The galaxies that quenched backwards

Today's paper is about contrarian galaxies that don't play by the rules. Most galaxies, left to their own devices, build up a big bulge of stars in the middle as the gas density is initially highest there. But this high star formation rate burns through its fuel very quickly and soon pitters out, "quenching" the star formation in the centre. Meanwhile the disc happily ambles along forming stars at a stately, steady pace, unhurried but longer-lasting.

There are some, though, which appear to do just the opposite. In clusters this is easy. Ram pressure preferentially removes the outer gas first, leaving the most stubborn gas remaining in the centre to carry on forming stars while the disc slowly reddens and dies. What's weirder is that there appear to be some galaxies like this in environments where ram pressure can't be playing any role. Ram pressure requires a hot intergalactic medium, which is pretty much only found at any significant levels in massive clusters. Everywhere else it should be far too weak to cause this kind of damage*.

* It probably doesn't have zero role to play outside clusters though. Very small satellite dwarfs can experience significant ram pressure from the hot gas of their parent galaxies, and there's some evidence that larger galaxies can experience at least a little gas-loss due to ram pressure in large-scale filaments. But probably not anywhere enough to account for galaxies like this. What they more likely experience is only starvation, where their outermost, thinnest gas is removed but nothing from within their denser discs. This means they can't replenish their gas as it gets eaten up by star formation. 

This paper is about a particular sort of these kind of galaxies, which they give the ugly name of "BreakBRDs" : Break Bulge Red Discs. The "break" refers to a particular spectral line indicating that there was star formation very recently in the bulges. "Blue Bulges Red Discs" might have been easier, but technically the bulges aren't actually blue, so this wouldn't be right. Even so, BBRDS would be a better acronym. Or heck, I'll just call 'em backwards galaxies.

I have to say I both like and dislike this paper. On the one hand it's very careful, thorough, and doesn't draw any overblown conclusions from the limited data. On the other, some of the discussion is long-winded, non-committal, and rather tedious considering that in the end the conclusion is so indecisive. I think there's some really great discussion on each individual scenario proposed to explain the backwards galaxies, but the collective whole becomes at times very confusing. It feels a bit like a paper written by committee. The main problem is nothing to do with the science at all, but the structure : there isn't a good unifying framework to tie all the different scenarios together. 

Here I shall try and simplify and disentangle things a bit. 

They begin with a nice overview of the observational difficulties of establishing what's going on. Some results support this classical picture of outside-in quenching where ram pressure (or other environmental effects) strip the outer gas first. But others find that this happens more in low density environments, exactly the opposite of what's expected ! Still others show something entirely different, where quenching happens irrespective of position in the galaxy, i.e. the whole disc quenches everywhere, all at once. Even simulations aren't much help, with galaxies similar to their BBRDs being found but with no clear mechanism responsible for what happened.

The particularly interesting thing about BBRDs, i.e. backwards quenchers, is that they exist over a wide range of stellar masses and apparently in all environments. Unfortunately they don't say anything else about the environments(s) of their particular sample, which is my only scientific quibble with the paper. Anyway what they do here is look at a sample of about a hundred or so which have HI gas measurements, using a combination of existing and their own new observations.

Their main result is actually quite simple, but you need to be aware of the colour-magnitude diagram first. Very simply, there are :

• Galaxies which are blue, have lots of gas, and are forming stars as per usual (the so-called star formation main sequence, or more simply the blue cloud).
• Galaxies which are red, don't have any gas, and aren't forming any stars : the red sequence.
• Galaxies which are intermediate between red and blue, have a bit of gas, and are forming stars more slowly : the green valley (or transition region).

Of course, regular readers here will know that it isn't as simple as that, but this'll do for what's needed here.

They show that BBRDs and GV (green valley) galaxies have about the same gas fractions regardless of their stellar mass, both considerably lower than those in the blue cloud. But given their star formation rates, BBRDs would burn through their current gas content very much faster than GV galaxies. That is, for the same amount of gas, BBRDs are consuming their gas far more efficiently than GV galaxies. They're forming stars at rates typical of blue cloud galaxies, even though they've got way less gas. GV galaxies, by contrast, are forming stars even slowly than those on the main sequence.

How do they do this ? Or, conversely, what suppresses the star formation in GV galaxies ?

This is not at all easy to answer. For the GV galaxies, their gas depletion times seem to be low because of their extremely low star formation rates, which more than compensates for their lower gas contents. For the BBRDs, their rapid gas depletion times are easier to explain, as their star formation rate remains normal despite their low gas contents.

But what's behind it all ? There are three main options.

1) The final stages of quenching

These BBRD galaxies could be experiencing the end stages of gas removal due to some external mechanism, as is usually the case in clusters. Perturbations to the gas could drive inflows into the centres of the galaxies, resulting in an increase in gas density and thus an increase in star formation rate. Unfortunately they don't have the detailed maps of the gas needed to say if this is the case or not, but here it would have been useful for them to say something more about environment.

2) Accretion

It could be that these galaxies were in fact already fully quenched and are experiencing something of a renaissance. Accretion is a somewhat controversial topic because it's hard to ever determine that it's really happening with any certainty, but clearly galaxies have got to get their gas from somewhere. One possibility is so-called cold accretion from the cosmic web, where the hot, diffuse extragalactic background cools and condenses, falling into the potential wells of galaxies along streams.

Reading between the lines I think this is their favoured scenario. There are several reasons to think the gas isn't doing what it's supposed to be doing. For a few galaxies they have well-resolved kinematics of the gas and stars, and here it tends to be either misaligned with and/or highly distorted compared to the stars. They've only got such data for the innermost regions, but the spectral profiles of the gas (measured over much larger scales) also suggest it's significantly asymmetric. And they're also systematically offset form the Tully-Fisher relation : that is, the gas kinematics has broader line widths that typical galaxies of the same mass. 

All this is what you'd expect if the gas had only recently arrived rather than evolving along with the galaxy throughout its history. What they can't say is whether this indicates a temporary revival of the galaxy or a full resuscitation. They might, potentially, be rejuvenating themselves back onto the star formation main sequence, or they may have just accumulated a little bit of gas and will soon burn out once again. My guess is that latter is more likely but this would mean galaxies would have extraordinarily complicated histories.

3) Mergers

The most dramatic way a galaxy can accumulate more gas is by gobbling up other galaxies whole. This is their least favoured scenario as there are no signs of the remnants of the encounter you'd expect, e.g. long stellar tails. They can't rule it out though, as it's possible the star formation is persisting long after the tails have dispersed, but resolved gas measurements could help verify this idea.

I've simplified the discussion here considerably but this is roughly what it boils down to; they themselves, in my opinion, are too focused on the details of the different processes rather than painting a clear picture of the main differences. Fortunately, they've got VLA time to resolve the gas in a pilot sample, so this looks like a solvable problem. As they say, "it is important to remember the results of McKay et al. (in preparation)"... well, asking to remember results that don't exist yet is a new one on me, but I'll try.

The ultimate in flattening the curve

It just refuses to go down...

Well, I'd play the innuendo card with this paper, at any rate. 

Galaxy rotation curves are typically described as flat, meaning that as you go further away from their centres, the orbital speeds of the gas and stars don't change. This is the traditional evidence for dark matter. You need something more than the visible matter to accelerate material up to these speeds, especially the further away you go : if you use the observed matter, the prediction is that orbital speeds should steadily decrease a la Kepler. Unseen dark matter is a neat way to resolve this dilemma, along with a host of other observational oddities.

This paper claims to have extended rotation curves considerably further than traditional measurements and find that the damn curves remain flat no matter how far they go. They reach about 1 Mpc, about the size of our whole Local Group, and still don't show any sign of a drop. This is not at all expected, because eventually the curve should drop as you go beyond the bulk of the dark mass. It is, they say, much more in-keeping with the prediction of modified gravity theories that do away with dark matter, such as MOND.

