arXiv:0708.3671v1 [astro-ph] 27 Aug 2007
A Multiwavelength Analysis of the Strong Lensing Cluster RCS
022434-0002.5 at z = 0.778
Department of Astronomy, University of Virginia, P.O. Box 400325, Charlottesville, VA
Center for Astrophysics and Space Astronomy, University of Colorado at Boulder, Campus
Box 389, Boulder, CO 80309
Department of Physics & Astronomy, University of Victoria, Elliott Building, 3800
Finnerty Rd, Victoria, BC, V8P 5C2
Department of Astonomy and Astrophysics, University of Chicago, 5640 S. Ellis Ave,
Chicago, IL 60637, USA
Department of Astronomy and Astrophysics, University of Toronto, 50 St. George St.,
Toronto, ON, M5S 3H4, Canada
MIT Kavli Institute for Astrophysics and Space Research, 77 Massachusetts Ave.,
Cambridge, MA 02139, USA
– 2 –
Department of Astronomy and Astrophysics, University of Toronto, 50 St. George St.,
Toronto, ON, M5S 3H4, Canada
McGill University Department of Physics, Rutherford Physics Building, 3600 Rue
University, Montreal, Quebec, Canada, H3A 2T8
Astronomy Technology Centre, Royal Observatory, Blackford Hill, Edinburgh EH9 3HJ
We present the results of two (101 ks total) Chandra observations of the
z = 0.778 optically selected lensing cluster RCS022434-0002.5, along with weak
lensing and dynamical analyses of this object. An X-ray spectrum extracted
within R2500(362 h−1
keV. The surface brightness profile of RCS022434-0002.5 indicates the presence
of a slight excess of emission in the core. A hardness ratio image of this object
reveals that this central emission is primarily produced by soft X-rays. Further
investigation yields a cluster cooling time of 3.3×109years, which is less than half
of the age of the universe at this redshift given the current ΛCDM cosmology. A
weak lensing analysis is performed using HST images, and our weak lensing mass
estimate is found to be in good agreement with the X-ray determined mass of
the cluster. Spectroscopic analysis reveals that RCS022434-0002.5 has a velocity
dispersion of 900± 180 km s−1, consistent with its X-ray temperature. The core
gas mass fraction of RCS022434-0002.5 is, however, found to be three times lower
than expected universal values. The radial distribution of X-ray point sources
within R200of this cluster peaks at ∼ 0.7R200, possibly indicating that the cluster
potential is influencing AGN activity at that radius. Correlations between X-ray
and radio (VLA) point source positions are also examined.
70kpc) results in an integrated cluster temperature of 5.1+0.9
Subject headings: cosmology:observations—X-rays:galaxies:clusters—galaxies:clusters:general
– 3 –
Clusters of galaxies which exhibit gravitational lensing have been actively sought af-
ter since their discovery in the mid-1980s (Soucail 1988; Lynds & Petrosian 1988). This
interest is due to the numerous scientific opportunities that these objects provide.
cause they magnify sources which lie at very high redshifts, they allow us to investigate
forming galaxies in the early epochs of the universe in great detail (e.g., Ebbels et al.
1996; Franx et al. 1997; Pettini et al. 2000; Swinbank et al. 2003). Gravitational lens-
ing can also be used as an independent probe of the mass distribution of clusters (e.g.,
Kneib et al. 1993; Tyson, Kochanski, & dell’Antonio 1998; Broadhurst et al. 2005). In ad-
dition, clusters with multiple lensed sources contain information on the underlying cosmol-
ogy of the universe (Golse, Kneib, & Soucail 2001). Until recently the cumulative sample of
strong lensing clusters was relatively small and mainly restricted to objects at redshifts of
z < 0.5 (Wu et al. 1998; Williams, Navarro, & Bartlemann 1999).
The Red-Sequence Cluster Survey (RCS; Gladders & Yee 2000, 2005) has recently dis-
covered eight strong lensing clusters at high redshift (Gladders et al. 2003). One of the most
striking objects discovered via the RCS survey is RCS022434-0002.5at z = 0.778 (Gladders, Yee, & Ellingso
2002). This cluster possesses a rich array of lensing arcs that originate from several back-
ground sources, one of which lies at a (spectroscopically confirmed) redshift of z = 4.88,
another at z = 3.66 (Swinbank et al. 2007), and a radial arc was found to be at z =
1.050 (Sand et al. 2005).
