David Harris at symmetry breaking points to a paper and accompanying commentary on the search for high-energy cosmic antiprotons by the PAMELA satellite experiment. (What one defines as “high-energy” depends on one’s upbringing; we’re talking about energies of up to 100 times the mass of the proton.) The impression is given that this is a brand-new result casting doubt on the earlier claims that PAMELA might have detected evidence for dark matter; that’s not really a correct impression, so it’s worth getting it all straight.
The PAMELA satellite, an Italian/Russian/German/Swedish collaboration, looks at high-energy cosmic rays from orbit, and pays particular attention to the presence of antimatter — basically, positrons (anti-electrons) and anti-protons. Part of the idea is that a high-energy matter particle can simply be a particle that had been lying around for a while and was accelerated to large velocities by magnetic fields or other astrophysical processes, whereas you need some pretty high energies to produce antiparticles in the first place. Say, for example, from the annihilation of dark matter particles with each other. There are certainly some high-energy collisions in the ordinary non-dark-matter world, so you expect to see a certain fraction of antimatter, but that fraction should noticeably diminish as you get to higher and higher energies.
So in October the experiment released two papers back to back:
A new measurement of the antiproton-to-proton flux ratio up to 100 GeV in the cosmic radiation
Authors: O. Adriani et al.
arXiv:0810.4994Observation of an anomalous positron abundance in the cosmic radiation
Authors: O. Adriani et al.
arXiv:0810.4995
If you look closely, you’ll notice the second paper has 10 trackbacks to its abstract on arxiv, while the first doesn’t have any (until now!). The reason is clear: the second paper has the word “anomalous” in the title. The PAMELA measurements of positrons deviate significantly from the theoretical expectation, while the measurements of anti-protons reported in the first paper are exactly what you might have predicted. Who wants to write about observations that fit theories we already have?
You might remember the PAMELA positron result as the one that created a stir when they gave a talk before submitting their paper, and theorists in the audience snapped pictures of the data with their cell phone cameras and proceeded to write papers about it. Those wacky theorists.
Here is the relevant positron plot, from paper 2 above:
 |
The vertical axis is the fraction of positrons in the total sample of electrons+positrons, plotted against energy. The red dots are the data, and the black curve is the theoretical prediction from ordinary astrophysical processes. Not the best fit, eh? At low energies that is not a surprise, as “weather” effects such as solar activity can get in the way of observing low-energy positrons. But at high energies the prediction should be more robust, and that’s where it’s the worst. Indeed, it’s pretty clear that the fraction of positrons is increasing with energy, which is pretty baffling, but could conceivably come from dark matter annihilations. See Resonaances for more discussion.
And here is the version for antiprotons, from paper 1 above:
 |
Now that’s what we call a fit to the data; again, fraction of antiprotons plotted versus energy, and the data go up and down just as predicted.
What happened is that the PAMELA collaboration submitted their second paper (anomalous positrons) to Nature, and their first paper (well-behaved antiprotons) to Physical Review Letters. The latter paper has just now appeared in print, which is why Simon Swordy’s commentary in Physics appeared, etc. Although the idea behind Physics (expert-level commentary on recently published articles) is a good one, it’s sponsored by the American Physical Society, and therefore pretends that the only interesting articles are those that appear in journals published by the American Physical Society. Which Nature is most surely not.
So one might get the impression that the antiproton result is a blow against the idea that we are seeing dark-matter annihilations. Which it is; if you didn’t know any better, you would certainly expect to see an excess of antiprotons in dark-matter annihilations just as surely as you would expect to see an excess of positrons. But it’s not a new blow; the papers appeared on arxiv (which is what really matters) at the same time!
