Insights into Extragalactic Background Light constraints with MAGIC archival data

TL;DR

This study investigates how assumptions about intrinsic gamma-ray spectra affect gamma-ray–based constraints on the Extragalactic Background Light (EBL). Using an open-source Monte Carlo framework and MAGIC archival data, the authors show that the usual Wilks' theorem-based uncertainties can be underestimated when systematics are present, and they quantify the resulting broader uncertainties. To mitigate spectral-model dependence, they introduce two methods: a generic concave Multiply Broken Power-Law (MBPWL) fit and a concave-EBL approach that isolates the EBL wiggle in the transmissivity, both designed to extract robust constraints from limited data. The work highlights the need for improved energy resolution and statistics (as anticipated with next-generation VHE facilities) to leverage these methods for tighter, more reliable EBL constraints.

Abstract

The Extragalactic Background Light (EBL) is the accumulated light emitted throughout the history of the universe, spanning the UV, optical, and IR spectral ranges. Stars and dust in galaxies are expected to be the main source of the EBL. However, recent direct measurements performed beyond Pluto's orbit (less affected by foregrounds than those performed from the Earth) hint at an EBL level in the optical band larger than the one expected from the integrated contribution of known galaxy populations. One approach that could solve this controversy uses Very High Energy (VHE) photons coming from sources at cosmological distances. These photons can interact with the EBL producing electron-positron pairs, a process that leaves an imprint on the observed VHE spectrum. This technique, however, requires assumptions on the intrinsic spectrum of the source, which can compromise the robustness of EBL constraints. In this contribution, we used Monte Carlo simulations and archival data of the MAGIC telescopes to study the impact that the assumptions adopted in the literature have on the estimate of the EBL density, and how using more generic ones would modify the results. Our results show how the EBL density constraints obtained highly depend on the intrinsic spectral shape assumed for the source. We have studied two different methods to reduce the assumptions on the intrinsic spectral shape to get more robust results. This will be especially important for upcoming with new VHE facilities, where systematic uncertainties are expected to play a more significant role compared to statistical ones.

Figures (5)

• Figure 1: Results from $10^4$ realizations of the simulation for the 15 Mrk421 spectra with the D11 model. In red the results of the real data fits.
• Figure 2: Cumulative distribution function of the simulation compared to a $\chi^2$ distribution. The vertical red line shows the point where the CDF equals $68.27\%$
• Figure 3: Combined profile likelihood of the 15 Mrk421 real data spectra. Uncertainties computed using Wilks' theorem are shown in light green and uncertainties obtained using our simulations are shown in red.
• Figure 4: Median of the profile likelihood results for 1ES1011+496 simulated by a PWL and fitted with a LP and a MBPWL using D11 EBL model. In dotted lines the region containing 68% of the realizations around the median.
• Figure 5: Profile likelihood of the 1ES1011+496 data fitted by a PWL and LP using the concave EBL method.