We studied the population of X-ray point sources in the elliptical galaxy Centaurus A, using archival Chandra data. Within a radius of 10 arcmin from the centre we detected 272 sources. Of these approximately half are LMXBs with the rest being CXB sources. The spatial distribution of the LMXBs is found to be consistent with the distribution of K-band light. We constrain the luminosity function (LF) of the X-ray sources down to ${\sim} 2\times 10^{36}$ erg s$^{-1}$. The X-ray LF of the LMXBs flattens significantly below $L_X\sim 5\times 10^{37}$ erg s$^{-1}$ and follows the $dN/dL\propto L^{-1}$ law in agreement with results from the bulges of spiral galaxies.

We present a study of the variable X-ray source populations in the Antennae galaxies and a small sample of nearby star-forming galaxies. We find that source variability does not affect the shape of their X-ray luminosity functions. Several of the sources detected in the Antennae exhibit a wide range of spectral variability patterns. Finally, from our study of a small sample of nearby star-forming galaxies, we find a weak indication for variations of the shape of their luminosity functions.

We analyze the upper-end X-ray luminosity function (XLF) observed in elliptical galaxies for point sources. We propose that the observed XLF is dominated by transient BH systems in outburst and the XLF shape reflects the black hole (BH) mass spectrum among old X-ray transients. The BH mass spectrum – XLF connection depends on a weighting factor that is related to the transient duty cycle and depends on the host-galaxy age, the BH mass and the donor type (main sequence, red giant, or white dwarf). We argue that the assumption of a constant duty cycle for all systems leads to results inconsistent with current observations. The type of dominant donors in the upper-end XLF depends on what type of magnetic braking operates: in the case of “standard” magnetic braking, BH X-ray binaries with red-giant donors dominate, and in the case of weaker magnetic braking prescriptions main sequence donors dominate.

We present preliminary results from a detailed analysis of the X-ray point sources in the XMM-Newton survey of M31 (e.g. Barnard et al. 2005; Pietsch et al. 2005). These sources are expected to be mostly X-ray binaries.

Although our Galactic Center harbors a black hole (Sgr A*) of a few million solar masses, it and its environments are very quiet at present. In X-rays however, the close vicinity of Sgr A* shows very unique and various phenomena mostly originated from young stellar populations. We report on the X-ray perspective on the young stellar populations which are related to our Galactic Center activities. The discussion is essentially based on the observational facts of new X-ray objects in the Galactic Center region in the $1^\circ \times 2^\circ $ area. They are;

1. Clusters of young high mass stars, which are Sgr B2, Arches, IRS 13 and Quintuplet.

2. X-ray reflections in the giant molecular clouds, such as Sgr B2, Sgr C, M0.01-0.09 and others.

3. New candidates of X-ray supernova remnants (SNRs), which are Sgr A East, G0.570-0.018 and G359.8-0.3.

4. Non-thermal Jets, Filaments and Shells, which are unique X-ray features in the GC region. These X-ray features may be closely related with each other, hence may have common origins. A unified picture is presented for the X-ray activity of our Galactic Center comparing with the X-ray spectra from other type of galaxies such as;

5. Star burst galaxy (NGC 253), low luminosity AGN (M 81) and Seyfert 2 (NGC 1068).

Population synthesis of binary stars with relativistic component is presented.

We have performed evolutionary population synthesis calculations to investigate the nature of the faint X-ray sources in the Galactic center region detected by recent Chandra surveys. Our results show that neutron star low-mass X-ray binaries contribute significantly to the observed sources, but the majority of the sources in Wang et al. (2002) survey are still beyond our expectation. We also point out that wind-accreting neutron stars and intermediate polars play a minor role in accounting for the faint X-ray sources in Wang et al. (2002) and Muno et al. (2003) surveys, respectively.

We propose that the observed difference in the formation rates of bright low-mass X-ray binaries in metal-rich and metal-poor globular clusters can be explained by taking into account the difference in the stellar structure of main sequence donors with masses between ${\sim}0.85 M_\odot$ and ${\sim}1.25 M_\odot$ at different metallicities. This difference is caused by the absence of an outer convective zone in metal-poor main sequence stars in this mass range. In the result, magnetic braking, a powerful mechanism of orbital shrinkage, does not operate and dynamically formed main sequence – neutron star binaries fail to start mass transfer or appear as bright low-mass X-ray binaries.

