back to article Stars say relativity still works

The Special Theory of Relativity may be under re-evaluation following CERN’s astonishing neutrino observations, but over in the world of astronomy, general relativity has had another reconfirmation from the Neils Bohr Institute at the University of Copenhagen. Radek Wojtak, Steen Hansen and Jens Hjorth have published in Nature …


This topic is closed for new posts.
  1. sayhi2yourmom4me

    You know, my best guess is that so much is known about galaxies, you can probably tell the mass to within a couple percent just by observing one. I wonder if they included dark matter in their calculations. Probably did.

  2. TheElder


    Some editing is required. The galaxies in question are a good deal farther than 8000 light years from us. That's just a stroll in the park. The 8000 number in the original article refers to the number of galaxy *clusters* that were tabulated and averaged to detect the influence of gravity on the light emitted by the clusters themselves.

  3. TheElder

    "El Reg would like to know how that’s accomplished, commenters"

    Count the galaxies in the cluster. Based on type of each we have a fair idea of how many stars any particular type contains. M31, for instance, has about 1 trillion stars (10^12) as determined by the Spitzer telescope. Since we have a very good idea how distant it is and how bright it is we can extrapolate what other similar looking galaxies will mass even when much further away. The same applies for other types (shapes and sizes) of galaxies that we can observe relatively close by for calibration.

    With this sort of information we can estimate the total mass fairly closely, maybe give or take 50%. Of course, there is still the question of "dark matter" and "dark energy".

    1. GitMeMyShootinIrons


      What's the weight in Elephants/Bulgarian funbags/*insert appropriate measure*?

      1. Thomas 4

        Easily measurable

        To the kitchen scales!

      2. hplasm Silver badge

        Er, -

        the weight is zero, in all units...

        1. wheel

          Gravity makes weight, mass makes gravity

          No, as there is mass within the clusters, there is obviously gravity. The gravity from any one massive body accelerates all other massive bodies, leading to weight. Consequently everything with mass has weight, the weight being affected by the gravitational frame of reference.

          So whatever the answer is, it is not zero. ;-)

    2. AndrueC Silver badge

      >we can estimate the total mass fairly closely, maybe give or take 50%

      If your accounts department only paid your salary 'give or take 50%' would you think that was 'fairly close'?

      1. G C M Roberts

        No, but in some areas of physics, 50% can be a great margin of error, some things have margins of error which are orders of magnitude.

      2. Chris Harden


        Depends on if they were giving or taking :)

        1. Naughtyhorse

          Takes me back to physics O level, part of the practical exam was estimating quantities - with a margin of error of 100%

          estimate everything to be 0 for a pass.


        2. The First Dave

          "No, but in some areas of physics, 50% can be a great margin of error, some things have margins of error which are orders of magnitude."

          Thing is, the difference between the measured speed of those neutrinos and the speed of light is a tiny percentage, so astronomical measurements are never going to come close to proving this one way or the other.

          1. Chemist

            "measured speed of those neutrinos"

            This has nothing to do with speed of light or neutrinos - it's another test of General Relativity

      3. mr.K


        If my payment were astronomical, yes I would.

      4. Naughtyhorse
        Thumb Up

        if it was give, id be happy enough

      5. Tom 13

        In astro, 50% is damn good.

        For most things, you're happy just to get in the right order of magnitude. One of the reasons they work so hard to add more digits on the end of an AU is so they can trim the error margins when you have to multiply it by the billion billions that is your baseline to the next star.

      6. puffspluslotion

        Irrelevant, 'give or take 50%' is good enough for things we know a lot more about like meteorology and investment economics so such a number on theoretical physics is indeed pretty close.

  4. Jolyon Smith

    How they did it...

    First, they observed the amount of red-shift. They then used this to calculate the mass. They then used this calculated mass to predict the degree of red-shift that should occur. They then observed the amount of red-shift and... oh.. wait... um ......

    (in related ponderings, use of trigonometry and solar parallax to calculate the distance of observable objects falls down at some quite close distance to the Earth - I haven't yet found a satisfactory explanation of how we can then know the distance of something that is further than that threshold distance... anyone ?)

    1. sayhi2yourmom4me

      You can tell distant stars by how dim they appear. But don't different size stars put out different amounts of light? Yes, but they also put out different wavelengths of light too. Biggers stars put out more light, but they are a lot hotter and throw off shorter wavelengths. Don't worry about the red giants. Anywy, you can tell how much light they should be putting out by looking at the wavelength. Then by compairing that to the amount of light you actually see, you can tell the distance. At least, that's how I think its done.

