Impacts of Voids, Line of Sight Interactions, and Local Emission Environment on Detectability of Gamma-Ray AGN

Abstract

Cosmic voids may have novel affects on the propagation of high-energy photons. We consider the fraction of the line of sight that intersect voids (termed \enquote{voidiness}). A previous study showed that active galactic nuclei (AGN) detected by \textit{Fermi} Large Area Telescope (LAT) lie along voidier lines of sight than redshift-matched populations of Sloan Digital Sky Survey (SDSS) optically detected quasars in the redshift range from $0.4 \leq z < 0.7$. We explore this difference and various astrophysical explanations for it. Weaker intergalactic magnetic fields in voids would naturally enhance the gamma-ray cascading flux within the \textit{Fermi}-LAT point-spread function. We find that line-of-sight interactions increasing the flux in the \textit{Fermi}-LAT energy band by $\sim$0.1\% per Mpc of void traversed may be sufficient to result in the observed difference in voidiness distributions. Voidiness comparisons between SDSS QSO and AGN detected by imaging atmospheric Cherenkov telescopes at very-high-energies (VHE) do not yield any conclusive statement, likely because of the limited VHE sample size, and therefore are inconclusive about the role of possibly weaker extragalactic background light within voids. Finally, we measure that $28 \pm 3 \%$ of gamma-ray detected sources exist within a void (consistent with random mock populations) compared to $19.1 \pm 0.3 \%$ of SDSS quasars. We do not find any significant local void effect for gamma-ray sources that would explain the voidiness difference between \textit{Fermi}-LAT gamma-ray and SDSS QSO sources. These results suggest that the observed difference in voidiness distributions may be due to line-of-sight interactions rather than the local emission environment of gamma-ray AGN.

Figures (6)

• Figure 1: CDFs of the intersecting void distance distributions of 4LAC sources (blue) and the mean intersecting void distance of redshift-matched SDSS QSOs (in red) along with one and two standard deviations from the mean. The left panel is for sources in the redshift range $0.1 \leq z < 0.4$ (containing 160 sources), the center panel for $0.4 \leq z < 0.7$ (containing 143 sources), and the right panel for $0.1 \leq z <0.7$ (containing the total 303 sources).
• Figure 2: Top row displays results for all 4LAC sources in the sample; bottom row for only BL Lac sources. Left: Significance of the difference between the voidiness distributions of 4LAC sources and SDSS QSOs (blue) and the KS statistics comparing the voidiness distributions of the two populations (purple). Right: For $z \geq 0.4$, the measured significance of the difference in voidiness distributions (blue) with the significance for 500 4LAC and SDSS QSO populations randomly removing sources (grey) to match the true population size in each redshift range. The median (black) and 1 sigma contours (purple) are also shown.
• Figure 3: Median KS p-values resulting from the KS comparison of voidiness distributions of cascade-corrected 4LAC gamma-ray sources with redshift-matched SDSS QSOs as a function of the assumed flux correction percentage per Mpc of void intersection, shown for both the nearby redshift range from $0.1 \leq z < 0.4$ (in blue) and the distant redshift range from $0.4 \leq z < 0.7$ (in green). The number of sources which remain after the removal of sources which fall below the sensitivity range of LAT is noted on the top axis for $0.4 \leq z < 0.7$. The gray band corresponds to the region where the voidiness distributions in each redshift range for 4LAC and SDSS QSOs are consistent within $2\sigma$, which is for a cascade correction of approximately $(0.10 \pm 0.02)\%$ per Mpc of void intersection. Uncertainty is estimated based on the region in which the two populations have consistent voidiness distributions in each redshift range.
• Figure 4: CDFs of the voidiness distributions in each redshift range for 4LAC sources (in blue) versus populations of redshift-matched SDSS QSO populations (in red) with contours for one and two standard deviations from the mean. From left to right the observed flux is cascade corrected by $0\%, ~0.1\%, \text{and} ~1\%$ per Mpc of void. The number of sources within each redshift range is provided in the upper left corner of each panel. We note that while cascade emission increases the observed flux compared to the emitted (intrinsic) GeV flux, to infer the detectability of AGN based on emitted flux alone, we reduce the observed flux.
• Figure 5: CDF of the voidiness distributions of 4LAC VHE-detected sources (in blue) and median voidiness of redshift-matched SDSS QSOs (in red), with the one and two standard deviation contours. The corresponding KS statistic and p-value are 0.23 and 0.31, respectively.