Free Access
Issue
A&A
Volume 551, March 2013
Article Number A8
Number of page(s) 18
Section Cosmology (including clusters of galaxies)
DOI https://doi.org/10.1051/0004-6361/201220389
Published online 08 February 2013

© ESO, 2013

1. Introduction

Clusters of galaxies are the largest gravitationally bound objects in the Universe and are tracers of the cosmic structure. Since their formation and evolution depends sensitively on the cosmological parameters, they are strong cosmological probes. Identification of clusters from X-ray surveys is currently the best method of constructing well defined samples of galaxy clusters for cosmological studies. The X-ray emission of the intracluster gas depends mostly on the square of the gas density. Owing to the peaked gas density distribution of clusters, the X-ray emission is more compact than the distribution of cluster galaxies, hence less affected by projection effects. For these reasons the X-ray luminosity is a good proxy of the total cluster mass.

The identification of ROSAT X-ray sources has resulted in large samples of local clusters (e.g. Böhringer et al. 2004, REFLEX) and a limited number of clusters at intermediate redshifts z > 0.4 (e.g. Vikhlinin et al. 1998). Only five distant clusters beyond z = 1 have been found among the ROSAT sources, most of them in the ROSAT Deep Cluster Survey (Rosati et al. 1998, RDCS).

Among the large number of extended sources serendipitously detected by XMM-Newton (Schwope et al. 2004; Lloyd-Davies et al. 2011; Takey et al. 2011) one can expect a significant fraction of distant clusters of galaxies. To date, about 40 clusters beyond redshift z ≳ 1 and about 10 z ≳ 1.4 are known, of which IDCS J1426.5+3508 at z = 1.75 (Stanford et al. 2012), ClG J0218.3-0510 at z = 1.62 (Papovich et al. 2010; Tanaka et al. 2010; Pierre et al. 2012), XMMU J0338.8+0021 at z = 1.49 (Nastasi et al. 2011), LH146 at z = 1.75 (Henry et al. 2010), XMMU J0044.0-2033 at z = 1.58 (Santos et al. 2011), and XMMU J1007.4+1237 at z = 1.56 (Fassbender et al. 2011b) are the most distant ones. There have also been claims of even more distant clusters at z ~ 2 (Andreon et al. 2009; Gobat et al. 2011); however, these sources lack either a clear X-ray detection or spectroscopic confirmation.

We are conducting a focussed project on the identification of distant (z ≥ 0.8) clusters from serendipitously detected XMM-Newton sources, the XMM-Newton distant cluster project (XDCP, Böhringer et al. 2005; Lamer et al. 2006; Fassbender et al. 2011a). An early success was our discovery of the most distant cluster of that time, XMMU 2235.3-2557 at z = 1.39 (Mullis et al. 2005). To date, the XDCP has been contributing half of the known sample of distant galaxy clusters. The sample contains 22 X-ray bright sources in the redshift range 0.9 < z < 1.6 alone (Fassbender et al. 2011a).

In the current paper we report on the extended X-ray sources found in the field of the broad absorption line (BAL) QSO LBQS 2212-1759. This object has been repeatedly targetted with XMM-Newton, so that the field features one of the deepest XMM-Newton EPIC (European Photon Imaging Camera) exposures taken at that point. This field also received deep optical imaging coverage by the ESO Imaging Survey (EIS field XMM-07, Dietrich et al. 2006; Mignano et al. 2007) and the Canada-France-Hawaii Telescope (CFHT) legacy survey (LS) deep field four (D4). Furthermore, we have performed extensive spectroscopic follow-up with VLT/FORS2 to confirm cluster membership. We supply a complete X-ray source list of the stacked X-ray images for point sources in Table 2 (available at the CDS). However, in this paper, we focus on the content of only extended sources (Table 3).

The paper is structured as follows. In Sect. 2 we describe the X-ray source detection and the X-ray spectral analysis of the brightest extended sources. Section 3 gives an overview of the optical data available for the field and describes the galaxies over-densities in photometric redshift space derived from them. The spectroscopic confirmation of the cluster candidates is presented next in Sect. 4. Section 5 presents the optical identifications of the extended X-ray sources and the cluster redshifts. In the same section, the identification of the individual objects is discussed. The cosmological interpretation of our results is finally presented in Sect. 6.

We adopt a cosmology of H0 = 70   km  s-1  Mpc-1, ΩM = 0.3, and ΩΛ = 0.7 and magnitudes are in the AB system.

2. XMM-Newton observations

2.1. X-ray data reduction and source detection

XMM-Newton has repeatedly targetted (yet never detected) the BAL QSO LBQS 2212-1759 (Clavel et al. 2006). The field received a total good exposure of 250 ks and is thus one of the deepest XMM fields. We retrieved theEPIC data sets from the public XMM archive (XSA) and processed the data from observation data files (ODFs) with the XMM science analysis software (XMM-SAS v. 6.5)1. Source detection runs with one of the latest versions XMM-SAS v. 11 show no major changes with respect to the source list used in this work. We therefore remain with the older version for consistency and comparison with the 2XMM catalogue.

To screen the data for periods of high particle background, we created light curves of the high energy (7–15 keV) events for each of the threeEPIC cameras. Periods where the count rate was more than 30% higher than the median rate were excluded from the later analysis. Table 1 shows the remaining exposure times of cleaned data for the individual pointing.

Table 1

Summary of XMM-Newton observations.

The event lists of each observation were transformed to a common astrometric frame (using the XMM-SAS task attcalc) and images in the five standard XMM-Newton energy bands (Watson et al. (2009), band 1: 0.2−0.5   keV, band 2: 0.5−1.0   keV, band 3: 1.0−2.0   keV, band 4: 2.0−4.5   keV, band 5: 4.5−12.0   keV) were binned for each camera and exposure. These bands differ from the energy bands (with a focus on 0.35−2.4   keV) usually used within the XDCP collaboration (Fassbender et al. 2011a).

Table 3

Extended X-ray sources properties from source detection.

For each of these images, an exposure map, background map, and detection mask were created using the respective XMM-SAS source detection tasks. Subsequently, the science images, exposure maps, and background maps of the individual exposures were added, creating one image for each of the three cameras and five energy bands.

In the softest band (0.2 − 0.5   keV) some of theEPIC detectors show spatially and temporally variable background features that can lead to spurious detections of extended sources. On the other hand, the thermal spectra of galaxy clusters only have a small fraction of their count rates in the hardest band beyond 4.5   keV. Therefore we restricted the detection of extended sources to the energy bands 2–4.

Source detection with the XMM-SAS tasks eboxdetect and emldetect was performed simultaneously on the nine images from three cameras and three energy bands. The task eboxdetect applies a sliding box detection algorithm to the input images. The resulting list is passed to the task emldetect, which fits the calibration PSF, optionally convolved with an extent model, to each input source. Extended sources were modelled using a King profile of the form (1)with core radii between rc =  [4.0′′ − 80.0′′]  as the model’s extent parameter space. x − x0 and y − y0 are pixel coordinates relative to the central pixel of the object considered. The thresholds for the detection likelihoods were set to L = 5.0 for eboxdetect and L = 6.0 for emldetect, with L =  −ln(Pfalse). This corresponds to probabilities of false detection of P = 6.7 × 10-3 and P = 2.5 × 10-3.

The emldetect output source list contains 264 sources2, of which nine are indicated to be extended. These objects are numbered in Fig. 1. Table 3 lists the extended sources with their basic X-ray parameters derived from emldetect and sorted according to source counts. Spectral X-ray properties normalised to physical scales are found in Table 4.

thumbnail Fig. 1

StackedEPIC image in the 0.2–2.0 keV band with detected point sources (small circles) and extended sources (large circles) labelled from 1 to 9.

Open with DEXTER

Table 4

Extended X-ray sources.

2.2. X-ray spectra and redshift estimates

Thanks to the deep exposure in the field, several of the extended sources were detected with more than 1000 source counts (see Table 3). We therefore attempted to estimate the redshifts of the brightest cluster candidates by fittingEPIC spectra with plasma spectra templates, keeping redshifts and temperatures as free parameters. With the same method we derived metallicities for those clusters where it was feasible.

We extracted spectra (using XMM-SAS v.10.0) for all the sources with >400 counts (i.e. sources ext#1-#6). Due to the low count rates (0.01 − 0.05ctss-1) and the extended nature of the sources the background fraction in the extracted spectra is relatively high. Therefore the regions for source and background extraction were manually selected to avoid systematic errors introduced by chip gaps, etc., and to avoid contamination by unrelated X-ray sources. The data were rendered and fitted through XSPEC using C statistics. We applied single mekal models corrected for galactic absorption (tbabs).

The spectral fits enable us to get a handle on the clusters redshift, which initially drove our follow-up programme. Some X-ray spectra enabled tight constraints on the distance, as in the case of source ext#1, which is the only source alone the iron line has been detected. Others merely confirmed the “high” redshift z ≳ 1 nature of the cluster as displayed in Figs. A.1(a)–e. An overview of spectral X-ray properties is presented in Table 4. The individual results are discussed both in Sect. 5 for each confirmed cluster individually and in Appendix A. The spectral properties in the context of the proper flux determination are discussed in the next section.

