Is it a bird, is it a plane ? No, it's a super spiral !

"Super spiral" is not a term I've ever heard used before but apparently this is what the cool kids are calling the largest and most massive of all spiral galaxies. Whereas the Milky Way rotates at about 220 km/s, the fastest rotation of all reaches a whopping 570 km/s.

Now normally in astrophysics all the key breakthroughs are made at the margins of what's technically possible. We build a new telescope to see a wavelength we couldn't see before, or we increase sensitivity to different sorts of features, and right at the limit of what we're able to accomplish we find something unexpected. And physics is usually better behaved on the largest of scales, where nice simple gravity tends to dominate. So to find giant galaxies doing something unexpected is extra-specially unusual, but that's just what this paper has found.

There's one small caveat. These beasties are so massive that they're incredibly rare, meaning you need a large search volume to find any. So measuring their rotation in the best possible way - using a radio telescope to measure their atomic gas - is difficult, because they're just a bit too far away for that (though I suspect it will be possible given a bit more time). But they're close enough to use other reliable measures, in this case via the ionised gas.

What the author's find is that the galaxies are rotating more quickly than the standard Tully-Fisher relation predicts. There's normally a tight correlation between the rotation speed of a galaxy and its combined mass of stars and gas, but these galaxies break that relation. And they deviate systematically, as though they were a different population obeying their own TFR with blackjack and hookers. In essence, it looks as though the galaxies are either rotating too quickly or have less gas and stars than expected.

Neither of these interpretations is easy to explain. The rotation looks very solid : the rotation curve shown is clear and with minimal scatter. It's even still rising at the edges, meaning that the figure of 570 km/s is a lower limit (though I would stress that since this is only a letter, not a full paper, only one rotation curve is shown). So not much chance the rotation has been measured incorrectly. As for the masses, stellar masses are relatively easy to compute, and although the ionised gas might not be the best tracer of the total gas mass, estimates would be have to be wrong by a factor 5-50 to explain the deviation - and that's just not credible.

But actually, it might not be too difficult to explain these whoppers in standard galaxy formation models. The authors say this points to an upper mass limit for galaxy formation, beyond which it's difficult for infalling gas to cool and form stars. To exceed this mass requires mergers instead, which build up galaxy mass and rotation in a different way, neatly explaining the change of slope of the TFR.

Where it gets really interesting is for modified gravity theories like MOND. The rotation signal is very clear and it doesn't seem at all sensible - as MONDian advocates usually do - to argue that the galaxies are out of equilibrium and so not representative of stable conditions. The rotation curve is frankly lovely and shows no signs of distriburbance, and getting anything to rotate at 570 km/s is frickin' hard anyway. And the parameters (size and rotation) of the galaxies are such that they should be showing strong signs of MOND's effects, which predict that they should rotate more slowly than expected, not more quickly ! Even MOND's beloved external field effect - the influence of nearby galaxies - doesn't seem to work, as that too would decrease the rotation speed, not increase it.

Can MOND survive this latest challenge ? It's seen off many a similar bold claim before, albeit often in ways that feel like it's rendered as a barely scientific theory at all, much less a decent one. But with these... I suspect there are too few data points here to be fully convincing, but the noose is tightening.

A Break in Spiral Galaxy Scaling Relations at the Upper Limit of Galaxy Mass

Super spirals are the most massive star-forming disk galaxies in the universe (Ogle et al. 2016, 2019). We measured rotation curves for 23 massive spirals and find a wide range of fast rotation speeds (240-570 km/s), indicating enclosed dynamical masses of 0.6 - 4E12 Msun.

I'm not angry

... I'm just disappointed. Why ? I shall tell you.

One of the biggest problem with the missing satellite problem is that it heavily relies on a sample of one. There aren't as many satellite galaxies detected around our own Milky Way galaxy as models predict, which is very annoying. But is this just because our own Galaxy happens to be weird, or does it reflect a much bigger problem with cosmological theory ? There are hints, through statistical analyses of large galaxy populations, that it's the latter, but it would be far more satisfying to point to another individual galaxy system and say, "Ahah ! This system does [or does not] have the right number of satellites." Only then can we really start commenting as to why our Galaxy is strange, if indeed that's the case.

This paper presents observations which look for satellites around 10 more potential host galaxies, finding a total of 153 candidate satellites of which 93 are new. Huzzah ! Right ? Well, no.

There's nothing much wrong with what they do, so far as I can tell. They have very deep imaging and carefully mask the brightest stars and galaxies, using a semi-automatic method to find candidates which are then subjected to a good old visual inspection. Totally fine. They do photometry on the candidates to measure their colours and brightness, and plot nice maps of their distribution on the sky - perhaps we could also see if there are any more satellite planes as well. They test their procedures with simulated galaxies to see how efficient their procedures should be at recovering real galaxies with different properties. They give full catalogues of everything they've done and comment in some detail on every system they examine. All well and good so far.

Where it starts to go a bit wrong is that they don't have distance measurements for the satellite galaxies. That means they can't say how bright they are and thus how well they (mis)match theoretical predictions, or say if they're found in planes or clouds. But "wrong" is really too strong a word, though, as they do say their data is enough for them to eventually be able to measure distances using surface brightness profiles. So this is just a preliminary catalogue. Fair enough, really, there's an awful lot of work gone into this, so it should be published.

No, where things get disappointing is when they try a preliminary examination of the data under the assumption that the dwarf galaxies are at the same distance as their potential hosts. In that case, the colours and basic structural parameters of the "dwarfs" don't much look anything like other dwarfs at all. In fact they are so different that, "From these plots, it is apparent that many of the candidates are background". So all of that work and they can't even say anything much about the satellite galaxies except that these candidates are pretty lousy.

Oh well, that's the way it goes. Eventually we will get better statistics to address the missing satellite problem, but blimey - getting there is going to be like pulling teeth.

