Dark Energy: No Longer a Surprise

By Sean Carroll | December 16, 2008 12:45 pm

A bit of science news: Alexey Vikhlinin and collaborators have used observations from the Chandra X-ray satellite to uncover new evidence for dark energy. (More info here, and the paper is here.) In particular, they simply count the number of galaxy clusters with various masses at various redshifts, and compare with the predictions of models with and without dark energy. If there were no dark energy, matter would keep clustering on larger and larger scales as the universe expanded, making new clusters all the way. But if dark energy eventually takes over, the creation of new clusters begins to turn off, as the dark energy provides an extra push of expansion beneath the feet of the particles that would like to cluster together, preventing them from doing so.

Just to guide the eye, here are plots of the number of clusters (vertical axis) as a function of their mass (horizontal axis) at two different redshift ranges — near is on top, far is at the bottom. The left plot, which fits the data, has an appreciable cosmological constant; the right one, which doesn’t, doesn’t. The graphs are a bit confusing, because dark energy affects not only the growth of structure, but also the relation ship between redshift and distance. But the point is that dark energy kills off cluster formation at late times.


You may ask the question: so? Didn’t we find dark energy ten years ago, and haven’t we confirmed its existence several times since? Yes, and yes. In a sense, this result doesn’t teach us anything we didn’t already know.

But we should resist the temptation to become too blase about the whole thing. (Notwithstanding that I’ve been guilty myself.) On the one hand, this is a new manifestation of dark energy: a dynamical effect on the evolution of matter, rather than simply a background effect on the expansion of the universe. This is of great interest to astronomers, and should help to constrain alternatives to the now-standard picture. But on the other, more important hand, it remains astonishing that we have this preposterous model that keeps fitting the data. We shouldn’t lose our sense of wonder that we’re able to understand as much of the universe as we do, or that the reality of cosmology is so much more interesting than simple theoretical models of the past would have predicted.

chandra-w-omegax.jpg Here is the graph from the paper showing limits on the equation-of-state parameter, w. Horizontal axis is the fraction of dark energy (about 75%, eventually I’ll have to stop using 70%), vertical axis is w (about -1, plus or minus 0.1). Looks pretty much like a cosmological constant (w=-1) from here, although there is obviously wriggle room.

  • Ryan Scranton

    Yes and no. It’s a very nice result, but the same physical mechanism (the formation of structure freezing out as dark energy becomes the dominant energy in the universe) gives rise to the integrated Sachs-Wolfe effect, which is near and dear to my heart. We’ve had measurements of that at the 2-3 sigma level since 2003 and two groups found detections in the 3.5-4.5 sigma range earlier this year by combining measurements from the previous detections.

    Still, I think the key in the next several years of dark energy science is going to be finding ways of measuring both the dynamical and growth function aspects of dark energy (and hopefully finding a signature that’ll tell us if we’ve really got dark energy or modified gravity). Having another tool for that purpose is definitely a good thing.

  • Tom

    No physicsit here. Just curious.
    You state “The left plot, which fits the data, has an appreciable cosmological constant; the right one, which doesn’t, doesn’t”.
    Is that a fair comparsion? What that tells me is that the right plot is likely incorrect or incomplete. The left plot represents one way to align observation and theory (i.e. dark matter as the cosmological constant).
    Are there any other theories (outside of dark matter) that are not ruled out that could explain the cosmological constant fit for the left plot?

  • http://blogs.discovermagazine.com/cosmicvariance/sean/ Sean

    Tom– it’s good to have people who are not physicists asking questions. However, as a rule of thumb it is best to assume that the physicists doing the research aren’t dumb. Physicists are doing everything they can to come up with alternative theories and test them against the data; that’s why I linked to alternatives in the post, and mentioned “just to guide the eye” in reference to the figures. Coming up with other theories and testing them is the fun part of the job!

  • John R Ramsden

    I’m sure I remember reading somewhere that an extremal black hole deflects and ejects exactly 75% of the mass that falls towards it. As I recall, this was a precise theoretical result rather than a rough estimate, or that was my impression.

    If this is so, it seems an intriguing coincidence that the measured/deduced proportion of dark energy seems to be converging towards exactly 75% (although of course there could be some other explanation, or my recollection could be at fault – wouldn’t be the first time).

