No Arabic abstract
The information content of the autocorrelation function (ACF) of intensity fluctuations of the X-ray background (XRB) is analyzed. The tight upper limits set by ROSAT deep survey data on the ACF at arcmin scales imply strong constraints on clustering properties of X-ray sources at cosmological distances and on their contribution to the soft XRB. If quasars have a clustering radius r_0=12-20 Mpc (H_0=50), and their two point correlation function, is constant in comoving coordinates as indicated by optical data, they cannot make up more 40-50% of the soft XRB (the maximum contribution may reach 80% in the case of stable clustering, epsilon=0). Active Star-forming (ASF) galaxies clustered like normal galaxies, with r_0=10-12 Mpc can yield up to 20% or up to 40% of the soft XRB for epsilon=-1.2 or epsilon=0, respectively. The ACF on degree scales essentially reflects the clustering properties of local sources and is proportional to their volume emissivity. The upper limits on scales of a few degrees imply that hard X-ray selected AGNs have r_0<25 Mpc if epsilon=0 or r_0<20 Mpc if epsilon=-1.2. No significant constraints are set on clustering of ASF galaxies, due to their low local volume emissivity. The possible signal on scales >6 deg, if real, may be due to AGNs with r_0=20 Mpc; the contribution from clusters of galaxies with r_0~50 Mpc is a factor 2 lower.
In this paper we present the two-point angular correlation function of the X-ray source population of 1063 XMM-Newton observations at high Galactic latitudes, comprising up to ~30000 sources over a sky area of 125.5 sq. deg, in three energy bands: 0.5-2 (soft), 2-10 (hard), and 4.5-10 (ultrahard) keV. We have measured the angular clustering of our survey and find significant positive clustering signals in the soft and hard bands, and a marginal clustering detection in the ultrahard band. We find dependency of the clustering strength on the flux limit and no significant differences in the clustering properties between sources with high hardness ratios and those with low hardness ratios. Our results show that obscured and unobscured objects share similar clustering properties and therefore they both reside in similar environments, in agreement with the unified model of AGN. We deprojected the angular clustering parameters via Limbers equation to compute their typical spatial lengths. From that we have inferred the typical mass of the dark matter haloes in which AGN at redshifts of ~1 are embedded. The short AGN lifetimes derived suggest that AGN activity might be a transient phase that can be experienced several times by a large fraction of galaxies throughout their lives.
A number of ultraluminous X-ray sources (ULXs) are physically associated with extragalactic globular clusters (GCs). We undertake a systematic X-ray analysis of eight of the brightest of these sources. We fit the spectra of the GC ULXs to single power law and single disk models. We find that the data never require that any of the sources change between a disk and a power law across successive observations. The GC ULXs best fit by a single disk show a bimodal distribution: they either have temperatures well below 0.5 keV, or variable temperatures ranging above 0.5 keV up to 2~keV. The GC ULXs with low kT have significant changes in luminosity but show little or no change in kT. By contrast, the sources with higher kT either change in both kT and $L_X$ together, or show no significant change in either parameter. Notably, the X-ray characteristics may be related to the optical properties of these ULXs, with the two lowest kT sources showing optical emission lines.
The angular correlation function is a powerful tool for deriving the clustering properties of AGN and hence the mass of the corresponding dark matter halos in which they reside. However, studies based on the application of the angular correlation function on X-ray samples, yield results apparently inconsistent with those based on the direct estimation of the spatial correlation function. The goal of the present paper is to attempt to investigate this issue by analysing a well defined sample. To this end we use the hard-band (2-10 keV) X-ray selected sources of the Chandra AEGIS fields, chosen because of the availability of accurately derived flux sensitivity maps. In particular we use the 186 hard-band sources with spectroscopic redshifts in the range z=0.3-1.3, a range selected in order to contain the bulk of the AGN while minimizing the contribution of unknown clustering and luminosity evolution from very high redshifts. Using the projected spatial auto-correlation function, we derive a clustering comoving length of 5.4+-1.0 Mpc (for gamma=1.8), consistent with results in the literature. We further derive the angular correlation function and the corresponding spatial clustering length using the Limbers inversion equation and a novel parametrization of the clustering evolution model that also takes into account the bias evolution of the host dark matter halo. The Limbers inverted spatial comoving clustering length of 5.5+-1.2 Mpc at a median redshift of z~0.75, matches the directly measured one, from the spatial correlation function analysis, but for a significant non-linear contribution to the growing mode of perturbations, estimated independently from literature results of x_0 at different redshifts. Therefore, using this sample of X-ray AGN and our clustering evolution parametrization we have found an excellent consistency between the angular and spatial clustering analysis.
We observed several nearby face-on spiral galaxies with the ROSAT PSPC. The apparent deficiency in soft X-ray surface brightness observed at the outer portion of their disks is consistent with the absorption of the extragalactic soft X-ray background by material associated with these galaxies, and allows us to place a lower limit on the intensity of this cosmologically important background. From the depth of the soft X-ray shadow observed in NGC 3184, a 95% confidence lower limit was derived to be $32 keV cm^{-2} s^{-1} keV^{-1}$ at 1/4 keV. This was obtained by assuming that there is no unresolved 1/4 keV X-ray emission from the outer region of the galaxy which may otherwise partially fill in the shadow: any such emission, or any unresolved structure in the absorbing gas, would imply a larger value. In the deepest exposure to date in this energy range, Hasinger et al. (1993) resolved about $30 keV cm^{-2} s^{-1} keV^{-1}$ at 1/4 keV into discrete sources; our current limit is therefore consistent with an extragalactic origin for all of these sources. Our results can also be directly compared with the corresponding upper limit derived from the ROSAT PSPC detection of soft X-ray shadows cast by high-latitude clouds in Ursa Major, $simeq 65 keV cm^{-2} s^{-1} keV^{-1}$ at 1/4 keV. The lower and upper limits are only a factor of 2 apart, and begin to provide a reasonable measurement of the intensity of the 1/4 keV extragalactic X-ray background.
The measured cosmic gamma ray background (CGB) spectrum at MeV energies is in reasonable agreement with the predicted contribution from type Ia supernovae (SNIa). But the characteristic features in the SNIa gamma ray spectrum, weakened by integration over source redshifts, are hard to measure, and additionally the contributions from other sources in the MeV range are uncertain, so that the SNIa origin of the MeV CGB remains unproven. Since different CGB sources have different clustering properties and redshift distributions, by combining the CGB spectrum and angular correlation measurements, the contributions to the CGB could be identified and separated. The SNIa CGB large-scale structure follows that of galaxies. Its rms fluctuation at degree scales has a characteristic energy dependence, ranging from $sim 1%$ to order of unity and can be measured to several percent precision by proposed future satellites such as the Advanced Compton Telescope. With the identification of the SNIa contribution, the SNIa rate could be measured unambiguously as a function of redshift up to $z sim 1$, by combining both the spectrum and angular correlation measurements, yielding new constraints on the star formation rate to even higher redshifts. Finally, we show that the gamma ray and neutrino backgrounds from supernovae should be closely connected, allowing an important consistency test from the measured data. Identification of the astrophysical contributions to the CGB would allow much greater sensitivity to an isotropic high-redshift CGB contribution arising in extra dimension or dark matter models.