Wide-Field Survey of Dwarf Satellite Systems Around 10 Hosts in the Local Volume

We present the results of an extensive search for dwarf satellite galaxies around 10 primary host galaxies in the Local Volume (D$<$12 Mpc) using archival CFHT/MegaCam imaging data. The hosts span a wide range in properties, with stellar masses ranging from that of the LMC to ${\sim}3$ times that of the Milky Way (MW).

Tantalising tales of terrific tails

Hot on the heels of that paper on measuring the atomic gas at relatively large distances, here's one about what happens to the ionised gas.

Detecting the gas at large distances is difficult, but detecting stars is relatively easy. This gives us a bit of a confusing picture. We know a lot about the stars in very distant galaxies, but not so much about the gas they form from. We also don't know too much about how their environment changes. In particular, the major reason galaxies lose their gas in clusters is thought to be ram pressure stripping, where galaxies plough through the hot, thin intracluster gas (ICM - M for "medium") fast enough for it to push their own gas away.

The strength of ram pressure is directly correlated with the density of the ICM and the speed galaxies move through it. Those two parameters, however, have complex dependencies on the structure of the cluster. Clusters assemble through the accretion of galaxies, so in the distant past, clusters would have been less massive (meaning galaxies would fall into them more slowly) and have less ICM. So it's thought that initially gravitational interactions would have driven galaxy evolution, which have stronger effects at lower speeds. On the other hand, it's thought that galaxies were entering clusters in larger numbers in the past.

Directly detecting signatures of ram pressure stripping at large distances would be extremely helpful. These authors make a very solid case, using the MUSE instrument on the VLT. That lets them detect the ionised gas with high precision measurements of its motions. Ionised gas is easier to detect and seems to be a better tracer of ram pressure stripping than neutral gas anyway. The downside is that because of the distance (redshift) they have to look at oxygen, which is much less common than hydrogen, but this should still give a nice view of what's happening.

And indeed it does. As part of their survey, they found two galaxies in a cluster with neatly parallel tails. They're pretty long features, one 30 kpc and the other 100 kpc - not astonishingly long, but spectacular nonetheless. By themselves, these results can't really say much about the overall importance of ram pressure in earlier epochs, but they do demonstrate that it's now possible with more data. They at least show that it was definitely happening, and that there hasn't been that much fundamental change in clusters in the ~6 billion years or so since the light left these galaxies.

The objects are perhaps more interesting in their own right. They're both moving at extremely high speeds relative to the cluster, ~800 and 1900 km/s. That means the cluster is massive and that they should indeed be experiencing extremely strong ram pressure. And ionised gas, left to its own devices, will recombine to form neutral gas very quickly. They estimate that in these conditions the recombination timescale is a mere 10,000 years, very much shorter than the ~100 million years or so needed to form the tails (given their lengths and the galaxy's velocities). But there aren't any stars in the tails, so what's keeping the gas ionised ? They suggest - in very hand-waving terms, "thermal conduction, magneto hydrodynamics waves, and shocks". Alternatively it could be that much of the gas is neutral (or even so hot it's not emitting at these frequencies), so the ionised gas we're seeing is just the small fraction which is in the process of cooling.

It's also interesting to note that the tails have very high random motions, ~80 km/s. The higher the velocity dispersion, the harder it is to detect neutral gas. That may be why atomic gas isn't such a good way to trace ram pressure in the nearby Universe - combined with the rapid ionisation timescales and long neutral tails should be rare things. I'll have more to say on that one I find a referee who's prepared to actually listen, but I digress...

Perhaps the oddest thing is that the tails are neatly parallel despite the high velocity difference between the galaxies. The different velocities suggest the galaxies are entering the cluster from different sides (at these redshifts, distance measurements are none too precise). So it's a bit of a surprise that the galaxies seem to be on such similar trajectories, but that could just be a neat coincidence. We won't be able to understand that one without many more observations. As they conclude, somewhat dramatically, MUSE is, "opening a new era in the study of the role of the environment on galaxy evolution."

Evidence for ram pressure stripping in a cluster of galaxies at z=0.7

MUSE observations of the cluster of galaxies CGr32 ($M_{200}$ $\simeq$ 2 $\times$ 10$^{14}$ M$_{\odot}$) at $z$ = 0.73 reveal the presence of two massive star forming galaxies with extended tails of diffuse gas detected in the [OII]$λλ$3727-3729 A emission-line doublet.

This is not (quite) the gas you're looking for

You can't study galaxies without studying star formation, and you can't understand star formation without understanding the gas. This is easy enough in the nearby Universe, but the radio emission from gas is very weak. You don't even have to go very far before it becomes impossible to detect at all. That's extremely annoying, because while we can see how the gas content of galaxies varies with environment, it's damn difficult to measure how it's changed over time.

This paper, like several others before it, attempts to overcome this through stacking. By observing hundreds of galaxies and combining their signals, it's possible to measure the average gas content of the whole sample. This has a lot of limitations : you won't know which galaxies have gas at all, or how much variation there is : some could have none and others lots, so long as the average is above your sensitivity limit. But it means you can detect the gas at much greater distances than is otherwise possible. The highest direct detection is of a galaxy at about 3 billion light years distance, whereas this stacked sample has an average distance of about 4 billion light years and with some as far as 5.

For this project they use India's Giant Metre-wave Radio Telescope. Previously this has had a reputation for giving rather hit-and-miss results, but its recent (under-reported) upgrade seems to have given significant improvements. While interferometers like this don't have the best sensitivity to low-density gas, they can still be perfectly good for detecting the dense, star-forming gas. And their large field of view means they can observe hundreds of distant galaxies at once.

