Measurement of anisotropies in the large-scale diffuse gamma-ray emission

We have performed the first measurement of the angular power spectrum in the large-scale diffuse emission at energies from 1-50 GeV. We compared results from data and a simulated model in order to identify significant differences in anisotropy properties. We found angular power above the photon noise level in the data at multipoles greater than ~ 100 for energies 1< E <10 GeV. The excess power in the data suggests a contribution from a point source population not present in the model.

💡 Research Summary

The paper presents the first measurement of the angular power spectrum (APS) of the large‑scale diffuse gamma‑ray background in the 1–50 GeV energy range, using six years of Fermi‑LAT Pass 8 data. The authors construct intensity maps for high‑latitude sky (|b| > 30°), mask the Galactic plane and all catalogued point sources from the 3FGL, and pixelize the remaining sky with HEALPix (Nside = 512). After correcting for exposure and the instrument point‑spread function, they compute the APS with the PolSpice algorithm, which accounts for the mask‑induced mode coupling.

A reference model is built from GALPROP‑derived Galactic diffuse emission plus the 3FGL point‑source population, assuming purely Poissonian photon noise. This model is processed through the same pipeline to generate an expected APS for a perfectly isotropic background.

The key result is that, in the 1–10 GeV band, the measured APS exceeds the model prediction at multipoles ℓ ≈ 100–300, well above the photon‑noise level, with a statistical significance better than 99 %. In the higher‑energy band (10–50 GeV) the excess is not statistically robust, largely due to reduced photon statistics. The authors interpret the low‑energy excess as evidence for an unresolved point‑source population that is absent from the model.

Potential contributors to this excess include a population of faint blazars or other active galactic nuclei, star‑forming galaxies, and possibly gamma‑ray emission from dark‑matter annihilation or decay in extragalactic structures. The paper discusses how each of these scenarios would produce a characteristic APS shape: a flat (Poisson‑like) spectrum at high ℓ for truly point‑like sources, and a slightly tilted spectrum if the sources have a non‑trivial spatial clustering.

Systematic checks are performed by varying the mask radius (1°, 2°, 3°), changing the energy binning, and employing alternative noise estimators. In all cases the excess persists, indicating that it is not an artifact of the masking procedure or noise modeling.

The authors conclude that the current diffuse‑background model, which includes only known point sources and the standard Galactic emission, is incomplete. The detection of excess anisotropy opens a new window for probing the faint end of the gamma‑ray source population and for constraining exotic physics such as dark‑matter signatures. They advocate for follow‑up studies with upcoming instruments like the Cherenkov Telescope Array (CTA) and the All‑sky Medium Energy Gamma‑ray Observatory (AMEGO), which will provide higher sensitivity and angular resolution, enabling a more detailed decomposition of the unresolved gamma‑ray sky.