[Paper Review] Luminosity-dependent evolution of soft X-ray selected AGN: New Chandra and XMM-Newton surveys

G. Hasinger, T. Miyaji+1 more|ArXiv.org|Jun 6, 2005

Galaxies: Formation, Evolution, PhenomenaPhysics and Astronomy67 references394 citations

TL;DR

This study analyzes soft X-ray selected type-1 AGN from Chandra and XMM-Newton surveys, combining multi-wavelength data to derive space density functions with rigorous treatment of optical identification completeness. It reveals luminosity-dependent evolution: low-luminosity AGN peak at z < 1, while high-luminosity AGN peak at z ≈ 2, with space density declining at high redshift for L_X < 10^45 erg s⁻¹.

ABSTRACT

We present new results on the cosmological evolution of unabsorbed (type-1) active galactic nuclei (AGN) selected in the soft (0.5-2 keV) X-ray band. From a variety of ROSAT, XMM-Newton and Chandra surveys we selected a total of ~1000 AGN with an unprecedented spectroscopic and photometric optical/NIR identification completeness. For the first time we are able to derive reliable space densities for low-luminosity (Seyfert-type) X-ray sources at cosmological redshifts. The evolutionary behaviour of AGN shows a strong dependence on X-ray luminosity: while the space density of high-luminosity AGN reaches a peak around z~2, similar to that of optically selected QSO, the space density of low-luminosity AGNs peaks at redshifts below z=1. This confirms previous ROSAT findings of a luminosity-dependent density evolution. Using a rigorous treatment of the optical identification completeness we are able to show that the space density of AGN with X-ray luminosities L_x &lt; 10^45 erg s^-1 declines significantly towards high redshifts.

Motivation & Objective

• To determine the cosmological evolution of unabsorbed (type-1) AGN selected in the soft X-ray band (0.5–2 keV).
• To measure space densities of low-luminosity AGN (Seyfert-type) at cosmological redshifts with unprecedented completeness.
• To investigate whether AGN evolution depends on X-ray luminosity, testing luminosity-dependent density evolution (LDDE).
• To correct for spectroscopic and photometric identification incompleteness in deep surveys to obtain reliable space density estimates.
• To compare the evolution of X-ray selected AGN with that of optically selected QSOs and assess consistency across wavebands.

Proposed method

• Combined data from ROSAT, Chandra, and XMM-Newton surveys covering five orders of magnitude in flux and six in solid angle.
• Selected 944 type-1 AGN with spectroscopic or photometric/NIR optical/NIR identification, achieving ~95% completeness for X-ray sources.
• Applied the N_obs/N_mdl estimator to correct for binning bias in binned luminosity functions, ensuring accurate bin-centering.
• Used individual 1/V_max values to derive luminosity functions free of binning bias, incorporating effective magnitude limits from incomplete redshift surveys.
• Employed a cosmology with H₀ = 70h₇₀ km s⁻¹ Mpc⁻¹ and (Ωₘ, Ω_Λ) = (0.3, 0.7), consistent with WMAP results.
• Treated optical identification incompleteness rigorously by modeling expected vs. observed source counts per redshift-luminosity bin.

Experimental results

Research questions

• RQ1 How does the space density of soft X-ray selected type-1 AGN evolve with redshift across different X-ray luminosities?
• RQ2 Does the peak of AGN space density shift with luminosity, indicating luminosity-dependent density evolution?
• RQ3 Is there a decline in space density for low-luminosity AGN at high redshift, and can this be robustly measured despite identification incompleteness?
• RQ4 How do the evolution patterns of X-ray selected AGN compare to those of optically selected QSOs?
• RQ5 What is the impact of survey depth and optical completeness on the reliability of space density estimates at high redshift?

Key findings

• The space density of high-luminosity AGN (log L_X ≈ 45–46 erg s⁻¹) peaks at z ≈ 2, consistent with the evolution of optically selected QSOs.
• The space density of low-luminosity AGN (log L_X ≈ 42–43 erg s⁻¹) peaks at z ≈ 0.5–0.7, significantly lower than the peak for high-luminosity AGN.
• For L_X < 10^45 erg s⁻¹, the space density declines toward high redshifts, a result confirmed with rigorous treatment of optical identification incompleteness.
• The amount of density evolution from z = 0 to peak space density is >100× for high-luminosity AGN, but <10× for low-luminosity AGN.
• The luminosity function shape changes with redshift, confirming luminosity-dependent density evolution (LDDE) for X-ray AGN, as previously found for optical quasars.
• The sample of 944 identified type-1 AGN and 57 unidentifiable sources provides the most complete and homogeneous soft X-ray AGN sample to date, filling the luminosity-redshift plane uniformly.