It's an interesting question as to which is the more efficient observing strategy : observing individual galaxies one by one with giant telescopes like Arecibo or FAST, or doing a ginormous GMRT survey and stacking the galaxies instead. Hard to say. Anyway, they spent almost 120 hours on a field containing more than 400 galaxies suitable for stacking, and none of them were directly detected. But combine their measurements and a pretty clear detection emerges. With a bit of smoothing to improve the sensitivity even further, it looks solid - certainly better than previous comparable claims using similar methods. They're also able to show that the gas is likely only coming from their target galaxies, and isn't a result of unwittingly combining whole groups of galaxies.

What does this tell us ? The average gas mass of the sample is about 5 billion solar masses. That's pretty high, but not extraordinarily so. It seems that the gas consumption timescale hasn't evolved much between when those galaxies were around and the present day, since estimates from local galaxies give similar results. On the other hand, it does support a change on longer timescales, adding another data point to the existing (indirect) estimates, but it's not a major development in itself.

So no revolutions here. But it's an impressive technical achievement, which ought to get the GRMT some attention as a seriously capable instrument. For understanding how the gas changes over time, though, we're going to have to wait until we have the capability to look even further back.

Atomic hydrogen in star-forming galaxies at intermediate redshifts

We have used the upgraded Giant Metrewave Radio Telescope to carry out a deep (117 on-source hours) L-band observation of the Extended Groth Strip, to measure the average neutral hydrogen (HI) mass and median star formation rate (SFR) of star-forming galaxies, as well as the cosmic HI mass density, at $0.2 < z < 0.4$.

When is peculiar not peculiar ?

More on those galaxies without dark matter. "But Rhys !" you may say. "Isn't it now widely accepted that the distance measurement was wrong and they're not that strange after all ?"
"Yes," say I. "But don't forget, other, similar candidates have also emerged, and it's much more difficult to explain them by dodgy distance measurements. So this new paper may still be relevant even if it's about the wrong galaxies."

How weird are these objects ? It depends on which model of cosmology you adopt. The vast majority of galaxies appear to be rotating so quickly that they ought to fly apart. The standard model says, "okay, that's weird, there must be some undetected enormous mass holding the galaxies together." Modified gravity theories say, "nope, there must be a problem with our current theory of gravity."

It's perfectly possible in standard cosmology for dark matter and stars to remain separate, albeit unusual. But in modified gravity cosmologies, if you have two galaxies which are basically identical (same mass and same mass distribution), then they should always show the same rotation speed since gravity should always work in same way. For one of them to be rotating much more slowly - as in these objects - is seriously weird. A significant caveat is that some modified gravity theories depend strongly on the large-scale distribution of material, so it's not really as simple as that - galaxies close to other galaxies can be expected to show the low rotation speeds.

Even though it's possible (in the standard model) to separate dark matter and stars, this is still really weird for objects as large as these. This has prompted claims that actually the distance measurement of these galaxies in incorrect : if they're far away then they'd be weird and lack dark matter, whereas if they were closer they'd be perfectly normal and dark matter dominated. I've become persuaded that the latter explanation is probably correct.

(Or, if you prefer, if they're close by then their dynamics are typical of galaxies regardless of which model you adopt, whereas if they're far away then their dynamics are much more unusual.)

The problem with this, though, is that if the galaxies are nearby then they must be moving with strong peculiar motions along our line of sight. That is, they deviate from the overall large-scale flow (the Hubble expansion of the Universe). This is common enough in massive groups or clusters, where the mass of other galaxies can accelerate them, but it's less obvious how this could happen for more isolated objects. So does that in itself challenge the standard model ?

According to this paper the answer is "yes", but I would tone this down to a cautious "maybe". They search the latest state-of-the-art simulations for similar galaxies and assess how frequently such objects are found. This is a common strategy, but as always, the devil's in the statistics : just what is a fair comparison in this case, and what does probability really mean ?

So far as I can tell, they assess probability by looking at the fraction of galaxies with similar parameters (size, velocity, and both together) of galaxies in a very generous mass range (40 million to 4 billion solar masses). On the one hand, this high range gives a huge scope to find similar objects, but on the other hand, if such objects are only common over a small mass range, then this could be misleading as to how common they really are. They explore this over a range of distances, correcting - I think - for the respective change in properties that implies. That's more reasonable, and a nice way to show which sorts of objects are really more common.

They find that if the parameters are restricted to just the structural properties, then such galaxies are indeed most common at the lower distance estimates. But if their peculiar velocities are used instead, then they're much more common at the higher distances. Overall, the latter dominates, so such objects are actually more compatible with the standard model if they're far away. In other words, normal objects moving weirdly are harder to explain than weird objects moving normally. Galaxies without dark matter are less contradictory to the standard model than normal galaxies moving against the general Hubble flow.

But then again, these galaxies have a weird population of globular clusters too. Taking that into account as well, the result switches back : galaxies with similar structures, velocities and globular clusters all together are more common in the simulations at low distances. So we're back to ordinary, dark matter dominated galaxies moving weirdly.

How improbable are such objects ? That's where it gets tricky, not least because the results are strongly dependent on which simulation is used. In the best case, such galaxies account for about 1 in 10,000 of the overall population, while in the worst case they make up more like one in a million.

Given that we're talking about a total galaxy population in the visible universe measured in the billions at the very least, are these numbers worrying ? Maybe. As far as I can tell, they examined only one fixed region in the simulations, extending to their maximum allowed distance of 20 Mpc. This means that if such objects are more common elsewhere then they won't be detected.

That's probably a small effect though. What's more problematic is that they don't discuss the formation scenarios of the rare matching objects that they do find. They could, I suppose, be formed purely by happenstance, but if there's a physical mechanism at work then they could be very much more common in certain conditions. So the low probabilities may not mean anything : if the real objects do indeed have similar environments, then this would be a powerful vindication of the models, not a refutation. Worse of all, because their are two (real) galaxies, they multiply the probabilities together to say that it's fantastically unlikely that we'd really detect a system like this. But you can only multiply probabilities if they're independent, which they have not demonstrated either for the simulation or observational reality.