I won't pretend I'm in any way expert in their methodology, however. A standard rotation curve directly measures the line of sight speed of gas and/or stars, which is relatively simple to convert into an orbital speed – and for qualitatively determining the shape of the curve, the corrections used hardly matter at all. But here the authors don't use such direct kinematic measurements, but instead use weak gravitational lensing. By looking at small distortions of background galaxies, the amount of gravity associated with a target foreground source can be determined. Unlike strong lensing, where distortions are easily and directly visible in individual sources, this is inferred through statistics of many small sources rather than from singular measurements.

Here they go even more statistical. Much more statistical, in fact. Rather than looking at individual lens galaxies they consider many thousands, dividing their sample into four mass bins and also by morphology (late-type spiral discs and early-type ellipticals). The lensing measurements don't give you orbital speed directly, but acceleration, which they then convert into velocity. 

1 Mpc is really quite a long way in galactic terms, and it wouldn't be at all uncommon to find another similar-sized galaxy within such a distance : in our Local Group, which is not atypical, there are three large spiral galaxies. Measuring the rotation curve out to such distances then becomes intrinsically complicated (even if you had a direct observational tracer like the gas) because it's hard to know which source is contributing to it. 

They say their sample is of isolated galaxies with any neighbours being of stellar mass less than 10% of their targets out to 4 Mpc away, but their isolation criterion uses photometric redshifts*. Here I feel on very much firmer footing in claiming that these are notoriously unreliable. Especially as the "typical" redshift of their lens galaxies is just 0.2, far too low for photometric measurements to be able to tell you very much. Their large sample means they understandably don't show any images, but it would have been nice if they'd said something about a cursory visual inspection, something to give at least some confidence in the isolation.

* These are measurements of the redshift based on the colour of the galaxy, which is extremely inexact. The gold standard are spectroscopic measurements, which can give precisions of a few km/s or even less.

If we take their results as given, they find that the rotation curves of all galaxies in all mass bins remain flat out to 1 Mpc, the limit of their measurement (although in one particular subset this doesn't look so convincing). They also show that in individual cases where they apparently can get good results from weak lensing, the results compare favourably with the direct kinematics they get from gas data.

As often with results questioning the dark matter paradigm, I'd have to describe the results as "intriguing but overstated". I don't know anywhere near enough about the core method of weak lensing to comment on the main point of the paper. But that this is normally in itself as result of statistical inference, and that here they use a very large sample of galaxies and convert the result from the native acceleration measurement to velocity, and that their isolation criteria seems suspect... I remain unconvinced. I'd need a lot more persuading that the weak lensing data is really giving meaningful results to such large distances from the lens galaxies.

What would have been nice to see is the results from simulations. If they could show that their photometric redshifts were accurate enough in simulated cases to give reliable results, and that the weak lensing should given something similar to this (or not, if the dark haloes in the simulations have a finite extent), then I'd find it all a lot more convincing. As it stands, I don't believe it. Especially given that so many galaxies are now known with significantly lower dark matter contents than expected : these "indefinitely flat" rotation curves seem at odds with galaxies with such low rotation speeds even in their innermost regions. Something very fishy's going on.

The shoe's on the other foot

My, how the tables have turned. The hunter has become the hunted. And various other cliché's indicating that the normal state of affairs have become reversed.

That is, as well as having to write an observing proposal, I find myself for the first time having to review them. Oh, I've reviewed papers before, but never observing proposals. This came about because ALMA has a distributed proposal review system : everyone who submits their own proposal has to review ten others. And since this year I finally submitted one, I get to experience this process first hand.

The ALMA DPR procedure

When you submit a proposal, you indicate your areas of expertise and any conflicts of interests  – collaborators and direct competitors who shouldn't be reviewing your proposal, either because they'd stand to benefit from it being accepted or would love to take you down a peg. It's a double-blind procedure : your proposal can't contain any identifying information and you don't know who the reviewers are. Some automatic checks are also carried out to prevent Co-Is on recent ALMA proposals being assigned as reviewers, and suchlike.

Then your proposal is sent off for initial checks and distributed to ten other would-be observers who also submitted observing proposals in the current cycle. You, in turn, get ten proposals to review yourself. Each document is four pages of science justification (of which normally one or even two pages are taken up with figures, references, and tables) plus an unlimited-length technical section containing the observing parameters for each source plus some brief justification on the specifics (in practise, in most proposals each of these so-called "science goals" are very similar, using the same observing setup on multiple targets). You then write a short review of each one, of a maximum of 4,000 characters but typically more like ~1,000 (or even less) describing both the strengths and weaknesses of each. You also rank them all relative to each other, from 1 (the strongest) to 10 (the weakest).

That's stage one. A few weeks later, in stage two you get to see everyone else's reviews for the same proposals, and can then change your own reviews and/or rankings accordingly, if you want to. So far as I know, each reviewer gets a unique group of ten proposals to review, so no two reviewers review the same set of proposals, meaning you can't see the others rankings. Exactly how their rankings are then all compared and combined, and ultimately, translated into awarded telescope time, remains a mystery to me. Those details I leave for some other time, and I won't go into the details of anonymity* here either : I seem to recall hearing that this gives a better balance of both experience and gender, but I don't have anything to hand to back this up.

* I will of course continue to respect the anonymity requirements here, and not give any information that could possibly identify me as anyone's reviewer.

Instead I want to give some more general reflections on the process. To be honest I went into this feeling rather biased, having received too many referee comments which were just objectively bollocks. I was quite prepared to believe the whole thing would be essentially little better than random, which is not a position without merit.

First thoughts

And my initial impressions justified this. It seemed clear to me that everyone had chosen interesting targets and would definitely be able to get something interesting our of their data, making this review process a complete waste of time.

But after I let things sink in a bit more, after I read the proposals a bit more carefully and made some notes, I realised this wasn't really the case. I still stand by (with one exception) that all proposals would result in good science, but the more I thought about it, the more I came to the conclusion that I could make a meaningful judgement on each one. I tried not to judge too much whether one would do better science than another, because who am I to say what's better science ? Why should I determine if studies of extrasolar planets are more important than active galactic nuclei ?

These aren't real examples, but you get the idea. Actually the proposals were all aligned very much more closely with my area of expertise. The length of four pages I would say is "about right", it gave enough background for me to set each proposal in its proper context as well as going into the specific objectives.

Instead, what I tried to assess was whether each project would actually be able to accomplish the science it was trying to do. I looked at how impactful this would be only as a secondary consideration. There isn't really any right or wrong answer as to whether it's better to look at a single unique target versus a statistical study of more typical objects, but I tried to judge how much impact the observations would likely have on the field, how much legacy value they would have for the community. But first and foremost, I considered whether I was persuaded the stated science objectives could actually be carried out if the observations themselves reached their design spec.

Judgement Day

And this I found was something I could definitely judge. Two proposals to me stood out as exemplary, perfectly stating exactly what they wanted to do and why, exactly what they'd be able to achieve with this. It was very clear that they understood the scientific background as well as anyone did. I initially ranked these essentially as a coin-toss as to who got first and who got second place; I couldn't meaningfully choose between them.

At the opposite extreme were two or three which didn't convince me at all. One of the principle objectives of one of them was just not feasible with the data they were trying to obtain, and they themselves presented better data in their proposal that they already had which would have been much more suitable for this. Lacking self-consistency is a black mark in just about any school of thought. Another looked like it would observe a perfectly good set of objects, but contained so many rudimentary scientific errors that there was no way I could believe they'd do what they said they would do. 