RCS022434-0002.5 (hereafter RCS0224-0002) was observed twice with the Chandra X-
ray Observatory, initially for 15 kiloseconds, and more recently for 89 kiloseconds, the deepest
single X-ray follow-up observation of an RCS cluster to date. Here we present our analysis
of these observations. From the data we derive a spatially resolved surface brightness profile,
cluster temperature, and mass. In addition, we compare high resolution X-ray flux and hard-
ness images of the cluster core with observations in the optical and radio. Weak lensing and
dynamical analyses are performed and compared to X-ray results. A point source analysis
of this field is also presented.
Unless otherwise noted, this paper assumes a cosmology of H0 = 70 km s−1Mpc−1,
ΩM= 0.3, and ΩΛ= 0.7. Using this cosmology, 1′′= 7.42 h−1
addition, all error bars represent 68% confidence levels.
70kpc at the cluster redshift. In
– 4 –
Chandra Observations and Data Reduction
RCS0224-0002 was observed by the Chandra X-ray Observatory on 15 November 2002
for 12,327 seconds, and 9 December 2004 for 89,202 seconds. The aimpoints of the obser-
vations were located on the S3 chip of the Advanced CCD Imaging Spectrometer (ACIS)
CCD array. Both observations were conducted in VFAINT mode, and reprocessed accord-
ingly. Upon obtaining the data, aspect solutions were examined for irregularities and none
were detected. Charged particle background contamination was reduced by excising time
intervals during which the background count rate exceeded the average background rate by
more than 20%. Bad pixels were removed from the resulting event files, and they were also
filtered on standard grades. After cleaning, the available exposure times were 12,051 seconds
and 88,973 seconds, providing a total exposure of 101,024 seconds.
Combined images, instrument maps and exposure maps were created in the 0.29-7.0
keV band using the CIAO 188.8.131.52 tools DMCOPY and MERGE ALL, binning the data by
a factor of 2. Data with energies above 7.0 keV and below 0.29 keV were excluded due
to background contamination and uncertainties in the ACIS calibration, respectively. Data
between 0.29 keV and 0.6 keV were included despite questionable low energy calibration for
the purpose of increasing the overall signal-to-noise ratio of the observation.
Flux images were created by dividing the image by the exposure map, and initial point
source detection was performed by running the tools WTRANSFORM and WRECON on the
0.29-7.0 keV image. These tools identify point sources using “Mexican Hat” wavelet functions
of varying scales. An adaptively smoothed flux image was created with CSMOOTH. This
smoothed image was used to determine the X-ray centroid of the cluster, which lies at an RA,
Dec of 02:24:34.161, -00:02:26.44 (J2000). The uncertainty associated with this position can
be approximated by the smoothing scale applied to this central emission (2.5′′). Determined
in this manner, the position of the X-ray centroid is 4.5 ± 2.5′′(33.4 ± 18.6 h−1
the brightest cluster galaxy (BCG) of RCS0224-0002, falling within the range of typical
offsets between BCGs and X-ray isophotal centers in X-ray selected samples (Patel et al.
2006). Figure 1 shows an HST image of RCS0224-0002 with adaptively smoothed X-ray flux
A radial surface brightness profile was computed from a combined, point source removed
image and exposure map over the range 0.29-7.0 keV in 1′′annular bins, then multiplied by
a factor of (1+z)3to correct for cosmological dimming. We find an excess of emission within
– 5 –
the central ∼ 4′′of the cluster. This central excess does not appear to be an AGN, since
it is extended (∼ 4× wider than the PSF) and no hard point source was detected at that
location. To test the validity of this central excess, a second surface brightness profile was
computed in 2′′annular bins.
Both profiles were then fit with both single and double β models. A single β model
takes the form:
I(r) = IB+ I0
where IBis a constant representing the surface brightness contribution of the background,
I0is the normalization and rcis the core radius.
