MASARYK UNIVERSITY
Faculty of Science
Department of Theoretical Physics and Astrophysics
HABILITATION
From supermassive black holes
to the large-scale structure of the Universe
Norbert Werner
2016
2
Abstract:
The Lambda Cold Dark Matter (ΛCDM) model has been shown to be remarkably successful
in describing many aspects of our Universe, which is dominated by dark energy
and dark matter. However, the physical processes associated with the evolution of the
baryonic matter are complex and still poorly understood. In the course of structure
formation, only a small fraction of the baryons turned into stars - most remain in a diffuse
intergalactic medium (IGM). The growth of galaxies is regulated by feedback processes,
such as energy and momentum input from supernovae, and jets and winds of
accreting supermassive black holes. These processes, collectively called galactic feedback,
can limit further gravitational collapse, and thus a detailed knowledge of how
they work is essential for our understanding of galaxy formation and evolution. Here
I present my study of the intimate connection between the IGM, star formation and the
growth of supermassive black holes in the most massive galaxies.
Keywords: clusters of galaxies, giant elliptical galaxies, large-scale structure, intergalactic
medium, active galactic nuclei, chemical enrichment, feedback, supernovae,
cosmology, X-ray astronomy
Contents
1 Introduction 1
2 Feedback under the microscope: heating, gas uplift, and mixing in the nearest
cluster core 9
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Data reduction and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.1 Chandra data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.2.2 XMM-Newton RGS data . . . . . . . . . . . . . . . . . . . . . . . . 16
2.2.3 Hα data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.3.1 Spatial distribution of the X-ray gas . . . . . . . . . . . . . . . . . 17
2.3.2 Morphology of the Hα nebulae: new clues about their powering 20
2.3.3 High-resolution XMM-Newton RGS spectra . . . . . . . . . . . . 20
2.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.4.1 Orientation of the X-ray arms . . . . . . . . . . . . . . . . . . . . . 21
2.4.2 Physical properties and origin of the X-ray arms . . . . . . . . . . 24
2.4.3 Implications for gas motions in the ICM . . . . . . . . . . . . . . . 26
2.4.4 The coolest X-ray emitting phase . . . . . . . . . . . . . . . . . . . 28
2.4.5 The nature of the Hα ﬁlaments . . . . . . . . . . . . . . . . . . . . 29
2.4.6 Implications for galaxy formation models . . . . . . . . . . . . . . 30
2.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3 Violent interaction between the AGN and the hot gas in the core of the galaxy
cluster S´ersic 159-03 33
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.2 Data reduction and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.1 Optical, Hα, and UV data . . . . . . . . . . . . . . . . . . . . . . . 35
3.2.2 Radio data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2.3 Chandra X-ray data . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2.4 XMM-Newton RGS data . . . . . . . . . . . . . . . . . . . . . . . . 37
3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.3.1 Optical and UV properties . . . . . . . . . . . . . . . . . . . . . . . 39
3.3.2 Radio morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.3.3 X-ray imaging: a disturbed morphology . . . . . . . . . . . . . . . 42
3.3.4 Thermodynamic properties of the core . . . . . . . . . . . . . . . . 45
4 CONTENTS
3.3.5 High resolution X-ray spectra . . . . . . . . . . . . . . . . . . . . . 45
3.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.4.1 Displacement of gas from the cD galaxy . . . . . . . . . . . . . . . 48
3.4.2 Cooling of the displaced gas and star-formation . . . . . . . . . . 49
3.4.3 The powerful radio mode AGN . . . . . . . . . . . . . . . . . . . . 51
3.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4 On the thermodynamic self-similarity of the nearest, most relaxed, giant ellipticals
55
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.2 Sample selection, data reduction, and analysis . . . . . . . . . . . . . . . 57
4.2.1 Sample selection and its properties . . . . . . . . . . . . . . . . . . 57
4.2.2 Chandra data reduction and analysis . . . . . . . . . . . . . . . . . 64
4.3 Thermodynamic measurements . . . . . . . . . . . . . . . . . . . . . . . . 65
4.3.1 2D distribution of thermodynamic properties . . . . . . . . . . . . 65
4.3.2 Azimuthally-averaged, deprojected thermodynamic properties . 66
4.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.4.1 Universal thermodynamic proﬁles . . . . . . . . . . . . . . . . . . 68
4.4.2 Heating in the innermost r 1 kpc . . . . . . . . . . . . . . . . . . 70
4.4.3 The origin of the hot gas . . . . . . . . . . . . . . . . . . . . . . . . 70
4.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
from resonant scattering 73
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.2 Sample and XMM-Newton observations . . . . . . . . . . . . . . . . . . . 77
5.3 Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
5.3.1 XMM-Newton RGS data analysis . . . . . . . . . . . . . . . . . . . 79
5.3.2 Chandra analysis of NGC 4636 . . . . . . . . . . . . . . . . . . . . 81
5.4 Observations of resonant scattering . . . . . . . . . . . . . . . . . . . . . . 82
5.5 Modelling of resonant scattering in NGC 4636 . . . . . . . . . . . . . . . 84
5.6 Discussion and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . 88
6 Deep Chandra observation and numerical studies of the nearest cluster cold
front in the sky 95
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.2 Observations and data analysis . . . . . . . . . . . . . . . . . . . . . . . . 99
6.2.1 Chandra data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.2.2 XMM-Newton data . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.4 Numerical simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
6.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
6.5.1 The width of the cold front and suppressed diffusion . . . . . . . 109
6.5.2 Kelvin-Helmholtz instabilities and ICM viscosity . . . . . . . . . 113
6.5.3 Gas mixing at the cold front . . . . . . . . . . . . . . . . . . . . . . 114
CONTENTS 5
6.5.4 Conduction across the discontinuity . . . . . . . . . . . . . . . . . 114
6.5.5 Ampliﬁed magnetic ﬁelds underneath the cold front . . . . . . . 115
6.5.6 The ﬂow of the sloshing gas . . . . . . . . . . . . . . . . . . . . . . 116
6.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7 Deep Chandra study of the truncated cool core of the Ophiuchus cluster 119
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
7.2 Observations and data analysis . . . . . . . . . . . . . . . . . . . . . . . . 121
7.2.1 Chandra data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
7.2.2 Radio JVLA data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3.1 X-ray imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
7.3.2 X-ray spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
7.3.3 The power-spectra of surface brightness ﬂuctuations . . . . . . . 132
7.3.4 Radio properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
7.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
7.4.1 Dynamical activity and the truncated cooling core . . . . . . . . . 134
7.4.2 The radio phoenix . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
7.4.3 The cold fronts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
7.4.4 Turbulence and thermal conduction in the hot ICM . . . . . . . . 137
7.4.5 The concave surface brightness discontinuity: giant AGN outburst
or merger related gas dynamics? . . . . . . . . . . . . . . . 138
7.4.6 Suppressed AGN feedback . . . . . . . . . . . . . . . . . . . . . . 139
7.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
8 The nature of ﬁlamentary cold gas in the core of the Virgo Cluster 141
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
8.2 Observations and data analysis . . . . . . . . . . . . . . . . . . . . . . . . 146
8.2.1 Far Infrared spectroscopy with Herschel PACS . . . . . . . . . . . 146
8.2.2 HST Hα+[N II] and FUV data . . . . . . . . . . . . . . . . . . . . . 146
8.2.3 Chandra X-ray data . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
8.2.4 Long slit optical spectra . . . . . . . . . . . . . . . . . . . . . . . . 147
8.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
8.3.1 Cold and warm gas phases . . . . . . . . . . . . . . . . . . . . . . 147
8.3.2 The soft X-ray emission . . . . . . . . . . . . . . . . . . . . . . . . 150
8.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
8.4.1 Magnetized ﬁlaments of multi-phase material . . . . . . . . . . . 151
8.4.2 The energy source of the ﬁlaments . . . . . . . . . . . . . . . . . . 153
8.4.3 X-ray line emission from cold gas due to charge exchange and
inner shell ionization? . . . . . . . . . . . . . . . . . . . . . . . . . 154
8.4.4 AGN uplift induced cooling instabilities or mixing? . . . . . . . . 156
8.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
6 CONTENTS
9 The origin of cold gas in giant elliptical galaxies and its role in fueling radiomode
AGN feedback 161
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
9.1.1 The galaxy sample . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
9.2 Observations and data analysis . . . . . . . . . . . . . . . . . . . . . . . . 165
9.2.1 Far-infrared spectroscopy with Herschel PACS . . . . . . . . . . . 165
9.2.2 Hα+[N II] imaging and spectroscopy . . . . . . . . . . . . . . . . . 166
9.2.3 Chandra X-ray data . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
9.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
9.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
9.4.1 Properties of the cold ISM . . . . . . . . . . . . . . . . . . . . . . . 177
9.4.2 Heating and ionization of the cold ISM . . . . . . . . . . . . . . . 180
9.4.3 Origin of the cold ISM . . . . . . . . . . . . . . . . . . . . . . . . . 182
9.4.4 Fueling the AGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
9.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
10 A uniform metal distribution in the intergalactic medium of the Perseus cluster
of galaxies 189
11 Summary and outlook 195
References 201
Chapter 1
Introduction
The Lambda Cold Dark Matter (ΛCDM) model has been shown to be remarkably successful
in describing many aspects of our Universe, which is dominated by dark energy
and dark matter. However, the physical processes associated with the evolution of the
baryonic matter are complex and still poorly understood. In the course of structure
formation, only a small fraction of the baryons turned into stars - most remain in a
diffuse intergalactic medium (IGM). My research focuses on the intimate connection
between the IGM, star formation and the growth of supermassive black holes in the
most massive galaxies.
The growth of galaxies is regulated by feedback processes, such as energy and momentum
input from supernovae, and jets and winds of accreting supermassive black
holes, also called active galactic nuclei (AGN). This galactic feedback stops the inward
ﬂow of gas onto galaxies, heats the interstellar medium, preventing it from cooling
and forming further stars, and ejects baryons back into the surrounding IGM, most of
which is relatively hot in the low redshift Universe. These processes can limit further
gravitational collapse, and thus a detailed knowledge of how they work is essential
for our understanding of galaxy formation and evolution. I study the stellar and AGN
feedback using data across a wide range of wavelengths, including the radio, infrared,
optical, ultraviolet, and with a particular emphasis on the X-ray band.
The IGM has very low densities and therefore it is mostly difﬁcult to observe. However,
in the gravitational potentials of giant elliptical galaxies, groups and clusters of
galaxies, the density and temperature of the IGM becomes high enough to emit observable
X-ray radiation. This hot IGM, that in galaxy clusters is called the intra-cluster
medium (ICM), has temperatures of Te ∼ 56–108 K (0.5–10 keV) and densities declining
from n ∼ 10−2 cm−3 near the cluster centers to 10−4 cm−3 in the outskirts. It consists
of completely ionized hydrogen and helium, with highly ionized heavier elements of
roughly the Solar abundance in the centre, decreasing to about one third of the Solar
metallicity in the outskirts. Most of its observed continuum emission is due to the interaction
of electrons and ions, which leads to the bremsstrahlung emission of X-ray
photons. De-excitation of the ions of heavier elements, leading to transitions to the Kand
L-shells, may result in observable line emission in the X-ray band. The ICM is in
a collisional ionization equilibrium, where the ionization and recombination rates are
entirely determined by collisions, and is optically thin (the radiation ﬁelds in clusters
2 Introduction
do not affect its ionization state). The shape of observed X-ray spectra is therefore completely
determined by the ICM temperature and metallicity. The normalization of the
observed spectrum is proportional to the emission measure EM = np ne dV, where
np and ne are the proton and electron number densities and V is the volume of the
emitting region.
The research presented in the enclosed papers is in a large part based on deep observations
of galaxy clusters with the US-led Chandra X-ray observatory, European-led
XMM-Newton satellite, and the Japan-led Suzaku mission, using X-ray spectroscopy as
the main tool to asses the physical properties of the hot plasma. I also present work
based on radio data from the Giant Metrewave Radio Telescope (GMRT) and the Jansky
Very Large Array (VLA), far-infrared data obtained by the Herschel Space Observatory,
and optical data from various facilities including the SOAR telescope and the
Hubble Space Telescope.
Using my expertise in high-resolution X-ray spectroscopy, and a broad experience
in both the acquisition and interpretation of multi-wavelength data, I address here the
following key questions:
1. How do black holes regulate the growth of structure?
It is now well established that black holes play a vital role in the evolution of their
host galaxy. There are two major mechanisms through which AGN can strongly affect
their surrounding environment: one is radiative AGN feedback, where the radiation
from vigorously accreting black holes couples with the cold gas in the host galaxy; the
other is the so-called radio-mechanical feedback, where jets from AGN accreting at a
modest rate heat or push the gas out.
Here, I discuss the radio-mechanical feedback mode, which is thought to dominate
the evolution of massive galaxies over the past 10 billion years. In the absence of heating,
the hot (∼ 107 K) X-ray emitting haloes surrounding giant elliptical galaxies (either
haloes of their own, or because they are part of a group or cluster of galaxies) would
cool radiatively and form stars, building much larger galaxies than are seen. However,
X-ray studies with Chandra and XMM-Newton have shown that the expanding,
AGN jet inﬂated, bubbles ﬁlled by relativistic plasma (seen as ‘radio lobes’ by radio
telescopes) displace the hot gas, creating cavities in the X-ray emitting plasma (e.g.
McNamara et al. 2005, see the left panel of Fig. 1) and drive weak shocks that heat the
surrounding medium isotropically, preventing it from cooling.
In the ﬁrst and second enclosed paper (Chapter 2 and 3), we show that in their
wakes, the rising bubbles ﬁlled by relativistic plasma uplift low entropy gas from the
innermost regions of their host galaxies mixing it with the ambient hot medium. The
total gas mass uplifted by the buoyantly rising relativistic plasma in M 87 is ∼ 6–
9 × 109 M (see the right panel of Fig. 1), which shows that the AGN is capable to
strip the galaxy of its lowest entropy gas, preventing star-formation (see Chapter 2 for
details). On the other hand, in Chapter 3 we show that the uplift and displacement of
the low entropy gas from the direct vicinity of the central AGN can temporarily break
3
Figure 1.1: Left panel: Composite optical, X-ray (blue), radio (red) image of the galaxy
cluster MS0735.6+7421. X-ray observations with Chandra have shown that jets from
the central AGN inﬂate bubbles ﬁlled by relativistic, radio-emitting plasma that displace
the hot gas, creating cavities in the X-ray images of galaxy clusters (McNamara
et al. 2005). Right panel: In its wake, the buoyantly rising relativistic, radio-emitting
plasma injected by the AGN jets, uplifts large amounts of low entropy X-ray emitting
gas (shown in white), preventing it from cooling and forming stars. (From our press
release by the Chandra X-ray Center)
the feedback cycle and lead to cooling without triggering an AGN feedback response
and resulting in star formation that is offset from the center of the galaxy. The observed
unchecked cooling where a black hole is offset from the cooling gas solidiﬁes the idea
that AGN play a key role in maintaining the cooling/heating balance in cluster cores.
Radio-mechanical AGN feedback appears to maintain a long-lived delicate balance
between heating and cooling in the hot X-ray emitting atmospheres of massive galaxies,
groups, and clusters of galaxies. However, there is very little consensus on how
this balance is maintained and how cooled or cooling gas fuels the AGN. To create
a feedback loop, the thermal state of the hot atmosphere must inﬂuence the power
output of the AGN. A particularly important outstanding question, addressed by my
research, is therefore: How does AGN feedback operate across spatial scales of over 8
orders of magnitude - from the immediate vicinity of the black hole to a scale of several
hundreds of kpc in clusters of galaxies? The feedback loop is maintained most
naturally if the supermassive black hole is fueled by gas from the hot atmosphere at a
rate that depends on the thermal state of the gas. The remarkable similarity of the hot
X-ray emitting atmospheres surrounding some of the nearest brightest giant elliptical
galaxies (see the enclosed paper in Chapter 4) is consistent with a picture of hot accretion
leading to jet heating, which creates an energy balance where heating and cooling
are in equilibrium, keeping the hot galactic atmospheres in a steady state.
As these jet inﬂated bubbles expand and rise buoyantly in the hot plasma, they drive
gas motions, including turbulence. A critical, missing piece in our picture of AGN
feedback have been observational measurements of gas motions, because the spectral
4 Introduction
Figure 1.2: The X-ray emitting hot plasma in the centers of galaxies where our Herschel
observations revealed cold gas (including NGC 4636 and NGC 5044) appears to
be much more disturbed than in the cold gas-free systems (such as NGC 4472 and
NGC 1399). This is a sign that outbursts from the central black hole have occurred,
possibly driven, in part, by clumpy, cold gas that has been pulled onto the black hole.
These outbursts dump most of their energy into the center of the galaxy, where the
cold gas is located. By stirring and partially destroying the cold gas, this energy deposition
may prevent the cold gas from cooling sufﬁciently to form stars. (From our press
release by the Chandra X-ray Center)
5
resolution of the CCD-type detectors used on the XMM-Newton, Chandra, and Suzaku
satellites is insufﬁcient to measure the velocity broadening of spectral lines. The intensity
of some spectral lines in the X-ray bright cores of elliptical galaxies, groups and
clusters of galaxies is suppressed by resonance scattering (line photons get absorbed
and quickly re-emitted in a different direction). Since the optical depth of a resonance
line depends on the characteristic velocity of small-scale motions, its measurement
allows us to determine turbulence independently of line broadening (Gilfanov et al.
1987). In the fourth enclosed paper (Chapter 5), we developed a new technique where
we measure the resonance scattering of X-ray spectral lines observed with the reﬂection
grating spectrometers on the XMM-Newton satellite to determine the AGN induced
turbulence in the hot plasma surrounding giant elliptical galaxies. This powerful technique
provided the ﬁrst observational constraints on velocities in the hot plasma in
these systems. Turbulence in the ICM continues to be of major interest for the understanding
of the evolution of galaxy clusters. Recently, we found strong indications
that turbulent heat dissipation may be of crucial importance for the heating of the ICM
(Zhuravleva et al. 2014a).
A missing key ingredient in our models is the lack of knowledge of the microphysical
properties of the hot IGM/ICM, which determine its thermal conductivity and
viscosity. This limits our understanding of how the energy from accreting black holes
couples with the hot diffuse ICM - e.g. how do gas motions driven by the rising bubbles
dissipate into heat and how is the heat distributed across the ICM. The hot ICM is
permeated by weak, tangled magnetic ﬁelds and as such its microphysics remains untreatable
in a purely theoretical way. New observational constraints are essential if we
are to make progress. The spectacular, sub-arcsecond imaging capabilities of the Chandra
X-ray Observatory will remain unsurpassed for at least the next 20 years, and ultradeep
observations with this instrument still provide opportunities for breakthroughs
in our understanding of the microphysics of the hot plasma. In the enclosed paper presented
in Chapter 6, I present a deep, 500 ks, legacy-class Chandra observation of the
nearest, best resolved cold front in the sky, which lies 90 kpc to the north of M87. Cold
fronts are believed to be due to the sloshing (or swirling) of the gas in the gravitational
potential of clusters and they are known to be remarkably sharp, with gas density and
temperature discontinuities at least several times sharper than the Coulomb mean free
path. Because this observation provided an extraordinary improvement in resolution,
resolving scales smaller than 80 pc, it allowed us to place crucial constraints on fundamental
physical processes shaping cold fronts, such as: growth of hydrodynamic
instabilities in the ICM (placing constraints on viscosity), suppression of thermal conductivity,
and structure of intra-cluster magnetic ﬁelds. The observation of the Ophiuchus
cluster described in Chapter 7 indicates that thermal conduction in the hot ICM
is not suppressed completely. The data also show evidence for both Rayleigh-Taylor
and Kelvin-Helmholtz instabilities, implying relatively low ICM viscosity. These observations
provide important constraints for the models of the evolution of baryons in
cosmological simulations.
However, many central group/cluster galaxies are also surrounded by spectacular,
ﬁlamentary nebulae containing large amounts of cold and ionized gas. Because of the
presence of many different gas and plasma phases in these systems, the use of multi-
6 Introduction
Figure 1.3: The combined energy of supernova explosions and accreting supermassive
black holes at redshifts 2–4 (before the period of cluster formation) must have been
strong enough to expel most of the metals from the galaxies at early times, and enrich
and mix the intergalactic gas. Exploding supernovae and accreting black holes were
the spoon that mixed the metals into the intergalactic medium. (From our JAXA press
release)
wavelength data is crucial, if we wish to understand how heating and cooling operate
throughout the putative AGN feedback loop. The enclosed papers in Chapters 8 and
9 report the results of our multi-wavelength survey of some of the nearest brightest
giant elliptical galaxies using far-infrared spectral imaging with Herschel PACS, optical
narrow band imaging and spectroscopy with the SOAR telescope, and X-ray spectral
imaging with Chandra. We ﬁnd a dichotomy, where some of the systems are cold-gasfree
and others harbor relatively large quantities of cold gas which is produced chieﬂy
by thermally unstable cooling from the hot phase (see Fig. 2). The jets in these systems
appear to couple efﬁciently to the cold gas phases and our observations indicate that
radio-mechanical AGN feedback is likely to play a crucial role in clearing giant elliptical
galaxies of their cold gas, keeping them red and dead. Intriguingly, our analysis
also revealed indications for an anti-correlation between jet power and cold molecular
gas content, favoring hot accretion models.
2. How did the star formation and chemical enrichment of the Universe proceed?
Supernovae inject large amounts of energy into the surrounding medium and enrich
it with heavy elements (collectively called metals). The deep gravitational potential
wells of galaxy clusters keep all of the metals ever produced by the stellar populations
7
of the member galaxies, making them archeological treasure troves to study the integrated
history of star-formation. The dominant fraction of metals in clusters currently
resides within the hot X-ray emitting IGM (for a review see Werner et al. 2008).
Most of the chemical elements with prominent line emission in the X-ray band, from
O up to Ni, are produced by supernovae. The observed equivalent widths of the emission
lines can be directly converted into elemental abundances. While core-collapse
supernovae produce large amounts of O, Ne, and Mg, Type Ia supernovae produce
large quantities of Fe, Ni, and Si-group elements, but very little O, Ne, and Mg.
Our groundbreaking studies of the outskirts of galaxy clusters (see Simionescu et al.
2011, 2013, 2015; Urban et al. 2011, 2014; Werner et al. 2013) have allowed us to measure
the ICM metallicity out to the virial radii of the Perseus and Virgo clusters. These observations
provide important hints that activity of accreting supermassive black holes
also played an important role in the distant past. In the enclosed paper in Chapter 10,
we show that the measured iron abundance has about a third of the Solar value and its
distribution is remarkably uniform both as a function of radius and azimuth out to the
cluster outskirts. This homogeneous distribution requires that most of the metal enrichment
of the intergalactic medium occurred before clusters formed, probably more
than ten billion years ago, during the period of maximal star formation and black hole
activity. The combined energy of supernova explosions and AGN must have been
strong enough to expel most of the metals from the galaxies at early times, and enrich
and mix the intergalactic gas. This gas was later accreted by clusters and virialized
(increasing its entropy through shock heating) to form the present ICM. Observations
of the Virgo Cluster, that also allow us to measure the abundances of Si, S, and Mg
out to the outskirts (Simionescu et al. 2015), show that the chemical composition of the
intra-cluster medium is constant on large scales and is generally consistent with that
of our own Galaxy.
8 Introduction
Chapter 2
Feedback under the microscope:
heating, gas uplift, and mixing in the
nearest cluster core
N. Werner1, A. Simionescu1, E. T. Million1, S. W. Allen1, P. E. J. Nulsen2, A. von der Linden1,
S. M. Hansen3, H. B¨ohringer4, E. Churazov5,6, A. C. Fabian7, W. R. Forman2, C. Jones2,
J. S. Sanders7, and G. B. Taylor8
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 382 Via
Pueblo Mall, Stanford, CA 94305-4060, USA
and SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA
94025, USA
2Harvard-Smithsonian Center for Astrophysics, 60 Garden St., Cambridge, MA 02138,
USA
3University of California Observatories & Department of Astronomy, University of
California, Santa Cruz, CA 95064, USA
4Max-Planck-Institut f¨ur extraterrestrische Physik, Giessenbachstr, 85748 Garching, Ger-
many
5Max-Planck-Institut f¨ur Astrophysik, Karl-Schwarzschild-Strasse 1, 85741 Garching,
Germany
6Space Research Institute (IKI), Profsoyznaya 84/32, Moscow 117810, Russia
7Institute of Astronomy, Madingley Road, Cambridge CB3 0HA
8Department of Physics and Astronomy, University of New Mexico, Albuquerque, NM
87131, USA
Published in the Monthly Notices of the Royal Astronomical Society, volume 407,
pages 2063–2074, 2010
10 AGN heating and gas uplift in M 87
Abstract
Using a combination of deep (574 ks) Chandra data, XMM-Newton high-resolution
spectra, and optical Hα+[N II] images, we study the nature and spatial distribution of
the multiphase plasma in M 87. Our results provide direct observational evidence
of ‘radio mode’ AGN feedback in action, stripping the central galaxy of its lowest
entropy gas and therefore preventing star-formation. This low entropy gas was entrained
with and uplifted by the buoyantly rising relativistic plasma, forming long
“arms”. A number of arguments suggest that these arms are oriented within 15◦–30◦
of our line-of-sight. The mass of the uplifted gas in the arms is comparable to the gas
mass in the approximately spherically symmetric 3.8 kpc core, demonstrating that the
AGN has a profound effect on its immediate surroundings. The coolest X-ray emitting
gas in M 87 has a temperature of ∼0.5 keV and is spatially coincident with Hα+[N II]
nebulae, forming a multiphase medium where the cooler gas phases are arranged in
magnetized ﬁlaments. We place strong upper limits of 0.06 M /yr (at 95 per cent conﬁdence)
on the amount of plasma cooling radiatively from 0.5 to 0.25 keV and show
that a uniform, volume-averaged heating mechanism could not be preventing the cool
gas from further cooling. All of the bright Hα ﬁlaments in M 87 appear in the downstream
region of the < 3 Myr old shock front, at smaller radii than ∼0.6 . We suggest
that shocks induce shearing around the ﬁlaments, thereby promoting mixing of the
cold gas with the ambient hot ICM via instabilities. By bringing hot thermal particles
into contact with the cool, line-emitting gas, mixing can supply the power and ionizing
particles needed to explain the observed optical spectra. Furthermore, mixing of
the coolest X-ray emitting plasma with the cold optical line emitting ﬁlamentary gas
promotes efﬁcient conduction between the two phases, allowing non-radiative cooling
which could explain the lack of X-ray gas with temperatures under 0.5 keV.
2.1 Introduction
If the energy radiated away at the centres of so-called “cool-core” clusters of galaxies,
which show sharp X-ray surface brightness peaks and central temperature dips, came
only from the thermal energy of the hot, diffuse intra-cluster medium (ICM), the ICM
would cool and form stars at rates orders of magnitude above what the observations
suggest (see Peterson & Fabian 2006, for a review). It is currently believed that the energy
which offsets the cooling is provided by the interaction between the active galactic
nuclei (AGN) in the central dominant galaxies and the ICM (see Churazov et al. 2000,
2001, 2002; McNamara & Nulsen 2007, for a review). Through a tight feedback loop, it
is thought that the AGN can provide enough energy to prevent catastrophic cooling.
Recent deep observations of nearby bright cooling core clusters such as Perseus, M 87,
Centaurus, and Hydra A revealed AGN induced weak shocks and sound waves with
sufﬁcient energy ﬂux to counterbalance the radiative cooling (Sanders & Fabian 2007;
Forman et al. 2005; Sanders & Fabian 2008; Nulsen et al. 2005; Simionescu et al. 2009a).
To the ﬁrst order it is thus understood where the energy comes from and how it gets
transported. However, the detailed physics of the AGN/ICM interaction is not yet
2.1 Introduction 11
clear.
M 87, the central dominant galaxy of the Virgo cluster, at a distance of only 16.1 Mpc
(Tonry et al. 2001) is the nearest, X-ray bright cool core. It is an ideal system for detailed
studies of the energy input from the AGN to the hot, cooling ICM. After the Perseus
cluster, M 87 is the second brightest extended extragalactic object in soft X-rays. The
most striking X-ray features in M 87 are its two “X-ray arms”, ﬁrst detected by Feigelson
et al. (1987) using Einstein Observatory data and later studied in detail by B¨ohringer
et al. (1995) using Rosat and VLA radio observations. The X-ray arms extend to the
East and Southwest from the centre of the galaxy and they spatially correlate with the
radio emission. B¨ohringer et al. (1995) found that the X-ray arms are due to cooler
gas, possibly uplifted from the centre of the galaxy. These results and the subsequent
high-quality radio data by Owen et al. (2000) led Churazov et al. (2001) to argue that
the X-ray and radio morphology can be explained by bubbles of radio-emitting plasma
rising buoyantly through the hot ICM. These rising bubbles entrain and uplift cooler
gas from the centre of the galaxy, which then further adiabatically cools. Churazov
et al. (2001) suggested that the buoyantly rising bubbles dissipate energy into sound
waves, internal waves, turbulent motion in the wake, and potential energy of the uplifted
gas, all of which could provide heating to the cooling core region (see also e.g.
Br¨uggen & Kaiser 2002; Kaiser 2003; Br¨uggen 2003; De Young 2003; Ruszkowski et al.
2004a,b; Heinz & Churazov 2005).
Observations with Chandra and XMM-Newton greatly enhanced our view of M 87.
Using XMM-Newton observations, Belsole et al. (2001), Matsushita et al. (2002), and
Molendi (2002) showed that the regions associated with radio arms have two-temperature
(in the ranges of kT ∼ 0.8–1 keV and kT ∼ 1.6–2.5 keV) or multi-temperature structure.
However, plasma below 0.8 keV is largely absent (B¨ohringer et al. 2002; Sakelliou et al.
2002). Young et al. (2002) analyzed an early, relatively short (40 ks) Chandra observation
which showed cavities and edges in the hot plasma. They compared the X-ray
morphology with the 6 cm radio (Hines et al. 1989) and Hα+[N II] emission (Sparks
et al. 1993) and showed that the optical ﬁlaments lie outside of X-ray cavities ﬁlled with
radio plasma; some are along the edges of the cavities, and several optical ﬁlaments coincide
with knots of cooler X-ray gas. Using a longer (∼150 ks) Chandra observation,
Sparks et al. (2004) studied the relation of the X-ray and optical ﬁlaments and concluded
that electron conduction from the hot X-ray emitting plasma to the cooler phase
provides a quantitatively acceptable energy source for the optical ﬁlaments. Using the
same Chandra data, Forman et al. (2005) identiﬁed shock fronts associated with an
AGN outburst about 1–2×107 yr ago. They argued that shock fronts may be the most
signiﬁcant channel for heating (i.e., entropy increase) of the ICM near to the AGN.
Subsequent Chandra observations increased the total exposure time to 500 ks, which
allowed Forman et al. (2007) to resolve a web of ﬁlamentary structure in both X-ray
arms. Both arms show narrow ﬁlaments, with a length-to-width ratio of up to ∼50.
However, while the Southwestern arm has only a single set of ﬁlaments, the Eastern
arm shows multiple sets of ﬁlaments and bubbles, increasing in scale with the distance
from the centre. The high spatial resolution of Chandra also highlights the fact that
while the Eastern X-ray arm is cospatial with the radio arm, the radio emission bends
around and avoids the Southwestern X-ray arm. A residual image of M 87 showing
12 AGN heating and gas uplift in M 87
Table 2.1: Summary of the Chandra observations. Colums list the observation ID, detector,
observation mode, exposure after cleaning, and observation date.
Obs. ID Detector Mode Exposure (ks) Obs. date
2707 ACIS-S FAINT 82.9 Jul. 6 2002
3717 ACIS-S FAINT 11.1 Jul. 5 2002
5826 ACIS-I VFAINT 126.8 Mar. 3 2005
5827 ACIS-I VFAINT 156.2 May 5 2005
5828 ACIS-I VFAINT 33.0 Nov. 17 2005
6186 ACIS-I VFAINT 51.5 Jan. 31 2005
7210 ACIS-I VFAINT 30.7 Nov. 16 2005
7211 ACIS-I VFAINT 16.6 Nov. 16 2005
7212 ACIS-I VFAINT 65.2 Nov. 14 2005
the inner cavities and the X-ray arms obtained using all Chandra ACIS-I data is shown
in Fig. 2.1 (from Million et al. 2010a), together with the 90 cm radio image by Owen
et al. (2000). The temperature structure and metallicity of the X-ray arms was studied
in detail by Simionescu et al. (2008) using deep (120 ks) XMM-Newton data. These
authors conﬁrmed that the arms are multiphase with a temperature distribution between
0.6–3.2 keV. This temperature distribution is also consistent with that found by
spectral ﬁts to the XMM-Newton Reﬂection Grating Spectrometer (RGS) data (Werner
et al. 2006), which show the presence of weak Fe XVII lines. Simionescu et al. (2008)
estimate the total mass of gas below 1.5 keV, most of which was uplifted by the AGN,
as 5 × 108 M .
In this paper, we study for the ﬁrst time the detailed nature and the spatial distribution
of the multiphase plasma associated with the X-ray arms. Chandra allows us to
study detailed structure in the cooler phases at a much higher spatial resolution than is
possible with XMM-Newton. Our data allow us to sample well the small substructure
seen in the Chandra images, map the spatial distribution of the individual temperature
components and compare it with new Hα images obtained at the Lick Observatory, and
archival data from the Hubble Space Telescope (HST). In Section 2, we describe the data
reduction and analysis; in Section 3, we present the results of the observations; in Section
4, we determine the physical properties of the cooler plasma phases and discuss
the uplift of the low entropy gas from the bottom of the gravitational potential well,
and effects of shocks and mixing. Finally, in Section 5, we summarize our conclusions.
A distance to M 87 of 16.1 Mpc (Tonry et al. 2001) implies a linear scale of 4.7 kpc arcmin−1.
All errors are quoted at the 68 per cent conﬁdence level for one interesting parameter.
2.1 Introduction 13
10.0 12:31:00.0 50.0 40.0 30.0 30:20.0
30:00.0
28:00.0
26:00.0
24:00.0
22:00.0
12:20:00.0
18:00.0
Right ascension
Declination
10 kpc
10.0 12:31:00.0 50.0 40.0 30.0 30:20.0
30:00.0
28:00.0
26:00.0
24:00.0
22:00.0
12:20:00.0
18:00.0
Right ascension
Declination
10 kpc
Figure 2.1: Left panel: Relative deviations of the surface brightness from a double beta
model ﬁt to the Chandra 0.6–2.0 keV image of M 87, obtained by coadding all ACIS-I
observations (see Paper II for details). The cavities in the core and the two X-ray arms
are clearly visible. The XMM-Newton RGS extraction region is overplotted. Right
panel: The 90 cm radio image by Owen et al. (2000). Two ﬂows (radio arms) emerge
from the inner-jet region, one directed Eastward (spatially coincident with the X-ray
arm) and the other directed to the Southwest. While the Eastern radio arm ends in
edge-brightened torus-like vortex rings, the Southwestern radio arm develops a Sshaped
Southward twist. Both radio arms are immersed in a pair of large, partially
overlapping radio lobes. Both panels show the same 14.6 ×15.8 ﬁeld.
14 AGN heating and gas uplift in M 87
2.2 Data reduction and analysis
2.2.1 Chandra data
The Chandra observations of M 87 were taken between July 2002 and November 2005
using the Advanced CCD Imaging Spectrometer (ACIS). The observations are listed
in Table 9.2. The total net exposure time after cleaning is 574 ks. We follow the data
reduction procedure described in Million et al. (2010a), hereafter referred to as Paper II
(see also Million & Allen 2009; Million et al. 2010b).
