The Origin of Gamma-Rays from Globular Clusters

Fermi has detected gamma-ray emission from eight globular clusters. We suggest that the gamma-ray emission from globular clusters may result from the inverse Compton scattering between relativistic electrons/positrons in the pulsar wind of MSPs in the globular clusters and background soft photons including cosmic microwave/relic photons, background star lights in the clusters, the galactic infrared photons and the galactic star lights. We show that the gamma-ray spectrum from 47 Tuc can be explained equally well by upward scattering of either the relic photons, the galactic infrared photons or the galactic star lights whereas the gamma-ray spectra from other seven globular clusters are best fitted by the upward scattering of either the galactic infrared photons or the galactic star lights. We also find that the observed gamma-ray luminosity is correlated better with the combined factor of the encounter rate and the background soft photon energy density. Therefore the inverse Compton scattering may also contribute to the observed gamma-ray emission from globular clusters detected by Fermi in addition to the standard curvature radiation process. Furthermore, we find that the emission region of high energy photons from globular cluster produced by inverse Compton scattering is substantially larger than the core of globular cluster with a radius >10pc. The diffuse radio and X-rays emitted from globular clusters can also be produced by synchrotron radiation and inverse Compton scattering respectively. We suggest that future observations including radio, X-rays, and gamma-rays with energy higher than 10 GeV and better angular resolution can provide better constraints for the models.

💡 Research Summary

The paper addresses the puzzling detection by the Fermi Large Area Telescope of GeV‑range γ‑ray emission from eight Galactic globular clusters (GCs). While the conventional explanation attributes this emission to curvature radiation from millisecond pulsars (MSPs) residing in the clusters, the authors argue that curvature radiation alone cannot reproduce the observed spectra and luminosity trends. They propose an additional, possibly dominant, contribution from inverse Compton scattering (ICS) of relativistic electrons and positrons (e±) that are injected by the pulsar winds of MSPs.

In their model, the e± population is assumed to follow a power‑law energy distribution and to diffuse outward from the cluster core to distances exceeding ∼10 pc. During this propagation the particles encounter several ambient soft‑photon fields: (1) the cosmic microwave background (CMB) or relic photons, (2) Galactic infrared (IR) photons from interstellar dust, (3) Galactic starlight in the optical/near‑IR band, and (4) the intrinsic stellar radiation field of the cluster itself. Each of these photon components provides a target for up‑scattering by the relativistic e±, producing γ‑rays with energies in the Fermi band.

The authors calculate the expected IC spectra for each GC, varying the dominant photon field. For 47 Tucanae they find that any of the three external photon fields (CMB, Galactic IR, or Galactic starlight) can reproduce the observed γ‑ray spectrum equally well, reflecting the relatively high e± density and the broad range of target photon energies present in this cluster. In contrast, the other seven clusters (including NGC 6624, NGC 6397, Terzan 5, etc.) are best fitted when Galactic IR or Galactic starlight photons dominate the scattering, consistent with their locations near the Galactic plane where the IR and optical photon energy densities far exceed the CMB.

A key empirical result is the strong correlation between the measured γ‑ray luminosity (Lγ) and the product of the stellar encounter rate (Γ) and the energy density of the ambient soft‑photon field (u). The encounter rate is a proxy for the number of MSPs formed in a cluster, while u quantifies the target photon density for IC scattering. The linear relationship Lγ ∝ Γ × u suggests that the γ‑ray output is jointly regulated by the MSP population and the surrounding radiation environment, supporting the IC scenario.

Beyond the GeV band, the model predicts that the same e± population should generate extended synchrotron radio emission (from magnetic fields in the cluster halo) and higher‑energy IC emission detectable in hard X‑rays and >10 GeV γ‑rays. Because the IC scattering region extends well beyond the core radius—potentially >10 pc—the resulting γ‑ray source would appear spatially extended if observed with sufficient angular resolution.

The paper concludes with a set of observational recommendations. High‑resolution radio interferometry (e.g., with the Square Kilometre Array) could map the predicted diffuse synchrotron halo. Deep X‑ray observations (e.g., with Athena) could detect the IC‑produced hard X‑ray component. Finally, next‑generation very‑high‑energy γ‑ray facilities such as the Cherenkov Telescope Array (CTA) could probe the >10 GeV tail of the spectrum and test the spatial extension of the emission. Together, these multi‑wavelength data would allow a decisive discrimination between pure curvature‑radiation models and the hybrid curvature + inverse‑Compton scenario advocated here, thereby advancing our understanding of particle acceleration and high‑energy radiation processes in dense stellar systems.