Interpretative analyses of the X-ray emission from the giant starformation region R 136 have concluded that several colliding-wind binaries are likely to contribute to its X-ray output. Using our dedicated high-energy stellar population synthesis programme, we try to reproduce the suggested number of colliding-wind binaries. It appears that only assuming a very high binary fraction for the cluster's stellar population we can reproduce the observed X-ray luminosity distribution, if also the two most luminous sources are in fact multiple sources.

We explore the X-ray diagnostics of cosmic star formation history that have possible form recent deep field X-ray studies. We summarize the status of our understanding of the X-ray evolution of galaxies. We indicate the lessons learnt so far from number count plots, and criteria for distinguishing between normal/straburst galaxies and AGNs in deep X-ray surveys. We summarize how the observed correlations between X-ray emission and that at other wavebands indicate the value of X-rays as a probe of cosmic star formation.

We present the results of population synthesis simulations aimed at exploring the evolution of the 2–10 keV luminosity of X-ray binaries in a young stellar population. The results are applicable to populations of extragalactic X-ray binaries in starburst galaxies and many LINERs. We find that the integrated 2–10 keV luminosity of the simulated population reaches a maximum of about $10^{40}\,{\rm erg~s}^{-1}$ after approximately 20 Myr (for a star-formation rate of 10 M$_{\odot}$ yr$^{-1}$) and remains significant even after the end of star formation and the demise of the luminous OB stars. The results of our simulation are in agreement with recently-derived correlations between the X-ray luminosity starburst galaxies and their star-formation rate. We also find that the cumulative luminosity function is initially fairly flat, in agreement with recent observational results, becoming steeper as the population ages and the high-mass X-ray binaries are succeeded by binaries with progressively lighter donor stars. Using the output of Hydrogen-ionizing far-UV photons from the stellar population, we can plot the track of a “post-starburst” system in the $L_{\rm X}-L_{\rm H\alpha}$ diagram. The system starts off in the starburst locus but quickly evolves to the AGN locus where it lingers for at least 1 Gyr.

This paper reviews the nature of normal and starburst galaxies in deep X-ray surveys, focusing on the observational issues. Normal and starburst galaxies may be divided from AGN via X-ray/optical flux ratios, optical spectroscopic identification, hardness ratio, and X-ray luminosity. Each of these is discussed, including the possible impact on derived X-ray-Star Formation Rate (X-ray/SFR) correlations. The measured differences in the normal galaxy X-ray Luminosity Functions (XLFs) by SED type at $z\approx0.3$–1.0 are also described.

The redshift frontiers of deep X-ray surveys are discussed, including those for individually detected accreting binary systems (Ultraluminous X-ray Sources at $z\approx0.1$–0.3) and that for the highest-redshift X-ray detection of star formation (stacking analyses of Lyman Break Galaxies to $z\approx4$). The paper closes with a discussion of normal galaxy studies with future X-ray missions such as Constellation-X, XEUS, and Generation-X.

We have investigated 136 Chandra extragalactic sources without broad emission lines, including 93 NELG and 43 ALG. Based on $f_X/f_O$, $L_X$, X-ray spectral hardness and optical emission line diagnostics, we have conservatively classified 36 normal galaxies (20 spirals and 16 ellipticals) and 71 AGNs. Their redshift ranges from 0.01 to 1.2, while normal galaxies are at z=0.01–0.3. Our sample galaxies appear to share similar characteristics with local galaxies in terms of X-ray luminosities and spectral properties, as expected from the X-ray binary populations and the hot ISM. In conjunction with normal galaxies found in other surveys, we found no statistically significant evolution in $L_X/L_B$, within the limited z range ($<$0.1). We have built our log(N)-log(S) relationship of normal galaxies in the flux range, $f_X$ (0.5–8.0) = $10^{-15}$–$10^{-13}$ erg sec$^{-1}$ cm$^{-2}$, after correcting completeness by a series of simulations. The best-fit slope is −1.5 for both soft (0.5–2.0 KeV) and broad (0.5–8.0 KeV) energy bands, which is considerably steeper than that of AGN-dominated cosmic background sources at faint fluxes, but slightly flatter than the previous estimate, indicating normal galaxies will exceed in number over the AGN population at $f_X<10^{-18}$ erg sec$^{-1}$ cm$^{-2}$.

