Energetic radiation pulsing from the belly of the Milky Way is the clearest signal yet of dark matter. That is according to a pair of astrophysicists in the US who reach this conclusion after scrutinising the public data collected by NASA's orbiting Fermi Observatory. "I certainly think it's the best evidence we've seen so far," says Dan Hooper, one half of the team, based at the University of Chicago.

It is a huge claim because for over 70 years astrophysicists have debated the existence of dark matter, which is thought to make up 80% of the universe's mass, yet they have failed to gather any definitive evidence, either direct or indirect, for its existence. But with several hints for dark matter published in recent years – all received with scrutiny by the wider astrophysics community – the US pair will have a hard time convincing others that their signal is what they think it is.

Hooper and his colleague Lisa Goodenough of New York University have analysed the spectra of gamma rays coming from the centre of our galaxy, as collected by the Large Area Telescope onboard the Fermi observatory. Although dark matter does not couple to light, it should annihilate with itself to produce gamma rays, and the amount of annihilation should increase rapidly towards the galactic centre as dark-matter density increases.

Excess gamma rays

Last year Hooper and Goodenough compared the Fermi spectra of gamma rays with a simple computer model of dark matter, and suggested that an excess of gamma rays coming from the galactic centre might be evidence of dark-matter annihilation. At that time other researchers weren't convinced because there were other possible origins for the signal, such as high-energy photons striking interstellar gas. In their latest analysis, however, Hooper and Goodenough have tried to allay these concerns using a far more complex methodology that looks at specific components making up the background of gamma rays.

The US pair break down the gamma-ray background into three parts: a narrow emission from the galaxy's disc; an emission from known point sources; and a spherical or "bulge" emission around the galactic centre. According to their model, no matter what parameters one chooses for dark matter, there should always be a threshold within the bulge emission where dark-matter annihilation begins to outshine other gamma-ray sources. This is because – unlike other sources – emission from dark-matter annihilation follows a square law, so that doubling the density increases the annihilation four-fold.

Hooper and Goodenough examined the Fermi spectra at many regions inside the gamma-ray bulge, and found the data always matched the model's prediction of normal emission – except right at the galactic centre. Here, in a narrow region spanning less than one-quarter of a degree, the emission was far stronger than the model predicted, and had a more lopsided spectrum. Those characteristics, the US pair claims, point to a dark-matter particle – a weakly interacting massive particle, or WIMP – in a mass range of 7.3–9.2 GeV.

A familiar mass

This light mass is partly what lends the analysis credence. For years physicists working on the DAMA experiment in Italy claim to have found WIMPs colliding with sodium-iodide nuclei, while those working on the CoGeNT collaboration in the US have tentatively revealed similar WIMP signals coming from germanium detectors – and many believe the only way to reconcile these signals is to assume a WIMP with a mass around 8 GeV.

"Until I had seen this latest paper from Hooper and Goodenough, I was kind of thinking with the light WIMP scenario – nah," says Alex Murphy, a particle astrophysicist who works on the ZEPLIN-III dark-matter experiment in the UK. "But now I've seen it, I'm starting to think – hmm, maybe. Perhaps now we should be looking at other ways to confirm or disprove this proposal."

Murphy voices scepticism about the strength of the claim, however, because he is not convinced Hooper and Goodenough understand the idiosyncrasies of the Fermi instrumentation sufficiently well. Although the Fermi team has published its own preprint revealing an excess of gamma rays near the galactic centre, it has so far stopped short of interpreting this as dark matter.

Still prone to misinterpretation

Ronaldo Bellazzini, the principal investigator on Fermi's Italian team, warns that Hooper and Goodenough's analysis of the galactic centre could still be prone to misinterpretation. "Unfortunately, this region, and whatever [Fermi] observes along the line of sight to it, is rich with astrophysical sources that can mimic signals similar to dark-matter annihilation, like pulsars and supernovae remnants" he says.

Meanwhile, Michael Kuhlen, a dark-matter theorist at the University of California at Berkeley, believes there is "probably a good reason" why the Fermi collaboration has held back from making conclusions on the gamma-ray excess. "They're certainly aware of it, but probably just haven't been able to convince themselves that they fully understand the instrument's behaviour, or the backgrounds, or the kinds of possible astrophysical sources that could produce the signal," he says.

But Kulen adds: "Really they're just trying to stir the pot, and get people to seriously consider the possibility that Fermi may have already detected a dark-matter annihilation signal. This is a good thing."

A preprint of the paper is available at arXiv: 1010.2752.