Gamma-Ray Observations of Galaxy Clusters Strongly Constrain Dark Matter Annihilation in Prompt Cusps

TL;DR

The paper addresses how prompt cusps formed in the early Universe influence present-day dark matter annihilation signals. It develops flux and morphology models for prompt cusps in seven galaxy clusters and tests them against 15 years of Fermi-LAT gamma-ray data using extended DM templates and a likelihood framework. The analysis finds no significant excess, deriving strong upper limits that exclude the thermal relic cross-section for $m_{\text{DM}} \lesssim 200$ GeV in the $b\bar{b}$ channel, with even stronger constraints than dwarf spheroidals or the IGRB in this mass range. These results underscore the utility of galaxy clusters in indirect DM searches when prompt cusps are considered and provide a sharper view on the viability of thermal WIMPs and related interpretations of Galactic Center signals.

Abstract

Thermal dark matter models generically include the prompt creation of highly-concentrated dark matter cusps in the early Universe. Recent studies find that these cusps can survive to the present day, as long as they do not fall into extremely dense regions of baryonic structure. In this work, we build models of dark matter annihilation within the prompt cusps that reside in galaxy clusters, showing that they dominate the total $γ$-ray annihilation signal. Using 15 years of Fermi-LAT data, we find no evidence for a $γ$-ray excess from these sources, and set strong constraints on annihilating dark matter. These constraints generically rule out the thermal annihilation cross-section to the $b\bar{b}$ channel for dark matter masses below $\sim$200~GeV.

Figures (9)

• Figure 1: Upper limits on the dark matter annihilation cross-section to a $b\bar{b}$ final state in the Virgo cluster. Our limits (magenta solid) rule out the thermal annihilation cross-section (black dashed) for dark matter masses below 200 GeV Steigman:2012nb, exceeding constraints from the stacking of dwarf spheroidal galaxies McDaniel:2023bju (blue dashed) and from the IGRB Delos:2023ipo (orange dashed). These results are in tension with dark matter explanations for the Galactic Center Excess (green region) Calore:2014xka. The light-shaded region indicates the excluded parameter space.
• Figure 2: Spatial extent of each cluster's annihilation signal. For each angle $\theta$ from the cluster center, we show the fractional contribution to the $J$ factor that comes from angles less than $\theta$.
• Figure 3: Cluster $J$ factors, shown as a function of the dark matter mass. Higher particle masses correspond to colder dark matter, so the prompt cusps are smaller and form at earlier times. The earlier formation makes these cusps more internally dense, resulting in a higher annihilation rate.
• Figure 4: Comparing different contributions to the annihilation signal of the Virgo cluster. As a function of the angle $\theta$ from the cluster center, we show the contribution $\mathrm{d} J/\mathrm{d}\ln\theta = 2\pi\theta^2 \mathrm{d} J/\mathrm{d}\Omega$ of each logarithmic angular interval to the integrated $J$ factor. The light teal solid curve shows the net predicted $J$ factor, which we use for our analysis. The navy dotted curve shows the $J$ factor without accounting for tidal stripping of prompt cusps (5--10% higher). The magenta dashed curve shows the $J$ factor from the smooth halo alone, neglecting contributions from cusps altogether. For comparison, the two gray lines show the contribution from subhalos (without prompt cusps) estimated according to two commonly used extrapolations (see the text).
• Figure 5: Upper limits on the annihilation cross-section from our seven selected galaxy clusters as a function of dark matter mass, based on the extended source models discussed in Section \ref{['sec:signal']}. We consider three annihilation channels: $b\bar{b}$ (magenta), $\tau^+\tau^-$ (navy), and $W^+W^-$ (blue). We compare our results (solid lines) with previous upper limits from the studies of the IGRB (dotted lines), from Ref. Delos:2023ipo. The canonical thermal annihilation cross-section is shown as black dashed line Steigman:2012nb. The light-shaded regions represent the parameter space excluded by galaxy cluster observations, while the hatched regions indicate exclusions based on the IGRB analysis. We find that Virgo constraints are typically the strongest, ruling out dark matter annihilating at the thermal cross-section for masses up to 200 GeV, 88 GeV, and 147 GeV for annihilations to $b\bar{b}$, $\tau^+\tau^-$, and $W^+W^-$, respectively.