Planck’s data is extraordinary, but will it teach us anything new?

On April 3, 2013

Planck's baby picture of the universe. Image: ESA

The recent results from the Planck satellite have garnered considerable interest among both scientists and the media. But from the discussions that have followed, I am learning at least as much about the nature of science as the origins of the universe.

But first, the science. (If you have heard all you want to about the cosmic microwave background, you may want to skip the next paragraph or two.)

Planck is designed to map and measure the cosmic microwave background (CMB) at higher resolution and over a wider range of frequencies than ever seen before. The CMB itself is a sea of photons left over from the astonishingly high temperature ‘fireball’ of the big bang. As the universe expands after the initial ‘kick’, it cools. When the CMB was formed, the universe was 380 000 years old. Its temperature was 2700C (at this point the universe would have glowed a red colour similar to an electric fire). In the billions of years since then, the expansion and cooling has continued, giving a temperature now of -270C (just 3 degrees above absolute zero) and a glow of microwaves. This glow is the CMB.

The CMB is important as it is the earliest light that it is possible for us to see. Before the CMB was created the universe was effectively opaque. However, imprinted on the map of the microwave sky is information about the universe. Not just at the time the CMB was formed, but much earlier – even well into the first second after the big bang. It is these echoes that Planck is studying. So, what has been discovered in this first data release? The headline numbers are impressive:

  •  Our basic model of the big bang and subsequent evolution of the universe fits well, with no need (or indeed much room) for any strange tweaks in the early stages.
  • The universe is 13.8 billion years old – slightly older than the previous best estimates (though within the uncertainties).
  • The universe appears to be made up of 5% ‘normal’ matter (the stuff of stars, planets and us); 26% mysterious, though much sought-after, dark matter; and 69% dark energy which nobody really has a good handle on at all.
  • There is an odd, albeit slight, north-south asymmetry and an unexplained cold spot in the south.

So, there is no question that Planck is working very well and the data quality is superb. However, none of these results are actually new. The precision is a significant improvement over previous measurements from the Cobe and WMAP satellites, but it could be argued that, while our understanding is more precise, it is no deeper.

It should be stressed that these are early days for Planck and much more analysis and calibration needs to be done before all the implications can be understood. However, it is important to face the question: Is Planck going to lead to any new science?

There is a useful comparison here with one of the biggest science stories of recent years: the discovery of a new particle at the Large Hadron Collider (LHC). Here, like Planck, a new area of ‘parameter space’ was set to be explored by better, more sensitive instrumentation. Also like Planck, aspects of the current best-bet theory would be tested to new limits and values of important parameters refined. However, arguably unlike Planck, there were also some specific new fundamental theories with clear predictions that could be tested. Some of these remain to be tested in future years, some have been relegated to history and some, such as the Higgs boson theory, are well on their way to passing their tests. Could the same be said for Planck?

Of course, the comparison is not a perfect one. Both the scale and scope (and cost) of the LHC are far larger than Planck. On one side, we have an unprecedented instrument that carries a significant fraction of the entire field of particle physics on its 27km shoulders. On the other side, a single telescope chipping away at a corner of astrophysics. Nevertheless, there are disgruntled mutterings that perhaps the resources that went into Planck could have been better spent on other, perhaps more risky or speculative missions.

This sort of discontent is natural when difficult decisions have been made in the allocation of limited resources, and hindsight provides an excellent fuel for it. Nevertheless, the balance between speculative science with a small chance of an enormous return, and the steady improvement of knowledge and understanding through incremental advances is a difficult one to strike, and some will always be unhappy.

Naturally, it is good to get such excellent confirmation that we are on the right track; that the work of the last decades has been well spent and we are moving steadily forward. Nevertheless there are many, myself included, who were hoping for something more. We were hoping that Planck would, at least, show new holes in our knowledge, new questions we should be asking. And, at the most, perhaps even show a crack in our basic understanding – the first hint that somewhere ahead of us is a major leap in our understanding.

Maybe these results are just around the corner and the further analysis from Planck will soon lead to just the sort of major result that I would like to see, but that is unlikely and I will probably remain disappointed for now.

However, that does not mean that I consider Planck in any way a failure. The quality of the data is extraordinary, the work of the science team in analysing it superb (and on time), and the improvements in the values of some fundamental properties are significant. The important thing now is not to ask what could have been done differently, but to see what we can do now. Maybe the results themselves do not (yet!) show anything unexpected, but they may be the first step towards it. And as the Chinese have known for many years:

“It is better to take many small steps in the right direction than to make a great leap forward only to stumble backward.”

