A2125 and its Environs: Evidence for an X-ray-emitting Hierarchical Superstructure

A2125 and its Environs:
Evidence for an X-ray-emitting Hierarchical Superstructure

Q. Daniel Wang
Dearborn Observatory, Northwestern University
2131 Sheridan Road, Evanston, IL 60208-2900
- - - , E-mail: wqd@nwu.edu

Q"^ I Andrew Connolly^ and Robert Brunner^

a^

X

Department of Physics and Astronomy
Johns Hopkins University, Baltimore, MD 21218

^ \ ajc@tiamat.pha.jhu.edu and rbrunner@hoplite.pha.jhu.edu

ABSTRACT

>

00

Based on a deep ROSAT/PSPC observation, we reveal an elongated complex of

^^ ' extended X-ray-emitting objects in and around the galaxy cluster A2125. Multicolor

(^ ■ optical imaging of galaxies in the field suggests that this complex represents a

C^^ ■ hierarchical superstructure spanning ~ 11 /i^q Mpc at redshift ~ 0.247. The multi-

Q^. at least three major subunits. The dynamical youth of this cluster is consistent with

its large fraction of blue galaxies observed by Butcher & Oemler. The superstructure
H I contains two additional clusters, projected at distances of only 3 and 4.3 h^Q Mpc

^ I from A2125. But the most interesting feature is the low-surface-brightness X-ray

emission from a moderate galaxy concentration not associated with individual clusters.

The emission likely arises in a hot (~ lO'^ K) intergalactic medium, as predicted in

N-body/hydro simulations of structure formation.

Subject headings: cosmology: observations — large-scale structure — galaxies: clusters:
individual (A2125) — galaxies: evolution — X-rays: general

1. Introduction

A substantial fraction, probably most, of the baryonic matter in the Universe is expected
to be in a hot phase of the intergalactic medium (IGM). This hot IGM is a definite prediction

^Guest observer at the Kitt Peak National Observatory, National Optical Astronomy Observatories, which is
operated by the Association of Universities for Research in Astronomy, Inc. (AURA) under cooperative agreement
with the National Science Foundation.

of recent N-body/hydro simulations of structure formation (e.g., Ostriker & Cen 1996). During
the structure formation, shocks induced by the cohisions between large-scale collapsing flows can
naturally heat the IGM to a temperature range of '^ 10^ — lO'^ K. The integrated emission from
the hot IGM (Cen et al. 1995) provides a reasonably good explanation of the diffuse extragalactic
~ 0.7 keV background inferred from the Magellanic Bridge X-ray shadowing experiment (Wang
& Ye 1996). The simulations further show that the hot IGM forms filamentary superstructures
(cosmic webs or caustics), on scales greater than individual clusters, and that the IGM emission
is greatly enhanced in regions near clusters, where gravitational potential wells are relatively
deep. Analytical modeling for the triaxial collapse of cosmological perturbations also predicts the
presence of X-ray-emitting filaments near clusters (Elsenstein & Loeb 1995). Properties (e.g.,
temperature, chemical abundance, axial ratio) of such filaments can provide important constraints
on theories of galaxy feedback and cluster formation.

Here, we report the detection of such an X-ray-emitting superstructure. Along with our
systematic X-ray study of Butcher & Oemler clusters (1984), we find that A2125, an optically rich
cluster of a high blue galaxy fraction (/^ = 19 it 3%) at z = 0.247, is associated with a complex of
galaxies, clusters and diffuse hot gas. Extending ~ 11 h^Q Mpc, {h^o — the Hubble constant in
units of 50 km s~^ Mpc~^; Qo = 0.1 is assumed throughout the paper), this hierarchical complex
closely resembles filamentary superstructures seen in the simulations (e.g., Fig. 10 in Bryan et al.
1994; Bryan 1996, personal communications).

