Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
A ghostly galaxy lacking dark matter (esahubble.org)
48 points by doener on Dec 12, 2021 | hide | past | favorite | 43 comments


From Wikipedia:

> A more recent study on NGC 1052-DF2 suggests the previously reported distance of the galaxy was greatly exaggerated. Consequently, the galaxy now looks "normal" in every way. Using five independent methods to estimate distances of heavenly bodies, a team of researchers from the Instituto de Astrofísica de Canarias (IAC) found the correct distance of NGC 1052-DF2 to be 42 million light years (13 MPc), not some 64 million light years (19 MPc) from the Earth. The total mass of the galaxy is around one-half of the mass estimated previously, but the mass of its stars is only about one-quarter of the previously estimated mass. This implies a significant part of NGC 1052-DF2 could be made up of dark matter, like any other galaxy.


Can someone elucidate how astronomers know this galaxy has no dark matter?


It’s not revealed in the article for this particular case, but the usual way of determining the dark matter content works by estimating the total mass or mass distribution, and subtracting the visible mass of stars.

The total mass can be estimated from gravitational lensing, and for rotating galaxies the distribution of mass within the galaxy can be estimated using the speed of rotation of stars at different distances from the center.


"Rotating" meaning sighted at an angle so you can use Doppler shifts of familiar spectra to figure out the tangential velocity of parts of the galaxy. In this case they used what they called "gas", meaning I suppose interstellar plasma, as the reference for the Doppler-shifted waves. They could check emission lines from the plasma, or reflections off it, or gaps in transmitted light representing absorption of that light by the plasma.


Dark matter is inferred by the difference in the amount of gravity a galaxy appears to have, and the gravity we'd expect from the mass we observe inside of it. So in this case, the observed gravity is what all the observed mass would predict it to be, instead of being far more, like most galaxies.


it's not actually star mass. only about 1% of the normal matter in a galaxy is from stars. the rest is low density dust.


That raises a question. If there's a second variable, the dust %, how can we solve for the amount of dark matter from the visible mass?


Not an astronomer but i think it would still glow at lower wavelengths (infra-red)


How do we measure the amount of dust?


Usually by how light passes through it from both stars within the galaxy and from galaxies occluded by it.


Couldn't dark matter just be dust that acts differently then? Like what if it were little clumps of dust instead, shouldn't that alter how light passes?


Cold baryonic dark matter candidates have been pretty much all but excluded.

The dark matter most people are looking for only interacts with gravity which means other than gravitational lensing it would not interact with light in any way (and the lensing is technically not an interaction either).

My own personal opinion is that whilst MOND is probably wrong I think the idea of it might be closer to reality that not.

I don’t think that dark matter as in matter that interacts only with gravity is going to be discovered.

Keep in mind this isn’t my field so it’s not a professional opinion it just feels like we are chasing elegance rather than going back to the drawing board.

We are still really bad at measuring distances and thus anything else at those scales the distance of the galaxy in question went from 20 something to 50 something to over 70M light years with consecutive measurements and I’m still not sure we got it right.

We are heavily reliant on “standard candles” for measuring distances and in the end it’s all about calibration which is somewhat still based on guess work as in we hope that we got the objects absolute luminance correct before comparing it to the absorbed luminance.

I should also mention that we not only using standard candles, parallax measurements are also used and those are probably relatively accurate however we can usually only use them for relatively near by stelar objects so they aren’t useful for measuring distances to galaxies or even remote stars within our own galaxy.

We did use parallax to check some of our standard candles but for galaxies many of our standard candles are supernovae not stars so there is still a bunch of variance there.

The galaxy in question makes it even harder to be confident in our ability to measure its distance accurately it’s very sparse, the stars are in it are mostly about 10B years old and are very dim and it doesn’t have much if any star formation.


That dust would invariably collide, heat up a bit, emit that heat as thermal radiation and end up with a bit less energy. Thus slowly but surely its orbit would decay and it would move towards the core like normal matter, because it is normal matter.

However to explain the rotation curves the extra mass must have a different density profile, and it has to be present far beyond the edge of the visible galaxy.

If one assumes a particle which can't emit thermal radiation as it does not interact with electromagnetism, then it won't cool down and will remain in a diffuse blob around the galaxy.

Turns out one can get good fits to galaxy rotation curves with a huge diffuse blob of such a particles around a galaxy.


Thanks, edited


At a guess, it’s mass (measured by lensing) is accounted for by its light output.


TFA suggests its mass was estimated by rotation curves, not lensing.


tldr is that the rate a haircut rotates is proportional to its mass (and mass distribution). most galaxies spin faster than they "should", so there has to be a lot of mass that we can't observe using telescopes. this one doesn't.


I though the haircut was an analogy at first ;D

Nice auto-correct



I don’t work in cosmology and can’t vouch for the site, but according to [0] the dark matter was essentially stolen by neighboring galaxies through a process called tidal disruption.

[0] http://www.sci-news.com/astronomy/tidal-stripping-lack-dark-...


TFA explains there are no plausible candidate galaxies.


Oh I see. The sci-news article is talking about the -DF4 galaxy, but the subject of the esahubble article is the -DF2 galaxy.


You’re talking about the original link from esahubble? It doesn’t mention that at all. That’s why I went digging.