The individual regions for the spectral analysis were determined using the Contour
Binning algorithm (Sanders 2006), which groups neighboring pixels of similar surface
brightness until a desired signal-to-noise threshold is met. In order to have small
enough regions to resolve substructure and still have enough counts to ﬁt a simple
multi-temperature model, we adopted a signal-to-noise ratio of 50. The total number
of regions in this analysis is ∼6000, with a total number of ∼15,000,000 net counts.
Background spectra for the appropriate detector regions were extracted from the
blank-sky ﬁelds available from the Chandra X-ray Center. These were normalized by
the ratio of the observed and blank-sky count rates in the 9.5 −12 keV band (the statistical
uncertainty in the observed 9.5 − 12 keV ﬂux is less than 5 per cent in all cases). The
background level in these observations is low and represents only a marginal source
of systematic uncertainty in the determined quantities.
Spectral modeling has been performed with the SPEX package (Kaastra et al. 1996,
SPEX uses an updated version of the MEKAL plasma model with respect to XSPEC).
We analyze data in the 0.6–7.0 keV band. To investigate the multi-temperature structure
and map the spatial distribution of plasmas at different temperatures, we ﬁt to
each bin a model consisting of collisionally ionized equilibrium plasmas at four ﬁxed
temperatures, with variable normalizations and common metallicity. This method was
introduced by Sanders et al. (2004) in the analysis of deep Chandra observations of the
Perseus cluster. For M 87, the temperatures are ﬁxed at 0.5, 1.0, 2.0, and 3.0 keV. The
temperatures of the coolest and hottest components were selected based on previous
multi-temperature analysis of XMM-Newton RGS and EPIC data (Werner et al. 2006;
Simionescu et al. 2008). We also searched for a 0.25 keV component in our spectral
ﬁts, but did not detect any emission at this temperature. The temperatures of the individual
spectral components are sufﬁciently far apart that the spectral analysis can
constrain their emission measures if they are simultaneously present in the same extraction
region. The overall metallicity in each region is free to vary. The O/Fe ratio
is ﬁxed at 0.59 Solar, Ne/Fe at 1.20 Solar, and the Mg/Fe at 0.60 Solar, as determined
from XMM-Newton RGS spectra (Werner et al. 2006). The Galactic absorption toward
M 87 is modelled as neutral gas with Solar abundances with a column density ﬁxed at
NH = 1.94 × 1020 cm−2, the value determined by the Leiden/Argentine/Bonn (LAB)
Survey of Galactic H I (Kalberla et al. 2005). Throughout the paper, abundances are
given with respect to the “proto-solar values” by Lodders (2003), which are for O, Ne,
and Fe approximately 30 per cent lower than the values given by Anders & Grevesse
(1989).
2.2 Data reduction and analysis 15
50 100 150 200
10.0 05.0 12:31:00.0 55.0 50.0 45.0 30:40.0 35.0
26:00.0
25:00.0
24:00.0
23:00.0
22:00.0
21:00.0
12:20:00.0
19:00.0
18:00.0
Declination
10 kpc
200 400 600
10.0 05.0 12:31:00.0 55.0 50.0 45.0 30:40.0 35.0
26:00.0
25:00.0
24:00.0
23:00.0
22:00.0
21:00.0
12:20:00.0
19:00.0
18:00.0
Declination
10 kpc
200 600 1000
10.0 05.0 12:31:00.0 55.0 50.0 45.0 30:40.0 35.0
26:00.0
25:00.0
24:00.0
23:00.0
22:00.0
21:00.0
12:20:00.0
19:00.0
18:00.0
Declination
10 kpc
0 100 200 300 400 500
10.0 05.0 12:31:00.0 55.0 50.0 45.0 30:40.0 35.0
26:00.0
25:00.0
24:00.0
23:00.0
22:00.0
21:00.0
12:20:00.0
19:00.0
18:00.0
Declination
10 kpc
Figure 2.2: Spatial distribution of the emission measure, Y = nHnedV, (in
1058 cm−3 arcsec−2) of the 0.5 keV (upper left), 1.0 keV (upper right), 2.0 keV (bottom
left), and 3.0 keV (bottom right) plasma detected at 99.7 per cent conﬁdence. On
the upper right panel, we over-plotted the contours of the 90 cm radio image (Owen
et al. 2000). All the panels show the same 10.45 ×9.75 ﬁeld. For a zoomed in version
of the spatial distribution of 0.5 keV plasma in the core see the right panel of Fig. 3.1.
16 AGN heating and gas uplift in M 87
2.2.2 XMM-Newton RGS data
The multi-temperature structure of the hot plasma in and around M 87 exhibits a complex
set of spectral lines that cannot be modelled by a single-temperature spectral
model (Sakelliou et al. 2002; Werner et al. 2006). In order to conﬁrm that our simple
four-temperature model describes the temperature structure of M 87 well, we ﬁt
the same model to the XMM-Newton RGS spectra.
The XMM-Newton RGS data were obtained in January 2005, with a net exposure
time of 84 ks. The data were processed as described in Werner et al. (2006). Spectra
were extracted from a 1.1 wide extraction region, centred at the core of the galaxy. Because
the RGS operates without a slit, it collects all photons from within the 1.1 × ∼12
ﬁeld of view. Line photons originating at angle ∆θ (in arcminutes) along the dispersion
direction are shifted in wavelength by ∆λ = 0.138∆θ ˚A. Therefore, every line is
broadened by the spatial extent of the source. To account for this spatial broadening in
our spectral model, we produce a predicted line spread function (LSF) by convolving
the RGS response with the surface brightness proﬁle of the galaxy derived from the
EPIC/MOS1 image along the dispersion direction. Because the radial proﬁle of a particular
spectral line can be different from the overall radial surface brightness proﬁle,
the line proﬁle is multiplied by a scale factor s, which is the ratio of the observed LSF
to the expected LSF. This scale factor is a free parameter in the spectral ﬁt.
We ﬁt the spectra in the 8–38 ˚A band with the same four-temperature model ﬁtted
to the Chandra data, with an additional 0.25 keV component included to determine an
upper limit on the amount of such cool plasma. We have also separately ﬁtted the RGS
data with a model consisting of four isobaric cooling ﬂows with elemental abundances
tied between the models (similar to one of the models used for the Centaurus cluster in
Sanders et al. 2008). In this case, we model separately gas cooling from 3 keV to 2 keV,
from 2 keV to 1 keV, from 1 keV to 0.5 keV, and from 0.5 keV to 0.25 keV. The central
AGN cannot be excluded from the RGS data of the core of the galaxy and also needs
to be accounted for. We model the AGN spectrum with a power-law of a photon index
γ = 1.95 and a 2–10 keV luminosity of LX = 1.15 × 1042 erg s−1 (Werner et al. 2006).
2.2.3 Hα data
Narrow-band Hα imaging (central wavelength 6606 ˚A, FWHM 75 ˚A) of M 87 was carried
out using the Shane 3 m telescope at Lick Observatory on March 26th, 2009. The
total integration time was 24 ks, split into 20 exposures using a dither pattern with
∼ 5 arcsec offsets. We also acquired 1200 seconds of R band imaging, also split into
20 exposures. The data were processed using the pipeline of Erben et al. (2005). A
constant background was subtracted, estimated from the region of the image with the
lowest counts. The images were registered with scamp (Bertin 2006), and resampled
and combined with swarp (Bertin et al. 2002). The seeing of the coadded images is
about 2 arcsec.
To study the morphology of the brighter Hα ﬁlaments in the cluster core, we have
also analyzed two Hα images available in the HST archive. Both images were taken
with WFPC2, placing the core onto the Planetary Camera. The Southeastern region
2.3 Results 17
was imaged with the Wide Field Camera for only 2700 seconds (proposal ID: 5122, PI
Ford), and is not as deep as our ground-based image. The Northwestern region was
imaged signiﬁcantly longer (13900 seconds, proposal ID: 6296, PI Ford).
2.3 Results
2.3.1 Spatial distribution of the X-ray gas
Fig. 7.4 shows the spatial distribution of the emission measures for the four individual
temperature components (0.5, 1.0, 2.0, and 3.0 keV) obtained from the Chandra data.
The top left panel shows an interesting result: the map reveals the presence of gas with
a temperature of kT ∼0.5 keV to the North of the jet in the core of the galaxy, in the
Southwestern arm, and in a horseshoe like ﬁlament to the Southeast of the core outside
of the mushroom shaped Eastern radio arm (radio contours are overplotted in the top
right panel of Fig. 7.4). No 0.5 keV gas with signiﬁcance higher than 3σ is present
within the Eastern radio arm. While all the detected 0.5 keV plasma in the core and
in the Southeastern horseshoe is within the RGS extraction region (see the left panel of
Fig. 2.1), the Southwestern arm is outside of the area covered by the present RGS data.
The spatial distribution of the ∼1 keV plasma seen in the top right panel of Fig. 7.4
correlates with the radio emission and is very similar to the morphology of the X-ray
arms seen clearly in Fig. 2.1. In the core, the 1 keV plasma forms a prominent ridge
to the North of the AGN, which is cospatial with 0.5 keV plasma and Hα+[N II] emission
(see below). Based on two-temperature spectral ﬁts to XMM-Newton data it has
been shown that the X-ray arms are relatively isothermal at ∼1 keV (Molendi 2002;
Simionescu et al. 2008) and our spectral ﬁts to the Southwestern X-ray arm conﬁrm
that its temperature is constant as a function of radius (see Fig. 2.3). Most of the Eastern
radio arm contains ∼1 keV plasma, a large amount of which is also present to the
South of the “stem of the radio mushroom” where the 0.5 keV horseshoe is observed.
The brightest part of the Southwestern radio arm is not spatially coincident with the
0.5 keV and 1 keV plasma, but it bends around the cooler X-ray gas. On the Southwest,
just outside of the core region, the 1 keV plasma forms a narrow and remarkably
straight and smooth ﬁlament, which broadens with increasing distance from the core.
At the projected radius of ∼3 , which is the approximate distance of the circular shock
front described in Paper II and by Forman et al. (2007), the distribution of 1 keV gas
broadens, and at 4.5 it starts to bend to the Southeast.
The spatial distribution of the plasma at ∼ 2 keV is, in comparison, remarkably
symmetric and only slightly elongated in the direction of the X-ray arms. A knot of
∼3 keV plasma is present at the inner edge of the X-ray cavity South of the core, at the
bottom of the bright radio cocoon, which Forman et al. (2005) identiﬁed as a “bud”.
This 3 keV plasma is most probably associated with a shock front at the radius of 0.6
(Forman et al. 2007, Paper II, see also the right panel of Fig. 2.5).
The fraction of the emission measure of cooler gas (1 keV and 0.5 keV) in the Xray
arms with respect to the total emission measure is 10–35 per cent, consistent with
the results obtained by ﬁtting a differential emission measure distribution to XMM-
18 AGN heating and gas uplift in M 87
1 2 3 4 5
0.511.522.5
kT(keV)
Radius (arcmin)
Figure 2.3: The temperature proﬁle of the Southwestern X-arm ﬁtted with a twotemperature
model. While the temperature of the hotter component shows a slight
radial increase (black data points) the radial distribution of the cooler temperature
component (red data points) looks remarkably ﬂat at around 1 keV. If the uplift were
adiabatic, the temperature of the cooler gas within the arm would be radially decreas-
ing.
2.3 Results 19
56.0 55.0 54.0 53.0 52.0 51.0 12:30:50.0 49.0 48.0 47.0 46.0
24:00.0
50.0
40.0
30.0
20.0
10.0
12:23:00.0
50.0
40.0
30.0
Right ascension
Declination
Ηα+[ΝΙΙ]
1kpc
56.0 55.0 54.0 53.0 52.0 51.0 12:30:50.0 49.0 48.0 47.0 46.0
24:00.0
50.0
40.0
30.0
20.0
10.0
12:23:00.0
50.0
40.0
30.0
Right ascension
Declination
0.5 keV component
1kpc
Figure 2.4: The Hα emission (left panel) and plasma at ∼0.5 keV (right panel; blue
contours in the left panel) spatially correlate in the core and form a horseshoe like
feature to the Southeast of the core. The right panel is a zoomed in version of the top
left panel of Fig. 7.4. Both panels show the same 2.7 ×1.85 ﬁeld.
52.0 12:30:50.0 48.0 46.0
20.0
10.0
12:24:00.0
50.0
40.0
30.0
20.0
10.0
23:00.0
50.0
Right ascension
Declination
Ηα+[ΝΙΙ]
1 kpc
0.007
0.006
0.005
0.004
0.003
0.002
Pressure
1 kpc
Figure 2.5: Left panel: Hα image of the innermost 2 ×2 region of M 87, divided by the
best ﬁt De Vaucouleurs proﬁle, obtained by HST. Right panel: Chandra pressure map
(in units of keV cm−3 × 1
2Mpc
−1/2
, see Paper II for details on producing the pressure
map) of the same inner 2 ×2 region overplotted with the 6 cm radio emission contours
from Hines et al. (1989). The pressure map clearly shows a discontinuity, a likely shock
front, at a radius of 0.6 . The Hα image shows that all the bright Hα ﬁlaments appear
in the downstream region of the shock. The arrows identify the bright Hα features
furthest from the centre, at the edges of apparently underpressured X-ray cavities, and
a bright Hα “knee” on the Southeast.
20 AGN heating and gas uplift in M 87
Newton data by Simionescu et al. (2008). The spatial distribution of the coolest gas
phases is also in good agreement with that found in the XMM-Newton data analysis
(Simionescu et al. 2008), but is revealed by Chandra with a strikingly better spatial
resolution.
The elemental abundances obtained from the ﬁts are likely to be biased due to the
limitations of our model, which has its temperatures ﬁxed at speciﬁc values (detailed
study of the metallicity will be presented in Million et al. in prep.; see also Paper II and
Simionescu et al. 2008).
2.3.2 Morphology of the Hα nebulae: new clues about their power-
ing
The left panel of Fig. 3.1 shows an excerpt of the Hα+[N II] image obtained at the
Lick Observatory after subtraction of the scaled R-band continuum image. Owing to
the large depth, our data reveal a previously undiscovered horseshoe-like ﬁlament of
Hα+[N II] emission to the Southeast of the core. The feature is detected at a mean
level of 5σ above the sky. Fig. 3.1 shows that the spatial distribution of the Hα+[N II]
emission is remarkably similar to that of the 0.5 keV plasma, both in the core and in the
Southeastern horseshoe like region. Except for a small knot of Hα+[N II] emission in
the “cap of the mushroom” (see Gavazzi et al. 2000), the Eastern radio arm is devoid of
observable Hα ﬁlaments. As mentioned above, the Eastern radio arm is also devoid of
any detectable 0.5 keV emission. Our Lick Observatory data (Fig. 3.1) and the HST data
(Fig. 2.5) unfortunately do not cover the Southwestern X-ray arm.
Fig. 2.5 shows an Hα image for the innermost 2 ×2 region of M 87 from HST next
to a pressure map obtained in Paper II by using Chandra data. In the pressure map
we see a discontinuity, likely shock front, at 0.6 and several underpressured regions
ﬁlled with radio emitting plasma, as shown by the contours of 6 cm radio emission
from Hines et al. (1989). Assuming that this inner shock propagates at an approximately
constant velocity of 1000 km s−1, it originated in an AGN outburst ∼2.8 Myr
ago. Interestingly, all the bright Hα emission is within the inner higher pressure region
surrounded by the shock front at 0.6 . Some ﬁlaments appear to be parallel to the shock
front, others seem to lie at the edges of cavities, while the ﬁlament to the Southeast of
the core is perpendicular to the shock front, bright in the downstream region and faint
in the upstream. Upstream of the shock front, the Southeastern Hα ﬁlament continues
at a much lower surface brightness and forms the prominent horseshoe. On the
Hα image, we identify with arrows the bright features furthest from the centre, at the
edges of X-ray cavities, and a bright Hα “knee” on the Southeast. The same arrows are
plotted in the pressure map.
2.3.3 High-resolution XMM-Newton RGS spectra
To conﬁrm the presence of the coolest gas phases indicated by ﬁtting the Chandra data
and to verify that the simple four-temperature model ﬁts the data sufﬁciently well, we
also analyze the XMM-Newton RGS high-resolution line spectra. The results of the
2.4 Discussion 21
ﬁve-temperature ﬁt to the RGS data are listed in Table 3.3 and the best ﬁt model to the
spectra is shown in Fig. 3.7.
The emission measures for the 2 keV and 1 keV phases obtained by the RGS are
consistent with those from the Chandra maps for the area covered by the gratings. The
emission measure of the 0.5 keV phase obtained using the RGS is higher by a factor of 3
than the value from the Chandra map. The sensitivity of Chandra in the 0.6–7 keV band
to 0.5 keV emission in these moderate S/N spectra is limited and if we only consider
regions where the 0.5 keV component was detected at larger than the 99.7% conﬁdence
level then we miss the cool emission in some regions. By including all regions detected
at the 68% conﬁdence the 0.5 keV features in the map become wider and the agreement
between the emission measures from Chandra and RGS improves.
The lack of O VII lines provides strong constraints on the amount of 0.25 keV plasma.
The 95 per cent upper limit for its emission measure is lower by a factor of 11 than the
emission measure of the 0.5 keV plasma.
To get constraints on the amount of radiative cooling, we also ﬁt the RGS spectra
with a set of cooling ﬂow models (see Table 3.3). If the plasma would cool isobarically
in the absence of heating, the best ﬁt mass deposition rates would be the same in the
different temperature ranges. However, the mass deposition rates are decreasing as a
function of temperature and for the gas cooling from 0.5 keV to 0.25 keV we determine
a tight 95 per cent upper limit of ˙M = 0.06 M /yr.
2.4 Discussion
2.4.1 Orientation of the X-ray arms
The inferred properties of the X-ray arms depend critically on their orientation with
respect to our line-of-sight. Based on HST observations of the superluminal motion
in the jet of M 87, Biretta et al. (1999) concluded that the position angle of the jet is
< 19◦ from our line-of-sight. The two large, partially overlapping, outer radio lobes
of M 87 (the diffuse structures which embed the more collimated, brighter radio arms
seen in the right panel of Fig. 2.1 and in Owen et al. 2000) are also probably oriented
close to our line-of-sight. It is likely that the Southern radio lobe is positioned towards
us and the Northern lobe is oriented away from us. This picture is supported by the
observation of polarized ﬂux from the Southern lobe (Andernach et al. 1979), which
shows that the Faraday depth towards it is likely to be small. The radius of the large
lobes/bubbles is ∼22 kpc and their approximate centres are in projection 26 kpc apart.
Assuming that the physical distance between the centres of the bubbles is at least twice
their radius, a conservative limit for the orientation of the axis along which the bubbles
rise is < 35◦ from our line-of-sight. The Southwestern radio arm appears to “merge”
into the Southern lobe indicating that its orientation is similar to that of the two large
bubbles. Although the relation of the Eastern radio arm to the outer lobe is less clear,
the most likely orientation of both X-ray and radio arms is approximately anti-parallel
and far from the plane of the sky.
22 AGN heating and gas uplift in M 87
Table 2.2: Best ﬁt parameters for a ﬁve-temperature ﬁt and a four-cooling-ﬂow model
ﬁt to the high-resolution RGS spectra extracted from a 1.1 wide region centred on the
core of M 87. Emission measures, Y = nHnedV, are given in 1063 cm−3. The scale
factor s is the ratio of the observed LSF to the expected LSF based on the overall radial
surface brightness proﬁle. The upper limits are quoted at their 95 per cent conﬁdence
level. Abundances are quoted with respect to the proto-solar values of Lodders (2003).
Parameter 5T-model 4-c.f. model
Y0.25keV < 0.07 –
Y0.5keV 0.79 ± 0.05 –
˙M0.5−0.25keV – < 0.06
Y1.0keV 7.1 ± 0.3 –
˙M1.0−0.5keV – 0.90 ± 0.03
Y2.0keV 56.3 ± 1.3 –
˙M2.0−1.0keV – 3.79 ± 0.25
Y3.0keV < 2.4 –
˙M3.0−2.0keV – 6.03 ± 0.21
s 2.21 ± 0.10 2.19 ± 0.11
C 0.75 ± 0.21 0.82 ± 0.22
N 1.9 ± 0.3 2.0 ± 0.3
O 0.79 ± 0.04 0.84 ± 0.04
Ne 1.92 ± 0.13 1.92 ± 0.13
Mg 1.25 ± 0.15 1.26 ± 0.13
Fe 1.33 ± 0.04 1.39 ± 0.05
Ni 0.9 ± 0.3 0.9 ± 0.3
2.4 Discussion 23
10 20 30
00.10.2
Counts/s/Å
wavelength (Å)
OVIIILα
FeXVII
NeX
FeXVIII/OVIIILβ
FeXVII
FeXXIII
FeXXII/
FeXX
FeXIX
FeXVIII
Figure 2.6: The ﬁrst order RGS spectrum extracted from a 1.1 wide region centred on
the core of M 87. The continuous line represents the best ﬁt model to the spectrum.
24 AGN heating and gas uplift in M 87
2.4.2 Physical properties and origin of the X-ray arms
We determine the properties of the X-ray arms for three different orientation angles
(15◦, 30◦, and 90◦) from our line-of-sight. For each orientation, we determine the radial
proﬁles of the gas mass, entropy, and cooling time for the different phases (see Fig. 4.6).
We assume that all the gas phases are in equilibrium with the ambient pressure at the
given distance from the centre. The radial distribution of the ambient pressure is determined
using the deprojected temperature and electron density proﬁles parametrized
by equations (6) and (7) in Churazov et al. (2008). Using this pressure proﬁle p(r), we
calculate the electron number density proﬁle ne(r) = p(r)/(kT) of the kT = 0.5 keV
and 1 keV plasmas. Using these density proﬁles and the best ﬁt emission measures,
Y = nHnedV, for each extraction region, we determine the volumes which the cooler
phases occupy: V = Y/(nHne), where nH = ne/1.2 is the hydrogen number density
for collisionally ionized plasma with Solar metallicity.
The line-of-sight depths, l, of the emitting volumes, V, shown in Fig. 2.8, are then
calculated as l = V/A, where A is the area on the sky of the given extraction region.
The corresponding gas masses are calculated as M = 1.4nHmpV (where mp is the proton
mass), only in the spatial regions where their presence was determined with better
than 3σ signiﬁcance. The cumulative radial gas mass proﬁles are shown in the bottom
panel of Fig. 4.6. The middle panel of this ﬁgure shows the radial entropy proﬁles calculated
as S = kT/n2/3
e . The top panel shows the cooling time proﬁles calculated as
the gas enthalpy divided by the energy lost per unit volume of the plasma:
tcool =
5
2(ne + ni)kT
neniΛ(T)
, (2.1)
where the ion number density ni = 0.92ne and Λ(T) is the cooling function for Solar
metallicity tabulated by Sutherland & Dopita (1993). Cooling functions based on more
up to date plasma codes (Schure et al. 2009) predict for the 0.5 keV plasma a 9 per cent
higher cooling rate.
If the arms lie in the plane of the sky, then the small line-of-sight depths of the
emitting volumes in Fig. 2.8 indicate that the 1 keV gas is not volume ﬁlling at most
radii, including the core of the galaxy. The 1 keV gas is then made of small blobs and
ﬁlaments with an entropy smaller than the lowest entropy of the ambient medium in
the centre of the galaxy (see Fig. 4.6). This would mean, as pointed out by Molendi
(2002), that the cool blobs/ﬁlaments can hardly originate from the adiabatic evolution
of the hot-phase gas entrained by buoyant radio bubbles, as suggested by Churazov
et al. (2001). However, if the arms are oriented close to our line-of-sight, the physical
picture changes.
For the likely angles of 15◦–30◦ from our line-of-sight, both the radial distribution
of the gas entropy and the distribution of the depths of the emitting volumes of the
1 keV component give a more consistent picture with other observables. The 1 keV
plasma appears to be mostly volume ﬁlling, except in regions where the X-ray images
reveal cavities. Its cooling time is >1.5 × 108 yr, which is comparable to or longer than
the uplift time. Moreover, by taking the projection effects properly into account, the
apparent large angle between the current direction of the jet and the Southwestern arm
2.4 Discussion 25
Figure 2.7: Cumulative gas mass, entropy (S = kT/n2/3), and radiative cooling time of
the 0.5 keV phase (red lines), 1.0 keV phase (blue lines), and the ambient ICM (black
lines) as a function of projected radius for orientation angles of 15◦ (dashed lines), 30◦
(dotted lines), and 90◦ (full lines) from our line-of-sight. For the deprojected density
and temperature proﬁles of the ambient ICM we assume the proﬁles parametrized by
equations (6) and (7) in Churazov et al. (2008). The cooler gas phases are assumed
to be in pressure equilibrium with the ambient medium (see text for details). The
vertical dotted lines denote the radii of the outer edge of the ridge of multiphase gas
North of the jet, the end of the Southeastern horseshoe, and the radius of the shock at
∼ 3 , which corresponds to the projected radius of the inner edge of the “cap of the
mushroom” of the Eastern arm.
26 AGN heating and gas uplift in M 87
translates into a much smaller physical angle. The inferred spatial distribution of the
depths of the emitting volumes (see Fig. 2.8) indicates that the orientation angle of the
ﬁlament is not constant and at certain positions the arms are folded along our line-ofsight.
The Southwestern arm appears to be folded at the projected radius of 3.4 , where
the arm divides into two ﬁlaments (see Fig. 2.1). The Eastern arm appears to have folds
at ∼2.1 and its line-of-sight depth increases in the cap of the mushroom. We caution
that because of the strongly simpliﬁed assumptions the values for the depths quoted
in the ﬁgures are approximate numbers only.
If the uplift of the low entropy plasma were strictly adiabatic, then the temperature
of the arms would decrease as a function of radius. However, as shown by Molendi
(2002), Simionescu et al. (2008), and in Fig. 2.3 using a two-temperature model, the
average temperature of the arms is remarkably constant. The constant temperature
implies rising entropy with radius and suggests that conduction plays an important
role in the thermal evolution of the arms.
The estimated total mass of 1 keV gas in the arms is between 6–9 × 108 M , which
given the uncertainties is remarkably similar to the total gas mass of 6.3 × 108 M
within the innermost 3.8 kpc region (0.8 in the plane of the sky), where the X-ray
emission is relatively spherically symmetric. Outside of this radius, all the ∼1 keV
gas is distributed along the two X-ray arms. If the entropy of the arms is increasing
due to heat conduction, then most of the 6–9×108 M of ∼1 keV plasma might have
been uplifted from within the innermost approximately spherical 3.8 kpc region of the
galaxy. While this is a large fraction of the total gas mass in the core of the cluster, it
is only a fraction of the uplifted gas mass observed in more massive clusters; e.g. the
inferred mass of the uplifted gas in Hydra A is 1.6–6.1 × 109 M (Simionescu et al.
2009b). Though seen with great clarity in our study, the physics of gas uplift in M 87 is
by no means extreme. Many brightest cluster galaxies can be expected to be similarly
efﬁcient in evacuating their lowest entropy plasma during AGN outbursts.
By assuming that the outer radio halo is a spherical bubble of hot plasma undergoing
a steady energy input from the jet, Owen et al. (2000) estimated its age to ∼108 yr.
However, if the radio halo consists of two bubbles rising buoyantly in the hot ICM,
then the outer radio lobes are likely to be older. The age of the radio and X-ray arms
estimated by Churazov et al. (2001) is a few times 107 yr. However, if the whole Southwestern
arm originated in a single uplift following an AGN outburst, then for an orientation
angle of <30◦ from our line-of-sight, its length is >60 kpc, which assuming an
uplift velocity of 400 km s−1 implies an age of >1.5 × 108 yr. The age of the observed
structures is thus greater than previously thought.
2.4.3 Implications for gas motions in the ICM
The NW edge of the Southwestern X-ray arm is remarkably smooth out to the radius
of ∼3.3 and the ﬁlament is surprisingly straight for a distance of over 2 . This strongly
suggests that the gas motions cannot be very turbulent in either phase. The lack of
strong turbulence in the ICM has been inferred previously based on the straightness
of the Hα ﬁlaments in the Perseus cluster (Fabian et al. 2003b), and strong upper limits
on turbulent velocities were placed based on resonance scattering in NGC 4636
2.4 Discussion 27
Figure 2.8: Maps of the line-of-sight depths in kpc for the emitting volumes of 1 keV
plasma for orientation angles of 90◦ (top left panel), 30◦ (top right panel), and 15◦
(bottom left panel) from our line-of-sight. The map on the bottom right panel shows
the line-of-sight depth distribution of the 0.5 keV component for an orientation angle
of 30◦ from our line-of-sight. All the panels show the same 10.36 ×9.25 ﬁeld.
28 AGN heating and gas uplift in M 87
(Werner et al. 2009) and turbulent line broadening in Abell 1835 (Sanders et al. 2010b).
The bending of the X-ray arms of M 87 at larger radii might be caused by gas sloshing,
which creates a spiral like shearing motion pattern in the ICM which is the most
obvious at the two opposite and staggered cold fronts seen at 7 to the Southeast
(Simionescu et al. 2007) and at 19 to the North (Simionescu et al. 2010). The sloshing
induced shearing motions might also generate mixing of the ∼1 keV gas with the
ambient ICM at large radii.
2.4.4 The coolest X-ray emitting phase
The total mass of ∼0.5 keV gas in M 87 is at least 7 × 106 M . Because we calculate
the mass of the 0.5 keV component only in spatial regions where it was detected at
higher than the 99.7 per cent conﬁdence, this value is a conservative lower limit. The
inferred small line-of-sight depths of the emitting volumes of the 0.5 keV component
imply that the coolest X-ray emitting gas is not volume ﬁlling, but forms a multi-phase
medium together with both the hotter X-ray emitting and the cold Hα emitting gas
phases. Soft band X-ray images (Forman et al. 2007) reveal that this cooler phase forms
a network of ﬁlaments, which spatially coincide with Hα ﬁlaments (Sparks et al. 2004,
see also Fig. 3.1). Relatively cool (∼ 0.5 keV) X-ray gas associated with Hα ﬁlaments
has also been observed in the Perseus cluster and in 2A 0335+096 (Sanders & Fabian
2007; Sanders et al. 2009).
One of the most prominent regions in M 87 occupied by multiphase plasma is a
ridge to the North of the jet, which contains 2 keV, 1 keV, 0.5 keV, and Hα gas. This
region is just outside of the inner radio cocoon and possibly contributes to the conﬁnement
of the radio plasma on the Northern side and to the depolarization seen there by
Owen et al. (1990). The depolarization of the radio emission in this region overlapping
with the ﬁlaments indicates that the magnetic ﬁelds are disordered on scales less than
0.1 kpc. Therefore, it is likely that the Hα ﬁlaments consist of many narrow magnetized
threads, which are not perfectly aligned and are small compared to the observed beam.
Depolarization of the radio emission at regions containing Hα and soft X-ray emission
has been previously observed in the Centaurus cluster (Taylor et al. 2007).
Limits on cooling and constraints on viable heating mechanisms
Even though the cooling time of the ∼0.5 keV gas is relatively short (see the top panel
of Fig. 4.6), no X-ray emitting plasma below 0.5 keV is observed in M 87. The 95 per
cent conﬁdence upper limit for the emission measure of 0.25 keV plasma based on the
non-detection of O VII lines with RGS is Y = nHnedV = 7 × 1061 cm−3, which is only
9 per cent of the emission measure of the 0.5 keV plasma. Therefore, assuming pressure
equilibrium between different gas phases, the mass of the 0.25 keV phase is at least 22
times lower than the mass of the 0.5 keV plasma. Assuming the 1 keV gas is isobarically
cooling to 0.5 keV, its best ﬁt mass deposition rate is ˙M = 0.9 M yr−1, which is 15
times larger than the 95 per cent conﬁdence upper limit for the mass deposition rate
from 0.5 keV to 0.25 keV. The upper limits on gas cooling below 0.5 keV rely on the
O VII lines and thus on the O abundance of the cool gas, which we assumed to be the
2.4 Discussion 29
same as that of the hotter phase which is fairly constant across the core (Werner et al.
2006). For signiﬁcant amounts of gas cooling radiatively below 0.5 keV to be present
in the core, without detectable O VII emission, the O abundance of the cooling plasma
would have to be unrealistically low. The lack of larger amounts of gas radiatively
cooling below 0.5 keV implies the presence of heating. A similar temperature ﬂoor at
∼0.5 keV is observed in the Perseus cluster, Centaurus cluster, 2A 0335+096, Abell 262,
Abell 3581, and HCG 62, (Sanders & Fabian 2007; Sanders et al. 2008, 2009, 2010a).
The energy emitted per unit volume in a blob of 0.5 keV gas, that is in pressure
equilibrium with the surrounding 1 keV and 2 keV gas, is 4.8 times larger than the
energy emitted by the neighboring volume of gas at 1 keV and 30 times larger than
for the gas at 2 keV. Therefore, any heating process preventing the coolest gas phase
from cooling below the ∼0.5 keV temperature ﬂoor and providing energy uniformally
in a volume-averaged way would overheat the surrounding gas and is ruled out by
the data. The mechanism that heats the coolest gas clearly provides more energy per
unit volume to cooler, denser phases than it does to the hotter phases.
An alternative explanation is that the coolest X-ray plasma cools non-radiatively by
mixing. Sanders et al. (2010a) reached similar conclusions for a sample of cool core
clusters. The coolest X-ray plasma in M 87 might cool non-radiatively by mixing with
the cold optical line emitting ﬁlamentary gas, allowing efﬁcient conduction between
the two phases (Fabian et al. 2002; Soker et al. 2004). As indicated by the fact that all
the bright Hα+[N II] ﬁlaments end at the innermost shock front, this mixing might be
promoted by the shock, which induces enhanced shearing motions, and might also
power the detected optical and UV line emission (see the next section).
2.4.5 The nature of the Hα ﬁlaments
The optical emission line ﬁlaments in M 87 are likely due to cooled gas, enriched by
dust from stellar mass loss (Sparks et al. 1993), which has been uplifted by the buoyant
relativistic plasma, together with the coolest X-ray emitting phase. The total mass of
the ∼104 K Hα emitting gas in the ﬁlament system in M 87 is estimated to be of the
order of 105–107 M (Sparks et al. 1993), which is comparable to the mass of ∼0.5 keV
gas. However, the total mass of cold neutral and molecular gas based on the upper limits
on CO emission within the bright Hα emitting regions is at most only a few times
106 M (Salom´e & Combes 2008). This implies that the Hα emitting gas is not a thin
skin on the underlying clouds of neutral and molecular gas as in the Perseus cluster,
where the inferred total mass of the cold gas is as high as ∼4×1010 M (Salom´e et al.
2006). Recently, Sparks et al. (2009) discovered ﬁlaments of C IV emission in M 87, with
a total power of 1.1 × 1039 erg s−1, arising from gas at a temperature of ∼105 K, which
spatially coincides with the Hα ﬁlaments on all scales. This detection indicates that
the cool and hot gas phases are in thermal communication. The emission strengths are
consistent with thermal conduction, which would create this intermediate temperature
phase at the interface of the hot ICM and the cooler gas. However, even though conduction
can quantitatively provide the energy seen in the form of line emission (Sparks
et al. 2004), it would have to proceed at close to the Spitzer or saturated levels to power
the ﬁlaments. The very thin and long Hα ﬁlaments are magnetized (see Sect. 2.4.4 and
30 AGN heating and gas uplift in M 87
Fabian et al. 2008) and the likely orientation of the magnetic ﬁelds parallel to the interface
of the cold and hot phase would largely suppress the level of conduction. Thus it
seems unlikely that thermal conduction alone can explain the optical line emission.