2. Observations and Data Reduction

The X-ray observation (rp800511) of A2125 was made with the ROSAT PSPC (Positional
Sensitive Proportional Counter; Triimper 1992 and references therein). The effective exposure
is 19359 s. We processed the data, following the standard procedure for studying diffuse X-ray
sources (Snowden et al. 1994). We limited our imaging analysis in the 0.5-2 keV band (PSPC
channels 13-30) to reduce both the foreground from the Galactic hot interstellar medium, which
dominates below ~ 0.3 keV, and the background variation caused by a differential foreground
X-ray absorption in the field. The spatial variation in the Galactic X-ray-absorbing gas column
density, inferred from the IRAS 100/im survey, is up to about 2 x 10^*^ cm^^ in the field and should
result in a differential absorption ^ 10% in the 0.5-2 keV image. Furthermore, our analysis was
based on data within 16' of the PSPC pointing axis. Outside this central field, the analysis might
otherwise be complicated by both the detector's window supporting structure (at ^^ 20' off-axis)
and the degraded point spread function (PSF; Hasinger et al. 1994).

We classified X-ray sources of signal-to-noise ratios greater than 3 into two classes, point-like
and extended (Fig. 1), based on maximum likelihood fits to the 2-D count distributions of
individual X-ray sources. Extended sources are those with their count distributions broader than
the off-axis dependent PSF at confidences ^ 95%. Some of these sources may represent peaks
of large-scale diffuse X-ray structures (e.g., clusters). We excised point-like sources by replacing

counts within 85% source flux-encircled radii with randomly generated events of intensities
interpolated from neighboring averages. The residual flux of the excised sources, spread over the
image due to the very extended wing of the PSF, produces no discernible local enhancement in
Fig. 2.

As a follow-up of this X-ray study, we obtained multicolor CCD images covering the extended
X-ray features in the field. Using the Prime Focus camera on the Kitt Peak Mayall 4-meter
telescope, we observed A2125 and the extended X-ray sources to the north and to the southwest
of the cluster. We also obtained a pointing just beyond the southwestern corner of Fig. 2.. All
these pointings were undertaken in the U, B, R and I passbands. With a pixel scale of 0'.'47, each
T2KB CCD pointing covered a 16' field of view (equivalent to 4.9 h^Q Mpc at z = 0.247).

We reduced these optical data following standard procedures in the IRAF image reduction
packages. We detected objects using the SExtractor package (Bertin and Arnouts 1996) with
a detection kernel of 1'.'2 (equivalent to the FWHM of the seeing disk). Aperture and Kron
magnitudes were measured for all detected sources. Colors of these sources were derived using a
20" diameter aperture. The detection limits for these photometric data were 22.7, 22.5, 23.5 and
22.4 in the U, B, R and I passbands respectively (for a 5 sigma detection of a point source). All
magnitudes were converted to the Johnson photometric system using Landolt standards.

3. Analyses and Results

Fig. 2 shows a complex of extended X-ray-emitting objects. Possibly limited by the size of the
field, this complex spans at least 35' (from the upper middle to the lower right). Fig. 3 presents
the apparent associations of the X-ray-emitting objects with concentrations of galaxies. The
optically known rich cluster A2125 (Abell richness 4) appears surprisingly irregular in the X-ray,
and is detected as a combination of six extended X-ray peaks. The two new clusters (Clusters B
and C hereafter) are evidently the optical counterparts of the two extended X-ray objects (Figs.
3b, c) north of A2125. There are three more extended X-ray peaks to the south and west of A2125
within the 16' off-axis (Fig. 1); their association with galaxy concentrations are less certain.
The double peaks at ~ 15 39™57, 66°6' has an extended radio counterpart, apparently in the 1.4
GHz NRAO Sky VLA survey. The radio and X-ray emissions may thus represent the interaction
between a jet from an active nucleus and its ambient medium (e.g.. Doe et al. 1995). We cannot,
however, exclude the possibility that such an apparently extended peak actually results from the
confusion between two or more point-like sources. Besides the clusters and peaks, the complex
contains substantial amounts of low-surface brightness X-ray emission (LSBXE). The emission is
most prominent in the region southwest to A2125 within a position angle between ~ 200° and
250° (anti-clockwise from North), relative to the centroid of the cluster. The emission between
~ —20° and 20° may also be considerable, but is difficult to quantify because of the confusion
with discrete sources. We concentrate on characterizing the three most prominent components:
A2125, Cluster B, and the southwestern LSBXE, which are all coincident with over-densities in

the galaxy distributions.

