Why ESO surveys aren't always published

Everyone bangs on about "publish or perish", but here's a look at the opposite case : astronomy surveys which never produce a publication. Specifically, surveys using European Southern Observatory facilities, about half of which don't produce a publication in the timeframe specified. However, about 36% of those (so about 18% overall) were still working on the data at the time the survey was conducted, and this turns out to be compatible with the average time between collecting data and producing a publication. And 10% had simply been misidentified, and had published a paper after all. Hence through lack of time I won't attempt to correct for this in the rest of this little summary.

There's sometimes talk of null/negative results not getting enough publication. This turns out to be a significant factor in the remaining cases, but not at all the dominant one. There isn't really a single dominant factor, it's a combination of things. That said, some reasons are essentially negligible : the authors just lost interest, they published a non-refereed paper instead, they had inadequate tools to process the data.

More significant are that the observers just didn't get as much or as good quality data that they required, they lacked the resources (e.g. manpower) to complete the analysis to get a meaningful result, or a variety of other reasons. That "other reasons" category sounds quite interesting :
...the most frequent being that the person leading the project left the field. Other recurrent explanations included: lack of ancillary data from other facilities, results not meeting expectations, lowered priority of the project because of more pressing activities, quicker results obtained by other teams and/or with better suited instruments, nondetections, etc.
Which sounds to me like the "results aren't interesting enough" slice of the pie could probably be increased a bit. We keep hearing about the importance of replication, but there's a motivational pressure against doing things people have done before - even if they might not have done it very well.
http://doi.eso.org/10.18727/0722-6691/5055

The life of a space jellyfish

Yesterday's jellyfish is allliiiiiiiive !

Space jellyfish !

This is a simulation in progress of a galactic disc being smashed with a wind from an external intracluster medium. Ram pressure stripping is a well-known phenomenon which is thought to be the main way gas gets stripped in galaxies in clusters. As galaxies fall through the hot, relatively thin gas of a galaxy cluster at high speed, the pressure of the external gas builds up to the point where it can push out the galaxy's own gas.

In this particular simulation, the gas in the galaxy's disc is especially thin (so as to prevent it forming stars, and match observations of some weird dark hydrogen clouds). This is a simple "wind tunnel" test, where we approximate the ram pressure by having the external gas move at a constant speed. Computationally this is much easier than setting up a galaxy falling through a cluster in a more realistic way. It's not too awful an approximation, but probably in this case the external gas is too thin and moving too quickly.

This simulation was done to burn up some CPU time we had left on a supercomputer. We didn't get as much as we'd like but we got enough to see something cool. The output files are still being converted into a format I can handle; expect a proper animation (this one is rotation only) in a few days.

Rendered in FRELLED : http://www.rhysy.net/frelled-1.html

Prestigious journals are not especially reliable

This is a nice, detailed, statistically careful piece.

Two main conclusions can be drawn: (1) experiments reported in high-ranking journals are no more methodologically sound than those published in other journals; and (2) experiments reported in high-ranking journals are often less methodologically sound than those published in other journals.

The prestige, which allows high ranking journals to select from a large pool of submitted manuscripts, does not provide these journals with an advantage in terms of reliability. If anything, it may sometimes become a liability for them, as in the studies where a negative correlation was found. This insight entails that even under the most conservative interpretation of the data, the most prestigious journals, i.e., those who command the largest audience and attention, at best excel at presenting results that appear groundbreaking on the surface. Which of those results will end up actually becoming groundbreaking or transformative, rather than flukes or frauds, is a question largely orthogonal to the journal hierarchy.

Well, a prestigious journal will only accept exciting claims, by definition. And exciting claims are much more common at low reliability levels. This isn't really news, but it's nice to have statistical confirmation of a hitherto largely anecdotal claim.

It is up to the scientific community to decide if the signal-to-noise ratio in these journals is high enough to justify the cost of serial scandals and, in the case of medical journals, loss of life, due to unreliable research.