All of the bright Hα emission in M 87 is observed in the downstream region of the
innermost <3 Myr old shock front, within a radius of ∼ 0.6 . This suggests that the
Hα emission in M 87 is somehow related to shocks in the ICM. The passage of a shock
accelerates the intra-cluster medium, but because of the large density contrast the relatively
cool Hα emitting gas is left behind, resulting in a sudden strong shear around
the ﬁlaments. It is likely that such shear promotes mixing of the cold gas with the ambient
hot ICM via instabilities. Ferland et al. (2009) recently showed that the spectra
of optical ﬁlaments observed around the central galaxies of cooling core clusters can
be explained if the ﬁlaments are heated by ionizing particles, either conducted from
the surrounding regions or produced in situ by MHD waves. By facilitating contact
between the hot thermal particles and the cold gas, mixing can supply the power and
the ionizing particles needed to explain the optical line emission spectrum. Mixing
of X-ray emitting plasma with the cold gas also naturally explains the presence of the
∼ 105 K intermediate temperature gas phase. If the coolest X-ray emitting gas cools
by mixing with the ﬁlamentary gas behind the shock front and conductively connects
with it (e.g. Begelman & Fabian 1990; Soker et al. 2004), then its thermal energy will
be radiated away in the UV band, in the optical Hα+[N II], and most likely at infrared
wavelengths in the presence of dust.
2.4.6 Implications for galaxy formation models
The series of recent ‘radio’ or ‘jet’-mode AGN outbursts in M 87, revealed in exquisite
detail by the Chandra data, by no means represent a unique or extreme phenomenon.
For example, a complete, ﬂux-limited study of eighteen optically and X-ray bright,
nearby galaxies, including M 87, by Dunn et al. (2010) shows that 17/18 sources exhibit
some form of current radio activity associated with the central AGN (see also
Best et al. 2005; Dunn & Fabian 2006, 2008; Sun 2009). Allen et al. (2006) also show
that the efﬁciency with which the accretion of hot gas powers jets in giant ellipticals,
including M 87, is remarkably uniform. Our results for the nearest, brightest cluster
core show that even the current, relatively modest level of radio mode AGN activity
has been sufﬁcient to remove a sizeable fraction of the coolest, lowest entropy gas from
the central galaxy. This is the material that would have cooled to form stars (e.g. Peterson
& Fabian 2006). Our results therefore provide direct evidence that radio mode
AGN input has a profound effect on the ability of large galaxies to form stars from their
surrounding X-ray emitting halos.
2.5 Conclusions
Using a combination of deep (574 ks) Chandra data, XMM-Newton high-resolution
spectra, and optical Hα+[N II] images, we have performed an in-depth study of AGN
2.5 Conclusions 31
feedback in M 87. Accounting for the fact that the long X-ray and radio “arms” are
seen in projection with likely angles of ∼ 15◦–30◦ from our line-of-sight, we ﬁnd that:
• The mass of the uplifted low entropy gas in the arms is comparable to the gas
mass in the approximately spherically symmetric 3.8 kpc core, demonstrating
that the AGN has a profound effect on its immediate surroundings. This result
has important implications for understanding AGN feedback in galaxy formation
models (e.g. Croton et al. 2006; De Lucia & Blaizot 2007; Sijacki et al. 2007).
• The coolest detected X-ray plasma, with a temperature of ∼0.5 keV, the Hα+[N II]
emitting gas, and the UV emitting gas are cospatial, and appear to be arranged in
magnetized ﬁlaments within the hot ambient atmosphere, forming a multi-phase
medium.
• We do not detect X-ray emitting gas with temperatures below ∼0.5 keV. To remain
in steady state and at constant pressure, the ﬁlaments of 0.5 keV gas would
require a 5 times higher heating rate than the 1 keV gas, and 30 times stronger
heating than 2 keV plasma. Thus, a uniform, volume-averaged heating mechanism
would overheat the surrounding gas and can be ruled out in M 87.
• All of the bright Hα and UV ﬁlaments are seen in the downstream region of the
< 3 Myr old shock front, within a radius of ∼0.6 . This argues that the generation
of Hα and UV emission in M 87 is related to shocks in the ICM.
Based on these observations we propose that shocks induce shearing around the
denser gas ﬁlaments, thereby promoting mixing of the cold gas with the ambient hot
ICM via instabilities. This process may both provide a mechanism for powering the
optical line emission and explain the lack of plasma with temperature under 0.5 keV:
• By helping to get hot thermal particles into contact with the cool, line-emitting
gas, mixing can supply the power and the ionizing particles needed to explain
the optical spectra of the Hα+[N II] nebulae (see Ferland et al. 2009).
• Mixing of the coolest X-ray emitting plasma with the cold optical line emitting
ﬁlamentary gas promotes efﬁcient conduction between the two phases, allowing
the X-ray gas to cool non-radiatively, which may explain the lack of gas with
temperatures under 0.5 keV.
Acknowledgments
Support for this work was provided by the National Aeronautics and Space Administration
through Chandra/Einstein Postdoctoral Fellowship Award Number PF8-90056
and PF9-00070 issued by the Chandra X- ray Observatory Center, which is operated by
the Smithsonian Astrophysical Observatory for and on behalf of the National Aeronautics
and Space Administration under contract NAS8-03060. This work was supported
in part by the US Department of Energy under contract number DE-AC02-76SF00515.
32 AGN heating and gas uplift in M 87
SMH is supported by the National Science Foundation Postdoctoral Fellowship program
under award number AST-0902010.
Chapter 3
Violent interaction between the AGN
and the hot gas in the core of the galaxy
cluster S´ersic 159-03
N. Werner1, M. Sun2, J. Bagchi3, S. W. Allen1, G. B. Taylor4, S. K. Sirothia5, A. Simionescu1,
E. T. Million6, J. Jacob7, M. Donahue8
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall, Stanford, CA 94305-4085, USA
and SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA
94025, USA
2Department of Astronomy, University of Virginia, P.O. Box 400325, Charlottesville,
VA 22904-4325, USA
3Inter University Center for Astronomy and Astrophysics (IUCAA), Post Bag 4, Ganeshkhind,
Pune 411 007, India
4Department of Physics and Astronomy, University of New Mexico, Albuquerque, NM
87131, USA
5National Centre for Radio Astrophysics, Tata Institute of Fundamental Research, Post
Bag 3, Ganeshkhind, Pune 411007, India
6Department of Physics and Astronomy, University of Alabama, Box 870324, Tuscaloosa,
AL 35487, USA
7Department of Physics, Newman College, Thodupuzha 685 585, India
8Physics&Astronomy Department, Michigan State University, East Lansing, MI 48824-
2320, USA
Published in the Monthly Notices of the Royal Astronomical Society, volume 415,
pages 3369–3379, 2011
34 AGN feedback in the cluster S´ersic 159-03
Abstract
We present a multi-wavelength study of the energetic interaction between the central
active galactic nucleus (AGN), the intra-cluster medium (ICM), and the optical emission
line nebula in the galaxy cluster S´ersic 159-03. We use X-ray data from Chandra,
high resolution X-ray spectra and UV images from XMM-Newton, Hα images from
the SOAR telescope, HST optical imaging, and VLA and GMRT radio data. The cluster
center displays signs of powerful AGN feedback, which has cleared the central regions
(r < 7.5 kpc) of dense, X-ray emitting ICM. X-ray spectral maps reveal a high pressure
ring surrounding the central AGN at a radius of r ∼ 15 kpc, indicating an AGN driven
weak shock. The cluster harbors a bright, 44 kpc long Hα+[N II] ﬁlament extending
from the centre of the cD galaxy to the north. Along the ﬁlament, we see low entropy,
high metallicity, cooling X-ray gas. The gas in the ﬁlament has most likely been uplifted
by ‘radio mode’ AGN activity and subsequently stripped from the galaxy due
to its relative southward motion. Because this X-ray gas has been removed from the
direct inﬂuence of the AGN jets, part of it cools and forms stars as indicated by the
observed dust lanes, molecular and ionized emission line nebulae, and the excess UV
emission.
3.1 Introduction
Supermassive black holes (SMBH) play a crucial role in galaxy formation. The correlation
between the mass of the central SMBH and the bulge mass of their host galaxies
suggests that their evolution is tightly coupled (Magorrian et al. 1998). Outbursts of
accreting SMBH (referred to as active galactic nuclei or AGN) disturb and heat the surrounding
gas, lowering the accretion rate and, in some cases, driving so much gas out
of the galaxy that the star formation is drastically reduced.
The most dramatic portrayal of the interaction between AGN and their surroundings
can be seen in the distribution of X-ray emitting gas in nearby giant ellipticals
and clusters of galaxies. Chandra X-ray images have revealed giant cavities and shocks
in the hot gas produced by repeated outbursts of the central AGN (e.g. Fabian et al.
2003a; Nulsen et al. 2005; Forman et al. 2005). These AGN outbursts in principle provide
enough power to offset radiative losses, suppress cooling, and prevent further star
formation in the dense cores of clusters of galaxies (see e.g. McNamara & Nulsen 2007).
Detailed X-ray imaging and 2D spectral mapping of cooling cores with cavities, X-ray
bright ﬁlaments, and shock fronts, provides the most reliable means of measuring the
energy injected into the hot intra-cluster medium (ICM) by AGN, and is a powerful
tool to study the physics of the AGN feedback in general (see e. g. Randall et al. 2011a;
Million et al. 2010a; Werner et al. 2010; Simionescu et al. 2009b).
Here we present a detailed multi-wavelength study of AGN feedback processes in
the core of the nearby (z=0.0564; Maia et al. 1987), relatively poor, low mass cluster of
galaxies S´ersic 159-03 (A S1101). The thermal properties and chemical enrichment of
the hot ICM in this cooling core cluster have been studied in detail using XMM-Newton
(Kaastra et al. 2001, 2004; de Plaa et al. 2006). Compared to other cooling core clusters,
3.2 Data reduction and analysis 35
Table 3.1: Summary of the radio observations. Columns list the central frequency, the
telescope array, the resolution of the beam, its position angle, the rms noise, and the
observing date.
Frequency (MHz) Array Resolution (arcsec) P.A. (deg) rms Noise (µJy beam−1) Obs. date
244 GMRT 20.04 × 8.92 11.0 1420 2007 November 2, 3
325 GMRT 17.67 × 9.08 -179.4 352 2009 May 16,17
617 GMRT 8.29 × 3.13 9.7 250 2009 May 09,11
1400 VLA-B 10 × 4 0.0 64 2006 August 8
8400 VLA-B 3.1 × 0.72 8.3 21 2006 August 8
where the central temperature drops to ∼1/3 of the peak ambient value, S´ersic 159-03
has a relatively modest (factor of 1.5) temperature drop in the core (Sun 2009). The
radio properties of this cluster have been reported by Bˆırzan et al. (2008), in the study
of a sample of 24 cooling cores. The bright optical emission line nebulae in the core of
the cluster have been studied in detail by Crawford & Fabian (1992) and more recently
by Oonk et al. (2010) using integral ﬁeld spectroscopy.
Throughout the paper we adopt a ﬂat ΛCDM cosmology with H0 = 70 km s−1 Mpc−1
and ΩM = 0.3, which implies a linear scale of 1.2 kpc arcsec−1 at the cluster redshift
of z = 0.0564 (Maia et al. 1987). Throughout the paper, abundances are given with
respect to the solar values by Grevesse & Sauval (1998). All errors are quoted at the 68
per cent conﬁdence level.
3.2 Data reduction and analysis
3.2.1 Optical, Hα, and UV data
To study the optical properties of the central cluster galaxy, we analyzed a 600 s observation
available in the Hubble Space Telescope (HST) archive (proposal ID: 8719). The
image was taken with WFPC2 using the F555W ﬁlter and placing the galaxy onto the
WF3 chip.
Narrow-band optical imaging was performed with the 4.1 m Southern Astrophysics
Research Telescope (SOAR) telescope on September 13, 2009 (UT) using the SOAR Optical
Imager (SOI). The night was photometric with a seeing of around 1 . We used the
6916/78 CTIO narrow-band ﬁlter for the Hα+[N II] lines, and the 6738/50 ﬁlter for the
continuum. We took four 1200 s exposures with the 6916/78 ﬁlter and four 750 s exposures
with the 6738/50 ﬁlter. We reduced each image using the standard procedures in
the IRAF MSCRED package using EG 274 as a spectroscopic standard. The pixels were
binned 2×2, for a scale of 0.154 per pixel. More details on the SOI data reduction can
be found in Sun et al. (2007).
We also analyzed near Ultra Violet data obtained by the XMM-Newton Optical Monitor
on 2002 November 20–21. Exposures were taken with the UVW1 (2400–3600 ˚A) and
UVW2 (1800–2400 ˚A) ﬁlters.
36 AGN feedback in the cluster S´ersic 159-03
3.2.2 Radio data
The summary of radio observations with the Giant Metrewave Radio Telescope (GMRT)1
and the Very Large Array (VLA), including the observation date, obtained beam, and
rms noise, is shown in Table 9.2. The strong, stable calibrator 3C 48 was used for absolute
ﬂux density and bandpass calibration at all frequencies. We tied the ﬁnal ﬂux
density scale to the standard ‘Baars-scale’ (Baars et al. 1977). The VLA data were obtained
in the B conﬁguration. Their calibration was performed using AIPS (Greisen
2003) in the standard fashion, while imaging and self-calibration were performed using
Difmap (Shepherd et al. 1995).
The GMRT data were reduced using AIPS++ (version: 1.9). After applying bandpass
corrections to the phase calibrator, gain and phase variations were estimated, and
ﬂux density, bandpass, gain and phase calibration were applied. The data for antennas
with high errors in antenna-based solutions were examined and ﬂagged over certain
time ranges. Some baselines were also ﬂagged based on closure errors on the bandpass
calibrator. Channel and time-based ﬂagging of data points corrupted by radio
frequency interference (RFI) was applied using a median ﬁlter with a 6σ threshold.
Residual errors above 5σ were also ﬂagged after a few rounds of imaging and self calibration.
The system temperature (Tsys) was found to vary with antenna, the ambient
temperature and elevation (Sirothia 2009). In the absence of regular Tsys measurements
for GMRT antennas, this correction was estimated from the residuals of corrected data
with respect to the model. The corrections were then applied to the data. The ﬁnal
images were made after several rounds of phase self calibration, and one round of amplitude
self calibration, where the data were normalized by the median gain for all the
data.
3.2.3 Chandra X-ray data
The Chandra observations of S´ersic 159-03 were taken in August 2009 using the Advanced
CCD Imaging Spectrometer (ACIS). The total net exposure time after cleaning
is 89.6 ks. We follow the data reduction procedure described in Million et al. (2010b)
and Million et al. (2010a). Background spectra were extracted from the appropriate recent
blank-sky ﬁelds available from the Chandra X-ray Center. These were normalized
by the ratio of the observed and blank-sky count rates in the 9.5–12 keV band.
Background subtracted images were created in 13 narrow energy bands, spanning
0.5–7.5 keV. These were ﬂat ﬁelded with respect to the median energy for each image.
The blank sky background ﬁelds were processed in an identical way to the source observation
and reprojected to the same coordinate system. The background subtracted
and exposure corrected images were co-added to create the broad band images.
For the spectral analysis, point sources were excluded. The individual regions for
the 2D spectral mapping were determined using the Contour Binning algorithm (Sanders
2006), which groups neighboring pixels of similar surface brightness until a desired
1The GMRT is a national facility operated by the National Centre for Radio Astrophysics of the TIFR,
India.
3.3 Results 37
signal-to-noise threshold is met. In order to have small enough regions to resolve substructure
and still have enough counts to achieve better than 7 per cent accuracy in
the temperature determination, we adopted a signal-to-noise ratio of 35 (∼1230 counts
per region). Separate photon-weighted response matrices and effective area ﬁles were
constructed for each region analyzed.
Spectral modeling has been performed with the SPEX package (SPEX uses an updated
version of the MEKAL plasma model with respect to XSPEC, Kaastra et al. 1996)
in the 0.6–7.0 keV band. To each bin we ﬁtted a model consisting of absorbed (Galactic
NH = 1.14 × 1020 cm−2, based on the LAB survey of H I by Kalberla et al. 2005)
collisionally ionized equilibrium plasmas with temperature, spectral normalization
(emission measure), and metallicity as free parameters. The new Galactic absorption
column density value for S´ersic 159-03 by Kalberla et al. (2005) is signiﬁcantly
lower than the older measurements by Dickey & Lockman (1990), which indicated
NH = 1.79 × 1020 cm−2. Using the new absorption values the X-ray spectra of the cluster
core do not show the excess soft X-ray emission reported by Bonamente et al. (2005)
and Werner et al. (2007).
3.2.4 XMM-Newton RGS data
The hot plasma in the core of S´ersic 159-03 exhibits a complex multi-temperature structure
(de Plaa et al. 2006). In order to determine the amount of low-temperature cooling
X-ray gas in the core and obtain upper limits on the mass deposition rate, we analyze
high spectral resolution XMM-Newton Reﬂection Grating Spectrometer (RGS) X-ray
data.
The XMM-Newton RGS data were obtained on November 20–21 2002, with a net exposure
time of 86.3 ks. The data were processed as described in de Plaa et al. (2006).
Spectra were extracted from a 4 wide extraction region, centred on the optical centre
of the galaxy. Because the RGS operates without a slit, it collects all photons from
within the 4 × ∼12 ﬁeld of view. Line photons originating at angle ∆θ (in arcminutes)
along the dispersion direction are shifted in wavelength by ∆λ = 0.138∆θ ˚A. Therefore,
every line is broadened by the spatial extent of the source. To account for this
spatial broadening in our spectral model, we produce a predicted line spread function
(LSF) by convolving the RGS response with the surface brightness proﬁle of the cluster
derived from the EPIC/MOS1 image along the dispersion direction. Because the radial
proﬁle of a particular spectral line can be different from the overall radial surface
brightness proﬁle, the line proﬁle is multiplied by a scale factor s, which is the ratio
of the observed LSF to the expected LSF. This scale factor is a free parameter in the
spectral ﬁt. We ﬁt the spectra in the 8–28 ˚A band.
38 AGN feedback in the cluster S´ersic 159-03
0 0.074 0.22 0.52 1.1 2.3 4.6 9.3 19 37 74
Northeast
nebula
filament
North-south
filament
Western
11.5 kpc
10 arcsec
Figure 3.1: Hα+[N II] image obtained with the 4.1 m SOAR telescope. The image shows
a large 44 kpc long north-south ﬁlament, a smaller western ﬁlament, and a separate
nebula at the northeast. The northern end of the large ﬁlament separates into two
parallel structures. The contours of the 8.4 GHz radio emission are over-plotted in
red and the beam size is indicated by the red ellipse in the bottom left corner. The
outermost radio contour is at 0.12 mJy beam−1 and the levels increase logarithmically
to the innermost contour which is at 4.0 mJy beam−1.
3.3 Results 39
0.01 0.013 0.02 0.034 0.062 0.12 0.23 0.44 0.88 1.8 3.5
59.6 59.4 59.2 23:13:59.0 58.8 58.6 58.4 58.2 58.0 57.8 57.6
30.032.034.036.038.0-42:43:40.042.044.046.048.0
4.8 kpc
4 arcsec
0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5
59.2 23:13:59.0 58.8 58.6 58.4 58.2 58.0
34.036.038.0-42:43:40.042.0
2.4 kpc
2 arcsec
0.037 0.13 0.22 0.31 0.41 0.5 0.59 0.68 0.77 0.87 0.96
11.5 kpc
10 arcsec
11.5 kpc
Figure 3.2: Left panel: HST image of the central cluster galaxy of S´ersic 159-03 with clear
dark dust bands. Middle panel: The HST image divided by the best ﬁt 2D elliptical de
Vaucouleurs model of the galaxy. The white “X” mark indicates the optical center of
the galaxy. A 4.5 kpc long dust lane displaced to the east and extending to the north
is clearly visible. The inner region of the galaxy is disturbed and not well described
by the de Vaucouleurs proﬁle. The green contours show the isophotes of the Hα+[N II]
ﬁlaments (see Fig. 3.1). Right panel: XMM-Newton Optical Monitor UVW2 (1800–
2400 ˚A) image of of the core of the galaxy with the Hα contours over-plotted. The UV
emission clearly extends along the brightest region of the Hα ﬁlament.
3.3 Results
3.3.1 Optical and UV properties
The cooling core of S´ersic 159-03 harbors a remarkable Hα+[N II] emission line ﬁlament
system shown in Fig. 3.1. This system of emission line nebulae consists of a large
44 kpc long north-south ﬁlament extending to the radius of 35 kpc in the north and
a smaller western structure extending out to 16 kpc. The total Hα+[N II] ﬂux is 5.9 ×
10−14 erg s−1 cm−2. Assuming [N II]
6583 ˚A
/Hα = 1 and [N II]
6548 ˚A
/[N II]
6583 ˚A
= 0.35,
the total Hα ﬂux is ∼2.5 × 10−14 erg s−1 cm−2. The total luminosity of the Hα line
emission (with the [N II] ﬂux excluded) is LHff = 2 × 1041 ergs s−1. Our measured
Hα ﬂux is higher than the ﬂux fHα = 1.79 ± 0.15 × 10−14 erg s−1 cm−2 measured by
McDonald et al. (2010), who used a ﬁlter with a 10 times narrower bandpass targeting
only the Hα emission lines. A separate nebula can be seen 25 kpc to the northeast of
the core.
Assuming a volume ﬁlling fraction of unity, a cylindrical geometry in the Hα bright
regions, optically thin ionized gas, isotropic radiation, and case B recombination at
104 K (Bland-Hawthorn & Maloney 1999), the total mass of the Hα emitting gas is
about 2.7 × 109 M . However, these optical ﬁlaments most likely consist of many thin
threads with very small volume ﬁlling fractions (e.g. Fabian et al. 2008). The inferred
mass is therefore likely to be signiﬁcantly overestimated.
The giant elliptical galaxy in the core of S´ersic 159-03 has its major axis aligned
in the northeast-southwest direction (left panel of Fig. 3.2). The HST data reveal a
40 AGN feedback in the cluster S´ersic 159-03
Table 3.2: Radio ﬂux densities. The ﬂux densities obtained by GMRT and VLA are from
this work, the values obtained at 408 MHz and 843 MHz by MOST are from the MRC
(Large et al. 1981) and SUMSS surveys (Bock et al. 1999; Mauch et al. 2003). The ﬂux
density at 4.75 GHz obtained by ATCA is from Hogan et al. in preparation.
Frequency Array Flux density
(MHz) (Jy)
244 GMRT 3.360 ± 0.20
325 GMRT 2.024 ± 0.20
408 MOST 1.390 ± 0.10
617 GMRT 0.786 ± 0.05
843 MOST 0.472 ± 0.014
1425 VLA-B 0.235 ± 0.020
4752 ATCA 0.0469 ± 0.007
8400 VLA-B 0.0313 ± 0.005
4.5 kpc long dust lane extending from the centre to the north, coincident with the bright
Hα emitting gas. It is displaced to the east from the core of the galaxy by about 1.2
(∼1.4 kpc). The dust lane is clearly visible in the middle panel of Fig. 3.2, where the
HST image is divided by the best ﬁt 2D elliptical de Vaucouleurs model. This ratio
image also uncovers a fainter dust lane to the south of the core. The over-subtraction of
the emission from center of the galaxy by the model shows that the surface brightness
of the stellar core is ﬂatter than the de Vaucouleurs proﬁle. We do not detect optical
emission from the AGN.
Near Ultra Violet images (NUV) from the XMM-Newton Optical Monitor show diffuse
emission which is co-spatial with the dust lanes and the emission line nebulae,
providing a strong indication for current star-formation (right panel of Fig. 3.2). After
correcting for coincidence, dead-time loss, and time sensitivity degradation, the total
UVW1 luminosity within a radius of 7 arcsec is 3.7×1042 ergs s−1. Based on Cardelli
et al. (1989), the Galactic extinction in the UVW1 band was assumed to be 0.07 mag and
the unknown internal extinction was set to zero. We measured the 2MASS J-band luminosity
within the same aperture and subtracted the contribution from the passive stellar
population from the empirical LUVW1 − LJ relation derived by Hicks & Mushotzky
(2005). We ﬁnd a net NUV luminosity excess of ∼2.3×1042 ergs s−1. Without correcting
for internal extinction, we obtain a UVW2-UVW1 color ∼0.4 mag.
3.3.2 Radio morphology
The contours of the 8.4 GHz radio emission in Fig. 3.1 and 3.3 resolve the source into an
S-shape, oriented primarily north-south, with a compact central feature of ﬂux density
6.3 mJy/beam, which is approximately the ﬁfth of the total radio ﬂux at this frequency.
The jets seem to start out in the northeast-southwest direction, before bending clock-
3.3 Results 41
11.5 kpc
5 arcsec
-3 -2.5 -2 -1.5 -1 -0.5
11.5 kpc
5 arcsec
1000 104
200 500 2000 5000
0.01
0.1
1
10
FluxDensity(Jy)
Frequency (MHz)
Figure 3.3: Left panel: Red+green color image, produced from the radio data at 617 MHz
(red) and 1.4 GHz (green), with the 8.4 GHz radio contours overplotted. The beam
sizes of the 8.4 GHz and 1.4 GHz radio emission are indicated by the white and green
ellipses in the bottom left corner. The outermost contour of the 8.4 GHz radio emission
is at 0.12 mJy beam−1 and the levels increase logarithmically to the innermost contour
which is at 4.0 mJy beam−1. Central panel: Spectral index map produced by combining
the 617 MHz GMRT and the 1.4 GHz VLA radio data. The radio maps show the innermost
0.5×0.78 arcmin2 region of the cluster. The radio maps at the two frequencies
have been matched in resolution. Right panel: The broad band radio spectrum between
244 MHz and 8.4 GHz. The radio spectrum between 244 MHz and 1.4 GHz can be
described by the overplotted power-law Sν ∝ ν−α with index α = 1.49.
42 AGN feedback in the cluster S´ersic 159-03
wise. The centers of the radio and optical emission are well aligned. At 1.4 GHz the
total ﬂux density is ∼230 mJy and shows a core with an extension ∼12 arcseconds to
the east (see the white contours in Fig. 3.4). In the 617 MHz GMRT radio map, the
morphology of the radio emission is nearly identical to 1.4 GHz. The extension to the
east is also detected at 617 MHz and seems to be divided into two ﬁlaments. The lower
frequency GMRT radio maps do not signiﬁcantly resolve the radio source.
In the left panel of Fig. 3.3 we show a red+green color image, produced from the
radio data at 617 MHz (red) and 1.4 GHz (green) with the contours of the 8.4 GHz
radio emission overplotted. This combined radio image clearly shows the extension
to the east. The central panel of Fig. 3.3 shows the spectral index map of the core of
S´ersic 159-03 produced by combining the 617 MHz GMRT and the 1.4 GHz VLA radio
data. Only data above 2 mJy/beam at 617 MHz and 0.2 mJy/beam at 1.4 GHz were
included. To the east, north-east and north-west of the core the radio emission has
a very steep spectrum described by a power-law Sν ∝ ν−α with index α > 2.2. The
spectrum of the central region is ﬂatter with index α ∼ 1.
In Table 3.2 we show the integrated radio ﬂux densities at ﬁve different frequencies
obtained with GMRT and VLA. We also list the published ﬂux densities obtained by
the MRC survey at 408 MHz (Large et al. 1981) and by the SUMSS survey at 843 MHz
(Bock et al. 1999; Mauch et al. 2003) performed with the Molonglo Observatory Synthesis
Telescope (MOST). The integrated ﬂux density at 4.75 GHz obtained by the Australia
Telescope Compact Array (ATCA) is from Hogan et al. in preparation. The broad band
radio spectrum is shown in the right panel of Fig. 3.3. Between 244 MHz and 1.4 GHz
the integrated spectrum can be described by a power-law Sν ∝ ν−α with α = 1.49. At
high frequencies, between 4.8 GHz and 8.4 GHz, however, the best ﬁt power-law is
signiﬁcantly ﬂatter with index α = 0.71.
3.3.3 X-ray imaging: a disturbed morphology
The X-ray data show no evidence for a point source associated with the central radio
source. Assuming a power-law like spectrum with a photon index Γ = 2.0, a conservative
upper limit on the X-ray ﬂux in the 2–10 keV band is fX = 2.3 × 10−15 erg
s−1 cm−2. The large scale X-ray emission from the cluster is elongated in the same
northeast-southwest direction as the major axis of the cD galaxy. The X-ray emission
does not show obvious sharp surface brightness discontinuities, indicative of cold
fronts due to sloshing gas.
While the large scale X-ray morphology of S´ersic 159-03 is relatively relaxed, its core
is strongly disturbed. The brightest, densest X-ray emitting gas is displaced northward
from the centre of the cD galaxy to a radius of r ∼ 8 kpc. In the top left panel of Fig. 3.4,
we show the smoothed Chandra image of the cluster core in the 0.5–7.5 keV band. The
black cross indicates the position of the central AGN. In the top right panel, we show
the same image, with the contours of the Hα+[N II] emission over-plotted in black, and
the contours of the 8.4 GHz and 1.4 GHz radio emission over-plotted in blue and white,
respectively. The lower panels of Fig. 3.4 show the same Chandra images divided by
their best ﬁt 2D elliptical double beta model. The images shows a bright ridge of dense
thermal X-ray gas displaced by about 8 kpc to the north of the AGN and a clumpy
3.3 Results 43
0 0.062 0.19 0.44 0.93 1.9 3.9 7.8 16 31 63
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
Western filament
cavity
Southern
elongated cavity
Eastern
of cooling gas
Northern ridge
northern filament
Clumpy
11.5 kpc
10 arcsec
11.5 kpc
10 arcsec
0 0.062 0.19 0.44 0.93 1.9 3.9 7.8 16 31 63
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
11.5 kpc
10 arcsec
11.5 kpc
10 arcsec
0.49 0.62 0.74 0.87 1 1.1 1.3 1.4 1.5 1.6 1.8
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
Western filament
cavity
Southern
elongated cavity
Eastern
of cooling gas
Northern ridge
northern filament
Clumpy
11.5 kpc
10 arcsec
11.5 kpc
10 arcsec
0.49 0.62 0.74 0.87 1 1.1 1.3 1.4 1.5 1.6 1.8
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
11.5 kpc
10 arcsec
Figure 3.4: Top left panel: Background subtracted, ﬂat-ﬁelded Chandra X-ray image of
the core of S´ersic 159-03 in the 0.5–7.5 keV band. The cross indicates the position of the
central radio source. Top right panel: The same Chandra image with over-plotted contours
of the 8.4 GHz (blue) and 1.4 GHz (white) radio emission, and with the contours
of the Hα+[N II] line emission (black). The beam size of the 1.4 GHz radio emission is
indicated by the white ellipse in the bottom left corner. The outermost radio contour
is at 0.3 mJy beam−1 and the levels increase logarithmically to the innermost contour
which is at 137.7 mJy beam−1. The contours of the 8.4 GHz radio emission are the same
as in Fig. 3.1 and 3.3. The dense cooling gas is displaced to the north of the centre of
the galaxy and the region within r 7.5 kpc is void of dense gas. Lower left panel: The
Chandra image divided by the best ﬁt 2D elliptical double beta model. Note the clear
structure to the north. Lower right panel: The same ratio image with the contours of the
radio and Hα+[N II] emission over-plotted. The images have been smoothed with a
Gaussian kernel.
44 AGN feedback in the cluster S´ersic 159-03
0.0016 0.0022 0.0027 0.0033 0.0038 0.0044 0.0049 0.0055 0.006 0.0066
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
11.5 kpc
10 arcsecdensity
1.8 2 2.2 2.4 2.5 2.7 2.9 3.1 3.2 3.4 3.6
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
temperature
11.5 kpc
10 arcsec
49 68 87 105 124 143 162 180 199 218 236
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
entropy
11.5 kpc
10 arcsec
0.0050 0.0060 0.0069 0.0078 0.0087 0.0097 0.0106 0.0115 0.0124 0.0134
02.0 01.0 23:14:00.0 59.0 58.0 57.0 56.0 13:55.0
43:00.010.020.030.040.050.0-42:44:00.010.0
pressure
11.5 kpc
10 arcsec
Figure 3.5: 2D maps of density (in units of cm−3 × l
2Mpc
−1/2
; top left panel), temperature
(in units of keV; top right panel), pressure (in units of keV cm−3 × l
2Mpc
−1/2
;
lower right panel), and entropy (in units of keV cm2 × l
2Mpc
1/3
; lower left panel).
The maps were obtained by ﬁtting each S/N∼35 region independently with a single
temperature thermal model, yielding 1σ fractional uncertainties of ∼7–10 per cent on
the temperature. The contours of the Hα+[N II] optical line emission are over-plotted
on the temperature and entropy maps. The contours of the 8.4 GHz and 1.4 GHz radio
emission are over-plotted on the density and pressure map, respectively.
3.3 Results 45
X-ray ﬁlament extending along the Hα ﬁlament to 31 kpc beyond this ridge. Another
X-ray ﬁlament extends to the west and coincides with the western optical emission line
nebula.
The ridge of dense, thermal X-ray emitting gas to the north of the AGN seems to
be interacting with, and conﬁning, the 8.4 GHz radio emitting plasma (blue contours
in Fig. 3.4). The jets appear distorted by the interaction with the dense cooling gas.
The 1.4 GHz radio plasma also appears deﬂected by the ridge of dense gas to the east,
where it ﬁlls a gap - an elongated cavity - in the X-ray surface brightness distribution.
This elongated cavity is possibly composed of two cavities. The X-ray surface brightness
drops sharply to the southeast of the AGN, forming an apparent cavity with a
radius of ∼11 kpc, ﬁlled by 1.4 GHz radio emission. This drop in surface brightness
is spatially coincident with the sharp southeastern edge of the bright emission line
nebulae.
3.3.4 Thermodynamic properties of the core
The disturbed morphology of the cluster core is also reﬂected in the 2D maps of density,
temperature, entropy, and pressure shown in Fig. 3.5. The ICM spatially associated
with the large north-south Hα ﬁlament has a relatively low projected temperature (top
right panel of Fig. 3.5). The high density (top left panel of Fig. 3.5) and low temperature
of this feature translate into low entropy (lower left panel of Fig. 3.5). The western
ﬁlament, on the other hand, has a signiﬁcantly higher temperature and entropy.
The projected thermal pressure peaks in an approximate ring surrounding the core
at a radius of r ∼ 15 kpc (lower right panel of Fig. 3.5). The pressure in this ring
consisting of over ﬁfteen independently ﬁtted adjacent regions is approximately 20 per
cent higher than in the center, indicating an AGN driven weak shock. The maps reveal
that the relatively over-pressured gas to the southeast of the core has approximately
1.2 times higher temperature than the surrounding gas, which is consistent with the
hotter gas having been compressed and heated by an AGN driven shock.
The metallicity of the ICM along the Hα+[N II] ﬁlaments is higher by ∼0.3 Solar
than that of the surrounding plasma (see Fig. 3.6). The metallicity increase is not signiﬁcant
in any single region, the error on the metallicity in every region is ∼0.20 Solar.
However, the fact that the increase is seen in 5 independently ﬁtted adjacent regions
along the ﬁlament makes this result interesting. The apparent metallicities of the bright
northern ridge of cooling plasma and of the core are, however, low. It has been shown
that the metallicity is sensitive to the modeling of the underlying temperature structure
and, if multi-temperature plasma is modeled with a single-temperature spectral
model, the metallicity of gas can be signiﬁcantly underestimated (e.g. Buote 2000a).