Again, deciding which one to rank lowest was essentially random, though I confess that one of them just wound me up the wrong way more than the other.

In the middle were a very mixed bunch indeed. Some had outstanding ideas for scientific discovery but were very badly-expressed, saying the same thing over and over again to the nth degree (I would say to these people, there's no obligation to use the full four pages, and we should stipulate this in the guidance to observers and reviewers alike. I tried to ignore the poor writing style of these and rank them highly because of the science). Some oversold the importance of what they'd do, making unwarranted extrapolations from their observations to much more general conclusions. Some had a basically good sample but claimed it was something which it clearly wasn't; others clearly stated what their sample was but the objects themselves were not properly representative of what they were trying to achieve.

This middle group... honestly here, a random lottery would work well. On the other hand, there doesn't seem any obvious reason not to use human judgement here either, because for me at least this felt like a random decision anyway. And if other people's judgements are similar then clearly there are non-random effects which probably should actually be accounted for, whereas if they are truly random then the effects will average out. So there's potentially a benefit in one case and no harm in the other, and in any case there almost certainly is a large degree of randomness at work anyway.

Reviewing the reviews

I went through a similar process of revising my expectations in stage 2, though to a lesser degree. At first glance I didn't think I'd need to change my reviews or rankings, but on carefully checking one of the other reviews, I realised this was not the case. One reviewer out of the ten had managed to spot a deeply problematic technical issue in one of the proposals that I otherwise would have ranked very highly. And on checking I was forced to conclude that they were correct and had to downgrade my ranking significantly. This alone makes the process worth doing : 1 out of 10 is not high, but with ~1,600 proposals in total, this is potentially a significant number overall.

Reading the other reviews turned out to be more interesting than I expected. While some did raise exactly the same issues with some of the proposals that I had mentioned, many didn't. Some said "no weaknesses" to proposals I thought were full of holes. One even said words to the effect that "no-one should doubt the observers will do good science with this", a statement I felt presumptuous, biased, and bordering on an argument-from-authority : it's for us the reviewers to decide this independently; being told what we should think is surely missing the point. 

The reverse of this is that some proposals I though were strong others thought were weak – very weak, in some cases. Everyone picks up on different things they think are important. There was one strange tendency for reviewers to point out that the ALMA data wouldn't be of the same quality as comparison data. This is fine, except that the ALMA data would usually have been of better quality, and downgrading it to the same standard is trivial ! I sort of wished I'd edited my reviews to point this out. Some also made comments on statistics and uncertainties that I thought were so generic as to be unfair, yes of course things might be different from expectations, but that's why we need to do observations !

What the DPR doesn't really do is give any chance for discussion. You can read the other reviews but you can't interact with the other reviewers. It might have been nice to have somewhere where we could enter a "comment to other reviewers", directed to the group, or at least have some form of alert system when reviews were altered. Being able to ask the observers questions might have been nice, but I do understand the need to keep things timely as well. On that front, reviews varied considerably in length; mine were on the longer and to be honest perhaps overly-long side (I think my longest was nearly 2,000 characters), while one was consistently and ludicrously short.

All this has given me very mixed feelings about my own proposal. On the one hand, I don't think it's anywhere near the worst, and I stand behind the scientific objectives. On the other, I think I concentrated overmuch on the science and not enough on the observational details. Ranking it myself with hindsight I'd probably have to put it in the lower third. It was always a long shot though, so I'll be neither surprised nor disappointed by the presumed rejection. One can but try with these things.

One thing I will applaud very strongly is the instruction to write both strengths and weaknesses of each proposal. All of them, bar none, had some really good points, but it was helpful to remind myself of this and not get carried away when reviewing the ones I didn't much like. Weaknesses were more of a mixed bag; one can always find something to criticise, although in some cases they aren't significant. Still, I found it very helpful to remember that this wasn't an exercise in pure fault-finding.

---

How in the world one judges which projects to actually undertake, though... that seems to me like the ultimate test of philosophy of science. Groups of experts of various levels have pronounced disagreements about factual statements; some notice entirely different things from others. There's the issue of not only will the science be significant, but also whether the data can be used in different ways from what's suggested. That to me remains the fundamental problem with the whole system, that one can nearly always expect some interesting results, but predicting what they could be is a fool's game.

Overall, I've found this a positive experience. Reading the full gamut of excellent to poor proposals really gives a clearer idea of what reviewers are looking for, something it's just not possible to get without direct experience. Not for the first time, I wonder a lot about Aumann's Agreement Theorem. If we the reviewers are rational, we ought to be persuaded by each other's arguments. But are we ? This at least could be assessed objectively, with detailed statistics possible on how many reviewers change their ranking when reading other reviews. 

And at the back of my mind is a constant paradoxical tension : a strong feeling that I'm right and others are wrong, coupled with the knowledge that other people are thinking the same thing about me. How do we reconcile this ? For my part, I simply can't. I formulate my judgement and let everyone else to the same, and hope to goodness the whole thing averages out to something that's approximately correct. The paradox is that this in no way makes me feel any the less convinced of my own judgments, even knowing that some fraction of them simply must be wrong.

Other aspects are much more tricky. This is a convergence of different efforts, both trying to asses what-is-true (what science claimed is factually correct, why do experienced experts still disagree on some points), what will likely benefit the community the most, and how we try and account for the inevitably uncertain and unpredictable findings. As I've said before many times, real, coal-face research is extremely messy. If it isn't already, then I would hope that telescope proposals ought to be an incredibly active field of research for philosophers of science.

The data won't learn from itself

Today I want to briefly mention a couple of papers about AI in astronomy research. These tackle very different questions from the usual sort, which might examine how good LLMs can be at summarising documents or reading figures and the like. These, especially the second, are much more philosophical than that.

The first uses an LLM to construct a knowledge graph for astronomy, attempting to link different concepts together. The idea is to show how, at a very high level, astronomical thinking has shifted over time : what concepts were typically connected and how this has changed. Using distributional semantics, where the meanings of words in relation to other words are encoded as numerical vectors, they construct a very pretty diagram showing how different astronomical topics relate to each other. And it certainly does look very nice – you can even play with it online. 

It's quite fun to see how different concepts like galaxy and stellar physics relate to each other, how connected they are and how closely (or at least it would be if the damn thing would load faster). It's also interesting to see how different techniques have become more widely-used over time, with machine learning having soared in popularity in the last ten years. But exactly what the point of this is I'm not sure. It's nice to be able to visualise these things for the sake of aesthetics, but does this offer anything truly new ? I get the feeling it's like Hubble's Tuning Fork : nice to show, but nobody actually does anything with it because the graphical version doesn't offer anything that couldn't be conveyed with text.

Perhaps I'm wrong. I'd be more interested to see if such an approach could indicate which fields have benefited from methods that other fields aren't currently using, or more generally, to highlight possible multi-disciplinary approaches that have been thus far overlooked.

The second paper is far more provocative and interesting. It asks, quite bluntly, whether machine learning is a good thing for the natural sciences : this is very general, though astronomy seems to be the main focus. 

They begin by noting that machine learning is good for performance, not understanding. I agree, but once we do understand, then surely performance improvements are what we're after. Machine learning is good for quantification, not qualitative understanding and certainly not for proposing new concepts (LLMs might, and I stress might, be able to help with this). But it's a rather strange thing to examine, and possibly a bit of a straw man, since I've never heard of anyone thinking that ML could do this. And they admit that ML can be obviously beneficial in certain kinds of numerical problems, but this is still a bit strange : what, if any, qualitative problems is ML supposed to ever help with ?