See also http://astrorhysy.blogspot.cz/2017/03/this-is-not-crisis-youre-looking-for.html


https://news.ycombinator.com/item?id=16427990
https://www.frontiersin.org/articles/10.3389/fnhum.2018.00037/full

The Clarke Exobelt : an exercise in silly numbers

This is a fun little paper considering SETI by looking for sufficiently dense belts of satellites around other planets.

This paper puts forward a possible new indicator for the presence of moderately advanced civilizations on transiting exoplanets. The idea is to examine the region of space around a planet where potential geostationary or geosynchronous satellites would orbit (herafter, the Clarke exobelt). Civilizations with a high density of devices and/or space junk in that region, but otherwise similar to ours in terms of space technology, may leave a noticeable imprint on the light curve of the parent star. In some cases, a Clarke exobelt with a fractional face-on opacity of ~1E-4 would be easily observable with existing instrumentation.

But, there's a catch, which I wish they'd properly quantify :

A particularly useful database is the compilation of data from public sources made by the Union of Concerned Scientists. Currently, the list contains parameters for 1738 satellites, of which approximately one third are in geostationary or geosynchronous orbits. Assuming a typical radius of 1 m, we have that [maximum] opacity [the fraction of light in our line of sight that is blocked by a surface element] = 3E-13.

In order to become visible from nearby stars with our current observational capabilities, the Clarke belt of our planet would need to be about 1E-4. However, our belt is becoming increasingly populated. Figure 3 shows that the Earth belt opacity opacity has been growing exponentially over the last 15 years. If this trend is extrapolated into the future, we would reach the “observable” threshold around the year 2200.

Assuming that opacity is a linear function of the number of satellites, keeping them all the same size, then the required opacity requires an increase in the number of satellites by a factor of 300 million, or a total of about half a trillion satellites. Maybe I'm wrong. I'm feeling very lazy. It works out to an average of 2.5 billion satellite launches per year, but of course because it's exponential the peak rate will be much higher... I'm not sure it's fair to label a civilization with a launch capacity exceeding 2.5 billion per year (80 per second) as "moderately" advanced.

Obviously, this extrapolation should not be viewed as a prediction. There is no reason to assume that the current exponential growth will be sustained for another 200 years. In summary, the 2200 date is not even a rough guess of when humanity will reach detectability threshold but rather an indication that this outcome is a reasonable expectation for the near future, given current trends.

But is it though ? What would be the average density of a belt of 500 billion satellites ? What would their average separation be ? Would it even be feasible to able to add any more at that point without them colliding in a cascading catastrophe ? Anyone else wanna try the maths ?

Later, they note that the observations would be able to determine the radius of the belt accurately to tell if the objects were in "geo"synchronous orbit or not. This is nice, because : "Geostationary orbits are very interesting for a society but are not preferred by any known natural process."

The simulations presented here show that CEBs may in some situations be detectable with existing instrumentation. The best candidates are planets around red dwarfs in tidal locking, in line with the optimal conditions for habitable exoplanet search. An initial difficulty would be how to distinguish between a CEB and a ring system. However, once a candidate has been identified, detailed follow-up observations may resolve this ambiguity from the shape of the light curve. In any case, the detection of a dense belt of objects at the distance of geostationary orbit would be a very strong evidence for the presence of ETI, especially considering that rings around habitable rocky planets are probably rather uncommon.

Oh, and then we get something fun :

The total mass of the entire belt for all the cases considered here is between 1E12 and 1E14 kg, assuming average object radius and mass of 1 m and 100 kg. This range is between the mass of a comet and that of a mountain. It is not an unreasonable requirement for a moderately advanced civilization.

Which gives a number of satellites from 10 billion to 1 trillion, in agreement with my earlier calculation. Yay ! I'd still like to know the density of the belt or mean free path though.

