Follow Slashdot stories on Twitter


Forgot your password?
Space Science

Measuring the Hubble Constant Better 102

eldavojohn writes "The Hubble Constant is used for many things in astrophysics: from determining how fast things are moving away from us, to the total volume of the universe, to predicting how our universe will end. The current best value for the Hubble Constant is 74.2 ± 3.6 (km/s)/Mpc according to recent conventional methods and the recently restored Hubble Telescope. Most astronomers agree that that's within 10% of its actual value. Researchers now claim that they might be able to get to 3% using water molecules in galactic disks to act as masers that amplify radio waves, to analyze galaxies seven times as far away as the current measurements. The further away the 'standard candle' is, the more assured they can be that local effects are not skewing the measurements. From one of the researchers: 'We measured a direct, geometric distance to the galaxy, independent of the complications and assumptions inherent in other techniques. The measurement highlights a valuable method that can be used to determine the local expansion rate of the universe, which is essential in our quest to find the nature of dark energy.' Once the Square Kilometer Array is completed, they hope to get even closer to the actual value."
This discussion has been archived. No new comments can be posted.

Measuring the Hubble Constant Better

Comments Filter:
  • From what I know, it's been discovered in the past decade or so to not be a constant. The expansion of the universe is accelerating. This is a minor nitpick, I know. :-)

    • by American Expat ( 1393429 ) on Tuesday June 09, 2009 @03:11PM (#28269959)
      It's just obeying the first rule of computer science: Constants aren't
      (second rule: Variables won't)
      • Didn't they already figure out how they screwed up creating the Hubble telescopes main mirror? Or is the mirror warping now?

        And didn't they just do the last fix on it just last month?

    • Re: (Score:1, Informative)

      Doesn't this constant place an additional limit on the size of the universe (or at least the part of the universe we're ever going to see) ?

      c / 74.2 km/s * Mpc = 300000 / 74.2 * 3 261 636.26 lightyear (1 Mpc = 3 261 636.26) or about 1.31872075 Ã-- 10^10 lightyear, about 13 billion lightyear.

      Because at that distance, the stars would be moving away from us at light speed, so in reality there's an event horizon between us and stars at that distance. Light from stars further away would never reach us, due

      • Universe expansion will create causal separation in the future, but not the past. It doesn't limit how far away you can see something, because you are looking at something in the past, but it does prevent you from going there. Because looking backward in time, the universe is shrinking, and you can see more and more of the universe going back. Looking forward in time, everything is getting more separated, and, for far regions of space, the rate of separation is higher than light can catch up to.

        In the scena

        • Re: (Score:2, Funny)

          by BizzyM ( 996195 )

          Universe expansion will create causal separation in the future, but not the past. It doesn't limit how far away you can see something, because you are looking at something in the past, but it does prevent you from going there. Because looking backward in time, the universe is shrinking, and you can see more and more of the universe going back. Looking forward in time, everything is getting more separated, and, for far regions of space, the rate of separation is higher than light can catch up to.

          You just blew my mind

    • Re: (Score:2, Interesting)

      by cheftw ( 996831 )

      And how come it's measured in some stupid space unit? It's a frequency so it wants hertz! []

      It's called SI. Get with the program dudes.

      • Re: (Score:1, Informative)

        by Anonymous Coward

        And how come it's measured in some stupid space unit? It's a frequency so it wants hertz!

        The Hubble constant tells you the speed that astronomic objects move away from us (or from any point in the universe, cf. Galilei invariance) depending on how far it already is, hence (km/s)/Mpc. An object at the distance of 1 Mpc moves away at approximately 74 km/s.

        Now what exactly does the value in Hz tell you? Nothing.

      • It's measured in "stupid space units" because it's used in astronomy, of course. "Kilometers per second per megaparsec" has a much more intuitive interpretation, when considering the speed at which distant galaxies are moving away from us, than does "cycles per second."

      • Hubble's constant is a measure of how the expansion velocity (in units of kilometers per second) of the universe changes with an object's distance from us (in units of megaparsecs), yielding units of km/s/Mpc. It is improper to simply cancel the distance units which would leave you with units of frequency (i.e., inverse seconds).