2.3. Flux determination

To enable comparisons of our results with previous findings from other works requires a flux measurement within a fixed radius. There, SΔ is the flux within RΔ, where Δ is defined as the average density of the intracluster medium (ICM) relative to the critical density of the Universe at a given redshift. The flux S500 is commonly indicated, however, in most cases not directly measurable, owing to the sensitivity of the instruments, the background level, point source contamination and chip gaps. The fluxes from Table 3 are the results of the source detection, and they correspond to fitted beta models integrated over a theoretically infinite radius.

To obtain S500, we apply an iterative approach. This method is described in detail in Šuhada et al. (2012) and Takey et al. (2011). To estimate M500 we apply the empirical scaling relation from Vikhlinin et al. (2009). Since these authors measure the flux in the energy band 0.5–2 keV as well, no conversion is required. Appropriately, the evolution of the LX − M relation deduced in Vikhlinin et al. (2009) is compatible with the findings from Reichert et al. (2011), which both allow for a non-evolution scenario; i.e., the luminosity of clusters does not change self-similarly with redshift. In the self-similar case, the evolution factor is E(z) − 7/4. The physical parameters used in the iterative approach are computed using the following equations: where ρ(z) is the critical density of the Universe, and h the normalised Hubble parameter. The evolution factor and ρ(z) are defined as Given the luminosity from the spectral fit, we can make an estimation for M500 by applying Eq. (1)and setting h to 0.70. The errors on the mass are the simple propagation of the luminosity uncertainties. From this quantity we derive R500 using Eq. (2). Since the cluster-specific core radius, which largely determines the shape of the beta model, is only mildly realistically described by the emldetect output, we use the empirical relation, Eq. (3), from the cosmic evolution survey (COSMOS, Finoguenov et al. 2007) to re-estimate rc. The essential step in this procedure is to determine what fraction of R500 we have covered in the area from which the spectrum has been extracted.

To do so, we synthesise beta models as in Eq. (7). The beta model is essentially a King profile from Eq. (1), and it describes the surface brightness profile of a cluster, determined uniquely by the core radius rc, provided β is known. By varying the radius R, which is computed using Eq. (2), we determine the fraction of the surface brightness enclosed, which equals the luminosity correction for a certain radius.

When this quantity converges, i.e. R = R500, after several iterations (typically 2) we assume we have obtained the true R500, and hence L500 proper. On average the spectral extraction radius has been 2.2 times smaller than R500. The resulting values are presented in Table 4 using spectral fluxes. (7)We fix the temperature at the value from the spectral fit. We emphasise, that all source detection chains in this work assume . The factor between L0.5 − 2.0   keV and L500 in Table 4 is applied to convert S0.5 − 2.0   keV into S500.

3. Optical imaging

The field of LBQS 2212-1759 has been the target of various imaging programmes. This section describes campaigns that have been launched as part of this study and the archival data analysed in supplement. An overview of the sky coverage is presented in Fig. 2. In this paper, only the area enclosed by the central XMM-Newton pointing (blue circle, solid line) is considered.

thumbnail Fig. 2

Sky coverage of imaging campaigns.Black: the optical 1   deg2 CFHTLS D4 field. Orange: the BVRI sky projection as imaged by the WFI. Red: the NIR WIRcam supplementary coverage. Cyan: the NIR observations made at Calar Alto. Blue: XMM-Newton pointings. The dashed circles represent the 2007 study (P.I. K. Nandra) covering the entire D4 in X-rays with  ~ 30 − 40ks exposures (Bielby et al. 2010). Green: VLT-FORS2 pointings used for initial detection. The crosses indicate the position of all clusters found.

Open with DEXTER

First, the field was selected for deep imaging observations in the B,V,R, and I bands with the Wide Field Imager (WFI) at the 2.2 m ESO/MPIA telescope as part of the ESO Imaging Survey (Dietrich et al. 2006). The exposure time in the R-band was augmented by a guest observer programme for the Bonn weak lensing survey (Dietrich et al. 2007, BLOX). In Oct. and Nov. 2006, a NIR campaign was run in the bands z and H with the Omega 2000 camera at the Calar Alto 3.5 m telescope. Both ESO data sets have been used for the initial clusters identification. For all final photometric analysis in this section, however, we rely solely on the data from the CFHTLS, due to its supremacy. We describe the data reduction in Sect. 3.1. Section 3.2 describes the retrieval of photometric redshifts, which we apply to find over-densities of galaxies in Sect. 3.3.

3.1. CFHT data

We make use of ugriz archival data taken with the MegaPrime/MegaCam instrument mounted at the prime focus at the 3.6 m CFHT. This instrument has a field of view (FoV) of 1 × 1 deg2 and a scale of 0.187 arcsec per pixel.

We made use of data located in the deep field D4 of the CFHT Legacy Survey (CFHTLS), which is part of the CFHT Supernova Legacy Survey (SNLS) and observations are still ongoing. We retrieved the data from the Elixir system3 in a preprocessed form and further processed it as described in Erben et al. (2009)4. This reduction is independent of the official releases by the CFHTLS collaboration (Ilbert et al. 2006; Coupon et al. 2009).

The photometry was performed with sextractor (Bertin & Arnouts 1996) in dual mode using the i-band image as a detection frame. We measured magnitudes in apertures of 1.86 arcsec of diameter in seeing-matched images. The derived magnitudes were used to derive colours, SED classification, and photometric redshifts (see below). We also made use of deep near-infrared data (JH & Ks bands) with the WIRcam (also at CFHT). The data is part of the WIRcam Deep Survey (see Bielby et al. 2010) and covers a significant portion of most of the deep XMM-Newton field. This data is important for obtaining precise photometric redshifts at high redshift as important continuum features (e.g. the 4   000    Å  break) are shifted to the near-infrared.

3.2. Photometric redshifts

In this paper we use the multi-band imaging (ugrizJHKs) and photometric galaxy redshifts from the CFHTLS to identify cluster counterparts of the extended X-ray sources and to measure their photometric redshifts. We used the code PHOTO-z (Bender et al. 2001) to estimate the photometric redshifts, with a set of 31 SED templates (see Brimioulle et al. 2008; Lerchster et al. 2011, for details).

We used the results from the spectroscopic programme (Sect. 4) to calibrate our photometric redshift estimates by applying zero-point offsets (typical values are  ~0.01...0.1 mag). A total of 106 spectra fulfil the quality criteria (i.e. objects with reliable photometric and spectroscopic redshifts). The fraction of catastrophic outliers is defined as η =  | zspec − zphot | /(1 + zspec) > 0.15, which is 2.8% in our case. The redshift accuracy is measured with the normalised median absolute deviation, σ = 1.48    ×    median( |  △ z | /(1 + z)) = 0.037, where  △ z = zspec − zphot and z is spectroscopic. The value of σ is identical to the standard deviation of  |  △ z | /(1 + z).

thumbnail Fig. 3

Photometric redshifts, plotted against the spectroscopic ones. The vertical dashes are the photometric redshift errors. The dashed lines are for zphot = zspec ± 0.15   (1 + zspec).

Open with DEXTER

3.3. Galaxy over-densities

From the photometric redshift information we measured the projected galaxy over-density in overlapping redshift intervals of Δz = 0.05. The purpose of this procedure was to qualitatively illustrate the completeness of the X-ray selection technique as a method to identify clusters of galaxies. Therefore we set a relatively high detection threshold in order to pick up only the most significant density peaks.

thumbnail Fig. 4

Over-density contours of galaxies within the field of view of the XMM-Newton pointing. The background image is the stacked image of all XMM-Newton observations of LBQS2212 (0.5–2.0 keV). The extended X-ray sources from Fig. 1 are indicated with their identifiers on the side (1 arcmin offset in RA). In smaller fonts the identifiers from other works are indicated either in black, if presented in this work as well, or in grey if not found by our method. Between round brackets the identifier from Bielby et al. (2010) is indicated and square brackets refer to the catalogue from Olsen et al. (2007). Curly brackets present the clusters from Adami et al. (2010). The red arrow points at an optical over-density, which has no clear counterpart in X-rays. The contours represent 5 Poissonian likelihood levels with 10 < L < Lmax, colour-coded, or rather colour-mapped, by redshift interval: purple z = 0.3 Lmax = 15, blue z = 0.35 Lmax = 20, green z = 0.45 Lmax = 20, dark pink z = 0.78 Lmax = 30, orange z = 0.98 Lmax = 15, red z = 1.18 Lmax = 20, heat z = 1.38 Lmax = 20, where z refers to , i.e. the middle of the bin.

Open with DEXTER

We computed the mean galaxy density in boxes of 400kpc a side, which is enlarged to 500kpc for redshifts z > 0.9, masking out regions affected by bright stars. At redshifts beyond z > 0.9 clusters are thought to be increasingly less relaxed. Therefore, the need arises to search for the associated galaxies in a larger physical volume. The values of Δz, the box size and the critical redshift where the box size changes have been tuned to maximise the detection of the already known clusters in this field from this work. We consider only galaxies with a photometric redshift error δz < 0.1. We compared the number of galaxies in the box to the assumed Poisson distributed background, which is taken to be the mean galaxy count in the field considered. The likelihood (Eq. (8)) was computed taking the negative natural logarithm of the incomplete gamma function P(a,x) as defined in Eq. (9). The incomplete gamma function is an integral, which is applied to computing the likelihood that a given distribution x (i.e. the number of galaxies is the box) is exceeding Poissonian noise a (Cash 1979). (8)with (9)where Γ(a) is the gamma function. This method of “source finding” is similar to the XMM-SAS task eboxdetect (Watson et al. 2009) also described in Sect. 2.1. The resulting map is displayed in Fig. 4 as contours on the stacked X-ray image. The redshift intervals are chosen to overlap in order to prevent a search bias that would occur when clusters fall exactly between two intervals. Since the redshift error cut is set to 0.1, some galaxy over-densities are expected to show up in multiple adjacent redshift bins. We have assumed a nominal redshift error of 0.001 for all galaxies, ensuring that each galaxy appears in one redshift bin only. All confirmed X-ray selected clusters from Table 7 are retrieved at the appropriate redshifts. That is, all spectroscopically confirmed clusters are found in a redshift interval around the nominal redshift taking the spread in the photo-z accuracy into account (see Fig. 3), of which typical values amount to about 5%.

