2018ApJ...861...27M

We conduct a stacking analysis using the combination of 1.4 GHz detections in the NRAO VLA Sky Survey (NVSS) and Planck all-sky maps to estimate the differential source counts down to the few 100 µJy level at 30, 44, 70, and 100 GHz. Using these source count estimates, we are able to measure the integrated extragalactic background light from discrete sources at these frequencies for comparison with the fit to the total radio sky measurements from ARCADE 2. By integrating down to a 1.4 GHz flux density of ≃2 µJy, we measure integrated, extragalactic brightness temperatures from discrete sources of 105.63 ± 10.56 mK, 21.76 ± 3.09 µK, 8.80 ± 0.95 µK, 2.59 ± 0.27 µK, and 1.15 ± 0.10 µk at 1.4, 30, 44, 70, and 100 GHz, respectively. Our measurement at 1.4 GHz is slightly larger than previous measurements, most likely due to using NVSS data compared with older interferometric data in the literature, but it still remains a factor of ≃4.5 below that required to account for the excess extragalactic sky brightness measured at 1.4 GHz by ARCADE 2. The fit to ARCADE 2 total extragalactic sky brightness measurements is also a factor of ≃8.6, 6.6, 6.2, and 4.9 times brighter than what we estimate from discrete sources at 30, 44, 70, and 100 GHz, respectively. The extragalactic sky spectrum (i.e., T_b ∝ ν^β) from discrete sources appears to flatten with increasing frequency, having a spectral index of β = -2.82 ± 0.06 between 1.4 and 30 GHz, flattening to β = -2.39 ± 0.12 between 30 and 100 GHz. We estimate that the spectral flattening most likely arises from a combination of gigahertz-peaked sources and the hardening of the spectra of radio-detected sources at higher frequencies, particularly at faint flux densities. However, the precise origin of a hard component of energetic electrons responsible for the emission remains unclear.