The low metallicities of these features thus indicate the presence of multi-phase gas.
3.3.5 High resolution X-ray spectra
In order to determine whether cooling X-ray gas with kT < 1 keV is present in the core
of the cluster, we examine the XMM-Newton RGS spectra (see Fig. 3.7) to search for low
46 AGN feedback in the cluster S´ersic 159-03
0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2
Figure 3.6: Metallicity map of the core of the cluster with the contours of the Hα+[N II]
optical line emission over-plotted. The metallicity of the ICM along the ﬁlament is
higher than that of the surrounding medium. The apparent metallicity of the bright
‘northern ridge’ and of the core is low, indicating the presence of multi-phase gas.
10 15 20
0.010.020.030.040.05
Counts/s/Å
wavelength (Å)
FeXVII
FeXVII
FeXVII+
OVIII
Figure 3.7: The ﬁrst order XMM-Newton RGS spectrum extracted from a 4 wide region
centred on the core of S´ersic 159-03. The continuous line represents the best ﬁt model
to the spectrum. Fe XVII lines emitted by plasma with kT < 0.9 keV are clearly visible
in the spectrum.
3.3 Results 47
Table 3.3: Best ﬁt parameters for a four-temperature ﬁt and a CIE+two-cooling-ﬂow
model ﬁt to the high-resolution RGS spectra extracted from a 4 wide region centred on
the core of S´ersic 159-03. Emission measures, Y = nHnedV, are given in 1066 cm−3.
Radiative cooling rates are given in M yr−1. The scale factor s is the ratio of the
observed LSF to the expected LSF based on the overall radial surface brightness proﬁle.
The upper limits are quoted at their 95 per cent conﬁdence level. Abundances are
quoted with respect to the values of Grevesse & Sauval (1998).
Parameter 4T-model CIE+c.f.+c.f. model
Y0.25keV 0.05 ± 0.04 –
Y0.75keV 0.19 ± 0.03 –
Y1.5keV 0.7 ± 0.3 –
Y3.0keV 12.1 ± 0.3 –
YCIE – 12.3 ± 0.2
˙M1.9−0.5keV – 82 ± 11
˙M0.5−0.1keV – < 25
kT (keV) – 3.05 ± 0.16
s 1.12 ± 0.13 1.13 ± 0.11
O 0.39 ± 0.04 0.43 ± 0.04
Ne 0.43 ± 0.12 0.44 ± 0.11
Fe 0.78 ± 0.04 0.82 ± 0.06
48 AGN feedback in the cluster S´ersic 159-03
temperature line emission (primarily Fe XVII) tracing rapidly cooling gas. We ﬁt the
RGS data with a model consisting of collisionally ionized equilibrium plasmas at four
ﬁxed temperatures (0.25 keV, 0.75 keV, 1.5 keV, and 3 keV) with variable normalizations
and common metal abundances. The abundances of O, Ne, Mg, and Fe are free parameters
in the ﬁt. Our ﬁt conﬁrms the presence of relatively cool, kT < 1 keV plasma.
Based on the lack of O VII line emission however, which is emitted at kT < 0.4 keV,
we place a strong upper limit on the amount of gas cooling radiatively to very low
temperatures. Our 90 per cent conﬁdence upper limit on the emission measure of gas
with kT = 0.25 keV is Y = nHnedV = 9 × 1064 cm−3, which is about 50 per cent of
the best ﬁt emission measure for the 0.75 keV gas.
In order to place constraints on the amount of cooling in the cluster core we ﬁt
the RGS spectra with a separate model consisting of thermal plasma in collisional
ionization equilibrium and two isobaric cooling ﬂow models. The ﬁrst cooling ﬂow
model is used to account for gas with temperatures between kTupper = 1.9 keV and
kTLower = 0.5 keV, which appears to be the ‘temperature ﬂoor’ in several cooling core
clusters (Sanders et al. 2009, 2010a; Werner et al. 2010). The second cooling ﬂow model
is used to place an upper limit on the radiative cooling rate from 0.5 keV down to cold
gas. While in the temperature range of 1.9–0.5 keV the data are formally consistent
with a radiative cooling rate of ˙M = 82 ± 11 M yr−1, the 95 per cent conﬁdence upper
limit for radiative cooling from 0.5 keV to lower temperatures is only 25 M yr−1.
3.4 Discussion
3.4.1 Displacement of gas from the cD galaxy
The core of S´ersic 159-03 has a remarkably complex and rich morphology at all wavelengths.
It displays signs of a powerful AGN feedback. The central regions of the
galaxy (r < 7.5 kpc) are cleared of the densest, X-ray emitting ICM and the cluster core
displays a massive, bright Hα ﬁlament extending northward from the centre of the cD
galaxy to a radius of 35 kpc. This long ﬁlament is reminiscent of that in Abell 1795
(Fabian et al. 2001; Crawford et al. 2005; McDonald & Veilleux 2009).
While the densest cooling X-ray plasma has been pushed away and uplifted from
the cooling core by the radio jets to a radius of at least r ∼ 8 kpc, the brightest core of
the Hα emission is not displaced from the galactic nucleus. The most likely reason is
that the Hα emitting gas is only a thin layer on an underlying large reservoir of dense
atomic and molecular gas at the base of the galaxy potential, which is difﬁcult to uplift
entirely. Observations of K-band emission lines of molecular and ionized hydrogen
with the SINFONI integral ﬁeld spectrograph reveal extended ﬁlaments in the core of
S´ersic 159-03, tracing closely each other and the Hα emission (Oonk et al. 2010). The
line-of-sight velocities, as measured by Oonk et al. (2010), show a striking east/west
dichotomy, suggesting that part of the velocity vector of the jets, which push and uplift
the gas, is oriented along our line of sight. The gas to the east of the stellar core has
a radial velocity of ∼200 km s−1 with respect to the cD galaxy, which seems to be
decreasing with the distance from the nucleus. The observed interaction between the
3.4 Discussion 49
radio jets and the Hα emitting gas (see Fig. 3.1) indicates that the gas is being pushed
out by the radio jets and is not falling in along our line of sight.
The radial velocity distribution of the gas along the western ﬁlament goes smoothly
from about -200 km s−1 to +50 km s−1 with increasing distance from the core, indicating
circular motion, which is consistent with the uplift scenario, but not with an
alternative scenario of a simple infall from the surrounding hot X-ray halo.
The morphology of the cooling ﬁlamentary gas indicates relative gas motions with
the ambient ICM moving toward the northwest, dragging and bending the Hα ﬁlament.
The gas uplifted by the jet in the southwestern direction has been dragged
to the west and the gas uplifted towards northeast has been displaced in the northnorthwestern
direction. These relative gas motions are most likely the result of the
south-southeastward peculiar motion of the cD galaxy.
Relatively cool X-ray gas is present along the whole northern ﬁlament of S´ersic 159-
03, but the correlation between the Hα and soft X-ray emission is not perfect. We
see clumps of dense, cooling X-ray emitting plasma at ∼8 kpc and at ∼17 kpc, but
no obvious Hα peaks at the same locations. The ‘northern ridge’ at r ∼ 8 kpc has
the coldest projected temperature and lowest entropy. The X-ray ﬁlament extends to
the north by about 3 kpc further than the observed Hα ﬁlament. The relatively low
temperature and entropy, and high metallicity of the ﬁlament compared to the ambient
plasma (see Fig. 3.5 and 3.6) indicate that the ﬁlamentary system has been uplifted and
stripped from the cD galaxy. The western Hα+[N II] ﬁlament is also surrounded by
bright, dense X-ray emitting gas, but its projected temperature is ∼0.8 keV higher. The
western ﬁlament might have been shock heated or its gas might have gotten mixed
with the hotter ambient plasma.
Gas uplift by buoyant radio-emitting plasma, supplied by the jets of the AGN, has
also been observed in other cooling core clusters. Detailed spectroscopic mapping of
the cooling core of the Virgo Cluster, centered on M87, shows clearly that radio mode
AGN feedback is highly efﬁcient in stripping the core of the galaxy of its lowest entropy
gas (Werner et al. 2010). The total mass of uplifted gas in M87 is 6–9×108 M , which is
similar to the current gas mass within its innermost r ∼ 3.8 kpc region. The disruption
of the core of S´ersic 159-03 is considerably larger than that of M87, but by no means
extreme. The AGN feedback in the Hydra A cluster is responsible for the uplift of a few
times 109 M of low entropy plasma (Simionescu et al. 2009a). Recently, Ehlert et al.
(2011a) presented a multi-wavelength study of the cluster MACS J1931.8-2634 where
extreme AGN feedback with a jet power of Pjet ∼ 1046 erg s−1 and the sloshing gas
disrupted the core, separating it into two X-ray bright ridges, which are currently at a
distance of ∼25 kpc from the core.
3.4.2 Cooling of the displaced gas and star-formation
To the north of the AGN, we see cooling X-ray plasma displaced by about 8 kpc from
the nucleus, producing a prominent bright ridge which conﬁnes the high frequency
radio plasma. The northern X-ray ﬁlament extends 31 kpc beyond this ridge. The relatively
dense uplifted X-ray emitting gas, which is removed from the direct inﬂuence of
the AGN jets, will cool in the absence of heating and eventually form stars. Multiphase
50 AGN feedback in the cluster S´ersic 159-03
cooling X-ray gas, displaced from the center of the cD galaxy and spatially coincident
with Hα emission, has also been seen in the Ophiuchus Cluster, in Abell 2052, and in
MACS J1931.8-2634 (Million et al. 2010a; Edwards et al. 2009; de Plaa et al. 2010; Ehlert
et al. 2011a).
Assuming that the ﬁlament has been uplifted at half of the sound speed (at 400 km s−1),
the age of the ﬁlament will be ∼ 108 yr. That is approximately equal to the cooling time
of 1 keV plasma in pressure equilibrium with the ambient ICM at the observed location.
The XMM-Newton RGS spectra clearly reveal Fe XVII line emission associated
with gas cooling to kT < 1 keV. The 95 per cent conﬁdence upper limit on radiative
cooling below 0.5 keV is 25 M yr−1.
The densest and coolest X-ray emitting clumps, in particular the bright multiphase
‘northern ridge’, will cool and form narrow Hα emitting ﬁlaments (e.g. see simulations
by Sharma et al. 2010), along which some of the gas might eventually fall back towards
the cluster centre. It is possible that a signiﬁcant fraction of the Hα emitting gas in
the northern ﬁlament at large radii is due to the cooling of the uplifted X-ray emitting
plasma. The cold gas is likely to go on to eventually form stars. The XMM-Newton
OM UVW2 and Galex images (McDonald et al. 2010) indicate that the UV emission is
extended along the brightest regions of the northern Hα ﬁlament, indicating ongoing
star formation.
The HST images show that the brightest regions of the ﬁlament are dusty. An infrared
survey with the Spitzer Space Telescope, however, did not detect the system at
70µm (Quillen et al. 2008) indicating that the ﬁlaments do not contain an exceptionally
large amount of warm dust. The excess UV emission from the cD galaxy corresponds
to a star-formation rate of ∼2.3 M yr−1 for the assumed Salpeter IMF. Assuming
the Kennicutt (1998) relation, SFR (M yr−1) = 7.9 × 10−42 (LHα/erg s−1), and neglecting
intrinsic extinction due to dust, the measured Hα luminosity corresponds to a
star-formation rate of 1.5 M yr−1. Accounting for internal extinction would further
increase the intrinsic Hα and NUV ﬂuxes. Star-formation, however, is most likely not
the only heating and ionizing source in the cluster center.
The large turbulent velocities and the elevated [N II]/Hα ratio of ∼1.5 in the nucleus
(Crawford & Fabian 1992) indicates that the interaction with the radio jets contributes
strongly to the heating of the gas in the center. The [N II]/Hα line ratios of 0.6 in the
western ﬁlament are relatively high as well (Crawford & Fabian 1992), indicating the
presence of an additional non-ionizing source of energy at larger radii (see e.g. Ferland
et al. 2009).
Based on the data for M87, Werner et al. (2010) proposed that the Hα ﬁlaments in
its core may be powered by shock induced mixing of cold gas with the surrounding
ICM (see Begelman & Fabian 1990). By bringing the hot thermal particles into contact
with the cool gas, mixing can supply the power and ionizing particles to explain
the observed spectra. Hot ICM electrons that penetrate into the cold gas excite the
molecular hydrogen and deposit heat. This scenario has been explored theoretically
by Ferland et al. (2008, 2009), who studied heating by cosmic-rays, which affect the
cooler ionized and neutral components in a similar way to hot ICM electrons. The fact
that the soft X-ray emission traces the Hα emitting gas suggests that this process could
be responsible for part of the Hα+[N II] line emission, part of the UV emission, and for
3.4 Discussion 51
the non-radiative cooling of the coldest X-ray gas in S´ersic 159-03.
The most similar known X-ray/Hα ﬁlament system is observed in Abell 1795. In
that cluster the ﬁlament extends for 50 kpc. The cD galaxy at the head of this ﬁlament
is moving with respect to the ICM and the emission line nebula may originate from a
runaway cooling of the hot X-ray emitting gas in the wake of that motion (Fabian et al.
2001; Markevitch et al. 2001; McDonald & Veilleux 2009). The ﬁlamentary system in
S´ersic 159-03, however, is most likely the result of AGN driven uplift and ram pressure
stripping from the cD galaxy. The ﬁlament in Abell 1795 is composed of a pair of thin
w < 1 kpc intertwined Hα+[N II] ﬁlaments (McDonald & Veilleux 2009), spatially coincident
with relatively cool X-ray emitting gas (Fabian et al. 2001), and with chains of
FUV-bright young star clusters condensing from the cooling gas in the ﬁlament (Crawford
et al. 2005; McDonald & Veilleux 2009). The northern end of the large Hα ﬁlament
in S´ersic 159-03 separates into two narrow, parallel structures. Given a better spatial
resolution, we would most likely resolve the ﬁlament into more thin threads. As discussed
by Fabian et al. (2008) for the emission line nebulae in the core of the Perseus
Cluster, the thin thread-like ﬁlamentary structures are most likely stabilized by magnetic
ﬁelds. These magnetic ﬁelds might be possible to detect using Faraday Rotation
against the polarized jet emission (as has been done for the ﬁlaments in the Centaurus
Cluster by Taylor et al. 2007).
3.4.3 The powerful radio mode AGN
The core of S´ersic 159-03 harbors a powerful radio mode AGN. Inﬂating the southern
cavity, which has a radius of ∼11 kpc and is approximately spherical, in the surrounding
medium with a thermal pressure of p = 0.065 keV cm−3, required a 4pV work
of about 7 × 1058 ergs, indicating powerful jets. However, no optical or X-ray point
source is presently seen at the position of the radio bright AGN. Using the ‘Black hole
fundamental plane’ relation of Merloni et al. (2003), for a 10 mJy core ﬂux at 4.75 GHz
and for a black hole mass of 6 × 108 M (determined from the K-band bulge luminosity
by Rafferty et al. 2006), the expected X-ray core ﬂux is 2.4 × 10−13 erg s−1 cm−2.
This value is two orders of magnitude higher than our conservative upper limit of
2.3 × 10−15 erg s−1 cm−2. The observed scatter around the ‘Black hole fundamental
plane’ is, however, large. The radio luminosities of some other well known brightest
cluster galaxies (e.g. NGC 1275) also show similar offsets with respect to the expected
relation.
The radio morphology is similar to 4C26.42 in Abell 1795 (Liuzzo et al. 2009) and
it is also reminiscent of PKS 1246-410 in the Centaurus Cluster (Taylor et al. 2007).
The distortion of the jets, and the change in the axis from N-S to E-W, is most likely
due to the strong interactions with the dense gas. The velocity dispersion of the NIR
emission lines, which sharply increases in the nuclear region (Oonk et al. 2010) shows
that the interaction of the jets with the cold gas is the strongest within the innermost
2 kpc region. The presence of cold, high density material in the cluster core is likely to
decelerate the jets, which might entrain thermal gas, slow down to subsonic velocities,
and continue to rise buoyantly. Such deceleration due to strong interaction with dense
gas on small scales has been seen using VLBA observations in Hydra A (Taylor 1996)
52 AGN feedback in the cluster S´ersic 159-03
and Abell 1795 (Liuzzo et al. 2009). While the 1.4 GHz and 617 MHz radio plasmas
appear to be deﬂected by the dense ‘northern ridge of cooling plasma’ to the east,
forming the ‘eastern elongated X-ray dark cavity’, the younger radio jet seen at 8.4 GHz
is being deﬂected to the northwest where it is likely to continue to buoyantly rise along
the short axis of the cluster. The 8.4 GHz southern jet is being deﬂected by the Hα
emitting gas to the southeast where it is partly ﬁlling the ‘southern X-ray cavity’.
Between 244 MHz and 1.4 GHz the integrated radio emission in the cluster core has
a remarkably steep power-law spectrum Sν ∝ ν−α with index α = 1.49. At higher
frequencies, however, the radio spectrum ﬂattens to index α = 0.71. At the 4.75 GHz
ATCA radio map (Hogan et al. in prep.), the radio morphology is very similar to that
at 8.4 GHz, indicating that the ﬂatter part of the integrated radio emission originates in
the ∼10 kpc scale inner lobes (see the contours in the left panel of Fig. 3.3) where active
jets are injecting relativistic particles. At the lower frequencies the radio emission is
more extended, indicating an older population of electrons. The spectral index map
produced from the 617 MHz and 1.4 GHz radio data (central panel of Fig. 3.3) shows
that while in the central regions α ∼ 1, to the east, where the radio plasma seems to
be ﬁlling the elongated cavity, the spectral index steepens to α > 2. The spectral index
also steepens in the northwest, where the plasma ﬁlls a region with a relative deﬁcit in
X-ray surface brightness. This radio plasma is most likely buoyant and exerting work
on the surrounding medium.
The steep integrated spectrum of the radio source is similar to radio mini-halos
found within the cooling radius of some clusters (e.g Ferrari et al. 2008). The morphology
of the extended radio emission, however, indicates that it is due to the plasma
from the radio jets which have been decelerated, deﬂected and conﬁned by the interaction
with the dense gas. A steep radio spectral index has been found in other cooling
core clusters (e.g., PKS 0745-191, A2029, A4059, A2597 Taylor et al. 1994; Pollack et al.
2005) and taken as an indicator of conﬁnement.
Our thermodynamic maps show that the projected thermal pressure peaks in an
approximate ring surrounding the AGN. The pressure in the ring is approximately 20
per cent higher than in the center, indicating the presence of an AGN driven shock
front. The high pressure region to the south of the AGN is hotter in multiple independently
ﬁtted adjacent regions than the plasma both outside and inside this feature. The
observed projected temperature jump of ∆T ∼ 1.2 suggests a shock with a Mach number
of M ∼ 1.5. The complex X-ray morphology with a cavity inside the shock front
unfortunately prevents a more precise modeling of the shock. The relatively modest,
factor of 1.5, temperature drop in the core of this cluster (Sun 2009) also points towards
a relatively strong AGN feedback activity. AGN induced shocks, which may be the
most signiﬁcant channel for heating of the ICM near to the AGN, have also been observed
around M87 in the center of the Virgo Cluster and in Hydra A, (Forman et al.
2005; Million et al. 2010a; Nulsen et al. 2005; Simionescu et al. 2009b).
3.5 Conclusions 53
3.5 Conclusions
We performed a multi-wavelength study of the energetic interaction between the central
AGN, the ICM, and the optical emission line nebula in the galaxy cluster S´ersic
159-03. We conclude that:
• Powerful ‘radio mode’ AGN feedback and possible ram pressure cleared the central
region of the cD galaxy (r < 7.5 kpc) of the cooling low entropy X-ray gas.
• This low entropy, high metallicity, relatively cool X-ray gas lies along the bright,
44 kpc long Hα+[N II] ﬁlament extending from the centre of the cD galaxy to the
north.
• As indicated by the observed dust lanes, molecular and ionized emission line
nebulae, and the excess UV emission, part of this displaced gas, which is removed
from the direct inﬂuence of the AGN, cools and forms stars.
• The pressure map shows evidence for an AGN induced weak shock at a radius
of r = 15 kpc and the X-ray images reveal cavities indicating past powerful AGN
activity with an energy of ∼ 7 × 1058 ergs. At low frequencies, the radio source
has an unusually steep spectrum with α = 1.5 indicating aging and conﬁnement
by the ambient gas.
Acknowledgments
We thank Alastair Edge and Michael Hogan for providing the reduced 4.75 GHz ATCA
radio data and for helpful discussions. We thank Seth Bruch and Emily Wang for their
help with the SOAR observations. Support for this work was provided by the National
Aeronautics and Space Administration through Chandra/Einstein Postdoctoral
Fellowship Award Number PF8-90056 and PF9-00070 issued by the Chandra X-ray
Observatory Center, which is operated by the Smithsonian Astrophysical Observatory
for and on behalf of the National Aeronautics and Space Administration under
contract NAS8-03060, and by the Chandra grants GO0-11019X and GO0-11139X. This
work was supported in part by the U.S. Department of Energy under contract number
DE-AC02-76SF00515. The Very Large Array is part of the National Radio Astronomy
Observatory. The National Radio Astronomy Observatory is a facility of the National
Science Foundation operated under a cooperative agreement by Associated Universities,
Inc. We thank the staff of the GMRT that made these observations possible. The
SOAR Telescope is a joint project of Conselho Nacional des Pesquisas Cientﬁcas e Tecnologicas
CNPq-Brazil, The University of North Carolina Chapel Hill, Michigan State
University, and the National Optical Astronomy Observatory.
54 AGN feedback in the cluster S´ersic 159-03
Chapter 4
On the thermodynamic self-similarity
of the nearest, most relaxed, giant
ellipticals
N. Werner1, S. W. Allen1, A. Simionescu1
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall, Stanford, CA 94305-4085, USA
and SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA
94025, USA
Published in the Monthly Notices of the Royal Astronomical Society, volume 425,
pages 2731–2740, 2012
Abstract
We present detailed spatially resolved measurements of the thermodynamic properties
of the X-ray emitting gas in the inner regions of the ﬁve nearest, X-ray and optically
brightest, and most X-ray morphologically relaxed giant elliptical galaxies known. Beyond
the innermost region at r 1 kpc, and out to r ∼ 6 kpc, the density, pressure,
entropy, and cooling time distributions for the X-ray emitting gas follow remarkably
similar, simple, power-law like distributions. Notably, the entropy proﬁles follow a
form K ∝ rα, with an index α = 0.92-1.07. The cumulative hot X-ray emitting gas
mass proﬁles and the gas-mass to stellar-light ratios of all ﬁve galaxies are also similar.
Overall the observed similarity of the thermodynamic proﬁles in this radial range argues
that, in these systems, relativistic jets heat the gas at a similar rate averaged over
time scales longer than the cooling time tcool 108 yr. These jets are powered by accretion
from the hot gas, or material entrained within it, onto the central super-massive
black hole. This jet heating creates an energy balance where heating and cooling are in
equilibrium, keeping the hot galactic atmospheres in a ‘steady-state’. Within r 1 kpc,
56 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
Table 4.1: Summary of the Chandra observations. Columns list the distances of the
galaxies, the angular scales per kpc at the corresponding distances, the B-band optical
and bolometric X-ray luminosities (O’Sullivan et al. 2001), radio luminosities at 1.4
GHz (Condon et al. 1998, 2002), the line-of-sight Galactic absorbing hydrogen column
densities, NH, in their directions (Kalberla et al. 2005), observation identiﬁers, and exposure
times after cleaning.
Galaxy Distance Scale LB LX LR NH Obs. ID Exposure
(Mpc) (arcsec kpc−1) (1010L ) (1041 erg s−1) (1038 erg s−1) (cm−2) (ks)
NGC4472 (M49) 16.7 12.4 8.7 2.96 1.20 1.53 321 19.1
5.8 11274 39.7
NGC4649 (M60) 16.5 12.5 2.05 0.13 2.04 8182 49.5
8507 17.5
NGC1399 20.9 9.9 4.4 5.68 1.52 1.50 319 49.9
4172 36.7
9530 59.3
NGC1407 28.8† 7.2 7.4 1.95 1.23 5.42 791 32.5
NGC4261 31.6† 6.5 5.1 1.63 329.52 1.75 9569 100.9
†Tonry et al. (2001)
Blakeslee et al. (2009)
this similarity breaks down: the observed entropy proﬁles show well resolved ﬂattening
and the values differ from system to system substantially. The accretion rate onto
the black hole and the AGN activity, heating the interstellar medium, must therefore
vary signiﬁcantly on time scales shorter than tcool = 107–108 yr.
4.1 Introduction
X-ray studies with Chandra and XMM-Newton have shown that relativistic jets, produced
by accreting supermassive black holes (SMBH) at the centres of galaxies, groups
and clusters, interact strongly with their environments, driving shocks (e.g. Forman
et al. 2005, 2007; Nulsen et al. 2005; Simionescu et al. 2009b; Million et al. 2010d) and
inﬂating bubbles of relativistic plasma in the surrounding X-ray emitting gas (e.g. Churazov
et al. 2000; Fabian et al. 2003a, 2006; Bˆırzan et al. 2004; Dunn et al. 2005; Dunn &
Fabian 2006, 2008; Forman et al. 2005, 2007; Rafferty et al. 2006; McNamara & Nulsen
2007). These bubbles rise buoyantly and can entrain and uplift large quantities of low
entropy gas from the innermost regions of their host galaxies (Simionescu et al. 2008,
2009a; Werner et al. 2010, 2011). All of this activity is believed to take place in a tight
feedback loop, where the hot interstellar medium (ISM), which is observable in the
X-ray band, cools and accretes onto the central SMBH (referred to as active galactic
nucleus or AGN), leading to the formation of jets which heat the surrounding gas and
drive the above phenomena (for a review see McNamara & Nulsen 2007; Gitti et al.
2012). This heating acts to lower the accretion rate, reducing the feedback, until the
accretion eventually builds up again. AGN feedback appears to play an important role
in regulating the ‘cooling cores’ of galaxy clusters and the formation and evolution of
massive galaxies (e.g. Peterson & Fabian 2006; Croton et al. 2006; De Lucia & Blaizot
4.2 Sample selection, data reduction, and analysis 57
2007; Sijacki et al. 2007). Although much has been learned about AGN feedback, many
questions regarding the remarkable balance between heating and cooling remain.
Nearby giant ellipticals are in many respects low redshift proxies for more distant
and more luminous cluster cooling cores, allowing us to study AGN feedback ‘in
closeup’. Another key advantage is that the binding energy per particle is lower in
galaxies: a given amount of AGN heating therefore affects the gas in galaxies more
obviously than in clusters. By observing a sample of nearby, X-ray bright galaxies with
a high degree of completeness, and with data of sufﬁcient quality to map the detailed
thermodynamic properties, we can hope to gain important insight into the physics of
AGN feedback, its duty cycle, and its role in galaxy formation.
One of the most useful thermodynamic quantities in studying the impact of cooling
and feedback in galaxies, groups, and clusters of galaxies is the gas entropy which,
by rewriting the adiabatic index in terms of X-ray observables, can be deﬁned as K =
kTn−2/3
e , where kT is the temperature and ne is the electron density of the X-ray emitting
gas. Because entropy can only be changed by gains and losses of energy, it traces
the thermal history of the gas. The radial proﬁles of entropy in statistical samples
of galaxy clusters and groups have been analyzed by Pratt et al. (2006, 2010), Cavagnolo
et al. (2009) and Sun et al. (2009). However, a detailed study of the thermodynamic
properties of giant ellipticals, reaching down to sub-kpc scales, has not been
performed.
Here we present detailed, spatially resolved measurements of the key thermodynamic
properties (density, temperature, pressure, entropy, cooling time, and gas mass)
of the X-ray emitting gas in the innermost regions of the ﬁve nearest, X-ray and optically
brightest, and most morphologically relaxed giant elliptical galaxies known.
All of these galaxies have deep Chandra observations, as well as exquisite optical and
multi-wavelength data.
In Sect. 9.1.1, we describe our sample, the reduction of the X-ray data, and the spectral
analysis. In Sect. 9.3, we present the measurements of both the 2D distributions
and the deprojected radial proﬁles of the key thermodynamic properties. We show, for
the ﬁrst time, that the azimuthally averaged gas density, pressure, cooling time, and
entropy proﬁles of relaxed giant elliptical galaxies are in the radial range 1 r 6 kpc
remarkably similar. Based on these results, in Sect. 9.4 we present some new insights
into the likely thermal history of the X-ray emitting atmospheres of the galaxies. In
Sect. 9.5, we summarize our conclusions.
4.2 Sample selection, data reduction, and analysis
4.2.1 Sample selection and its properties
Our target list is drawn from the parent sample of Dunn et al. (2010), who identiﬁed
the optically-brightest (BT ≤ 13.5 mag) and X-ray brightest (FX > 10−12 erg cm−2 s−1
in the 0.1–2.4 keV band) giant elliptical galaxies within a distance d ≤ 100 Mpc and
with declination dec ≥ −45 degrees. Motivated by a goal to resolve the innermost
thermodynamic structure in detail, from this list we select only those systems with
58 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
X-ray 2 kpc
NGC 4472
Near-infrared 2 kpc
NGC 4472 NGC 4472
Radio 2 kpc
6.4 6.5 6.6 6.8 7.2 8.1 9.8 13.2 20.0 33.5 60.5
NGC 4472
Pressure 2 kpc
3.4 6.9 10.1 13.4 16.6 20.0 23.3 26.5 29.9 33.2 36.4
NGC 4472
Entropy 2 kpc
0.68 0.73 0.79 0.84 0.90 0.96 1.01 1.07 1.12 1.18 1.23
NGC 4472
kT 2 kpc
Figure 4.1: Top panels: Chandra X-ray image of NGC 4472 in the 0.5–2 keV band (left
panel); K-band near-infrared image from the Two Micron All-Sky Survey (2MASS;
central panel); VLA 1.4 GHz radio map obtained in the C-conﬁguration from Dunn
et al. (2010) with the X-ray contours over-plotted (right panel). The images are
shown in on a logarithmic scale. Bottom panels: 2D map of pressure (in units of
keV cm−3 × l
20kpc
−1/2
; left panel), entropy (in units of keV cm2 × l
20kpc
1/3
; central
panel), and temperature (in units of keV; right panel). The maps were obtained by ﬁtting
each region independently with a single temperature thermal model, yielding 1σ
fractional uncertainties of ∼3–5 per cent in temperature, entropy, and pressure. Point
sources were excluded from the spectral analysis and therefore appear as black circles
on the maps.
4.2 Sample selection, data reduction, and analysis 59
X-ray 2 kpc
NGC 4649
Near-infrared 2 kpc
NGC 4649
2 kpcRadio
NGC 4649
2.0 2.1 2.3 2.6 3.3 4.7 7.4 12.8 23.8 45.4 88.6
2 kpcPressure
NGC 4649
3.2 6.7 10.3 14.0 17.5 21.1 24.6 28.0 31.7 35.3 38.8
2 kpcEntropy
NGC 4649
0.69 0.72 0.75 0.78 0.81 0.84 0.88 0.91 0.94 0.97 1
2 kpckT
NGC 4649
Figure 4.2: As Fig. 4.1, but for NGC 4649. The VLA 1.4 GHz radio map was obtained
in the B-conﬁguration (Dunn et al. 2010).
60 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
X-ray 4 kpc
NGC 1399
Near-infrared 4 kpc
NGC 1399
4 kpcRadio
NGC 1399
5.0 5.1 5.4 6.0 7.1 9.3 13.8 22.6 40.4 75.7 146
4 kpcPressure
NGC 1399
0 7.3 14.7 22.2 29.5 37.1 44.4 51.7 59.2 66.6 73.9
4 kpcEntropy
NGC 1399
0.74 0.88 1.0 1.2 1.3 1.4 1.6 1.7 1.8 2.0 2.1
4 kpckT
NGC 1399
Figure 4.3: As Fig. 4.1, but for NGC 1399. The VLA 1.4 GHz radio map was obtained
in the H-conﬁguration (hybrid A and B conﬁguration; Dunn et al. 2010).
4.2 Sample selection, data reduction, and analysis 61
3 kpcX-ray
NGC 4261
3 kpcNear-infrared
NGC 4261
5 kpcRadio
NGC 4261
0 0.1 0.2 0.5 1.0 2.2 4.4 8.8 17.7 35.2 70.2
3 kpcPressure
NGC 4261
0 3.0 5.8 8.8 11.8 14.9 17.7 20.7 23.7 26.5 29.5
3 kpc
NGC 4261
3 kpcEntropy
0.56 0.61 0.65 0.69 0.73 0.78 0.82 0.86 0.90 0.94 1.0
3 kpckT
NGC 4261
Figure 4.4: As Fig. 4.1, but for NGC 4261.
62 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
X-ray 4 kpc
NGC 1407
Near-infrared 4 kpc
NGC 1407
0 0.1 0.3 0.6 1.3 2.7 5.5 11.0 22.1 44.1 87.8
4 kpc
NGC 1407
Pressure
0 8.8 17.9 26.9 35.8 44.8 53.6 62.5 71.5 80.6 89.4
4 kpc
NGC 1407
Entropy
0.67 0.75 0.83 0.90 0.98 1.06 1.14 1.22 1.30 1.38 1.46
4 kpc
NGC 1407
kT
Figure 4.5: As Fig. 4.1, but for NGC 1407.
4.2 Sample selection, data reduction, and analysis 63
distance d < 35 Mpc. From the resulting sample of thirteen galaxies, we select those
systems that appear the most morphologically relaxed at X-ray wavelengths (i.e. symmetric
and undisturbed in their Chandra images). The highly symmetric morphologies
of these galaxies ensure that their deprojected thermodynamic properties can be determined
as robustly as possible. Five galaxies meet these criteria. Their X-ray, nearinfrared,
and radio images (except for NGC 1407, for which we do not have spatially
resolved radio data) are shown in Figs. 4.1–4.5 and their properties are summarized in
Table 9.2.
As seen in the X-ray images (the reduction of the X-ray data is described in Sect. 4.2.2),
NGC 4649 and NGC 1407 are highly symmetric. Although NGC 1399, NGC 4472,
and NGC 4261 do display slight departures from spherical symmetry, they are still
among the most X-ray morphologically relaxed giant elliptical galaxies known. The
faintest galaxy in our sample is NGC 4261, with a bolometric X-ray luminosity of
LX = 1.63 × 1041 erg s−1 and ﬂux of 1.36 × 10−12 erg s−1 cm−2. Our relatively high
luminosity limit restricts the contamination of the detected X-ray emission by an unresolved
population of X-ray binaries (see Sect. 4.2.2 for details). The relatively high
X-ray ﬂuxes of our sources, combined with their proximity ensure that we can measure
the thermodynamic properties down to sub-kpc radii.
The processed near-infrared K-band images obtained with the Two Micron All-Sky
Survey (2MASS) shown in the upper central panels of Figs. 4.1–4.5 were downloaded
from the NASA/IPAC Infrared Science Archive1 (Jarrett et al. 2003). They show that
the stellar light and by implication the star-mass distribution in all of the galaxies is
highly symmetric with no indications for disturbance or star-formation. The galaxies
do not, in general, show detectable dust lanes or H I emission and their stellar populations
are old and red (Serra & Oosterloo 2010). NGC 4261 does show a nuclear disk
of gas and dust in the innermost r ∼ 100 pc (see Jaffe et al. 1993; Jaffe & McNamara
1994) containing (5.4 ±1.8)×104 M of molecular and atomic hydrogen (Ferrarese et al.
