Key points are not available for this paper at this time.
We present the SEDs of a hard X-ray selected sample containing 136 sources with F_ (2-10 keV) > 10^ (-14) erg cm^ (-2) s^ (-1) ; 132 are AGNs. The sources are detected in a 1 deg² area of the XMM-Newton Medium Deep Survey where optical data from the VVDS and CFHTLS and infrared data from the SWIRE survey are available. Based on a SED fitting technique we derive photometric redshifts with σ (1 + z) = 0. 11 and 6% of outliers and identify AGN signatures in 83% of the objects. This fraction is higher than derived when a spectroscopic classification is available. The remaining 17^ (+9) _ (-6) % of AGNs show star-forming galaxy SEDs (SF class). The sources with AGN signatures are divided in two classes, AGN1 (33^ (+6) _ (-1) %) and AGN2 (50^ (+6) _ (-11) %). The AGN1 and AGN2 classes include sources whose SEDs are fitted by type 1 and type 2 AGN templates, respectively. On average, AGN1s show soft X-ray spectra, consistent with being unabsorbed, while AGN2s and SFs show hard X-ray spectra, consistent with being absorbed. The analysis of the average SEDs as a function of X-ray luminosity shows a reddening of the infrared SEDs, consistent with a decreasing contribution from the host galaxy at higher luminosities. The AGNs in the SF classes are likely obscured in the mid-infrared, as suggested by their low L_ (3-20 μm) /L^ (corr) _ (0. 5-10 keV) ratios. We confirm the previously found correlation for AGNs between the radio luminosity and the X-ray and the mid-infrared luminosities. The X-ray-radio correlation can be used to identify heavily absorbed AGNs. However, the estimated radio fluxes for the missing AGN population responsible for the bulk of the background at E > 10 keV are too faint to be detected even in the deepest current radio surveys.
Building similarity graph...
Analyzing shared references across papers
Loading...
M. Polletta
M. Tajer
L. Maraschi
The Astrophysical Journal
Centre National de la Recherche Scientifique
University of California, San Diego
Imperial College London
Building similarity graph...
Analyzing shared references across papers
Loading...
Polletta et al. (Fri,) studied this question.
www.synapsesocial.com/papers/69d7b6b50a5b166600f307e6 — DOI: https://doi.org/10.1086/518113