        Another example in astronomy is the unit given to monochromatic flux, which typically has units of Joules/meter^2/second/Hertz. Note that the unit has both seconds and Hertz.
        • by cheftw ( 996831 )

          I have no wish to start an internet argument, nor is astrophysics my department, but I would bet you a shiny penny that if you did cancel them nothing bad would happen. Life would go on and all your calculations would be correct. (Assuming you got them right in the first place).

          Counterexamples welcome.

          • Is something "bad" going to happen? No. Does it make interpretation of the quantity more confusing? Absolutely.

          • Think of Gauss's formula: e to the i x pi power +1 = 0.
            This usually gets taught as part of second semester calculus or so. It gives some students headaches, because it emphasizes so strongly how raising a number to a power isn't really best understood as self multiplication once we get beyond the integers (It's fairly simple to see e to the 4th as e x e x e x e, but harder to imagine what e to the i or pi power involves). Just doing one of the simplest possible operations to the equation, making it read: e

            • by cheftw ( 996831 )

              I think you need haskell to determine e that way :)

              But where would you differentiate a constant? Like 2 is a constant, but d/dx(2^x)=(ln2)(2^x)!=0

    • by selven ( 1556643 )
      Technically, the only way the Hubble constant can be a constant is if the universe is expanding exponentially.
      • Just the opposite. "Hubble Constant" generally refers to Ho, which is the current rate of expansion.

        After the Big Bang, mutual gravitation of mass in the Universe slowed the rate such that expansion went at about t^(2/3) -- see [] for the scale factor (rate of change of expansion) in a matter-dominated universe.

        Einstein's General Relativity says that space contains energy -- called dark energy or vacuum energy, which has the effect of causing inflation. A
        • by selven ( 1556643 )
          I was talking about the km/sec/megaparsec value. If the rate of change (speed they're flying away from us, aka derivative) is directly proportional to the current value (the distance, if this was a graph it would be the y coordinate), it's an exponential function.
          • You are correct. Currently the scale factor, a(t), goes as e^kt. The Hubble parameter can be defined defined H(t) = (da/dt)/a.

            Therefore we take the derivative of e^kt and get k*e^kt. Substituting, H(t) = k * e^kt / e^kt = k.

            Thus H(t) is constant in an exponentially expanding universe.
    • Some of the inflationary models also suggest the universe is very, very much bigger than the part we could theoretically see, a factor of about 10e30 times or more. (We could theoretically look back close to the total age of the universe, and because of expansion, the total distance would actually be at least a bit more than 2x larger than that roughly 12 Billion years would seem to allow, say 26-30 Billion LY radius.).
      For those models where the total size of the universe is so

    • The expansion rate of the Universe at the present time is called the Hubble constant. Astronomers more generally refer to the Hubble parameter to describe the expansion rate as a function of cosmic time. We have known that the Hubble parameter changes with time for at least 60 years, but, as you note, we've only known that the rate is currently accelerating for the past decade.
  • by Alzheimers ( 467217 ) on Tuesday June 09, 2009 @03:03PM (#28269865)

    Yep, he's still dead.

    • Re: (Score:3, Funny)

      by Red Flayer ( 890720 )
      So is the value 0 or 1?

      I'd assume the dead state is 0, and the live state is 1 -- except Hubble was living while he calculated the value, so he may have assigned 0 to the live state, and 1 to the dead state. Or he might have foreseen my current problem and switched the values just to trick me.

      Speaking of which (my current problem), it appears my doomsday machine has entered into a positive feedback loop, and I'll only know how to fix it and save the planet if I have the correct value. I'd appreciate a
    • by Chris Burke ( 6130 ) on Tuesday June 09, 2009 @03:45PM (#28270523) Homepage

      Yep, he's still dead.

      But that measurement is only accurate to within 10%.

    • 'E's not dead! 'E's pinin' for the nebulas!

  • How can something of infinite size have a volume?
    • Re: (Score:2, Insightful)

      by Yokaze ( 70883 )

      AFAIK, the universe is not infinite in size, it is just infinite. The very same way a circle is infinite, but has a length, or a ball or torus a surface.

      • That makes no sense. Thank you for playing. Maybe you mean a circle has an infinite number of points?

        • by vlm ( 69642 )

          Maybe you mean a circle has an infinite number of points?