A double β model is represented as:
I(r) = IB+ I1
where each component has fit parameters (In,rn,βn).
The central excess does persist in the more highly binned surface brightness profile, yet
the reduced χ2s of the single vs. double β model fits are almost identical in both cases, and
there is no clear statistical preference between the two models. β model fits to both the
1′′and 2′′bin surface brightness profiles are shown in Figure 2 and details of all four fits are
given in Table 1. Unless otherwise noted, the single β model fit to 2′′bin data is used in all
4.Spectral Analysis and R2500
Using the centroid position indicated in Section 2, a point source removed spectrum
was extracted from each observation in a circular region with a 300 h−1
spectra were then jointly analyzed in XSPEC (Arnaud 1996), using weighted response ma-
trices (RMF) and effective area files (ARF) generated with the CIAO tool SPECEXTRACT
and the Chandra calibration database version 3.2.2. Backgrounds were extracted from the
aimpoint chip in regions free of cluster emission. The resulting spectra, one extracted from
the short observation and one from the long observation, containing 183 and 1651 source
counts respectively. Both spectra were grouped to include at least 20 counts per bin.
70kpc radius. These
– 6 –
The spectra were jointly fit with a single temperature spectral model including fore-
ground absorption. The absorbing column was fixed at its measured value of 2.91×1020cm−2(Dickey & Loc
1990), and the metal abundance was fixed at 0.3 solar (Edge & Stewart 1991). Data with
energies below 0.29 keV and above 7.0 keV were excluded from the fit.
The results of these fits, combined with the single β model parameters from Section 3,
were then used to estimate the value of R2500 for this cluster. This is accomplished by
combining the equation for total gravitating mass
Mtot(< r) = −kT(r)r
?δ ln ρ
δ ln r+δ ln T
δ ln r
(where µmpis the mean mass per particle) with the definition of mass overdensity
where z is the cluster redshift, and ∆ is the factor by which the density within r∆exceeds
ρc(z), the critical density at z. Here ρc(z) is given by ρc(z) = 3H2
[Ωm(1 + z)3+ ΩΛ]1/2. These equations are then combined with the density profile implied
from the β model (assuming hydrostatic equilibrium, spherical symmetry, and isothermality)
0E(z)2/8πG, where E(z) =
ρgas(r) = ρ0
resulting in the equation
− 1, (6)
(Ettori 2000; Ettori et al. 2004).
After the initial estimation of R2500, additional spectra were extracted from within
that radius, and spectral fitting was performed again. This iterative process was continued
until the fitted temperature and calculated value of R2500was consistent with the extraction
radius. The final estimate of R2500is 362+61
region was Tx= 5.1+0.9
spectrum, the unabsorbed bolometric luminosity of RCS0224-0002 within R2500 is LX =
70kpc (48.8′′), with an extrapolation to R200
70Mpc. The best fit temperature of the spectra extracted from within this
−0.5keV, with a reduced χ2=0.75 for 77 degrees of freedom. Using this
– 7 –
on the data.
−0.3×1044erg s−1. Figure 3 is a plot of these spectra with the best fitting model overlaid
Images, instrument maps and exposure maps were created for both a soft X-ray band
(0.5-2.0 keV) and a hard band (2.0-8.0 keV) by the method described in Section 2. These
files were then used to create flux images in the two bands.
Using the smoothing scale produced in the analysis of the entire energy band, smoothed
flux images were created in each of the two bands. The hard band image was subtracted
from the soft band image, and the result of that calculation was divided by the smoothed flux
image of the entire energy band (i.e.,IS−IH
are overlayed on a VLA 1.4 GHz radio image in Figure 4. Soft emission is concentrated in a
region containing both the X-ray centroid and the BCG, a radio galaxy to the southwest of
the X-ray centroid (Figure 4). There is no clear emission peak that coincides with the BCG,
a radio galaxy at z = 0.7773.