1996); however, cool gas or dust has not been detected elsewhere in NGC 4261 (Serra
& Oosterloo 2010; Combes et al. 2007). All ﬁve galaxies are essentially ‘red and dead’.
All ﬁve galaxies have central, active radio jets, but their observed radio powers and
morphologies span a wide range, as indicated by the radio luminosities listed in Table
9.2 and by the radio maps in Figs. 4.1–4.4 (see also Dunn et al. 2010). NGC 4261
is the most powerful radio source, with jets that puncture through the X-ray emitting
galactic atmosphere to form giant lobes in the surrounding intra-group medium (IGM).
NGC 1399 has ∼10 kpc long radio jets with lobes that also appear to be depositing energy
away from the centre of the galaxy. The radio lobes of NGC 4472, and of the 10
times less radio luminous NGC 4649, are seen at smaller radii. For NGC 1407, we do
not presently have high quality radio data, but its radio luminosity is similar to those
of NGC 4472 and NGC 1399 (see Table 9.2), clearly indicating the presence of jets.
1http://irsa.ipac.caltech.edu/
64 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
4.2.2 Chandra data reduction and analysis
Data reduction
Our analysis of the Chandra data follows the data reduction procedures described in
Million et al. (2010c) and Million et al. (2010d). The standard level-1 event lists were
reprocessed using the CIAO (version 4.3) software package, including the latest gain
maps and calibration products. Bad pixels were removed and standard grade selections
applied. The data were cleaned to remove periods of anomalously high background.
The net exposure times after cleaning are summarized in Table 9.2. Background
images and spectra were extracted from the blank-sky ﬁelds available from the
Chandra X-ray Center. These were cleaned in an identical way to the source observations,
reprojected to the same coordinate system, and normalized by the ratio of the
observed to blank-sky count rates in the 9.5–12 keV band.
Background subtracted images were created in 6 narrow energy bands, spanning
0.5–2.0 keV. These were ﬂat ﬁelded with respect to the median energy for each image
and then co-added to create the broad band X-ray images shown in Figs. 4.1–4.5.
Identiﬁcation of point sources was performed using the CIAO task WAVDETECT.
These were excised from the images in Figs. 4.1–4.5 and substituted by local backgrounds.
Point sources were also excluded from all regions used for spectral analysis.
Separate photon-weighted response matrices and effective area ﬁles were constructed
for each spectral region.
Spectral analysis
Our spectral analysis was separated into two parts. In the ﬁrst part, 2D maps of projected
thermodynamic quantities were made. The individual regions for the 2D spectral
mapping were determined using the Contour Binning algorithm (Sanders 2006),
which groups neighboring pixels of similar surface brightness until a desired signal-tonoise
threshold is met. In order to have small enough regions to resolve substructure,
yet still have enough counts to achieve better than 5 per cent accuracy in the temperature
measurements, we adopted a signal-to-noise ratio of 18 (∼320 counts per region).
We modeled the spectra extracted from each spatial region with the SPEX2 package
(Kaastra et al. 1996, SPEX is faster, uses an updated version of the MEKAL plasma
model with respect to XSPEC, though provides consistent results). Because the spectral
band above 2 keV is dominated by background in low temperature galaxies, the spectral
ﬁtting was performed in the 0.6–2.0 keV band. The spectrum for each region was
ﬁtted with a model consisting of an absorbed single-phase plasma in collisional ionization
equilibrium, with the temperature and spectral normalization (emission measure)
as free parameters. The line-of-sight absorption column densities, NH, were ﬁxed to
the values determined by the Leiden/Argentine/Bonn radio survey of H I (Kalberla
et al. 2005, see Table 9.2).
After the point source removal, the contribution from an unresolved population
of low mass X-ray binaries (LMXB) to the X-ray emission is insigniﬁcant. We per-
2www.sron.nl/spex
4.3 Thermodynamic measurements 65
formed ﬁts, where we included power-law emission components with photon indices
Γ = 1.56 − 1.80, which were shown to describe well the population of low-luminosity
LMXB (Irwin et al. 2003; Kim & Fabbiano 2004). Such additional power-law emission
is not detected signiﬁcantly in the galaxies and including it in the spectral ﬁtting does
not signiﬁcantly affect the measured thermodynamic properties of the hot ISM (we
note that this conclusion does not change if we extend the ﬁtted band to 7 keV).
The largest systematic uncertainty in the measured densities stems from the metallicity
measurements. When ﬁtting the spectra of thermal plasmas with kT in the range
∼0.5–1.0 keV, the emission measure and metallicity typically anti-correlate. In regions
with complicated temperature structure the metallicity is often biased low and
consequently the emission measure is biased high (e. g. Buote 2000b; Werner et al.
2008). Moreover, for spectra with relatively low numbers of counts, the statistical uncertainties
on the metallicity are high and multi-temperature model ﬁts are unfeasible.
Therefore, our 2D maps of thermodynamic properties were produced with the
metallicity ﬁxed to 0.5 Solar (throughout the paper, metal abundances are given with
respect to the Solar values by Grevesse & Sauval 1998). If the metallicity is underestimated/overestimated
by a factor of 2 in the model, the density will be overestimated/underestimated
by a factor of ∼1.35.
In the second stage of our analysis, we measured azimuthally-averaged, deprojected
radial proﬁles of thermodynamic quantities. We extracted spectra from concentric annular
regions at least 1 arcsec wide, with a signal-to-noise ratio of at least 18 (∼320
counts). The deprojected proﬁles of thermodynamic properties were obtained with the
XSPEC (version 12.5 Arnaud 1996) spectral ﬁtting package, using the projct model. The
combined set of azimuthally averaged spectra extracted from concentric annuli was
modeled in the 0.6–2.0 keV band simultaneously to determine the deprojected electron
density (ne) and temperature (kTe) proﬁles. The emission from each spherical shell was
modeled with a photoelectrically absorbed (Balucinska-Church & McCammon 1992)
APEC thermal plasma model (Smith et al. 2001, using AtomDB v2.0.1). Deprojected
densities were determined in 14–27 shells, and temperatures in 7–10 regions, depending
on the galaxy. For each galaxy we ﬁtted two metallicity values, one for the shells
outside the radius of ∼1 kpc and another value for the shells in the central regions of
the galaxies.
For all spectral ﬁtting, we employ the extended C-statistics available in XSPEC and
SPEX. All errors are quoted at the 68 per cent conﬁdence level for one interesting parameter
(∆C = 1).
4.3 Thermodynamic measurements
4.3.1 2D distribution of thermodynamic properties
The lower panels of Figs. 4.1–4.5 show 2D maps of projected electron pressures (nekTe),
entropies (kTe/n
2
3
e ), and temperatures (kTe). The projected electron densities were determined
assuming a ﬁxed line of sight of column depth l = 20 kpc.
66 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
The 2D distributions of the pressures are approximately spherically symmetric in
the cores of all ﬁve galaxies, consistent with the X-ray emitting gas being in approximate
hydrostatic equilibrium. The most obvious departure from spherical symmetry
is seen in NGC 4472 in the areas spatially coincident with the radio lobes at radii
r ∼ 4 kpc, where the projected thermal pressure drops by 10–25 per cent.
The spatial distribution of the projected entropy appears spherically symmetric in
4/5 systems, the exception being NGC 1399 where the entropy distribution beyond
r ∼ 1.5 kpc is elongated, with lower entropy gas extended along the radio jets. Overall,
however, in all cases, the lowest entropy gas resides in the innermost cores of the
galaxies, as expected for convective stability. The regions of relatively high entropy
at r > 6 kpc, to the north of NGC 4472 and NGC 1407, and to the east of NGC 1399,
indicate that these galaxies are moving through the ambient medium.
The observed temperature distributions of the galaxies appear less symmetric. In
NGC 4472 we see cooler plasma extended along the eastern radio lobe, and in NGC 1399
the low temperature ISM is distributed along the radio jets. The lowest projected temperatures
of the galactic atmospheres are in the range 0.6–0.75 keV. Such low ISM temperatures
are not, however, always seen in the densest, central regions. Three out of
ﬁve galaxies (NGC 4649, NGC 1407, and NGC 1399) show a signiﬁcant temperature
increase in their cores. The clearest central temperature increase is seen in NGC 4649,
arguably the most morphologically relaxed galaxy in our sample.
4.3.2 Azimuthally-averaged, deprojected thermodynamic properties
The relatively undisturbed X-ray morphologies and high degree of azimuthal symmetry
observed in the distributions of projected thermodynamic quantities conﬁrms that
the ﬁve systems in this study are among the most dynamically relaxed giant elliptical
galaxies in the local Universe. All systematic uncertainties associated with deprojection
analyses, carried out under the assumption of spherical symmetry and of a singlephase
nature at each radius, are therefore minimized for these systems. We have determined
the azimuthally-averaged deprojected radial proﬁles of electron density (ne),
temperature (kTe), and metallicity (Z) from the Chandra data. Using these quantities
we also determine the deprojected radial proﬁles for the cumulative gas mass, entropy
(K = kTe/n
2
3
e ), electron pressure (Pe = nekTe), and cooling time. We deﬁne the cooling
time as the gas enthalpy divided by the energy radiated per unit volume of the plasma:
tcool =
5
2(ne + ni)kT
neniΛ(T)
, (4.1)
where the ion number density ni = 0.92ne, and Λ(T) is the cooling function for Solar
metallicity tabulated by Schure et al. (2009). Cooling functions based on older plasma
codes (Sutherland & Dopita 1993) predict a 9 per cent lower cooling rate for 0.5 keV
plasma.
Our main result from the deprojection analysis is that beyond the central r ∼ 1 kpc,
and out to the radii r 6 kpc, the cumulative gas mass, density, pressure, entropy,
and cooling time distributions for the X-ray emitting gas follow remarkably similar
4.3 Thermodynamic measurements 67
radial proﬁles (see Fig. 4.6). The observed cumulative gas mass within r = 8 kpc is
∼ 5.8 × 108 M with a system-to-system dispersion of 10 per cent. The entropy proﬁles
follow a simple, power-law form K ∝ rα with an index α = 0.92-1.07. The weighted
system-to-system dispersions in entropy at r = 1 kpc and at r = 5 kpc are only 14 per
cent and 17 per cent, respectively. The density proﬁles follow the form ne ∝ r−β with an
index β = 1.1–1.46 and with a system-to-system dispersion of 10 per cent at r = 5 kpc.
The pressure proﬁles have an approximate power-law form of Pe ∝ r−γ with an index
γ = 0.93–1.43 and with a system-to-system dispersion of 13 per cent at r = 5 kpc. The
thermodynamic proﬁles of NGC 4261 can be well described with power-law relations
all the way down to the smallest radii measured.
In the absence of heating, cooling in the central regions of these galaxies would be
strong: within the radius r 4 kpc, the cooling times in all ﬁve galaxies are tcool <
1 Gyr. Within r 1 kpc of the center, the cooling times are tcool < 108 yr, and less than
a few 107 yr within 0.5 kpc.
All ﬁve galaxies show a ﬂattening of their central entropy proﬁles within r 1 kpc.
In contrast to the remarkable similarity in the thermodynamic proﬁles at larger radii
(1 r 6 kpc), within this innermost r 1 kpc region the entropies of the galaxies
also show a signiﬁcant system-to-system scatter of 40 per cent at r = 0.2 kpc. The
system with the least central ﬂattening of the entropy distribution is NGC 4261, where
the entropy drops to K0 = 0.88 ± 0.02 keV cm2 at 0.23 kpc. The largest core entropy of
K0 = 4.1 ± 0.5 keV cm2 at r = 0.27 kpc is observed in NGC 1407.
The temperature proﬁles of the galaxies exhibit signiﬁcant system-to-system scatter.
Interestingly the temperature proﬁles for NGC 4649, NGC 1399, and NGC 1407
show clear temperature increase in the central regions, which may be associated with
repeated shock activity increasing the entropy of the gas. In contrast, the radial temperature
distributions in the centres of NGC 4472 and NGC 4261 are relatively ﬂat.
The best ﬁt metallicities of the hot ISM within r ∼ 1 kpc are Z ∼ 0.5 Solar, except
for NGC 1399 which has a core metallicity of Z ∼ 0.85 Solar. Beyond this innermost
region (1 < r < 6 kpc) the metallicities of NGC 4649 and NGC 4261 are Z ∼ 0.3 Solar;
NGC 1407 has Z ∼ 0.75 Solar, and the metal abundances of NGC 1399 and NGC 4472
are Z ∼ 1 Solar. The systematic uncertainties on the metallicity measurements of kT
1 keV plasma with CCD type detectors are, however, unfortunately still signiﬁcant.
To verify that the spectroscopically measured temperatures, and other thermodynamic
quantities, determined in azimuthally averaged annular regions are approximately
unbiased with respect to the true average values at a given radius, we have
compared their results with the average values determined from all bins at the same
radii in the 2D maps. We ﬁnd no signiﬁcant differences between the values found
by ﬁtting spectra extracted from circular annuli and the azimuthally averaged values
determined from 2D maps at any radius.
The obtained deprojected density and temperature proﬁles are broadly consistent
with the previously published results in the literature (e.g. Humphrey et al. 2006, 2009;
Churazov et al. 2010).
68 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
4.4 Discussion
Beyond the innermost core, at r 1 kpc and out to at least r ∼ 6 kpc, the X-ray emitting
hot gas mass proﬁles and thermodynamic proﬁles exhibit remarkable self-similarity.
The density, entropy, pressure, and cooling time distributions follow simple, powerlaw
forms. The total gravitating mass at these radii is also similar in all ﬁve galaxies,
and is dominated by their stellar populations (Humphrey et al. 2006; Das et al. 2010).
The mass fraction of hot X-ray emitting gas in this radial range is small, of the order of
fgas 2 × 10−3. Because the hot ISM represents a minor and essentially dynamically
insigniﬁcant fraction of the total mass of the galaxies, the observed similarity of the
hot gas mass and its thermodynamic proﬁles is surprising and reﬂects a deep and
long term stability of energy input from the AGN and from the stellar populations.
Interestingly, the ratio of the cumulative gas mass within r < 8 kpc to the total B-band
luminosity for the galaxies in our sample spans a relatively small range of Mgas/LB =
7.1 − 11.3 × 10−3 M /L .
4.4.1 Universal thermodynamic proﬁles
The similarity of the thermodynamic properties of the gaseous atmospheres of our
galaxy sample has important implications for the evolution of giant ellipticals. The
cooling time of the hot gas in the central regions is short (see left panel of Fig. 4.6)
and in the absence of heating this gas will cool and form stars (e.g. Peterson & Fabian
2006). The X-ray spectra of these galaxies, however, lack the expected strong O VII
lines associated with gas cooling below T ∼ 5 × 106 K (see e.g. Sanders & Fabian
2011). Moreover, their stellar populations are old, showing no evidence for recent starformation,
and no signiﬁcant central atomic or molecular reservoirs of cooled gas are
detected (Serra & Oosterloo 2010). All of this indicates that the gas has been kept from
cooling.
Type Ia supernovae are expected to explode in giant elliptical galaxies at a rate of
0.166 per 100 yr per 1010 LB (Cappellaro et al. 1999), releasing ∼ 1051 ergs of energy
per supernova. In principle, this can provide a heating rate of 1.6 × 1041 erg s−1 within
r 6 kpc, which is of the same order of magnitude as the X-ray luminosity within
the same region. However, to approximately balance the radiative cooling, the energy
released by supernovae would have to couple to the X-ray emitting gas with a 100 per
cent efﬁciency, which seems unlikely. The dominant source of heating is instead likely
to be the central AGN.
As indicated by the observed radio morphologies, the jets and lobes driven by the
central AGN interact with and mechanically heat the ISM out to a radius of several
kpc. A complete, ﬂux-limited study of 18 nearby, X-ray bright, optically bright giant
elliptical galaxies by Dunn et al. (2010) shows that 17/18 systems exhibit some form of
current radio activity associated with the central AGN (see also Best et al. 2005; Dunn
& Fabian 2006, 2008; Sun 2009). Such observations imply that active ‘radio-mode’ AGN
feedback is not a rare and sporadic phenomenon, but rather represents the default state
for large elliptical galaxies. Allen et al. (2006) demonstrate a tight correlation between
the Bondi accretion rates of hot gas and the observed jet power in such ellipticals. This
4.4 Discussion 69
Figure 4.6: Left panel: Radial distributions of cooling times, entropies, and cumulative
gas masses for our sample of 5 nearby giant elliptical galaxies. Right panel: Radial distributions
of the deprojected temperatures, electron densities, and electron pressures.
argues that AGN feedback is likely controlled by the accretion of hot gas, or multiphase
gas entrained within it. The similarity of the observed thermodynamic proﬁles
reported here supports this, and argues that relativistic jets, produced by accretion onto
the central SMBH, heat the gas at a similar and constant rate averaged over time scales
longer than the cooling time, tcool 108 yr. This jet heating creates an energy balance
where heating and cooling are in approximate equilibrium, keeping the hot galactic
atmospheres in a near ‘steady-state’. Importantly, the observed ‘steady-state’ also indicates
that the time averaged accretion rate onto the SMBH is intimately connected to
the conditions of the hot X-ray emitting gas phase on spatial scales of several kpc.
The jet-powers estimated from the energies and timescales required to inﬂate the
bubbles of relativistic radio-emitting plasma are Pjet ∼ 8 × 1042 ergs s−1 in NGC 4472
(Allen et al. 2006); Pjet ∼ 1.3 × 1042 ergs s−1 in NGC 4649 and Pjet ∼ 2.2 × 1042 ergs s−1
in NGC 1399 (Shurkin et al. 2008); and Pjet > 1043 ergs s−1 in NGC 4261 (O’Sullivan
et al. 2011). In NGC 4472, NGC 4649, and NGC 1399 the energy required to keep the
hot ISM within r 6 kpc from cooling is only 2.5–10 per cent of the estimated jetpower,
and in NGC 4261, which is the system with the most powerful radio source
in our sample, this energy is less than 1 per cent of the power output of the AGN.
In NGC 4261 the jets penetrate through the hot galactic corona and form lobes which
deposit most of their energy into the surrounding intergalactic medium (O’Sullivan
et al. 2011).
Although we do not see evidence for shocks currently propagating through the Xray
atmospheres of the galaxies in our sample, repeated AGN induced shocks most
likely do contribute to the heating of the hot ISM. Detailed studies of the giant elliptical
galaxy NGC 5813 revealed that the total energy in shocks driven by the expand-
70 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
ing bubbles can be 1/2–3 times the total internal energy of the cavities (Randall et al.
2011b). The total jet power in these systems is therefore most likely even higher than
the estimates provided above. Interestingly, despite the fact that the fraction of the estimated
‘current’ jet-power that is needed in each case to keep the hot ISM from cooling
varies from galaxy to galaxy by an order of magnitude, the observed similarity of the
thermodynamic properties, including the gas mass and entropy, argues that the energy
deposited into the ISM within r 6 kpc must be similar and steady, to keep the hot
galactic atmospheres in a ‘steady-state’.
4.4.2 Heating in the innermost r 1 kpc
As shown in Sect 9.5, the entropy and cooling time proﬁles show well resolved ﬂattening
at radii r 1 kpc. This ﬂattening, combined with the short cooling times of
tcool < 108 yr, implies that the gas at these radii has been heated by AGN activity relatively
recently. Radiative cooling will reduce the entropy of the central gas, and in
order to offset such cooling and ﬂatten the proﬁles a heating mechanism is required.
Heating by repeated AGN induced weak shocks is likely to be the most signiﬁcant
channel for entropy increase of the hot plasma near to the AGN. Shock heating, which
increases the entropy by ∆S, will offset the radiative heat loss by ∆Q = T∆S (where T
is the temperature of the gas). Examples of pronounced, ongoing AGN induced shock
heating have been studied in detail in M 87 (Forman et al. 2005, 2007; Million et al.
2010d) and in NGC 5813 (Randall et al. 2011b).
The steepest central entropy distribution is seen in NGC 4261, where the entropy
continues to drop all the way in to the centre of the system. The fact that the central
entropies at r < 1 kpc, where tcool < 108 yr, are substantially different from system to
system indicates that the accretion rate onto the black hole and the resulting jet-power,
heating the ISM, varies signiﬁcantly on time scales shorter than a few 107 yr.
Given the remarkable similarity of the proﬁles at r > 1 kpc our analysis suggests
that the application of hydrostatic mass analyses to the regions of those galaxies from
1–6 kpc are likely to be relatively accurate (e. g. Humphrey et al. 2006; Churazov et al.
2008, 2010). Humphrey et al. (2008, 2009) extend their hydrostatic mass analyses all
the way into the cores of their target galaxies, arguing that in some systems, most
notably in NGC 4649, relative hydrostatic equilibrium applies all the way to small
radii allowing to infer directly the mass of the central SMBH. The increased systemto-system
scatter in our observed central gas entropy indicates that caution should
be employed when extending such analyses to the inner 1 kpc, where feedback may
induce signiﬁcant thermodynamic variations on timescales as short as a few 106 yr.
4.4.3 The origin of the hot gas
There are three possible explanations for the origin of the hot gas in giant ellipticals:
ﬁrst, it could be leftover baryonic material from the process of galaxy formation. In this
case, the similarity of the gas mass to stellar light ratios in the galaxies would indicate
similar star formation and feedback histories. Second, some fraction of the X-ray emit-
4.5 Conclusions 71
ting gas could have been accreted from the surrounding large scale environment. If this
process is important, then the gas mass to stellar light ratios would depend strongly on
the large scale environments of the galaxies and would therefore be expected to exhibit
signiﬁcant system-to-system variation, which is not seen in our sample3. A third possibility
is that stellar mass loss contributes signiﬁcantly to the X-ray halos (for a review
see Mathews & Brighenti 2003). In this process, gas from the outer layers of evolved
stars is mixed and assimilated into the hot phases of the galaxies (Mathews 1990). Hydrodynamic
simulations of the gas ejected by red giant stars and planetary nebulae
have been performed by Parriott & Bregman (2008) and Bregman & Parriott (2009).
These simulations predict that about 75 per cent of the ejecta produced by red giant
stars moving supersonically relative to the ambient medium will be shock heated to
approximately the temperature of the hot ISM. However, the products of stellar mass
loss evidently do not thermalize universally in all environments. The presence of dust
and PAHs in the dense cooling cores of some galaxy clusters suggests that in those systems
much of the ejected stellar gas remains cool (Voit & Donahue 2011; Donahue et al.
2011). The high ambient pressures in the cores of galaxy clusters with massive cooling
cores might be promoting more rapid radiative cooling of the products of stellar mass
loss. However, none of our galaxies lie at the centers of massive cooling cores.
While we cannot at present discriminate clearly whether original baryonic matter
left over from the process of galaxy formation or the products of stellar mass loss dominate
the X-ray emitting material within 1 < r < 8 kpc, the remarkable similarity of
the thermodynamic and gas mass proﬁles, and gas mass to stellar light ratios in the
systems presented here provides an important, new constraint.
4.5 Conclusions
We performed a detailed spatially resolved study of the thermodynamic properties
of the X-ray emitting gas in the inner regions of the ﬁve nearest, X-ray and optically
brightest, and most morphologically relaxed giant elliptical galaxies at X-ray wavelengths.
Our main conclusions may be summarized as follows:
• Beyond the innermost central region, at r 1 kpc and out to r ∼ 6 kpc, the
radial proﬁles of density, pressure, cooling time, and entropy of these galaxies
follow remarkably similar, simple, power-law like distributions. The entropy
proﬁles follow a form K ∝ rα with an index α = 0.92-1.07, with a system to
system dispersion at a given radius of only ∼15 per cent. The cumulative hot
X-ray emitting gas mass proﬁles and the gas-mass to stellar-light ratios of all ﬁve
galaxies are also similar.
• The observed similarity of the thermodynamic proﬁles in this radial range argues
that, in these systems, relativistic jets heat the gas at a similar rate averaged over
3The gas mass fractions around the brightest cluster galaxies of cooling core clusters are signiﬁcantly
higher than those of the galaxies investigated in this paper (e. g. Allen et al. 2008). Most of the gas in
those systems likely originates from the surrounding intra-cluster medium.
72 On the thermodynamic self-similarity of relaxed, nearby, giant ellipticals
time scales longer than the cooling time tcool 108 yr. This jet heating creates
an energy balance where heating and cooling are in equilibrium, keeping the hot
galactic atmospheres in a ‘steady-state’.
• The entropy proﬁles show well resolved ﬂattening at radii r 1 kpc. In contrast
to the remarkable similarity at larger radii (1 r 6 kpc), the central entropy
value differs from system to system, with a scatter of 40 per cent at 0.2 kpc. The
accretion rate onto the black hole and the AGN activity, heating the ISM, must
therefore vary signiﬁcantly on time scales shorter than tcool = 107–108 yr.
• Our results support the picture in which the jets associated with the central AGN
are powered by the accretion of hot gas, or the material entrained within it.
Acknowledgments
We thank Robert Dunn for providing us the radio data for NGC 4472, NGC 4649, and
NGC 1399. We thank Payel Das for providing us stellar mass and dark matter proﬁles
for the galaxies. We thank Paul Nulsen and Mark Voit for inspiring discussions.
Support for this work was provided by the National Aeronautics and Space Administration
through Chandra/Einstein Postdoctoral Fellowship Award Number PF8-90056
and PF9-00070 and through the Chandra Award Number GO9-0088X issued by the
Chandra X-ray Observatory Center, which is operated by the Smithsonian Astrophysical
Observatory for and on behalf of the National Aeronautics and Space Administration
under contract NAS8-03060. SWA acknowledges support from the U.S. Department
of Energy under contract number DE-AC02-76SF00515
Chapter 5
Constraints on turbulent pressure in the
X-ray halos of giant elliptical galaxies
from resonant scattering
N. Werner1, I. Zhuravleva2, E. Churazov2,3, A. Simionescu4, S. W. Allen1, W. Forman5, C.
Jones5, J.S. Kaastra6,7
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall/mc 4085, Stanford, CA 94305, USA
2Max-Planck-Institut f¨ur Astrophysik, Karl-Schwarzschild-Strasse 1, 85741 Garching,
Germany
3Space Research Institute (IKI), Profsoyznaya 84/32, Moscow 117810, Russia
4Max-Planck-Institut f¨ur Extraterrestrische Physik, Giessenbachstr, 85748 Garching,
Germany
5Harvard-Smithsonian Center for Astrophysics, 60 Garden St., Cambridge, MA 02138,
USA
6SRON Netherlands Institute for Space Research, Sorbonnelaan 2, 3584 CA Utrecht,
the Netherlands
7Sterrenkundig Instituut, Universiteit Utrecht, P.O. Box 80000, 3508 TA Utrecht, the
Netherlands
Published in the Monthly Notices of the Royal Astronomical Society, volume 398,
pages 23–32, 2009
Abstract
The dense cores of X-ray emitting gaseous halos of large elliptical galaxies with temperatures
kT 0.8 keV show two prominent Fe XVII emission features, which provide
a sensitive diagnostic tool to measure the effects of resonant scattering. We present
here high-resolution spectra of ﬁve bright nearby elliptical galaxies, obtained with the
Reﬂection Grating Spectrometers (RGS) on the XMM-Newton satellite. The spectra for
74 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
the cores of four of the galaxies show the Fe XVII line at 15.01 ˚A being suppressed by
resonant scattering. The data for NGC 4636 in particular allow the effects of resonant
scattering to be studied in detail and to prove that the 15.01 ˚A line is suppressed only
in the dense core and not in the surrounding regions. Using deprojected density and
temperature proﬁles for this galaxy obtained with the Chandra satellite, we model the
radial intensity proﬁles of the strongest resonance lines, accounting for the effects of
resonant scattering, for different values of the characteristic turbulent velocity. Comparing
the model to the data, we ﬁnd that the isotropic turbulent velocities on spatial
scales smaller than ≈1 kpc are less than 100 km s−1 and the turbulent pressure support
in the galaxy core is smaller than 5% of the thermal pressure at the 90% conﬁdence
level, and less than 20% at 95% conﬁdence. Neglecting the effects of resonant scattering
in spectral ﬁtting of the inner 2 kpc core of NGC 4636 will lead to underestimates
of the chemical abundances of Fe and O by ∼10–20%.
5.1 Introduction
The hot intra-cluster medium and the hot halos around giant elliptical galaxies are
usually assumed to be optically thin. Although this assumption is valid for most of
the emitted X-ray photons, at the energies of the strongest resonant transitions, the
hot plasma can be optically thick (Gilfanov et al. 1987). The transition probabilities of
strong resonance lines are large and, if the column density of the ion along a line of
sight is sufﬁciently high, photons with the energies of these resonance lines will get
absorbed and, within a short time interval, reemited in a different direction. Because
of the short time between absorption and emission, this process can be regarded as
scattering. Since resonant scattering in clusters and elliptical galaxies will cause the
radial intensity proﬁle of an emission line to become weaker in the centre and stronger
outside, it can also lead to an underestimate of metal abundances in the dense cores
and a corresponding overestimate in the surrounding region. This effect is, however,
not expected to be large. Sanders & Fabian (2006) found that metallicities in cluster
cores could be underestimated by at most 10% due to resonant scattering.
Gilfanov et al. (1987) pointed out that since the optical depth τ in the core of a resonance
line depends on the characteristic velocity of small-scale motion, measurements
of τ give important information about the turbulent velocities in the hot plasma. Measuring
the level of resonant scattering in clusters and giant elliptical galaxies is thus
a good way to constrain the energy in turbulence and the turbulent pressure support.
The ﬁrst constraints on turbulent velocities using resonant scattering were obtained
for the Perseus cluster by Churazov et al. (2004). Using XMM-Newton EPIC data, they
compared the relative ﬂuxes of the 1s–2p and 1s–3p He-like Fe lines in the core and
in an annulus surrounding the core of the Perseus cluster. The expected optical depth
of the 6.7 keV 1s–2p Fe XXV resonance line is much larger than that of the 1s–3p line,
therefore the ratio provides information about the level of resonant scattering. Churazov
et al. (2004) found no evidence for resonant scattering in Perseus, indicating that
differential gas motions on scales smaller than ∼100 kpc in the core of the cluster must
have a range of velocities of at least half of the sound speed. Independently, using the
5.1 Introduction 75
00.20.40.60.81
FeXVIIionicfraction
NGC1404
NGC4636
NGC5813
NGC4649
NGC4472
0.2 0.4 0.6 0.8 1
11.21.41.61.8
(I17.05
+I17.10
)/I15
kT (keV)
Figure 5.1: Fraction of Fe in the form of Fe XVII and the theoretical line ratio (Iλ17.05 +
Iλ17.10)/Iλ15.01 for an optically thin plasma as a function of the plasma temperature.
The vertical lines mark the RGS measured temperatures of the ﬁve galaxies in our
sample.
same data, Gastaldello & Molendi (2004) reached similar conclusions.
The most sensitive spectral lines to determine the level of resonant scattering are the
Fe XVII lines at 15.01 ˚A (2p–3d) and the unresolved blend of the same ion at 17.05 and
17.10 ˚A (2p–3s). The optical depth is directly proportional to the column density of
the ion and the oscillator strength of the given transition. While the oscillator strength
of the 15.01 ˚A line is f = 2.73, and thus the line is expected to have a relatively large
optical depth, the oscillator strength of the 17.05 ˚A line is f = 0.12 and the optical
depth of this line blend is negligible (the oscillator strength of the 17.10 ˚A line is of the
order of 10−8)1. Because of this dramatic difference in expected optical depths, and
the fact that both lines originate from the same ion of the same element, the Fe XVII
lines provide, in principle, an excellent diagnostic tool to measure the magnitude of
resonant scattering. At the relatively low temperatures of hot halos around giant elliptical
galaxies, ∼0.6–0.8 keV, both lines are very strong and, in this temperature range,
their expected intensity ratios have only a very weak dependence on temperature (see
Fig. 5.1, Doron & Behar 2002). Therefore, any observed spatial dependence of the intensity
ratios is unlikely to be due to gradients in temperature or Fe abundance in the
1http://cxc.harvard.edu/atomdb/WebGUIDE/index.html
76 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
Table 5.1: List of the galaxies in the sample, with their distance (luminosity
distances, obtained from the NASA/IPAC Extragalactic Database,
http://nedwww.ipac.caltech.edu/), corresponding linear scale per arcminute, Galactic
NH value (Kalberla et al. 2005), average temperature measured by ROSAT. The last
two columns give the total and ﬁltered RGS exposure times, respectively.
galaxy distance scale NH kT Exp. time Clean time
Mpc kpc/arcmin 1020 cm−2 keV ks ks
NGC 4636 17.5 5.1 1.90 0.55∗ 64406 58205
NGC 5813 29.9 8.7 4.37 0.52† 36708 29899
NGC 1404 25.5 7.4 1.51 0.60∗ 55032 17231
NGC 4649 19.8 5.8 2.1 0.78∗ 54210 46096
NGC 4472 18.4 5.4 1.53 0.88∗ 111033 81850
∗ O’Sullivan et al. (2003)
† Reiprich & B¨ohringer (2002)
hot plasma, making these lines especially clean diagnostic tools. The Fe XVII line ratios
in extended objects can be measured with the XMM-Newton Reﬂection Grating Spectrometers
(RGS, den Herder et al. 2001). Xu et al. (2002) reported the ﬁrst and so far
only radial proﬁle of the (Iλ17.05 + Iλ17.10)/Iλ15.01 ratio for the X-ray luminous elliptical
galaxy NGC 4636, which indicates the presence of resonant scattering. Attempts to
measure resonant scattering in NGC 5044 and M 87, which are hotter and more massive
systems with weaker Fe XVII lines, were not successful (Tamura et al. 2003; Werner
et al. 2006).
Here we follow up on the work by Xu et al. (2002), by reporting the Fe XVII (Iλ17.05 +
Iλ17.10)/Iλ15.01 line ratios measured in the cores of ﬁve nearby bright elliptical galaxies
observed with XMM-Newton RGS: NGC 4636, NGC 5813, NGC 1404, NGC 4649, and
NGC 4472. In Sect. 2 we describe the sample and observations; in Sect. 3 we discuss
the details of the data analysis; and in Sect. 4 we present the results of the observations.
We focus our attention on NGC 4636, for which data are of very high quality. Because
NGC 4636 is X-ray bright, has a relatively long exposure and a favorable temperature,
for this system we can measure and compare the observed line ratios in the core and
surrounding regions. In Sect. 5 we present a model for the radial proﬁle of the 15.01 ˚A
line in NGC 4636 for different values of turbulent velocities, derived using deprojected
density and temperature proﬁles from Chandra data. Comparing this model to the data,
we place constraints on the characteristic velocity of isotropic turbulence in the core of
the galaxy. Implications of the results are discussed in Sect. 6.
Throughout the paper, abundances are given with respect to the “proto-solar values”
by Lodders (2003). All errors are quoted at the 68% conﬁdence level for one interesting
parameter (∆C = 1; for C statistics).
5.2 Sample and XMM-Newton observations 77
Figure 5.2: Chandra image of NGC 4636 with the RGS extraction regions over-plotted.
The central extraction region is 0.5 wide (indicated by full lines), the two extraction
regions next to the core (between the full and dashed lines) are 2.25 wide. The image
was extracted in the 0.5–2.0 keV band, cleaned and adaptively smoothed.