What I think would maybe be more interesting to know is what it would take to be able to detect satellite belts of density much more comparable to that of our own - say, no more than 1,000 times greater. Does the sensitivity requirement scale linearly ? Will we be able to make significant progress with, say, the ELT or TMT ?

Ah screw it, let's have a stab at estimating the average distance between satellites. They say the maximum inclination of geo_synchronous_ satellites from the geo_stationary_ orbit is 15 degrees. Geostationary's at a radius of 36,000 km, so 36,000*tan(15) = 10,000 km (rounding terribly, because this ain't gonna need to be exact). They also say the allowed spread in altitude is 150 m. So we have a hollow cylinder, inner edge 36,000 km, outer edge 36,000 km + 150 m, height 10,000 km. Total volume = 3.4E11 cubic metres.

Total volume of satellites = 5E11, since their average diameter is given as 1m and assuming them to be cubes because why not. So the fraction of space occupied by the satellites is... greater than one.

Ahh, crap. Someone else can have a go.
https://arxiv.org/abs/1802.07723

Understanding your own ideas is hard work

More recent strata lie on top of older strata, except when they lie beneath them. Radiometric dates obtained by different methods always agree, except when they differ. And the planets in their courses obey Newton's laws of gravity and motion, except when they depart from them.

There is nothing that distinguishes so clearly between the scientific and the dogmatic mindset as the response to anomalies. For the dogmatist, the anomaly is a "gotcha", proof that the theory under consideration is, quite simply, wrong. For the scientist, it is an opportunity. If an idea is generally useful, but occasionally breaks down, something unusual is going on and it's worth finding out what. The dogmatist wants to see questions closed, where the scientist wants to keep them open.

All well and good.

This is perhaps why the creationist denial of science can often be found among those professions that seek decision and closure, such as law and theology.

Umm, what ? Is there an anti-science agenda in the law courts that no-one's told me about ? And throwing theology in there is just asking for trouble... but we may dismiss this careless throwaway comment, because the rest of the article is very nice.

No theory exists on its own, as the philosopher-scientist Duhem pointed out over a century ago, and when a theory fails an observational test there are two kinds of possible explanation. The fault may lie with the theory itself, or with the assumptions we make while testing it. More specifically, as Lakatos pointed out in 1970, every application of a theory involves ancillary hypotheses, which can range from the grandiose (the laws of nature are unchanging) to the trivial (the telescope was functioning correctly). When a theoretical prediction fails, we do not know if the fault is in one of these, rather than the core theory itself. Much of the time, we are not even aware of our ancillary hypotheses, which is one reason why we need philosophers of science.

I don't like the term "ancillary hypothesis". I'd have gone with something like "underlying / hidden / implicit assumption." But I digress.

No scientific theory is rejected simply on the basis of its anomalies. It is rejected only when a superior theory is put forward, and the new theory is superior if it explains as much as the old theory, and more besides. Thus we should not even see theories as existing in isolation, but as part of a sequence or research programme. You are bound to be wrong, but don't let that worry you unduly, because error is opportunity, and the way science progresses is by being less wrong about more things. I find this viewpoint liberating.

... The anomaly in the orbit of Mercury, however, could not be resolved in this way, and remained unexplained until the formulation of Einstein's General Theory of Relativity. In the case of Uranus, the anomaly was associated with the ancillary hypothesis that we had a complete list of planets, and it was this ancillary hypothesis that was overthrown. In the case of Mercury, however, the shortcoming was in the theory itself.

... From the perspective of this essay, Rayleigh's initial thinking included the ancillary hypothesis that all the components of air had been identified. This was not true, and (as readers with our knowledge of chemistry will be aware) the additional component was to play a vital role in explaining chemical bonding.