Itot). Contours of the resulting hardness ratio image
6. Mass Determinations
6.1.Gas Density and Cooling Time
The gas mass density distribution of a cluster of galaxies whose surface brightness profile
is well described by a β model can be shown to follow Equation 5. This relationship is
dependent upon the assumptions of hydrostatic equilibrium, spherical symmetry, and gas
isothermality. The central density in this equation can be determined by an expression
relating the observable cluster X-ray luminosity to gas density. This calculation requires the
experimental determination of both gas temperature and emission measure, which can be
performed in XSPEC.
With the results of spectral analysis (Section 4) and using the best fitting surface
brightness parameters for the inner luminosity peak (Section 3), a central gas density of
7.6 ± 0.09 × 10−3cm−3is obtained for RCS0224-0002.
Recall that the characteristic time that it takes a plasma to cool isochorically through
an increment of temperature δT can be written
– 8 –
(Sarazin 1988), where n is the electron density, Λ(T) is the total emissivity of the plasma
(the cooling function), and k is Boltzmann’s constant. In the case of isobaric cooling, 3/2 is
replaced by 5/2.
Using this formula combined with the cooling function appropriate to a 0.3 solar abun-
dance plasma, the cooling time of RCS0224-0002 is 3.3× 109years. The age of the universe
in our adopted cosmology at z = 0.778 is 6.8 ×109years (e.g., Thomas & Kantowski 2004).
Therefore the cooling time of the innermost region of RCS0224-0002 is less than the
age of the universe at z = 0.778. This, combined with the luminosity excess in the central
few arcseconds, as well as the existence of a soft X-ray peak at the same location, indicates
that this cluster may contain a proto-cooling core. The observation does not contain enough
source photons to obtain spatially resolved temperature information for this cluster, therefore
longer exposures will be required to completely characterize the central emission.
At least one other study has detected a high redshift cluster with the signatures of
a small cooling core. Grego et al. (2004) in their analysis of a Chandra observation of
MS1137.5+6625 (at z = 0.783) detect a small central surface brightness excess, and an
excess of soft emission in the core of this cluster. They also calculate a central cooling time
that is shorter than the age of the universe at that epoch.
6.2. X-ray Mass Estimate
Using the results of the β model fits (Section 3) and spectral fit (Section 4), along with
Equation 5 and the equation of hydrostatic equilibrium (Equation 3), it can be shown that
Mtot(< r) =3β
1 + (r/rc)2. (8)
Using this equation (with µ = 0.62), gas mass and total mass were calculated out to R2500
(∼ 362 h−1
density, gas mass, total mass, and gas mass fraction were determined through statistically an-
alyzing a random sampling (N = 10,000) of values drawn from the 68% confidence intervals
of relevant parameters, assuming that their errors are uncorrelated. Mass determinations
resulted in M2500,tot= 1.6 ± 0.2 × 1014h−1
70kpc) and R200(∼ 1.47 h−1
70Mpc) for RCS0224-0002. One sigma errors on central
70M⊙, M2500,gas= 0.05 ± 0.005 × 1014h−5/2
– 9 –
M200,tot= 8.3 ± 0.8 × 1014h−1
temperature relation of Vikhlinin et al. (2006) predicts M2500,tot= 1.2±0.05×1014h−1
for z = 0.778 and TX= 5.1 keV, which is within 2σ (25%) of our observed value.
70M⊙and M200,gas= 0.24 ± 0.02 × 1014h−5/2
M⊙. The mass-
The core gas mass fraction of RCS0224-0002 is fg,2500= 0.031±0.005 h−3/2
is significantly lower than that determined via the same method for a sample of 14 medium
redshift X-ray selected clusters (Hicks et al. 2006) and is three to four times lower than
both theoretically and observationally expected “universal” values (Evrard 1997; Allen et al.
2004). This is a phenomenon that is also seen in groups, poor clusters, and other high red-
shift clusters (Dell’Antonio, Geller, & Fabricant 1995; Sanderson et al. 2003; Sadat et al.
2005), as well as high redshift cluster simulations (Ettori et al. 2004) and additional RCS
clusters (Hicks et al. 2007, in prep).
. This value
6.3. Weak Lensing Analysis
The central region of RCS0224-0002 was observed in the F606W and F814W filter
with the WFPC2 camera onboard the Hubble Space Telescope, with the center of the cluster
located on WFC3. The observations consist of 2 orbits in each filter, yielding total integration
times of 6600s in F606W and F814W.