    2. Dalen

      I am not an astronomer, but if I remember correctly, the "standard candle" method is used. This compares the luminosity of known stellar objects at various distances.

      Speaking of which, what are the El Reg official units for stellar luminosity?

      1. BristolBachelor Gold badge

        Standard units of stars

        I think probably something like:

        Brown dwarf; very dim. Units like 1 ashton kutcher or 3 Paris Hitons.

        Um that's as far as I can take this. I'm a nerd and like Sheldon Cooper, I fall down on knowedge of film stars and popular culture.

        The fail is for me.

        1. Anonymous Coward
          Anonymous Coward

          what are the El Reg official units for stellar luminosity?

          Measured in Milkman-eye-glints assuming a standard British 30 year old housewife in a baby-doll nightie circa 1972.

        2. Alfie

          Presumably going all the way up the scale to supernova == 1 Brangelina


        3. Chris Harden

          Brown Dwarf = 1 Kutcher

          Really bright, hot star (guess my knowlage is invert of yours - fail for me) = 1 Vorderman?

      2. Yag

        Speaking of which, what are the El Reg official units for stellar luminosity?

        I would suggest the "Nuke", but this unit would be a bit too large : The sun would be a bit less than one milliNuke according to some american and japanese observers in 1945.

    3. Anonymous Coward
      Anonymous Coward

      Measure distance

      It's quite simple to work out the distances really.

      1. The universe is expanding in all directions.

      2. Redshift correlates to speed differential (it's the doppler effect)

      Therefore, objects further away will be going away faster and will thus have more redshift.

      Calibrate by measuring a few nearby stars (which can be triangulated) and you're good to go.

      Redshift is calculated by knowing that specific elements in a star will absorb light at very specific wavelengths, which then tells you both the chemical makeup and the size.

      All quite clever really!

      1. Chris Miller

        Not quite (AC 06:47)

        You have to establish the value of the Hubble constant, which you can't do by measuring stars in the Milky Way using parallax (which can only be done out to ~100ly anyway) - their motion is determined by individual factors and the rotation of the galaxy - or even the closest (few million ly) galaxies - Andromeda is blue-shifted, moving towards us ready for collision in a few billion years.

        For nearby galaxies (~100 million ly) we can use Cepheid variables, which are very bright variable stars whose absolute luminosity is directly related to their period - the astrophysics of this is well understood. For more distant galaxies, we can use a specific type of Supernova (1a), which is the result of a white dwarf accumulating hydrogen from a neighbouring star until a critical mass is achieved and they should therefore all have the same absolute luminosity. We can identify type 1a by the decay spectrum (the current Whirlpool galaxy supernova is of this type), but supernova physics isn't as well understood, so opinions of how bright these events actually are may vary.

        The Hubble constant determines the age of the universe (1/H with some fudging for the way H is now believed to have changed over time), and the current value of close to 70 kilometers per second/Megaparsec fits the results of the cosmic microwave background quite well (phew!)

      2. John Mangan
        Thumb Down

        One slight problem with that . .

        is that you have some circular logic there. The redshift gives you the distance if you know the Hubble constant. But you need to know the distance of some (non-triangulated galaxies to calculate the Hubble constant.

    4. G C M Roberts

      There is a similar problem in particle physics, from what I remember from Uni 11 yrs ago, whereby detectors measure energy and thus the velocity and mass have to be intelligently guessed.

    5. John Mangan

      If I remember correctly . . .

      the parallax method gives you a 'local' population of satrs with directly measured sitances. This allow you to calculate absolute magnitudes (how bright the *really* are) for a range of star types. Within this group are a class of stars known as Cepheid variables whose brightness changes with a period dependent on their size. These standard candles then allow you to calculate distances for more distant stars and 'local' galaxies. From the population of galaxies you get a scale of mass/brightness which can then be used to calculate equivalent values for more distant galaxies where you can't pick out the standard candles.

      If I remember correctly . . . .

    6. Tom 13

      The parallax calculations depend on the baseline,

      so it depends on how you take your baseline. The use of photographic plates and keeping precise records on time, location, and orientation of the telescope has allowed us to extend the baseline a fair piece. Not only can we get the AU as a baseline, but because we can approximate the sun's orbit around the center of the galaxy, we can extend the baseline beyond an AU.