          Getting closer ... infinite number of tangents at a constant radius from one central point.

          Infinite number of points is just any ole line or squiggle.

        • Finite means limited; infinite means unlimited.

          Now, take a circle. Put a pencil tip on one point and start moving it along the circumference (clockwise or counterclockwise, it doesn't matter). Now, get back to me when you've reached the end and can't go any further. You can continue infinitely far along a curve of finite length.

          The ideas I've seen about an infinite but not infinitely-sized Universe tend to be more complicated, but it's the same general principle.

      • Who modded "a circle is infinite" as insightful? A circle with a finite radius has a finite area, only a circle with infinite radius has infinite area. As the other responses to this say a circle does have some qualities which are infinite, but that doesn't make it infinite (anymore than 1 is infinite because it belongs to the natural numbers which are infinite is a good argument).

      • I know what you're saying, but I think you're just confusing it.

        The standard way to say this is that the universe is "finite but unbounded," in the same way that the [i]surface[/i] of a sphere is.

    • Whoever said the universe had an infinite size?
      • by rm999 ( 775449 )

        The size of the Universe is entirely an unknown. As such, scientists don't talk about it much.

        • There are some WMAP data in which the low quadrupole moment of the CMB patterns suggests that the universe might, in fact, be finite. But it's REALLY iffy in my opinion, and other than that, there's no evidence (AFAIK) for a finite universe, despite ubiquitous claim.

        • by dmartin ( 235398 )

          The size of the Universe is entirely an unknown. As such, scientists don't talk about it much.

          Not entirely unknown. We have some pretty good lower bounds on what it can be =).

    • When people talk about whether the universe is infinite or not, they are referring to the whole thing.

      When people talk about the volume, they are referring to the observable universe. The observable universe is about 93 light years across

    • by djp928 ( 516044 )

      It isn't infinite in size. The size is approaching infinity, though. The universe is finite, but unbounded--meaning it is finite in volume at any given time, but is constantly increasing in size as space expands.

      We may never know exactly "how big" the universe really is, since we are effectively cut off from whatever is beyond the edge of the observable universe. Anything that might be beyond that is expanding "away" from us faster than light--so we can never see it from here, and can likely never go the

    • by JustinOpinion ( 1246824 ) on Tuesday June 09, 2009 @04:46PM (#28271329)
      We have to be more careful with what we mean by 'size' and 'volume' and such.

      The observable universe [] is the region of space we can see. The universe has a finite age, so there is a finite distance over which we can see. Any further than that, and light literally hasn't had enough time to reach us. So there is indeed a boundary beyond which we cannot observe. This boundary recedes as time goes on. The universe is ~13.5 billion years old, but because the universe was expanding during all that time, the observable universe is bigger than just 13.5 billion light-years (see comoving distance [])... in fact it is 46.5 billion light-years in radius.

      Now there is every indication that the universe extends beyond the cosmological horizon. So as the universe ages, we see more and more of the full universe, which is much larger than our observation volume. So how big is the universe as a whole? Our best understanding right now is based on the curvature of spacetime []. If spacetime at large scales is curved, then the universe can loop back upon itself and thus the universe is finite. If spacetime is perfectly flat on cosmological scales, then in fact the universe as a whole is infinite in size.

      Our best measurements indicate the universe is flat, within error. Our best theories of the origin of the universe, coupled with available data, generically predict that the universe is infinite. So our current best answer is that the universe is infinite in size/volume. A strange result, perhaps, but that's our best understanding of the current data. Now there are indeed errors on our measurements, so our universe could be smaller. But the curvature is so small that it implies our universe contains at least [] 1000 Hubble volumes [] (the Hubble volume is the surrounding space beyond which nothing is accessible since matter is receding faster than light). Others have analyzed the night-sky looking for 'repeat patterns' that would be expected for smaller closed universes, and no such patterns have been found.

      So the observable universe is finite (but ever-expanding), and the full universe is considerably larger (infinite according to our current best data and theories).
      • Who says the universe has a finite age? The observable universe might have an age if defined in such a way as creation by some large scale event (aka big bang) What we know of the universe is just one "big bang event" out of a possible infinite number of others over an infinite amount of time. I think the problems begin when astrophysicists make the assumption that universe was created in the first place.
  • Once the Square Kilometer Array is completed

    The name sounds impressive, but how big will it be?