The great sensitivities of the Chandra X-ray Observatory and XMM-Newton are allowing us to explore the X-ray emission from galaxies at moderate to high redshift. By using the stacking method with CXO data, we show that we can detect the ensemble emission from normal elliptical, spiral and irregular galaxies out to redshifts approaching one. The average X-ray luminosity can then be compared with the results of models of the evolution in the numbers of X-ray binaries and can possibly be used to constrain models of star formation. In order to account for the increasing luminosity of spiral galaxies from low to moderate redshift, AGN components may need to be invoked.

Deep field observations are an essential tool to probe the cosmological evolution of galaxies. In this context, X-ray deep fields provide information about some of the most energetic cosmological objects: active galactic nuclei (AGN). Astronomers are interested in detecting sufficient numbers of AGN to probe the accretion history at high redshift. This talk gives an overview of the knowledge resulting from a highly complete soft X-ray selected sample collected with ROSAT, XMM–Newton and Chandra deep fields. The principal outcome based on X-ray luminosity functions and space density evolution studies is that low–luminosity AGN evolve in a dramatically different way from high–luminosity AGN: The most luminous quasars perform at significantly earlier cosmic times and are most numerous in a unit volume at cosmological redshift $z\sim2$. In contrast, low–luminosity AGN evolve later and their space density peaks at $z\sim 0.7$. This finding is also interpreted as an anti–hierarchical growth of supermassive black holes in the Universe. Comparing this with star formation rate history studies one concludes that supermassive black holes enter the cosmic stage before the bulk of the first stars. Therefore, first solutions of the so-called hen–egg problem are suggested. Finally, status developments and expectations of ongoing and future extended observations such as the XMM–COSMOS project are highlighted.

We present results of the X-ray spectral analysis of the $\sim$370 ksec deep survey obtained with XMM-Newton on the Chandra Deep Field South (CDFS). Using sample of 127 sources with redshift identifications and with a raw count limit of 100 (pn detector) we explored both the physical properties of individual sources, and the general properties of two AGN classes. The corresponding flux limits in the (0.5–2) and (2–10) keV bands are 1 $\times 10^{-16}$ and 9 $\times 10^{-16}$ erg cm$^{-2}$ s$^{-1}$, respectively. The average photon index is $\Gamma \sim$1.9 and $\Gamma \sim$1.8 for type-1 AGN and type-2 AGN, respectively. Although the properties of the spectra of more that 90% of AGN are in good agreement with the unified model, a fraction of ‘atypical’ objects (absorbed type-1 and unabsorbed type-2 AGN) was detected.

We present results from the hard X-ray sky survey being performed with INTEGRAL observatory. During two years of operation the significant sky area was covered with moderate sensitivity which now allows us to estimate surface number density of Active Galactic Nuclei (AGN) in hard (17-60 keV) energy band. Our catalog of detected sources comprises $>60$ AGN with wide range of intrinsic absorption. We used $24000 deg^{2}$ sky coverage (high Galactic latitude observations) with total exposure $>2$ Msec to derive source number-flux relation. The estimated surface density of extragalactic hard X-ray sources with flux $>10^{-11}$ erg s$^{-1}$ cm$^{-2}$ is $(1.0 \pm 0.5) 10^{-2} deg^{-2}$.

3C 273 is one of the brightest and best studied quasars. It has been observed for 770 ks with the imager IBIS (FoV 12 deg) on board INTEGRAL. To achieve the best possible S/N the dataset has been screened using several criteria indicating the quality of the data (i.e., number of good time intervals, etc). We describe the necessary tools and methods to analyze data of deep fields.

Population studies of faint X-ray sources (e.g. galaxies, AGN, $\gamma$-ray bursts and their afterglows) have become more and more common. The problem of obtaining an average spectrum in the rest-frame of the sources is non-trivial. We show that conventional methods for averaging low signal-to-noise X-ray spectra, when applied to sources at different redshifts, result in a mean rest-frame spectrum that can exhibit artificial features. These include the broadening and weakening of emission/absorption lines, a broad dip in the continuum above an emission line, and a spectral hardening at the highest energies. All of these effects have been observed in real data and have been given fairly weighty astrophysical and/or cosmological interpretations. We present a new method of averaging which considerably reduces the severity of the problems.

Chandra and XMM detect X-rays emitted during accretion onto supermassive black holes, even when they are highly obscured. I review what has been learned about the cosmic evolution of the X-ray luminosity functions and the reconstruction of the accretion history of supermassive black holes from extensive follow-up observations of both deep and wide-area Chandra X-ray surveys.