HN user
Measurements of pretty much any time in the universe can test predictions made by models of the early universe. One of the main reasons we think there was inflation is from late time (near today) observations of matter density (see https://en.wikipedia.org/wiki/Flatness_problem).
Inflation happened during the first tiniest fractions of a second post big bang. No telescope is going to make direct observations of the inflationary period so I'm not sure what you mean by this.
I worked with someone (this is in astronomy) who said that papers in nature were the most likely to be wrong. They are in nature because they have a dramatic (new/unexpected) result. One good reason for a new/unexpected result is a mistake somewhere.
Edit: I'm not saying anything about this paper. I know nothing about this. Just a meta comment that, in really hard to get published in journals, there might be a bi-modality of papers. Really important and really wrong :)
Small stars live a very long time. A star of the mass of our sun has a lifespan of ~10 billion years which gets you almost back to the beginning of the universe (~13 billion). This star is smaller (0.8 solar masses), so it can live even longer, so we don't need to look at distant objects to see early, low mass stars.
The other issue is that we just can't see individual stars at cosmological distances except in incredibly rare cases. See for example [1] where they discovered a single star at redshift 1 (roughly 6 billion years ago). Basically, if the star gets lensed in just the right way, it can be hugely magnified. This is a strong contender for the coolest observation that I know about. Galaxies are hard to see at redshift 1, to get a single star is crazy.
On your second question, generally people think that galaxies formed "inside-out". I.e. the inner region forms first, then the outer region. See [2]. Stars almost never collide with things (except maybe at the very center of the galaxy, but still super rare there), so survival isn't really a function of position in the galaxy.
That said, this observation is in stripe 82 (a very famous section of the SDSS. They observed it to greater depth than the rest of their area and many other surveys have since also observed it) which I'm pretty sure is away from the galactic plane (so not straight toward the center).
[1] https://iopscience.iop.org/article/10.3847/1538-4357/ab2888 [2] https://www.nasa.gov/mission_pages/WISE/multimedia/pia17554....
That's a weird distinction to try to make. Metal does mean the elements, not the spectral lines. The presence of those elements is usually inferred from the spectral lines, but if another method was used we would still say "metal poor/rich".
It seems significant that the lens galaxy has little dark matter
I don't see that in the article? They centered on galaxy clusters which contain huge amounts of dark matter. DM is needed to account for lensing. There isn't enough mass in the stuff we can see to account for the lensing signal.
How is DM supposed to cool, anyway
It can't! The way that normal matter cools is by radiating (giving off light). Dark matter doesn't radiate (if it did, we would be able to see it). So, while normal matter cools and settles to the center of the gravity well, dark matter doesn't and (as you say) forms physically larger structures.
I think it was taken at a particular time. If it were the closest, venus would be closer than the sun.
Skimming the paper it looks like it was the positions in Aug 2003
The early universe's expansion was not accelerating, as the early universe was not dominated by dark energy. I don't know off the top of my head exactly when that changed, but z=0.76 seems about right. So I'm pretty sure that is what that is.
The reason the dominant form of matter changes is fairly simple. As the universe expands, the amount of matter doesn't change, so the matter density goes as 1/r^3. The amount or radiation goes as 1/r^4 as there is an extra loss of energy due to redshifting (E = hc/wavelength). Dark energy though is (we think) a constant, so as the universe expands, the amount of it stays constant.
There's no strong lensing there. If the background galaxies were lensed enough to have multiple images they would also be incredibly distorted.
But that does look like a small group/cluster.
Edit: Here is an example of strong lensing - see the very curved object just below the bottom bright galaxy https://viewer.legacysurvey.org/?ra=39.9717&dec=-1.5822&laye... This is https://en.wikipedia.org/wiki/Abell_370
Eek! I guess I'm more OK with this now...
The best suggestion I have is https://astrobites.org/
These are paper summaries, written by people in the field, where you are probably the target audience. I don't read it myself, but give it a go!
Yup, by fringe I meant "On the edge of what reasonable people are working on". But maybe it's not a great term if people misunderstand.