Not that quantitative and qualitative are entirely separable. Sometimes once you obtain a number you can robustly exclude or confirm a particular model, so in that sense the qualitative requires the quantitative. But, as they rightly point out, as I have myself many times, interpretation is a human thing : machines know numbers but nothing else. More interestingly they note :  

The things we care about are almost never directly observable... In physics, for example, not only do the data exist, but so do forces, energies, momenta, charges, spacetime, wave functions, virtual particles, and much more. These entities are judged to exist in part because they are involved in the latent structure of the successful theories; almost none of them are direct observables. 

Well, this is something I've explored a lot on Decoherency (just go there and search for "triangles"). But I have to ask, what is the difference between an observation and a measurement ? For example we can see the effects of electrical charge by measuring, say, the deflection of a hair in the static field of a balloon, but we don't observe charge directly. But we also don't observe radio waves directly, yet we don't think they're less real than optical photons, which we do. Likewise some animals do appear to be able to sense charge and magnetic fields directly. In what sense, then, are these "real" and what sense are they just convenient labels we apply ?

I don't know. The extreme answer is that all we have are perceptions, i.e. labels, and no access to anything "real" at all, but this remains (in some ways) deeply unsatisfactory; again, see innumerable Decoherency posts on this, search for "neutral monism". Perhaps here it doesn't matter so much though. The point is that ML cannot extract any sort of qualitative parameters at all, whereas to humans these matter very much – regardless of their "realness" or otherwise. If you only quantify and never qualify, you aren't doing science, you're just constructing a mathematical model of the world : ultimately you might be able to interpolate perfectly but you'd have no extrapalatory power at all.

Tying in with this and perhaps less controversially are their statements regarding why some models are preferred over others :

When the expansion of the Universe was discovered, the discovery was important, but not because it permitted us to predict the values of the redshifts of new galaxies (though it did indeed permit that). The discovery was important because it told us previously unknown things about the age and evolution of the Universe, and it confirmed a prediction of general relativity, which is a theory of the latent structure of space and time. The discovery would not have been seen as important if Hubble and  Humason had instead announced that they had trained a deep multilayer perceptron that could predict the Doppler shifts of held-out extragalactic nebulae.

Yes ! Hubble needed the numbers to formulate an interpretation, but the numbers themselves don't interpret anything. A device or mathematical model capable of predicting the redshifts from other data, without saying why the redshifts take the values that they do, without relating it to any other physical quantities at all, would be mathematical magic, and not really science.

For another example, consider the discovery that the paths of the planets are ellipses, with the Sun at one focus. This discovery led to extremely precise predictions for data. It was critical to this discovery that the data be well explained by the theory. But that was not the primary consideration that made the nascent scientific community prefer the Keplerian model. After all, the Ptolemaic model preceding Kepler made equally accurate predictions of held-out data. Kepler’s model was preferred because it fit in with other ideas being developed at the same time, most notably heliocentrism.

A theory or explanation has to do much more than just explain the data in order to be widely accepted as true. In physics for example, a model — which, as we note, is almost always a model of latent structure — is judged to be good or strongly confirmed not only if it explains observed data. It ought to explain data in multiple domains, and it must connect in natural ways to other theories or principles (such as conservation laws and invariances) that are strongly confirmed themselves.  

General relativity was widely accepted by the community not primarily because it explained anomalous data (although it did explain some); it was adopted because, in addition to explaining (a tiny bit of new) data, it also had good structure, it resolved conceptual paradoxes in the pre-existing theory of gravity, and it was consistent with emerging ideas of field theory and geometry.

Which is a nice summary. Some time ago I'd almost finished a draft of a much longer post based on this this far more detailed paper which considers the same issues, but then blogger lost it all and I haven't gotten around to re-writing the bloody thing. I may yet try. Anyway the need for self-consistency is important, and doesn't throttle new theories in their infancy as you might expect : there are ways to overturn established findings independent of the models. 

The rest of the paper is more-or-less in line with my initial expectations. ML is great, they say, when only quantification is needed : when a correlation is interesting regardless of causation, or when you want to find outliers. So long as the causative factors are well-understood (and sometimes they are !) it can be a powerful tool for rapidly finding trends in the data and points which don't match the rest. 

If the trends are not well-understood ahead of time, it can reinforce biases, in particular confirmation bias by matching what was expected in advance. Similarly, if there are rival explanations possible, ML doesn't help you choose between them if they don't predict anything significantly different. But often, no understanding is necessary. To remove the background variations in a telescope's image it isn't necessary even to know where all the variations come from : it's usually obvious that they are artifacts, and all you need to is the mathematical description of them. Or more colourfully, "You do not have to understand your customers to make plenty of revenue off of them." 

Wise words. Less wise, perhaps only intended as a joke, are the comments about "the unreasonable effectiveness of ML", that it's remarkable that these industrial-grade mathematical processes are any good for situations to which they were never designed. But I never even got around to blogging Wigner's famous "unreasonable effectiveness" essay because it seemed worryingly silly. 

Finally, they note that it might be better if natural sciences were to shift their focus away from theories and more towards the data, and that the degeneracies in the sciences undermine the "realism" of the models. Well, you do you : it's provocative, but on this occasion, I shall allow myself not to be provoked. Shut up and calculate ? Nah. Shut up and contemplate.

Red, not dead, but who knows how well fed ?

In the last post I looked at a probably-nonsense claim about massive red but gas-rich galaxies that for some reason aren't forming stars. It's not totally rubbish by any means : some of the sample almost certainly are really interesting objects. It's just that many of them clearly are forming stars, you can see it very clearly just by looking, and if their numbers say otherwise, then I bet against their numbers.

Today's paper makes for a nice counterpoint. Here the authors make essentially the exact opposite claim, that all early-type galaxies (popularly supposed to be "red and dead") are forming stars. Not very many stars necessarily, but some.

One of the problems in establishing whether such galaxies really have star formation activity is that there's a degeneracy in the methods used for estimating star formation rates. Typically, H-alpha observations tell you about star formation occurring right now, that is, on timescales of at most a few tens of millions of years. UV, on the other hand, tells you about both ongoing and recent (few hundred millions of years) activity. You can use estimates from other wavelengths, but they tend to be less accurate, so H-alpha and UV tend to be the go-to preferred choices. And H-alpha is relatively difficult to do, so UV can be a bit easier.

Here's where the degeneracy comes in. UV emission can be from hot young stars but it can also be form much older objects : asymptotic giant branch stars and even stellar remnants like white dwarves. So while UV emission is known in early-type galaxies (ellipticals and lenticulars, both are which are smooth and structureless, with the latter being more disc-like), that doesn't automatically mean it's indicating star formation.

The best way to break the degeneracy, say the authors, is by high resolution observations. If the UV is just from the normal old stellar population that dominates early-type galaxies, then it should appear to have the same basic morphology, smooth and symmetrical, as the bulk of the red stars. Star formation, by contrast, is by definition something that happens in distinct structures, so in this case the UV should appear clumpy, maybe disc-like and otherwise distinct from the rest of the stars in the galaxy.

What they do here is get a sample of 32 galaxies with high-resolution UV observations from the UVIT instrument on AstroSat. I wasn't even aware of this before so this alone is quite interesting. They are really quite careful in defining their sample to make sure it's representative and that the observations are sufficiently deep. They seem to cull this in half a dozen different ways, making it sometimes quite tedious to follow, and they actually miss out on what to me seem like the obvious and most interesting selection criteria (which I'll get back to). And the sample is, of course, very small indeed for claiming that all such galaxies are forming stars – this is one hell of an extrapolation.

Even so, it's careful and meticulous work. They measure the structure in the UV using three different parameters : concentration (essentially how much of the total emission is within some radius), asymmetry (how different the emission is comparing two arbitrary opposite sides), and clumpiness (how much of the emission is in small-scale structures compared to larger-scale emission). This too I found interesting, as these appear to be much more readily-quantifiable morphological parameters than the standard Hubble type, which tells you whether a galaxy is a spiral, irregular, or other sort of object. 

Exactly how well these work in practise I'm not sure, so I'm going to take their word for it. I would imagine that a clumpy galaxy would also be asymmetrical since the clumps on one side won't be at the same location as on the opposite – whether the asymmetry involves some level of smoothing to account for this, I don't know.

Anyway, the nifty thing about their observations is that they have much higher resolution observations than typical UV data (i.e. from the GALEX satellite that did an all-sky survey), in fact about six times sharper. So while their overall findings in terms of the other parameters are quite similar to previous studies, they show quite definitively that indeed the UV emission is clumpy. Previous studies also showed the UV was structured, so already indicating it was from star formation, except that it didn't appear to be clumpy at all. This paper, then, has an an incremental but still important new finding, with that higher resolution being quite definitive. They also show images of the galaxies and it's quite obvious that yes, GALEX could not possibly have seen this because it's too blurry, whereas UVIT is definitely up to the mark.

Okay so this is quite nice. It's more incremental than I was expecting because I'm not familiar with star formation in ellipticals, but it's still definitely progress. As I said, they're hardly forming stars at the same rate as spirals, but still it's ticking over, which is in marked contrast to the popular "red and dead" moniker. And over cosmic time, this looks to have been significant : they did not form all their stars early on, as it commonly supposed.

What about those selection criteria I mentioned ? Well, that's a bit disappointing. Star formation needs gas, so the obvious question is : how much gas do they have and where did it come from ? This they don't mention. This is intimately connected with the next question : where are these particular galaxies ? Ellipticals in the general field are known to have significant amounts of gas, so it would be interesting but unsurprising to discover that they're forming stars. By contrast, those in clusters such as Virgo appear to have virtually no gas whatsoever. Not merely very little gas, but orders of magnitude less. So if those are forming stars, that would be very much more interesting. Hence the two unmentioned selection criteria of gas content and environment could have made the paper all the more exciting.

But no matter. Despite the small sample size and exaggerated extrapolation to "all" early-type galaxies, it seems like a much surer finding than the previous paper. And it would certainly be interesting to see what UVIT reveals about those supposedly-quenched galaxies. I'd bet heavily that some of them at least are really-UV bright and heavily structured... time will tell, perhaps.

Red, dead, but very well-fed

What stops galaxies from forming stars ? Normally I'm most interested in this for objects which have hardly any stars at all : ultra diffuse galaxies, ultra faint dwarves, dark galaxies and suchlike. But there are also cases of objects which have managed to form tonnes and tonnes (literally) of stars in the past, still have oodles of gas today, but are doing diddly-squat right now. What's going on ?

Today's paper is an attempt to catalogue some of these "quenched HI-rich" galaxies, "quenched" just being standard jargon for "not forming many stars any more". It's not surprising that when galaxies run out of gas they stop forming stars, it's the whole "gas rich and quenched" thing that's surprising. The details get rapidly complicated. For instance, the atomic gas doesn't seem to (much) participate directly in star formation, but has to condense into a molecular state first, which makes the correlation between star formation and atomic gas hard to interpret. And most gas loss removal mechanisms involve initially compressing the HI, i.e. transforming it to molecular and so potentially triggering a burst of star formation before quenching.

We also know that in most normal galaxies the HI extends well beyond the stellar disc. Generally in the outskirts the gas density is very much less than in the inner regions, so low density can certainly mean a lack of star formation. But it seems extraordinarily rare to encounter cases where the gas density is low everywhere. After all, if you have a massive stellar disc, gravity ought to help the gas to collapse to higher densities in the central regions, ensuring star formation keeps ticking over.

The great bulk of this paper is given over to proving that such objects do in fact exist. Most gas-rich objects are spirals and irregulars, which are happily churning out stars, but it's not so unusual to find at least some gas and star formation in ellipticals. These are often known as "red and dead" because their (much older) stellar population tends to be reader, and star formation rates tend to be much lower than in spirals. Especially so in clusters, where most ellipticals are found, where they very often appear to have no gas at all.

So what they're looking for are gas-rich red galaxies. Previous studies, they note, have been widely criticised because they get the star formation rates wrong, so many claimed quenched gas-rich galaxies may simply not be as quenched as thought. They try and improve things by using more robust selection criteria, especially using UV emission to quantify colour as this is more sensitive to ongoing star formation than traditional optical wavelengths : UV photons are higher energy and preferentially emitted by hot, young stars. Most of their HI data comes from the large ALFALFA survey, plus a couple of other Arecibo projects.

This is a very nice, easy-to-follow, tremendously narratively linear paper that spells out what they did in great detail. Basically they define their sample in terms of HI mass and colour and make different cuts to have both a putative gas-rich red sample but also different comparison samples of the same stellar mass, colours, etc. And they show that yes, their main sample is weird, being mostly elliptical galaxies which are very red but have shedloads of gas. They're clearly outliers from the trends in a way that their control samples certainly aren't. Compared with previous findings these ones are definitely unusual in their gas content, as they confirm that most galaxies with similar stellar parameters tend to be very gas poor.

This did make me wonder why, if these red/dead/fed objects are so structurally similar to ordinary galaxies, what changed ? That is, they must have been forming stars in the usual way in the past, so why is the star formation in these objects only now being suppressed ? This isn't such a problem for objects which have never managed to form many stars at all, but I'll get back to that.

Because the paper is so concerned with establishing the existence of the objects, it doesn't go much into detail as to what might be causing the quenching. It can't be gas loss, but it might still be gas density. There are some tentative hints that the gas might be more extended in these objects than others, but this is based on scaling relations – only one galaxy in their sample actually has resolved HI observations, and this is only a bit more extended than normal. They suggest this could be the result of high angular momentum, though it's a bit disappointing that they don't do anything with the HI kinematic data to see if this is consistent. And they note that the galaxies tend to live in low-density environments and in relatively small dark matter halos, but this doesn't address what happened to stop them forming stars when they clearly used to have no such problems. Well, we all have problems as we get older...

Most of the paper is rife with statistics and endless graphs. Normally my eyes glaze over with papers like this but I found their presentation style really very good indeed : for a potentially dry statistical analysis, it's remarkably un-put-downable. 

Except... it contains not a single image of any galaxy [but see below for a major caveat !].

Here's where I get seriously worried. When you actually look at the sample, yes, some of them do indeed appear to be your classical ellipticals : red, smooth, structureless, and ultimately aesthetically boring. But even going through just the first ten, it's clear that these are almost the exception rather than the norm. Their very first object is clearly a spiral, with a dominant central bulge but enormous, blue spiral arms ! Their second has a funky interaction going on, their sixth and eighth appear to be basically normal spirals, and their tenth appears to be a freakin' polar ring galaxy. Some of them are really spectacular oddballs that defy description, like this one. Many of their objects have tens of references in the literature but they don't remark on any objects individually : it's purely statistical.

This has me concerned that all their very careful work is for nought. Rather than being the globally-quenched objects they claim, many of their sample might well just have little or no star formation in their central regions, which is not so unusual for spiral galaxies. Visually, at least, that's certainly what it looks like : big red smooth bulges surrounded by complex, blue, likely star-forming extended arms. Not mentioning the ring, in particular, is a bit of a red flag, because these are such interesting systems in their own right that you'd think this is surely worth reporting.

STOP PRESS : At the last minute I realised I'd read an older version of the paper prior to its passing peer review. The accepted version contains an appendix which has cutout images of their whole sample. I don't want to re-read the whole paper, but their section 4.3 does now acknowledge that about half their sample contains rings or other extended features. From what I can tell, though, they haven't changed their major conclusions or adjusted the morphological classifications, or made any significant changes to any of the other figures.