Image: Planck’s baby picture of the universe. Credit: ESA

About Andy Newsam

Andy Newsam is Professor of Astronomy Education and Engagement at Liverpool John Moores University.  His research involves turning sets of astronomical data into a collection of useful numbers that can be used to do science. He is Director of the National Schools' Observatory, which gives school children the opportunity to make their own observations alongside professional astronomers on top-quality telescopes, and also runs a suite of distance learning courses in astronomy for the general public.

10 Responses to Planck’s data is extraordinary, but will it teach us anything new?

  1. Excellent point, Andy – thanks for the post. As someone who worked on the CMS experiment on the LHC, I was amazed at how quickly some people started decrying even the Higgs boson as boring – given that it was still “just” part of the Standard Model. It’s important to think about what one defines as “new physics” – something I blogged about here.

    Out of interest, where do you think we might get the best “new physics” bang for our buck? Until the LHC starts up again in 2015, my money would be on Dark Matter experiments like DarkSide-50. AMS may have something interesting to say later this afternoon though…

    PS: I’d also argue that, in many ways, Planck is sort of like the LHC anyway

  2. I would say that it’s a little unfair that there was *nothing* new from Planck. Sure, there was refinement of current theories, in terms of narrowing error bars and tweaking numbers, but there were some new results – though in fairness they are somewhat subtle and so not really picked up on in the press. By far the easiest result to get a handle on was that the Universe is slightly older than we thought – though as you say it’s within the old error bars (which are now much smaller), and in fact scientifically that’s probably one of the least interesting results – largely because the calculation of the age of the Universe requires a number of assumptions.

    It’s incredibly impressive that a theory that essentially uses just six numbers (and a bunch of equations, of course) can match the data to such high precision – data that is only obtainable with Planck. One of the really nice things is that some of these numbers are now obtainable from the Cosmic Microwave Background data alone, without requiring the addition of other data sets (which normally requires some assumptions and can introduce additional uncertainties). Not that Planck is trying to work in isolation – the combination of lots of data sets (when done properly) is essential for a proper analysis.

    For example, the very strong evidence (where there had not really been anything firm before) that the properties of the fluctuations on small scales compared to large ones (the “spectral index”) agree with the predictions of inflation. This, along a few other statistics of the fluctuations (e.g. whether there are more hot spots than cold ones in the CMB – the so-called “non-Gaussianity”) are predictions of inflation, and are being tested by Planck. These measurements are only possible because Planck looks at the whole sky at high resolution and sensitivity – previous experiments have looked at the whole sky at poorer resolution/sensitivity, or looked at small patches at very high resolution.

    On the flip-side, the Planck results have ruled out a few “exotic” theories about weird properties of the Universe. These theories made predictions which simply don’t show up in the Planck analysis. For example, the data has also fairly conclusively shown that there are no currently unknown types of neutrinos.

    Even more subtle(!), is the ability to map the history of mass in the Universe through the distortions of CMB light (“weak lensing”) (see e.g. ESA Science & Technology and this BBC News article)

    The stuff about the “anomalies”, such as an apparent asymmetry, is possibly on the verge of requiring new physics, though not as you say that might become clearer after further analysis – due out some time next year. And the evidence that the expansion rate appears different needs some explaining. There is certainly something “fishy” about it all, but it’s not clear what yet.

    I guess what I’m getting at is the distinction is what you mean by “new science”. If you mean the disproving of generally accepted theories and the requirement of brand new physics, then I guess not. If you mean new strong evidence supporting theories that only had weaker evidence before, and ruling out other theories, then Planck has done something new. Does all “new science” have to be headline-grabbing? I would argue not – though it’s very good if science generates discussion (such as this one!)

    As with the LHC, it would be very exciting if Planck were to produce results that required a complete re-writing of the text-books, but they may end up being in the same boat in the end. For example, the LHC might prove that the Higgs Boson exists as predicted, and there’s nothing new required in that theory. Likewise, Planck might prove that inflation is correct, and nothing new is required. Of course, there’s always details, such as what kind of Higgs Boson, and what flavour of inflation, but these rarely show up in the headlines.

    George Efstathiou (Planck Survey Scientist, Cambridge University) said that Planck has taken us from the era of “precision cosmology” (where we were refining our model of the Universe) to an era of “discordance cosmology” (where we do our best to prove that model wrong in an effort to find new physics).

    That’s turned into a rather long reply, probably because I think your arguments are fair, but I thought someone should stand up for Planck!

  3. While the first results on cosmology from Planck might look like it’s just confirming what we already thought, there’s a huge amount of new stuff that still needs to be sorted through. At the Planck ETLAB meeting yesterday George Efstathiou made it clear that the small but real discrepancies with other data sets (eg. the Planck Hubble parameter value vs. the HST programme value) might provide hints of new physics, or might just be the result of poor understanding of the detailed physics of galaxies. At this point we don’t know, and more analysis of both Planck and other datasets, as well as theoretical modelling, is needed.