That's a fair point. I love reading computer hardware rumours, most of which are probably total garbage (and probably obviously so to anyone in the field). And in this case, whether the general public thinks the universe in MOND or LCDM really doesn't matter at all.
My real issue is when this reporting is on things where the general public's opinion does matter. Things that the general public might vote on. Economics, medicine, etc. Having seen this type of reporting in a field that I do know something about (and a field where there is no real incentive to mislead, again MOND vs LCDM, who cares), I'm a lot more distrustful of science reporting in fields I don't know much about (and where there are incentives to mislead).
If they had published the article exactly as is, giving you all the excitement, but just added a single line somewhere saying "this is new work that is up against a large body of previous work that points in the opposite direction. Let's see what happens, but its a cool idea" I'd be totally fine with it.
No-one is making an argument by authority. Here is a nice popular article that outlines some of the issues with MOND https://www.forbes.com/sites/startswithabang/2018/03/06/only...
I just think that it is very hard to understand the scientific consensus (average view of people who spend a lot of time thinking about this) when all you read are popular science articles that tend to focus on the exciting/new/possibly game changing edges. I'm just here letting people know what the consensus is.
The astrophysics journal (ApJ) is a really good journal. A majority of good astronomy papers are published in ApJ or in MNRAS. Big things go in Nature/Science + then there are other smaller journals.
(At least this is what I tend to find. This may just reflect my biases - US based, in the cosmology field.)
But just because something gets published doesn't mean it is right :) I think non-scientists don't know what "peer review" actually entails! First, as this was a MOND paper, it could well have been reviewed by someone who favours MOND. Second, even if the reviewer doesn't favour MOND, if the steps taken and the arguments given seem reasonable, I expect they would suggest it should be published.
I have no issue with the paper being published. It is important that theories have advocates who put forward the best argument for them. What I do take issue with is the skewed presentation in popular science. What sells is exciting and new, not slow and steady. And 99% of science is slow and steady.
I only skimmed the paper past the abstract, but for what it is worth.
The MOND favouring group is a fringe of the cosmology community. The vast majority feel that there is enough evidence to rule it out.
I mostly mention this because I don't like that fact that popular science magazines (or at least their content that I see posted here!) has a bias towards "new and possibly exciting" or "controversial" research. Which I understand - revolution is more interesting that "physicist reduces error bars by 50%. Big picture unchanged". But, if all you read is these articles, you will get a very skewed idea of what the consensus is.
Just so there is no confusion, modified gravity as an explanation for Dark Energy is very possible, as an explanation for Dark Matter, the consensus is that it is ruled out.
Edit: I clarified my point below but will do it here too so everyone see it. I don't have a problem with this paper, I'm glad people are writing papers with alternate explanations to the consensus, that is how science is done. But, your conclusion from reading this article shouldn't be, "ahh damn, I guess MOND is right and LCDM is wrong" and I think that is how pop-sci articles tend to frame these things.
I'm not suggesting that at all! I'm just saying it took 30 odd years to take off/break out of the nuclear physics world.
Though rereading my first comment that wasn't entirely clear...
Kringing is the same thing as gaussian process regression, and astronomers use GPs a fair bit. I'm not sure whether they are used more widely.
My favourite forgotten/isolated statistical method is MCMC. These were first used by nuclear physicists at Los Alamos in the 40s/50s, but weren't really recognized more widely until the 80s. This is probably partly because only people working on bombs had access to the computing power before then, but still.
I didn't know about that, that's really cool! Thanks for letting me know.
I don't think that is right, though I'm not a GW expert so please tell me if I'm wrong.
I think they know the difference just by looking at the mass. i.e. we think it is hard to form black holes smaller than ~3.3ish solar masses and we don't think neutron stars can be more massive than 2.2 solar masses.
This is why we get articles like [1] where there is an issue when we think we've found something between those numbers.
And yes, the foolproof way of checking whether a NS was involved is to follow up with telescopes. But the constraints on position from GW aren't always good and so you can't always find it.
[1] https://www.sciencemag.org/news/2020/06/gravitational-waves-...