So overall I'm not at all convinced these galaxies are going to be a challenge for the standard model. It's true that the peculiar velocities are worthy of further investigation, but those who challenge the standard model have this odd habit of saying, "the model can't explain this" without actually checking if this is true. Which is a shame, since they've already identified similar objects in the models, so it would be relatively straightforward to investigate them.

The model-dependent result is also important. These simulations are very fancy, but that doesn't mean they're flawless. Indeed, their staggering complexity is a weakness. It's entirely possible and credible that their input physics will need substantial improvements before we can even use them to make sensible comparisons to reality, without needing to change the basic dark matter paradigm. I don't think this result is wrong so much as I think that it's simply premature.

The ultra-diffuse dwarf galaxies NGC 1052-DF2 and 1052-DF4 are in conflict with standard cosmology

Recently van Dokkum et al. (2018b) reported that the galaxy NGC 1052-DF2 (DF2) lacks dark matter if located at $20$ Mpc from Earth. In contrast, DF2 is a dark-matter-dominated dwarf galaxy with a normal globular cluster population if it has a much shorter distance near $10$ Mpc.

Gas off a plane

Galaxies are pretty disgusting things, and understandably so. They whirl around at hundreds of kilometres per second, with their inner parts moving at different speeds to their outer regions, occasionally bumping into each other and even swallowing other galaxies accidentally. It's no wonder they feel a bit queasy.

If two galaxies have a close encounter, they can rip off parts of each other's gas. Just as it's easier to launch a rocket into orbit from the equator than the pole, so it's easier for galaxies to disturb each other if the interaction is aligned with their planes of rotation. Their gas is already rotating rapidly, so it only needs a slight shove to remove it completely - whereas to remove it vertically requires a powerful kick, since the gas is barely moving vertically at all.

But some galaxies show gas which is quite clearly well above their main rotating plane, so how does it get there ? There are several ways. A sufficiently strong interaction can do it, but would usually be heavily disruptive to the stars as well as the gas. In certain environments, galaxies can move through external gas which can build up a sufficient "ram pressure" to push the gas out of the disc. If a galaxy gobbles up a much smaller companion, the unlucky dwarf can be ripped apart without much affecting the giant. And even galaxies that live in isolation can belch out enormous amounts of gas through stellar winds and supernovae explosions, producing "galactic fountains" of vomited gas that eventually rain back down elsewhere in the galaxy's disc.

All this is an important part of the baryon cycle of how matter is arranged in the Universe. If we want to understand how galaxies form stars, then we have to study their gas. And if we want to estimate for how long they're likely to continue to form stars, we need to look at this extraplanar gas. It's often difficult to work out if we're seeing gas that's being expelled or accreted onto a galaxy's disc.

Actually, it's often hard to even decide where the galaxy's disc begins and the extraplanar gas begins. This paper attempts to model that using high resolution gas measurements taken with the Westerbork radio telescope. Most galaxies are not seen directly face-on or edge-on to us but at an angle. The radio data gives us information not only about the position on the sky but also how fast the gas is moving along our line of sight. By modelling how the main gas disc is expected to move, the authors believe they can disentangle the disc and extraplanar material even when the two are similarly aligned on the sky.

Since the data is of very high resolution, at any pixel the spectral profile displays a simple Gaussian shape in the disc (that is, the brightness peaks at a certain velocity and falls off smoothly at higher and lower velocities). Extraplanar gas is found in distinct "wings" at higher and/or lower velocities. So by fitting the Gaussian to the brightest parts of the profile, this can be removed to leave behind only the gas outside of the disc, which is moving quite differently to the disc material. This is pretty ingenious because otherwise it would be very difficult to remove only the disc material without also taking a big chunk of the extraplanar gas along with it.

One would imagine that at this point the disc can simply be masked, leaving behind a nice image of only the extraplanar stuff. But that doesn't seem to be the case, as they go on to model the shape and rotation of the extraplanar material. I cannot say I fully understand how or why they do this (it wasn't clear to me if this is somehow a necessary part of the process or just a nice bonus to extract more information), and there are clearly a lot of uncertainties in how they do this, but they claim that this means they can even determine if the gas is flowing towards or away from the disc.

They find that the typical extraplanar gas in this sample accounts for about 15% of the parent galaxy's gas disc. That's a fair old chunk, and it seems that this material is a common feature of galaxies - so it's unlikely to be due to some external, environmental effect. Galactic fountains therefore appear to be normal. More unexpectedly, material tends to be found only one one side of the galaxy and is always flowing into it. They say that this is because they're looking at neutral gas, while gas ejected by supernovae explosions should be ionised. But this isn't convincing, as there are plenty of large neutral bubbles known. As to why the material should be lopsided, I don't know.

This is definitely a very clever paper with an interesting technique and results. I do wish they'd shown nice simple maps of the extraplanar gas though; they show it in rotation curves but a simple map would have been better (unless I've misunderstood something fundamental about their method). I imagine that this can also be applied to many more galaxies which have suitable observations, so it'll be fun to see this applied more widely.

HALOGAS: the properties of extraplanar HI in disc galaxies

We present a systematic study of the extraplanar gas (EPG) in a sample of 15 nearby late-type galaxies at intermediate inclinations using publicly available, deep interferometric HI data from the HALOGAS survey. For each system we mask the HI emission coming from the regularly rotating disc and use synthetic datacubes to model the leftover "anomalous" HI flux.

I told you so

Over two years ago I reported that there's something weird about Ultra Diffuse Galaxies : they're rotating too slowly. When you plot these objects on the Tully Fisher relation, i.e. comparing their rotational speed to total mass, you see a clear, enormous offset. They are solid, unambiguous outliers from what's normally a pretty tight relation.

Finally the team I met in Tenerife have formally published this result as a Letter. This is not as detailed as a full article (which I presume is forthcoming), but all the major points are addressed. In fact it's even more interesting than I was expecting.