And it’s not a blow that can’t be recovered from. All you have to do is declare that your dark matter candidate is “hadrophobic,” and likes to annihilate into electrons and positrons rather than protons and antiprotons. Not an easy task, but that’s why theorists get paid the exorbitant salaries we do. (Without ready access to champagne and caviar, we can hardly be expected to justify unusual branching ratios in WIMP annihilations.) The favorite model out there right now belongs to Arkani-Hamed, Finkbeiner, Slatyer, and Weiner, featuring a new gauge force that is broken at relatively low energies. But there are various models on the market, and the number is only going to grow.
Most likely the PAMELA positron excess is coming from something that can be fit quite nicely into the Standard Model of particle physics, like pulsars. That’s my guess, anyway. Happily, there’s all sorts of data coming down the pike that will help us sort it out.
Pingback: symmetry breaking » Blog Archive » PAMELA signal probably not dark matter
Sean wrote:
“All you have to do is declare that your dark matter candidate is “hadrophobic,””
Can’t we all just get along?
e.
The favorite model out there right now belongs to Arkani-Hamed, Finkbeiner, Slatyer, and Weiner, featuring a new gauge force that is broken at relatively low energies.
Whose favorite? It’s appealing in some ways. On the other hand, there is generically a tension in models of annihilating dark matter with observations of gamma rays (which tend to give stronger constraints than the antiprotons), so it seems that a lot of people are more intrigued by decaying dark matter explanations. (I don’t have a horse in this race, I’m just summarizing my impression of the state of things.)
“Without ready access to champagne and caviar, we can hardly be expected to justify unusual branching ratios in WIMP annihilations.”
I use chocolate. It is superior in every way, and almost as expensive if you get the good stuff.
Gee, last time I checked theorists got paid pretty much what experimentalists got paid. And that is LOTS more than all the unemployed in this country……
Two bucks and a theorist’s postdiction will get you a cup of coffee (but not a venti latte). I’d love to see an experimentalist present some hoax data (e.g. at a conference on April 1) and see how many theorists take the bait and invent a theory that accommodates the hoax data: “The data clearly indicates a super-natural heterotic gauge coupling on a Calabi-Yau manifold.”
What percentage of a star’s mass can be non-baryonic dark matter before it goes noticeably out of whack with the current stellar models?
Arun, I don’t know. Probably a fairly small percentage. But the predicted percentage is extremely small; most dark matter candidates are essentially collisionless, and would not collect inside a star.
Hi Sean, great post!
I want to ask for your advice, hopefully not too tangent to the current topic. To begin, I am a particle physicist, but I am very interested in particle astrophysics. Specifically, I am interested in two topics,
1) the predictions made by fundamental particle physics that affect/constrain our macroscopic theories of the universe (e.g. the amount of dark matter/energy present in the universe, inflation, cosmology, CP violation, etc.) and
2) astrophysics measurements that constrain our microscopic particle physics theories (detection of dark matter particles, supernovae measurements, detection of cosmic particles, etc.).
In short, I am interested in understanding the interplay between the largest and smallest scales in our universe — what particle physics has to say about the universe, and what measurements from space have to say about particle physics models. My question is: can you recommend any research, papers, books, physicists, websites, or anywhere to start, where I might learn more about this exciting (and recent) intersection of fields?
I hope my question is not too poorly-worded as to be confusing… if so I apologize, and thank you ahead of time for any advice you may provide!
Hi Enrico– that’s a pretty broad question. Here is a list of review articles, but I’m sure there are more. There is a book by Bergstrom and Goobar, and another by Perkins, that you might want to check out.
Thanks for posting on this Sean! This afternoon my students and I were just looking at some new work/plots for the positron excess and we are going to tackle the anti-proton flux next…we’ll have to find a fudge factor if the anti-protons don’t come in small enough.
JoAnne, you wrote “find a fudge factor” where you probably meant “improve the sophistication of our theoretical predictions.” Must have been a typo.
Hi Sean,great post!
I’m currently working on this topic from the AMS perspective.Certainly , astrophysical objects are no longer the dark horse in the race, but measurements in other channels will tell. Hope we will improve on this from the experimental side.
Thanks for this article, especially after all the blog entries about (ugh) politics.