The exposures in each passband were split into two sets (by the orbit the data were
taken), resulting in two images.These two images were designed to be offset by 5.5
pixels, to allow the construction of an interlaced image (e.g., van Dokkum et al. 1999;
Hoekstra, Franx, & Kuijken 2000), which has a sampling that is a factor
the original WFPC2 images. This same approach was used by Hoekstra, Franx, & Kuijken
(2000) in their study of MS1054-03 and we refer to this paper for more details. For the anal-
ysis presented here we omitted data from the Planetary Camera because of the the brighter
√2 better than
The next step is to use the interlaced images for our weak lensing analysis. The analysis
technique is based on the one developed by Kaiser, Squires, & Broadhurst (1995) with a num-
ber of modifications described in Hoekstra et al. (1998) and Hoekstra, Franx, & Kuijken
(2000). We note that this implementation has been tested extensively (e.g., Hoekstra et al.
1998; Heymans et al. 2006) and has been shown to be accurate at the few percent level. We
follow the procedure outlined in Hoekstra et al. (1998, 2000) which results in a catalog of
ellipticities for the faint galaxies that we use in our lensing analysis. These shapes have been
corrected for PSF anisotropy and the size of the PSF.
The measurement of the shape of an individual galaxy provides only a noisy esti-
– 10 –
mate of the weak gravitational lensing signal. We therefore average the measurements of
a large number of galaxies as a function of distance to the cluster center. As discussed in
Hoekstra, Franx, & Kuijken (2000) we weight each object with the inverse square of the un-
certainty in the distortion, which includes the contributions of the intrinsic ellipticities and
the shot noise in the shape measurement.
Figure 5 shows the resulting average tangential distortion as a function of distance from
the cluster center, which was taken to be the location of the brightest cluster galaxy. We
selected 462 galaxies with 22.5 < F814W < 25.5 and F606W − F814W < 1.5 (i.e., bluer
than the cluster early types) as source galaxies. We detect a significant lensing signal. The
lower panel of Figure 5 is the signal, gX, when the phase of the distortion is increased by
π/2. If the signal observed in the upper panel is due to gravitational lensing, gX should
vanish, as is observed. The solid line is the best fit singular isothermal sphere model which
yields an Einstein radius of 4.7 ± 0.9 arcseconds.
To relate the observed lensing signal to an estimate of the mass requires an estimate of
the mean source redshift distribution. As in Hoekstra, Franx, & Kuijken (2000), we use the
results from Fernandez-Soto, Lanzetta, & Yahil (1999) which are based on the Hubble Deep
Fields. This allows us to determine an equivalent line of sight velocity dispersion using the
best fit isothermal sphere model. We obtain a value of σ = 777+69
Alternatively we can fit an NFW model to the measurements. We assume that the
density profile is only a function of the virial mass and use the relation between concentration
and mass from Bullock et al. (2001). This yields a virial mass of 5.4+4.1
corresponding M2500= 1.3+1.0
agreement with both the dynamical and X-ray analysis results.
−2.7× 1014M⊙and a
−0.6× 1014M⊙(and r2500= 375 kpc). This result is in excellent
As shown in Hoekstra et al. (2002) mass estimates based on single HST pointings tend
to be biased low, because of substructure in the cluster center. However, the distribution
of galaxies in RCS0224-002 is relatively regular and does not show significant substructure.
Furthermore, because of its high redshift, the lensing signal can be measured sufficiently far
out to be less affected by substructure.
Optical multi-object spectroscopic observations of galaxies in this field were performed
by several telescopes and instruments; dates and observational parameters are listed in Ta-
ble 2. Both galaxies and gravitational arc images were targeted, over a field of view of 5
arcminutes. Most cluster galaxies had magnitudes of R∼ 20-22, though a few galaxies as
– 11 –
faint as R∼ 24 were identified via their [OII] emission lines. Several slits targeting cluster
arcs were tilted or curved (see Gladders et al. 2003), but these were not used for cluster
galaxy redshifts. Overall, 39 galaxy redshifts at 0.70 < z < 0.85 were measured by the
wavelength of individual lines or via cross-correlation against galaxy spectra, with a range
of spectral types. We find no significant systematic differences in the velocities measured
by the different telescopes and instruments, either in velocity centroid, or in the 3 objects
whose redshifts were measured accurately by more than one telescope.