  • Tom

    Sean… my apologies!
    Next time I will phrase my comment better – I certainly don’t think you and your peers haven’t thought through all this. I am pretty sure I don’t understand most of your questions, much less your answers! Just me trying to keep up.
    I actually have the utmost respect for physicists. In some parallel universe I think I might have been one myself :-)

    I did fail to follow the alternatives link – thank you for providing that.

  • John L. Winters

    So, looks like we are IN a blackhole. That was our big bang.

  • John R Ramsden

    Or was it 25% 😉

  • http://blogs.discovermagazine.com/cosmicvariance/sean/ Sean

    John– the dark energy fraction is not converging to 75%; it’s growing monotonically, and will approach 100%. Our measurements of the current dark energy fraction seem to be converging on 75% or thereabouts.

  • http://lablemminglounge.blogspot.com/ Lab Lemming

    OK, I’m confused.
    You say that the two plots show the same data, compared to different models, one with and one without dark energy. But on these plots, the model curves seem to stay constant, while the data appears to move. What gives?

  • Aaron F.

    Lab Lemming — Just a guess (I haven’t read the original paper, and I am not a cosmologist), but as Sean says, changing the amount of dark energy in your model affects the relationship between redshift and distance. In particular, it affects the volume you calculate for the region A < z < B, where z is redshift.

    In the plots, the vertical axis is "number density of clusters," and it's given for two different redshift regions: 0.025 < z < 0.25 and 0.55 < z < 0.90. You can find the number density of clusters in a given redshift region by counting the clusters whose redshifts are in that range and then dividing by the calculated volume of the region. When you change your model, the calculated volume of the region will change, and the vertical coordinates of your data points will change as well!

    This explains why the data points in the two plots have different vertical coordinates… but why do they also have different horizontal coordinates? The horizontal axis is "cluster mass," and it's given in units of "M_500." If I recall correctly, M_500 is the mass of the highest-density region of the cluster—specifically, the region whose density is over 500 times the critical density of the universe. Depending on how you measure this, I think it could conceivably be model-dependent, although I'm not sure how.

    So, in summary: the amount of dark energy you assume there is can change a lot of things in cosmology, including the values of some of your measurements!

    On the other hand, I could be completely wrong.

  • Aaron F.

    Sean – Come to think of it, where did that Omega_Lambda = 0.70 estimate come from in the first place? It doesn’t seem to be supported by any of the blobs on that plot, except maybe the Type Ia supernova data.

    But the Type Ia supernova data came first… so maybe I just answered my own question… 😛

  • http://blogs.discovermagazine.com/cosmicvariance/sean/ Sean

    Aaron is exactly right. The background cosmology is different for the second graph, so the interpretation of the observations in terms of number densities is different.

    The 70% estimate is a holdover from the old-timey days of dark energy, indeed.

  • Lawrence Crowell

    We seem to have about 75% for DE. Interesting that to capture a cluster right at the tipping point of formation vs expansion. This might be compared to “weighing the universe,” where the scales compare the relative “gravity” of DE and the cluster.

    One thing I did spot, and I have not looked at the original paper yet, is the right hand plot has Omega_{/} = 0 and Omega_m = .25. Is Omega_{total} = .25 for DM and ordinary matter?

    Lawrence B. Crowell

  • JC

    “We shouldn’t lose our sense of wonder that we’re able to understand as much of the universe as we do …”

    Uh? last time i looked physicists had no idea what 90% of the universe is made of. Giving something a name (dark energy, dark matter) doesn’t mean you understand it (see Feynman’s conversation as a child with his father). Nor does fiddling with some equations to “fit the data” automatically mean that you solved the riddle.

  • http://engti.blogspot.com Robin


    10% of the universe is quite a fascinating place…..

    don’t be a grouch :)

    and no nobody claims that they know what dark matter is, but the first steps towards understanding it has been taken.

  • Lawrence Crowell

    There is a difference between dark energy and dark matter. Dark matter is a cold nonluminous gravitating source which clumps around galaxies and clusters of galaxies. Dark energy is a homogeneous, or so it is thought, effect (I call it that because it might not be “energy” so much as some generalization of gravitation) that fills the universe.

    Lawrence B. Crowell

  • Low Math, Meekly Interacting

    What’s the status of the idea that space-time has some intrinsic curvature that acts like lambda? If I understand correctly, the hypothesis is that space is “automatically” de Sitter with no extra “stuff” gravitating or anti-gravitating required. DOA? Gaining interest? Does it leave us still asking why space is that way (i.e. would our universe’s intrinsic curvature still seem fine-tuned)?

