Tracing black hole accretion with SED decomposition and IR lines: from local galaxies to the high-z Universe

TL;DR

This paper develops a method to decompose galaxy spectral energy distributions to estimate AGN and star-formation luminosities, linking IR lines to galaxy activity, and applies it to local and high-redshift galaxies to predict future IR observations.

Contribution

It introduces a detailed SED decomposition approach including IR spectral data to accurately estimate AGN and star-formation contributions, extending local relations to high redshift for future IR surveys.

Findings

IR line luminosities strongly correlate with star-formation and AGN bolometric luminosities.

Relations between IR lines and galaxy activity vary with AGN contribution.

Predicted IR line luminosity functions for high-redshift galaxies for future IR observatories.

Abstract

We present new estimates of AGN accretion and star-formation luminosity in galaxies obtained for the local 12-$\mu$m sample of Seyfert galaxies (12MGS), by performing a detailed broad-band spectral energy distribution (SED) decomposition including the emission of stars, dust heated by star formation and a possible AGN dusty torus. Thanks to the availability of data from the X-rays to the sub-millimetre, we constrain and test the contribution of the stellar, AGN and star-formation components to the SEDs. The availability of Spitzer-IRS low resolution mid-infrared (mid-IR) spectra is crucial to constrain the dusty torus component at its peak wavelengths. The results of SED-fitting are also tested against the available information in other bands: the reconstructed AGN bolometric luminosity is compared to those derived from X-rays and from the high excitation IR lines tracing AGN activity like [Ne V] and [O IV]. The IR luminosity due to star-formation (SF) and the intrinsic AGN bolometric luminosity are shown to be strongly related to the IR line luminosity. Variations of these relations with different AGN fractions are investigated, showing that the relation dispersions are mainly due to different AGN relative contribution within the galaxy. Extrapolating these local relations between line and SF or AGN luminosities to higher redshifts, by means of recent Herschel galaxy evolution results, we then obtain mid- and far-IR line luminosity functions useful to estimate how many star-forming galaxies and AGN we expect to detect in the different lines at different redshifts and luminosities with future IR facilities (e.g., JWST, SPICA).