Figure 6 shows a histogram of galaxy velocities. There are two secondary peaks close
to the cluster, one at z∼ 0.750 and another at z∼ 0.805. Both of these comprise galaxies
spread over the field of view and are not associated with an easily identified isolated clump.
There are substantial gaps between them and the cluster peak at z=0.778, and hence we do
not identify these galaxies as cluster members.
Correcting in quadrature for an estimated uncertainty of 100 km s−1for individual
velocity measurements, we find that the cluster is located at a mean redshift of 0.7777 and
has a rest-frame velocity dispersion of 900 ± 180 km s−1, based on 24 cluster members. This
estimate is very close to that expected from a standard σ-TXrelationship (Voit et al. 2003;
Xue & Wu 2000), where an X-ray temperature of 5.1 keV predicts a velocity dispersion of
872 km s−1. Our velocity dispersion measurement is also consistent with both weak lensing
analysis (6.3) and the value of 920 km s−1reported by Swinbank et al. (2007) based on
their strong lensing analysis of this cluster.
7.1. X-ray Source Population
Studies of the AGN population in clusters of galaxies are an important probe of galaxy
evolution in these environments (Martini et al. 2002). While AGN can be difficult to detect
at optical wavelengths due to obscuration, they are much more easily seen in the X-ray,
where they comprise the dominant contribution to the hard X-ray point source population.
In this section we perform an X-ray point source analysis of the Chandra observation of
Using the hard and soft band images and exposure maps referred to in Section 5, point
sources in these bands were identified using the CIAO tools WTRANSFORM and WRECON.
Only sources with a detection significance above 3σ were retained for further analysis. In
the case of faint sources, our numbers may represent a lower limit, due to both off-axis
degradation of the PSF (Kim et al. 2004), and the possible masking of very faint sources by
– 12 –
diffuse X-ray emission in the cluster center.
Using PIMMS (Portable Interactive Multi-Mission Simulator), assuming an absorbed
powerlaw spectrum with a hydrogen column density fixed at the galactic value and a pow-
erlaw index of 1.6, X-ray fluxes were calculated for each source after correcting for exposure
variations across the chips. LogN-logS information was then calculated for each energy
band, inclusive of all sources within R200. These data were then compared to the logN-
logS background values of Moretti et al. (2003). The results of these comparisons with 1σ
error bars (Gehrels 1986) are shown in Figure 7. The data are generally consistent with
the measured X-ray background, however the background data do not extend to very low
flux sources, and there may be a very slight excess of low-flux soft X-ray sources within
R200. “Cosmic variance”, however, can become significant at low-flux as well, showing
variations in source counts as high as 3.9σ for 2.0 − 8.0 keV sources with fluxes below
< 1 × 10−15erg cm−2s−1(Brandt & Hasinger 2005)
A list of point sources within R200is given in Tables 3 and Table 4 for the 0.5-2.0 keV and
2.0-8.0 keV bands, respectively. These tables include both flux and luminosity information
(also achieved with PIMMS, assuming a redshift of z = 0.778) for each source.
To compare the distribution of sources in the RCS0224-0002 field to those reported
in Ruderman & Ebeling (2005, based on a statistical study of ∼ 50 clusters), radial his-
tograms containing the number of sources per square megaparsec (assuming a redshift of
z = 0.778) were created from data in each energy band (Figure 8). These plots indicate
a significant excess of point sources within R200, particularly in the 0.6 − 0.8 R200 bins.
The Ruderman & Ebeling (2005) study was conducted in the 0.5-2.0 keV energy band, and
our soft band data are roughly consistent with their results.
7.2. Radio Comparisons
RCS0224-0002 was observed with the National Radio Astronomy Observatory’s (NRAO)
Very Large Array (VLA)1at 1.4 GHz for a total of 15 hours over three days (Webb et al.