5.2 Sample and XMM-Newton observations
We have analyzed a sample of ﬁve nearby, giant elliptical galaxies observed with
XMM-Newton RGS. The sample is listed in Table 5.1. The galaxies were selected based
on their relatively low core temperatures, at which a signiﬁcant fraction of Fe is expected
to be in the form of Fe XVII (see Fig. 5.1), their large X-ray ﬂuxes, and strongly
peaked surface brightness distributions, which are necessary to obtain RGS spectra
with sufﬁcient statistics. The sample contains both relatively relaxed galaxies and systems
showing very clear signs of interaction between the hot halo gas and central Active
Galactic Nucleus (AGN). In Figs. 5.2 and 9.6 we show 0.5–2.0 keV Chandra images
of the galaxies in our sample with the RGS extraction regions overplotted. The images
show that the cores of NGC 4649, NGC 4472, and NGC 1404 are relatively relaxed (although
on larger scales NGC 1404 shows a prominent cold front to the Northwest, and
NGC 4649 and NGC 4472 harbor central radio sources, which exhibit interaction with
the hot gas, Shurkin et al. 2008; Biller et al. 2004). NGC 4636 and NGC 5813, on the
other hand, look much more disturbed, with obvious signs of recent episodes of interaction
between the hot plasma and the AGN. NGC 4636 has a very dense core and
78 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
Figure 5.3: Chandra images of the giant elliptical galaxies in our sample with overplotted
RGS extraction regions (see also Fig. 5.2). The extraction regions are 0.5 wide.
While the core of NGC 5813 is strongly disturbed, the cores of the other three galaxies
are relatively relaxed. The images were extracted in the 0.5–2.0 keV band, cleaned and
adaptively smoothed.
5.3 Data analysis 79
Table 5.2: The best ﬁt parameters for a single-temperature, optically thin plasma model
ﬁtted to the XMM-Newton RGS spectra extracted from 0.5 wide regions centred on
the cores of the galaxies. For NGC 4636 we also show the results from ﬁts to spectra
extracted from two 2.25 wide regions surrounding the core (see Fig. 5.2). The 13.8–
15.5 ˚A part of the spectrum, where the strongest Fe XVII and Fe XVIII resonance lines
are present was initially excluded from the ﬁts. Fluxes are given in the 0.3–2.0 keV
band. The emission measure is deﬁned as Y = nenHdV. The scale factor s is the
ratio of the observed LSF width to the expected LSF for a ﬂat abundance distribution.
Abundances are quoted with respect to the proto-solar values of Lodders (2003). The
last three rows list the best ﬁt line ratios in the full spectral band (after the Fe XVII
ion was set to zero in the model and replaced by gaussian lines), the theoretical line
ratios predicted for an optically thin plasma, and the derived level of suppression of
the 15.01 ˚A line, (I/I0)15.01 ˚A
.
galaxy NGC 4636 core NGC 4636 outer reg. NGC 5813 NGC 1404 NGC 4649 NGC 4472
ﬂux (10−12 erg cm−2) 1.75 ± 0.08 2.57 ± 0.17 1.47 ± 0.12 1.08 ± 0.11 1.63 ± 0.10 1.44 ± 0.08
Y (1064 cm−3) 0.47 ± 0.02 0.46 ± 0.03 1.01 ± 0.10 0.50 ± 0.05 0.31 ± 0.03 0.34 ± 0.02
kT (keV) 0.606 ± 0.006 0.695 ± 0.004 0.645 ± 0.008 0.608 ± 0.009 0.774 ± 0.007 0.781 ± 0.006
s 0.40 ± 0.04 1.02 ± 0.04 0.87 ± 0.11 0.97 ± 0.22 0.69 ± 0.21 0.79 ± 0.12
N 1.3 ± 0.3 1.5 ± 0.4 2.0 ± 0.8 2.3 ± 0.8 1.3 ± 0.7 1.3 ± 0.5
O 0.44 ± 0.05 0.61 ± 0.06 0.53 ± 0.09 0.58 ± 0.10 0.61 ± 0.15 0.53 ± 0.07
Ne 0.31 ± 0.08 0.39 ± 0.18 0.33 ± 0.19 0.81 ± 0.22 1.31 ± 0.35 1.18 ± 0.22
Fe 0.52 ± 0.03 0.92 ± 0.06 0.75 ± 0.09 0.67 ± 0.08 0.87 ± 0.18 0.83 ± 0.08
[λ17/Iλ15]observed 2.04 ± 0.21 1.28 ± 0.13 1.99 ± 0.34 1.98 ± 0.29 1.25 ± 0.28 2.24 ± 0.34
[λ17/Iλ15]predicted 1.31 1.31 1.30 1.31 1.27 1.27
I15.01/I0 15.01 0.64 ± 0.07 1.02 ± 0.10 0.65 ± 0.11 0.66 ± 0.10 1.02 ± 0.23 0.57 ± 0.09
X-ray bright arm-like structures which might have been produced by shocks (Jones
et al. 2002). The core of NGC 5813 is highly disturbed with most of the hot gas being
pushed into dense sheets by the rising bubbles of relativistic plasma.
Currently the best data set for the study of resonant scattering is that for NGC 4636.
The X-ray halo of this galaxy has a conveniently low temperature, at which ∼26% of
Fe is in the form of Fe XVII, a very dense core and a deep XMM-Newton observation,
which also allows us to extract spectra from outside the core region with reasonably
good statistics in the Fe XVII lines. Using a deep Chandra observation of this galaxy
we derive deprojected density and temperature proﬁles which we use to model radial
intensity proﬁles for the strongest resonance lines, taking into account the effects of
resonant scattering for different assumed values of turbulent velocities.
5.3 Data analysis
5.3.1 XMM-Newton RGS data analysis
We processed the RGS data using version 8.0.0 of the XMM-Newton Science Analysis
System (SAS), and extracted spectra using the method described by Tamura et al.
(2001). To minimize the contamination by soft-protons from Solar ﬂares, we extracted
80 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
10 12 14 16 18 20
00.010.020.03
Counts/s/Å
wavelength (Å)
OVIII LαFeXVIINeX
FeXVIII
OVIII Lβ
FeXVII
FeXVIII
NGC 5813
FeXIX
FeXVIII
10 12 14 16 18 20
00.010.02
Counts/s/Å
wavelength (Å)
OVIII LαFeXVII
NeX
FeXVIII
OVIII Lβ
FeXVII
FeXVIII
NGC 1404
FeXIX
FeXVIII
10 12 14 16 18 20
00.010.02
Counts/s/Å
wavelength (Å)
OVIII LαFeXVIINeX
FeXVIII
OVIII Lβ
FeXVII
NGC 4649
FeXVIII
FeXIX
FeXVIII
FeXX
10 12 14 16 18 20
00.010.02
Counts/s/Å
wavelength (Å)
OVIII LαFeXVIINeX
FeXVIII
OVIII Lβ
FeXVII
NGC 4472
FeXVIII
FeXIX
FeXVIII
FeXX
Figure 5.4: RGS spectra for NGC 5813, NGC 1404, NGC 4649, and NGC 4472. The
spectra were extracted from 0.5 wide extraction regions centred on the cores of the
galaxies. The full line indicates the best-ﬁt optically-thin single-temperature plasma
model. The 13.8–15.5 ˚A part of the spectrum, where the strongest Fe XVII and Fe XVIII
resonance lines are present, was excluded from these spectral ﬁts.
5.3 Data analysis 81
a light curve for each dataset using events on CCD 9 of the RGS, outside the central
region, with a distance larger than 30 from the dispersion axis, and excluded time
intervals with elevated count rates. The original exposure time and the total time after
cleaning are listed in Table 5.1. The ﬂares in the observation of NGC 1404 were weak
and we obtain the same best ﬁt spectral parameters for both the ﬁltered and unﬁltered
data set. Therefore, for this galaxy we used the full observation. Because emission
from the galaxies ﬁlls the entire ﬁeld of view of the RGS, the background cannot be
estimated from a region on the detector away from the source. Therefore, we modeled
the background using the standard RGS background model for extended sources
available in SAS (rgsbkgmodel, Gonzal´ez-Riestra 2004).
The effects of resonant scattering are most signiﬁcant in the centermost parts of the
galaxies, where the density is highest. Therefore, we extract spectra from narrow 30
wide bands centred on the cores of the galaxies, probing regions within radii of 1.3–
2.2 kpc (depending on distance) in the cross dispersion direction of the RGS. We extract
and ﬁt both ﬁrst and second order spectra. Because the RGS operates without a slit, it
collects all photons from within the 0.5 × ∼12 ﬁeld of view. Line photons originating
at angle ∆θ (in arcminutes) along the dispersion direction, will be shifted in wavelength
by
∆λ = 0.138 ∆θ ˚A. (5.1)
Therefore, every line will be broadened by the spatial extent of the source. To account
for this broadening in our spectral model, for each RGS spectrum, we produce a predicted
line spread function (LSF) by convolving the RGS response with the surface
brightness proﬁle of the galaxy derived from the EPIC/MOS1 image in the 0.8–1.4 keV
band along the dispersion direction. Because the radial proﬁle of a particular spectral
line can be different from the overall radial surface brightness proﬁle (e.g. in the presence
of metallicity gradients), the line proﬁle is multiplied by a scale factor s, which is
the ratio of the observed LSF width to the expected LSF for a ﬂat abundance distribution.
This scale factor is a free parameter in the spectral ﬁt.
For the spectral modeling of the RGS data we use the SPEX package (Kaastra et al.
1996). Spectral ﬁtting of the RGS data is done in the 10 ˚A to 28 ˚A band. We model
the emission of the observed galaxies as optically thin plasmas in collisional ionization
equilibrium, absorbed by neutral Galactic gas with Solar abundances. For each object
we ﬁx the Galactic column density to the value determined by the Leiden/Argenitine/Bonn
(LAB) Survey of Galactic H I (Kalberla et al. 2005). For the emission of the elliptical
galaxies we assume a single-temperature plasma. Using more complicated multitemperature
models does not improve the ﬁts and differential emission measure models
always converge to a simple single-temperature approximation. The spectral normalization,
plasma temperature, and the abundances of N, O, Ne, and Fe are free parameters
in the ﬁt. For the spectral ﬁtting we use C-statistics.
5.3.2 Chandra analysis of NGC 4636
We reprocessed the Chandra observations of NGC 4636 (OBSID 3926 and 4415) applying
the latest CTI and time-dependent gain calibrations using standard methods (see
82 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
Vikhlinin et al. 2005, for more details). In brief, we performed the usual ﬁltering by
grade, excluded bad/hot pixels and columns, removed cosmic ray ‘afterglows’, and
applied the VF mode ﬁltering. We excluded data with anomalously high background.
The remaining exposure time for NGC 4636 is 150.4 ks. The background ﬁles were
processed in the same manner as the observations. An additional background component
is produced during the 41 ms of the 3.2 s nominal exposure readout of the ACIS
CCDs, while the chips are exposed to the sky. This small contribution of the source
ﬂux, 1.3%, is uniformly re-distributed along the readout direction and is subtracted
using the technique described by Markevitch et al. (2000). The blank-ﬁeld background
is also renormalized by reducing its integration time by 1.3% to account for this additional
subtraction.
As the ﬁnal step in data preparation (see Churazov et al. 2008, for more details), we
computed, for each X-ray event, the ratio of the effective area (a function of photon
energy and position) to that at a predeﬁned energy and position:
η = A(E, xd, yd)/A0(E), (5.2)
where A(E, xd, yd) includes mirror and detector efﬁciencies (including non-uniformity
of the detector quantum efﬁciency and the time and spatially dependent contamination
on the optical blocking ﬁlter). Finally, for each event list, we make an exposure map
that accounts for all position dependent, but energy independent, efﬁciency variations
across the focal plane (e.g., overall chip geometry, dead pixels or rows, and variation of
telescope pointing direction). We use these data in Sect. 5.5 to derive deprojected gas
density and temperature proﬁles.
5.4 Observations of resonant scattering
The results from the spectral ﬁtting of the XMM-Newton RGS data with an optically
thin single-temperature plasma model are shown in Table 5.2. The 13.8–15.5 ˚A part
of the spectrum, where the strongest Fe XVII and Fe XVIII resonance lines are present,
was initially excluded from this ﬁt. These lines are expected to be suppressed by resonant
scattering and this suppression, which is not accounted for in the plasma model,
could slightly bias the best ﬁt spectral parameters. After obtaining the best ﬁt, we
freeze the temperature and Fe abundance of the thermal component, set the abundance
of the Fe XVII ion to zero in the model, add narrow Gaussians at the wavelength
of the strongest Fe XVII lines (restframe wavelengths of 15.01 ˚A, 17.077 ˚A, 15.30 ˚A, and
16.80 ˚A) and re-ﬁt the full 10–28 ˚A spectral band. Freezing the temperature and the Fe
abundance is necessary because by setting the Fe XVII ion to zero and replacing it with
Gaussians in the model we lose important constraints on these parameters. The line
normalisations, the continuum level (normalisation of the thermal component), the
abundances of the other elements and the broadening factor s are free parameters. In
the error calculation, we also marginalise over the uncertainties of the temperature and
Fe abundance as determined in the initial ﬁt. We determine the best ﬁtting normalisations
for these Gaussians and the value of (Iλ17.05 + Iλ17.10)/Iλ15.01 (the 17.05 and 17.10
˚A lines are blended and we ﬁt them with a single Gaussian at 17.077 ˚A). We divide the
5.4 Observations of resonant scattering 83
10 12 14 16 18 20
00.010.020.030.04
Counts/s/Å
wavelength (Å)
OVIII Lα
FeXVII
NeX
FeXVIII
OVIII Lβ
FeXVII
NGC 4636
FeXVIII
FeXIX
FeXVIII
10 12 14 16 18 20
05×10−30.010.0150.02
Counts/s/Å
wavelength (Å)
OVIII LαFeXVIINeX
FeXVIII
OVIII Lβ
FeXVII
NGC 4636
FeXVIII
FeXIX
FeXVIII
Figure 5.5: XMM-Newton RGS spectra extracted from a 0.5 wide extraction region centered
on the core of NGC 4636 (left panel) and from two 2.25 wide extraction regions
surrounding the core (right panel). The ratios of the Fe XVII lines at 15.01 and ∼17 ˚A
change. The full line indicates the best ﬁt optically thin single-temperature plasma
model. The 13.8–15.5 ˚A part of the spectrum, where the strongest Fe XVII and Fe XVIII
resonance lines are present, was excluded from these spectral ﬁts.
theoretical ratio (Iλ17.05 + Iλ17.10)/Iλ15.01 by the measured value, which, assuming the
lines at ∼17 ˚A are optically thin, gives us the level at which the 15.01 ˚A line is suppressed,
(I/I0)15.01 ˚A
. These values are listed in Table 5.2. In four out of ﬁve galaxies
the line ratios are signiﬁcantly higher (observed values ∼ 2.0) than the expected theoretical
value for an optically thin plasma of 1.27–1.31, indicating that the 15.01 ˚A line is
optically thick. In NGC 4649, the line ratios are consistent with the 15.01 ˚A line being
optically thin.
A number of factors argue that the observed high (Iλ17.05 + Iλ17.10)/Iλ15.01 line ratio
is due to resonant scattering. The (Iλ17.05 + Iλ17.10)/Iλ15.01 line ratio in an optically thin
plasma can reach values ∼ 1.9 only for temperatures as low as 0.2 keV (see Fig. 5.1,
Doron & Behar 2002), which is much lower than the observed temperatures of the
galaxies. Moreover, if the plasma in the cores of the galaxies were multiphase with
an additional 0.2 keV component, which would alter the Fe XVII line ratios, then we
would also observe other strong lines (e.g. O VII), which we do not. Another possibility
to obtain such high line ratios in an optically thin plasma is non-equilibrium
ionization (NEI). However, NEI effects are unlikely in the dense plasma in cores of giant
elliptical galaxies because, due to the high density, the equilibration time is short.
If the higher-than-expected line ratios were due to incomplete atomic physics in the
current spectral models, then they should not change as a function of radius. This
can be directly veriﬁed in NGC 4636, which has the best combination of X-ray ﬂux,
temperature, and exposure. The data for this galaxy allow us to extract spectra and
determine the (Iλ17.05 + Iλ17.10)/Iλ15.01 line ratios in the combined spectrum from two
regions surrounding the core at 15 –150 from the centre of the galaxy, in the crossdispersion
direction of the RGS (see Fig. 5.2). We follow here the same ﬁtting procedure
as for the cores of the galaxies. The best ﬁtting results are shown in the third column of
Table 5.2. The off-centre line ratios are consistent with the 15.01 ˚A line being optically
84 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
Table 5.3: Oscillator strengths and optical depths for the strongest X-ray lines in the
spectrum of NGC 4636 for Mach numbers 0.0, 0.25, 0.5, 0.75. The expected suppression
due to resonant scattering (I/I0)cir gives the value integrated within a radius of 2 kpc
from the centre of the galaxy. The approximate values (I/I0)RGS were obtained by
integrating within an “effective extraction region” which is 0.5 wide and 3 long. The
metallicity is assumed to be constant with the radius.
Ion λ ( ˚A) f M = 0.0 M = 0.25 M = 0.5 M = 0.75
τ (I/I0)cir(I/I0)RGS τ (I/I0)cir(I/I0)RGS τ (I/I0)cir(I/I0)RGS τ (I/I0)cir(I/I0)RGS
Fe XVII 15.01 2.73 8.8 0.47 0.69 3.6 0.59 0.79 1.9 0.68 0.85 1.3 0.76 0.89
Fe XVII 17.05 0.12 0.5 0.89 0.95 0.2 0.95 0.98 0.1 0.97 0.99 0.07 0.98 0.99
Fe XVIII 14.20 0.57 1.3 0.72 0.87 0.5 0.86 0.94 0.3 0.91 0.96 0.2 0.95 0.98
Fe XVIII 16.08 0.005 0.01 0.99 1.00 0.005 0.99 1.00 0.003 1.00 1.00 0.002 1.00 1.00
O VIII Lα 18.96 0.28 1.2 0.74 0.87 0.8 0.81 0.91 0.5 0.88 0.94 0.3 0.91 0.96
thin in this region. The difference in the spectra of the core and surrounding regions
of NGC 4636 can be seen clearly in Fig. 5.5. While the 15 ˚A line is suppressed in the
centre, it may be slightly enhanced outside the dense core.
The best ﬁt line broadening scale factor s indicates that in NGC 4636 the distribution
of the line producing ions is more peaked than the surface brightness, suggesting that
the distribution of metals is centrally peaked. In the other systems this effect is smaller
and less signiﬁcant.
5.5 Modelling of resonant scattering in NGC 4636
To gain a better understanding of the observed line ratios, we simulate the effects of
resonant scattering on the radial intensity proﬁles of the strongest X-ray emission lines
in NGC 4636. For this we use the deprojected temperature and density proﬁles measured
with Chandra.
In our deprojection analysis, we follow the approach described by Churazov et al.
(2008). We assume spherical symmetry, but make no speciﬁc assumption about the
form of the underlying gravitational potential. For a given surface brightness proﬁle
in na annuli, we choose a set of ns (ns ≤ na) spherical shells with the inner radii
r(i), i = 1, . . . , ns. The gas emissivity E is assumed to be uniform inside each shell, except
for the outermost shell, where the gas emissivity is assumed to decline as a power
law of radius: E = Eoutr−6βout, where βout is a parameter. The deprojection process
is a simple least square solution to determine the set of emissivities in the set of shells
(along with the emissivity normalization Eout of the outer layers) that provides the best
description of the observed surface brightness. The emissivity of each shell can then be
evaluated as an explicit linear combination of the observed quantities. Since the whole
procedure is linear, the errors in the observed quantities can be propagated straightforwardly.
With our deﬁnition of η (see subsection 5.3.2), the projection matrix does not
depend on energy. Therefore, the deprojection in any energy band is precisely that used
to deproject the surface brightness. Thus, we can accumulate a set of spectra (corrected
for background and readout) for each of the na annuli, and apply the deprojection to
5.5 Modelling of resonant scattering in NGC 4636 85
Figure 5.6: Observed radial proﬁles of the electron density and temperature used to
model resonant scattering in NGC 4636. The blue circles and red squares indicate the
deprojected proﬁles determined from Chandra data with metallicity as a free parameter
in the ﬁt and with metallicity ﬁxed to 0.68 Solar, respectively. The magenta triangles
show the projected radial temperature proﬁle with the metallicity as a free parameter in
the ﬁt. The solid lines show the parametrization of the gas density and gas temperature
proﬁles given in eqs. 5.3 and 5.4.
86 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
determine the emissivities of each shell in each of the ACIS energy channels.
The Chandra data were modeled using XSPEC V12 (Arnaud 1996) and the APEC
model (Smith et al. 2001). The gas temperature and normalization were free parameters
in the model. The heavy element abundance was either a free parameter in the ﬁt or
ﬁxed to 0.68 Solar (as determined from a spectral ﬁt to Chandra data extracted within
the radius of 1 from the centre of NGC 4636). The metallicity proﬁle in NGC 4636
has large uncertainties, because the metal abundance distribution in the deprojected
Chandra spectra has a large scatter as a result of noise enhancement during deprojection
coupled with the limited spectral resolution of the CCDs.
The electron density proﬁle was derived from the spectral normalization, ﬁxing the
proton to electron ratio to 0.83. Fig. 5.6 shows the deprojected electron density (lower
panel) and the temperature (upper panel) as a function of distance from the centre of
the galaxy. The circles and squares show the deprojected proﬁles for metallicity as a
free parameter and for metallicity ﬁxed to 0.68 Solar, respectively. The triangles show
the projected temperature proﬁle with the metallicity as a free parameter.
Based on the observed radial distributions of electron density ne and temperature
kTe in NGC 4636, we adopt the following approximate forms for the deprojected density
and temperature proﬁles:
ne = 1.2 × 10−1
1 +
r
0.25
2 −0.75
cm−3
, (5.3)
and
kT = 0.56
1 + 1.4(r/1.2 )4
1 + (r/1.2 )4
keV. (5.4)
We consider two possible radial behaviors for metallicity: a simple constant abundance
of 0.68 Solar; and a centrally peaked abundance distribution, which is more consistent
with the best ﬁt XMM-Newton RGS line proﬁle (s = 0.40 ± 0.04). In the latter
case we approximate the metallicity proﬁle with the following function:
Z = 0.95
2 + (r/0.8 )3
1 + (r/0.8 )3
− 0.4. (5.5)
We calculate the optical depth of the 15 ˚A line from the centre of the galaxy to inﬁnity,
τ = niσ0dr, where ni is the ion concentration and σ0 is the cross section at the line
centre, which for a given ion is
σ0 =
√
πhrecs f
∆ED
, (5.6)
where the Doppler width is given by
∆ED = E0
2kTe
Ampc2
s
+
V2
turb
c2
s
1/2
. (5.7)
5.5 Modelling of resonant scattering in NGC 4636 87
In these equations re is the classical electron radius, f is the oscillator strength of a
given atomic transition, E0 is the rest energy of a given line, A is the atomic mass of the
corresponding element, mp is the proton mass, cs is the sound speed, and Vturb is the
characteristic velocity of isotropic turbulence. Vturb ≡
√
2V1D,turb, where V1D,turb is the
velocity dispersion in the line of sight due to turbulence. Using the adiabatic sound
speed, cs = γ k T /µ mp, the expression for the broadening can be rewritten as
∆ED = E0
2kTe
Ampc2
s
(1 +
γ
2 µ
AM2
)
1/2
, (5.8)
where µ = 0.6 is the mean particle mass, γ is the adiabatic index, which for ideal
monatomic gas is 5/3, and M = Vturb/cs is the corresponding Mach number. The line
energies and oscillator strengths were taken from ATOMDB2 and the NIST Atomic
Spectra Database3.
The resonant scattering effect has been calculated using Monte-Carlo simulations, as
described by Churazov et al. (2004). We model the hot halo as a set of spherical shells
and obtain line emissivities for each shell using the APEC plasma model (Smith et al.
2001). We account for scattering by assuming a complete energy redistribution and
dipole scattering phase matrix. The line of Fe XVII has a pure dipole scattering phase
matrix. In Fig. 5.7 we show the optical depth of the 15.01 ˚A Fe XVII line as a function of
radius, for various turbulent velocities, assuming the peaked Fe abundance distribution.
For higher turbulent velocities, the optical depth becomes smaller. In Fig. 5.8 we
show the ratio of the 15.01 ˚A line intensity calculated including the effect of resonant
scattering to the line intensity without accounting for resonant scattering for various
turbulent velocities, both for constant and peaked iron abundance distributions. The
ﬁgure clearly shows the effect of the resonant scattering: in the core of the galaxy the
intensity of the line is suppressed, whereas in the surrounding regions the intensity
of the line rises. The effect is more prominent in the case of the centrally peaked Fe
abundance distribution, where the central suppression is larger and the enhancement
in the outer part is stronger.
We compare the models of resonant scattering with the observations and determine
the systematic uncertainties involved by simulating spectra corresponding to each of
these models, which are then ﬁt in the same way as the observed data. To this end, we
multiply the surface brightness proﬁle of NGC 4636 with the theoretical suppression
proﬁles for the 15.01 ˚A line (Fig. 5.8) to determine the predicted RGS line proﬁle for
the four considered characteristic turbulent velocities, for both assumed abundance
distributions. We simulate spectra with photon statistics comparable to that of the actual
observation, using the best ﬁt values for the core of NGC 4636 from Table 5.2 as
input parameters, and convolving the 15.01 ˚A line through these predicted line proﬁles.
By ﬁtting the data simulated with no turbulence, we obtain (I/I0)15.01 ˚A
= 0.69
and 0.71 for peaked and ﬂat abundance proﬁles, respectively. The ratios of (I/I0)15.01 ˚A
for isotropic turbulent velocity of M = 0.25 are 0.76 and 0.78, and for M = 0.50 they
2http://cxc.harvard.edu/atomdb/WebGUIDE/index.html
3www.physics.nist.gov/PhysRefData/ASD/index.html
88 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
are 0.80 and 0.82, for peaked and ﬂat abundance proﬁles, respectively. The systematic
uncertainty on these values due to our lack of knowledge about the actual Fe abundance
distribution is ±0.02. We add this uncertainty in quadrature to the uncertainties
associated with our modeling of the line broadening, which is at most 0.03, resulting
in a total uncertainty of 0.04. Thus we conclude that the turbulent velocities in
NGC 4636 are relatively small and that isotropic turbulence with a characteristic velocity
of M > 0.25 can be ruled out at the 90% conﬁdence level; turbulent gas motions at
M > 0.5 can be ruled out at the 95% conﬁdence level. (The sound speed in NGC 4636
is cs ∼ 400 km s−1.)
In Table 5.3, we list the strongest lines observed in the spectrum of NGC 4636, their
oscillator strengths and optical depths for different Mach numbers. Clearly, Fe XVII at
15.01 ˚A is the most optically thick line in the spectrum, but O VIII Lα and Fe XVIII at
14.2 ˚A also have relatively large optical depths. For M = 0, the depth of the Fe XVII line
at 17.05 ˚A line is τ = 0.5, meaning that our assumption that the 17.05 ˚A and 17.10 ˚A line
blend is optically thin is not completely correct, and our observed values of (I/I0)15.01 ˚A
are biased slightly high, which further increases the signiﬁcance of our upper limits
on the turbulent velocities. In Table 5.3, we also list the predicted suppression for the
strongest emission lines, (I/I0)circ = IdA / I0dA, integrated within a circular region
of radius 2 kpc, and within the RGS extraction region (I/I0)RGS. The suppression
for the RGS is determined in a simpliﬁed way, integrating within our 0.5 wide, and
for line emission effectively about 3 long, RGS extraction region. These predicted ratios
in Table 5.3 were determined assuming a ﬂat abundance distribution. By adopting
an optically thin plasma in the analysis, the best ﬁtting O abundance within the 2 kpc
radius can be underestimated by ∼10–25%, and the Fe abundance by ∼10–20%.
5.6 Discussion and conclusions
High-resolution spectra obtained with XMM-Newton RGS reveal that the Fe XVII line
at 15.01 ˚A in the cores of the elliptical galaxies NGC 4636, NGC 1404, NGC 5813, and
NGC 4472 is suppressed by resonant scattering. The effects of resonant scattering can
be investigated in detail for NGC 4636. The comparison of the measured suppression
of the 15.01 ˚A Fe XVII line in the spectrum of NGC 4636 with the simulated effects of
resonant scattering reveals that the characteristic velocity of isotropic turbulence in the
core of the galaxy is smaller than 0.25 and 0.5 of the sound speed at the ∼90% and 95%
conﬁdence levels, respectively. The energy density in turbulence is:
turb =
3
2
ρ V2
1D,turb =
3
4
ρ c2
s M2
(5.9)
where ρ is the density of the plasma. The thermal energy of the plasma is given as:
therm =
3
2
ρ
µmp
kT. (5.10)
By combining these equations we obtain the ratio of turbulent to thermal energy:
turb
therm
=
γ
2
M2
. (5.11)
5.6 Discussion and conclusions 89
Figure 5.7: Optical depth of the 15.01 ˚A line calculated from the given radius to inﬁnity,
for isotropic turbulent velocities corresponding to Mach numbers 0.0, 0.25, 0.5, and
0.75. A centrally peaked Fe abundance distribution is assumed.
90 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
Figure 5.8: Radial proﬁles of the ratio of the 15.01 ˚A line intensity with and without the
effect of resonant scattering, for isotropic turbulent velocities corresponding to Mach
numbers 0.0, 0.25, 0.5, and 0.75. The upper panel shows the proﬁles for a constant
metallicity, and the lower panel for a centrally peaked Fe abundance distribution.
5.6 Discussion and conclusions 91
0.1 1 10
11.52
kT(keV)
Radius (kpc)
NGC4649
Figure 5.9: The deprojected temperature proﬁle of NGC 4649 determined using Chandra
data. It shows an unusual central peak.
For the 90% and 95% conﬁdence level upper limits of M 0.25 and 0.5 on the characteristic
velocity of isotropic turbulence in NGC 4636, the upper limit on the fraction of
energy in turbulent motions is 5% and 20%, respectively. This is in agreement with the
upper limits on the non-thermal pressure component in elliptical galaxies obtained on
larger spatial scales by Churazov et al. (2008) and with the properties of AGN induced
turbulence in three-dimensional simulations, that use adaptive mesh hydrodynamic
code with a subgrid model of turbulence and mixing (Scannapieco & Br¨uggen 2008).
A small value for the average turbulent velocity dispersion was also inferred in the
initial analysis of the RGS spectra for NGC 4636 by Xu et al. (2002). They estimated
that the turbulent velocities in the galaxy do not strongly exceed 1/10 of the sound
speed. Caon et al. (2000) studied the kinematics of line emission nebulae in NGC 4636
and found an irregular radial velocity curve which indicates turbulent motions in the
core. The measured radial velocities of the emission line nebulae are mostly within
about ∼100 km s−1. The kinematics of emission line nebulae in the sample of elliptical
galaxies studied by Caon et al. (2000) indicates chaotic gas motions with velocities
of about M = 0.2–0.4. In cooling cores, optical emission line nebulae often spatially
coincide with soft X-ray emission (e.g. Sparks et al. 2004; Fabian et al. 2006), which indicates
a coupling, suggesting that the two gas phases should share the same turbulent
velocities.
The dense cores of NGC 4636 and NGC 5813 show strong evidence of recent violent
interaction between the hot plasma and the AGN. This interaction is expected to induce
turbulence through the mechanical activity of the outﬂows (Churazov et al. 2001). Our
92 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
results indicate that the typical velocities of the AGN induced turbulent motions in the
hot plasma on sub-kpc spatial scales are less than 100–200 km s−1. Gas motions on
larger spatial scales likely have larger velocities. Observations of the Perseus cluster
suggest both turbulent and laminar gas motions with velocities as high as 700 km s−1
(Fabian et al. 2003b,a). The observed lack of resonant scattering in the 6.7 keV Fe-K
line within the inner ∼100 kpc (Churazov et al. 2004) of Perseus conﬁrms the presence
of such strong gas motions. AGN induced gas motions with large range of velocities
along our line of sight, perhaps even on spatial scales smaller than ≈1 kpc,
could explain the observed lack of resonant scattering in the 15.01 ˚A Fe XVII line in
M 87 (the analysis of a 0.5 wide extraction region centred on the core of M 87 gives
(Iλ17.05 + Iλ17.10)/Iλ15.01 = 1.3 ± 0.3, Werner et al. 2006). We note that the central AGN
in Perseus and in M 87 are more active in X-rays than the AGN in the target galaxies
of this study.
The dense cores of galaxies showing smaller disturbance by AGN might be expected
to exhibit less turbulence than the heavily disturbed systems. Therefore, we might expect
the 15.01 ˚A line to be more suppressed in the most relaxed galaxies in our sample.
While there is a weak indication of the suppression increasing from the least to the
most relaxed systems: from NGC 5813 and NGC 4636, to NGC 4472, the large errorbars
on the line ratios and the small number of galaxies in the sample do not allow us
to draw ﬁrm conclusions.
The best ﬁt line ratio in the apparently relaxed galaxy NGC 4649 shows no indication
for resonant scattering in its core. The lack of observable optical depth effects in this
galaxy can be explained by the unusual temperature proﬁle of its gaseous halo, which
exhibits a sharp temperature peak in its centre (see Fig. 5.9, Allen et al. in prep.).
Within the central 0.1 kpc of the galaxy, the temperature increases by about a factor of
2 to ∼1.5 keV, at which no Fe XVII is present in the gas, implying that all the observed
Fe XVII emission is from lower density regions surrounding the hot core. The high
temperature might be due to shock heating by the central AGN (e.g. Shurkin et al.
2008). None of the other galaxies in our sample shows such an unusual temperature
proﬁle (Allen et al. 2006, Allen et al. in prep.).
Unfortunately, the high temperatures of cluster cores do not allow us to use Fe XVII
lines as a diagnostic tool to measure the effects of resonant scattering and thus place
ﬁrm constraints on turbulent velocities in the intra-cluster medium. No other lines
have a comparable sensitivity to the turbulent velocities. However, the dense cores
of elliptical galaxies, are in terms of the dominant physical processes like cooling and
AGN feedback, sufﬁciently similar to cooling cores of clusters of galaxies to argue that
the turbulent velocities on small spatial scales, and thus the turbulent pressure support
in these systems, are also likely to be low. This is encouraging for cosmological studies
based on X-ray observations of galaxy clusters (e.g. Allen et al. 2008; Mantz et al. 2008;
Henry et al. 2008; Vikhlinin et al. 2008). Simulations predict that turbulent gas motions
in the intra-cluster medium can provide 5%–25% of the total pressure support through
the virial region of clusters, potentially biasing low hydrostatic mass estimates, even
in relaxed systems (e.g. Nagai et al. 2007). However, these predictions depend sensitively
on the gas physics assumed. Our results argue that such turbulence may be
suppressed in nature, at least within the inner regions of such systems. For the merg-
5.6 Discussion and conclusions 93
ing unrelaxed Coma cluster, a ∼10% lower limit on turbulent pressure support was
obtained by Schuecker et al. (2004) from a Fourier analysis of pressure maps. In the
range 40–90 kpc, the turbulent spectrum was found to be well described by a projected
Kolmogorov/Oboukhov-type model.
Future, deeper (of the order of ∼500 ks) observations of large nearby elliptical galaxies
with the XMM-Newton RGS should provide even tighter constraints on turbulent
velocities from resonant scattering. X-ray calorimeters on future missions such as
Astro-H and IXO will enable direct measurements of the velocity-broadening of emission
lines in the hot gas in galaxies, groups, and clusters of galaxies (Sunyaev et al.