Table 1 lists X-ray spectral properties of the individual components, including the centers
and sizes of the regions from which individual spectra are extracted. A background spectrum is
estimated in a partial annulus between 8' and 16' centered on A2125, excluding the two pie-like
regions of the LSBXE mentioned above. We fit the spectra with the standard Raymond & Smith
thermal plasma model, plus a foreground absorption with an assumed metal abundance of the
X-ray-absorbing gas as 100% solar. The fitted absorptions (Nh) toward the two clusters are
consistent with the 21 cm measured Galactic neutral hydrogen column density of 2.9 x lO^'' cm^^
of the field (Stark et al. 1992). We fix the abundance of the ICM as 20% solar. Varying
the abundance within its expected uncertainty of a factor 2 (Mushotzky et al. 1996) changes
the obtained spectral parameters by less than 10%. The abundance of the X-ray-emitting gas
responsible for the LSBXE component is unknown. Our best-fit is 2% solar, but no meaningful
90% confidence limits are obtained. The component also provides little constraints on the
absorption. We thus fix the X-ray absorbing-gas column density as the 21 cm measured value.

From the best-fit temperature kT ~ 0.85 keV of the LSBXE component, one can estimate the
hot gas emission measure as EM ~ (3.2 x 10^^ cm^® pc)(5'o/10^^ counts s^^ arcmin^^),
using the observed surface brightness So (Fig. 2). The mean gas density is then
~ (1.8 X 10^^ cm^'^)(S'o/10^'^ counts s^^ arcmin^^)^/^(/Mpc)^^ , where Impc is the effective depth
of the gas in units of Mpc. If the X-ray-emitting gas fills uniformly a cylinder of diameter 2.5
h^Q Mpc (8') and length 3.4 /i^q Mpc (11'), for example, the luminosity in Table 1 then suggests
a total gas mass Mg ~ 7 x 1O^^M0.

We characterize the azimuthally-averaged surface brightness profiles of the two clusters,
using the standard (3 model (Cavaliere &: Fusco-Femiano 1976). A fit of the model (including
a background parameter) to the profile of Cluster B {x^ /d.o.f. = 29.6/36) gives the (3 value,
the core radius Tc and the central electron density ne^c as 1.3(0.86-4.5), 0.65(0.43-1.5) /15Q Mpc,
and 1.1(0.88 — 1.3) x IO^'^/i^q cm~^ (90% confidence limits). The two parameters (5 and Vc
are strongly correlated with each other. We have also assumed an approximately spherical and
isothermal state of the ICM in deriving Ue^c- The total ICM mass within an 1.5 /i^q Mpcradius
is ~ 5.4 X IQ^^Mq. a direct 13 model fit to the A2125 profile {x^/d.o.f = 57.6/36) can be ruled
out at the 99% confidence. The deviation of the profile from the model is primarily due to the
presence of the three distinct peaks in the central region of the cluster (Fig. 2). To improve the
fit we remove the data within a circular region around the centroid of each peak. The radius of
this region is chosen to be the 85% source-flux-encircled radius of a point-like source. The fit to
the peak-excised profile is acceptable (x^/d-o./. = 44.3/36). The obtained /3 model parameters are
(3 = 0.77(0.62 - 1.1), Tc = 0.52(0.37 - 0.76) h^^^ Mpc, and n^^c = 1.0(0.83 - 1.2) x lO'^/ig/^ cm'^.

*50

The estimated ICM mass is 8.4 x IO^^Mq within the 1.5 /i^g Mpc radius.