We define the photometric redshift zphot of the clusters as the mean of all galaxies meeting the above-mentioned selection criteria (i.e. photometric accuracy and spatial distribution), having , where Δz1 =  | z ∗  − zmid | . zmid is the middle of the redshift bin containing the largest number of galaxies, whereas z ∗  is the photometric redshift of each individual galaxy. redshift. The relative difference (in %) in Table 7 is defined as , where Δz2 = zphot − zspec. For the nearby clusters in our sample, zphot tends to be slightly overestimated, while the high redshift ones the photometric redshift is somewhat lower than the spectroscopic one. The redshift bin, in which the galaxy signal is strongest (i.e. highest likelihood) is selected for display in Fig. 4. This quantity is identical neither to zmid nor to zphot due to different treatments of background galaxies. Nonetheless, they remain similar.

The clusters are indicated with colour-coded contours reflecting the redshift and the over-density likelihood. We display only significant5 over-densities having L > 10, within the FoV of the XMM-Newton pointing. A selection of clusters found in previous publications (Bielby et al. 2010; Olsen et al. 2007; Adami et al. 2010) are also referred to in Fig. 4. The most distant (Stanford et al. 2006) cluster is detected by all groups except for Adami et al. (2010), whereas some other structures, i.e. system ext#6 and the new, possibly spurious structure (red arrow in Fig. 4), are found solely in this work. Source ext#4 is also found by Adami et al. (2010) at zphot = 0.95. Finally, we retrieve all X-ray selected clusters as optical over-densities. Additionally, we find a significant over-density, that is idicated by the arrow in Fig. 4, which could not be associated with an X-ray source at an off-axis angle  ~ 14 arcmin. The optical data suggest a detection due to an projection effect, since no conglomeration of red galaxies is evident.

Olsen et al. (2007) have applied the Postman matched filter (MF, Postman et al. 1996) to search for optically selected clusters of galaxies in the CFHTLS deep fields. In the overlap of field D4 with the XMM-Newton FoV they found eight clusters with redshifts ranging from z = 0.3 to z = 1.1. These structures are partially retrieved by us as seen in Fig. 4. Grove et al. (2009) have extended the sample by including the z′-band. This has led to an enhanced detection of distant clusters, which, however, are not discussed further in this work.

In a similar fashion Adami et al. (2010) detect ten clusters within the common area from galaxy density maps resulting from an adaptive kernel algorithm working on slices in (photometric) redshift space. The papers from Olsen et al. (2007), Grove et al. (2009), and Adami et al. (2010) are comparable to one another in various ways, and for the same reasons different from our study. First of all, all base their density map on a single band detection (for instance the i′ band), secondly, use a similar detection algorithm for optical over-densities, and thirdly, none perform an X-ray source detection, which automatically points them to a larger quantity of less massive optical clusters. Last, the works addressed above are restricted to the optical u ∗ griz′ data and therefore are expected to remain less sensitive to the more distant clusters in the field.

Gavazzi & Soucail (2007) use the i′-band images for weak lensing mass reconstruction. In the D4 field they find one significant mass peak, which is outside the XMM-Newton FoV.

A more direct camparison to our work is given by Bielby et al. (2010), who applied a red sequence finder to their wavelet-detected extended X-ray sources in D4. This team focussed on groups and clusters at high redshift z ≳ 1.1, disgarding all sources at lower redshift. The number density of sources (40perdeg2) they find is roughly equal to the results from our search algorithm using XMM-SAS. These authors report on three distant X-ray luminous groups within the area covered in this paper, of which two (see Fig. 4) have been retrieved by us as well both as X-ray and optical over-density structures. Interestingly, Bielby et al. (2010) make no reference to the source ext#4, described in Sect. 5.4. One of their sources, ID-32 or WIRDXC J2216.4-1748, could not be confirmed by us. Likewise, we have been able to spectroscopically refine the two photometric cluster redshifts found by these authors.

4. Optical spectroscopy

The spectroscopic results used in this publication were obtained with FORS2 mounted on the VLT through observing programme 079.A-0369(C) (Aug./Sept. 2007) and 080.A-0659(A) (Oct./Nov. 2007 and Jun. 2008). The total integration time of roughly 19 hours was used to target the nearby cluster candidates ext#1, ext#3 and ext#6 with three masks through grism 150I and the distant cluster candidates ext#4 and ext#5 with four masks through grism 300I grism. As spectroscopic targets we prioritised those objects that lie close to the X-ray emission peak and/or have colours close to the red sequence as determined from the ESO-VLT imaging. We also targetted some X-ray point sources.

The science data was flat-fielded and wavelength-calibrated with the appropriate calibration files supplied by the FORS2 database. The bias was read from the chip overscan region. Flux calibration has been applied with a single response curve derived from only one observation. No atmospheric correction has been performed. The recorded spectra have a spectral coverage from 6000−10 000 Å  (ℜ = 660) and 4000−10 000 Å  (ℜ = 260) for the 300I and 150I respectively, where ℜ denotes the resolving power.

From 289 slits we were able to extract 231 science-grade spectra, plus 19 double recordings, having sufficient signal-to-noise to classify the spectra and measure redshifts, of which six are QSOs and 42 (late-type) stars. Their redshift distribution is shown in Fig. 5. A total of 66 spectra are of cluster members, excluding ten double recordings. The remaining 118 spectra belong to unrelated field galaxies. The complete catalogue can be viewed as in Table 5 (available at the CDS). The list of cluster members is presented in Table 6. Galaxy cluster members have been selected within an redshift interval Δz ≲ 0.015 around the mean redshift within a projected radius of  ≲ 1   Mpc around the X-ray source.

thumbnail Fig. 5

Histogram of all 231 science grade spectra. The insets show a zoom-in on the low (top) and high (bottom) redshift range.

Open with DEXTER

Redshifts were obtained by correlating the spectra with the elliptical and oldest starburst template from Kinney et al. (1996), both with the EZ software (Garilli et al. 2010) for passive galaxies and the iraf fxcor task for some galaxies with strong emission lines.

Galaxy spectra were processed from 4500    Å  up to 10   000    Å , depending on spectral coverage and faulty regions. Regions with strong sky absorption have been masked out. In most case this range samples the following most prominent features: [OII] λλ3726,3729 emission lines, the CA H+K λλ3934, 3969 absorption lines, the 4000    Å -break, the G-band around λ ~ 4300, the hydrogen lines Hδ λ4102 and Hβ λ4861, and [OIII] λλ4959,5007. Strong emission line galaxies were sampled in a region around the most prominent emission lines. Only the two mentioned templates were used to avoid systematic redshift shifts due to resolution between templates.

No redshift errors are indicated in Table 6, since the variance in redshift resulting from using different templates (i.e. other templates than the two mentioned above) is generally large (~0.001) because of the spectral resolution of the instrument, which is limited to δz ~ 0.01. The formal 1σ confidence intervals resulting from the χ2 template fit, however, are much smaller, typically <0.001. For emission line galaxies the errors are smallest and are given by our spectral wavelength calibration (~0.1Å), which results in δz ~ 0.00001. Line feature centroids can be determined with much higher precision than the instrumental resolving power. In Table 5, however, we merely present the formal 1σ errors.

Table 6

Overview of spectroscopic redshifts for all confirmed cluster members with distances from the X-ray centre  ≲1   Mpc/ ≲ 1.7   Mpc/>2.5   Mpc (white/grey/dark grey background).

Table 7

Optical identifications of extended sources.

5. Identifications of XMM-Newton cluster candidates

thumbnail Fig. 6

a) Colour–magnitude diagram for the extended X-ray source ext#1. CFTHLS D4 z′ and H-band images have been used. The small dots show all galaxies within D4 having cluster-concordant photometric redshifts, i.e. within the area between the dashed lines in Fig. 3. Open circles show all galaxies with good photometry <1 Mpc from the X-ray peak, of which the filled symbols have cluster-concordant photometric redshifts. Vertical dashes indicate the error in the colour. Star symbols indicate spectroscopic members. All symbols (apart from the dots) have been scaled, inversely, to their physical distance from the X-ray emission peak. In each panel, the horizontal line indicates the expected colour of luminous, red sequence galaxies at the given spectroscopic redshift, as obtained from the simple stellar population models calibrated on data available in Fassbender et al. (2011a). b) CFTHLS D4 z′-band image centred on the extended X-ray sources in the field. The blue and red circles indicate the positions of the spectroscopically confirmed star forming and passive cluster members, respectively. The contours overlay represent linearly spaced photon count levels of the X-ray sources. North is up, east is to the right. The image is 1.2 arcmin across. c) Selected spectra for cluster ext#1 in the observers’ frame in black and their noise spectra in grey. Either an elliptical or emission template is overplotted in red. The positions of expected absorption and emission features have been indicated with blue dashed and red dotted lines, respectively. Vertical grey lines indicate regions affected by sky absorption. The spectra have been smoothed for clarity. The capital letters correspond to the entries in Table 6 and to the positions in Fig. 6(b). The redshift z and the distance d (arcsec) to the formal peak of the X-ray source are indicated in the plot.