That's an interesting question. I've never thought about it or seen that taken into account (its not something we fit to when constraining cosmology)
I just looked around and LIGO has constrained Omega_GW < 1e-7 (they find no evidence, but that is the limit of their sensitivity). This is at least an order of magnitude smaller than Omega_Radiation and so will have a negligible effect on cosmology/total energy content.
I highly recommend these books by Alan Hirshfeld [1][2]. The first is about the discovery of parallax, and the instrument that was used to do this in 1838 was built by Fraunhofer. The precision of the instrument required to do this is incredible - the movement of a star due to parallax is, over the course of a year, about a fifth of the angular size of jupiter's great red spot. Unless you've been lucky enough to look through a fairly big telescope, you've probably never seen the great red spot and two centuries ago Bessel (using Franhofer's instrument) was able to detect a change in position of a small fraction of this.
The second book is about the technological developments of ~1850-1920 (cameras, spectroscopes, big telescopes) and is also really interesting. Fraunhofer also makes an appearance here - he was one of the first people to measure the absorption spectrum of the sun. These absorption features are still called Fraunhofer lines [3]. Later Kirchhoff (of the circuit laws) and Bunsen (of the burner) worked out what caused these.
The personal histories of all these people are also really cool, and these books discuss the people as much as the science/tech. It's amazing for how many people a small bit of luck was needed (or in Fraunhofer's case with the house collapsing -> meeting the elector, a big bit of luck)!
[1] https://www.amazon.com/Parallax-Measure-Alan-W-Hirshfeld/dp/... [2] https://www.amazon.com/Starlight-Detectives-Astronomers-Ecce... [3] https://en.wikipedia.org/wiki/Fraunhofer_lines
We have!
A constraint on dark matter is that it needs to be around at very early times. Before the time of the CMB (400 000 years after the Big Bang). If there wasn't dark matter at this time, the under/overdensities of baryons alone are not large enough to produce the large scale structure (galaxy groups/clusters) that we see today.
What this means is that black holes, formed in the conventional process (stars dying), cannot be dark matter. The first stars only formed much later. However, primordial black holes (formed at very early times, before the CMB) were still a possibility.
This possibility has mostly been ruled out though. The main way we have done that is through microlensing. If there were lots of reasonably sized black holes floating around, they would magnify background stars as the passed in front of them. It's a pretty cool effect. Here's a nature paper from a couple of years ago that investigated it [1]. The abstract is very readable and figure 5 shows how people have been slowly ruling out black holes as a major component of DM.
While yes, better launch capability == cheaper to put telescopes in orbit == good for astronomy, its not that simple.
Interferometry is HARD. The Keck telescopes which sit 100m apart on the surface of the earth were designed to be used as an interferometer and never lived up to the expectations. The interferometry abilities were shut down about a decade ago. Here's an article which includes a quote from one of the people who designed it who talks about spending 100s of nights trying to get this to work [1]. This only works because these telescopes are physically connected (see the discussion of the VLT in [2]). You are cavalierly talking about getting this working with 1000s of telescopes in space. If you are wondering how we got the event horizon interferometer if it is as hard as I am claiming, things become much easier at long wavelengths [2]. That's why we have lots of radio (ALMA, SKA) interferometers and almost no optical.
At ~1.2Eur for the ELT, that would imply $45m per "hubble".
One of these is on a mountain, the other is in space. One you can plug into the mains, the other you need batteries and solar panels etc. One you know how it is oriented (its on the earth) the other you need gyroscopes and control systems and etc. One you can plug an ethernet cable into to get the data, the other you need some sort of transmitting receiving system. One I can go fix with a spanner if something goes wrong, the other costs another X if it does.
Seems doable
If you ignore all the complexity of being in space, all the complexity of working with an array of telescopes, the fact that interferometry is way harder in optical, yeah it sounds great!
[1] https://skyandtelescope.org/astronomy-news/closure-looms-for... [2] https://www.eso.org/public/usa/teles-instr/technology/interf...
The argument is actually really really simple. If it started from within the solar system, it would not (barring some three body interaction) have enough energy to escape. But, we know from observations that it does have enough kinetic energy to escape. There are two options,
1. It may have stolen some energy from another object (an accidental gravity assist), but we know that it didn't come close enough to anything big on its way though.