Well, this explains a lot about why ongoing research is so messy. It's because when a researcher formulates an idea, they only consider a limited number of parameters that have drawn their attention to something interesting. They want to suggest some mechanism to explain the data. During that process, it's the mechanism itself that they're interested in - not in the implicit assumptions it requires. Those may well be interesting in themselves, maybe more so than the theory. But not at that particular moment. The point is that for a complex problem, the researcher will not even be immediately aware of the assumptions they're making. Figuring out what those are takes time - usually, years. And this is why it can often look as though scientists are flogging a dead horse, adding yet more and more adjustments to an "obviously" broken idea; adding more crystal spheres and epicycles. They're not. What they're actually doing is figuring out (and testing) the assumptions behind the idea that they already made years ago, but were unaware of at the time.

http://www.3quarksdaily.com/3quarksdaily/2018/02/in-praise-of-fallibility-why-science-needs-philosophy-.html

The Ones That Got Away

Since I follow anything relating to dark extragalactic hydrogen clouds with the obsession of a rabid dog, I'm amazed that I missed these. But I did.

This paper is about optical follow-up observations of two extremely massive hydrogen clouds near the obscure galaxy IC 5270. These were previously reported back in 2015. Most such clouds are thought to be debris, removed from parent galaxies either by tidal encounters or other gas stripping processes. There are a handful of dramatic, very extended hydrogen streams known, and quite a few more smaller, discrete clouds, which tend to be low-mass.

Not so these two stonkin' great clouds, both of which have about a billion times the mass of the Sun in hydrogen (a few times less than that of a giant spiral like the Milky Way). They're very close to IC 5270 itself, no more than a few times its optical diameter, which has about 8 billion solar masses of hydrogen. One other galaxy in this group shows a disturbed hydrogen disc, but IC 5270 itself does not. And while it's possible to tidally strip the outermost, thinnest parts of a gas disc without doing too much damage to the rest of the galaxy, it's very difficult to imagine what process could remove so much gas while leaving the galaxy itself unscathed. In other environments, ram pressure stripping (the effect of a galaxy moving through a hot, low density, external gas belonging to the group itself) might be viable, but this group is too small to likely possess much of this - and anyway the motion of the galaxy is just too slow for it have any effect.

Similar amounts of more obviously stripped material are known in other cases, but the key difference is the size. These two clouds are both relatively compact - at about 20 kpc in extent, they're not so different from an a typical spiral galaxy. Stripped material of this mass tends to be very much larger. This means the density of the clouds appears to be comparable to that of standard spiral galaxies, well above the threshold where star formation is normally occuring. Yet they're almost entirely optical dark. The new observations presented in this paper identify some truly pathetic optical counterparts, but these don't even well-match the positions of the centres of the clouds (both look to be in parts of the same cloud - the other remains entirely dark). They may well be associated, but they still challenge the expected star formation efficiency. The authors speculate that the clouds might be more extended along the line of sight, giving them a greater volume and so making the gas less dense than it appears.

So what the hell are these things ? It's very difficult to say. Based on earlier data, there could be a significant amount of extra hydrogen over a more extended area. This is a great weakness of most new major radio telescopes : they're interferometers, which are hundreds of times less sensitive to extended structures - they're limited to the densest gas. This is still very interesting since in these cases the clouds are so dense that they ought to be forming stars, but it means we might be seeing only a small part of the system.

Frustratingly, there's no measurement of the velocity width of the clouds. And that's crucial, because we've shown with extensive simulations that while it's relatively easy to produce a long stream with a high velocity gradient, it's really difficult to produce isolated clouds exactly(!) as small as these with high gradients. They say that one of them has a "slight" gradient and the other is "homogeneous", but they don't quantify this; figures in the earlier paper lack labels for the velocity colour bar. Which is mmwwwwwaaaaargh because this is something for which numbers really matter - double the gradient and it gets vastly more difficult to produce, at least according to our models.

Furthermore, the clouds are both on the same side of the galaxy. Most tidal encounters produce a double tail (just as the Moon produces high tides on opposite sides of the Earth). High velocity encounters can produce one-sided features, on occasion, but this shouldn't happen here because the velocity dispersion of the group is so low - any anyway there are very few other group members. Low velocity encounters can also do it, but it takes a long time for the system to evolve to that point - and the clouds remain suspiciously close to their assumed parent galaxy.