Open with DEXTER

In the deep CFHT imaging six out of the nine extended X-ray sources can be identified with clusters of galaxies. All mentioned redshifts are mean averaged spectroscopically determined redshifts, unless stated otherwise. Table 7 gives the likely identifications with their photometric redshifts if available. Column “redshift X-ray” gives the 1σ range of the X-ray spectroscopic redshift estimates derived from kT − z contours (see Figs. A.1(a) − e). The cross-reference redshifts listed in column “redshifts publ.” refer to the optically selected catalogues from previous publication. In Figs. 6(a)24 the colour–magnitude diagrams Hmag vs. (z′ − H)mag for all X-ray positions are displayed, which are discussed qualitatively for each position individually. The terms red sequence and the expected colour of a cluster are used invariably, since we assume, as a zeroth-order approximation, a zero gradient of z′ − H (mag) as a function of relative brightness. The values of the empirically determined cluster colour as a function of redshift are taken from Fassbender et al. (2011a). We chose to remain with a single colour, i.e. z′ − H, in order to illustrate its effectiveness thereof for the redshift’s range 0.3 < z < 1.5 and to be able to relate our results to Fassbender et al. (2011a; their Fig. 3). Yet, a single colour does not always optimally sample the 4000    Å  break, however, the chosen filter combination is a monotonic function of the red sequence of passively evolving galaxies, hence sufficient for our purpose.

5.1. XMMU J221536.8-174534 (ext #1) z = 0.41

Nearly on axis we find the brightest extended source at a redshift of z = 0.408. The position of brightest galaxy coincides with the X-ray emission peak, as can be expected for relaxed clusters at relatively low redshifts. The extended X-ray source has been described as alikely cluster of galaxies by Clavel et al. (2006). This is confirmed by our optical imaging. The X-ray spectrum (see Fig. A.1(a)) is consistent with the photometric redshift z = 0.4 as found in the MF searches by Olsen et al. (2007) and Dietrich et al. (2007). The CFHT z′-band image with X-ray contours and spectroscopic members is shown in Fig. 6(b). The cluster has a well-evolved red sequence as seen in Fig. 6(a) consistent with the empirically expected z′ − H colour. Six of its 11 spectra are presented in Fig. 6(c).

In Fig. 4 system ext#1 is associated with two over-density contours: the purple ones at z ~ 0.3 and the greens ones at z ~ 0.45. It is possible, therefore, that the X-ray emission is a superposition of two clusters at different redshifts. From the X-ray redshift-temperature contours and from the positioning of galaxies with respect to the X-ray emission peak, however, we are able to conclude that the X-ray emission originates in the more distant redshift of the two. The structure at z ~ 0.3 was detected only as galaxy over-density and is not regarded as contributing to the X-ray emission. Additionally, we retrieved two spectroscopic redshifts at z ~ 0.34 at about 50 arcsec from the peak of source ext#1, which are listed as object O and P in the ext#3 column in Table 6.

5.2. XMMU J221558.6-173810 (ext #2) z = 1.45

The most distant galaxy cluster in this field has already been reported by Stanford et al. (2006). It has been the topic of several dedicated studies since then, e.g. studying star formation at high redshift (Hilton et al. 2010; Hayashi et al. 2011). This is one of the most distant known X-ray selected cluster of galaxies with unambiguously strong X-ray emission, albeit with a confirmed contamination from point sources (Hilton et al. 2009). In addition to earlier publications, we find several galaxies with photometric redshifts compatible with z ~ 1.45. The over-density contours in Fig. 4 for this cluster are displayed with the heat colour scheme. An CFHT z′-band image is shown in Fig. 8 with X-ray contour overlays and the positions of three new quasars indicated, of which two reside at the cluster redshift. The colour − magnitude diagram in Fig. 7 shows that a red sequence is already present at this high redshift, following the empirically expected colour. We have not obtained additional galaxy spectra in our observation runs.

However, we serendipitously identified two quasars at the nominal cluster redshift on the very outskirts of the cluster (see Fig. 8). The spectra are presented in Fig. 9 (identifiers A and B). For future spectroscopic X-ray analysis, these X-ray sources do not need to be included as power laws contributing to the X-ray flux since they lay beyond 1 arcmin, i.e. farther than the spectral extracting radius. In a joint Chandra-XMM-Newton analysis, Hilton et al. (2010) report a temperature of 4.1 kT with the inclusion of three power laws to account for point source contributions. In contrast, our analysis, as stated before, assumes the emission to result from only a single hot plasma. The extraction region used by Hilton et al. (2010) is only mildly (7 arcsec) larger than the one we used.

5.3. XMMU J221500.9-175038 (ext #3) z = 0.34

At a mean redshift z = 0.338 cluster ext#3 is the closest galaxy cluster in the field. This cluster was also detected in the MF searches by Olsen et al. (2007) (z ~ 0.3) and Dietrich et al. (2007). We show the CFHT z′-band image with X-ray emission contours and labelled spectroscopic members in Fig. 11. The red sequence is easily visible in Fig. 10 at the colour expected for its redshift. Its selected 14 spectroscopic members are within  | z − zcluster |  ≲ 0.006. Figure 12 shows six of the confirmed members.

In the same redshift bin as cluster ext#3 (green contours in Fig. 4) we find an over-density of galaxies virtually at the same projected position on the sky as the central cluster ext#1. One interpretation is to consider these over-densities to be a sign of knots in the cosmic web  ~3 Mpc apart; that is, the two galaxy over-densities are part of the same large-scale structure. There is no X-ray detection from a group of galaxies at z = 0.338 at the coordinates of cluster ext#1. Namely, the X-ray flux is compatible with only coming from source ext#1 as discussed in Sect. 5.1.

5.4. XMMU J221557.5-174029 (ext #4) z = 1.00

On the very edge of the redshift unity barrier we find a new cluster with 13 members. This cluster has already been referred to by Adami et al. (2010) as a photometric cluster candidate; however, it appears in no other catalogue. Two members are at the very centre of the X-ray emission, and both can be considered BCG, since their fluxes are similar.

The red sequence is visible in Fig. 13 without deviations from the expected z′ − H colour. Striking, however is the absence of a clear magnitude gap between the BCGs and the other galaxies, which could indicate a recently formed system. The spectra, a selection of six from 13, are presented in Fig. 15. The two BCGs have spectroscopic redshifts Δz = 0.003, which is well beyond their formal (1σ) errors. The cluster redshift of z ~ 1.0 is consistent with the estimate from the X-ray spectrum. An image of the cluster is shown in Fig. 14 with X-ray contours and confirmed members. The apparent point-like morphology of the X-ray emission is purely due to chip gaps masking out the cluster edges.

5.5. XMMU 221556.6-175139 (ext #5) z = 1.23

The optical imaging shows a conglomeration of both nearby and distant galaxies. The CFHT z′-band image is shown in Fig. 17 with X-ray contour and spectroscopically confirmed cluster members. This cluster has already been hinted at by Bielby et al. (2010), however, at a photometric redshift of z = 1.17. The colour–magnitude of the cluster is somewhat peculiar as seen in Fig. 16. The galaxy closest to the X-ray emission peak is blue and star-forming. Its membership is confirmed both spectroscopically and photometrically. Although the colour of the red members is as expected, we note there seems to be a significant ongoing build up of the red sequence. These properties are similar to the cluster found at z = 1.56 by Fassbender et al. (2011b). This second farthest cluster in the field has 12 concordant redshifts with  | z − zcluster |  ≲ 0.013. Figure 18 shows a selection of six members. Mask offsets6 during observation caused the signal-to-noise to remain very low. However, due to the existence of single emission lines in the very centre of the clusters, the redshift could be established as z = 1.227. A redshift based only on one emission line remains tentative; however, the morphology of the galaxy in combination with its extent on the sky does not render it a likely possibility that we are seeing a background high-z Lyα emitter.

For this source the X-ray spectrum indicates a redshift lower limit of z > 0.9. Both the X-ray and galaxy over-density contours show an NW-SE elongation.

thumbnail Fig. 7

As in Fig. 6(a) for source ext#2 at z = 1.45.

Open with DEXTER

thumbnail Fig. 8

As in Fig. 6(b) for source ext#2. The cluster is NE of the image centre. No new spectra have been obtained for this publication, apart from the three labelled X-ray point sources, whose optical spectra are shown in Fig. 9.

Open with DEXTER

thumbnail Fig. 9

Spectra (black) of the quasars discovered in the vicinity of cluster ext#2. In red a quasar template is overplotted. Objects a and b can be regarded as cluster members or rather active galaxies in the cluster outskirts. Object c is an unrelated distant X-ray luminous quasar serendipitously found in this study (Lamer et al. 2011). The labels correspond to the objects identified in Fig. 8. Spectral features, etc., are coded as in Fig. 6(c). In the case of a the emission lines are Lyα and CIV.