      That does still limit to relatively close stars, but from there you can start to work with relative/absolute magnitude to determine the distance to the star. The initial work there was done with binary pairs where you have an independent means of determining mass. But once that is established you work more with red-shifting. It all keeps feeding back on itself, but you get further and further away and become more and more comfortable that you assumptions and calculations are correct.

  5. TheElder

    "I haven't yet found a satisfactory explanation of how we can then know the distance of something that is further than that threshold distance... anyone ?)"

    The overall redshift is directly correlated to the distance. What was measured in the above article is differences in the redshift from center to edges. Those differences are very much smaller than the average redshift of the cluster. It is the average redshift of each of the galaxies being similar that identifies the members of the cluster.

    I must get back to my astrophotography. It's the first good night for same in quite a long while.

    1. Steve Ives

      Not done any astraphotography, but I did once get a good picture of a Vectra...

      1. Your Retarded

        A good picture of a Vectra?

        Are you sure...?!

  6. Sander van der Wal

    You can estimate the mass of a cluster by looking at the velocities of the galaxies and the gas clouds in that cluster. These velocities also result in a shift of all spectral lines, to the blue when they move towards us and to the red when they move from us.

    1. Tom 13

      Except that all spectra are red-shifted.

      This is one of the observations which leads to the hyper-exapnding universe hypothesis. So you first have to get the average red shift, and then look at the relative red shifts to find what is blue-shifted compared to the average.

  7. Michael Dunn

    IIRC, didn't the brightness of Cepheid variables provide some measure of the distances of galaxies containing them? Personally I find sky photographs rather confusing: individual stars, galaxies, clusters of galaxies all just projected onto a 2-d surface. Who is to say that a particular object in the photo isn't just a smudge on the telescope lens/mirror?

    Need a good warm coat for astrophotography!

  8. Full Mental Jacket

    No mention of

    No mention of red shift due to absorbtion/retransmission through more galactic dust. Was that factored in?

    1. Allen Versfeld

      Reply to Full Mental Jacket

      Reddening due to absorption by galactic dust is not the same as redshift. Reddening is a filtering effect - blue light is more likely to scatter, red light less so. More of the original red light reaches us than of the original blue light. In this case, information is lost because light of specific wavelengths never reaches us.

      But redshift changes the colour of the light, by lengthening its wavelength. The original information is not lost, merely shifted down the spectrum.

    2. Ru

      Probably not

      Career astrophysicists are well known to be an ignorant, lazy, workshy bunch who only went into academia because the real world seemed too stressful. They probably don't know a whole lot about astrophysics, and they probably don't know anyone who does and even if they did they wouldn't ask them.

      They probably just got drunk, scrawled down a bunch of random numbers in vomit and posted it off to the nearest journal, without even asking the omniscient polymath denizens of the commentardia to do any proofreading.

  9. Eponymous Cowherd
    Thumb Up

    I really hope CERN's observations are correct.

    Largely for the spectacle of watching Prof Al-Khalili eat his own boxers.!/jimalkhalili/status/117160630527594496

    1. Primus Secundus Tertius Silver badge

      Boxer shorts

      He may organise a pair made of rice paper, worn outside an ordinary set of y-fronts.

  10. Geoff May

    Calculating cluster weight

    The weight of the individual Galaxies in the cluster is known as it's printed on the side of each bar.

  11. taggedcattle


    Niels Bohr.

  12. Andus McCoatover

    Only way to sort this argument.. to shoot yourself. Then, ask God directly. Sorry if I can't bring an answer sharpish...

    1. Anonymous Coward
      Anonymous Coward

      can i suggest then....

      ..that you try to encourage as many religious types as possible to try and find the answer for us..


  13. David Pollard

    The Virial Theorem

    Initial estimates of galactic mass are based on the Virial Theorem.

    Assuming that the only force in play is gravity and assuming that stellar velocities have more or less evened out ('virialised') from the galaxy's initial formation, then the Virial Theorem shows that the total kinetic energy is equal to half the potential energy. Total kinetic energy is proportional to total mass. Total potential energy (in a given configuration) varies as square of mass. So if the relative motion of the stars about the galaxy centre of mass is known - and information about their velocities is available from spectroscopic observations - by making reasonable assumptions about their distribution within the galaxy it is possible to estimate the total mass.