So I have no idea how these things formed, and that's why I like 'em.
https://arxiv.org/abs/1802.05279

Explore the Virgo Cluster in VR

Haven't you always wanted to explore the Virgo Cluster in VR ? Of course you have, and now you can ! It's rather crude but it basically works, though it's more of a test than a finished product. There's a bit of flickering in some regions which I don't understand and the galaxies looks weird from certain angles because I'm using the old multiple-billboards trick. I should probably have all the images track the camera instead, but oh well.

Also the resolution is lower than I'd like because headset can't play back anything higher. It's quite fun despite its many deficiencies. At least thanks to YouTube you can still view it in 2D 360, or with anaglyph glasses if you have them.

More info on how the galaxies are rendered available here.

https://www.youtube.com/watch?v=Q-UQjm7equk

Computers are logical, my foot

They say insanity is doing the same thing over and over again and expecting to get a different result. Well, my code is currently taking a bunch of files, doing a fixed operation on those files, and spitting out a different number of files each time.

Yay.

Stealth in space : define the problem first !

It's OK to be sensitive

I'm prompted to write this by discussions on stealth in space, but really it could apply to absolutely anything. I don't have time to write this up more fully, so this'll have to do.

When we search for something, what do we mean by the "sensitivity" level of our survey ? There are three basic ideas everyone should be aware of.

1) Sensitivity. Any survey is going to have some theoretical hard limit. Astronomical surveys always have noise, political surveys always have errors. If a source is below your noise level you have no chance of detecting it; if your question was flawed there are answers you won't be able to obtain. You might be able to improve this by doing a longer integration or asking more people, but with the data you've actually got, you're always limited. Above this limit, you might be able to detect something. Here's where it gets more subtle and often complicated - it's a very bad idea to take a given "sensitivity limit" and apply this without thinking more deeply about what it means.

You're going to have some procedure for extracting the data that you're interested in, e.g. number of stars in a given region. This procedure, like the data itself, will have its own errors. You're going to find some sources which aren't really stars at all, and miss some thing which are real stars completely. In general, the closer a source (be that a star or anything else that's detected, even if it isn't real) comes to the noise level, the more problems which will result. In particular :

2) Reliability. Some of what you detect will be real, but some of it won't. Reliability is defined as the fraction of things you find which are really what you were looking for. If you go looking for elephants and find 100, but when you take a closer look at your photographs later on you realise than 25 of them were actually cardboard cut-outs that looked like elephants, then your survey is 75% reliable.
For the definition itself, the total number of elephants actually present is irrelevant. In practise, if you're got 9,00 real elephants and 100 fake elephants in a dense forest, this is going to be a lot harder than 9 real elephants and 1 fake elephant in an open field. However, reliability can be quantified relatively easily : you have to go back, examine each source more carefully and if necessary get better data for each one. There's usually some perfect test you can do to distinguish between an elephant and a stick or a very large horse. One should keep in mind that this will vary depending on the particular circumstances; an automatic elephant-finder might be 50% reliable on average, but 99% in open sandy deserts and 2% reliable in dark grey rocky canyons. So even reliability figures must be handled carefully.

3) Completeness. This refers to the fraction of real sources you're interested in that you actually detect. Say you survey a volume of space containing 100 stars and you detect 99 of them. Then your survey is 99% complete. Easy peasy, except that it isn't. In practise, you very rarely actually know how many sources are really present. Quantifying completeness can be much harder than quantifying reliability, because you can't measure what you haven't detected. You can make some approximations based on those things you do detect, and hope that the Universe isn't full of stuff you haven't accounted for, but you can't be certain.
Galaxies are a bit of a nicer example than stars because they're extended on the sky. Consider two galaxies of the same total brightness but with one 10 times bigger than the other. You might think that the larger one is easier to detect because it's bigger, but this is not necessarily so - its light will be much more spread out, so its emission will be everywhere closer to the theoretical sensitivity limit. Whether you detect it or not will depend very strongly indeed on your survey capabilities and your analysis methods. And at the other extreme, if the galaxy was much further away it could look so small you'd confuse it for a star.

