Friday, July 20, 2012

Mostly good news on the Fermi \(130\GeV\) line

Dark matter may be starting to shine in front of our eyes

Someone at the Physics Stack Exchange asked about the recent status of the \(130\GeV\) line located in the Fermi gamma-ray data by Christoph Weniger in April 2012 (a confirmation).

It's one of the most exciting and potentially emerging signals of new physics – signals of dark matter. I answered as follows.

Another very fresh paper presented at Dark Attack yesterday, one by Hektor et al.,
also claims that the signal is there – not only in the center of the Milky Way but also in other galactic clusters, at the same 130 GeV energy. This 3+ sigma evidence from clusters is arguably very independent. All these hints and several additional papers of the sort look very intriguing.

There are negative news, too. Fermi hasn't confirmed the "discovery status" of the line yet. Puzzles appear in detailed theoretical investigations, too. Cohen at al.
claim that they have excluded neutralino – the most widely believed identity of a WIMP – as the source because the neutralino would lead to additional traces in the data because of processes involving other Standard Model particles and these traces seem to be absent. The WIMP could be a different particle than the supersymmetric neutralino, of course.

Another paper also disfavors neutralino because it is claimed to require much higher cross sections than predicted by SUSY models:
But one must be careful and realize that the status of the "5 sigma discovery" here isn't analogous to the Higgs because in the case of the Higgs, the "canonical" null hypothesis without the Higgs is well-defined and well-tested. In this case, the \(130\GeV\)-line-free hypothesis is much more murky. There may still exist astrophysical processes that tend to produce rather sharp peaks around \(130\GeV\) even though there are no particle species of this mass. I think and hope it is unlikely but it hasn't really been excluded.

Everyone who studies these things in detail may want to look at the list (or contents) of all papers referring to Weniger's original observation – it's currently 33 papers (update on Friday: 36):
Papers referring to Weniger
Addition for TRF readers...

Yesterday, I also wrote in a comment thread that I find it puzzling when there are many independent hints and each of them seems to be around 3 sigma. Given the a priori possible significance level or relative strength of the signal between \(10^{-10}\) and \(10^{+10}\), it seems surprising when all the significance levels of relevant observations manage to hit the narrow 3-5 sigma window.

However, when I am thinking about the same signal seen in the different galaxy clusters, I am more impressed by it than I was two days ago. The energy \(130\GeV\) is very high. If it happened to be a peak of some astrophysical processes, they would probably have to be processes that depend on some typical star size, star density, and other things, and it seems likely that the peak which depends on these astrophysical parameters would be shifted to different energies in different galaxy clusters.

So it seems increasingly likely that some new physics is associated with \(130\GeV\) photons. If this is where the evidence is going, we must determine what the particle is made of.

Do you remember that we used to call the satellite "GLAST" and then "Fermi, formerly GLAST"? The satellite has brought us so many wonderful insights that we can't even remember when we exactly switched to "Fermi" and internalized this new name. But it's here.

Meanwhile, in the dark matter direct search experiments, a new powerful explosion was detonated by the dark-matter-is-not-seen axis. XENON100 has published its new exclusion limits.

Click to zoom in.

All combinations of (mass, cross section with the nucleon) above the blue line at the bottom – the line inside the green-yellow Brazil band – are said to be forbidden by XENON100's data. This of course strengthens the conflict with the dark-matter-is-seen experiments such as DAMA, CoGeNT, CRESST, and perhaps others that claim that the dark matter apparently does show up and it lives at various islands in the XENON100-forbidden sea.

It's increasingly unclear which side is right. The conflict is getting sharper. The methodologies of all the experiments are inequivalent so it is in principle conceivable that the dark matter particle is behaving in very different ways in various experiments and all of them are compatible with each other. However, it seems more likely that some of the experiments are doing something wrong.

The tension exists not just in between the different dark matter direct search experiments. There is a tension with the 130 GeV Fermi line, too. While XENON100's exclusion curve is just starting to "bite" into the region predicted by various models of a neutralino, these models (and the XENON100-allowed cross sections) seem to predict way too low cross sections to explain the rather strong Fermi line located by Weniger.

You may decide that the right reaction is to dismiss XENON100: there are too many hints that dark matter exists. But that won't really reduce the discrepancies much because the direct dark-matter-is-seen experiments indicate the dark matter particle mass close to \(10\GeV\) which is much lower than Weniger's \(130\GeV\). So something could be completely wrong about our models or the experiments – and things could be much more complex or very different from what the experimenters think.

BTW I spent some time reading an impressive proof of the formula for all the tree-level amplitudes in \(\NNN=8\) SUGRA. Equipped with this proof, it seems increasingly clear to me that this is just a version of the sphere-surface tree-level stringy scattering amplitude that may be extracted from type II string theory with some way to rewrite the degrees of freedom to the twistor variables, ignore some "compactified ones", and using some analytical continuations so that the path integrals get reduced to residues. I would love to see the reduction of the formula to tree-level string theory in some more explicit form but even without that, the status of the twistor formulae looks clearer to me – it's string theory with reparameterized world sheet variables.


  1. I already "liked" these good news at Physics SE, it is cool :-)
    Rewriting the formula for these tree-level scattering amplitudes according to your ideas could probably lead to an awsome paper ... ;-) ?

  2. Thanks for the post.

    First, I want to say, that the signal can't be 20,000 sigma, I disagree :-) It is strictly between 0 and ~6-7 sigma for all sources, because if it were above ~6-7 sigma it would have been discovered already :-)

    Second, something can be learned from the fact it isn't 0 sigma in the clusters. If this were due to dark matter decay rather than dark matter annihalation, the signal would be 0 toward the clusters. That is because the signal from dark matter decay scales as the density, whereas the signal from dark matter annihalation scales as the square of the density -- the latter benefits from substructure. I think decay of dark matter was already disfavoured from theory, but it is still useful to show that this line cannot be due to dark matter decay.

