A stacked analysis of brightest cluster galaxies observed with the Fermi Large Area Telescope

We present the results of a search for high-energy γ-ray emission from a large sample of galaxy clusters sharing the properties of three existing Fermi Large Area Telescope detections (in Perseus, Virgo and A3392), namely a powerful radio source within their brightest cluster galaxy (BCG). From a parent, X-ray flux-limited sample of 863 clusters, we select 114 systems with a core-dominated BCG radio flux above 50/75 mJy (in the National Radio Astronomy Observatory Very Large Array Sky Survey and the Sydney University Molonglo Sky Survey, respectively), stacking data from the first 45 months of the Fermi mission in three energy bands, to determine statistical limits on the γ-ray fluxes of the ensemble of candidate sources. For a >300 MeV selection, the distribution of detection significance across the sample is consistent with that across control samples for significances 4σ signals which are not associated with previously identified γ-ray emission. Modelling of the data in these fields results in the detection of four non-2FGL Fermi sources, though none of these appear to be unambiguously associated with the BCG candidate. Only one is sufficiently close to be a plausible counterpart (RXC J0132.6−0804) and the remaining three appear to be background active galactic nuclei. A search at energies >3 GeV hints at emission from the BCG in A2055, which hosts a BL Lac object. There is no evidence for a signal in the stacked data, and the upper limit derived on the γ-ray flux of an average radio-bright BCG in each band is at least an order of magnitude more constraining than that calculated for individual objects. F1 GeV/F1.4 GHz for an average BCG in the sample is <15, compared with ≈120 for NGC 1275 in Perseus, which might indicate a special case for those objects detected at high energies. The tentative suggestion that point-like beamed emission from member galaxies comprise the dominant bright γ-ray sources in clusters implies searches for evidence of dark matter annihilation or large-scale merger shock signatures, for example, need to account for a significant level of contamination from within each cluster that is both highly stochastic and varies significantly over time.