    Unfortunately, observations of the variation of stellar velocities with distance from the centre are not in accord with this simple model. The prediction is that outside the main concentration of stars, velocities should fall off with the inverse square of distance. In fact, velocities deduced from spectographic observations appear pretty much constant to a considerable distance from the centre. Equally puzzling is the observations that stellar (linear) velocities are in the range of 150 to 350 km/S irrespective of large differences in the size of the galaxy they inhabit.

    It is the discrepancy between the 'Keplerian model' (stars in a galaxy behaving in a similar way to planets around a star under the influence of inverse square law gravity only) and the observation of more or less constant velocity which gave rise to the idea of dark matter. This dark matter is assumed, by some miracle or other, to sit in the shape of a halo which is in precisely ihe right place to cause the observed pattern of stellar velocities.

    A current rationalisation of the dilemma which some cosmologists favour comes from MOND - Modified Newtonian Dynamics. Here it is assumed that the inverse square law does not hold at very large distances.

    For reasons which largely escape me, the alternative explanation proposed half a century ago by Hannes Alfvèn that electromagnetic processes are in play is generally treated as heresy.

    1. David Pollard

      Viral Theorem - Crekshun

      "... outside the main concentration of stars, velocities should fall off with the inverse square of distance ..."

      replace with

      ... velocities should fall off with the inverse square root of distance ...

    2. John Brookes

      Alfven was an electrical engineer and (part-time) plasma physicist....

      He proposed electromagnetic processes as the solution to virtually every problem he looked at.....

      This partly explains why the 'physics establishment' were a little sniffy with him.

      1. Andus McCoatover

        Abraham Maslow...

        "When the only tool you have is a hammer, every problem begins to resemble a nail"

      2. Tom 13

        No, John was correct.

        One of the few things I can trust Wiki for:

  14. Mephistro Silver badge

    "CERN silicon questioned"

    Now if someone could explain to me clearly and slowly how this affects the results of the neutrino studies at CERN... I mean, the CERN thing makes reference to neutrinos, not photons, and they are quite different beasts.

  15. sisk Silver badge

    Ok, I may be way off base with this, but isn't relativity well known to break down at the quantum level? And if so wouldn't a quantum particle like nutrinos breaking light speed be not so much proof that Einstein was wrong as another bit of quantum wierdness to try to wrap our minds around? That's been bugging me ever since CERN first came out with this.

  16. TheElder

    I will point out that the CERN results may require modification of Einstein's Special Theory of Relativity whereas the results of the study of gravitational "drag" on photons is a matter of General Relativity.

    There are a number of distant possibilities other than experimental error that might explain the CERN results. Quantum effects could conceivably play a role if some sort of mechanism similar to "tunneling" is taking place. The problem with that is the possibility of a neutrino interacting twice or more times with matter over such a short distance is vanishingly small. A neutrino can travel through light years of lead before it has an even chance of interaction. Also, since any sort of quantum interaction will be ruled by probability it should show a characteristic curve superposed on the emission curve of the burst.

    The CERN experiment depends on matching the emission burst curve to the detector burst curve so anything that alters that curve along the path will upset the correlation. The curves appear similar so it is unlikely that any interactions took place.

    Re the distance to the galaxy clusters, they are too distant to use the cephid variable star "standard candle" method. They are also too distant to make use of stellar orbital velocity methods of mass determination. As well, the orbital velocity method is confounded by the issue of dark matter. The Hubble constant (red shift) is most easily applied in this case to give approximate distance to the clusters. That, combined with the the apparent luminosity for particular types of galaxies, serves to give galactic cluster mass well enough for the purposes of this experimental determination.

  17. The last doughnut
    Thumb Up

    Thank you

    Some actually informative comments there.

    Well done chaps.

  18. Anonymous Coward
    Anonymous Coward

    Missing a few zeros?

    Interesting article, but I'm having a hard time understanding how a "distant galactic cluster" can be only 8,000 light years away. That's actually seems pretty close (like, Eta Carinae close), relatively speaking . . . .

  19. mhenriday

    Kudos to Richard Chirgwin

    for his article - and, for that matter, the one on possible general relativistic explanations for the results of the OPERA measurements - and to many knowledgeable commentators here for good clarifications. Hope the Reg will vouchsafe us more of this sort of thing !...


This topic is closed for new posts.

Biting the hand that feeds IT © 1998–2019