The Alpha Magnetic Spectrometer, bolted to the International Space Station on 19 May 2011, has produced exquisitely precise measurements, but nobody knows precisely what they show.

The Alpha Magnetic Spectrometer, bolted to the International Space Station on 19 May 2011, has produced exquisitely precise measurements, but nobody knows precisely what they show.


Space station detector reports more hints of dark matter—or not

New reports of further evidence for dark matter have been greatly exaggerated. Yesterday, researchers working with the Alpha Magnetic Spectrometer (AMS), a $2 billion cosmic ray detector attached to the International Space Station, reported their latest data on a supposed excess of high-energy positrons from space. They contended—at least in a press release—that the new results could offer new hints that they’ve detected particles of dark matter, the mysterious stuff whose gravity binds the galaxies. But several cosmic ray physicists say that the AMS data are still perfectly consistent with much more mundane explanations of the excess. And they doubt AMS alone will resolve the issue.

The leader of the AMS team, Nobel laureate Samuel Ting of the Massachusetts Institute of Technology in Cambridge, takes care to say that the new results do not prove that AMS has detected dark matter. But he also says the data lend more support to that interpretation than to some others. "The key statement is that we have not found a contradiction with the dark matter explanation," he says.

The controversy centers on AMS's measurement of a key ratio, the number of antimatter positrons to the sum of positrons and electrons. In April 2013, AMS confirmed early reports that as the energy of the particles increased above about 8 gigaelectron Volts (GeV), that ratio, or "positron fraction," increased, even as the individual fluxes of electrons and positrons were falling. That increase in the relative abundance of positrons could signal the presence of dark matter particles. According to many theories, if those particles collide, they would annihilate each other to produce electron-positron pairs. That would alter the balance of electrons and positrons among cosmic rays, as the usual source such as the cloudlike remnants of supernova explosions produce far more electrons than positrons.

However, that interpretation was hardly certain. Even before AMS released its measurement of the ratio, astrophysicists had argued that the excess positrons could potentially emanate from an undetected nearby pulsar. In November 2013, Eli Waxman, a theoretical astrophysicist at the Weizmann Institute of Science in Rehovot, Israel, and colleagues went even further. They argued that the excess positrons could come simply from the interactions of "primary" cosmic rays from supernova remnants with the interstellar medium. If so, then the positrons were just "secondary" rays and nothing to write home about.

However, AMS team researchers see two new features that are consistent with the dark matter interpretation, they reported online yesterday in Physical Review Letters. First, the AMS team now sees that after rising with energy, the positron fraction seems to level off and may begin to fall at an energy of 275 GeV, as would be expected if the excess were produced by colliding dark matter particles, as the original particles' mass would put an upper limit on the energy of the positron they spawned. AMS researchers say the leveling off would be consistent with a dark matter particle with a mass of 1 teraelectron volt (TeV). (Thanks to Albert Einstein’s famous equivalence of mass and energy, the two can be measured in the same units.)

Second, the AMS team measured the spectra of electrons and positrons individually. They found that the spectra have different shapes as energy increases. "It's really surprising that the electrons and positrons are so different," Ting says. And, he argues, the difference suggests that the positrons cannot be secondary cosmic rays produced by primary cosmic ray electrons, as such production should lead to similar spectra.

But some cosmic ray physicists aren't convinced. For example, in AMS's graph of the electron fraction, the error bars at the highest energies are large because the high-energy particles are so rare. And those uncertainties make it unclear whether the positron fraction really starts to drop, says Stéphane Coutu, a cosmic ray physicist at Pennsylvania State University, University Park. And even if the positron fraction does fall at energies higher than AMS reported, that wouldn't prove the positrons come from dark matter annihilations, Coutu says. Such a "cutoff" could easily arise in positrons from a pulsar, he says, if the spatial region in which the pulsar accelerates particles is of limited size. All told, the new results are "probably consistent with anything," Coutu says.

Similarly, Waxman questions Ting's claim that the new data suggest the positrons aren't simply secondary cosmic rays. If that were the case, then the electrons and positrons would be coming from different places and there would be no reason to expect their spectra to be similar, Waxman says. Moreover, he notes, AMS's measurement of the positron fraction seems to level out just at the limit that he and colleagues predicted would be the maximum achievable through secondary cosmic rays. So, in fact, the new data support the interpretation that the positrons are simply secondary cosmic rays, he says. "To me this is a very strong indication that we are seeing cosmic ray interactions.”

Will the argument ever end? AMS is scheduled to take data for 10 more years, which should enable scientists to whittle down the uncertainties and extend their reach toward higher energies, Ting says. "I think we should be able to reach 1 TeV with good statistics," he says, and that should be enough to eventually settle the dispute. But Gregory Tarlé, an astrophysicist at the University of Michigan, Ann Arbor, says, "I don't think that's a legitimate claim." Higher energy cosmic rays arrive at such a low rate that even quadrupling the data set would leave large statistical uncertainties, he says. So, Tarlé suspects, years from now the AMS results will likely look about as ambiguous they do now.