Measuring the Infrared Background with Gamma Rays
The cosmic microwave background, the remaining glow of the big bang, was recently mapped in great detail, but astronomers have had much more trouble measuring the extragalactic infrared (IR) background, mainly because it’s hard to separate from sources inside our galaxy. Thirty years ago, an indirect approach to the measurement, free of such contamination, was suggested: look at the effects of the IR background photons on gamma rays arriving from distant sources. In the 6 April PRL, a team describes the most detailed analysis of this type, which uses the latest high energy gamma ray observations to set the most stringent, model-independent limits yet proposed for the IR background. They also apply their IR limits to two specific problems–very massive dark matter candidates and the possibility of massive neutrinos.
Within the past decade, astronomers have begun to reliably detect the atmospheric particle showers from very high energy gamma rays in the range where the IR background should have the most effect (above , or ). According to theory, a gamma photon from a distant source should combine with a background IR photon to produce an electron-positron pair with high probability, assuming enough IR photons exist between Earth and the source. The reaction requires a minimum gamma photon energy, and theorists expect a steep reduction in the number of photons reaching Earth with higher energies because this process should absorb them. But the energy of this cutoff is strongly dependent on the density of IR photons, so even the detection of a small number of high energy photons from a source of known distance implies an upper limit on the IR photon density–with an IR background above this limit, the gamma rays would never have arrived at Earth.
One of the greatest difficulties in this approach to studying the IR background is the uncertainty in the shapes of both the IR spectrum and the gamma spectrum at its source. Steven Biller, of Oxford University in the UK, and his colleagues, attempted to address this difficulty in their PRL paper by assuming very little about the spectra. “We were surprised,” recalls Biller, by the stringency of the upper limits they derived from simply assuming a gamma spectrum steeper than and requiring consistency with satellite-based IR observations and recent ground-based gamma ray observations. Biller and his colleagues essentially relied on the satellite data as an upper limit for the IR levels at energies between 0.3 and and used it to scale their predictions for lower IR energies, which were binned into several energy ranges. In the range between 0.03 and , their results reduced the IR upper limit by more than a factor of 10 below the limit based directly on satellite observations.
The analysis agrees with other recent work which suggests that the IR background is not much larger than would be expected from “conventional” sources, such as the stars and dust in galaxies. Perhaps 90% of the Universe is composed of mysterious “dark matter” which we can’t observe directly, but some dark matter candidates would contribute to the IR background, increasing it above the “conventional” level. Biller and his colleagues claim to essentially rule out one of those candidates, called “very massive objects” (VMO’s)–ancient stars up to solar masses that would have become black holes long ago, but would have contributed IR radiation during their lifetimes. The authors raise one caveat to this result, however: If there was a large amount of IR-absorbing dust reradiating at other wavelengths, VMO’s would not be ruled out. They also used their IR background data to limit the parameters for proposed massive neutrinos, which could contribute IR photons through a radiative decay mechanism if the neutrino mass were in a range near 0.1 eV.
A new gamma telescope array, called VERITAS, is scheduled to begin construction within the next several years, and the team eagerly awaits the wider range of energies that will be detectable. “There’s no reason the [IR] limits will not improve,” says Biller.