How to Go After Dark Energy?

By Sean Carroll | October 28, 2009 12:42 pm

It’s well known that dark energy is a mystery — both for scientists, and apparently for funding agencies who are trying to figure out how best to learn more about this stuff that makes up about 73% of the energy of the universe. I haven’t been paying close attention to the ins-and-outs of this saga (there are more rewarding ways to give yourself an ulcer), but last I had heard the National Academy of Sciences had given very high priority to a satellite observatory meant to pin down the properties of dark energy. This was the JDEM idea — Joint Dark Energy Mission, where “joint” indicates a partnership between NASA and the Department of Energy. (They don’t always play well together, but the Fermi satellite is a notable recent success.)

Now, via Dan Vergano’s Twitter feed, I see a story in Nature News to the effect that things have become murky once again. The proposals got too expensive, so NASA turned to the European Space Agency for help, but ended up giving away things the DOE thought were in their domain, so they threatened to take their toys and go home, giving up on the idea of a satellite altogether.

The story is complicated by disagreement over how important it is to measure the dark energy equation-of-state parameter, the number characterizing how quickly the energy density changes (if at all). It’s frequently said that “we know nothing” about dark energy, but that’s not true; we know that it’s smoothly distributed and nearly-constant in density through time. We even have a very natural candidate for what it is: the vacuum energy. There is of course the problem that the vacuum energy is much smaller than it should be, but that problem is there whether it’s strictly zero or just really small. Other models still have that problem, and tend to add other fine-tunings on top. It would be great, and we would certainly learn a lot, if the dark energy were not simply vacuum energy; but right now we have no compelling reason to think it’s not, so it’s a bit of a long shot.

  • Phillip Helbig

    Sean, with the massive resources of all your social networks, books, conference appearances
    etc, please start calling it “smooth tension” again!

  • Ben

    more rewarding ways to give yourself an ulcer

    I prefer yelling at the dang television.

  • Pingback: Supernovae nicht die Antwort: Wie soll man die Dunkle Energie ergründen? « Skyweek Zwei Punkt Null()

  • Tod R. Lauer

    DE was a discovery provided by those who simply went out to observe the history of the Hubble expansion. Had they not bothered to make the measurements, we would have not known about it. Expect for a few speculative papers, an accelerating expansion was not part of the conversation pre-1998. It cannot be forgotten that this is an unexpected observational result.

    OK, so now we have a some notions of what the EOS “should” be. So should this keep you from going back to the telescope to have a crack at it? If you believe in “once burned – twice shy,” the answer is clear. With no clear theory, DE remains an observational game, and we can improve our knowledge of the DE EOS to accuracy significantly beyond what we know today for resources that are substantial but are well within a minority portion all the other things we want to do and are embarking on.

    It seems ironic after a revolution in cosmology brought by people armed with telescopes that we would decide to walk down from the mountains.

  • Sean

    Tod, that is is little beside the point. Of course every sensible person would love to gather as much data as possible about the dark energy equation of state. If it’s not equal to -1, or if it’s somehow varying, that would be a momentous discovery, of the same order as the original discovery of dark energy. That’s not in dispute. The question is, how do we best go after it in a time of very finite resources? Is a satellite the way to go, or can we do it on the ground? If we do need a satellite, how do we prioritize different missions with very different goals? I don’t know any of the answers, but we have to be able to ask the questions without being painted as anti-scientific or anti-experimental.

  • Ellipsis

    Being that, after about 1000 SNIa, the limiting uncertainty is photometric precision, _not_ supernova statistics, having an atmosphere in the way is more than a bit of a disadvantage. You’re completely limited by the fact of being on the ground. As well as the fact that you can’t get significant statistics beyond z=1.2 or so. Not much of a choice.

  • Tod R. Lauer

    Dear Sean, please excuse my perhaps excessively emphatic answer – it is a reaction to too many conversations that I’ve participated in on this issue. The issue is indeed mute for ground-based programs, as there are plenty of teams who have already or will mount the resources for frontier work – there is no question that observations will be made!

    As you note, it’s the $ ~1G space probe that gets the discussion going. A concrete answer to how to decide the priorities is that we have a process for this. JDEM did rise to the top of the BEPAC survey (which by definition was limited to the NASA Beyond Einstein line), and now we are waiting for the decadal panel. Appealing to process I admit sidesteps conversation, but yet, a decision has to be made and this is how we have chosen to do it. I do want to emphasize, however, again where JDEM sits in the space portfolio. It is cheaper than HST, JWST, LISA, IXO and so on. It still may be too pricey for a risky return, but in context, it does not compare to a “flag-ship” mission.

  • Eugene

    Tod sez : With no clear theory, DE remains an observational game, and we can improve our knowledge of the DE EOS to accuracy significantly beyond what we know today for resources that are substantial but are well within a minority portion all the other things we want to do and are embarking on.