Open with DEXTER

thumbnail Fig. 10

As in Fig. 6(a) for source ext#3 at z = 0.34.

Open with DEXTER

thumbnail Fig. 11

As in Fig. 6(b) for source ext#3. The associated optical spectra are found in Fig. 12.

Open with DEXTER

thumbnail Fig. 12

Selected spectra for cluster ext#3 in the observers’ frame. See Fig. 6(c) for explanation. The positions are seen in Fig. 11.

Open with DEXTER

thumbnail Fig. 13

As in Fig. 6(a) for cluster ext#4 at z = 1.00.

Open with DEXTER

thumbnail Fig. 14

As in Fig. 6(b) for source ext#4. The associated optical spectra are found in Fig. 15.

Open with DEXTER

thumbnail Fig. 15

Selected spectra for cluster ext#4 in the observers’ frame. Colours and labels are as in Fig. 6(c). Labels refer to identified objects in Fig. 14. The spectra have been smoothed.

Open with DEXTER

thumbnail Fig. 16

As in Fig. 6(a) for cluster ext#5 at z = 1.23.

Open with DEXTER

thumbnail Fig. 17

As in Fig. 6(b) for source ext#5. The associated spectra are seen in Fig. 18.

Open with DEXTER

thumbnail Fig. 18

Selected spectra for cluster ext#5 in the observers’ frame. Colours and labels are as in Fig. 6(c). The positions are seen in Fig. 17.

Open with DEXTER

thumbnail Fig. 19

As in Fig. 6(a) for source ext#6 at z = 0.41.

Open with DEXTER

5.6. XMMU 221503.6-175215 (ext#6) z = 0.41

The cluster with the weakest X-ray emission at a redshift of z = 0.410 can easily be understood as a sign of large-scale structure. Cluster ext#6 possibly is a knot in the filaments around the more luminous, hence more massive structure ext #1. This work provides the first reference to this cluster. The colour–magnitude diagram in Fig. 19 shows a well-evolved red sequence, which both confirms the maturity of the system and rejects the possibility of a chance-alignment. Since the colour is virtually identical to that of cluster ext#1, we have established further proof for the common redshift of the two clusters. We find 14 spectroscopic members (see Fig. 20) for the clusters of which six are plotted in Fig. 21.

thumbnail Fig. 20

As in Fig. 6(b) for source ext#6. The associated spectra are found in Fig. 21.

Open with DEXTER

thumbnail Fig. 21

Selected spectra for cluster ext#6 in the observers’ frame. Colours are as in Fig. 6(c). Labels refer to identified objects in Fig. 20.

Open with DEXTER

The claim to perceive cluster ext#6 (Fig. 4) as part of the larger structure around system ext#1 is supported by the fact that we see emission line galaxies close to the centre of the cluster. Three of the five galaxies nearest to the BCG (<1′) show at least Hα in emission. Evolved clusters are not expected to show strong signs of star formation, yet in the outskirts of the largest dark matter potentials, such as cluster ext#6, we do expect galaxies to be more active.

5.7. XMMU 221546.2-174002 (ext#7) and XMMU 221551.7-173918 (ext#8)

No clear cluster identification for these X-ray sources can be found in the optical data. Their vicinity on the sky makes it is natural to discuss these structures in symbiosis. Remarkably, neither in the very deep CFHT imaging nor in the WIRCam NIR data do we find a plausible visible counterpart for these sources as seen in Fig. 23, which rules out the possibility of a very distant cluster of galaxies within any reasonable redshift limit (i.e. z ≲ 2). The colour–magnitude diagrams as displayed in Figs. 22(a) and b do not show any hint of a red sequence.

Source ext#7 has the lowest value for EXT_LIKE (extent likelihood in Table 3) in this sample and could be a blend of faint point sources and/or associated with source ext#8. The NVSS radio catalogue (Condon et al. 1998) contains a faint radio source (3.3 mJy at 1.4 GHz) at 18 arcsecond distance of source ext#8. The projection of the two object on the sky (NW versus SE with respect to the centre of Fig. 23) raises the idea of an associated bipolar phenomenon.

Furthermore, that both sources were not detected in the cluster’s typical energy band (0.35–2.4 keV, see Sect. 2.1) suggests they may possess a relatively hard spectrum, which in turn favours another mechanism at work rather than bremsstrahlung.

Based on the extended nature of the sources, the radio detection, the hardness of the spectrum, as well as the absence of a clear optical counterpart(s), little alternative is left but to identify source ext#8 (and source ext#7) as a Compton ghost (Fabian et al. 2009).

5.8. XMMU 221624.3-173321 (ext#9)

This source is excluded from further (cosmological) analysis, both for its being at a large off-axis angle, namely 16 arcmin (i.e. covered by theEPIC PN camera only), and its likely being a point source. Nonetheless, at about 1 arcmin from source ext#9 there is a complex X-ray structure (at an off-axis angle of 16 arcmin as well), which was not flagged as extended by our source detection. The two sources are displayed in Fig. 25, centred on the second source, while source ext#9 is SW towards the bottom. The structure at RA 22:16:20.0 Dec  − 17:32:25 has also been found by our photometric over-density search. It was proposed that it is at z ~ 0.7 in the MF sample of Olsen et al. (2007) and in the weak lensing sample of Dietrich et al. (2007).

The precise coordinates of the cluster centre (hence the BCG) are problematic to constrain. The position of the X-ray emission peak is not known exactly due to contaminating sources and because it is at the veryEPIC camera’s chip edge. Furthermore, all optical over-density methods have an arcmin-like spatial resolution (see dark pink contours in Fig. 4). Therefore, a tentative BCG is selected at z = 0.71 ± 0.07, with the need to be confirmed by spectroscopy in the future. The tentative colour–magnitude diagram in Fig. 24 is consistent with a redshift of z = 0.7.

6. Cluster counts

6.1. X-ray sensitivity function

The advantage of using only X-rays to find clusters of galaxies is the grasp one has on the selection function. In a rather straightforward manner the X-ray behaviour of a cluster sample can be mimicked and convolved with a detector response, thereby obtaining the detection limits for various physical parameters. In this study, the survey completeness is determined using a sensitivity function, which is generated by means of Monte-Carlo simulations. The details of the simulations are presented in Mühlegger (2010) and discussed in Fassbender et al. (2011a). An abbreviated description is found in this section.

The probability of detecting extended X-ray sources at a (1) certain flux limit and (2) a given distance from the optical axis, is computed for ten simulated beta models () inserted randomly on concentric rings in the stacked XMM-Newton pointings. During insertion, regions that already contain an extended source are avoided. If this were not done, we would introduce an unphysical overlap between clusters at this faint end of the flux scale, which will result in an over-estimation of the source density when correcting for this unrealistic projection effect. The simulated clusters cover a range of 25 bins in both core radii and photon counts. This results in 3   125 simulations, since the procedure is repeated five times. The core radii distribution assumed is taken from Vikhlinin et al. (1998). The cluster sample in Vikhlinin et al. (1998) consists of 200 clusters observed with ROSAT with redshift information. The redshifts range from 0.015–0.73 with  ⟨ z ⟩  = 0.28. The advantage of this particular sample is that the core radius is computed by applying β = 0.67 as in our analysis. The core radii derived are scaled to how they would appear at z = 1 using the geometrical angular distance dependency on redshift. No further evolution is assumed. The mean core radius at z = 1 is  ~ 14 arcsec.

thumbnail Fig. 22

Colour magnitude diagrams around the two extended X-ray sources showing no galaxy over-density or obvious red sequence. Open circles denote all galaxies with good photometry within 1 arcmin from the X-ray centroid, scaled inversely with their distance to this centroid. Likewise, the dots represent objects within 2 arcmin.

Open with DEXTER

Source detection is performed for every simulation with XMM-SAS v.6.5 in the multiple energy bands simultaneously with the same parameters as mentioned in Sect. 2.1. The resulting image displaying the survey completeness level as a function of limiting flux (Slim) and off-axis angle (θ) is shown in Fig. 26(a). We notice that at the indicated completeness levels, the curve is rather flat between ca. 2–12 arcmin and in particular the 50% curve, which we take as the completeness level for our survey. Towards the centre of the pointing, however, an extended structure (cluster ext#1) affects the sensitivity in this observation, while on the outskirts the vignetting and the very end of the chip catastrophically affect the detectability of clusters. At the 50% level, corresponding to a flux of 2.5 × 10-15ergcm-2   s-1, we are complete. All clusters detected have (source detection) fluxes above this level. For our log N − log S analysis, none of the objects ext#1-#6 needs to be excluded: (10)Dividing the LBQS 2212-1759 observation into rings, we can fold the geometric area with the survey sensitivity at each off-axis angle θ. We limit the radius to  ≤ 12 arcmin. Beyond this radius the sensitivity drops significantly as seen in Fig. 26(a). This area restriction also ensures the coverage by all threeEPIC cameras. One extended source (ext#9) is excluded as a result of this angle limit. The cumulative effective surface area, or detection probability weighted area A(Slim), computed following Eq. (10)is displayed in Fig. 26(b). Owing to our limit on the maximum radius, we obtain an survey area of  ~0.13   deg2. Slim denotes the limiting flux. This is the correction curve to be convolved with the clusters’ number counts in the purely geometrical survey area, resulting in the survey-independent cosmological cluster count in flux bins.

