Neutrinos in IceCube/KM3NeT as probes of Dark Matter Substructures in Galaxy Clusters

Galaxy clusters are one of the most promising candidate sites for dark matter annihilation. We focus on dark matter with mass in the range 10 GeV - 100 TeV annihilating to muon pairs, neutrino pairs, top pairs, or two neutrino pairs, and forecast the expected sensitivity to the annihilation cross section into these channels by observing galaxy clusters at IceCube/KM3NeT. Optimistically, the presence of dark matter substructures in galaxy clusters is predicted to enhance the signal by 2-3 orders of magnitude over the contribution from the smooth component of the dark matter distribution. Optimizing for the angular size of the region of interest for galaxy clusters, the sensitivity to the annihilation cross section of heavy DM with mass in the range 300 GeV - 100 TeV will be of the order of 10^{-24} cm^3 s^{-1}, for full IceCube/KM3NeT live time of 10 years, which is about one order of magnitude better than the best limit that can be obtained by observing the Milky Way halo. We find that neutrinos from cosmic ray interactions in the galaxy cluster, in addition to the atmospheric neutrinos, are a source of background. We show that significant improvement in the experimental sensitivity can be achieved for lower DM masses in the range 10 GeV - 300 GeV if neutrino-induced cascades can be reconstructed to approximately 5 degrees accuracy, as may be possible in KM3NeT. We therefore propose that a low-energy extension “KM3NeT-Core”, similar to DeepCore in IceCube, be considered for an extended reach at low DM masses.

💡 Research Summary

The paper investigates the potential of using high‑energy neutrino telescopes, IceCube and KM3NeT, to search for dark‑matter (DM) annihilation signals originating from galaxy clusters. Galaxy clusters are attractive targets because they host large amounts of dark matter and, according to ΛCDM simulations, contain abundant sub‑halos (substructures) that can boost the annihilation signal by two to three orders of magnitude relative to the smooth halo component. The authors focus on DM masses between 10 GeV and 100 TeV and consider four annihilation channels: χχ → μ⁺μ⁻, χχ → νν̄, χχ → t t̄, and χχ → 2 ν 2 ν̄. For each channel they compute the neutrino energy spectra using PYTHIA 8, taking into account the full decay chains (especially for the t t̄ channel).

The dark‑matter distribution in a cluster is modeled as the sum of a smooth Navarro‑Frenk‑White (NFW) profile and a population of sub‑halos. The sub‑halo mass function follows dN/dM ∝ M⁻¹·⁹ down to a minimum mass of 10⁻⁶ M⊙, and their spatial distribution is taken to be more concentrated toward the cluster centre. Integrating the squared density over the line of sight yields the J‑factor, which is enhanced by a factor of 10²–10³ when substructures are included.

Neutrino fluxes are then propagated to Earth, and the expected event rates in IceCube and KM3NeT are calculated. The analysis optimizes the region of interest (ROI) for each detector based on the angular size of the cluster and the detector’s point‑spread function. IceCube, with its excellent track reconstruction (≈1° angular resolution) at TeV energies, is most sensitive to high‑mass DM, while KM3NeT benefits from its location in the Northern Hemisphere, allowing longer exposure to many clusters and better performance at lower energies.

Backgrounds consist of atmospheric neutrinos and neutrinos produced by cosmic‑ray interactions with the intra‑cluster medium. The latter becomes dominant at low energies (≲300 GeV). The authors model both components and show that, by tailoring the ROI and employing energy cuts, the signal‑to‑background ratio can be maximized. Sensitivity is derived using a profile‑likelihood method with Poisson statistics, yielding 90 % confidence‑level upper limits on the velocity‑averaged annihilation cross section ⟨σv⟩.

Key results: for DM masses between 300 GeV and 100 TeV, the combined IceCube/KM3NeT exposure of ten years can reach ⟨σv⟩ ≈ 10⁻²⁴ cm³ s⁻¹ for the considered channels, roughly an order of magnitude better than limits obtained from observations of the Milky Way halo. In the lower mass range (10–300 GeV), the sensitivity is limited by the background from intra‑cluster neutrinos, but the authors demonstrate that if cascade events can be reconstructed with an angular accuracy of about 5°, the limits improve dramatically, approaching ⟨σv⟩ ≈ 10⁻²⁵ cm³ s⁻¹.

To achieve such angular precision at low energies, the paper proposes a dedicated low‑energy extension of KM3NeT, analogous to IceCube’s DeepCore, tentatively called “KM3NeT‑Core”. This would involve a denser array of optical modules and enhanced photon‑detection efficiency, enabling better reconstruction of cascade topologies and thus extending the reach of neutrino‑based DM searches to lighter particles.

In conclusion, the study shows that galaxy‑cluster neutrino observations are a powerful probe of dark‑matter substructure and can significantly improve indirect detection limits, especially when sub‑halo boosts are accounted for. The suggested KM3NeT‑Core upgrade would further broaden the accessible DM mass range, making neutrino telescopes competitive with gamma‑ray and cosmic‑ray experiments in the quest to uncover the particle nature of dark matter.