    I think this is a straw man. JDEM is basically sucking up all the $$ for other experiments…remember the Cage-o-Doom where all the Beyond Einstein probes get thrown into and only one emerges? Well JDEM emerges, with LISA limping along and CMBpol/Constellation-X dead in the water. So JDEM is not taking the “minority portion”, it took basically the whole pie.

  • Chris Hirata


    First, there’s an underlying difference of philosophy here of what to do given that one model for DE (Lambda) stands out as the “simplest”. Some people would argue that this means it should be de-prioritized because the “probability” of alternatives to Lambda is small. An alternative viewpoint is that having one theoretically favored model that makes a definite prediction (i.e. where the theory says you shouldn’t need additional fudge parameters) warrants increased priority.

    I also think we may have different notions of what “nearly constant in time” means. Exactly constant in time means rho[DE]~a^0 (where a is scale factor). A 0.07 (1 sigma) error in w means that we know the exponent of a to an uncertainty of 0.2 (1 sigma) or 0.4 (2 sigma). The Nature article repeats the common oversimplification (sometimes made by me too!) that this means DE “is constant with a precision of 7%”. Well, not quite. In particular, I disagree with the notion that Lambda is already well enough tested to “check the box”.

    As for how we go after DE: if you really do want an ulcer, “my cosmological probe is better than your cosmological probe” is a great argument to have :)


  • Brian Gerke

    The implication here that BigBOSS is the DOE’s threat to give up on a satellite altogether strikes me as a misunderstanding.

    As I understand it, BigBOSS is a smallish team of scientists, not all from DOE, claiming that a particular observational strategy for JDEM (acoustic oscillations with photometry, and nothing else), which had been gaining traction, can be done around 10x more cheaply from the ground than from space. To drive home the point, they then proposed to do the ground-based survey.

    But this isn’t the same as DOE “giving up” on a satellite, it’s just some scientists arguing, correctly, that the science needs to justify the cost: if you’re going to to to space, the science return had better be waaay better than what you can get from the ground. If the BigBOSS claims are correct (and I’m not in a position to say), then JDEM will only be worth it with a more comprehensive approach in terms of the number of DE probes it considers. (The fact that DOE folks had such a strategy in mind is possibly relevant here, though.)

  • Doug


    JDEM emerged from BEPAC because LISA et al were untested (and in many cases yet to be developed) technologies which would cost well in excess of JDEM. As BEPAC was tasked with deciding which of the Beyond Einstein probes should be the first to be built in a budget situation where you could only build one in the next x years (I think 5, I forget the actual number), they concluded (wisely imo) that the one that uses known technology is probably the best bet.

    Any multi-satellite mission is unproven technology, so would likely require a pre-LISA mission to develop that technology before starting the LISA design. If JDEM were to be funded, you’d be draining a portion of LISA’s development budget, nowhere close to the full budget. Really the only argument worth having (among the BE designs) for a near term mission is a JDEM vs a descoped SIM vs a descoped con-X (or whatever the current name for it is, as it can’t be a multi-satellite constellation mission).

  • Doug

    I should also add, if you use weak lensing as one of the probes of DE in JDEM, you get several thousand square degrees of space quality imaging along with redshift information for all of the galaxies in that area (photo-z’s at least). Even if you don’t think DE is worth studying, the amount of additional science which such a dataset would produce would be immense, and probably worth the cost of the mission by itself.

  • Brian Gerke


    Thanks for making my oblique implications more explicit. :)

    I can’t imagine how any space-based mission would be worth the cost without enabling lots of ancillary science besides DE. Thousands of square degrees of Hubble-like imaging? Yes, please.

  • Timon of Athens

    “, if the dark energy were not simply vacuum energy; but right now we have no compelling reason to think it’s not”

    Actually, we have no compelling reason to believe either way: see for example

    There are degeneracies all over the place, and breaking them is all-important.

  • Sili

    Is there more to this mission than ‘just’ very accurate measurements of SN1As? If not, why is a dedicated probe necessary – it would seem to be a job a more ‘all-round’ telescope could do. I’m not saying we need a new Hubble (but we need a new Hubble), but a satellite that could do two or three things at once might do the trick.

    How thick on the ground are super novae? Would it be possible to use the space telescope in question to do one kind of ‘boring’ observation most of the time, and then just turn it around whenever the fireworks go off?

    Of course, I’m ignorant and have my biases – I’m not really fond of Kepler, even if it does what it does very well. To be fair I probably should look up how much that cost. (And I’m being unfair since I lurrrrrrve the dedicated missions of WMAP and Planck.)

  • Eugene


    I understand the arguments pro/con of BEPAC. I am not saying I disagree with BEPAC’s choice, I am just calling out Tod’s statement that JDEM is a small portion of the pie, which is of course, not true.


  • likaifang

    Oh,That’s bad news,I think it is a mission worth to do.

  • shantanu

    talking of LISA, anyone know the status of LISA path-finder? It was supposed to be launched in Oct. 2009?


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] .


See More

Collapse bottom bar