6.2. log N–log S

The computed sensitivity function is used to predict the number of X-ray luminous clusters in a certain survey area. We convolve the probability of detecting clusters in concentric rings around the focal axis with the enclosed area (Fig. 26(b)). The observed flux distribution is normalised to the probability weighted survey area to obtain the expected number of galaxy clusters per flux bin. The log N − log S graph is presented in Fig. 27. The empirical values are compared to the expected numbers simulated by Rosati et al. (2002, 1998), which in turn proved to be a good approximation for the extended X-ray source distribution in the COSMOS (Finoguenov et al. 2007) and the Subaru-XMM-Newton deep field (SXDF, Finoguenov et al. 2010).

When making the link between the derived flux limits and our measured fluxes, we have to bear in mind that the simulated clusters cover a range of core radii as they appear at z = 1. Therefore we compute S500 for hypothetical clusters at z = 1 with rc = 14 arcsec, using the iterative approach, which spans the flux range from our simulations. The mean conversion factor between Slim and S500 is 0.8, which we apply to our limiting fluxes.

The intrinsic errors resulting from the method used to obtain S500 do not enter into Fig. 27, but only the spectral flux errors. Furthermore, since the distribution of clusters on the sky is Poissonian, we add an additional error of on the number counts.

The number counts in flux bins from this study are, among others and within errors, compatible with the findings from the SXDF and COSMOS. Likewise, using spectral or source detection fluxes has no dramatic effect on the results. For low-flux objects, the scenario in which there is no evolution in the XLF for extended sources (Rosati et al. 2002) is favoured.

thumbnail Fig. 23

Source ext#7 and source ext#8. Stack of NIR HJK bands to enhance depth for the two unknown structures. Source ext#8 is NW and source ext#7 SE of the image centre as pointed out by the arrows. The grey contours display the NVSS radio source. Dimension and orientation is as in Fig. 6(b).

Open with DEXTER

thumbnail Fig. 24

As in Fig. 6(a) for centre coordinates as described in Sect. 5.8 close to source ext#9. The difficulty to pin-point the cluster centre results in an only vague red sequence.

Open with DEXTER

thumbnail Fig. 25

As in Fig. 6(b), however, the image is centred on the cluster of galaxies found close to source ext#9 (see Sect. 5.8). The extended source from our source detection is SE of the image centre. The black line indicates the end of the XMM-Newton EPIC chip.

Open with DEXTER

thumbnail Fig. 26

a) Survey completeness as a function of off-axis angle and limiting flux. The 10%, 50% and 90% levels are indicated. The dashed red line indicates the restriction put on the off-axis angle. b) Cumulatively sensitivity-weighted surface areas as a function of limiting flux, integrated over all off-axis angles. The short-dashed red line indicates the deduced limiting flux of 2.5 × 10-15ergcm-2   s-1 at which 50% completeness is reached as indicated in a).

Open with DEXTER

thumbnail Fig. 27

Cosmological log N − log S for clusters of galaxies. The black discrete curve shows the confidence interval resulting from measured source detection flux errors and an additional on the number count from this work. A comparison to comparable surveys is made. The curves show the model from Rosati et al. (2002) for no evolution (dashed). The dot-dashed and dotted lines show the measured ROSAT brightest cluster sample (BCS, Ebeling et al. 1998) clusters and their extrapolation, respectively. The discrete lines display the findings from COSMOS (solid) (Finoguenov et al. 2007), SXDF (dashed) (Finoguenov et al. 2010), and XMM-BCS (dotted) (Šuhada et al. 2012).

Open with DEXTER

7. Summary

This work presents the detection and study of extended X-ray sources in a deep (250 ks) XMM-Newton field covering a FoV of  ~ 14 arcmin across. The nature of the sources has been identified. We have valiantly achieved “five at one blow”, since this single field has allowed us to discover five new spectroscopically confirmed clusters. In the following we summarise the most essential findings from this study.

  • We performed a robust source detection on deep, stackedXMM-Newton observations targettingLBQS 2212-1759, revealing nine extended sources withconservative detection thresholds. Six of these sourcesprove to be clusters of galaxies, five of which are new spec-troscopically confirmed discoveries from this work alone.The sixth detected cluster was already known from Stanfordet al. (2006). The first uncon-firmed extended source, at >14 arcmin off-axis, is a confusedpoint source. The remaining two sources could not be identi-fied. Since the detection of these sources is dominated by thehard energy bands, this suggests that another mechanism thanbremsstrahlung is responsible for the X-ray emission fromthese two sources, which possibly leads at least one of them to be aCompton ghost.

  • A complementary qualitative optical/IR completeness study, allowed by very deep CFHTLS data (i.e. ), has confirmed that all X-ray luminous clusters are associated with an optical over-density, in agreement with the spectroscopic redshift. One new, significant over-density could not be correlated with any X-ray emission, hinting at a projection effect in the optical search method.

  • The X-ray selection function has been computed through extensive Monte-Carlo simulations. The approach of retrieving generated β models inserted into the observations has given us a 50% completeness flux limit of  ~ 2.5 × 10-15   ergs-1 over most of the FoV. We constructed a log N − log S based on a complete sample of spectroscopically confirmed sources down to weak fluxes in a precisely determined survey area with a known selection function. Our result agrees with the findings from the COSMOS and the SXDF. Comparing the number counts with the models from Rosati et al. (2002), our data seems to favour no evolution in the XLF in the faint flux range we covered.

  • By spectroscopic follow-up of some X-ray point sources, we serendipitously discovered three new AGN. Two of these objects reside at the concordant redshift of the cluster (Stanford et al. 2006) at z = 1.45. The other quasar has a redshift of z = 4.20, so is among the handful of most distant X-ray-selected AGN known today (Fiore et al. 2012).

In this deep field study, good photometric data have allowed us to scan the sky in multiple wavelengths to study the galaxy counterparts of extended X-ray emission. We emphasise that photometric redshifts with errors Δz < 0.1 are still too expensive to date for routinely performing this kind of analysis. With respect to the upcoming X-ray mission eROSITA (Predehl et al. 2011), future optical surveys such as LSST (Ivezic et al. 2008), the ESO/Vista project VHS (Arnaboldi et al. 2007)7, and PanStarrs (Kaiser et al. 2002) will obtain sufficient depth and cover enough area to identify a wide range of distant and faint X-ray clusters. This work, therefore, could very well be extended to a much larger area of the sky within the next decade.

In the near future, as the photometric and spectroscopic follow-up campaigns for the XDCP is becoming complete, we are already compiling catalogues of distant clusters to extend the present work.


1

XMM-SAS version 6.5 was the prototypical release used to construct the 2XMM catalogue (Watson et al. 2009).

2

The complete source list of point source is found in Table 2 (available at the CDS). The extended source, however, are listed in Table 3.

4

The full procedure of data reduction and calibration will be described in Brimioulle et al. (in prep.), see also Brimioulle et al. (2008).

5

L = 10 corresponds to a significance >99.9% or roughly 5σ in Gaussian statistics.

6

Due to accidental pointing errors.

7

Data sharing agreement with DES (Dark Energy Survey).

Acknowledgments

A.d.H. was supported by the DFG under fund no. Schw536/24-2. GL acknowledges support by the Deutsches Zentrum für Luft- und Raumfahrt (DLR) under contract no. FKZ 50 OX 0201 and 50 QR 0802. J.P.D. is supported by by the German Science Foundation (DFG) through the Transregio 33 and the Cluster of Excellence “Origin and Structure of the Universe”, funded by the Excellence Initiative of the Federal Government of Germany, EXC project number 153. D.P. acknowledges the kind hospitality at the MPE. This research made use of (1) APLpy, an open-source plotting package for Python hosted at http://aplpy.github.com and (2) IPython, an interactive python shell (Pérez & Granger 2007). We are grateful to the CFHT survey team for conducting the observations and the TERAPIX team for developing software used in this study. We acknowledge use of the Canadian Astronomy Data Centre, which is operated by the Dominion Astrophysical Observatory for the National Research Council of Canada’s Herzberg Institute of Astrophysics. It is based on observations obtained with WIRCam, a joint project of CFHT,Taiwan, Korea, Canada, France, at the Canada-France-Hawaii Telescope (CFHT) which is operated by the National Research Council (NRC) of Canada, the Institute National des Sciences de l’Univers of the Centre National de la Recherche Scientifique of France, and the University of Hawaii. This work is based in part on data products produced at TERAPIX, the WIRDS (WIRcam Deep Survey) consortium, and the Canadian Astronomy Data Centre. This research was supported by a grant from the Agence Nationale de la Recherche ANR-07-BLAN-0228. This research has made use of the NASA/IPAC Extragalactic Database (NED), which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. M.L. thanks the European Community for the Marie Curie research training network “DUEL” doctoral fellowship MRTN-CT-2006-036133.