There are two main ways to explain a deviation from the standard TFR. The first is that the rotation speed could have been incorrectly estimated. This was my main concern when I saw the original data, which comes from the ALFALFA survey. Arecibo has extremely high sensitivity but doesn't have the resolution to map exactly how the gas is moving. It can place a lower limit on how fast the gas in a galaxy is moving along our line of sight, but that's not the same as measuring how it's rotating. In principle, the gas could actually be rotating much more quickly than the basic estimate implies. And because of selection effects, it's possible that the sample was unexpectedly biased towards detecting galaxies which looked like they were rotating more slowly than they actually are.

This was a real possibility, but not a very likely one as even the first paper had some observations taken with less sensitive but higher resolution telescopes. The authors here do the same for three more galaxies, giving them a sample of six. Once you can resolve a galaxy, you can determine its rotation much more accurately : you can correct for the effect of viewing it at an angle, and you can see if its motions display the pattern characteristic of ordered rotation or if it's doing something more chaotic. All six galaxies here show ordered rotation (though it would be nice to see the full maps), and the viewing angle definitely isn't enough to explain the deviation. This effectively demolishes the argument that the sample is biased.

The second way to explain the offset is that the total mass (here meaning the combination of gas and stellar mass) is incorrect. The problem with telescopes that give you nice resolution is that they, for technical reasons, give you crappy sensitivity. But in this case that can't be a problem, since they measure about the same amount of gas as for the earlier, higher sensitivity observations. The only other way the mass could be wrong is if the galaxies were much closer than expected. But that simply isn't possible for these objects - it would require them to have extremely high peculiar velocities, and since they're all pretty isolated objects this is just not credible.

So the galaxies really do seem to deviate. They've got rather high gas masses, around a billion times the mass of the Sun - not far off how much there is in a typical massive spiral. Whereas the Milky Way rotates at 220 km/s, these guys rotate at just 30 km/s or so. They don't have high velocity dispersions so there can't be much in the way of stellar winds. They seem to be in equilibrium, since they have nice ordered motions, so they can't have experienced some recent event that could have disturbed their kinematics. They're also isolated : at least 350 kpc from the nearest galaxy with a mean distance of 1 Mpc. So there's not much chance of them being produced in tidal encounters between galaxies. They really do seem to be primordial galaxies that rotate very, very slowly.

There's been a lot of hoo-hah about galaxies lacking dark matter recently. The consensus seems to be that those objects have incorrect distance estimates. But that's not credible for these objects (although surely people will investigate this too), so it really does seem that such weird galaxies do exist after all : their position on the TFR is exactly consistent with them completely lacking dark matter. They also seem to have no missing baryons.

What does all this mean ? Potentially, lots. No galaxy formation theory I've ever heard of predicts the existence of objects like this, so on the one hand, this is a major and unexpected development. The main question is why objects with similar gas masses apparently have radically different levels of star formation activity. Why do some gas-rich objects, the normal galaxies we're familiar with,convert most of their gas into stars, while others, like these, barely form any at all ?

On the other hand, the good news for the standard model is that these dark matter deficient galaxies could be strong evidence that dark matter really does exist. That sounds ironic, but in modified gravity theories the appearance of dark matter (i.e. high rotation speeds) arises solely from the gas and stars. Thus, any object of the same mass and distribution should show the same fast rotation. That these objects don't is likely to prove a big headache. It's easy in the dark matter model to claim that some galaxies simply don't have dark matter, whereas it's very difficult for modified gravity theories to say that identical objects should rotate differently. And the low scatter in the TFR has long been a puzzle, so this may finally solve this too.

Of course, watch this space. Hopefully this will be the start of the next interesting controversy !

EDIT : There's a second, quite similar paper here. I'm not giving it it's own post for several reasons : it doesn't cite the original Leisman paper, it uses an outdated version of the ALFALFA catalogue, some sentences are garbled, and one graph doesn't have proper axes labels (and it's unclear to which journal - if any - it's been submitted). It also uses only the ALFALFA data and doesn't have resolved observations. Still, the population of galaxies it describes is different to those presented in the other papers and their deviation from the TFR, and I can find no reason to dispute its main result.

Off the baryonic Tully-Fisher relation: a population of baryon-dominated ultra-diffuse galaxies

We study the gas kinematics traced by the 21-cm emission of a sample of six HI$-$rich low surface brightness galaxies classified as ultra-diffuse galaxies (UDGs). Using the 3D kinematic modelling code $\mathrm{^{3D}}$Barolo we derive robust circular velocities, revealing a startling feature: HI$-$rich UDGs are clear outliers from the baryonic Tully-Fisher relation, with circular velocities much lower than galaxies with similar baryonic mass.

Being edgy for the sake of it

There are no less than three papers about Ultra Diffuse Galaxies on astroph today. The one I'm most interested in is an attempt to find UDGs which are edge-on with HI. That means the HI line width will give a very accurate indication of their true rotation speed (and thus total mass) without needing to do any inclination corrections, which may have irritating selection effects. In other words, this could give a good indication of whether these things are massive galaxies with absurdly low star formation rates, or just very extended dwarf galaxies. This is the central mystery and controversy of UDGs.

The selection effect at work with inclination is a little complicated, but it's worth being aware of. Suppose that we have a galaxy with a given mass of hydrogen. The signal to noise of the hydrogen signal will be given by the mass, rotation speed, and viewing angle. If the viewing angle is face-on, then the S/N will be highest because all the gas will be moving at the same velocity relative to us. If it's edge-on, then the S/N will be lowest because the galaxy is rotating, spreading out the hydrogen flux into multiple frequencies. So we will preferentially detect face-on galaxies if we look for them using hydrogen.

