Galaxy properties of type 1 and 2 X-ray selected AGN and comparison among different classification criteria

TL;DR

This study compares host galaxy properties of type 1 and 2 X-ray selected AGNs, demonstrating that SED fitting with X-CIGALE effectively classifies AGN types and reveals differences in host galaxy mass.

Contribution

The paper introduces a comprehensive analysis combining optical to far-infrared SED fitting with X-CIGALE to classify AGN types and compare their host galaxy properties, validating the method against spectroscopic classifications.

Findings

Type 2 AGNs reside in more massive galaxies than type 1s.

X-CIGALE accurately identifies AGN types based on SED fitting.

Optical/mid-IR colour criteria are less effective for low to moderate luminosity AGN identification.

Abstract

We present analyses of host galaxy properties of type 1 and type 2 X-ray selected AGNs in the XMM-XXL field, which have available optical spectroscopic classification. We model their optical to far-infrared spectral energy distributions (SEDs) using the X-CIGALE code. X-CIGALE allows the fitting of X-ray flux and accounts for the viewing angle of dusty torus and the attenuation from polar dust. By selecting matched type 1 and 2 subsamples in the X-ray luminosity and redshift parameter space, we find that both types live in galaxies with similar star formation. However, type 2 AGN tend to reside in more massive systems ($10.87^{+0.06}_{-0.12}\,\rm M_\odot$) compared to their type 1 counterparts ($10.57^{+0.20}_{-0.12}\,\rm M_\odot$). In the second part of our analysis, we compare the spectroscopic classification with that from the SED fitting. X-CIGALE successfully identifies all spectroscopic type 2 sources either by estimating an inclination angle that corresponds to edge on viewing of the source or by measuring increased polar dust in these systems. $\sim 85\%$ of spectroscopic type 1 AGN are also identified as such, based on the SED fitting analysis. There is a small number of sources ($\sim 15\%$ of the sample), that present broad lines in their spectra, but show strong indications of obscuration, based on SED analysis. These, could be systems that are viewed face on and have an extended dust component along the polar direction. The performance of X-CIGALE in classifying AGN is similar at low and high redshifts, under the condition that there is sufficient photometric coverage. Finally, usage of optical/mid-IR colour criteria to identify optical red AGN (${\it{u}}-\rm W3$), suggests that these criteria are better suited for IR selected AGN and their efficiency drops for the low to moderate luminosity sources included in X-ray samples.