I find that very strange. When you look at the images, it's pretty obvious they cannot possibly be as quenched as they claim : there are some really striking blue structures here which just must be forming stars. For sure, some of their objects are indeed classical, smooth ellipticals, and those will be interesting. The rest may well be interesting as well but I doubt very much as examples of fully-quenched galaxies. My guess is that the gas is indeed extended, but it's experiencing local star formation in the usual way with the central regions being quenched through simple gas depletion. I'm also extremely skeptical of the deep learning paper that presented the original classifications, though I should try and read that. Another factor might be the relatively low sensitivity of the GALEX UV observations, which might not have picked up the faint outer star-forming regions.

In short, there may be a few interesting objects here, but most of them don't look all that impressive or unusual from a quenching point of view.

Back from the grave ?

I'd thought that the controversy over NGC 1052-DF2 and DF4 was at least partly settled by now, but this paper would have you believe otherwise. 

To recap, these are two small, extremely faint galaxies that appear to have no dark matter. This claim caused a prolonged furore when it was announced, most contentious of which was the distance dilemma. If they were at 20 Mpc distance, as originally claimed, then they'd indeed lack dark matter and be quite strange in other aspects too, like how many globular clusters they have. This is weird because galaxy formation models don't predict objects like this. If on the other hand they were at about 13 Mpc, then they'd be pretty normal and uninteresting.

After a lengthy ping-pong of claims and counter-claims over the distance, this seemed to have been settled decisively with a wholly-independent measurement placing DF2 at 22 Mpc and DF4 at 20 Mpc. Very clearly then, they have no or negligible amounts of dark matter.

So, exciting stuff ? New physics ? Alas, not so fast. Simulations showed that such objects could be produced by a rare but perfectly normal process. It turns out that the direct collision of a dwarf galaxy with a larger object can almost completely remove its dark matter, a process made easier because dark matter particles should orbit radially around the centre of their parent object. That is, if a particle happens to be deep inside the galaxy at some point, then eventually it will wander to the outskirts where it can be more easily removed. This is quite different to the stars and gas, which remain much more tightly confined to the inner regions at all times, making them much harder to disturb.

All well and good. A very few oddball galaxies having experienced a weird but entirely conventional process : fun, but not revolutionary for our understanding of physics or even just plain old galaxy evolution. Except, of course, that we now know of much greater numbers of galaxies which seem to at least be significantly deficient in dark matter, if not lacking it entirely. And many of those are isolated, where encounters where other galaxies wouldn't be a likely explanation.

Anyway, today's paper goes back to the original DF2 and DF4. They use really, seriously deep optical images to search for the signatures of tidal encounters which are expected to be present if the galaxies really had lost their dark matter thanks to interactions with other galaxies. They note that simulations show that these features should be surprisingly long-lasting, on timescales of several gigayears. So even if it happened in the distant past, something should still be visible.

Most of the paper is a very technical description of exactly how they reduced their data. I skipped over this; it's likely only of interest to anyone planning to actually do it for themselves. The bottom line is that they say they confirm earlier evidence that DF4 does have signatures of tidal encounters, whereas DF2 shows no indication of any disturbance.

First, DF4. As I describe in the last link, I was very skeptical about this claim, which looks like a marginal variation in the image to me, and not at all like the S-shaped feature they say they've found. The new, deeper images, which show much larger and stranger features... well, I don't know what to make of them. I'll take the authors at their word and assume they've processed the data correctly; certainly they understand this much better than I do. What I think would have been really valuable would have been to actually reduce the sensitivity by using shorter segments of their exposures, to get to the same sensitivity as the previous images. If these had then also shown the same claimed feature as that earlier paper, that would have least been a bit more convincing that they were real.

But instead (understandably) they go straight for the jugular of getting the highest sensitivity possible. To me the resulting features in the image look like a rather haphazard variation in the data that's not at all indicative of the tidal features expected. The extremely small size of the image is a major limitation here, though, showing little beyond the target galaxy itself. So it's difficult to tell if these apparently-random variations are really part of some larger, more coherent structures that would be far more persuasive for the claim of detecting tidal features.

DF2, however, is a case where you can prove a negative. The image shows a classic case of a non-detection. What was supposed to be there, according to theory, undeniably isn't there, and the authors don't dispute this.

Does this mean the tidal hypothesis is done, at least for DF2 ? They say yes, and venture to suggest that maybe there is a distance tension after all, given that some of the major galaxies in the group have been found to be a bit closer (17 Mpc). This feels like a weak argument to me. By all means dispute the distance, but I think that you need to do so based on direct measurements, not those of galaxies which are nearby in projection. Close sky-alignments of galaxies which are at radically different distances happen all the time. 17 Mpc is anyway not 13 Mpc.

A slightly better argument is that if DF2 were associated with NGC 1042 instead (which is presumably closer), their projected separation would be enough that no tidal features would be expected. But again this is circumstantial, and not really a direct argument in favour of a different distance. It also ignores all those other dark matter deficient galaxies, making the whole thing feel just a tad... petty.

In yet another post about these objects, I mentioned that observations showing a neat line of galaxies on larger scales was consistent with a tidal origin, describing this as another nail in the coffin for anyone hoping these would turn out to be seriously weird, physics-challenging objects. BUT :

Not the final nail by any means, and it's still just about possible they could burst back out like an enraged vampire, but it's definitely making it harder for any would-be children of the night to go on a killing spree.

So do I have to say that now the vampire is slowly rising from the tomb after all, ready to drain the blood of mainstream physics ? Nah. Even leaving aside the explanation that they're actually just closer than first thought, exotic physics doesn't seem necessary to me. As the authors of today's paper themselves note, the same data has been interpreted differently by different teams even when they don't dispute the numerical values. The clear lesson from this storm in a teacup is just how much the details matter. What at first seems like a Eureka moment can, on repeat analysis, turn out to be erroneous or incomplete, which is why having multiple analyses, multiple lines of attack, varying perspectives and teams digging into the dirt, really matters. Discovery in this case is a slow and unglamorous process of slowly chipping away at best, and more often a case of two-steps-forward-one-step-sideways-oops-I've-tripped-and-broken-my-ankle.

For now, my guess remains on the tidal encounter scenario. There are so many parameters in simulations like this that the possibilities are vast; I would not be inclined to take a lack of observed tidal features in this case as being decisive. But what continues to interest me more are those similar objects in isolation. DF2 and DF4 may be emblematic, but as members of a whole wider population, they themselves no longer matter quite so much. Regardless of their own particular origins, there are plenty of other fish to fry.

Taking the galactic paternity test

Even though my backlog of unread papers is a mile long, this one from today's arXiv was so pertinent I had to read it immediately. 

You might remember that a few years ago I was co-author on a paper that re-examined an enormous gas cloud in the Virgo Cluster that looks a bit like a rhino. This was already known from ALFALFA observations, and with no clear parent galaxy but an enormous amount of gas, and no obviously-associated optical emission, it was undeniably weird. Most bits of atomic fluff floating around in galaxy clusters tend to be small, but this was was enormous : almost 200 kpc long and with an HI mass of over a billion times the mass of the Sun. Originally seen as a complex of HI clouds, our more sensitive observations showed that there was even more gas present and all the individual clouds, which originally looked like separate objects, are actually connected by fainter gas structures.

We showed that this made the already-suspicious candidate galaxies for the origin of the Complex even more implausible. To lose that much gas implies a very large galaxy indeed. Now galaxies can and do, of course, lose huge amounts of gas in clusters through ram pressure stripping : as they move through the cluster's own, much thinner gas, pressure builds up and eventually pushes gas right out of the galaxy's disc. But this process is also prone to dispersing and dissolving much of the gas, leaving it undetectable. This means that the gas we see today might only be a small part of that which was originally present, requiring an even more massive parent to supply the gargantuan amounts of gas needed.

