Prospects for dark matter observations in dwarf spheroidal galaxies with the Cherenkov Telescope Array Observatory

TL;DR

The study forecasts Cherenkov Telescope Array Observatory (CTAO) sensitivity to gamma-ray signals from dark matter annihilation or decay in Milky Way dwarf spheroidal galaxies (dSphs). By unifying photometric/spectroscopic data with Jeans-Jeans+CLUMPY modeling, it computes robust astrophysical factors ($J_{\rm ann}$, $J_{\rm dec}$) and identifies eight promising dSph targets. Through a CTAO-ready analysis pipeline, including template-background background modeling and profiling over $J$-factor uncertainties, the work derives 95% CL upper limits on $\langle\sigma v\rangle$ (down to ~5×10^{-25} cm^3 s^{-1} for multi-TeV DM) and lower limits on DM lifetimes (up to ~10^{27} s) and discusses combined-target strategies. The results show that dSphs offer strong, relatively background-free DM probes at TeV scales, with uncertainties dominated by DM content modeling; upcoming surveys will be crucial to improve these constraints and fully exploit CTAO’s potential.

Abstract

The dwarf spheroidal galaxies (dSphs) orbiting the Milky Way are widely regarded as systems supported by velocity dispersion against self-gravity, and as prime targets for the search for indirect dark matter (DM) signatures in the GeV-to-TeV $γ$-ray range owing to their lack of astrophysical $γ$-ray background. We present forecasts of the sensitivity of the forthcoming Cherenkov Telescope Array Observatory (CTAO) to annihilating or decaying DM signals in these targets. An original selection of candidates is performed from the current catalogue of known objects, including both classical and ultra-faint dSphs. For each, the expected DM content is derived using the most comprehensive photometric and spectroscopic data available, within a consistent framework of analysis. This approach enables the derivation of novel astrophysical factor profiles for indirect DM searches, which are compared with results from the literature. From an initial sample of 64 dSphs, eight promising targets are identified -- Draco I, Coma Berenices, Ursa Major II, Ursa Minor and Willman 1 in the North, Reticulum II, Sculptor and Sagittarius II in the South -- for which different DM density models yield consistent expectations, leading to robust predictions. CTAO is expected to provide the strongest limits above $\sim$10 TeV, reaching velocity-averaged annihilation cross sections of $\sim$5$\times$10$^{-25}$ cm$^3$ s$^{-1}$ and decay lifetimes up to $\sim$10$^{26}$ s for combined limits. The dominant uncertainties arise from the imprecise determination of the DM content, particularly for ultra-faint dSphs. Observation strategies are proposed that optimise either deep exposures of the best candidates or diversified target selections.

Figures (19)

• Figure 1: DM $\gamma$-ray spectra as a function of the DM mass for pure WIMP annihilation into specific channels, obtained with gammapygammapy and based on the PPPC parametrization by Cirelli:2010xx. We compare three values of $m_{\mathrm{DM}}=0.5,5,50$ TeV ( gray solid lines) to show that the photon yield and spectral differences as a function of the mass are minor, with the exception of the $W^+W^-$ where electroweak loop corrections and final-state radiation significantly affect the high-energy tail of the spectrum. The $b\bar{b}$ and $\tau^+\tau^-$ channels are selected as representative examples of the theoretical DM photon yields, the former being the softest with a signal peak at $m_{\mathrm{DM}}/20$ and the latter being the hardest with a peak at $m_\mathrm{DM}/3$. In the case of DM decay, the major difference is that the cut-off happens at $m_\mathrm{DM}/2$ -- while preserving the spectral shape.
• Figure 2: Sky distribution of known MW satellites and Local Group dSphs, superimposed to the Fermi-LAT $\gamma$-ray background (credits: NASA/DOE/ Fermi-LAT Collaboration). All of the targets from \ref{['tab:dsph_excluded']} ( gray dots) are reported, along with the sources from \ref{['tab:dsph_firstcut']} (i.e. those passing the first selection cut on distance; blue triangles) and those objects passing the second selection (availability of good spectral and photometric data; green squares). The optimal targets ( red stars) are highlighted with symbols of increasing size, proportional to the value of their $\log{J_{\rm ann}(<0.1^\circ)}$ (see \ref{['tab:dsph_jfactors']}).
• Figure 3: DM density profiles of the 14 optimal dSphs for both cored bur95 and cuspy models ein65, along with the corresponding uncertainties at 68% CL ( pink and gray shaded areas). In all panels, the tidal radius (see text for details) for the Einasto ( blue solid line) and Burkert profile ( blue dashed line) is also indicated, along with the typical uncertainty ( blue dotted line) for the representative cases of CBe and RetII.
• Figure 4: Astrophysical factors for DM annihilation $J_{\rm ann}(<\alpha_{\rm int})$ as functions of the integration angle $\alpha_{\rm int}$ (or the equivalent integration distance from the dSph centroid $r_{\rm int}$) for the best Northern and Southern dSphs. In all panels, the median astrophysical factor profiles for both cuspy (Einasto; black solid lines) and cored (Burkert; purple dashed lines) DM density profiles are plotted alongside the relative uncertainties at 1$\sigma$ CL ( gray shaded areas and pink shaded areas). The integration angles corresponding to an instrumental PSF of $0.06^\circ$ ( blue dotted lines) are also indicated. Each profile is truncated at the corresponding dSph tidal radius for the Einasto profile (see \ref{['eq:tidal_radius']} and \ref{['tab:dsph_parameters']}). The Burkert profiles for GruII, Seg1, SgrI and TriII are not reported because no finite integration could be obtained.
• Figure 5: Same as \ref{['fig:dsph_profile_ann']}, but for the case of DM decay. In this case, also the Burkert profile for RetII is excluded due to a non-finite integration.