A TeV-based Determination of the Local Extragalactic Background Light and its Consistency with Galaxy Counts and Direct Measurements

Abstract

The extragalactic background light (EBL), the cumulative radiation from all extragalactic sources, traces galaxy formation and cosmic evolution. High-energy ${\hat{I}}^3$ rays attenuated via pair production with EBL photons are a powerful probe of the EBL. In this work, we use very-high-energy (VHE; $E_{\hat{I}}^3>100\mathrm{GeV}$) ${\hat{I}}^3$ rays to measure the local EBL intensity and test its consistency with galaxy counts and direct measurements. Our analysis employs a sample of 268 spectra from 45 sources observed with Imaging Atmospheric Cherenkov telescopes. A model-dependent study shows seven EBL templates require only $\le 10\%$ rescaling to fit the observed ${\hat{I}}^3$-ray attenuation. The galaxy-count-anchored model gives the closest match. We then derive template-marginalized TeV optical depths from a representative model subset. We combine them with \textit{Fermi}-LAT GeV measurements to reconstruct the EBL at $z=0$ using empirical and physically motivated models. The two reconstructions agree and follow the integrated galaxy light to within $2$--$3\mathrm{nW}{\mathrm{m}}^{-2}\mathrm{s}{\mathrm{r}}^{-1}$ (typically $<25\%$) over $0.5$--$30\hat{I}¼$m. Both are consistent with low-zodiacal-light observations, including outer solar system and dark cloud measurements. In contrast, the near-IR excess reported by IRTS and CIBER exceeds our reconstructed intensity by $3$--$5\ddot{I}ƒ$, implying an additional $\gtrsim 5$--$10\mathrm{nW}{\mathrm{m}}^{-2}\mathrm{s}{\mathrm{r}}^{-1}$ incompatible with the ${\hat{I}}^3$-ray optical depths. Combined with GeV constraints on EBL evolution to $z\simeq 4$, these TeV optical depths provide a VHE-anchored determination of the local EBL intensity. The agreement with galaxy counts and deep-space measurements indicates that known galaxy populations account for most of the optical and near-IR background, leaving limited room for an additional diffuse component.

Accepted for publication in ApJL, 12 pages, 6 figures