References

  1. Adami, C., Durret, F., Benoist, C., et al. 2010, A&A, 509, A81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  2. Andreon, S., Maughan, B., Trinchieri, G., & Kurk, J. 2009, A&A, 507, 147 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  3. Arnaboldi, M., Neeser, M. J., Parker, L. C., et al. 2007, The Messenger, 127, 28 [NASA ADS] [Google Scholar]
  4. Bender, R., Appenzeller, I., Böhm, A., et al. 2001, in Deep Fields, eds. S. Cristiani, A. Renzini, & R. E. Williams, 96 [Google Scholar]
  5. Bertin, E., & Arnouts, S. 1996, A&AS, 117, 393 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Bielby, R. M., Finoguenov, A., Tanaka, M., et al. 2010, A&A, 523, A66 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  7. Böhringer, H., Schuecker, P., Guzzo, L., et al. 2004, A&A, 425, 367 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  8. Böhringer, H., Mullis, C., Rosati, P., et al. 2005, The Messenger, 120, 33 [NASA ADS] [Google Scholar]
  9. Brimioulle, F., Lerchster, M., Seitz, S., Bender, R., & Snigula, J. 2008, unpublished [arXiv:0811.3211] [Google Scholar]
  10. Cash, W. 1979, ApJ, 228, 939 [NASA ADS] [CrossRef] [Google Scholar]
  11. Clavel, J., Schartel, N., & Tomas, L. 2006, A&A, 446, 439 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  12. Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ, 115, 1693 [NASA ADS] [CrossRef] [Google Scholar]
  13. Coupon, J., Ilbert, O., Kilbinger, M., et al. 2009, A&A, 500, 981 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  14. Dietrich, J. P., Miralles, J.-M., Olsen, L. F., et al. 2006, A&A, 449, 837 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  15. Dietrich, J. P., Erben, T., Lamer, G., et al. 2007, A&A, 470, 821 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  16. Ebeling, H., Edge, A. C., Böhringer, H., et al. 1998, MNRAS, 301, 881 [NASA ADS] [CrossRef] [Google Scholar]
  17. Erben, T., Hildebrandt, H., Lerchster, M., et al. 2009, A&A, 493, 1197 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  18. Fabian, A. C., Chapman, S., Casey, C. M., Bauer, F., & Blundell, K. M. 2009, MNRAS, 395, L67 [NASA ADS] [Google Scholar]
  19. Fassbender, R., Böhringer, H., Nastasi, A., et al. 2011a, New J. Phys., 13, 125014 [NASA ADS] [CrossRef] [Google Scholar]
  20. Fassbender, R., Nastasi, A., Böhringer, H., et al. 2011b, A&A, 527, L10 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Finoguenov, A., Guzzo, L., Hasinger, G., et al. 2007, ApJS, 172, 182 [NASA ADS] [CrossRef] [Google Scholar]
  22. Finoguenov, A., Watson, M. G., Tanaka, M., et al. 2010, MNRAS, 403, 2063 [NASA ADS] [CrossRef] [Google Scholar]
  23. Fiore, F., Puccetti, S., Grazian, A., et al. 2012, A&A, 537, A16 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  24. Garilli, B., Fumana, M., Franzetti, P., et al. 2010, PASP, 122, 827 [NASA ADS] [CrossRef] [Google Scholar]
  25. Gavazzi, R., & Soucail, G. 2007, A&A, 462, 459 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  26. Gobat, R., Daddi, E., Onodera, M., et al. 2011, A&A, 526, A133 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  27. Grove, L. F., Benoist, C., & Martel, F. 2009, A&A, 494, 845 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  28. Hayashi, M., Kodama, T., Koyama, Y., Tadaki, K.-I., & Tanaka, I. 2011, MNRAS, 415, 2670 [NASA ADS] [CrossRef] [Google Scholar]
  29. Henry, J. P., Salvato, M., Finoguenov, A., et al. 2010, ApJ, 725, 615 [NASA ADS] [CrossRef] [Google Scholar]
  30. Hilton, M., Stanford, S. A., Stott, J. P., et al. 2009, ApJ, 697, 436 [NASA ADS] [CrossRef] [Google Scholar]
  31. Hilton, M., Lloyd-Davies, E., Stanford, S. A., et al. 2010, ApJ, 718, 133 [NASA ADS] [CrossRef] [Google Scholar]
  32. Ilbert, O., Arnouts, S., McCracken, H. J., et al. 2006, A&A, 457, 841 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  33. Ivezic, Z., Tyson, J. A., Acosta, E., et al. 2008, unpublished [arXiv:0805.2366] [Google Scholar]
  34. Kaiser, N., Aussel, H., Burke, B. E., et al. 2002, in SPIE Conf. Ser. 4836, eds. J. A. Tyson, & S. Wolff, 154 [Google Scholar]
  35. Kinney, A. L., Calzetti, D., Bohlin, R. C., et al. 1996, ApJ, 467, 38 [NASA ADS] [CrossRef] [Google Scholar]
  36. Lamer, G., Schwope, A., Böhringer, H., et al. 2006, in The X-ray Universe 2005, ed. A. Wilson, ESA SP, 604, 729 [NASA ADS] [Google Scholar]
  37. Lamer, G., de Hoon, A., Fassbender, R., et al. 2011, in The X-ray Universe 2011, eds. J.-U. Ness, & M. Ehle, 239 [Google Scholar]
  38. Lerchster, M., Seitz, S., Brimioulle, F., et al. 2011, MNRAS, 411, 2667 [NASA ADS] [CrossRef] [Google Scholar]
  39. Lloyd-Davies, E. J., Romer, A. K., Mehrtens, N., et al. 2011, MNRAS, 418, 14 [NASA ADS] [CrossRef] [Google Scholar]
  40. Mignano, A., Miralles, J.-M., da Costa, L., et al. 2007, A&A, 462, 553 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  41. Mühlegger, M. 2010, Ph.D. Thesis, Technischen Universität München [Google Scholar]
  42. Mullis, C. R., Rosati, P., Lamer, G., et al. 2005, ApJ, 623, L85 [NASA ADS] [CrossRef] [Google Scholar]
  43. Nastasi, A., Fassbender, R., Böhringer, H., et al. 2011, A&A, 532, L6 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  44. Olsen, L. F., Benoist, C., Cappi, A., et al. 2007, A&A, 461, 81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  45. Papovich, C., Momcheva, I., Willmer, C. N. A., et al. 2010, ApJ, 716, 1503 [NASA ADS] [CrossRef] [Google Scholar]
  46. Pérez, F., & Granger, B. E. 2007, Comput. Sci. Eng., 9, 21 [CrossRef] [Google Scholar]
  47. Pierre, M., Clerc, N., Maughan, B., et al. 2012, A&A, 540, A4 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  48. Postman, M., Lubin, L. M., Gunn, J. E., et al. 1996, AJ, 111, 615 [NASA ADS] [CrossRef] [Google Scholar]
  49. Predehl, P., Andritschke, R., Becker, W., et al. 2011, in SPIE Conf. Ser., 8145 [Google Scholar]
  50. Reichert, A., Böhringer, H., Fassbender, R., & Mühlegger, M. 2011, A&A, 535, A4 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  51. Rosati, P., della Ceca, R., Norman, C., & Giacconi, R. 1998, ApJ, 492, L21 [NASA ADS] [CrossRef] [Google Scholar]
  52. Rosati, P., Borgani, S., & Norman, C. 2002, ARA&A, 40, 539 [NASA ADS] [CrossRef] [Google Scholar]
  53. Santos, J. S., Fassbender, R., Nastasi, A., et al. 2011, A&A, 531, L15 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  54. Schwope, A. D., Lamer, G., Burke, D., et al. 2004, Adv. Space Res., 34, 2604 [NASA ADS] [CrossRef] [Google Scholar]
  55. Stanford, S. A., Romer, A. K., Sabirli, K., et al. 2006, ApJ, 646, L13 [NASA ADS] [CrossRef] [Google Scholar]
  56. Stanford, S. A., Brodwin, M., Gonzalez, A. H., et al. 2012, ApJ, 753, 164 [NASA ADS] [CrossRef] [Google Scholar]
  57. Takey, A., Schwope, A., & Lamer, G. 2011, A&A, 534, A120 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  58. Tanaka, M., Finoguenov, A., & Ueda, Y. 2010, ApJ, 716, L152 [NASA ADS] [CrossRef] [Google Scholar]
  59. Šuhada, R., Song, J., Böhringer, H., et al. 2012, A&A, 537, A39 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  60. Vikhlinin, A., McNamara, B. R., Forman, W., et al. 1998, ApJ, 502, 558 [NASA ADS] [CrossRef] [Google Scholar]
  61. Vikhlinin, A., Burenin, R. A., Ebeling, H., et al. 2009, ApJ, 692, 1033 [NASA ADS] [CrossRef] [Google Scholar]
  62. Watson, M. G., Schröder, A. C., Fyfe, D., et al. 2009, A&A, 493, 339 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]

Appendix A: X-ray properties and the quality of the spectral fits

In this section we present the contour plots displaying the confidence intervals of the redshift-temperature degeneracy. This method has driven the optical photometric follow-up programme.

thumbnail Fig. A.1

Confidence contours (68%, 90%) in redshift-temperature space for source ext#1 a), source ext#2 b), source ext#3 c), source ext#4 d), and source ext#5 e).

Open with DEXTER

All Tables

Table 1

Summary of XMM-Newton observations.

Table 3

Extended X-ray sources properties from source detection.

Table 4

Extended X-ray sources.

Table 6

Overview of spectroscopic redshifts for all confirmed cluster members with distances from the X-ray centre  ≲1   Mpc/ ≲ 1.7   Mpc/>2.5   Mpc (white/grey/dark grey background).

Table 7

Optical identifications of extended sources.

All Figures

thumbnail Fig. 1

StackedEPIC image in the 0.2–2.0 keV band with detected point sources (small circles) and extended sources (large circles) labelled from 1 to 9.

