What can be learnt from UHECR anisotropies observations - I. Large-scale anisotropies and composition features

International audienceContext. In recent years, evidence for an anisotropic distribution of ultra-high-energy cosmic rays (UHECRs) has been claimed, notably a dipole modulation in right ascension has been reported by the Auger collaboration above the 5σ significance threshold.Aims. We investigate the implications of the current data regarding large-scale anisotropies, including higher order multipoles, and we examine to what extent they can be used to shed some light on the origin of UHECRs and constrain the astrophysical and/or physical parameters of the source scenarios. We investigate the possibility of observing an associated anisotropy of the UHECR composition and discuss the potential benefit of a good determination of the composition and of the separation of the different nuclear components. We also discuss the interest and relevance of observing the UHECR sky with larger exposure future observatories.Methods. We simulated realistic UHECR sky maps for a wide range of astrophysical scenarios satisfying the current observational constraints, taking into account the energy losses and the photo-dissociation of the UHE protons and nuclei, as well as their deflexions by intervening magnetic fields. We investigated scenarios in which the UHECR source distribution follows that of the galaxies in the Universe (with possible biases), varying the UHECR source composition and spectrum, as well as the source density and the magnetic field models. For each of them, we simulated 300 realizations of independent datasets corresponding to various assumptions for the statistics and sky coverage, and we applied similar analyses as those used by the Auger collaboration for the search of large-scale anisotropies.Results. We find the following. First, reproducing the amplitude of the first-order (dipole) anisotropy observed in the Auger data, as well as its evolution as a function of energy, is relatively easy within our general assumptions. Second, this general agreement can be obtained with different sets of assumptions on the astrophysical and physical parameters, and thus it cannot be used, at the present stage, to derive strong constraints on the UHECR source scenarios or draw model-independent constraints on the various parameters individually. Third, the actual direction of the dipole modulation reconstructed from the Auger data, in the energy bin where the signal is most significant, appears highly unnatural in essentially all scenarios investigated, and this calls for their main assumptions to be reconsidered, either regarding the source distribution itself or the assumed magnetic field configuration, especially in the Galaxy. Fourth, the energy evolution of the reconstructed dipole direction contains potentially important information, which may become constraining for specific source models when larger statistics is collected. Fifth, for such high-statistics datasets, most of our investigated scenarios predict a significant quadrupolar modulation, especially if the light component of UHECRs can be extracted from the all-particle dataset. Sixth, except for protons, the energy range in which the GZK horizon strongly reduces is a key target for anisotropy searches for each given nuclear species. Seventh, although a difference in the average composition of the UHECRs in regions having a different count rate is naturally expected in our models, it is unlikely that the composition anisotropy recently reported by Auger can be explained by this effect, unless the reported amplitude is a strong positive statistical fluctuation of an intrinsically weaker signal.Key words: astroparticle physics / cosmic rays / magnetic field