We also suggested a new, but unlikely, possible parent : NGC 4522. This is one of the most famous examples of ram pressure stripping. It's got a nice clear gas tail, and even in the optical you can see signs of disturbances in the dust. Our observations showed a second gas tail, with different kinematics to that which was already known, lining up quite nicely with the Kent Complex (which Brain Kent prefers to call the boring name of the ALFALFA Complex 7). But the velocities of the galaxy and Complex are so different that it's difficult to believe the one could be the source of the other.

This latest paper changes the game. They find that there is actually some faint optical emission in one (small) part of the Complex, comparable in many ways to the putative ram pressure dwarfs the same authors proposed a couple of years ago. In those objects, it seems the stars form directly in the gas stripped due to ram pressure, forming stellar structures that might be gravitationally self-bound or might just be passing, transient features that will soon disperse. As with the others, the stars in this object are all young, with no evidence for a stellar population more than a few tens of megayears old. That's basically instantaneous for these sorts of features. It's worth also pointing out that this little star-forming region is very small in comparison with the rest of the Complex, maybe ~10" across compared to the 40' of the whole Complex (a factor of 240 difference in size !).

What they get with the new observations is not just the discovery of this faint patch of starlight, which by itself would be unconvincing : such little smudges aren't that uncommon, and it's just too small to be obviously related. No, they go much further than this, getting very nice resolved observations with Hubble but also a stellar velocity which is a perfect match for the Kent Complex. That makes the association about as secure as it's going to get. And they also measure the chemical composition (a.k.a. metallicity), showing that it's similar to the other potential ram pressure dwarfs but also allowing for comparisons to the other galaxies in the vicinity.

And they estimate, however roughly, the stellar mass. This is small, a few tens of thousands of times the mass of the Sun. Even only considering the particular clump of gas nearest the stars, the gas content is at least three thousand times higher than the stellar mas, which is extraordinary even in comparison to the other ram pressure dwarf candidates.

What does this mean for the origin of the complex ? Well, like our earlier paper, they conclude there aren't any fully convincing candidate galaxies, but they open the door a little. While we largely dismissed NGC 4522 because of its huge velocity difference with respect to the Complex (1,800 km/s, which is high even in a chaotic place like the Virgo Cluster), they're a bit more charitable. This extreme velocity with respect to the cluster means that it could have experienced incredibly strong ram pressure stripping and very recently too, recently enough that most of the cloud would remain intact and detectable. That would make NGC 4522 a plausible candidate because it wouldn't have needed to have had such a huge gas content, as not so much would have become undetectable yet. 

It's also a good match in metallicity, though they're careful to point out that limited metallicity data for other galaxies in the vicinity makes this not such a strong diagnostic tool as we might like. There's also a large gap between NGC 4522 and the Complex, but this could simply imply two stripping episodes, and they say that simulations have shown that such features can indeed happen.

Of our preferred candidate, NGC 4445... well, "preferred" is too strong a word. It was really only the one we disliked the least. Anyway they raise the same objection that we do, that other observations show that NGC 4445 has a tail pointing in exactly the wrong direction. So again, two stripping episodes would be needed, with some weird geometry, but this one is kinematically closer to the complex, and could more easily account for enormous gas lost.

My immediate impression is that their arguments are quite persuasive : NGC 4522 could be the parent after all. We also objected because of the particular kinematics of the structure (one of the sub-clumps is at the wrong velocity), but without a dedicated numerical simulation, this is a weaker argument. The huge kinematic difference of the Complex and galaxy remains, as they say, problematic, especially given the rather low velocity dispersion in this part of the cluster, but it might not be fatal. Still, I wouldn't rule out NGC 4445 either.

What's really interesting about this object, I think, are three things besides the obvious how-did-the-damn thing-form in the first place. First, why is only part of it forming stars, and why has that part only started forming stars right now ? What's special about this particular section of the Complex – why aren't other bits of it forming stars as well ? What happened recently to trigger star formation ? Secondly, more generally, how does this relate to the other ram pressure dwarfs, given that it's so much more massively gas rich than all of the others ? Does it point to a similar origin or is this one a coincidence ? And thirdly, why is this structure so damnably complex ? Why does it have all these fiddly little details rather than being one big long stream ?

The most likely avenue of progress at this point would seem to be detailed, dedicated numerical simulations. More observational data might help of course, but I think if we could show that certain passages of galaxies in this part of the cluster would (or more likely, would not) produce even vaguely similar structures to this one, we'd be able to formulate a convincing argument for and against individual candidate galaxies. Not at all easy to do, but possible. 

Regardless of its origin, it's a spectacular and fascinating object, and I'm gratified to see someone not only having properly read our previous work but also genuinely understanding it too. Were I there referee, this one would likely have gone through on a nod. In fact the other real question I'd have would be how it can be submitted as a letter when it's fifteen pages long and pretty fully fleshed out – may as well skip the letter phase and make it a full paper at this point.

Another one for the collection

Another paper about so-called "almost dark" galaxies, a term I intensely dislike. Just call them dim ! "Almost dark" sounds lame, like being an "almost professional" snooker player or something. Although I suppose being called a "dim" snooker player would be even worse...

With a stellar mass of about 400 million times the mass of the Sun, it would be a bit of a stretch to call this one especially faint anyway. But what it is is quite dramatically extended to compared even to other, otherwise similar Ultra Diffuse Galaxies. Its stellar mass profile is much more extended than the famous DF44, emblematic of these especially fluffy systems, and this one clearly deserves some attention.

But I'm getting ahead of things. From the outset, this paper is about how this galaxy can help inform us about the nature of dark matter rather than merely saying, "look, here's a system almost entirely dominated by dark matter, isn't that neat" rather than addressing any actual science. Kudos to them for that, this is much needed.

They seem to have found this object accidentally and then gone and got some seriously impressive follow-up data : with an optical surface brightness magnitude of 31 magnitudes per square arcsecond, this is a sensitivity I don't think I've heard quoted outside of Hubble papers before. Their 12 hours of integration on the GBT, however, has only given at best a marginal HI detection (even though it's detected in both polarisations), and it's somewhat annoying that they don't seem to quote its linewidth clearly anywhere. With single-object papers I think it's always a good idea to put the major parameters in a nice clear table, but never mind. I wouldn't have too much confidence in the measurement anyway given how weak the signal is. This doesn't affect their main analysis in any case.

[See edits below for major corrections about these points !]

From their optical data, they find that the colour profile is flat, not varying at all from the innermost regions to the outskirts. So just one single stellar population everywhere with no regions being particularly susceptible or resistant to star formation, which would seem to fit with UDGs in general. The stellar density profile does show a steady decline, indicating that they've likely reached the edge of the object and even deeper observations probably wouldn't reveal anything else.

There seem to be multiple possible ways to form faint, extended galaxies in clusters, but in isolation nobody (so far as I know) has come up with a convincing explanation. Here they ask the obvious question as to whether it could be the result of tidal interactions but the nearest other galaxy is bloody miles away so that seems unlikely; it also has a chemical composition much more typical of larger galaxies whereas tidal dwarfs tend to be even more metal-rich. Diligently, they concede that it could be a very old tidal dwarf that's had billions of years to enrich its metals and could probably have lasted this long quite happily given its isolation, but its extremely high dark-to-light ratio (several tens) and rather low gas content all but rule this out.