Open with DEXTER
In the text
thumbnail Fig. 2

Sky coverage of imaging campaigns.Black: the optical 1   deg2 CFHTLS D4 field. Orange: the BVRI sky projection as imaged by the WFI. Red: the NIR WIRcam supplementary coverage. Cyan: the NIR observations made at Calar Alto. Blue: XMM-Newton pointings. The dashed circles represent the 2007 study (P.I. K. Nandra) covering the entire D4 in X-rays with  ~ 30 − 40ks exposures (Bielby et al. 2010). Green: VLT-FORS2 pointings used for initial detection. The crosses indicate the position of all clusters found.

Open with DEXTER
In the text
thumbnail Fig. 3

Photometric redshifts, plotted against the spectroscopic ones. The vertical dashes are the photometric redshift errors. The dashed lines are for zphot = zspec ± 0.15   (1 + zspec).

Open with DEXTER
In the text
thumbnail Fig. 4

Over-density contours of galaxies within the field of view of the XMM-Newton pointing. The background image is the stacked image of all XMM-Newton observations of LBQS2212 (0.5–2.0 keV). The extended X-ray sources from Fig. 1 are indicated with their identifiers on the side (1 arcmin offset in RA). In smaller fonts the identifiers from other works are indicated either in black, if presented in this work as well, or in grey if not found by our method. Between round brackets the identifier from Bielby et al. (2010) is indicated and square brackets refer to the catalogue from Olsen et al. (2007). Curly brackets present the clusters from Adami et al. (2010). The red arrow points at an optical over-density, which has no clear counterpart in X-rays. The contours represent 5 Poissonian likelihood levels with 10 < L < Lmax, colour-coded, or rather colour-mapped, by redshift interval: purple z = 0.3 Lmax = 15, blue z = 0.35 Lmax = 20, green z = 0.45 Lmax = 20, dark pink z = 0.78 Lmax = 30, orange z = 0.98 Lmax = 15, red z = 1.18 Lmax = 20, heat z = 1.38 Lmax = 20, where z refers to , i.e. the middle of the bin.

Open with DEXTER
In the text
thumbnail Fig. 5

Histogram of all 231 science grade spectra. The insets show a zoom-in on the low (top) and high (bottom) redshift range.

Open with DEXTER
In the text
thumbnail Fig. 6

a) Colour–magnitude diagram for the extended X-ray source ext#1. CFTHLS D4 z′ and H-band images have been used. The small dots show all galaxies within D4 having cluster-concordant photometric redshifts, i.e. within the area between the dashed lines in Fig. 3. Open circles show all galaxies with good photometry <1 Mpc from the X-ray peak, of which the filled symbols have cluster-concordant photometric redshifts. Vertical dashes indicate the error in the colour. Star symbols indicate spectroscopic members. All symbols (apart from the dots) have been scaled, inversely, to their physical distance from the X-ray emission peak. In each panel, the horizontal line indicates the expected colour of luminous, red sequence galaxies at the given spectroscopic redshift, as obtained from the simple stellar population models calibrated on data available in Fassbender et al. (2011a). b) CFTHLS D4 z′-band image centred on the extended X-ray sources in the field. The blue and red circles indicate the positions of the spectroscopically confirmed star forming and passive cluster members, respectively. The contours overlay represent linearly spaced photon count levels of the X-ray sources. North is up, east is to the right. The image is 1.2 arcmin across. c) Selected spectra for cluster ext#1 in the observers’ frame in black and their noise spectra in grey. Either an elliptical or emission template is overplotted in red. The positions of expected absorption and emission features have been indicated with blue dashed and red dotted lines, respectively. Vertical grey lines indicate regions affected by sky absorption. The spectra have been smoothed for clarity. The capital letters correspond to the entries in Table 6 and to the positions in Fig. 6(b). The redshift z and the distance d (arcsec) to the formal peak of the X-ray source are indicated in the plot.

Open with DEXTER
In the text
thumbnail Fig. 7

As in Fig. 6(a) for source ext#2 at z = 1.45.

Open with DEXTER
In the text
thumbnail Fig. 8

As in Fig. 6(b) for source ext#2. The cluster is NE of the image centre. No new spectra have been obtained for this publication, apart from the three labelled X-ray point sources, whose optical spectra are shown in Fig. 9.

Open with DEXTER
In the text
thumbnail Fig. 9

Spectra (black) of the quasars discovered in the vicinity of cluster ext#2. In red a quasar template is overplotted. Objects a and b can be regarded as cluster members or rather active galaxies in the cluster outskirts. Object c is an unrelated distant X-ray luminous quasar serendipitously found in this study (Lamer et al. 2011). The labels correspond to the objects identified in Fig. 8. Spectral features, etc., are coded as in Fig. 6(c). In the case of a the emission lines are Lyα and CIV.

Open with DEXTER
In the text
thumbnail Fig. 10

As in Fig. 6(a) for source ext#3 at z = 0.34.

Open with DEXTER
In the text
thumbnail Fig. 11

As in Fig. 6(b) for source ext#3. The associated optical spectra are found in Fig. 12.

Open with DEXTER
In the text
thumbnail Fig. 12

Selected spectra for cluster ext#3 in the observers’ frame. See Fig. 6(c) for explanation. The positions are seen in Fig. 11.

Open with DEXTER
In the text
thumbnail Fig. 13

As in Fig. 6(a) for cluster ext#4 at z = 1.00.

Open with DEXTER
In the text
thumbnail Fig. 14

As in Fig. 6(b) for source ext#4. The associated optical spectra are found in Fig. 15.

Open with DEXTER
In the text
thumbnail Fig. 15

Selected spectra for cluster ext#4 in the observers’ frame. Colours and labels are as in Fig. 6(c). Labels refer to identified objects in Fig. 14. The spectra have been smoothed.

Open with DEXTER
In the text
thumbnail Fig. 16

As in Fig. 6(a) for cluster ext#5 at z = 1.23.

Open with DEXTER
In the text
thumbnail Fig. 17

As in Fig. 6(b) for source ext#5. The associated spectra are seen in Fig. 18.

Open with DEXTER
In the text
thumbnail Fig. 18

Selected spectra for cluster ext#5 in the observers’ frame. Colours and labels are as in Fig. 6(c). The positions are seen in Fig. 17.

Open with DEXTER
In the text
thumbnail Fig. 19

As in Fig. 6(a) for source ext#6 at z = 0.41.

Open with DEXTER
In the text
thumbnail Fig. 20

As in Fig. 6(b) for source ext#6. The associated spectra are found in Fig. 21.

Open with DEXTER
In the text
thumbnail Fig. 21

Selected spectra for cluster ext#6 in the observers’ frame. Colours are as in Fig. 6(c). Labels refer to identified objects in Fig. 20.

Open with DEXTER
In the text
thumbnail Fig. 22

Colour magnitude diagrams around the two extended X-ray sources showing no galaxy over-density or obvious red sequence. Open circles denote all galaxies with good photometry within 1 arcmin from the X-ray centroid, scaled inversely with their distance to this centroid. Likewise, the dots represent objects within 2 arcmin.

Open with DEXTER
In the text
thumbnail Fig. 23

Source ext#7 and source ext#8. Stack of NIR HJK bands to enhance depth for the two unknown structures. Source ext#8 is NW and source ext#7 SE of the image centre as pointed out by the arrows. The grey contours display the NVSS radio source. Dimension and orientation is as in Fig. 6(b).

Open with DEXTER
In the text
thumbnail Fig. 24

As in Fig. 6(a) for centre coordinates as described in Sect. 5.8 close to source ext#9. The difficulty to pin-point the cluster centre results in an only vague red sequence.

Open with DEXTER
In the text
thumbnail Fig. 25

As in Fig. 6(b), however, the image is centred on the cluster of galaxies found close to source ext#9 (see Sect. 5.8). The extended source from our source detection is SE of the image centre. The black line indicates the end of the XMM-Newton EPIC chip.

Open with DEXTER
In the text
thumbnail Fig. 26

a) Survey completeness as a function of off-axis angle and limiting flux. The 10%, 50% and 90% levels are indicated. The dashed red line indicates the restriction put on the off-axis angle. b) Cumulatively sensitivity-weighted surface areas as a function of limiting flux, integrated over all off-axis angles. The short-dashed red line indicates the deduced limiting flux of 2.5 × 10-15ergcm-2   s-1 at which 50% completeness is reached as indicated in a).

Open with DEXTER
In the text
thumbnail Fig. 27

Cosmological log N − log S for clusters of galaxies. The black discrete curve shows the confidence interval resulting from measured source detection flux errors and an additional on the number count from this work. A comparison to comparable surveys is made. The curves show the model from Rosati et al. (2002) for no evolution (dashed). The dot-dashed and dotted lines show the measured ROSAT brightest cluster sample (BCS, Ebeling et al. 1998) clusters and their extrapolation, respectively. The discrete lines display the findings from COSMOS (solid) (Finoguenov et al. 2007), SXDF (dashed) (Finoguenov et al. 2010), and XMM-BCS (dotted) (Šuhada et al. 2012).

Open with DEXTER
In the text
thumbnail Fig. A.1

Confidence contours (68%, 90%) in redshift-temperature space for source ext#1 a), source ext#2 b), source ext#3 c), source ext#4 d), and source ext#5 e).

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.