2010ApJ...720..786L

Active galactic nuclei (AGNs) consist of an accretion disk around a supermassive black hole which in turn is surrounded by an obscuring torus of dust and gas. As the resulting geometry of this system affects the observable properties, quantifying isotropic indicators of intrinsic AGN luminosity is important in selecting unbiased samples of AGNs. In this paper, we consider five such proxies: the luminosities of the [O III]λ5007 line, the [O IV]25.89 µm line, the mid-infrared (MIR) continuum emission by the torus, and the radio and hard X-ray (E > 10 keV) continuum emission. We compare these different proxies using two complete samples of low-redshift Type 2 AGNs selected in a homogeneous way based on different indicators: an optically selected [O III] sample and an MIR-selected 12 µm sample. To assess the relative merits of these proxies, we have undertaken two analyses. First, we examine the correlations between all five different proxies, and find better agreement for the [O IV], MIR, and [O III] luminosities than for the hard X-ray and radio luminosities. Next, we compare the ratios of the fluxes of the different proxies to their values in unobscured Type 1 AGNs. The agreement is best for the ratio of the [O IV] and MIR fluxes, while the ratios of the hard X-ray to [O III], [O IV], and MIR fluxes are systematically low by about an order of magnitude in the Type 2 AGNs, indicating that hard X-ray-selected samples do not represent the full Type 2 AGN population. In a similar spirit, we compare different optical and MIR diagnostics of the relative energetic contributions of AGN and star formation processes in our samples of Type 2 AGNs. We find good agreement between the various diagnostic parameters, such as the equivalent width of the MIR polycyclic aromatic hydrocarbon features, the ratio of the MIR [O IV]/[Ne II] emission lines, the spectral index of the MIR continuum, and the commonly used optical emission-line ratios. Finally, we test whether the presence of cold gas associated with star formation leads to an enhanced conversion efficiency of AGN ionizing radiation into [O III] or [O IV] emission. We find that no compelling evidence exists for this scenario for the luminosities represented in this sample (L_bol{&x2248;}10⁹ - 8x10¹¹ L_☉).