2005). Using this data, an investigation of the coincidence of X-ray point sources with 1.4
GHz sources was performed. Figure 9 consists of the 1.4 GHz radio image of RCS0224-0002
with 0.2-5.0 and 2.0-8.0 X-ray point source regions overlaid. Obvious correlations are noted
in Tables 3 and 4, and can be seen in Figure 9. None of the detected X-ray point sources
1The National Radio Astronomy Observatory is a facility of the National Science Foundation operated
under cooperative agreement by Associated Universities, Inc.
– 13 –
(in either energy band) is coincident with the cluster BCG.
RCS0224-0002 was also observed at both 450µm and 850µm with the Submillimeter
Common-User Bolometer Array (Holland et al. 1998, SCUBA) in 2001 and 2002. The re-
sults of these observations are reported in Webb et al. (2005), and have resulted in five
submillimeter source detections in the RCS0224-0002 field. Only one of the five sources is
coincident with an X-ray detected point source, smm-RCS0224.2 (source 6 in Table 3 and
source 3 in Table 4).
8. Discussion and Summary
This paper describes the results of the most in depth X-ray analysis of an optically
selected high redshift RCS cluster to date, the 101 ks of Chandra observations of the lensing
cluster RCS0224-0002 at z = 0.778. Imaging analysis reveals that the peak of X-ray emission
is slightly displaced (4.5 ± 2.5′′) from the center of the lensing arcs. This displacement falls
well within the common range of X-ray centroid offsets (Patel et al. 2006).
Single temperature spectral fitting of the X-ray emission within a 362 h−1
radius indicates that this cluster has an ambient temperature of 5.1+0.9
brightness profile of RCS0224-0002 exhibits a small central peak of emission which can be
modeled via the inclusion of a second β component. The β model used in our analysis has a
core radius of 213+14
and surface brightness fitting, R200is determined to be 1.47+0.1
−0.5keV. The surface
70kpc and β value of 0.88+0.07
−0.04. Using the results of both temperature
70Mpc for this cluster.
Hardness ratio maps were created from soft (0.5-2.0 keV) and hard band (2.0-8.0 keV)
flux images. A peak in soft X-ray emission is seen to coincide with the central peak in surface
brightness. This peak is elongated toward a 1.4 GHz VLA radio source which is centered on
the BCG of this cluster, as seen in the HST image (Figure 4).
Using the results of both spectral and surface brightness fitting, we have calculated a cen-
tral density for RCS0224-0002 of 7.6×10−3±9 × 10−4cm−3, which results in a central cooling
time of 3.3×109years. Without spatially resolved temperature information we are unable to
confirm the presence of a cool component in the center of RCS0224-0002, however the cen-
tral surface brightness excess, corresponding soft peak of emission, and lack of a central hard
point source support the idea that RCS0224-0002 may contain a small proto-cooling core.
Similar evidence has been detected in another z ∼ 0.8 cluster, MS1137.5+6625 (Grego et al.
2004). These two clusters are the highest redshift systems in which central cooling has yet
been detected (albeit indirectly), and their existence provides evidence that massive clusters
are in place at early times (Mullis et al. 2005).
– 14 –
Mass estimates of RCS0224-0002 result in total masses of M2500,tot = 1.6 ± 0.2 ×
of total gravitating mass is consistent within 68% errors with weak lensing mass estimates
of M2500 = 1.3+1.0
cluster of 900 ± 180 km s−1, which is consistent with that predicted by the σ − TX rela-
tionship (Voit et al. 2003; Xue & Wu 2000), as well as both weak (6.3) and strong lensing
estimates of the velocity dispersion (Swinbank et al. 2007). The consistency of these mass
estimators is an indication that RCS0224-0002 is a relatively relaxed system.