  • by Colonel Korn ( 1258968 ) on Tuesday June 09, 2009 @03:29PM (#28270271)

    1 - Distance measurements are currently kludged together very carefully using bridging. We use one measurement, for instance parallax based on the Earth's movement over 6 months, to show us the distance to a star that has some particular properties and which our models say should always be a certain luminosity. The parallax measurement has error bars.

    2- Then we find a much more distant star of that same type that is near a particular type of supernova, and measure its brightness, comparing that to the brightness of our first star to give the distance to the distant star, and thus the supernova as well. That has bigger error bars.

    3- Then we look for that type of supernova in very very distant galaxies. Supernovae are brighter than the rest of their galaxy put together while they're burning hot, so we can see them at tremendous distances. We use the measured brightness of that supernova to determine the distance to its galaxy.

    4- Then we pair the knowledge of its distance with its velocity with respect to us, which we can determine through redshifting of something with a familiar spectrum. More error bars. That becomes a single point for the determination of the Hubble Constant (and yes, the "constant" is changing).

    With only a cursory glance at TFA, it looks to me like this is a way to skip to step 3 or 4, thereby avoiding the need to bridge these length-scales using several techniques.

    • Re: (Score:3, Informative)

      by JohnFluxx ( 413620 )

      More details here: []

    • Yes it looks like they are measuring the distance by paralax using the fact that galaxies are in fact quite wide really. From the article it looks like this has been done before for galaxies close by which is not very usefull for measuing hubbles constant but that they have found way of amplifying the signal from distant galaxies. What the article doesn't say is how they measure both the linear and angular size of the gallaxy which is required to gauge the distance it just says that they did. If so then th
    • The trouble is that these galaxies aren't that far away (despite the article summary says). They're quite a bit further away than the previous measurements of water masers, but you still need to use Type Ia supernovae to actually get to the distances where this discussion gets interesting. The cool thing about the water masers is that they might allow us to get out a bit further without using another "rung" on the distance ladder, but there is no way that they are going to replace the (much, much more dista

  • Halton C. Arp, a professional astronomer was Edwin Hubble's assistant, says otherwise ... []

    • Maybe you should present a more credible source, if you wish to be taken seriously.
      • Edwin Hubble's assistant isn't credible enough?!

        lol !


        William A. Tiller, Materials Sciences Department, Stanford University wrote: "The present scientific establishment has grown somewhat fossilized by its current world picture and is locked into a view of reality that has outlived its usefulness. It has begun to limit mankind's growth and has so increased its sense of specialization, separateness, materiality, and mechanical computer-like functioning that it is in real danger of self-extermination."

        • The website is what I was questioning, really, although another reply hits on the fact that simply being an assistant to Hubble does not necessarily make a person credible. is a proponent of the "electric universe" "theory" -- which has been thoroughly rejected over and over, but remains a favourite of the conspiracy theorist type.
          • > is a proponent of the "electric universe" "theory" -- which has been thoroughly rejected over and over

            NASA begs to differ... []

            The TSS-1R electrodynamic tether experiment: Scientific and technological results

            N. H. Stonea, W. J. Raittb and K. H. Wright, Jr. c
            a Space Sciences Laboratory, NASA Marshall Space Flight Center, Huntsville, AL 35812, USA
            b Center for Atmospheric and Space Science, Utah State University, Logan, UT 84322, USA
            c Center for Space Plasma and Aeronomic Research, Univer

            • I am disappointed that this was the best you could come up with to support your theory. The existence of cosmic plasmas such as those encountered by the TSS-1R mission is quite widely accepted. Proving that they have the effects claimed by the theory you apparently advocate is quite another matter.