    Then there is the Planck polarization data, which isn’t out yet, but which early tests presented here looks of very high quality. This will at least constrain inflation models even better, and might provide new physics.

    But there’s also a whole load of other results from Planck from the rest of the universe between here and the CMB. There’s a fantastic new SZ selected cluster catalog, all sky submm surveys for the kind of dusty galaxy that I work on, thanks to CMB lensing we now have a map of the large scale mass distribution in the universe, and there’s lots of stuff that can be done on the Galaxy and processes closer to home.

    The true value of Planck will only become clear over the next 10 years as the data is studied in detail and compared to multitudinous other data sets.

  4. It’s good to see that there is indeed so much that Planck is bringing to the table. As I say, I also believe that not only is the current data extraordinary, but the best is yet to come. I guess the problem, as Chris says, that most of the boundary-stretching stuff is not very media-friendly. I found that this gave rise to a bit of a mis-match between the pre-release excitement and the details that made it into the press.

    My main point is a more general one, though (and I have rather picked on Planck). There is a tension when designing (and trying to funding for) missions/instruments/experiments between confident, incremental science gains, and speculatively looking for for the truly unexpected and revolutionary. To my mind, Planck was mainly about the former, as indeed most experiments must be. However, there is strong pressure on every mission to do both.

    That is rarely going to be possible, and will largely be luck when it does, but while the pressure exists, any mission that does not at least appear to will seem, to some, weak.

    How do we deal with that pressure? And is it sustainable in the long term?

    I have no answers, but I think it is important to ask the questions, especially when science budgets around the world are under severe pressure.

    (Sorry for any typos. I’m trying to compose this on my phone in a field in Cumbria and while that is splendidly high-tech and exciting, it is freezing my fingers…).

  5. Here we have the LHC and the Planck intrument. With the LHC, we are almost there with the SM Higgs boson and this completes the particle SM as such. However, the LHC with its higher energy MAY bring some thing beyond the SM. Yes, it may with of all sorts of theories glaring at it for their redemption.
    As to to the Planck, the most interesting thing will be the polaristion results that will directly tease the inflation phenomenon and via it, the large scale isotropy and homogeneity of the univers, which are the basic elements of its SM.

  6. I’d say Planck has made a major discovery in large-scale inhomogeneity, but the penny hasn’t dropped yet. See the ESA website and note “We see an almost perfect fit to the standard model of cosmology, but with intriguing features that force us to rethink some of our basic assumptions.” People are usually pretty hopeless about examining their basic assumptions. But look at the Lambda-CDM model on wikipedia and note that “the model uses the FLRW metric”. Then follow the link to that and note this: “The FLRW metric starts with the assumption of homogeneity and isotropy of space”. Only Planck says it’s not homogeneous and isotropic! Which means that not only is the fine structure constant not constant, but the cosmological constant isn’t constant either. And that’s “equivalent to an energy density in otherwise empty space”. That energy has a mass equivalence, so any dark-energy inhomogeneity will act as dark matter. See for example “Inhomogeneous and interacting vacuum energy” on arXiv by Josue De-Santiago, David Wands, and Yuting Wang, and expect more papers along those lines. We could end up with dark matter solved thanks to Planck.

  7. The current funding environment in the UK at least will mean that large collaborations going after evolutionary not revolutionary science will become the rule. This was always the case with space experiments (there’s a distinction with space observatories but it’s true there as well), but with the thinning down of the UK’s access to ground-based telescopes, it gets harder and harder for a small team to get time for something radical.

    Under the REF, too, large teams that produce a large number of papers, on which there are many authors, will likely trump people producing papers with a small number of authors but in smaller numbers. And someone spending a long time sorting something fundamental properly (examples from the past might include DNA structure or the confocal microscope) is going to be totally screwed.

    With funding restrictions and the HEFCE looking over everybody’s shoulder we will inevitably end up with more conservative science.

    • Yes, ” the big project science” seems to be the case for the physical sciences such as particle physics and cosmology, but for other disciplines such as biology, solid-state physics, material sciences, the thing still keep their “human” scale.

    • Sadly, I think you are right that funding pressures combined with increasingly detailed and narrow-criteria scrutiny are currently leading to more conservative approaches, but is it really inevitable? The eternal optimist in me thinks that there must be an alternative. However, that same optimist doesn’t have a particularly good track record…

    • Dave, I think you’ve pretty much hit the nail on the head there. That’s precisely what worries me about the future astronomy in the UK. Maybe Andy is right that there is an alternative. Not sure what it is, but maybe it would help if we could convince policy makes that science isn’t just about press releases.