The photino birds got clobbered by the Xeelee in that galaxy.


Some people seem to think this is a worthless comment, I prefer to see it as a good book recommendation. It also jives with my feeling that dark matter has something to do with life being present in a galaxy, the effect it has on the structure of galaxies seems almost... deliberate.


Like a Kardashev type III civilisation(s).

I love that idea.

Edit: thinking about it more, it's even more interesting because this particular galaxy is far less dense than say the Milky Way, so perhaps it's basically unsuitable for colonisation/exploitation.

It also means that the Milky Way contains a Type III civilisation.


It also means almost all galaxies contain similar Type III civilizations. 13 billion years is a long time, how long did Fermi calculate it would take to colonize the galaxy? 10 million years?


No intelligent design or superstitious nonsense here, thanks.


Gatekeeping aside, I took the meaning of the comment to be that there is something in the galaxy that has the ability to construct/manipulate their surroundings in such a way that it appears to us as “intentional, from living beings”


Yes, thank you for thinking outside the box.


Yeah, well. That is an observation that is pretty bloody difficult to explain with MOND type theories.

Any MOND proponents want to have a try?


Not a MOND proponent but why does it exactly contradict it?

It’s a UDG with very little to no star forming gas and very low overall mass and density, looking up the galaxy it’s about the same size Milky Way but has less than 0.5% of its stars and much less than that when it comes to overall mass.

It seems that there are some papers that attempt to reconcile the observations with MOND and on the face of it the velocities expected if MOND holds true and are not far off from the current observations.

https://arxiv.org/abs/1804.04167

Considering the distance of the galaxy (72m light years) and the difficulty of establishing the actual mass of UDGs and even globular clusters at that distance it doesn’t look like a MOND killer not nearly as much as gravitational waves and the overall challenges to generalize MOND and make it relativistic.


Let me preface this by saying that I am by NO means an expert in any of this.

My thought process was simply along the lines of:

1) If MOND is true, it should be evident in the rotational structure of every galaxy observed.

2) Here is a galaxy that doesn't (at first glance, at least -- I need to read the paper you pointed out) show such rotational structure.

3) Hence, this is an "outlier" observation in the panoply of the MOND vs dark matter debate -- one in which MOND seems unsupportable.

Nothing deeper than that, and again this is ABSOLUTELY NOT my field!


I think one should run simulations of MOND first; MOND does not directly make a statement about rotational structures; it makes a statement about the gravity law. This galaxy clearly has unique structural considerations that could yield results that you don't expect from a trivial glance at higher order factors.

Edit, after digging: someone has, apparently.

https://www.researchgate.net/publication/336982682_A_new_for...

I can't speak to whether or not the paper is any good, but my quick understanding is that: the galaxy is consistent with MOND if the distance to the galaxy has been misestimated; at least one published estimate (that disagrees with most estimations) of how far it is to 1052 falls inside the range where it's explainable by mond. My (limited; I last took astro a very long time ago) understanding is that how we measure distances to foreign galaxies (slam-dunk methods like cepheids don't work if there are no cepheids) could be very sketchy for galaxies of different structures. Here is part of the ongoing debate about distance. Linked is a paper from the perspective that would disfavor MOND (the Trujillo estimate is what needs to be true for 1052 to be consistent with MOND according to the first paper):

https://www.pietervandokkum.com/ngc1052-df2


MOND has very little to say about gravity. It claims that f=ma breaks down at sufficiently low values. That, in turn, implies that the rules should still be consistent with these modifications, and that galaxies with apparently different proportions of dark matter (as it's understood) shouldn't exist - all galaxies with similar visible sizes and structures should be essentially identical in gravitational behaviour, and that simply hasn't been demonstrated to be the case.


> with similar visible sizes and structures

This galaxy is very much not in the normal range of "sizes and structures", and is highly dissimilar to almost all other galaxies observed.

the stuff about changing F=ma vs changing Fg is kind of pedantic and immaterial to the topic of simulation, given that the accleration due to gravity is the only thing that is being simulated.


I’ve linked above to a paper that attempts to establish what to rotational speed of the clusters should be based on MOND and it’s close enough to the observations that in isolation it still the case that the galaxy either has 1/400th of the dark matter density a normal galaxy would have vs we don’t need dark matter due to MOND.


FWIW, NGC-1052-DF2 has it's own Wikipedia page, that discusses (and more importantly cites some literature about) the relation between it and MOND.

https://en.wikipedia.org/wiki/NGC_1052-DF2


>the galaxy it’s about the same size Milky Way but has less than 0.5% of its stars and much less than that when it comes to overall mass.

What's the mechanism behind the formation of such a sparsely populated galaxy? Was there just way less gas in that region of space for stars to form in the first place? How is there even enough gravity to bind those stars into a galaxy rather than just a forming a series of rogue stars that move independently of one another?


Nobody knows. Galaxies like that were not supposed to be possible.


(What is meant here is Modified Newton Dynamics, see https://en.m.wikipedia.org/wiki/Modified_Newtonian_dynamics )


https://tritonstation.com/2018/04/04/the-dwarf-galaxy-ngc105...

A post by MOND proponent. Not sure what if anything changed in the data since then.


Dark matter is a very weak hypothesis and this is a good reminder that MOND explain more data.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: