Wednesday, April 03, 2013 ... Deutsch/Español/Related posts from blogosphere

AMS-02 seems to overcautiously censor solid evidence for dark matter

The case for a \(350\GeV\) neutralino strong but probably made look weaker

Quick links: the new AMS-02 paper (open access); liveblog from the CERN talk (US LHC blogs); a remarkably accurate popular overview of the situation and the implications of SUSY models (The Economist!)
At 5 pm CERN Summer Time, I was watching a CERN talk by Sam Ting of MIT, a Nobel prize winner for J/ψ and the boss of the AMS Collaboration, via I was impressed by the complexity of the problem to do such an experiment in space, by the professional attitude by Ting and lots and lots of collaborators, and by the number of subtleties and tests they had to be careful about.

The AMS Collaboration released a press release. Just to be sure you know what we're talking about, the 7.5-ton gadget on the board of the International Space Station measured the number of electrons, positrons, and protons at various energies between \(1\) and \(350\GeV\) coming from various directions of the heavenly sphere.

Ting hasn't shown the absolute numbers yet – they will appear in a paper that will be out within days – but we've seen the positron fraction. The graph is incomparably more accurate than the corresponding graphs by PAMELA and Fermi. The positron fraction (of the positron+electron "whole") starts at about \(0.10\) near \(E=1\GeV\) on a plateau that smoothly begins to drop, reaching a minimum of \(0.05\) near \(9\GeV\) or so, another plateau.

Then it begins to increase again, reaching something that looks as another local extremum (well, a global maximum) near \(200-350\GeV\) where the positron fraction is close to \(0.15\).

The number of detected particles becomes too low – the error margin is pretty high – for these very high energies although it's still comparable to the "normal" number of particles that PAMELA built their claims upon. In the last "unclassified" bin, \(260-350\GeV\), they see just \(72\) positrons and they decided not to show the data above \(350\GeV\), something that a female physicist with common sense aptly asked about (well, so did one young man before her but Ting vehemently pretended he didn't understand the question; incidentally, despite his clear East Asian accent, I understood his English perfectly).

Ting has explained that he is an A* experimenter who would never publish anything wrong so all of us had to wait. It took 18 years for AMS to get to this point so be ready for a few more decades. However, your humble correspondent has performed a thorough analysis of Ting's words and facial expressions. I have determined, with the statistical significance of 4.5 sigma, that the facial expressions are the same as those of Sheldon Cooper who is trying to keep a very exciting secret (example video). For example, he had more nervous ticks than a lime disease research facility.

So I think that the number of positrons above \(350\GeV\) is very, very low. Ting admitted that the number of electrons is still very high but he didn't say anything explicit about positrons. ;-) My quantification of Ting's facial expressions indicate that there are \(O(100)\) electrons above \(350\GeV\) but only \(O(1)\) positrons, perhaps one positron (an event we were shown). The fraction probably drops well below 0.15 – perhaps to 0.01 etc. – very sharply. Ting is afraid to tell us because he would run the risk of claiming a bogus discovery.

Let me mention that if there are 100 electrons and 1 positron above 350 GeV, the probability is really small that the fraction is above 0.05, for example. And this drop – a drop that was totally hidden in Ting's official words and encoded in his facial expressions only – is the smoking gun for dark matter. Note that dark matter particles such as neutralinos may just annihilate \(\chi\chi\to e^+e^-\) but this decay rather abruptly stops above a certain energy related to \(m_\chi\).

Without this crucial censored information, theorists are obliged to wait. The fluxes of the electrons and positrons may be parameterized by simple smooth functions as they're lacking any discontinuities or other sharp structures. In this form, ignoring the information conveyed via facial expressions, it's plausible that the positrons and some electrons are emitted by pulsars (the astrophysical sources of choice here), too. Pulsars are spinning neutron stars. Those in the galactic plane would have to be the key here. I personally think that the pulsar theory is in bad shape due to the angular distribution etc.

To summarize, the data he has shown are compatible with some theories of WIMP particles of dark matter – perhaps some new force with force messengers \(1\GeV\) heavy in the dark sector could be handy to explain the drop of the fraction above \(1\GeV\) which would be very interesting if proven more reliably. This compatibility is encouraging by itself but I feel that the AMS folks actually have much stronger a piece of evidence supporting the dark matter interpretation than what they showed us – and it has something to do with the (missing) very high-energy positrons.

That's what I wrote after the CERN lecture. The rest of the blog entry is older.
Watch the AMS dark matter webcast now!

A press release is out now: positron excess, the slope does decrease later (but the fraction itself stabilizes: but the drop could be above the cutoff they imposed due to a growing error), consistent with WIMP annihilations in somewhat natural enough models (!) that probably require a new DM-specific \(1\GeV\) force messenger, other explanations (in particular, pulsars) not yet ruled out. Sadly, even a model with \(e^+e^-\) fluxes, each composed of diffuse and power law components.
Older text written on April 2nd

Wednesday, April 3rd, is the day when the Alpha Magnetic Spectrometer will announce its first, potentially interesting results that will have some implications for our knowledge about dark matter.

First, at 5 pm Prague Summer Time (11 am Boston Daylight Time), you will be recommended to listen to Samuel Ting at

At 1:30 pm Washington DC Daylight Saving Time i.e. 7:30 pm Prague Summer Time i.e. 2.5 hours later i.e. 90 minutes after the talk above is over, you may open NASA TV at which will air a briefing from NASA headquarters.

I am a bit confused about the tomorrow's location of Samuel Ting. Will he be present at CERN, in Washington DC, will he be communicating as an otherwise located person, or will he be teleported from the Old World to the New World within 90 minutes? Oops, yup, the question is actually answered in the press release about the NASA briefing: Ting will be present in DC only via a video link while 3 extra people from DOE and NASA will talk from DC.

This blog entry will be updated once the information becomes available.

Add to Digg this Add to reddit

snail feedback (26) :

reader billy said...

What's required to show if those Dark matter particles are supersymmetric ? Do we need more data about the behaviour of the excess positrons which is not available now?

reader anna v said...

Well, it is an interesting plot . I find hard reading Ting's facial expressions. Have you any rumors in addition to your reading his expression to support your expectation of a drop? At the moment the plot reminds me of the global temperature anomaly :).

reader Luboš Motl said...

It's really up to a combination of the facial expressions and my theoretical considerations. I agree it's like "the global warming temperature" and just like in that case, I think it's pretty obvious that the graph doesn't continue with a big additional *increase*. ;-)

reader Shannon said...

Lubos, do you think supersymmetry is around the corner ? :)

reader Luboš Motl said...

Dear Billy, even with a perfect knowledge, SUSY can't be really established just from these positron data. What it can establish - once the fast drop is shown - is that the spectrum results from pair annihilation of WIMP particles.

SUSY makes WIMPs very natural. They're stable because they're the lightest particles with the negative R-parity. The lightest particle with certain type of charge - even this discrete charge - is stable. Non-SUSY theories face extra questions why the particle doesn't decay e.g. to neutrino+photon etc. But in principle, of course that a non-SUSY theory may always imitate these particular results.

To really establish SUSY, you have to check that all particles have their superpartners with the same strength of interactions etc. It's clear that beyond a certain level, the non-SUSY explanations would become very awkward.

But so far, they haven't officially excluded even the "very different" explanation of the positron excess that it's coming from pulsars.

reader Luboš Motl said...

:-) I think that a clear proof of SUSY isn't around the corner - I've never believed around-the-corner expectations because they seem fine-tuned and a wishful thinking to me - but it could come by Summer 2015.

On the other hand, big new hints of SUSY could be around the corner and this WIMP is one of them.

reader Robert Rehbock said...

Are the error bars with increasing energy sufficiently large to justify the effort at restraint?
I am quite eager to see this confirmation of a single dark matter

reader Luboš Motl said...

Hi, I don't really "know" how large they are. It depends on Nature - on what they saw. ;-)

But it would still be helpful if they gave us the raw data (energies of the high-energy positrons), without attempts to interpret them. I suspect that a part of the secrecy is an effort to "give meaning" to the next 10-20 years of the AMS mission. Only 10% of the planned data have been used to draw the today's graphs.

In this "careful dosing" of the excitement per unit time, I suspect that Ting/AMS have "undershot" what they should have done and people will view their today's results as insufficiently new.

reader Phil Jones said...

What would be the evidence that would allow them to rule out the pulsar source of the excess positrons?

reader Luboš Motl said...

Hi! Insufficiently quickly decreasing positron fraction at very high energies.

reader Phil Jones said...

Ah right thanks! I guess those background estimates are based on some pretty complicated models of production processes plus whatever happens to the pairs en route to the detector. Quite an impressive bit of calculation.

reader Physics Junkie said...

I noticed that there is a bump at about 90 Gev. Could that be evidence of dark matter annihilation, is it the Z0 boson or just a statistical fluctuation?

reader tome said...

If Lubos reads Sam's expressions correctly, a 2nd Nobel is looking like a very real possibility.

reader anna v said...

How about this estimate: In the last bin of the table, there are 72 positrons and the fraction is 0.16. This means there are about 450 electrons+ positrons. Subtracting the "signal" which should have an equal number of electrons leaves about 350 background electrons. Estimating the slope of the background electrons by the previous bin:101?.15=673 ; 673-102=471 ;
within statistics it seems that the electron background is flat at these energies. I think you are right, they are hiding something , like Mann." Hide the decline".

reader anna v said...

just noticed they are using a sort of logarithmic scale for the energy axis, so the last bin should be correspondingly larger.

reader Luboš Motl said...

Dear Anna, exactly although my absolute numbers are less optimistic than yours.

Because the total number of particles decreases with energy, I would guess that between 350 and 500 GeV, there could be 200 electrons-or-positrons. With the fraction indicated, that may still translate to 30 positrons plus minus five so if they're seeing 1 or 2, it's close to 5 sigma.

If it were remaining flat, like 30 or 20, they would still show this extra point. ;-)

reader billy said...

Lubos , does SUSY determine accurately what the mass of neutralinos should be?

reader Luboš Motl said...

No, SUSY is a symmetry, a principle, and it doesn't determine any masses.

A particular well-defined model respecting SUSY does but there are many of them.

reader cb said...

Dear Lubos : if you can find dark matter in "facial expressions" do you mean it is some kind of quantum Cheshire cat ;-)

reader anna v said...

We should not forget the 130GeV gamma peak that is now not as strongly supported cf: . Probably this is adding to AMS caution .

reader Luboš Motl said...

Good point.

Unless I am missing something, AMS-02 has pretty safely ruled out a 150 GeV-like LSP, assuming that what it sees is linked to dark matter at all.

reader anna v said...

Well, if the 130 gamma peak is real , it could be like what we called a Jacobean peak in the p_t spectrum, which starts at a p_t half the mass of the putative particle and goes lower, from higher energy/different direction decaying particles. We could suppose it is a cold particle decaying into a gamma and something else in the cms the particle should have 260 GeV mass. at least. One could speculate that the WIMP can annihilate with an antiWiMP ( suppose violating R parity?) and the same WIMP decay to two gammas through some complicated very higher order diagrams. ?

reader Luboš Motl said...

Anna, playful - but at any rate, I think that AMS excludes a 260 GeV WIMP, too.

reader Kjeld said...

And another question ... what if a dark matter particle annihalates below 350 GeV? Is this a possibility? If so then it would be extremely difficult to see the difference between pulsar positrons or dark matter positrons?

reader anna v said...

Well, from what I have seen in the discussions with statistical methods of binning for temperatures ( always current in the skeptical blogs) an uptick at the end of the plot is easy to come by. I expect HEP physicists to be sophisticated enough in statistics not to fall in the trap, but it is only one bin that "excludes", and how sharp can exclusions be is something not shown yet.

reader Dilaton said...

Ha ha Lumo,

thanks for this funny as hell post about these quiete promising results :-)

Your comments about Ting's facial expressions made me LOL and burst with laughter several times :-D

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');