Study of the Gamma-ray Spectrum from the Galactic Center in view of Multi-TeV Dark Matter Candidates

TL;DR

This study investigates whether dark matter annihilation could explain the complex gamma-ray spectrum observed from the Galactic Center, using multi-instrument data and spectral modeling to constrain potential dark matter properties.

Contribution

The paper provides a detailed spectral analysis combining Fermi LAT and HESS data, exploring dark matter models with specific annihilation channels and boost factors, and constrains the parameter space for dark matter explanations.

Findings

Good spectral fits with power-law and pulsar-like spectra plus dark matter contributions.

Dark matter models with annihilation to b-bbar and tau+ tau- channels are marginally compatible with data.

Constraints on dark matter parameters show only a small overlap at high confidence levels.

Abstract

Motivated by the complex gamma-ray spectrum of the Galactic Center source now measured over five decades in energy, we revisit the issue of the role of dark matter annihilations in this interesting region. We reassess whether the emission measured by the HESS collaboration could be a signature of dark matter annihilation, and we use the {\em Fermi} LAT spectrum to model the emission from SgrA*, using power-law spectral fits. We find that good fits are achieved by a power law with an index $\sim 2.5-2.6$, in combination with a spectrum similar to the one observed from pulsar population and with a spectrum from a $\gsi10$ TeV DM annihilating to a mixture of $b{\bar b}$ and harder $\tau^+ \tau^-$ channels and with boost factors of the order of a hundred. Alternatively, we also consider the combination of a log-parabola fit with the DM contribution. Finally, as both the spectrum of gamma rays from the Galactic Center and the spectrum of cosmic ray electrons exhibit a cutoff at TeV energies, we study the dark matter fits to both data-sets. Constraining the spectral shape of the purported dark matter signal provides a robust way of comparing data. We find a marginal overlap only between the 99.999% C.L. regions in parameter space.