2003; Inogamov & Sunyaev 2003; Rebusco et al. 2008), thereby allowing an independent
check on our results as well as a critical extension to higher mass systems. However,
we note that even the excellent (2–5 eV) spectral resolution of these instruments
will not allow measurements of turbulent motions with velocities signiﬁcantly smaller
than 100–200 km s−1. Even though, the broad point spread function of the Astro-H mirrors
will prohibit investigations of the effects of resonant scattering on spatial scales
smaller than presented in this paper, it will for the ﬁrst time allow us to directly measure
the gas motions and spatially map them out to larger radii. The next planned
X-ray observatory able to produce a real breakthrough by mapping the gas motions in
groups and clusters of galaxies at high spatial resolution, will be IXO.
An important consequence of resonant scattering is the bias it can introduce in measurements
of radial proﬁles of abundances of chemical elements. As we show in Table
5.3, the integrated emission of several strong resonance lines is signiﬁcantly suppressed
within the inner 2 kpc of NGC 4636. By ignoring the effects of resonant scattering,
the Fe and O abundance will be underestimated by 10–20% within this region.
This conclusion is in line with the results of Sanders & Fabian (2006). However, while
resonant scattering makes abundances appear to be 10–20% lower than the real value,
it cannot explain the much stronger abundance dips seen in the cores of several clusters
of galaxies (e.g. Sanders & Fabian 2002).
Acknowledgements
Support for this work was provided by the National Aeronautics and Space Administration
through Chandra Postdoctoral Fellowship Award Number PF8-90056 issued by
the Chandra X-ray Observatory Center, which is operated by the Smithsonian Astrophysical
Observatory for and on behalf of the National Aeronautics and Space Administration
under contract NAS8-03060. The work of EC is supported by the DFG grant
CH389/3-2. SWA acknowledges support from Chandra grant AR7-8007X. This work
is based on observations obtained with XMM-Newton, an ESA science mission with
instruments and contributions directly funded by ESA member states and the USA
(NASA). This work was supported in part by the U.S. Department of Energy under
contract number DE-AC02-76SF00515.
94 Constraints on turbulent pressure in the X-ray halos of giant elliptical galaxies
Chapter 6
Deep Chandra observation and
numerical studies of the nearest cluster
cold front in the sky
N. Werner1,2, J.A. ZuHone3, I. Zhuravleva1,2, Y. Ichinohe4,5, A. Simionescu4, S.W. Allen1,2,6,
M. Markevitch7, A.C. Fabian8, U. Keshet9, E. Roediger10,11, M. Ruszkowski12,13, J.S. Sanders14
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall, Stanford, CA 94305-4085, USA
2Department of Physics, Stanford University, 382 Via Pueblo Mall, Stanford, CA 94305-
4060, USA
3MIT Kavli Institute for Astrophysics and Space Research, Massachusetts Institute of
Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139, USA
4Institute of Space and Astronautical Science (ISAS), JAXA, 3-1-1 Yoshinodai, Chuo-ku,
Sagamihara, Kanagawa, 252-5210, Japan
5Department of Physics, Graduate School of Science, University of Tokyo, 7-3-1 Hongo,
Bunkyo, Tokyo 113-0033, Japan
6SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025,
USA
7X-ray Astrophysics Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD
20771, USA
8Institute of Astronomy, Madingley Road, Cambridge CB3 0HA
9Physics Department, Ben-Gurion University of the Negev, Be’er-Sheva 84105, Israel
10Hamburger Sternwarte, Universit¨at Hamburg, Gojensbergsweg 112, D-21029 Hamburg,
Germany
11Dublin Institute for Advanced Studies, Astronomy and Astrophysics Section, 31 Fitzwilliam
Place, Dublin 2, Ireland
12Department of Astronomy, University of Michigan, 500 Church Street, Ann Arbor,
MI 48109, USA
13Michigan Center for Theoretical Physics, 3444 Randall Lab, 450 Church St, Ann Arbor,
MI 48109, USA
96 Resolving the nearest cold front in the sky
14Max-Planck-Institut f¨ur extraterrestrische Physik, Giessenbachstrasse 1, 85748 Garching,
Germany
Published in the Monthly Notices of the Royal Astronomical Society, volume 455,
pages 846–858, 2016
Abstract
We present the results of a very deep (500 ks) Chandra observation, along with tailored
numerical simulations, of the nearest, best resolved cluster cold front in the sky, which
lies 90 kpc (19 ) to the northwest of M 87. The northern part of the front appears the
sharpest, with a width smaller than 2.5 kpc (1.5 Coulomb mean free paths; at 99 per
cent conﬁdence). Everywhere along the front, the temperature discontinuity is narrower
than 4–8 kpc and the metallicity gradient is narrower than 6 kpc, indicating that
diffusion, conduction and mixing are suppressed across the interface. Such transport
processes can be naturally suppressed by magnetic ﬁelds aligned with the cold front.
Interestingly, comparison to magnetohydrodynamic simulations indicates that in order
to maintain the observed sharp density and temperature discontinuities, conduction
must also be suppressed along the magnetic ﬁeld lines. However, the northwestern
part of the cold front is observed to have a nonzero width. While other explanations
are possible, the broadening is consistent with the presence of Kelvin-Helmholtz instabilities
(KHI) on length scales of a few kpc. Based on comparison with simulations, the
presence of KHI would imply that the effective viscosity of the intra-cluster medium is
suppressed by more than an order of magnitude with respect to the isotropic Spitzerlike
temperature dependent viscosity. Underneath the cold front, we observe quasilinear
features that are ∼ 10 per cent brighter than the surrounding gas and are separated
by ∼ 15 kpc from each other in projection. Comparison to tailored numerical
simulations suggests that the observed phenomena may be due to the ampliﬁcation of
magnetic ﬁelds by gas sloshing in wide layers below the cold front, where the magnetic
pressure reaches ∼ 5–10 per cent of the thermal pressure, reducing the gas density between
the bright features.
6.1 Introduction
One of the ﬁrst important results obtained with the Chandra X-ray Observatory was
the surprising discovery of cold fronts - remarkably sharp surface brightness discontinuities,
with relatively dense and cool gas on the inner (bright) side and low density
hot gas on the outer (faint) side. The ﬁrst discovered cold fronts in Abell 2142 and
Abell 3667 were interpreted as signs of mergers (Markevitch et al. 2000; Vikhlinin et al.
2001a), where the sharp surface brightness discontinuity separates the hotter intracluster
medium (ICM) from the low entropy core of a merging subcluster.
6.1 Introduction 97
However, as more clusters were observed with Chandra, cold fronts were also discovered
in the centers of cool-core clusters (at r =10–400 kpc) with no signs of recent
major mergers (Markevitch et al. 2001). Indeed, such cold fronts are observed in most
relatively relaxed cool-core clusters (Ghizzardi et al. 2010), often with several, oppositely
placed arcs curved around the central gas density peak, forming a spiral-like
pattern (see e.g. Markevitch & Vikhlinin 2007; Owers et al. 2009; Simionescu et al. 2012;
Rossetti et al. 2013; Ichinohe et al. 2015). Considering projection effects and the weak
density jumps, possibly all cool-core clusters may host some cold fronts.
Cold fronts in cooling cores are believed to be due to the sloshing (or swirling) of
the gas in the gravitational potential of the cluster. Numerical simulations show that
sloshing can be induced easily by a minor merger where a subcluster falls in with a
nonzero impact parameter (Tittley & Henriksen 2005; Ascasibar & Markevitch 2006).
Due to this disturbance, the density peak of the main cluster swings on a spiral-like
trajectory relative to its center of mass. As the central ICM is displaced from the corresponding
dark matter peak, opposite and staggered cold fronts are created where
cooler and denser parcels of gas from the centre come into contact with the hotter gas
at larger radii. The large-scale coherent ﬂows associated with sloshing and the corresponding
cold fronts can persist for Gyrs. By redistributing the ICM, sloshing plays
an important role in shaping the cluster cores and may even help prevent signiﬁcant
cooling ﬂows (ZuHone et al. 2010; Keshet 2012).
Markevitch et al. (2001) showed that sloshing cold fronts require signiﬁcant acceleration
of the gas beneath the contact discontinuity. This can be explained as centripetal
acceleration of a tangential bulk ﬂow just below the cold front (Keshet et al.
2010), which causes a strong tangential shear along the discontinuity. This strong shear
ﬂow should promote the growth of hydrodynamic instabilities, which would broaden
and eventually destroy the cold front on short sub-dynamical time scales. But cold
fronts are remarkably sharp, both in terms of the density and the temperature jumps:
the gas density discontinuity in e.g. the merging cluster Abell 3667 is at least several
times narrower than the Coulomb mean free path (Vikhlinin et al. 2001a; Markevitch
& Vikhlinin 2007) and the observed temperature jumps require thermal conduction
across cold fronts to be suppressed by a factor of order ∼ 102 compared to the collisional
Spitzer or saturated values (Ettori & Fabian 2000; Xiang et al. 2007). A layer of
magnetic ﬁeld parallel to the discontinuity could stabilize the front, prevent the growth
of instabilities, and suppress transport across the cold front (Vikhlinin et al. 2001b).
Vikhlinin et al. (2001b) and Vikhlinin & Markevitch (2002) used the stability of the
merger induced cold front in Abell 3667 to derive a lower limit on the magnetic ﬁeld
of B >7–16 µG at the discontinuity. Magnetic ampliﬁcation in sloshing cold fronts is
achieved through the shear due to the bulk ﬂow of plasma beneath the front, which
naturally sustains the strong magnetic ﬁelds parallel to cold fronts (Keshet et al. 2010;
ZuHone et al. 2011). In the layers underneath cold fronts, ﬁelds with initial strengths of
0.1–1 µG may be ampliﬁed to tens of µG, and the magnetic pressures may be ampliﬁed
to values reaching a signiﬁcant fraction of the thermal pressure, leading to gas depleted
bands (ZuHone et al. 2011).
One of the most interesting aspect of cold fronts is that they provide a relatively
clean experimental setup. This allows us to address questions about the physics of the
98 Resolving the nearest cold front in the sky
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
50 kpc
Figure 6.1: XMM-Newton EPIC/MOS mosaic image of the central r ∼ 150 kpc region
of the Virgo cluster. The exposure corrected image has been divided by the best-ﬁtting
radially symmetric beta model (see Simionescu et al. 2010; Roediger et al. 2011). The
white square indicates our Chandra ACIS-I pointing shown in Fig. 2. The “+” sign indicates
the center of M 87 at the bottom of the global gravitational potential of the Virgo
Cluster. The “X” sign indicates the center of curvature (RA: 187.676, DEC: 12.5064) of
the part of the cold front imaged by our Chandra pointing.
ICM in a more exact way than in cluster cores with complex astrophysical substructure.
The closest example of a cluster cold front is the spectacular surface brightness
discontinuity that lies 90 kpc (19 ) to the north of M 87 in the Virgo Cluster and extends
140◦ in azimuth (see Fig. 6.1, Simionescu et al. 2010). At the distance of 16.1 Mpc (Tonry
et al. 2001), 1 arcsec corresponds to only 78 pc, making this the best resolved cluster
cold front in the sky. The gas density jump across the discontinuity is ρin/ρout ≈ 1.4
and the gas in the inner, bright side is ∼40% more abundant in Fe than the ICM outside
the front (Simionescu et al. 2010). Simionescu et al. (2007) also detected a much weaker
cold front ∼ 17 kpc to the south-east of the core. The opposite and staggered placement
of these fronts, as well as the absence of a remnant sub-cluster, make gas sloshing
the most viable explanation for the presence of these edges (Tittley & Henriksen 2005;
Ascasibar & Markevitch 2006).
Using constrained hydrodynamical simulations, Roediger et al. (2011) showed that
a sloshing scenario can reproduce the radii and the contrasts in X-ray brightness, projected
temperature, and metallicity across the observed cold fronts in the Virgo Cluster.
By comparing synthetic and real observations, they estimate that the original minor
merger event that triggered the sloshing took place about 1.5 Gyr ago, when a subcluster
of 2–4 × 1013 M passed the Virgo core at 100 to 400 kpc distance nearly perpendicular
to our line-of-sight. The resulting gas sloshing, with motions primarily in
the plane of the sky, creates a contact discontinuity with a surface that is predominantly
perpendicular to our line of sight, maximizing the contrast and minimizing projection
6.2 Observations and data analysis 99
Table 6.1: Summary of the Chandra observations. Columns list the observation ID,
observation date, and the exposure after cleaning.
Obs. ID Obs. date Exposure (ks)
15178 2014 Feb. 17 47.1
15179 2014 Feb. 24 41.9
15180 2013 Aug. 1 140.5
16585 2014 Feb. 19 45.6
16586 2014 Feb. 20 49.8
16587 2014 Feb. 22 37.8
16590 2014 Feb. 27 38.1
16591 2014 Feb. 27 23.8
16592 2014 Mar. 1 36.1
16593 2014 Mar. 2 38.1
effects. In subsequent work, Roediger et al. (2013) demonstrated the capability of deep
Chandra observations to reveal Kelvin-Helmholtz instabilities in the Virgo cold front
by detecting multiple edges and shallower than expected surface brightness proﬁles
across the discontinuity. Based on relatively short XMM-Newton observations, they
also present tentative indications that the northeastern portion of the front is disturbed
by instabilities.
Here, we present a very deep (500 ks) Chandra observation of the sharpest, northwestern
part of the Virgo cold front, designed to test the theoretical predictions in unprecedented
detail.
6.2 Observations and data analysis
6.2.1 Chandra data
The Chandra X-ray observatory observed the prominent cold front to the northwest of
M 87 (see Fig. 6.1) between 2013 August and 2014 March using the Advanced CCD
Imaging Spectrometer (ACIS-I). The standard level-1 event lists were reprocessed using
the CIAO (version 4.7, CALDB 4.6.5) software package, including the latest gain
maps and calibration products. Bad pixels were removed and standard grade selections
applied. The data were cleaned to remove periods of anomalously high background.
The observations, along with the dates, identiﬁers, and net exposure times
after cleaning (total net exposure of 500 ks), are listed in Table 1. Background images
and spectra were extracted from the blank-sky ﬁelds available from the Chandra X-ray
Center. These were cleaned in an identical way to the source observations, reprojected
to the same coordinate system and normalized by the ratio of the observed to blank-sky
count rates in the 9.5–12 keV band.
100 Resolving the nearest cold front in the sky
1.78 2.11 3.45 8.77 30.0 114
10 kpc
0.9 0.96 1.02 1.08 1.14 1.2 1.26 1.32 1.38 1.44 1.5
Quasi-linear features
10 kpc
Figure 6.2: Left panel: Chandra image of the cold front in the 0.8–4.0 keV energy band.
The image was smoothed with a Gaussian function with a 1.5 arcsec window. Right
panel: The residual image obtained by dividing the image on the left with a beta-model,
and patched on large scales with an 80 arcsec scale smoothing window (see Zhuravleva
et al. 2015, and the text for details), reveals three X-ray bright quasi-linear features
separated from each other by ∼ 15 kpc in projection. The dark band outside the cold
front is an image processing artifact due to the sharp surface brightness discontinuity.
Point sources were removed from the residual image.
6.2 Observations and data analysis 101
Image analysis
Background-subtracted images were created in nine narrow energy bands (0.8–1.0 keV,
1.0–1.2 keV, 1.2–1.5 keV, 1.5–2.0 keV, 2.0–2.5 keV, 2.5–2.75 keV, 2.75–3.0 keV, 3.0–3.5 keV,
and 3.5–4.0 keV), spanning 0.8–4.0 keV. The nine narrow-band images were ﬂat ﬁelded
with respect to the median energy for each image and then co-added to create the X-ray
images shown in Fig. 6.2. In the 0.8–4.0 keV band the signal over background ratio is
optimal and the emissivity is nearly independent of the gas temperature, maximizing
the contrast at which we detect the contact discontinuity. Identiﬁcation of point sources
was performed using the CIAO task WAVDETECT. The point sources were excluded
from the subsequent analysis.
In order to emphasize structure in the image, we also produced a beta-model subtracted
image, shown in the right panel of Fig. 6.2. In order to remove the underlying
large-scale surface brightness gradient, we ﬁtted a beta-model to an azimuthally
averaged X-ray surface brightness proﬁle extracted from the XMM-Newton mosaic
image (see Sect. 6.2.2 and Fig. 6.1) and divided our Chandra image by this model,
patched on large scales with an 80 arcsec smoothing window (see Zhuravleva et al.
2015, for details). Our patched model with which we divide our image is deﬁned as
Ipm = IβSσ[IX/Iβ], where Iβ is the best ﬁt beta-model to the azimuthally averaged Xray
surface brightness proﬁle, Sσ[.] is Gaussian smoothing with window size σ, and IX
is the X-ray surface brightness image.
Spectral analysis
Individual regions for the 2D spectral mapping were determined using the contour
binning algorithm (Sanders 2006), which groups neighbouring pixels of similar surface
brightness until a desired signal-to-noise ratio threshold is met. We adopted a
signal-to-noise ratio of 75 (approximately 5600 counts per region), which provides approximately
3 per cent precision for density, 4 per cent precision for temperature, and
10–20 per cent precision for metallicity measurements, respectively. We modelled the
spectra extracted from each spatial region with the SPEX package (Kaastra et al. 1996).
The spectral ﬁtting was performed in the 0.6–7.5 keV band. The spectrum for each
region was ﬁtted with a model consisting of an absorbed single-phase plasma in collisional
ionization equilibrium, with the spectral normalization (emission measure),
temperature, and metallicity as free parameters. The deprojected spectra are obtained
following the method described by Churazov et al. (2003) and are ﬁtted using XSPEC
and the APEC plasma model (Smith et al. 2001; Foster et al. 2012). The line-of-sight
absorption column density was ﬁxed to the value, NH = 2 × 1020 cm−2, determined by
the Leiden/Argentine/Bonn radio survey of H I (Kalberla et al. 2005). Metallicities are
given with respect to the proto-Solar abundances by Lodders (2003).
6.2.2 XMM-Newton data
A mosaic of six XMM-Newton pointings, each with an exposure time of ∼ 20 ks, was
performed in June 2008. An additional pointing was obtained in November 2008.
102 Resolving the nearest cold front in the sky
Rf = 59.76±0.37
nj = 1.48±0.07
𝜒2 = 1.16r
Rf = 59.72±0.09
nj = 1.62±0.07
𝜒2 = 1.50r
Rf = 57.42±0.20
nj = 1.43±0.04
𝜒2 = 2.12r
Rf = 57.38±0.47
nj = 1.39±0.04
𝜒2 = 1.37r
Rf = 57.75±0.28
nj = 1.42±0.05
𝜒2 = 1.71r
Rf = 59.30±0.23
nj = 1.50±0.07
𝜒2 = 0.99r
Rf = 59.58±0.19
nj = 1.65±0.08
𝜒2 = 0.86r
Rf = 59.86±0.37
nj = 1.61±0.07
𝜒2 = 1.39r
Rf = 56.25±0.28
nj = 1.25±0.03
𝜒2 = 1.01r
Rf = 58.59±0.10
nj = 1.53±0.05
𝜒2 = 1.78r
Rf = 59.01±0.14
nj = 1.41±0.05
𝜒2 = 1.40r
Rf = 59.76±0.19
nj = 1.48±0.07
𝜒2 = 1.28r
Rf = 57.80±0.14
nj = 1.52±0.05
𝜒2 = 1.00r
Rf = 58.55±0.15
nj = 1.46±0.05
𝜒2 = 1.45r
Rf = 57.61±0.13
nj = 1.66±0.07
𝜒2 = 1.07r
Rc (kpc) Rc (kpc) Rc (kpc)
Figure 6.3: Surface brightness proﬁles across the cold front measured along 15 narrow,
5 degree wedges, each corresponding to ∼ 5 kpc along the front. The azimuth angles
are measured counterclockwise from the west. Radii are measured from the center of
curvature of the cold front (see Fig. 6.1). Over-plotted are the best ﬁt broken power-law
models for the ICM density proﬁles. For each wedge we indicate the best-ﬁt radius of
the discontinuity (Rf, measured from the center of curvature of the cold front), density
jump (nj), and the reduced χ2 for 92 degrees of freedom.
6.3 Results 103
These observations cover an annulus around the deep (net exposure time of 120 ks)
pointings on M 87. The full exposure corrected mosaic of XMM-Newton observations
divided by the best ﬁt radially symmetric beta-model is shown in Fig. 6.1. See
Simionescu et al. (2010) for the detailed discussion on the data and the analysis.
6.3 Results
The most prominent sloshing cold front in the Virgo Cluster extends at r ∼ 90 kpc
north of M 87 (see Fig. 6.1). Our Chandra observation (indicated by the white square in
the ﬁgure) targets the part of the front with the largest surface brightness contrast. The
deep Chandra image, on the left panel of Fig. 6.2, shows clearly this surface brightness
discontinuity.
The Chandra image also reveals unexpected, relatively bright quasi-linear features
underneath the discontinuity. To our knowledge, this is the ﬁrst time that such substructure
has been identiﬁed with a cold front. Three of these quasi-linear features can
be seen clearly on the residual image on the right panel of Fig. 6.2. They are ∼ 10 per
cent brighter than the surrounding gas and are separated by ∼ 15 kpc from each other
in projection. These features are below the detection limit of the existing XMM-Newton
observation.
To study the detailed cold front interface, we extracted surface brightness proﬁles
along 15 narrow, 5 degree wedges, each corresponding to ∼ 5 kpc along the front (see
Fig. 6.3). To optimize the sharpness of the surface brightness edge in our proﬁles, we
aligned our annuli with the front interface, choosing the vertex of the wedges at the
center of curvature of the highest-contrast part of the cold front (RA: 187◦.676, DEC:
12◦.5064). The data were binned into 4.92 arcsec radial bins. We ﬁt the surface brightness
proﬁles by projecting a spherically symmetric discontinuous, broken power-law
density distribution (see e.g. Owers et al. 2009). The ﬁtting was performed using the
PROFFIT package (Eckert et al. 2011) modiﬁed by Ogrean et al. (2015). We ﬁt the data
in the radial range of 8–16 arcmin, corresponding to 38–75 kpc from the center of curvature.
The free parameters of our model are the normalization, the inner and outer
density slopes, and the amplitude and radius of the density jump. We veriﬁed that the
best ﬁt parameters are not sensitive to the utilized radial range. The surface brightness
proﬁles over-plotted with their best ﬁt models, along with the values for the best ﬁt
density jumps, radii of the discontinuity, and reduced χ2s (for 92 degrees of freedom)
are shown in Fig. 6.3.
Our model provides the best ﬁt to the northern part of the front, at azimuthal angles
of 70–95 degrees, indicating that this is the sharpest part of the cold front. In particular,
our model provides the best ﬁt to the proﬁle extracted at the azimuthal range of 80–85
degrees. In order to place an upper limit on the width of the front, we ﬁt this proﬁle
with a model that includes Gaussian smoothing as an additional free parameter. We
ﬁnd that the 99 per cent upper limit on the smearing of this part of the cold front is 2.5
kpc, which is 1.5 Coulomb mean free paths (see Sect. 6.5.1 for details). Given the quasispherical
geometry along our line of sight, in this nearby system we observe a steep,
few kpc wide surface brightness gradient at the front (in the more distant systems
104 Resolving the nearest cold front in the sky
where the gradient cannot be resolved we only see a sharp surface brightness jump).
Because of this resolved gradient, our upper limit on the width is signiﬁcantly larger
than Chandra’s resolution.
On the other hand, at azimuths of 25–70 degrees, our model mostly provides a
poorer ﬁt to the data. Fig. 6.3 shows that around the brightness edge most proﬁles
display deviations from the model, with shallow, smeared discontinuities. The parameters
of the best ﬁt model and the structure seen in the proﬁles also show azimuthal
variation. In particular, the northwestern part of the front, at azimuths below 55 degrees,
shows signiﬁcant variation both in the value of the best ﬁt density jump and
radius. For the sectors with the least sharp edges the ﬁt signiﬁcantly improves after
including Gaussian smoothing in the model. For example, for the proﬁle extracted in
the 65–70 degree wedge, Gaussian smoothing with a best ﬁt width of 4.3 kpc improves
the χ2 from 138 (for 92 degrees of freedom) to 101, for one additional free parameter.
According to the F-test, the likelihood that this is a chance improvement is 1.06 × 10−7.
This highly signiﬁcant improvement strongly suggest that at these azimuths the front
has a ﬁnite width.
To study the projected thermodynamic properties and the metallicity of the ICM
across the cold front, we extracted spectra along three 22.5 degree wide wedges (labeled
North, Northwest, and West), the vertex of which is at the same center of curvature
of the cold front as for the surface brightness proﬁles. The best ﬁt projected
temperature and metallicity proﬁles along the three wedges, determined in narrow
2 kpc wide regions, are shown in Fig. 6.4. The projected temperature as a function of
radius increases from ∼ 2.7 keV to ∼ 3.6 keV across the interface, and the metallicity
decreases from ∼ 0.6 Solar to ∼ 0.3 Solar. The projected temperature distribution
appears to be smeared across the cold front. The observed gradient of the projected
temperature is between ∼ 4 kpc (West) and 8 kpc (North) wide.
The 2D maps of the projected thermodynamic properties shown in Fig. 7.4 conﬁrm
these trends. While the temperature clearly increases across the front, the best ﬁt metallicity
drops from ∼ 0.6 Solar to ∼ 0.3 Solar. The projected temperature and entropy
distributions appear slightly smeared at the edge. The smearing is likely to be at least
in part due to projection effects (see Sect. 6.5.1 for more discussion). As expected, the
cold front is not visible in the pressure map, indicating that the sloshing gas is in approximate
hydrostatic equilibrium within the cluster potential. The linear structures
identiﬁed in the images (see Fig. 6.2) do not show any signiﬁcant difference in temperature
or metallicity from the surrounding ambient medium.
In order to study the temperature difference of the quasi-linear features and the ambient
plasma in more detail, we extracted ﬁve spectra from rectangular regions: three
spectra from regions centered on the quasi-linear features, and two spectra from the
surrounding regions. Based on ﬁtting the combined spectra, the best ﬁt temperature of
the quasi-linear features is kT = 2.77 ± 0.03 keV and the best ﬁt temperature of the surrounding
gas is kT = 2.92 ± 0.03 keV, qualitatively consistent with the trend expected
from the simulations we present in Sect. 6.4 (see Fig. 6.9). However, we caution that
because the surface brightness excess associated with these features is only ∼ 10 per
cent, the measured temperature difference could easily be due to the underlying spatial
variation in the temperature of the ambient gas.
6.3 Results 105
North
Northwest
West
Rc (kpc)
Figure 6.4: Projected temperature and metallicity proﬁles along three 22.5 degree wide
wedges, extracted from the northern direction westward. The proﬁles extracted from
the north, northwest, west are shown in black, blue, and red, respectively. The best
ﬁt radius of the front determined by ﬁtting the surface brightness proﬁles is indicated
by the dashed line (the uncertainty on the front radius is ±2 kpc). Radii are measured
from the center of curvature of the cold front.
106 Resolving the nearest cold front in the sky
Figure 6.5: Deprojected electron density, temperature, metallicity, pressure, and entropy
proﬁles (from the bottom to the top), respectively. The red datapoints were determined
in a 30 degree wedge extending from the north to the west, centered on M 87,
using Chandra data and the blue datapoints are averaged over the other directions (330
degree region that excludes the wedge) using XMM-Newton data. The radius of the
cold front is indicated with the dotted line. While the deprojected pressure is continuous
across the cold front, the density, temperature, entropy, and metallicity distributions
show relatively strong discontinuities. Note that under the cold front (∼ 20 kpc)
the ICM entropy is smaller and above the front it is higher than in the other directions.
6.4 Numerical simulations 107
In order to determine the deprojected spectral properties of the ICM across the cold
front, we also extracted spectra from a wedge centered on M 87 at the bottom of the
gravitational potential of the Virgo Cluster. Centering the series of partial annuli used
for spectral extraction at the bottom of the global gravitational potential enables us
to perform the most robust deprojection of the stratiﬁed cluster atmosphere. The 30
degrees wide wedge used for extracting the Chandra spectra extends from the north
towards the west. For this particular wedge, the width of the cold front is the least
affected by the choice of the different center. Fig. 6.5 shows the comparison of the
deprojected electron density, temperature, metallicity, pressure, and entropy proﬁles
determined in the wedge using Chandra data (red datapoints) with the properties determined
in the other directions (averaged over a 330 degree region that excludes the
wedge) using XMM-Newton spectra (blue datapoints). While the deprojected pressure
appears continuous across the cold front, the density, temperature, entropy, and
metallicity distributions show clear discontinuities. The deprojected entropy proﬁle
extracted across the cold front shows a remarkably ﬂat distribution underneath the
discontinuity that extends across a radial range of ∼ 50 kpc. The comparison also
shows that under the cold front (∼ 20 kpc from the discontinuity) the ICM entropy is
smaller and above the front it is higher than in the other directions. The pressure appears
slightly higher under the cold-front than in the other directions. We veriﬁed that
the deprojected proﬁles determined using Chandra and XMM-Newton data in the 30 degree
wedge show consistent results. We caution that while our deprojection procedure
assumes spherical symmetry centered on M87, the distribution of the gas uplifted by
sloshing is likely to be different and therefore our inferred thermodynamic properties
underneath the discontinuity may be biased.
6.4 Numerical simulations
We compare our long exposure of the Virgo cold front to magnetohydrodynamic (MHD)
simulations. These simulations were performed using the Athena 4.1 astrophysical
simulation code (Stone et al. 2008). The technical details of the simulation setup are
discussed in ZuHone et al. (2015) and the initial conditions, based on the setup from
Roediger et al. (2011) and Roediger et al. (2013), were chosen to match the global Virgo
Cluster potential and the ICM properties. The cluster is modeled as a cooling-core system
with a mass of ∼ 2 × 1014 M and kT ∼ 2 keV. The sloshing in the simulation is
due to the passage of a gas-less dark-matter subcluster with a mass of 2 × 1013 M at
r ∼ 100 kpc at the time of t ∼ 1 Gyr after the beginning of the simulation. We note that
simulations show that similar cold front features are produced by gas-less perturbers
approaching at close distances or gas-ﬁlled perturbers approaching at larger distances,
indicating that the choice of perturber is not as important as the shape of the potential
well and the thermodynamic proﬁles of the main cluster (ZuHone et al. 2010; Roediger
et al. 2011; Roediger & Zuhone 2012). The time at which the properties of the Virgo
cold front most resemble the results of the simulation is ∼ 1.7 Gyr after the passage
of the subcluster. In Fig. 6.7, we show the synthetic X-ray surface brightness residual
distributions for the simulation at this time for two pure hydrodynamic runs with
108 Resolving the nearest cold front in the sky
2.4 2.6 2.8 2.9 3.1 3.3 3.4 3.6 3.8 3.9 4.1
Temperature 10 kpc
0.15 0.24 0.34 0.44 0.53 0.63 0.72 0.81 0.91 1 1.1
Fe abundance 10 kpc
1.4 1.7 2 2.2 2.5 2.8 3.1 3.4 3.6 3.9 4.2
Pressure 10 kpc
190 230 270 310 350 400 440 480 520 560 600
Entropy 10 kpc
Figure 6.6: Projected thermodynamic maps. Each spatial region contains 5600 net
counts (S/N of 75). To compare the pseudo-pressure and pseudo-entropy distributions,
we assume the gas uniformly distributed along the line-of-sight depth of l = 1
Mpc over the entire ﬁeld of view. The units of temperature (top left), Fe abundance
(top right), pressure (bottom left) and entropy (bottom right) are keV, Solar (Lodders
2003), 1000 keV cm−3(l/1 Mpc)1/2 and keV cm2(l/1 Mpc)1/3, respectively. The fractional
1σ statistical errors are ∼ 4 per cent for temperature, 5 per cent for pressure and
entropy, and ∼ 10–20 per cent for metallicity. Point sources were excluded from the
regions shown in grey.
6.5 Discussion 109
different viscosities (top row), and for two MHD runs (bottom row). The pure hydrodynamic
simulations were performed assuming either inviscid ICM (µ = 0), or 10 per
cent of the isotropic Spitzer-like temperature dependent viscosity (µ = 0.1µSp). Both
MHD runs started with a tangled magnetic ﬁeld distribution (set up as in ZuHone
et al. 2011), one with an initial ratio of thermal to magnetic pressure of β = 1000 and
the other with β = 100.
The two pure hydrodynamic simulations are visibly inconsistent with the data: the
inviscid run shows large, well-developed Kelvin-Helmholtz instabilities (KHI) that are
not seen at the Virgo cold front image (the surface brightness proﬁles extracted from
the mock image show a distorted front with multiple strong edges; see also Fig. 4
in Roediger et al. 2013); and the run with 10 per cent of Spitzer viscosity produces a
discontinuity that is much smoother than the proﬁles extracted across the northwestern
part of the cold front, seen in Fig. 6.3. Importantly, pure hydrodynamic simulations do
not reproduce the bright linear features seen in the right panel of Fig. 6.2, though such
features appear in the MHD run with the initial plasma β = 100; we will discuss this
in Section 6.5.5.
6.5 Discussion
6.5.1 The width of the cold front and suppressed diffusion
The observed surface brightness proﬁles across the cold front show considerable azimuthal
variation. While at certain azimuth angles (particularly in the north) the proﬁles
are relatively well described by the model of a spherically symmetric cloud with
a power-law density distribution, in other directions they deviate from the model, are
wider, and the best ﬁt front parameters show signiﬁcant azimuthal variation.
For the best ﬁt deprojected temperatures and electron densities, the Coulomb mean
free paths inside and outside of the cold front are λin = 0.77 kpc and λout = 2.2 kpc,
respectively. For unsuppressed diffusion, particles diffusing from the bright, dense
side of the discontinuity to the low density outer ICM will be especially efﬁcient at
broadening the cold front. The Coulomb mean free path for diffusion from the inner
bright side to the outer faint side of the discontinuity is:
λin→out = 15
Tout
7 keV
2
ne,out
10−3 cm−3
Tin
Tout
G(1)
G(
√
Tin/Tout)
= 1.7 kpc, (6.1)
where ne,out is the electron density outside the cold front, Tin and Tout are the temperatures
inside and outside the cold front, and G(x) = [φ(x) − xφ (x)]/2x2, where φ(x)
is the error function (Spitzer 1962; Markevitch & Vikhlinin 2007). Our conservative
upper limit of 2.5 kpc on the intrinsic width of the cold front, determined in the sector
with the sharpest brightness edge, corresponds to about 1.5 Coulomb mean free paths.
Part of the apparent broadening of the cold front in the Virgo Cluster may be caused
by projection effects and the true intrinsic width of the discontinuity is also likely to
be smaller than the Coulomb mean free path. Because the front is seen along the surface
in projection, any deviations from the ideal spherical shape will smear the edge.
110 Resolving the nearest cold front in the sky
B = 0, µ = 0
20 kpc
B = 0, µ = 0.1µSp, isotropic
β = 100, µ = 0 β = 1000, µ = 0
−0.40 −0.32 −0.24 −0.16 −0.08 0.00 0.08 0.16 0.24 0.32 0.40
Fractional Residuals
Figure 6.7: Simulated X-ray surface brightness residual distributions for the Virgo
Cluster simulation at t = 1.7 Gyr after the closest passage of the subcluster for two
pure hydrodynamic simulation runs with different viscosities (top row), and two MHD
runs with different initial plasma beta parameters (bottom row).