For optical wavelengths it should be the opposite. If they're face on, then their optical light is spread out over the largest possible area, minimising the S/N. If they're edge-on, then we're seeing all their stars compressed into the smallest area, maximising the S/N. So edge-on galaxies should be easy to detect in the optical but difficult using hydrogen. In reality it will be a bit more complicated than that : a large, faint thing may be easier (in some circumstances) to detect than a small faint thing, but as a general guideline it's reasonable.

The other complication is that surface brightness (i.e. no. stars per unit area on the sky) has to be corrected. You can't just measure it directly off the sky unless the galaxy is face-on. So correcting this when the galaxy is edge-on, and has all sorts of internal extinction effects (e.g. absorption by dust and gas), is tricky - but possible.

The first attempt to find UDGs in an HI catalogue looked at isolated objects and didn't particularly concern itself with viewing angles. These authors take the exact opposite approach, but still find that their galaxies are primarily in low density environments. That's not too surprising, since in a higher density environments there are effects which can trigger star formation that consumes the gas and/or directly remove the gas.

What I don't understand is their selection criteria. For the optical data they use SDSS DR7, but we're now on DR15. For the HI data they use the ALFALFA 40% complete catalogue, but the 100% complete has been out for months and the 70% one for ages and ages. Then - I think - they do surface brightness profile measurements to find objects with similar shapes to UDGs, although this isn't clear to me. Even the 40% catalogue has over 10,000 members so doing this in an automatic way is tough. So I don't understand how, but somehow they reduce this down to a catalogue off 11 candidate edge-on UDGs, after correcting the surface brightness for inclination. I found this section to be a bit odd and not at all clear as to what they did or how and why they did it.

Regardless, they find that their galaxies look relatively normal. They're very blue, edge-on discs, but nothing extraordinary. That suggests to me that there could be large numbers still undetected. They are, however, extremely gas rich, with HI mass-to-light ratios of between 5 and 30. So their mass really is dominated by gas, strongly suggestive of low star formation efficiency. Their most interesting plot shows that in color-magnitude space, UDGs found in isolation and those found in groups occupy distinctly different regions of parameter space, consistent with the idea that they could be formed by different mechanisms.

What about the all important dynamics ? Amazingly they say little about this - I don't know why people keep doing that. But their HI spectra look quite normal : broad velocity widths with double-horn shapes. Line widths are not extraordinarily high though, with a maximum of 150 km/s. That's in the upper range for dwarfs, but definitely not giants. I did a quick inspection of the Tully-Fisher relation, and their line widths and gas masses are, unlike some other UDGs, consistent with normal galaxies. That lends some credence to the idea that UDGs with unexpectedly low line widths are the result of a selection effect and their inclination angle hasn't been corrected properly. On the other hand, since they were here looking for edge-on galaxies, then they might have excluded those with low line widths by definition.

Overall, the general view that UDGs are mostly dwarves looks to be consolidating its position. It's still too early to say if this is true for all of them though, and the difference between UDGs in low and high density environments has yet to be explained.

Edge-on HI-bearing ultra diffuse galaxy candidates in the 40% ALFALFA catalog

Ultra-diffuse galaxies (UDGs) are objects which have very extended morphology and faint central surface brightness. Most UDGs are discovered in galaxy clusters and groups, but also some are found in low density environments. The diffuse morphology and faint surface brightness make them difficult to distinguish from the sky background.

The complexities of the Kent complex

I'll do a proper outreachy post with memes and stuff about this soon. For now, a brief summary.

AGES was such a fun survey we decided we didn't want to stop, so in effect we've kept going. The new survey, WAVES (Wide-field Arecibo Virgo Extragalactic Survey) will only cover the Virgo cluster, but with the ultimate goal of giving full cluster coverage to the AGES sensitivity level or nearly so. That will, amongst many other things, let us work out how common all those optically dark clouds really are, where they're found etc., and so give much better constraints on what the the little blighters actually are.

But there's much more to Virgo than poxy little clouds. For one thing, there's the massive Kent complex, the main target of the first WAVES field. This was discovered using the shallower ALFALFA survey and had some follow-up with the VLA. As we all know, interferometers just aren't as sensitive as single-dish telescopes. So since this is a big, extended feature, surveying it with Arecibo but to a higher sensitivity level (a factor of four) seemed like a good idea. And it was.

The Kent complex has a unique set of properties. It's 150 kpc in projected length, though there are other clouds known which are longer. It's got a gas mass comparable to a large galaxy, but other clouds of similar masses are also known. It's got complex kinematics, but so do others. It doesn't have an obvious parent galaxy, but neither do other clouds. But none of the other features have all of these properties - in Virgo at least, this is unique.

Our observations show that it's probably weirder than Kent gave it credit for. They show that the mass is more than a billion times the Sun, more than doubling the value from Kent. In our data it looks for all the world like a collection of galaxies, except that there's nothing visible at optical wavelengths at all. Importantly, Kent's VLA data rules out the dark galaxy hypothesis because there's no sign of ordered motions either of the complex as a whole or within individual clumps. We also found that much of the gas is in diffuse, extended emission, as well as in the discrete clouds that Kent found. We show that these clouds are indeed directly connected to each other, embedded in a sort of common envelope. And the whole thing looks a lot like a rhino.

How does this change how we interpret the cloud ? Kent leaned towards either NGC 4445 or NGC 4424, both reasonably nearby spiral galaxies, as the possible parents. Either ram pressure or tidal harassment (or both) could potentially have removed the gas. But honestly, neither of these candidates are great. They don't line up with the major axis of the cloud at all - in fact, they're basically orthogonal to it. That doesn't match any other similar features, and the fact that they're completely detached is also weird : ram pressure doesn't seem to do that, and removing this much has through tides is extremly difficult. We considered a scenario where one of the galaxy lost all its gas very rapidly, which then formed an expanding cloud, but while that might work (actually we tried quite a few models which didn't make it into the paper - we might do a separate paper in future), it's hard to physically justify. And NGC 4424 has a tail which points in exactly the wrong direction, so it's almost certainly not that one.