What they do next is a pretty neat thing to dry and I'm glad they did it, but personally I wouldn't have had the audacity. They speculate that because such objects don't appear to be compatible with the Standard Model, they maybe the point to something different about the nature of dark matter. They say that something called ultralight axions might be responsible. Now here I want to say YES ! Thank you for trying this and using these objects to probe fundamental physics, that's what's interesting about them ! And then I immediately want to say NO ! You can't provide any meaningful constraints from just one object !

I'm exaggerating somewhat. By no means do they claim to have overturned the Standard Model or anything else, they just venture an intriguing idea. Good for them. But a couple of paragraphs about how incompatible this object is with simulations seems to me to be a bit of a stretch to then go immediately for exotic physics as a viable explanation. I think what's needed here is much more about context. What are the other galaxies in the vicinity like ? Is there deep optical data for any of them ? Can we really be confident about its isolation ?

I don't think there's anything wrong with posing more radical alternatives, but I'd need a lot more persuading that the other possibilities can be so firmly dismissed. To my mind there's still enough uncertainties in baryonic physics that we shouldn't be overly-concerned that simulations aren't predicting objects like this. Regardless, it's good to see someone firmly pointing out the continuing weirdness of isolated UDGs. 

EDIT : After a group meeting it's clear that I've given the authors too much credit here. True, they did include everything in a nice clear table, at which my eyes glazed over so I didn't notice their velocity width measurement. Fair enough, they do state all the major parameters then – my mistake ! But their line width estimate is just 34 km/s, with an error bar of 11 km/s... after smoothing their native velocity resolution from an exquisite 0.7 km/s to a ghastly 25 km/s ! Lots of numbers, but the bottom line is that their velocity width measurement isn't terribly meaningful. And the line width dictates total dynamical mass, so what they can really say about the dark matter mass here is... not much.

Let me break that down a bit more. Almost certainly, they smoothed their initially very precise resolution to increase sensitivity. Nothing wrong with that, and in any case, the HI line itself is seldom less than 10 km/s in width due to the temperature of the gas anyway. So you don't really need 0.7 km/s resolution. But 25 km/s is getting pretty bad, and to try and claim any sort of accuracy that the true width is 34 km/s which is barely higher than the resolution, especially with a signal which is anyway marginal... nah. This is really stretching credibility. Any claims that this galaxy is dark matter dominated a la DF44 need to be viewed very suspiciously until somebody obtains better data.

No stacks please, we're simulationists

Back in 2017 there was a Nature paper claiming to have detected declining rotation curves in galaxies at high redshift. This would mean that galaxies in the distant early Universe have significantly less dark matter that contemporary nearby galaxies, whose flat rotation curves are one of the principle signatures of dark matter in the first place.

I was rather skeptical of this. None of their individual measurements looked the least bit convincing to me, with the fitted curves highly dependent on single data points : the slightest error could have thrown them off (and some curves just don't go through the points at all). True, the stacked curve was much more convincing, but any systematic error in estimating the rotation velocity will only compound this error rather than averaging it out.

On the other hand none of this actually would be evidence against dark matter. As with the now-plethora of Ultra Diffuse Galaxies and the like (in the nearby Universe, with relatively precise, sensitive data) which seem to have a significant and sometimes total deficit of dark matter, all this really indicates is that dark and visible matter can be separated. This is very much harder to do with modified gravity theories. If the rotation curve arises only as a result of baryonic matter, then if you have two systems in which the baryons have similar distributions (and are suitably isolated), then they should always have similar rotation curves. The only reasonable way in which they can differ is if one has dark matter and the other doesn't.

The real question is whether, according to cosmological theory, we expect galaxies in the early Universe to be less dark matter-dominated than today's. Ethan Siegel simply says "yes", that these declining rotation curves are indeed expected in standard models of galaxy formation.

The author's of today's paper, however, say No. And this is certainly more intuitive. If dark matter is mass-dominant, then it seems odd that it would actually become more important over time. Surely it should be gas falling into dark matter haloes that describes the process of galaxy assembly, and if that's the case – if dark matter makes up the bulk of the mass of galaxies from the word go – then they should always have similar (though not necessarily identical) rotation curves to those of the present day.

Now I should mention that the first author was my Master's project supervisor. You can read about this in some detail here, but in brief, we ran a galaxy formation simulation without dark matter and found that it just didn't work (see also my latter efforts failures to build a stable disc without dark matter). Anyway, the arguments that seemed quite compelling to younger me no longer have the same appeal; I'm delighted to see that others are still pursuing this line of inquiry and I wish them well, but I don't see this as likely to be anything more than a dead end. An interesting one to be sure, but I'm not convinced there's light at the end of the tunnel, so to speak.

What they do is run a series of simulations of the monolithic collapse scenario of galaxy formation. This is far simpler than the standard paradigm of hierarchical merging, in which galaxies assemble from a multitude of mergers in the early Universe (which can be a surprisingly efficient process, but then the early Universe was a lot smaller than the modern one). They consider different initial geometries, in which the initial dark matter halo is the same size, smaller, differently-structured, or spatially separated from its associated gas cloud.

They get the same result in all cases. The collapsing monolith experiences violent relaxation which essentially wipes out its initial conditions : the collapse proceeds at ever-increasing speed, the gas shocks and triggers star formation, and the galaxy forms during the re-expansion phase. This means that it doesn't really matter how you start, you get the same thing regardless.

And in this scenario you always get a galaxy with a flat rotation curve. The only way they could get a declining curve is to take out the dark matter altogether*.

* I'm intrigued that this is much more successful than my attempts at the same scenario, which gave something completely unphysical. I'll have to try and follow up on that.

This, as they've argued before, suggests that the initial conditions are irrelevant. But as I understand it, the hierarchical merging scenario is just more radically different than they give it credit for. There's just no reason in modern cosmology to assume the presence of any sort of collapsing monoliths, certainly at the very least not at all as the norm for galaxy formation in the early Universe. I don't think you even get much or any in the way of violent relaxation in this scenario, but rather a continuous assembly of tiny, already stabl-ish- proto-galaxies. So I really don't think it's fair to compare the failure of reproducing declining rotation curves in a monolith collapse scenario with the expectations from standard cosmology; this is comparing apples with oranges.

I also wonder just how similar these objects are to the observations. They don't compare the gas masses at all. Their gas radii are in some gases huge at 70 kpc : not outlandish by any means, but definitely on the larger side. They don't give this comparison to the observations, preferring to normalise their results to more accurately compare the shape rather than the size of the rotation curves.

This isn't a mistake in and of itself. But as in the press release that accompanied the original claim, there are a multitude of different factors in play. The galaxies detected might not actually be the progenitors of modern-day spirals but rather more spheroidal systems, which are anyway known to have less dark matter. There may indeed be significantly less dark matter in early galaxies overall, and those earlier systems might be more dominated by turbulence than rotation – which significantly affects the interpretation of the rotation curve.

This is not to say that there aren't questions to answer. The authors of this study point to other findings of more typical simulations saying that the turbulence isn't enough to account for all this, but they also note other observations of individual galaxies showing flat rotation curves (though they dispute this result) at earlier epochs. And they note that some galaxies might form in dark halos and others not, which seems very likely given all the hoo-hah about UDGs.

All this is very reasonable. The whole thing is a mess with many different variables to juggle. It just seems to me that we're a very long way indeed, further than ever in fact, for needing to invoke other physics like magnetic fields (as the authors do) to explain modern flat rotation curves. It seems far more likely that a combination of different effects are likely to explain early declining rotation curves than it is that they undermine the whole paradigm, which is otherwise supported by a vast array different and independent considerations, both in theory and observation, on a whole set of enormously different scales. 

Do we fully understand galaxy assembly ? Absolutely not. But it's way to drastic to point to a complex result like this and say it calls the whole basis of modern theory into question, and it's just not fair to compare the results of a monolithic collapse scenario with the radically different processes postulated by mainstream cosmology.