We further characterize the 2-D X-ray morphology of A2125 and Cluster B. We first fit the
X-ray morphology of each cluster with a series of ellipses, using iteratively-calculated moments of

the PSPC count distribution (Buote & Canizares 1994; Wang, Ulmer, & Lavery 1997). Cluster
B is statistically consistent with being circular; the ellipticity is ^ 0.31 (95% confidence). A2125
has an ellipticity of 0.38(0.29-0.48) and a North-to-East position angle of 132(124-140) deg,
on scales between 2'-6', and shows multi-peaks on smaller scales. We then calculate the the
peak-excised surface brightness profile of the cluster as a function of the semi- major axis in the
best-fit elliptical coordinates. The /3 model fit to the profile is satisfactory {x^ /d.o.f. = 31.3/36),
and yields the model parameters as /3 = 0.77(0.59 — 1.1), Vc = 0.51(0.33 — 0.84) /i^q Mpc, and
ne,c = 1.0(0.80 — 1.3) X lO'^h^Q cin~^ (assuming an oblate shape of the ICM).

We follow an approach similar to that described by Neumann &; Bohringer (1996) to reveal
the substructure of A2125. We first subtract the best-fit elliptical /5 model {Sm) from the data
(So), and then smooth the residual map by convolving it with a Gaussian (g), i.e., {So — Sm) © 9-
We further calculate the noise map as [max(5o/t ® g^, Sm/t © g"^)]^'"^, where t is the exposure
map. The residual-to-noise ratio gives a significance measure of surface brightness residuals. Fig.
4 shows both the residual and significance maps. The three central peaks, standing out with
their residual-to-noise ratios greater than 3, are of comparable sizes ~ 10" it 4" (after the PSF
contribution is subtracted in quadrature) and luminosities ~ 1.6 — 2.3 x IO^^/i^q ergs s~^.

The association of the X-ray components with the galaxy over-densities are evident in Fig.
3. While we are yet to make a spectroscopic redshift mapping of galaxies in the field, we can use
the colors of the galaxies to determine whether the E/SO sequences in each of the components
are consistent with their being at the same redshift. For an elliptical galaxy at a redshift of
0.247 the Johnson B - R color is ~ 2.5 (Fukugita et al. 1995). We find that A2125 and the
Cluster B have distinct E/SO sequences in their B — R vs R color magnitude diagram. The
B — R color of these sequences is 2.54 and 2.45 for A2215 and cluster B respectively. Given the
uncertainty of approximately 10% in the colors, the two systems are consistent with a redshift
of 0.247. Furthermore, if we isolate those galaxies with 2.4 < B — R < 2.6, we find that the
LSBXE correlates with an over-density of galaxies. We thus tentatively identify the complex as
a coherent superstructure, which spans ~ 11 h^Q Mpc at the redshift z = 0.247. The moderate
galaxy over-density appears to extend further to the north, beyond the field covered by Figs. 1-2,
but not to the southwest. We will present more detailed analysis of the optical observations later.

4. Discussion

The X-ray-emitting complex associated with A2125 is morphologically very similar
to hierarchical superstructures seen in various N-body/hydrodynamic simulations (e.g.,
CDM-I-A universe; Cen & Ostriker 1994; one may find movies of the simulations at

http://dept.physics.upenn.edu/bode/TVD/ ). Such superstructures represent caustics where the
initial density fluctuation peaks and large-scale collapsing gas flows intersect. In addition to the
shock-heating of the gas, both the thermal and chemical feedback from stars may also play an
important role. The heating and chemical enrichment of the IGM affect our understanding the

structure and evolution of clusters (e.g., Kaiser 1991; David, Jones, & Forman 1996). The hot IGM
may also be responsible for the positive cross-correlation between Abell clusters and the X-ray
background surface brightness at ~ 1 keV, as detected by Soltan et al. (1996).