What's that ? You say your fancy algorithm can overcome this ? You're wrong. These issues apply equally to humans and algorithms searching data. Now you can, to some extent, improve the quality of the data to improve your completeness and reliability. For example in astronomy it's far more subtle than just doing a deeper survey - different observing methods produce very different structures in the noise, which can sometimes create features which are literally impossible to distinguish from the noise without doing a second, independent observation - no amount of clever machine learning will ever get around that. So choosing a better type of survey or asking better questions can get you a much better result than just taking a longer exposure or asking more questions to find out what the answer is. But even these improvements have limits.

A good example is the recent claim of a drone which automatically detects sharks (http://www.bbc.com/news/av/world-australia-41640146/a-bird-s-eye-view-of-sharks), which apparently has a 92% reliability rate. The problem is that this tells you absolutely nothing (assuming the journalists used the term correctly) about its completeness ! It might be generating a catalogue of 100 shark-shaped objects, of which 92 turn out to be real sharks, but there could in principle be thousands of sharks it didn't spot at all. Of course that's very unlikely, but you get the point.

Completeness and reliability both vary depending on the type of thing you're trying to observe and the method you're using. For example, the drone might detect 92% of all Great Whites but miss 92% of all tiger sharks (for some reason). Or your survey might be great at detecting stars and other point sources but be miserable at finding extended sources. For any survey, the closer the characteristics of your target are to the theoretical limit, the more problems you'll have for both completeness and reliability. In short, the fact that something is theoretically detectable tells you very little at all about whether it will be detected in reality.

So for stealth in space, it's imperative to define very carefully what you mean by stealthy. Do you mean you want a Klingon battle cruiser that can sneak up and poke you in the backside before you notice it ? Or do you just want to hide a lump of coal in the next star system across where, hell, it's difficult enough to detect an entire planet ? What level of risk are you willing to accept that it won't be detected - or might be detected, but not actually flagged as an object of interest ? Because whatever survey is looking for your stealth ships is gonna have some level of completeness and reliability which will depend very strongly on the characteristics of what it's searching for. It might very well record photons from the ship, but that tells you nothing about whether anyone will actually notice it.

Statistics on a postcard

Very nice article on understanding statistics, summarised on a postcard :
https://www.ft.com/content/ba4c734a-0b96-11e8-839d-41ca06376bf2

Or if you prefer, an audio interview is here :
https://www.ft.com/content/850fa787-b463-4156-a11d-da70c8620eb3

"But wait !", I hear you cry. "Don't you already have a nearly identical blog post about this ?"
"Yes," I reply, "I do. And since I'm resharing something of near-identical content to my own, I suppose this means I'm guilty of confirmation bias. Also, talking to myself."
http://astrorhysy.blogspot.cz/2015/11/sense-and-sensible-statistics.html


“It is better to be vaguely right than exactly wrong,” wrote Carveth Read in Logic (1898), and excessive precision can lead people astray. On the eve of the US presidential election in 2016, the political forecasting website FiveThirtyEight gave Donald Drumpf a 28.6 per cent chance of winning. In some ways that is impressive, because other forecasting models gave Drumpf barely any chance at all. But how could anyone justify the decimal point on such a forecast? No wonder many people missed the basic message, which was that Drumpf had a decent shot. “One in four” would have been a much more intuitive guide to the vagaries of forecasting.

