[Paper Review] XMM-Newton observation of the Lockman Hole II. Spectral analysis

V. Mainieri, J. Bergeron+6 more|CaltechAUTHORS (California Institute of Technology)|Jul 8, 2002

Astrophysical Phenomena and ObservationsPhysics and Astronomy1 references137 citations

TL;DR

This study presents a deep X-ray spectral analysis of 98 XMM-Newton sources in the Lockman Hole field, focusing on intrinsic absorption and spectral indices. It finds that the mean spectral index ⟨Γ⟩ ≈ 2 is independent of absorption, and the hardening of faint source spectra is primarily due to increasing intrinsic absorption, identifying a significant population of obscured AGN, including 18 X-ray-selected EROs and six Type-2 QSO candidates with L_X > 10^44 erg s⁻¹ and log(N_H) > 22 cm⁻², yielding a density of ~69 such objects per square degree.

ABSTRACT

We present the results of the X-ray spectral analysis of the first deep X-ray survey with the XMM-Newton observatory during Performance Verification. We restrict the analysis to the sample of 98 sources with more than 70 net counts (flux limit in the [0.5-7] keV band of 1.6 10^{-15} erg cm^{-2} s^{-1}) of which 61 have redshift identification. We find no correlation between the spectral index Gamma and the intrinsic absorption column density N_H and, for both the Type-1 and Type-2 AGN populations, we obtain ~2. The progressive hardening of the mean X-ray source spectrum with decreasing flux is essentially due to an increase in intrinsic absorption. The marked separation between the two AGN populations in several diagnostics diagrams, involving X-ray colour, X-ray flux, optical/near IR colour and optical brightness, is also a consequence of different absorption column densities and enables the classification of optically faint obscured AGN. About 27% of the subsample with R-K' colour are EROs (R-K>5) and most of these 18 X-ray selected EROs contain anobscured AGN as revealed by their high X-ray-to-optical/near IR flux ratios. There are six sources in our sample with L_X[0.5-10]>10^44 erg s^{-1} and log(N_H)>10^22 cm^{-2}: which are likely Type-2 QSOs and we thus derive a density of ~69 objects of this class per square degree.

Motivation & Objective

• To understand the physical nature of faint X-ray sources in the Lockman Hole using X-ray spectral analysis.
• To determine whether the progressive hardening of X-ray spectra with decreasing flux is due to intrinsic absorption or intrinsically flat spectra.
• To classify obscured AGN using X-ray and optical/near-IR diagnostics, particularly R−K′ colours and X-ray-to-optical flux ratios.
• To identify and characterize high-luminosity, highly absorbed AGN, including potential Type-2 QSOs.
• To assess the contribution of host galaxies to optical and near-IR colours in obscured versus unobscured AGN.

Proposed method

• Spectral fitting of 98 XMM-Newton EPIC-pn sources with >70 net counts in the [0.5–7] keV band.
• Measurement of spectral index (Γ) and intrinsic absorption column density (N_H) using X-ray energy distribution modeling.
• Use of hardness ratios and X-ray colour-colour diagrams to separate source populations.
• Cross-identification with optical and near-IR data to derive R−K′ colours and optical magnitudes.
• Comparison of X-ray and optical/near-IR fluxes to identify obscured AGN via high f_X/f_R ratios.
• Use of redshift information and evolutionary tracks to distinguish between AGN and host galaxy contributions to observed colours.

Experimental results

Research questions

• RQ1 Is the observed hardening of X-ray spectra in faint sources due to increasing intrinsic absorption or intrinsically flat spectra?
• RQ2 How do X-ray spectral properties correlate with optical and near-IR colours, particularly R−K′, in AGN populations?
• RQ3 Can X-ray-selected EROs (R−K′ ≥ 5) be reliably identified as obscured AGN based on their X-ray and photometric properties?
• RQ4 What is the density of highly absorbed, luminous Type-2 QSOs (L_X > 10^44 erg s⁻¹, log(N_H) > 22 cm⁻²) in the deep XMM-Newton survey?
• RQ5 To what extent do host galaxy light and K-corrections affect the observed R−K′ colours of X-ray sources?

Key findings

• The mean spectral index ⟨Γ⟩ ≈ 2 is independent of intrinsic absorption, indicating that spectral hardening in faint sources is not due to intrinsically flat spectra.
• The progressive hardening of the mean X-ray spectrum with decreasing flux is primarily driven by increasing intrinsic absorption, not by a change in the power-law slope.
• About 27% of the R−K′ colour-selected sample are EROs (R−K′ ≥ 5), and most of these 18 X-ray-selected EROs are obscured AGN with high X-ray-to-optical/near-IR flux ratios.
• Six sources with L_X[0.5–10] > 10^44 erg s⁻¹ and log(N_H) > 22 cm⁻² are identified as likely Type-2 QSOs, yielding a density of ~69 such objects per square degree at the flux limit of 1.6×10⁻¹⁵ erg cm⁻² s⁻¹.
• Unidentified, optically faint sources (R > 24 for ~80%) and red R−K′ colours (≥4 for ~90%) are predominantly absorbed, with log(f_X[2–10]/f_R) > 1, indicating they are likely Type-2 AGN.
• Optical spectroscopic follow-up confirms that most newly detected, faint XMM sources are at z < 1 and are obscured AGN, with two X-ray-bright, optically normal galaxies also showing strong X-ray absorption.