Our detection of the LSBXE in the vicinity of A2125 provides probably the first direct
observational evidence for hot gas outside individual clusters. Our measured temperature of
kT ~ 0.85 keV is at the high end of the range predicted for the hot IGM. This is natural because
the X-ray emission, proportional to the square of gas density, traces deep gravitational potential
valleys in the Universe. Because no bright nearby early-type galaxy is present in the field, the
LSBXE cannot be due to the intragroup hot gas of a nearby group of galaxies. Most likely, both
the hot gas and the galaxy concentration represent part of a hierarchical superstructure that
includes Clusters B and C, and A2125 as well. The scale and morphology of both the hot gas and
galaxy distributions indicate that this superstructure is not in a relaxed equilibrium state. The
entire superstructure can, however, collapse within a few 10^ years to form a single cluster.

A2125 itself also appears to be a young system. First, the presence of at least three major
X-ray subcomponents indicates that the A2125 is a coalescence of subunits of comparable masses.
Such a coalescent occurs typically only at the initial stage of rich cluster formation. Second, this
coalescence scenario explains both the strong elongation of A2125 and its gross misalignment with
the orientation of the overall superstructure. Mergers of a rich cluster with small subclusters,
taking place at later stages of cluster evolution, tend to produce a cluster elongation along the
superstructure. The measured ellipticity of A2125 is the largest in a sample of 10 Butcher &
Oemler clusters observed with the PSPC (Wang & Ulmer 1997 in preparation). Third, the
relatively low temperature and luminosity of A2125 (Table 1), compared to relaxed clusters of
similar optical richness (e.g.. Edge & Steward 1992), indicate that the ICM has not yet virialized
to the gravitational potential of the system. Fourth, the high fraction of blue galaxies observed in
A2125 can be a natural consequence of the coalescence of spiral-rich poor clusters (Kauffmann
1995; Schindler & Wambsganss 1996; Wang, Ulmer, & Lavery 1997). Therefore, A2125, as a rich
assembly of galaxies, is still at its early evolutionary stage to become a rich X-ray-emitting cluster.

Further observations of the A2125 field can yield more quantitative information on the IGM,
galaxies, and dark matter in the superstructure. Ongoing optical photometry and spectroscopic
surveys of the field will provide information on both the galaxy population and dynamics. The field
is also ideal for probing the large-scale mass distribution by observing weakly lensed background
galaxies, and for measuring the hot gas column density by mapping out microwave background
distortions through the Sunyaev-Zel'dovich effect (Persi et al. 1995).

We thank G. Bryan and J. Ostriker for valuable comments on the work, and acknowledge the
support by NASA under grants NAG-5-2716, NAG-5-2717 (QDW), and AR-06394 (AJC) as weh
as the Graduate Student Researchers Program (RJB).

Table 1: Spectral Properties of the Three Major Components^

Parameter

A2125

Cluster B

LSBXE

R.A. (J2000)

15 41 06.3

15 41 10.6

15 40 2

Dec. (J2000)

+66 16 10

+66 26 26

+66 10

Size (arcmin)

5.0

2.5

~8xll

Count rate (10^^ counts s^-*^)

42.9

20.6

18.9

xVd.o.f

27.4/28

8.6/18

4.8/9

kT (keV)

2.3(1.6-4.5)

1.6(1.2-2.8)

0.85 (0.46-1.3)

Nh{1^'^^ cm-2)

2.5(1.7-3.4)

3.6(2.5-6.0)

2.9 (fixed)

Luminosity (10^^/i^q ergs s"^)

15

6.4

4.0

^Uncertainties in parameter values, as presented in parentheses, are at the 90% confidence leveL The luminosities are
estimated with the best-fit spectral parameters and in the z — 0.247 rest-frame energy range of 0.5-2 keV.