  • http://whenindoubtdo.blogspot.com/ Eugene

    I was about to post a comment about Ryan’s results, when I saw that he is already on the beat.

  • Shameless Booster of Friends

    For the technical types in the audience:
    A related paper appeared on the arxiv on the same day as the one discussed in this post. Given that we know the expansion history of the universe, ie we know that it’s 73% dark energy, 23% dark matter, etc, we can ask if the growth of structure shows evidence for departures from General Relativity, such as DGP or f(R) theories. [We’re not talking MOND here]. These theories predict different behaviors for the ‘growth index’, a way of parameterizing modified GR theories. By counting the number of clusters at a given mass as a function of time, the same technique as the above paper, but with an independent dataset [which presented the above results last year], the authors were able to put the tightest constraints yet on the growth index. Breathe easy…GR is consistent with the data.


    Definitely worth a read if you have a pet theory of modified GR!

  • Sven

    Have they taken the inertia of the matter into consideration? Just like static planets would fall into the star which they are situated in close proximity to, inertia could also potentially explain why the whole system hasn’t collapsed in on itself.

  • http://www.pipelin.com/~lenornst/index.html Leonard Ornstein


    The only observational evidence of the existence of black holes is the influence of the gravitational fields, of the otherwise unobservable matter ‘consumed’ by black holes, on matter on ‘our side’ of event horizons.

    It’s often argued that since we can’t access putative events that might occur beyond the cosmological horizon or the event horizons of black holes with any kind of ‘light signal’, any discussion of “multi-verses”, or of phenomena or ‘existence’ on the ‘other side’ of such a horizon, belongs to science fiction. But, as noted, gravitational effects regularly extend across such horizons. Therefore theories involving massive ‘objects’ on the ‘other side’ of the cosmological horizon may constitute meaningful physics, if they predict observable effects within our universe that depend on the mass of such ‘objects’.

    For example, appropriately-distributed mass, beyond the cosmological horizon, might be responsible for that excess acceleration of the galaxies nearest to the cosmic horizon which has been attributed to “dark energy”.

    Have such possibilities been discussed? Do the Vikhilinin results conflict with such a possibility?

  • Lawrence Crowell

    Loaw math: I am in a bit of a minority on this, but I think the Lambda which is a pure Ricci curvature term may not be so much determined by a vacuum energy density and pressure, but rather this curvature determines the vacuum. We might expect on cosmological scales that any frame we impose, such as our Hubble frame, does not hold globally.

    Lawrence B. Crowell

  • Low Math, Meekly Interacting

    Am I wrong, or would such an intrinsic curvature (where I guess a cosmological constant is an analogous term to the “de Sitter radius” of the universe) solve some problems? Namely, some theorists would like extra symmetries to exactly cancel out SM contributions to the vacuum energy so that it’s zero, i.e. there is no cosmological constant. Since that doesn’t appear to be working, folks are stumped as to why it appears to be so very close to zero, when it should be much MUCH bigger if there is a non-zero value at all, short of invoking the Anthropic Principle. Well, if space has an intrinsic curvature, maybe, one could argue, these extra symmetries really DO cancel everything, and there really IS no c.c.

    Of course, even then, maybe we’re still left stumped as to why the value of the de Sitter radius is what it is, and hence there’s no refuge here from anthropism. Hence, maybe it’s not the most well-motivated idea out there.

    Just my idle curiosity. I can’t claim much of a legitimate opinion one way or the other.

  • Eyes Squared

    John L. Winters Said:
    December 16th, 2008 at 4:16 pm
    So, looks like we are IN a blackhole. That was our big bang.

    I’ve been saying this for years!!! I spoke about it with JC and at Fermi earlier this year. Apply the disintegration of matter crossing the event horizon of this much larger system to the progenesis of our universe and we have a full model of the system.

    Big TOE here we come!!!

  • Pingback: Some Stuff You Might Want to Read « IBY’s Island Universe()

  • Pingback: A New Challenge to Einstein? | Cosmic Variance | Discover Magazine()


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Cosmic Variance

Random samplings from a universe of ideas.

About Sean Carroll

Sean Carroll is a Senior Research Associate in the Department of Physics at the California Institute of Technology. His research interests include theoretical aspects of cosmology, field theory, and gravitation. His most recent book is The Particle at the End of the Universe, about the Large Hadron Collider and the search for the Higgs boson. Here are some of his favorite blog posts, home page, and email: carroll [at] cosmicvariance.com .


See More

Collapse bottom bar