It was just a thought – there is a lot of dark matter, a lot of billions of years for it to “relax” into stars, and a lot of matter in stars for nearly-collisionless stuff to interact with. I guess another way to ask the question is – for the set of dark matter parameters such that stars won’t gather any noticeable quantity of dark matter, does that leave any room in parameter space for our instruments, where dark matter will be directly detectable?
Thanks for this Sean..I see Fermilab has picked up your comments but not SLAC.
Do many physicists share Sean’s remark that what really matters is the arxiv? Science journalists are often berated by physicists for delving into the arxiv, rather than waiting for the peer reviewed publication – which in PRL’s case can take an age.
Thanks Sean, I appreciate the recommendations. It certainly is a broad topic… that list of reviews is a nice find!
Pingback: theba.hu » Blog Archive » Ezt olvastam - 2009-02-05
i suspect he meant the date of arxiv posting is what really matters.
journals can take a long time to publish a paper.
Valerie, the process of peer review helps separate the sensible papers from the crazy ones. (Not with perfect fidelity, unfortunately.) But what matters is whether a paper is sensible or crazy, not whether it is peer-reviewed or not. (That’s my view; others may disagree. Levels of disagreement likely correlate with age.)
Scientists could legitimately be annoyed if a crazy paper that never would have passed peer review were picked from the arxiv and widely publicized. But if a paper is sensible, I personally don’t see what difference it makes that it’s been peer-reviewed or not. The idea that a result doesn’t really “exist” until it appears in a peer-reviewed journal is a fiction that is becoming increasingly difficult to maintain.
Of course, I am a theorist, and most of my experience is with theoretical papers. Experimentalists might feel differently, and are welcome to chime in.
Also, physicists just like to berate people, so don’t take it personally.
Valerie: there are some theorists who don’t “bother” to submit all their papers to journals for publication and thus only disseminate through the arXiv. That’s fine for well-known senior folks as their papers will get read anyway, but it’s the kiss of death for young authors.
Sean: Nope, I meant “find a fudge factor” – it is astrophysics afterall. 😉
Sean, thanks for your detailed comments on this paper.
I had precisely the same reaction as Valerie about your comment that arXiv publication is what matters. All of us who write about science these days do write stories based on what comes from the arXiv but we all receive negative comments from scientists for doing it. It doesn’t stop us from doing it! However, having something to peg the story to, like journal publication, helps us choose what stories to cover among the many possible options, especially if we didn’t catch it first time around.
I also agree that the Physics piece could have done with some more context about the Nature paper. Apparently Physics takes letters to the editor although I don’t think I’ve seen any yet. Perhaps it would be worth something brief…
I still think it was worth pointing out this paper now, especially given that, as you pointed out, it hadn’t had any coverage previously but is an important part of the story. Cosmic Variance had talked about the positron paper but not the anti-proton paper, for example. I see that PhysicsWorld has just done a piece on this paper as well: http://physicsworld.com/cws/article/news/37665
It might not be new, but it still seems to be news.
Arun –
Isn’t it better to think of stars collecting in pools of dark matter?
Or really, since they both influence and follow spacetime curvature, DM and ordinary matter ought to collect into wells together, with more massive objects tending to sift (or at least be found) “downward” eventually.
Are there degenerate phases of dark matter? How much denser is DM near the centre of large scale structures than in galactic halos or in inter-galactic space?
Arun’s question got me thinking. It’s probably a crazy idea, but could photons with very long wavelengths, e.g. hundreds of light years long, be trapped by the slight gravitational field of a galaxy, i.e. reflected off the slightly curved space and confined to rattle about inside the galaxy?
With such long wavelengths they wouldn’t interact much if at all with matter, but if there were enough of them (produced in the Big Bang shortly after the Universe dropped below the elecroweak unification temperature?) they would exert a significant gravitational force.
(It may be the Bullet cluster is enough to put paid to this idea, if dark matter remains localized after being bumped aside from its host galaxy.)