REFERENCES

Bertin, E., & Arnouts, S. 1996, A&A, 117, 393

Bryan, G., et al. 1994, ApJ, 428, 405

Buote, D. A., & Canizares, C. R. 1994, ApJ, 427, 86

Butcher, H., & Oemler, Jr. A. 1984, ApJ, 285, 426

Cavaliere, A., & Fusco-Femiano, R. 1976, A&A, 49, 137

Cen, R., & Ostriker, J. P. 1994, ApJ, 429, 4

Cen, R., Kang, H., Ostriker, J. P., & Ryu, D. 1995, ApJ, 451, 436

David, L. P., Jones, C, & Forman, W. 1996, ApJ, 473, 692

Doe, S. M., Ledlow, M. J., Burns, J. O., k White, R. A. 1995, AJ, 110, 46

Edge, A. C, & Stewart, G. C. 1991, MNRAS, 252, 414

Elsenstein, D. J., & Loeb, A. 1995, ApJ, 439, 520

Fukugita, M., Shimasaku, K., & Ichikawa, T. 1995, PASP, 107, 945

Hasinger, G., et al. 1994, Legacy, 4, 40

Kaiser, N. 1995, ApJ, 383, 104

Kauffmann, G. 1995, MNRAS, 274, 161

Mushotzky, R. F., et al. 1996, ApJ, 466, 686

Neumann, D. M., Sz Bohringer, H. 1996, preprint astro-ph/960706^

Ostriker, J. P., k Cen, R. 1996, ApJ, 464, 27

Persi, F. M., et al. 1995, ApJ, 442, 1

Schindler, S., k Wambsganss, J. 1996, A&A, 313, 113

Snowden, S. L., McCammon, D., Burrows, D. N., k Mendenhall, J. A. 1994, ApJ, 424, 71

Soltan, A. M., Hasinger, G., Egger, R., Snowden, S., Triimper, J. 1995, A&A, 303, 17

Stark, A. A., et al. 1992, ApJS, 79, 77

Triimper, J. 1992, QJRAS, 33, 165

Wang, Q. D., Ulmer, M., k Lavery, R. 1997, MNRAS, submitted

Wang, Q. D., k Ye, T. 1996, New Astronomy, 1, 245

This preprint was prepared with the AAS lATp^X macros v4.0.

Fig. 1. — PSPC image of A2125 and its vicinity in the 0.5-2 keV band. The image, corrected
for exposure, is smoothed adaptively with a Gaussian, the size of which is adjusted at each pixel
to achieve a count-to-noise ratio of 4. Each contour is 50% {2a) above its lower level; the lowest
contour is at 1.8 x 10^^ counts s^^ arcmin^^. Detected X-ray source positions are represented with
crosses; those extended ones are further marked with squares.

Fig. 2. — The image is the same as in Fig. 1, except it is smoothed after point-like sources are
excised.

Fig. 3. — Overlays of the diffuse X-ray surface brightness contours (Fig. 2) on blue-band CCD
images covering A2125 and the LSBXE region (a), Cluster B (b), and Cluster C (c).

Fig. 4. — X-ray substructures of A2125. The upper panel shows the PSPC surface brightness
residuals after the best-fit elliptical (3 model is subtracted; the contours are at -6.8, -2.8, 5.2,
9.2, 17.2, and 33.2 xlO~^ counts s~^ arcmin"^. The lower panel presents the significance of the
residuals, measured with the residual-to-noise ratio; the contours are at -2, -1, 1, 2, 3, 4, 5a.
Negative contours outline regions of possible over-subtraction of the model.

This figure "figl.gif ' is available in "gif ' format from:

http://arXiv.org/ps/astro-ph/9702018vl

This figure "fig2.jpg" is available in "jpg" format from:

http://arXiv.org/ps/astro-ph/9702018vl

This figure "fig3a.jpg" is available in "jpg" format from:

http://arXiv.org/ps/astro-ph/9702018vl

This figure "fig3b.jpg" is available in "jpg" format from:

http://arXiv.org/ps/astro-ph/9702018vl

This figure "fig3c.jpg" is available in "jpg" format from:

http://arXiv.org/ps/astro-ph/9702018vl

This figure "fig4.gif" is available in "gif ' format from:

http://arXiv.org/ps/astro-ph/9702018vl