2. It came from outside the solar system and so entered with some velocity, and therefore will leave with roughly the same velocity it came in with.
We have well known mechanisms to eject objects from star systems so it isn't crazy to have things passing through. No-one questions it because high school level physics is enough to show why it is the only reasonable explanation.
I read a lot of astronomy journalism that I think is absolutely awful (one of the main reasons I made this account was to point out garbage articles/comments). I actually think that this is pretty good.
It mostly follows the paper (here's what looks like an earlier version of this paper that is easily available [1]), doesn't engage in too much hyperbole, isn't publicizing something widely outside of the consensus.
Disclaimer that I don't work on planetary formation/dynamics so am not an expert on this, but still, pretty impressed.
[1] https://www.hou.usra.edu/meetings/apophis2020/pdf/2018.pdf
Famously, Hubble's 1931 paper that detected the expansion of the universe found H0 ~ 500 km/s/Mpc (fig 5 in [1]). The distances that he was using were way off...
Through most of the late 1900s the uncertainty was between 50 and 100 km/s/Mpc.
Now we know it at least as well as most other things, but this history of uncertainty means it is treated differently. Most annoyingly, simulations often work in units of Distance/h (where h = H0/100). This causes anyone who uses them incredible annoyance as you need to get your factors of little h right. Someone even wrote a paper called "Damn you little h" [2]. It's a total pain...
[1] http://articles.adsabs.harvard.edu/pdf/1931ApJ....74...43H
That sounds a little like parameter fitting.
You're exactly right. We have a model (Lambda CDM + GR + a few other details). We have ways to generate descriptions of how the universe would look given certain parameters (H0=70, Omega_Lambda=0.73, etc, etc) and we basically just see what range of parameters gives a universe that looks like (quantified using some statistics) the one we see through our telescopes.
But this is just phenomenology. The next step is working out the physics. For example, let's say we know there is X amount of something that looks like a cosmological constant - but what is that. This is what e.g. the search the dark matter particle is about - we know there is something that is cold + collisionless but what particle is it.
So then you argue that Λ isn't constant. Maybe it depends on time. Or on distance, which I guess just makes it a polynomial. Something like that?
Yup, I'm pretty sure the only models we have tested are wCMD which allows w (the equation of state of DE) to be something other than -1 (which is what the cosmological constant is). There is also w(a) which parameterizes the equation of state of dark energy as a linear function of scale factor (just think of it as time, a=1 now a=0 at the big bang). So linear rather than constant. We haven't gone to higher order than that.
The downside to adding parameters though is that, while you can always fit your data better (or at least as well) with more parameters,
1: Your error bars often blow up
2: Getting from phenomenology to physics might become hard. There are some models people have proposed that might allow us to fit the data, but then you need to explain why w changed in a very particular way at a very particular time. Basically it starts to look a little like overfitting.
Is it simplistic to conclude that it's this? Dark energy, the model presumes, takes the form of a cosmological constant Λ ...
Yup quite possibly, and there are people investigating it! The extended model is "Time dependent dark energy" See [1] or many papers [2]
But what other discrepancies would that create?
This is kinda the crux - modifying something to fix the current problems causes other problems. An example of this is the proposal that DE is just a result of us having the wrong model for gravity (GR) and that gravity is different at cosmological distances (note that this is not MOND which was proposed to not require dark matter and pretty universally unfavoured). However, gravity is really really well measured at solar system distances so you somehow need a theory of gravity that looks a lot like GR at small ranges and quite different at long ones, and that it hard.
[1] https://www.forbes.com/sites/startswithabang/2017/05/30/is-t... [2] https://ui.adsabs.harvard.edu/search/q=title%3A%22time%20dep...
I'm 99% sure this isn't true. Can you point to a single paper that mentions it?
Edit: actually I'm 100% sure this isn't true. See for example the all sky map from planck (http://www.bbc.co.uk/news/special/2013/newsspec_5106/img/pla...). And a paper discussing how they will measure the CMB dipole using planck https://ui.adsabs.harvard.edu/abs/2002A%26A...393..359P/abst...