              I read the first chapter of the book on, and it mostly seems to be jeering at the complexity and unintuitive nature of current theories, while also heavily emphasizing the "unprovability" of an

              • Maybe you can explain why these PhD's are so gullible for "buying" into a "currently unacceptable theory":


                "I really love this book. It is causing me to rethink a great deal of my own work. I am convinced that The Electric Sky deserves the widest possible readership.... I felt genuine excitement while reading and felt I was delving into a delicious feast of new ideas."
                - Gerrit L. Verschuur, PhD, University of Manchester. A well-known radio astronomer and wri

  • Presumably "Erm, around 100" isn't good enough then?
  • When I was doing university physics with a slide rule, three significant figures ( 74.2 ± 3.6 (km/s)/Mpc) was good enough for anything. Is our next probe going to miss M31? Oh yeah, get off my lawn too. :-)
    • Re: (Score:3, Informative)

      by John Hasler ( 414242 )

      > When I was doing university physics with a slide rule, three significant figures ( 74.2
      > ± 3.6 (km/s)/Mpc) was good enough for anything.

      When I was doing university chemistry with a book of log tables four significant figures was barely good enough for my homework.

    • Indeed. This constant may not be a constant. It may not be the same everywhere in the universe. So by observing things very far away (distance and time), we may actually end up with a less accurate number for a `local' variable.

  • Bad Labrador (Score:2, Interesting)

    I've been following Alexander F Mayers work on Minkowski's (Einsteins Maths Teacher) space time mathematics which Einstein, who didn't understand them, called "superfluous erudition'. Mayer derives a model for the universe that does not require the universe to be expanding, let alone accelerating expansion, does not require "Dark matter" nor "Dark energy", that makes a damn sight more cosmological sense than the "Big Bang" and fits the current observations, much, much better, with no free variables like "qu
    • Well since Minkowski latter came to be a major contributor to the development of the theory of relativity he clearly decided that they weren't worth the effort either. Any model the does not require the universe to be expanding must really take some work to avoid the fact that it clearly is. The fact the dark matter has nothing to do with the universe's expansion well that is to say the evidence for dark matter does not rely on it. Proposing that dark matter does not exist would require rethinking not just
      • The only evidence for expansion is that the further away a galaxy is from us, the more red shifted it is. Mayer/Minkowskis equations show that the red shift may be an artifact of curved space time and not a radial recessionary velocity at all. His equations appear to fit the observed data much better and explain such things as the Pioneer anomaly much better. If Mayer is correct, the Hubble constant measurements are merely measuring the change of timelines due to the curvature of spacetime. Needless to sa
        • Re: (Score:2, Insightful)

          by anarchyboy ( 720565 )
          Why not? If he's theory is right surely that would be a good thing? interestingly I read the book you linked to and found it quite hard going, there was little explanation of the ideas presented and seemed to have many descriptive quotes from people like minkowski that were then interpruted, woryingly these 'sound bites' were offered as support of the theory presented.

          The multi-dimensional description of time was woefully under explained, probably due to a lack of a concise mathematical description but w
  • According to the article "the astronomers determined that the galaxy UGC 3789 is 160 million light-years from Earth". This translates to 49 Mpc. According to NED [], the velocity (in the Cosmic Microwave Background frame) is 3385 km/s.

    Therefore this measurement of the Hubble parameter is then 3385/49 = 69 km/s/Mpc.

    (Unfortunately the article does not quote an uncertainty on the 49 Mpc measurement. Because of peculiar velocities, I would estimate that there is at least a 300 km/s uncertainty on the 3385 km/s ve

    • It's also possible that this galaxy is not totally in the Hubble Flow. In other words, it might be pulled around by other nearby galaxies/galaxy clusters. All galaxies are affected by this to some extent, but with nearby galaxies (like this one), these gravitationally-caused velocities can be significant compared to the Hubble expansion-caused velocities.

  • The current best value for the Hubble Constant is 74.2 ± 3.6 (km/s)/Mpc according to recent conventional methods and the recently restored Hubble Telescope. Most astronomers agree that that's within 10% of its actual value.

    10% of 74 is 7.4, corresponding to ± 3.7; meaning that in the very worst case, where the true value is at one end of the interval, we can only get about 10% away. What the astronomers agree on is that the estimate of the uncertainty on the measurements is something like ± 3.6. This is not as trivial a matter as it would seem - it can be quite complex to calculate and is a source of many of the more embarrasing errors in science.

  • The Perimeter Institute recently gave a lecture on this, by Brian Schmidt, Australian National University - "The Universe From Beginning to End". I understand they will EVENTUALLY make these lectures available on their website, after they've made a bit of money by showing them on Discovery etc: []

"Yeah, but you're taking the universe out of context."