There's another galaxy in this region which is one of the iconic images for ram pressure stripping : NGC 4522. This is already known from VLA data to have a short tail pointing towards the Kent complex. We found it has a second tail, of similar alignment, several times longer, but at a different velocity. The direction alignment towards the complex is awfully tempting... but the velocity of the galaxy is 1,800 km/s different to the cloud. Even in a cluster, where galaxies swarm like angry bees, that velocity difference is just too high for the two features to be associated.

The final weird thing is just how massive the Kent complex is. In almost all galaxies with hydrogen streams, the mass in the tail is only a few percent or so of the mass in the disc. For reasons we don't fully understand, the gas appears to rapidly evaporate and/or disperse once removed from its host, likely because of the very hot intracluster medium. The problem is that for any of the plausible donor galaxies, hardly any gas at all would have evaporated. And there's no obvious reason at all why the Kent complex should retain almost all of its gas as detectable atomic hydrogen whereas in all other cases the majority of it becomes undetectable.

So what the hell is it ? We don't know. NGC 4445 remains the least worst candidate parent galaxy, but we mean that literally : it's only the best of a bad bunch, and not at all convincing. As for how the cloud formed, we just don't know.

The Widefield Arecibo Virgo Extragalactic Survey I: New structures in the Kent cloud complex and an extended tail on NGC 4522

We are carrying out a sensitive blind survey for neutral hydrogen (HI) in the Virgo cluster and report here on the first 5° x 1° area covered, which includes two optically-dark gas features: a five-cloud HI complex (Kent et al. 2007, 2009) and the stripped tail of NGC 4522 (Kenney et al.

IAU Symposium 355 : The Realm of the Low Surface Brightness Universe

I've decided to do everyone a favour and not combine the science and pretty pictures of landscapes from the latest conference. You can read about the travel experience of Tenerife here. Although I've already mentioned some of the outcomes, I thought I'd also give a more executive summary of the stuff I found most interesting.

The theme of this conference was the low surface brightness universe - the faint fuzzy stuff. It covered the whole range of such features, focusing on galaxies but also including zodiacal light, the challenges of observing (including a very nice talk from an "amateur") and data processing, philosophy of science issues and reproducibility (it's tricky to get really deep images, and sometimes they give remarkably different results), and was just generally awesome. So here are my highlights.


Mike Disney's introductory talk 

Mike is essentially the conference godfather, who predicted the existence of large numbers of low surface brightness galaxies way back when. He began by noting just how vicious the field can be - and he's right. The conference in Cardiff back in 2007 was the first I ever went to, and by far the most bitter and acrimonious. I'm not sure why it went down that way, but this one didn't. People said plenty of controversial things (especially Mike !) without ever seeming like they were about to start beating each other up.

Mike's major points were that there should be a large population of hitherto unseen faint galaxies, though he tried earnestly to present arguments both in favour and against this. While everyone agrees that galaxies certainly get a lot more interesting the deeper you look (in particular elliptical galaxies tend to look much more interesting), it's more controversial as to how many brand new galaxies this will pick up (more on that later). I've covered some of these arguments before, but I'll be revising some of them thanks to this conference.

One of the issues I don't think people quite got nailed down was the significance of the low surface brightness galaxies. Many good arguments were presented that such hidden galaxies cannot contribute much to the total amount of starlight. This is probably true, but misses the point that such galaxies could still be dynamically massive and dark matter dominated. So perhaps both sides are right, depending on whether one thinks of stars or dark matter as being more important.

Mike's other controversial point didn't get so much attention, probably because the conference wasn't really focused on it : the galaxies seen in the spectacular Hubble Deep Field, he says, cannot possibly be the progenitors of today's galaxies because they're too bright. I'm skeptical, but there wasn't time to go into this much.


Johan Knapen's data talk

This was essentially a philosophy of science talk from the perspective of the sheer data size that's coming our way very soon. This is a very real and serious challenge, with the SKA expected to produce exobytes of data per second. While Mike raised the point about defining the scientific method being very difficult, Johan took this a but further. He listed four possible paradigms of science that have changed over time :

1. Experiment-driven, as in the days of Newton and Gallileo
2. Theory-driven, as in Einstein and other analytic theoreticians
3. Numerical simulations
4. Data exploration

Other suggested that the fifth paradigm could be A.I. while the sixth would be letting Facebook do everything.

Johan's point (if I remember correctly) was that astronomy was going to have to move away from the traditional hypothesis-testing method we learn in schools and towards this "fourth paradigm" of being a data-driven approach. He's not wrong, but those who I talked to seemed to agree that this already the case - and maybe always has been. I've made the point before at length : there's more than one right way of doing science, and the data usually tell you something very interesting but utterly unrelated to what you were interested in. I mean this very literally. And I believe it was Simon Driver who, in discussions afterwards, said that in experiment proposals it should be absolutely legitimate to describe which area of parameter space you wanted to explore and why it was new, without needing to say exactly what you expect to find there. With this I fully agree. It's largely a waste of time describing observational results before you've got them.

Mohammad Akhlaghi's reproducibility talk

Mohammad doesn't like the fact that papers often contain highly vague instructions as to how experiments were carried out and how data was analysed. He's trying to tackle the latter issue by developing a package that makes it very easy to document the full details of the software used, by including the exact software name, version, and all its associated dependencies (libraries etc.), in a way that makes it simple to include in a paper. It'll also let you automatically update any numbers if you change the software without having to redo the calculations or edit the paper yourself, and it doesn't require any special software modules to install : the point is that the user has full control over what packages they use.

I was a bit skeptical listening to this, being acutely aware that many aspects are entirely subjective, but I came around to it afterwards. You'll have to take care that if you make changes, your new numbers are still consistent with your original conclusions. And I'm a bit doubtful that there are that many cases where changing software actually changes the answer. But the basic idea that papers should be as reproducible as possible is something I can definitely get behind - provided we remember that objective, repeatable measurements can still be absolutely wrong. Reproducibility means you can find the errors, not say, "my method is objective and therefore objectively correct", which is an easy trap to fall in to.

