Disentangling cosmic-ray and dark-matter induced γ-rays in galaxy clusters

¹ Laboratoire de Physique Subatomique et de Cosmologie, Université Joseph Fourier Grenoble 1/CNRS/IN2P3/INPG, 53 avenue des Martyrs, 38026 Grenoble, France
e-mail: dmaurin@lpsc.in2p3.fr
² Aix Marseille Université, CNRS, LAM (Laboratoire d’Astrophysique de Marseille), UMR 7326, 13388 Marseille, France
³ CNRS, IRAP, 14 avenue Edouard Belin, 31400 Toulouse, France
⁴ Université de Toulouse, UPS-OMP, IRAP, 31400 Toulouse, France

Received: 8 February 2012
Accepted: 12 September 2012

Abstract

Context. Galaxy clusters are among the best targets for indirect dark matter detection in γ-rays, despite the large astrophysical background expected from these objects. Detection is now within reach of current observatories (Fermi-LAT or Cerenkov telescopes); however, assessing the origin of this signal might be difficult.

Aims. We investigate whether the behaviour of the number of objects per “flux” bin (log N − log F) and that of the stacked signal could be used as a signature of the dominant process at stake.

Methods. We use the Clumpy code to integrate the signal from decaying or annihilating dark matter and from cosmic rays along the line of sight. We assume the standard Navarro-Frenk-White (NFW) profile for the dark matter density and rely on a parametrised emissivity for the cosmic-ray component. In this context, the consequences of stacking are explored using the MCXC meta-catalogue of galaxy clusters.

Results. We find the value of the slope of the log N − log F power law (or the increase of the signal with the number of stacked objects) to be a clear diagnosis to disentangle decaying dark matter from cosmic-ray induced γ-rays. For dark matter annihilation, depending on the signal boost from the substructures, it is either similar to the cosmic-ray signal (no boost) or similar to the decay case (large boosts). The shift between the brightest object and its followers also depends on the signal origin. For annihilation, this shift and the stacked signal are poorly constrained because of the large uncertainty affecting the boost. We also underline that the angular dependence of the annihilation signal is not universal because of the substructure contribution.