density. They used a 2D parameter, which is not perfect since this can include galaxies at very different distances. So that there don't appear to be any strongly deficient galaxies in the really dense environments could just be a combination of projection and selection effects.

What of asymmetry ? That's even worse. There's a tentative, weak trend in the closest galaxies for the most asymmetrical galaxies to become even more asymmetrical at higher densities, but this isn't visible at all in any of the other bins. There's no trend between asymmetry and deficiency either : asymmetrical galaxies aren't especially gas-deficient.

Basically the conclusion of the paper appears to be, "we did this so you don't have to - please don't try this again, the data just isn't good enough". Logically, the correlations should exist. But we'll have to wait for better surveys before we can really test this - for now, this looks like another example of Simpson's Paradox. It's not that the trends aren't there, necessarily, it's just that the data can't show them. Which begs the question as to why they didn't use the far more sensitive, higher resolution ALFALFA survey... follow-up paper, perhaps ?

HI Deficiencies and Asymmetries in HIPASS Galaxies

We present an analysis of the sky distribution of neutral hydrogen (HI) deficiency and spectral asymmetry for galaxies detected by the HI Parkes All-Sky Survey (HIPASS) as a function of projected environment density. Previous studies of galaxy HI deficiency using HIPASS were sensitive to galaxies that are extremely HI rich or poor.