AGN obscuration in optical and X-rays: Host properties and the interplay of nuclear and galactic gas and dust in a combined SDSS-XMM sample

Abstract

We investigate the link between optical obscuration and X-ray absorption in active galactic nuclei (AGN) by combining X-ray spectroscopy from 4XMM-DR11 with SDSS DR16Q spectroscopy. Bayesian X-ray spectral fits were obtained within the XMM2Athena project, and host-galaxy properties were derived via \textsc{CIGALE} SED fitting. Our final sample comprises 241 X-ray AGN at $z<1.9$. For 172 sources ($\sim70\%$), the optical broad-line (BL) or narrow-line (NL) classification agrees with their X-ray obscuration based on $N_{\rm H}$, but two mismatched populations emerge. Eleven BL AGN show signs of X-ray absorption (BLAbs) and elevated gas-to-dust ratios compared to BL AGN, consistent with dust-free or host-scale absorbers. Conversely, 58 NL AGN appear unobscured in X-rays (NLUnabs) and low gas-to-dust ratios. Nearly half are assigned type~1 properties by SED fitting, suggesting diluted or intrinsically weak broad-line regions, host contamination, or variability. Optical line diagnostics support this picture: NL AGN show higher Balmer decrements than NLUnabs, indicating stronger extinction and different ionization conditions. Host diagnostics further reinforce the contrasts: at $\rm z<0.8$, NLUnabs show star-formation rates and accretion efficiencies that are comparable to BL AGN, whereas NL AGN reside in more quiescent hosts with lower star formation and less efficient black-hole growth. BLAbs match BL AGN in host and accretion properties, with their peculiarity lying in excess X-ray absorption. These findings demonstrate that obscuration arises not only from orientation but also from multi-scale distributions of gas and dust. Identifying such mismatched populations will be crucial for interpreting AGN demographics in ongoing and upcoming surveys such as \emph{Euclid} and VRO/LSST.

Figures (8)

• Figure 1: Distribution of the FWHM for the 241 sources included in the analysis. The plotted value corresponds to the emission line used for each source (e.g., H$\beta$, Mg II), depending on redshift (see text for more details).
• Figure 2: Distribution of the 241 sources in the $L_{\mathrm{X}}$--$z$ plane. The definitions of Absorbed Broad (BLAbs) and Unabsorbed Narrow (NLUnabs) are given in Sect. \ref{['sec_optical_vs_xray']}.
• Figure 3: Distribution of [O III]$\lambda5007$ FWHM values for the AGN sample.
• Figure 4: FWHM versus X-ray absorption (N$\rm _H$). Different AGN populations are plotted using the colours and symbols indicated in the legend. The horizontal dashed line marks the FWHM threshold used to separate sources by the width of their optical lines, while the vertical dashed line indicates the N$\rm _H$ threshold adopted to identify X-ray–absorbed sources (see text for details).
• Figure 5: Classification based on SED fitting for the four AGN populations, using the updated SED fitting criteria (see text for more details).