70M⊙and a gas mass of M2500,gas= 0.050 ± 0.005 × 1014h−5/2
M⊙. This estimate
−0.6× 1014M⊙. Similarly, we find a rest-frame velocity dispersion for this
An investigation of the point source population in the field of RCS0224-0002 indicates
a possible slight excess of low-flux soft energy sources within R200of the cluster center. The
high energy source population of this cluster is generally consistent with the measured X-ray
background. Both measurements of point source density represent lower limits because of
a selection effect which makes it more difficult to detect point sources close to the cluster
center (due to the surface brightness of the cluster). Radial distributions of these sources
exhibit a significant peak within R200in both the hard and soft bands, possibly indicating
that the cluster potential is influencing AGN activity at that radius. Spectral information
on cluster member galaxies is needed, however, to test that hypothesis. The radial profile of
soft X-ray point sources is consistent with the results of Ruderman & Ebeling (2005). X-ray
point sources within the field of RCS0224-0002 have been compared to 1.4 GHz VLA flux
information, and coincident sources are listed in Tables 3 and 4.
The core gas mass fraction of this cluster, fg,2500= 0.031±0.005 h−3/2
times lower than expected universal values (Evrard 1997). This is a phenomenon that is also
seen in groups, poor clusters, and other high redshift clusters (Dell’Antonio, Geller, & Fabricant
1995; Sanderson et al. 2003; Sadat et al. 2005), as well as high redshift cluster simula-
tions (Ettori et al. 2004). It is possible that a comparatively higher fraction of the baryonic
mass of this object was converted into stars (Vikhlinin et al. 2006; Nagai, Kravtsov & Vikhlinin
2007); however additional analysis is required to confirm or rule out this possibility. In ad-
dition, the AGN component of RCS0224-0002 may provide a significant contribution to the
entropy of this system, however it is likely that other non-gravitational processes (i.e. galaxy
formation) are also involved.
, is more than three
Support for this work was provided by the National Aeronautics and Space Administra-
tion through Chandra Award Numbers GO2-3158X and GO4-5153X issued by the Chandra
X-ray Observatory Center, which is operated by the Smithsonian Astrophysical Observa-
tory for and on behalf of the National Aeronautics Space Administration under contract
NAS8-03060. Erica Ellingson also acknowledges support from NSF grant AST-0206154.
– 15 – Download full-text
A portion of this work was based on observations made with the NASA/ESA Hubble
Space Telescope, obtained at the Space Telescope Science Institute, which is operated by
the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS
5-26555. These observations are associated with program 9135.”
In addition, some of the data presented herein were obtained at the W.M. Keck Obser-
vatory from telescope time allocated to the National Aeronautics and Space Administration
through the agency’s scientific partnership with the California Institute of Technology and
the University of California. The Observatory was made possible by the generous financial
support of the W.M. Keck Foundation.
Finally, this paper was also based on observations obtained at the Gemini Observatory,
which is operated by the Association of Universities for Research in Astronomy, Inc., under
a cooperative agreement with the NSF on behalf of the Gemini partnership: the National
Science Foundation (United States), the Particle Physics and Astronomy Research Coun-
cil (United Kingdom), the National Research Council (Canada), CONICYT (Chile), the
Australian Research Council (Australia), CNPq (Brazil) and CONICET (Argentina).
We would also like to thank Phil Armitage, Webster Cash, John Houck, Richard
Mushotzky, Craig Sarazin, and Michael Wise for their contributions and input.
Allen, S.W., Schmidt, R.W., Ebeling, H., Fabian, A.C., & van Speybroeck, L. 2004, MNRAS,
Arnaud, K. A. 1996, ADASS, 101, 5
Brandt, W.N. & Hasinger,G. 2005, ARA&A, 43, 827
Broadhurst, T., Takada, M., Umetsu, K., Kong, X., Arimoto, N., Chiba, M., Futamase, T.,
2005, ApJ, 619, 143
Bullock, J.S., Kolatt, T.S., Sigad, Y., Somerville, R.S., Kravtsov, A.V., Klypin, A.A., Pri-
mack, J.R., & Dekel, A. 2001, MNRAS, 321, 559
Dell’Antonio, I.P., Geller, M.J., & Fabricant, D.G. 1995, AJ, 110, 502
Dickey, J.M. & Lockman, F.J. 1990, ARA&A, 28, 215
Ebbels, T.M.D., Le Borgne, J.F., Pello, R., Ellis, R.S., Kneib, J.P., Smail, I., & Sanahuja,
B. 1996, MNRAS, 281, 75