The XMM-Newton/2dF Survey -- VI. Clustering and bias of the soft X-ray point sources

Abstract

We study the clustering properties of X-ray sources detected in the wide area (∼2 deg 2 ) bright, contiguous XMM-Newton/2dF survey. We detect 432 objects to a flux limit of 5 x 10 -15 erg cm -2 s -1 in the soft 0.5-2 keV band. Performing the standard angular correlation function analysis, a ∼3σ correlation signal between 0 and 150 arcsec is detected: w(θ < 150 arcsec) ≃ 0.114 ± 0.037. If the angular correlation function is modelled as a power law, w(θ) = (θ 0 /θ) γ-1 , then for its nominal slope of y = 1.8 we estimate, after correcting for the integral constraint and the amplification bias, that θ 0 ≃ 10.4 ±1.9 arcsec. Very similar results are obtained for the 462 sources detected in the total 0.5-8 keV band (θ 0 ≃ 10.8 ± 1.9 arcsec). Using a clustering evolution model which is constant for comoving coordinates (e = -1.2), a luminosity-dependent density evolution (LDDE) model for the X-ray luminosity function and the concordance cosmological model (Ω m = 1 - Ω Λ = 0.3) we obtain, by inverting Limber's integral equation, a spatial correlation length of r 0 ∼ 16 h -1 Mpc. This value is larger than that of previous ROSAT surveys as well as of the optical two-degree quasar redshift survey. Only in models where the clustering remains constant for physical coordinates (e = -3), do we obtain an r 0 value (∼7.5 h -1 Mpc) which is consistent with the above surveys. Finally, comparing the measured angular correlation function with the predictions of the concordance cosmological model, we find for two different bias evolution models that the soft X-ray sources at the present time should be biased with respect to the underline matter fluctuation field with bias values in the range (which depends on the biasing model used): 1.9 ≤b 0 ≤ 2.7 for e = -1.2 or 1 ≤ b 0 ≤ 1.6 for e = -3.