Multi-Component Dark Matter as a Solution to the Galactic Center GeV Excess

TL;DR

This study investigates multi-component dark matter models to explain the Galactic Center Excess, finding that two-component models with exclusive annihilation channels provide a better fit and may indicate a complex dark sector.

Contribution

It introduces a systematic analysis of multi-component dark matter scenarios, demonstrating their improved fit to the GCE and potential to resolve existing tensions with constraints.

Findings

Two-component dark matter models are statistically favored for explaining the GCE.

Exclusive annihilation channels into specific final states improve the fit.

Multi-component models suggest a light-plus-heavy mass hierarchy in dark matter.

Abstract

The Galactic Center Excess (GCE) is a compelling signature of dark matter annihilation, but its spectral morphology is difficult to reconcile with the traditional paradigm of a single particle species. In this work, we perform a systematic investigation of multi-component dark matter sectors, exploring scenarios with two ($N=2$) and three ($N=3$) distinct particle species while considering both exclusive and mixed annihilation channels. Using the Akaike Information Criterion (AIC) to rigorously penalize model complexity, we find that the GCE data statistically favors an $N=2$ scenario where each dark matter component annihilates exclusively into a single final state. Our results reveal that the preferred solutions naturally follow a light-plus-heavy mass hierarchy, and that specific final states such as $t\bar{t}$, $ZZ$, and $hh$, which are individually unable to explain the excess are effectively ``resurrected'' by the improved morphological fit provided by the multi-component framework. Furthermore, we show that these scenarios may mitigate the tension with current constraints, reaching compatibility within existing uncertainties. Our results suggest that the GCE may be the first evidence of a diverse dark sector, favoring a multi-scale solution over the minimal WIMP paradigm.