Compton thick AGN in the NuSTAR era

TL;DR

This study analyzes 30 candidate Compton thick AGN using NuSTAR data, revealing that nearly half are actually Compton thin, which lowers the estimated fraction of true Compton thick AGN in the local universe.

Contribution

It provides a refined estimate of the true fraction of Compton thick AGN by incorporating NuSTAR data, highlighting the importance of high-quality spectra for accurate classification.

Findings

47% of candidate CT-AGN are reclassified as CT-thin with NuSTAR data

The overall observed fraction of CT-AGN is lower than previous estimates, around 4%

Spectral quality impacts the accuracy of N_H measurements and classification

Abstract

We present the 2-100 keV spectral analysis of 30 candidate Compton thick (CT-) active galactic nuclei (AGN) selected in the Swift-BAT 100-month survey. The average redshift of these objects is $\langle z\rangle\sim$0.03 and they all lie within $\sim$500 Mpc. We used the MyTorus (Murphy et al. 2009) model to perform X-ray spectral fitting both without and with the contribution of the NuSTAR data in the 3-50 keV energy range. When the NuSTAR data are added to the fit, 14 out of 30 of these objects (47% of the whole sample) have intrinsic absorption N$_{\rm H}<10^{24}$ cm$^{-2}$ at the $>$3$\sigma$ confidence level, i.e., they are re-classified from Compton thick to Compton thin. Consequently, we infer an overall observed fraction of CT-AGN with respect to the whole AGN population lower than the one reported in previous works, and as low as $\sim$4%. We find evidence that this over-estimation of N$_{\rm H}$ is likely due to the low quality of a subsample of spectra, either in the 2-10 keV band or in the Swift-BAT one.