Thomas Sedgwick's hunt for dark galaxies with supernovae

This was one of the most interesting and novel methods proposed for finding very faint galaxies. As long as the galaxy is forming some stars, it will have a few supernovae, and these can be detected. This is not something I would ever have thought possible - even though supernovae surveys are now decades old, I just don't think of stochastically exploding stars as something you can do a survey of. But you can. And you can even work out how many galaxies your survey implies, given some very reasonable, justified assumptions about the survey completeness and star formation rate.

Interestingly, it turns out that these corrections imply a galaxy distribution that's fully compatible with the standard model : that there are indeed large numbers of very faint galaxies out there, just as models have predicted but observations failed to find. This is a really cool result, but it relies on an enormously large statistical extrapolation, so it's probably safe to assume the problem isn't solved just yet (and kudos to the speaker for saying as much).

A somewhat similar talk was given by Raja Guhathakurta on looking for faint galaxies by searching for their globular clusters. The difference here is it should be possible to get much better completeness of the sample. We're planning to do a search for such features for the Virgo clouds I've been working on.

Nushkia Chamba's new definition of the size of galaxies

Nushkia knew me as "that guy with the hilarious blog", which absolutely made my day. She's come up with a new parameter for the size of galaxies, which is extremely interesting. If correct, it'll dramatically change how we think about ultra diffuse galaxies. But she asked that this not go on twitter, and I assume that includes other social media so I'll say no more about it. Expect to hear a lot more when it's published.

Daniel Prole's talk on the abundance of UDGs in the field

Are ultra diffuse galaxies a very common galaxy component that until recently went largely undetected, or are they just a smattering of exotic objects ? Several people made the point that UDGs were already know but it's their abundance in new surveys that's got people excited about them again. Daniel (who has the same PhD supervisor as I did) is attempting to estimate how abundant they are in the field, which is much harder than in clusters and has a far large volume. He's got a number, but rather surprisingly hasn't compared this to other numbers so it doesn't mean much yet. I would have thought there are already numbers for more typical galaxies, but perhaps getting a fair comparison (e.g. correcting for survey biases) is harder than you might think.

Pavel Mancera-Piña's talk on the dynamics of UDGs

I was so glad to see this talk. You may remember that I've commented several times on the weird line widths of UDGs with HI detections, which tend to be much less than expected (scroll to "things are getting weird" in that link for a plot). I emailed a couple of people about it - I got a cautious response from one and nothing from another. I've shown several people and they all think it's interesting, but I never have time to work on this myself. Thankfully Pavel does, and he's done a much better job than me of demonstrating that this weird result probably isn't due to observational constraints : these galaxies do seem to be weird. While it looks unlikely that certain famous candidates are not actually galaxies without dark matter, some of these UDGs might soon resurrect this possibility.

My one major concern, which I think should be relatively easy to address, is survey incompleteness. At any given mass, galaxies of low line width are easier to detect. If galaxies are rotating discs, it so happens that means we'll preferentially detect ones which are close to face-on from our perspective, which makes it hard to estimate their true line width (details here). So it might be that the survey is biased towards nearly face-on galaxies - and because they're so damn faint, it's hard to measure their inclination angle directly. In principle one could test this by calculating the line width they'd need to escape detection and the corresponding inclination angle required to reach this. However, I very much doubt this will explain all the objects. Many of those galaxies with reasonably clear detections look quite convincingly close to edge-on, implying a negligible velocity correction.

Freeke van de Vort's talk on simulations the circumgalactic medium

Freeke has done some spectacular simulations of the gas structures around galaxies. Normally I think of this as probably very diffuse, fluffy stuff, interesting but not particularly photogenic. But Freeke's simulations really are spectacular - they look like the sort of thing you'd get if you told a Marvel CGI artist to "make some pretty gas - really go nuts with this". She notes that the simulations aren't yet converged. While they get the major galactic structures right, increasing the resolution keeps changing the results for the CGM. She's also found some gas clouds without stars or dark matter, but of course the resolution dependence makes the significance of this hard to assess.

Honourable mentions

There are too many to mention properly but I can't avoid a few others :

• Eva Grebel made the point that there are a few red, isolated UDGs known in the field, while several people (but particularly Anna Ferré-Mateu) noted that at least some UDGs may indeed be giants even if they aren't the majority. Most people seemed happy with the notion that there may be several different ways to make a UDG. 
• The Dragonfly team defended their data reduction procedures in the face of their failure to detect the double arms of NGC 5907, although no-one seems to know what happened. Another talk showed us that the double arms had been detected independently so they're almost certainly real.
• Gaspar Galaz said that there's a large linear stream extending from Malin 1, which is just weird. How you get a linear stream intersecting a stellar disc, I just don't know.
• Everyone agreed that galaxies look much nicer with deeper imaging but that it's jolly hard to do. 
• Bärbel Koribalski gave a very nice overview of extended optically dark gas features. Very nice to know that it's not just me working on this !
• Sarah Pearson showed how we'll soon have detections of large numbers of globular cluster streams around galaxies. I don't think of streams in statistical terms, but this could be an interesting way to constrain the behaviour of galaxies and their dark matter content.
• Anna Saburova, a collaborator of mine, sounded like she was about to kill everyone (it's the Russian accent) but described how difficult it is to explain giant low surface brightness discs. They most likely have different formation mechanisms - some by catastrophic collisions, others through slow accretion.
• Sebastiano Cantalupo explained dark galaxies at high redshift, which I was surprised to hear may be not all that dissimilar to the candidates at low redshift, with similar masses and dynamics. Definitely one I need to read up on more as I'd assumed the high redshift objects would be very different.

Which just about wraps it up. Plenty of background reading to do until the conference proceedings are released.