    The substructure leads to a predicted parameter called "boosting". Boosting can be predicted from theory if you assume negligible self-interactions in the dark sector. They say that the boosting their result is consistent with is ~1800-3800, and that cosmological models of structure growth predict a boosting factor of around that quantity.

    I am sure other astronomers will check this. I'm also cautious and optimistic.

    As another note, Galaxy clusters are on average a few thousand times further away than the Galactic Center (2.g. 20 megaparsecs versus 8200 kiloparsecs). They are also a few thousand times more massive, so you'd expect the signal to be 1000x weaker. However, the "Galactic center", represents only a few percent of the signal from the Galaxy, they're using 6 clusters, and the signal to noise for these clusters is ~5x smaller than that for the Galactic center, so it might come close to working out.

    Another potential issue is that this mass is very different than the 5 GeV dark matter reported by some of the direct detection experiments. I'm not sure who is more reliable at this point, but in any case I don't think it's certain dark matter has to be a single particle even though it's an unwritten assumption in most papers on the subject.


  3. Does this have anything to do with the volume of polytopes talked about by Jacob Bourjaily on the Parks Taylor formula after 13:00 in this lecture?

  4. So there is some good work being done after all the Perimeter Institute..

  5. I meant at the Perimeter Institute. Is it possible to edit the Disqus comments?

  6. There's a lot of good work at the Perimeter Institute these days.

    I see "edit" before my comments, next to reply. Don't you?

  7. When I first became interested in the fundamental laws of physics, some six decades ago, the world was very different. Astronomy (cosmology) was totally empirical and could, in no way, be considered a science; it was observation accompanied only by wild speculation. Now, our observations of the cosmos are adding immensely to our understanding of the basics of our existence. I really think that this is the most exciting period that I have experienced vis-a-vis probing the depths of our knowledge concerning how things really are and how they came to be.

    I just don't see how Sabine Hossenfelder and others can be discouraged and pessimistic about what is going on. I have never seen better times than these.

  8. I agree, Gene, and I think that even during my childhood, we lived through conditions you suggested.

    It would have been normal for kids like me to be really into astronomy. Let me admit, I have never been an astronomy geek. Astronomy looked like botany a little bit. People would be happy about looking at lots of different objects - usually just points - and their random properties that weren't related to anything else, and so on.

    Of course, astronomy is a human activity of this kind even today.

    On the other hand, there were far-reaching philosophical speculations about universe and God - those things were usually presented together with religious ideas etc. - and they were disconnected from science and a tighter network of ideas, too.

    So already as a kid, I was more intrigued by elementary particles and the laws that hold everywhere and that one may study locally. But I learned lots of related things from astronomy-centered articles and TV programs, like from those by Dr Grygar but not only those, just to be sure.

    Only in recent decade or so, those things got really connected with each other. Cosmology in particular became a precision science with data whose precision and statistical structure is analogous to the collider data and that are equally tightly and directly connected with some fundamental equations. It's of course exciting. The possible ability to find a complete new dominant kind of matter filling the galactic neighborhoods and possibly create it artificially on Earth, and we may really be 1-2 years from making those things clear, is something that even the famous decades half a century ago we may often be nostalgic about just couldn't have.

    These are places where the humans have made some genuine progress and it is fascinating to watch and even more fascinating to be at least a remote part of this process.

  9. I don't but I probably need to register with Disqus, will do that now.

  10. I agree with you and Gene, but I used to think that cosmology became a science almost a century ago, when Friedmann and Lemaitre proposed their model of a hot, expanding universe and Hubble's observations confirmed it (around 1922 - 1930).
    Certainly, there was a long wait after during which not much happened (as opposed to the situation in particle physics), until the discovery of the microwave background radiation in 1964.

  11. The transition of cosmology from speculation to science has occurred (and is still occurring) as things get quantitative; i.e. when the numbers actually agree. I did not mean to disparage speculation, which is, necessarily, the start of everything. Besides, it's great fun!

  12. When I first learned about the Kerr Metric, my first question was "how come this was only discovered in 1964?". And I got the same answer from multiple physicists -- a lot of people could have solved the problem before, but gravity, cosmology, black holes, etc were not attracting the best minds for a long time. It was an intellectual backwater.

  13. OK, the existence of exact solutions like that is fun but I never considered them and I still don't consider them (and their form in particular) too important.

    Einstein was satisfied with perturbation theory. He figured out how to derive the Newtonian limit of GR; and he figured out various intrinsic new GR corrections such as the bending of light and precession of Mercury perihelion.

    If it were up to folks like Einstein, we would be treating GR in this way maybe even today and we wouldn't really lose that much. Maybe we would be heavily using computers and they would know the Kerr and similar solutions - no one would just realize that they have a simple algebraic form.

    Schwarzschild was the guy who started this sport of analytic solutions but of course that it only became a mass sport once people decided it was relevant for something. So they had to really believe that the solutions were physical up to the horizon - and perhaps beyond - and then there was a reason to go through the work of finding the exact solutions.

    Today, it's a mass sport to the extent that people are looking for exact solutions even if they know that they're inconsequential pretty much everywhere.

  14. I think low energy susy does not exist. More likely it enters only at much higher energies, or only in some dual description of our universe. Which is probably more interesting and fundamental than the way phenomenologists describe it.

  15. Dear, again in the Buy WOW Goldvalentine's day. I don't know Buy RS Gold how many of you have me to your chocolate, but I see clearly...Buy Diablo 3 Gold !year I can send--to reduce weight tea?