Exaggerated precision has another cost: it makes numbers needlessly cumbersome to remember and to handle. So, embrace imprecision. The budget of the NHS in the UK is about £10bn a month. The national income of the United States is about $20tn a year. One can be much more precise about these things, but carrying the approximate numbers around in my head lets me judge pretty quickly when — say — a £50m spending boost or a $20bn tax cut is noteworthy, or a rounding error.

I want to encourage us all to make the effort a little more often: to be open-minded rather than defensive; to ask simple questions about what things mean, where they come from and whether they would matter if they were true. And, above all, to show enough curiosity about the world to want to know the answers to some of these questions — not to win arguments, but because the world is a fascinating place.

Satellite planes still not a thing

A while back I was raving on about claims that satellite galaxies lie in narrow planes around their host galaxies. This is potentially really interesting, since the standard models of galaxy formation, flawed as they are, predict that satellites should be found in roughly-spherical clouds rather than narrow planes. You can find a full write up here (which is uber-long, but you can read section 2 in isolation - I'll try and make a shorter, more bloggy version soon).

In short, claims for such planes use dodgy statistical analysis and in one case border on the nonsensical. There are three main claims for these systems :
- The Milky Way. This one is solid and probably unimpeachable. And it's genuinely interesting and deserves an explanation.
- Andromeda. This one is marginal - it was found by deliberately selecting the thinnest structure, which is guaranteed to find a thin structure even in a spherical cloud ! There's a hint of a structure when you consider the motions of the galaxies, but I wouldn't call it any more than a hint.
- Centaurus A. The claim was that this elliptical galaxy had not one but two planes of satellites, which to me looked to be simply absurd. The paper didn't even explain how the planes were selected, they just arbitrarily decided there were two "distinct" planes because why not.

This latest paper, which has passed peer review, now says that the previous claims for two planes were wrong and that there's only one plane. But there isn't, as the gif below shows. There just isn't. At best, there's a marginal hint that the cloud is elongated, but take away a mere two galaxies and even that vanishes (equally, if the surveys have missed any galaxies, that apparent elongation could easily disappear).

You can watch the author's own video here : https://www.youtube.com/watch?v=f1GoVAyHH3E
But I don't like it. Halfway through, they select the structure they want you to see - I defy anyone to claim they would have spotted this by themselves in the gif !

What is a bit better is that here they have measurements of the line of sight motions of the galaxies, and that is consistent with rotation. But they don't know the true 3D motion of the sources across the sky, and in any case it's still a very thick structure and not at all plane-like. Furthermore, other models (which I know via private communication that the authors are aware of but don't cite) have shown that interactions between galaxies can perturb their satellite clouds into planes and preferentially destroy satellites moving in certain directions. So even the apparent rotation of this system isn't particularly impressive.

EDIT : That's "not impressive" in the sense that it contradicts CDM predictions. I don't dispute that the system is consistent with rotation or is even rotation dominated, which is, admittedly, definitely interesting. But a plane ? No, I can't accept that. I just don't see it.

The paper suffers from all the same problems as the previous satellite plane papers. They make a very strong claim from this spheroidalish cloud of points, claiming that it's in "serious tension with the expectations from the standard model" and that finding "three such systems is extremely unlikely." But it isn't and they haven't. Various mechanisms have been proposed to explain the planes, which they largely ignore. Their estimate of how rare the planes in the standard models smells extremely fishy - they don't comment on the leading paper which does find such planes, they demand a very precise match between the simulations and observations (down to the number of satellites, rather than, say, the fraction in a plane; also the isolation of the system seems - maybe - excessive). And then they make no comment on the systems which do match the observations in this model. That is a serious error : if there is a physical mechanism for the plane formation, it makes no sense to keep insisting they only form by chance - firstly because if Cen A is in a similar situation to the galaxies in the simulations, then you should absolutely expect it to have a plane, and secondly because you can't multiply probabilities if they're not independent. And of course, they insist that the Andromeda plane is definitely real, despite being marginal.

Even Snakes on a Plane was more believable than this.


https://arxiv.org/abs/1802.00081