6.5 Discussion 111
-100 -50 0 50 100
x (kpc)
-50
0
50
100
y(kpc)
No Conduction Conduction
10−7
10−6
SX(1−4keV)(countss−1cm−2arcsec−2)
5 10 15 20 25 30 35 40
radius (kpc)
10
3
10
4
SX(ctss−1cm−2)
No Conduction
Conduction
5 10 15 20 25 30 35 40
radius (kpc)
2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
T(keV)
Figure 6.8: Mock surface brightness images (left two panels), and surface brightness
and temperature proﬁles (right two panels) extracted from the black wedges for numerical
MHD simulations performed assuming no conduction and full Spitzer conduction
along the magnetic ﬁeld lines. Spitzer conduction along the magnetic ﬁeld
lines would wash out the temperature discontinuity to a degree that is inconsistent
with the observations. The observed temperature and surface brightness proﬁles (see
Fig. 6.3, 6.4) are consistent with the conduction along the magnetic ﬁeld lines being
suppressed by a factor 10.
Therefore, the observed width is a conservative upper limit on the actual width of the
discontinuity. Based on comparisons to numerical simulations, Roediger et al. (2011,
2013) argue for sloshing in the Virgo Cluster close to the plane of the sky, ruling out
orbital planes that are further than 45 degrees from the plane of the sky. We veriﬁed
that the mock X-ray images of sloshing in our simulations, produced for various lines
of sight, are consistent with those in Roediger et al. (2011, 2013). For sloshing taking
place predominantly in the plane of the sky projection effects are minimized. Their
smearing effect gets more important as the shape of the front departs from spherical
symmetry and as the inclination of the orbital plane to the plane of the sky gets larger
(until the inclination approaches 90 degrees, when sloshing is along the line-of-sight).
Unsuppressed diffusion would smear the density discontinuity by several mean free
paths. Our upper limit on its intrinsic width therefore indicates that diffusion is suppressed
across the cold front. This is consistent with the results obtained for the cold
front identiﬁed with a subcluster merger in Abell 3667, where Vikhlinin et al. (2001a)
showed that the gas density discontinuity is several times narrower than the Coulomb
mean free path (see Markevitch & Vikhlinin 2007, for updated results). Churazov &
Inogamov (2004) argue that a small but ﬁnite width of the interface set by diffusion,
that is of the order of a few per cent of the curvature radius (∼ 2 kpc for the Virgo
cold front), could strongly limit the growth of perturbations such as KHI (see also
Chandrasekhar 1961). Given projection effects that are expected to be broadening the
observed width of the discontinuity, the intrinsic width of the Virgo cold front is likely
to be too small to be suppressing the development of KHI.
112 Resolving the nearest cold front in the sky
0 10 20 30 40 50 60 70 80
x (kpc)
0
10
20
30
40
50
60
70
80
y(kpc)
1
2
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
ne(cm−3)
1e−3
0 10 20 30 40 50 60 70 80
x (kpc)
0
10
20
30
40
50
60
70
80
y(kpc)
1
2
1.8
2.0
2.2
2.4
2.6
2.8
3.0
3.2
3.4
kT(keV)
0 10 20 30 40 50 60 70 80
x (kpc)
0
10
20
30
40
50
60
70
80
y(kpc)
1
2
10
-1
10
0
10
1
MagneticFieldStrength(µG)
0 10 20 30 40 50 60 70 80
x (kpc)
0
10
20
30
40
50
60
70
80
y(kpc)
1
2
−0.30
−0.24
−0.18
−0.12
−0.06
0.00
0.06
0.12
0.18
0.24
0.30
SurfaceBrightnessResiduals(1.0-4.0keV)
Figure 6.9: A zoom in on a 80 × 80 kpc2 region around the cold front located northwest
of the cluster center for the MHD simulation run with an initial magnetic to thermal
pressure ratio β = 100 (see the bottom right panel of Fig. 6.7). Top-left: slice of gas
density; top-right: slice of gas temperature; bottom-left: slice of magnetic ﬁeld strength;
bottom-right: projected surface brightness residuals. The solid cyan lines indicate the
extraction regions for the proﬁles shown in Fig. 6.10.
6.5 Discussion 113
0.0025
0.0030
0.0035
0.0040
0.0045
ne(cm−3)
Electron Density
2
4
6
8
10
12
B(µG)
Magnetic Field Strength
20 25 30 35 40 45 50 55 60
x (kpc)
0.010
0.012
0.014
0.016
0.018
0.020
0.022
p(keVcm−3)
Pressure Thermal Pressure
Total Pressure
0.0030
0.0035
0.0040
0.0045
0.0050
ne(cm−3)
Electron Density
1
2
3
4
5
6
7
B(µG)
Magnetic Field Strength
40 45 50 55 60
x (kpc)
0.014
0.016
0.018
0.020
0.022
p(keVcm−3)
Pressure
Thermal Pressure
Total Pressure
Figure 6.10: Proﬁles of electron density, magnetic ﬁeld strength, and pressure along the
cyan lines indicated in Fig. 6.9 as ‘1’ (left panel) and ‘2’ (right panel). The difference
between the total and thermal pressure is due to magnetic pressure support. Dashed
lines indicate approximate boundaries of density enhancements along the proﬁle.
6.5.2 Kelvin-Helmholtz instabilities and ICM viscosity
The primary effect of projection is the broadening of the surface brightness edge and
possibly a relatively smooth azimuthal change of the broadening. However, the nonzero
width of the cold front, combined with the observed substructure and the azimuthal
variations of the best ﬁt radius and density jump seen in Fig. 6.3, require
the presence of additional processes, such as KHI. The temperature proﬁles shown
in Fig. 6.4 are consistent with the presence of instabilities on a scale of a few kpc.
Constrained hydrodynamic simulations by Roediger et al. (2012, 2013) demonstrate
the efﬁciency of viscosity at suppressing KHI and show that for viscosities of about 10
per cent of the Spitzer value or larger, at this cold front, KHIs should be suppressed.
Our pure hydrodynamic simulation run with 10 per cent Spitzer viscosity is consistent
with these conclusions. Based on comparison to these numerical simulations, the
presence of KHI would imply that the effective viscosity of the ICM is suppressed by
more than an order of magnitude with respect to the isotropic Spitzer-like temperature
dependent viscosity. Our conclusions, however, depend on the correctness of the
assumed merger model.
The growth of KHIs may be facilitated by the presence of magnetic ﬁelds which
lower the effective viscosity of the plasma by preventing the diffusion of momentum
and heat perpendicular to the ﬁeld lines. Also, micro-scale MHD plasma instabilities
may set an upper limit on the viscosity that is much lower than that expected from
ion collisionality (Kunz et al. 2014). However, depending on the ﬁeld geometry and
strength, magnetic ﬁelds may also protect cold fronts against instabilities. A layer of
strong enough magnetic ﬁelds aligned with the cold front could suppress the growth
114 Resolving the nearest cold front in the sky
of KHIs (Vikhlinin & Markevitch 2002; Keshet et al. 2010). However, the suppression of
KHI in the simulations of ZuHone et al. (2011) is often partial, indicating that magnetic
ﬁelds cannot always stabilize cold fronts. Our example in the bottom-right panel of
Figure 6.7 shows that if magnetic ﬁelds are initially weak (β = 1000 in our example),
they will be unable to suppress KHI. Stronger magnetic ﬁelds (see bottom-left panel of
Fig. 6.7 for an example with initial β = 100) suppresses most of the KHI but not all.
Other possible explanations for the broadening and the observed substructure in
the surface brightness proﬁles include plasma depletion due to regions of ampliﬁed
magnetic ﬁelds below the cold front, jet inﬂated bubbles entrained by the sloshing gas,
or clumpy, inhomogeneous ICM distribution.
6.5.3 Gas mixing at the cold front
Well developed, long lasting, large KHI would mix the high metallicity ICM (∼ 0.6
Solar) on the bright side of the interface with the low metallicity gas (∼ 0.3 Solar) on
the faint side. The relatively sharp (narrower than 6 kpc) metallicity gradient indicates
that if the surface brightness substructure is due to KHI, they are on relatively small,
few kpc, scales and are unable to efﬁciently mix metals across wider regions around
the interface. Since cold fronts are transient wave phenomena, the lack of signiﬁcant
mixing across the front is not entirely surprising. Numerical simulations indicate that
metallicity jumps remain preserved even for inviscid ICM (Roediger et al. 2011) and
they get washed out only on very small scales, consistent with our observation and
observations of sloshing cold fronts in other more distant clusters which are also associated
with metallicity discontinuities (see e.g. Ghizzardi et al. 2014).
6.5.4 Conduction across the discontinuity
ZuHone et al. (2013) show that while sloshing approximately aligns the magnetic ﬁeld
lines with the cold front surface, perfect alignment is not expected. KHI will re-tangle
the ﬁeld lines, restoring the heat ﬂow between the gas phases above and below the
cold front. ZuHone et al. (2015) also performed simulations for the Virgo Cluster with
anisotropic conduction, assuming full Spitzer conductivity along the magnetic ﬁeld
lines. In these, the temperature and surface brightness gradients are washed out to
such a degree that is inconsistent with our observation (see Fig 6.8). The relatively narrow
temperature gradients across the interface (see Fig. 6.4) suggest that the effective
conductivity along the ﬁeld lines must be reduced by effects not modeled by the simulation,
such as strong curvature of the ﬁeld lines, stochastic magnetic mirrors on small
scales, and/or microscopic plasma instabilities (e.g. Chandran et al. 1999; Malyshkin
2001; Narayan & Medvedev 2001; Schekochihin et al. 2005, 2008). While the observed
temperature and surface brightness proﬁles are consistent with no conduction along
the magnetic ﬁeld lines (see Fig. 6.8), conductivity at or below 0.1 of the Spitzer value
along the ﬁeld lines (which would still result in a steep gradient across the discontinuity,
see ZuHone et al. 2013) cannot be excluded and the exact suppression factor
remains unknown. Like our conclusions on viscosity, this result also depends strongly
6.5 Discussion 115
on the correctness of the assumed merger model.
6.5.5 Ampliﬁed magnetic ﬁelds underneath the cold front
The Chandra images in Fig. 6.2 show quasi-linear features that are ∼ 10 per cent brighter
than the surrounding gas. The simulations of ZuHone et al. (2011) and ZuHone et al.
(2015) show that the magnetic ﬁeld layers ampliﬁed by sloshing may be wide and extend
relatively far below cold fronts. This can be seen clearly in the Virgo simulation
with the initial plasma β = 100 in Fig. 6.9, which shows slices through a 80 ×80 kpc2 region
around the cold front for gas density, temperature, and magnetic ﬁeld strength, as
well as residuals of projected surface brightness. Along the lines indicated as “1” and
“2”, we calculate the distributions of the density, temperature, magnetic ﬁeld strength
and pressure and show them in Fig. 6.10. The quasi-linear features seen in the simulation
are very similar to those in our observation, both in terms of width and apparent
surface brightness excess. We ﬁnd that regions of high density/surface brightness correspond
to regions of relatively low magnetic ﬁeld in the map. The thin regions of
high density are surrounded by wider bands which are highly magnetized. Here, the
increased magnetic pressure has pushed gas out, decreasing its density. This opens up
the intriguing possibility that the observed bright bands may be due to denser ICM
layers between bands of more magnetized plasma, where the magnetic pressure is ampliﬁed
to ∼ 5–10 per cent of the thermal pressure. Though the magnetic ﬁelds in this
simulation are strong, they do not completely prevent the development of KHI, consistent
with the indications of KHI in our observation.
By extrapolating the pressure proﬁles outside of cold fronts inwards, up to the edge,
Reiss & Keshet (2014) infer that in the inner parts of the spirals most sloshing cold
fronts show a pressure jump of Pin/Pout ∼ 0.8, which may correspond to the ampliﬁcation
of magnetic ﬁelds mainly below the discontinuity. The Virgo cold front does
not show any signiﬁcant pressure jump, which may be due to the ampliﬁcation of the
magnetic ﬁelds both above and below the front. Fig. 6.9 shows that such a scenario
does occur in our simulation. The magnetic ﬁelds above the cold front get ampliﬁed
due to the shear in the large-scale velocity ﬁeld produced by the encounter with the
subcluster. The same large scale gas motions then draw the ampliﬁed magnetic ﬁelds
inward, towards the cold front.
The observed quasi-linear features may potentially also be due to compressed thermal
gas formed in front of bubbles entrained by the magnetized sloshing gas, spiraling
outwards below the cold front. Such bubble entrainment was previously suggested by
Keshet (2012). Future numerical simulations, that include both jet inﬂated bubbles and
sloshing, will test this possible interpretation.
Alternatively, for geometries where the sloshing occurs along our line-of-sight, KHI
would develop predominantly in the plane of the sky, forming rolls projected on the
bright side of the front. For such geometries, Kelvin-Helmholtz rolls projected on the
sloshing surface could, in principle, appear as quasi-linear features. However, based
on the comparison of the numerical simulations to the observations, such geometry
is unlikely. None of the simulations that we performed show KHI in projection that
would look like the observed quasi-linear features.
116 Resolving the nearest cold front in the sky
6.5.6 The ﬂow of the sloshing gas
Under the cold front interface, the ICM entropy is lower and above the interface it
is higher than the radial average. The low entropy gas under the interface also has
roughly two times higher metallicity than the gas above the interface. This entropy and
metallicity distribution indicates convergent gas ﬂows at the cold front interface, with
the low entropy metal rich gas moving outward and, in the same time, the high entropy
gas moving inward. Such convergent ﬂows are also seen in numerical simulations (e.g.
Ascasibar & Markevitch 2006; Roediger et al. 2011).
The entropy proﬁle shows a remarkably ﬂat distribution underneath the cold front,
extending across a radial range of ∼ 50 kpc. The ﬂat entropy distribution may indicate
that, in this region, the gas uplifted by sloshing originated at the same radius. Similar
entropy plateaus have also been seen in e.g. the Perseus and Abell 2142 clusters
and interpreted as evidence for gas uplift by sloshing (Simionescu et al. 2012; Rossetti
et al. 2013). However, if the distribution of the uplifted gas deviates signiﬁcantly from
spherical symmetry centered on M87 (as assumed by our deprojection analysis), then
the shape of the entropy proﬁle measured underneath the cold front may be biased.
6.6 Conclusions
We have analyzed a new, very deep (500 ks) Chandra observation, together with complementary
archival XMM-Newton data, and performed tailored numerical simulations
to study the nearest cluster cold front in the sky. We ﬁnd that:
• The northern part of the front appears the sharpest, with a width smaller than
2.5 kpc (1.5 Coulomb mean free paths; at 99 per cent conﬁdence). Everywhere
along the front, the temperature discontinuity is narrower than 4–8 kpc and the
metallicity gradient is narrower than 6 kpc, indicating that diffusion, conduction
and mixing are suppressed across the interface. Such transport processes can
be naturally suppressed by magnetic ﬁelds aligned with the cold front. We also
ﬁnd that the temperature jump across the cold front is inconsistent with the temperature
jumps produced in simulations with anisotropic thermal conduction,
indicating that conduction is also suppressed along the magnetic ﬁeld lines.
• However, the northwestern part of the cold front is observed to have a nonzero
width. The broadening is consistent with the presence of projected KHI eddies on
length scales of a few kpc. Based on comparison with simulations, the presence
of KHI would imply that the effective viscosity of the ICM is suppressed by more
than an order of magnitude with respect to the isotropic Spitzer-like temperature
dependent viscosity.
• Underneath the cold front, we observe linear features that are ∼ 10 per cent
brighter than the surrounding gas and are separated by ∼ 15 kpc from each
other in projection. Comparison to tailored numerical simulations suggests that
the observed phenomena may be due to the ampliﬁcation of magnetic ﬁelds by
6.6 Conclusions 117
gas sloshing in wide layers below the cold front, where the magnetic pressure
reaches ∼ 5–10 per cent of the thermal pressure, reducing the gas density between
the bright features.
• The deprojected entropy distribution and the metallicity of the gas indicate convergent
gas ﬂows at the cold front interface, with the low entropy metal rich gas
moving outward and, in the same time, the high entropy gas moving inward.
Acknowledgments
NW thanks G. Ogrean, P. Nulsen, O. Urban, and R. Canning for discussions. Support
for this work was provided by the National Aeronautics and Space Administration
through Chandra Award Number GO3-14142A issued by the Chandra X-ray Observatory
Center, which is operated by the Smithsonian Astrophysical Observatory for
and on behalf of the National Aeronautics Space Administration under contract NAS8-
03060. JAZ acknowledges support from NASA though subcontract SV2-8203 to MIT
from the Smithsonian Astrophysical Observatory. Analysis of the simulation data was
carried out using yt, a visualization and analysis software suite for simulations in
astrophysics (http://yt-project.org, Turk et al. 2011). YI is ﬁnancially supported by
a Grant-in-Aid for Japan Society for the Promotion of Science (JSPS) Fellows. E. R.
acknowledges the support of the Priority Programme Physics of the ISM of the DFG
(German Research Foundation) and a visiting scientist fellowship of the Smithsonian
Astrophysical Observatory. U.K. is supported by the European Union Seventh Framework
Programme, by an IAEC-UPBC joint research foundation grant, and by an ISFUGC
grant. This work was supported in part by the U.S. Department of Energy under
contract number DE-AC02-76SF00515.
118 Resolving the nearest cold front in the sky
Chapter 7
Deep Chandra study of the truncated
cool core of the Ophiuchus cluster
N. Werner1,2, R. E. A. Canning1,2, I. Zhuravleva1,2, S. W. Allen1,2,3, A. King1,2, J. S. Sanders4,
A. Simionescu5, G. B. Taylor7, R. G. Morris1,3, A. C. Fabian6
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall, Stanford, CA 94305-4085, USA
2Department of Physics, Stanford University, 382 Via Pueblo Mall, Stanford, CA 94305-
4060, USA
3SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025,
USA
4Max-Planck-Institut f¨ur extraterrestrische Physik, Giessenbachstrasse 1, 85748 Garching,
Germany
5Institute of Space and Astronautical Science (ISAS), JAXA, 3-1-1 Yoshinodai, Chuo-ku,
Sagamihara, Kanagawa, 252-5210, Japan
6Institute of Astronomy, Madingley Road, Cambridge CB3 0HA, UK
7Department of Physics and Astronomy, University of New Mexico, Albuquerque, NM
87131, USA
Submitted to the Monthly Notices of the Royal Astronomical Society in March 2016
Abstract
We present the results of a deep (280 ks) Chandra observation of the Ophiuchus cluster,
the second-brightest galaxy cluster in the X-ray sky. The cluster hosts a truncated cool
core, with a temperature increasing from kT ∼ 1 keV in the core to kT ∼ 9 keV at
r ∼ 30 kpc. Beyond r ∼ 30 kpc the intra-cluster medium (ICM) appears remarkably
isothermal. The core is dynamically disturbed with multiple sloshing induced cold
fronts, with indications for both Rayleigh-Taylor and Kelvin-Helmholtz instabilities.
The sloshing is the result of the strongly perturbed gravitational potential in the cluster
core, with the central brightest cluster galaxy (BCG) being displaced southward from
120 Chandra study of the Ophiuchus cluster
the global center of mass. The residual image reveals a likely subcluster south of the
core at the projected distance of r ∼ 280 kpc. The cluster also harbors a likely radio
phoenix, a source revived by adiabatic compression by gas motions in the ICM. Even
though the Ophiuchus cluster is strongly dynamically active, the amplitude of density
ﬂuctuations outside of the cooling core is low, indicating velocities smaller than ∼ 100
km s−1. The density ﬂuctuations might be damped by thermal conduction in the hot
and remarkably isothermal ICM, resulting in our underestimate of gas velocities. We
ﬁnd a surprising, sharp surface brightness discontinuity, that is curved away from
the core, at r ∼ 120 kpc to the southeast of the cluster center. We conclude that this
feature is most likely due to gas dynamics associated with a merger and not a result
of an extraordinary active galactic nucleus AGN outburst. The cooling core lacks any
observable X-ray cavities and the AGN only displays weak, point-like radio emission,
lacking lobes or jets, indicating that currently it may be largely dormant. The lack of
strong AGN activity may be due to the bulk of the cooling taking place offset from the
central supermassive black hole.
7.1 Introduction
The degree to which cool cores are disrupted by merger events is an outstanding question
in cluster physics, with implications for galaxy cluster evolution and cosmology
(e.g. Burns et al. 2008; Allen et al. 2011). Some simulations have suggested that cluster
cool cores are very difﬁcult to destroy once formed, and are therefore expected to survive
major mergers (Burns et al. 2008; Poole et al. 2008). However, these conclusions
are sensitive to the hydrodynamical/magnetohydrodynamical treatments employed.
Our previous analysis of a short 50 ks Chandra observation of the nearby Ophiuchus
cluster also challenges this view (Million et al. 2010a).
The Ophiuchus cluster (z = 0.0296, Durret et al. 2015) is the second-brightest galaxy
cluster in the 2–10 keV sky (Edge et al. 1990). It is the most massive structure in the
Ophiuchus Supercluster and is surrounded by many in-falling smaller clusters and
groups (Wakamatsu et al. 2005; Hasegawa et al. 2000). Spatially resolved spectroscopy
using the previous Chandra observation (Million et al. 2010a) revealed remarkably
steep temperature and iron abundance gradients within the cool core: in the central
30 kpc, the temperature plunges by more than an order of magnitude, and the metallicity
varies by a similar amount. Beyond this radius, however, and out to the edge of
the Chandra ﬁeld of view, near constant temperature and metallicity are observed. The
cool core of the cluster appears to have been truncated.
The previous Chandra observations also revealed a series of cold fronts (Ascasibar
& Markevitch 2006; Million et al. 2010a), suggesting substantial ‘sloshing’ of the X-ray
emitting gas (Markevitch et al. 2001; Tittley & Henriksen 2005; Ascasibar & Markevitch
2006; Markevitch & Vikhlinin 2007). The X-ray emission associated with the cool core is
very sharply peaked and the X-ray morphology of the inner 10 kpc region is complex.
The X-ray and stellar optical/near-IR brightness peaks are offset by ∼ 4 arcsec (∼ 2.2
kpc; Hamer et al. 2012). These observations indicate that the cool core of the Ophiuchus
cluster has been stripped by the violent gas sloshing in the center of the cluster.
7.2 Observations and data analysis 121
Table 7.1: Summary of the Chandra observations. Columns list the observation ID,
observation date, and the exposure after cleaning.
Obs. ID Obs. date Exposure (ks)
3200 2002 Oct. 21 50.5
16142 2014 Jul. 1 32.6
16143 2014 Jul. 13 14.9
16464 2014 Aug. 8 31.7
16626 2014 Jul. 3 30.8
16627 2014 Jul. 5 34.6
16633 2014 Aug. 4 9.5
16634 2014 Jul. 11 22.4
16635 2014 Jul. 12 18.9
16645 2014 Aug. 9 33.6
Here, we present a deep (280 ks) Chandra observation of the Ophiuchus cluster along
with new high resolution Jansky Very Large Array (JVLA) 1–2 GHz radio data, to study
the effects of strong gas dynamics on the evolution of the cool core in detail. At a redshift
of z = 0.0296 (Durret et al. 2015), assuming the concordance ΛCDM cosmology,
one arcsecond corresponds to 0.59 kpc.
7.2 Observations and data analysis
7.2.1 Chandra data
The Chandra X-ray observatory pointed at the Ophiuchus cluster in July and August
2014 for 230 ks using the Advanced CCD Imaging Spectrometer (ACIS) in the VFAINT
mode, using CCDs 1–4 and 6. We also included in our analysis a 50 ks observation
performed in October 2002 using the ACIS S3 chip. The standard level-1 event lists
were reprocessed using the CIAO (version 4.7, CALDB 4.6.5) software package, including
the latest gain maps and calibration products. Bad pixels were removed and
standard grade selections applied. The data were cleaned to remove periods of anomalously
high background. The observations, along with the identiﬁers, dates, and net
exposure times after cleaning (total net exposure of 280 ks), are listed in Table 1. Background
images and spectra were extracted from the blank-sky ﬁelds available from the
Chandra X-ray Center. These were cleaned in an identical way to the source observations,
reprojected to the same coordinate system and normalized by the ratio of the
observed to blank-sky count rates in the 9.5–12 keV band.
Image analysis
Background-subtracted images, using both the native resolution of the Chandra CCDs
and using 2 × 2 pixel binning, were created in 13 narrow energy bands (0.6–0.8 keV,
122 Chandra study of the Ophiuchus cluster
0.00e+00 7.66e-08 4.34e-07 2.09e-06 9.81e-06
100 kpc
0.8 0.81 0.84 0.9 0.98 1.1 1.2 1.3 1.5 1.7
100 kpc
Concave
discontinuity
Southern
surface brightness
excess
Southern outer
cold front
x BCG
Eastern inner
cold front
Figure 7.1: Left panel: Chandra X-ray image of the Ophiuchus cluster in the 0.6–7.5 keV
energy band. The image was smoothed with a Gaussian function with a 1.0 arcsec
window. Right panel: The residual image obtained by dividing the image on the left
with an elliptical double beta-model reveals a prominent surface brightness excess in
the south and a surprising, sharp discontinuity southeast of the cluster center, curved
away from the core. North is up and east is left in all images.
7.2 Observations and data analysis 123
Concave
discontinuity
Southern
surface brightness
excess
Southern outer
cold front
x BCG
Eastern inner
cold front
0.00e+00 3.28e-07 1.86e-06 8.94e-06 4.20e-05
30 kpc
Eastern inner
cold front
Southern outer
cold front
0 0.00e+00 1.26e-07 7.12e-07 3.42e-06 1.61e-05
GGM filter scale 4+8
30 kpc
Figure 7.2: The same Chandra image as Fig. 7.1, zoomed in on the central part of the
cluster (left panel). The right panel shows the GGM ﬁltered image, using σ = 4 and
8 pixels (a pixel corresponds to 0.984 arcsec), which highlights multiple edges and the
disturbed, broken-up cold front to the southeast of the cluster core.
0.8–1.0 keV, 1.0–1.2 keV, 1.2–1.5 keV, 1.5–2.0 keV, 2.0–2.5 keV, 2.5–2.75 keV, 2.75–3.0 keV,
3.0–3.5 keV, 3.5–4.0 keV, 4.0–6.0 keV, 6.0–7.0 keV, 7.0–7.5 keV), spanning 0.6–7.5 keV.
The narrow-band images were ﬂat ﬁelded with respect to the median energy for each
image and then co-added to create the X-ray images shown in Fig. 7.1. Identiﬁcation
of point sources was performed using the CIAO task WAVDETECT. The point sources
were excluded from the subsequent analysis.
To look for sub-structure in the cluster, we removed the underlying large-scale surface
brightness gradient by ﬁtting an elliptical double beta-model to the image using
the Sherpa package, and subsequently divided the image by the best-ﬁt model (see
the right panel of Fig. 7.1).
To highlight the sharp features in the image, we also used a Gaussian Gradient Magnitude
(GGM) ﬁlter (see Sanders et al. 2016). The GGM ﬁlter uses Gaussian derivatives
to determine the magnitude of surface brightness gradients in the image. We applied
the GGM ﬁlter to the exposure-corrected, background subtracted image, from which
point sources have been removed, for various Gaussian widths, using σ = 1, 2, 4, and
8 pixels (a pixel corresponds to 0.984 arcsec). Regions with large gradients, such as discontinuities,
are bright, while ﬂatter surface brightness areas are darker (see the right
panel of Fig. 7.2).
To study the detailed properties of the surface brightness discontinuities in our Xray
images, we extracted surface brightness proﬁles at the native resolution of the
Chandra CCD detectors, in 0.492 arcsec bins. We optimized the sharpness of the surface
124 Chandra study of the Ophiuchus cluster
0.00e+00 2.13e-07 1.21e-06 5.82e-06 2.73e-05
10 kpc
x BCG
Eastern inner
cold front
Tongue-like
extension
Western
discontinuity
K-H
instability (?)
0.00e+00 1.41e-06 7.99e-06 3.84e-05 1.80e-04
GGM filter scale 1+2
10 kpc
Figure 7.3: The same Chandra image as Fig. 7.1, zoomed in on the very central part
of the cluster core (left panel). The right panel shows the GGM ﬁltered image, using
σ = 1 and 2 pixels, which highlights the remarkably sharp eastern inner cold front with
a ‘tongue’-like extension to the north, and the western discontinuity with a possible
Kelvin-Helmholtz roll. The X-ray surface brightness peak is offset from the center of
the BCG by about 2.2 kpc.
brightness edges in the proﬁles by aligning our circular annuli with the discontinuities.
We ﬁt the surface brightness proﬁles by projecting a spherically symmetric discontinuous,
broken power-law density distribution (see e.g. Owers et al. 2009). The ﬁtting was
performed using the PROFFIT package (Eckert et al. 2011) modiﬁed by Ogrean et al.
(2015). In order to place upper limits on the widths of the edges, we included in our
models Gaussian smoothing as an additional free parameter.
Spectral analysis
Individual regions for the 2D spectral mapping are determined using the contour binning
algorithm (Sanders 2006), which groups neighboring pixels of similar surface
brightness until a desired signal-to-noise ratio threshold is met. The Ophiuchus cluster
is a very hot system, with kT ∼ 10 keV (Matsuzawa et al. 1996; Fujita et al. 2008a;
Million et al. 2010a), which makes its temperature and metallicity measurement relatively
difﬁcult (at such high temperatures the cut-off of the thermal bremsstrahlung
spectrum is outside of the energy window of the telescope), requiring a large number
of counts. Therefore, we adopt a relatively high signal-to-noise ratio of 200 (approximately
40,000 counts per region), which provides approximately 2 per cent precision
for density, 4 per cent precision for temperature, 12 per cent precision for metallicity,
and 3 per cent precision for the line-of-sight absorption column density measurements,
respectively. To obtain a high resolution 2D spectral map of the cooling core, we also
7.3 Results 125
produced a map of the innermost r = 10 kpc region with a signal-to-noise of 50 (approximately
2500 counts per region), which provides 3–9 per cent precision for temperature,
and 18–25 per cent per cent for metallicity. The errorbars are the smallest for the
lowest temperature regions and grow bigger with increasing temperature. We model
the spectra extracted from each spatial region with the SPEX package (Kaastra et al.
1996) in the 0.6–7.5 keV band. The spectrum for each region is ﬁtted with a model consisting
of an absorbed single-phase plasma in collisional ionization equilibrium. The
Ophiuchus cluster lies close to the plane of our Galaxy, and therefore its line-of-sight
absorption column density is high, with NH, according to the Leiden/Argentine/Bonn
radio survey of H I, of NH = 2 × 1021 cm−2 (Kalberla et al. 2005). Radio surveys of H I
usually underestimate the total absorption for systems with such high NH, because
they are not sensitive to molecular and dust absorption. Therefore, the NH has to be a
free parameter in our model. Other free parameters include the spectral normalization
(emission measure), temperature, and metallicity.
The deprojected spectra are obtained using the PROJCT model in XSPEC and the
APEC plasma model (Smith et al. 2001; Foster et al. 2012). The line-of-sight absorption
column density is a free parameter in the ﬁt and for a given azimuth it is assumed to
have the same value in all annuli except the innermost region where it is allowed to
vary independently. Metallicities are given with respect to the proto-Solar abundances
by Grevesse & Sauval (1998).
7.2.2 Radio JVLA data
We observed the Ophiuchus cluster with the Jansky Very Large Array (JVLA) in the L
band for four hours on 2015 July 30. The observation was taken in the A conﬁguration,
resulting in a restoring FWHM beam size of 1.17 ×2.48 . We spent 16 minutes
on 3C286 to calibrate the ﬂux, 22 minutes on J1734+0926 to calibrate the polarization
leakage, 18 minutes on J1714-2514 to calibrate the phase, and a total of 150 minutes on
source. The total bandpass was 1 GHz.
The data analysis was performed using the CASA package (McMullin et al. 2007). After
the ﬂux and phase calibrations were transferred to the source and RFI excluded, we
used the CLEAN routine to image the cluster. We used a Briggs weighting of Robust =
0.7 to increase the image sensitivity, while still keeping the high resolution to detect
any possible extended structure near the BCG. We created the images ﬁtting for both
spectral index and correcting for the primary beam attenuation. The ﬁnal image has
an rms of 10 µJy.
7.3 Results
7.3.1 X-ray imaging
The Ophiuchus cluster shows a strongly centrally peaked X-ray surface brightness distribution,
elongated north-northwestward (see the left panel of Fig. 7.1). By ﬁtting an
elliptical double beta-model to the image, with the cluster center as a free parameter,
126 Chandra study of the Ophiuchus cluster
2 3 4 5 6 7 8.1 9.1 10 11
kT 60 kpc
3.6 3.7 3.9 4.0 4.1 4.2 4.3 4.4 4.6 4.7
N_H 60 kpc
0.22 0.36 0.5 0.65 0.79 0.93 1.1 1.2 1.4 1.5
Fe abundance 60 kpc
Figure 7.4: 2D maps of the best ﬁt projected temperature, line of sight absorbing hydrogen
column density, and iron abundance. Each spatial region contains 40,000 net
counts (S/N of 200). The units of temperature, absorbing column density, and Fe abundance
are keV, 1021 cm−2, and Solar (Grevesse & Sauval 1998). The fractional 1σ statistical
errors are ∼ 4 per cent for temperature, and ∼ 12 per cent for metallicity. Point
sources were excluded from the regions shown as black circles.
0 0.91 1.8 2.7 3.7 4.6 5.5 6.4 7.3 8.2
5 kpckT
0 1.12 2.24 3.36 4.49
H 5 kpcN
0 0.28 0.56 0.85 1.1 1.4 1.7 2 2.3 2.5
5 kpcFe abundance
Figure 7.5: The 2D maps zoomed-in on the innermost r = 10 kpc core of the Ophiuchus
cluster show the best ﬁt projected temperature (left panel), line of sight absorbing hydrogen
column density (central panel), and iron abundance (right panel) for spatial
regions with 2500 counts (S/N of 50). The units of temperature, absorbing column
density, and Fe abundance are keV, 1021 cm−2, and Solar (Grevesse & Sauval 1998).
The fractional 1σ statistical errors are 3–9 per cent for temperature, and 18–25 per cent
per cent for metallicity. The errorbars are the smallest for the lowest temperature regions
and grow larger with increasing temperature.
7.3 Results 127
0 36 72 108 145 181 217 253 289 325
60 kpcEntropy
0 0.018 0.036 0.054 0.072 0.09 0.11 0.13 0.14 0.16
60 kpcPressure
Figure 7.6: 2D maps of the projected entropy and pressure. To study their azimuthal
distribution, we assume the gas uniformly distributed along the line-of-sight depth of
l = 100 kpc over the entire ﬁeld of view. The units of the entropy and pressure are
keV cm2(l/100 kpc)1/3 and keV cm−3(l/100 kpc)1/2, respectively. The fractional 1σ
statistical errors are about 5 per cent.
1 10 100
10−3
0.010.1
ne
(cm−3
)
Radius (kpc)
east
north
south
west
southeast
1 10 100
11025
kT(keV)
Radius (kpc)
east
north
south
west
southeast
1 10 100
0123
Fe(Solar)
Radius (kpc)
east
north
south
west
southeast
deprojected
Figure 7.7: Radial proﬁles of the deprojected electron density and temperature distribution
determined along ﬁve different azimuths (the azimuth angles are measured
counterclockwise from the west): west (305–30 degress), north (30–110), east (110–205),
southeast (205–260), and south (260–305). For the iron abundance distribution, in the
right panel, we show the projected measurements and the azimuthally averaged deprojected
proﬁle.
128 Chandra study of the Ophiuchus cluster
1 10 100
0.010.11
P(keVcm−3
)
Radius (kpc)
east
north
south
west
southeast
1 10 100
101001000
kT/n2/3
(keVcm2
)
Radius (kpc)
east
north
south
west
southeast
Figure 7.8: Using the electron density ne and temperature kTe proﬁles shown in Fig. 7.7,
we determine the radial proﬁles of the total pressure (P = nkT; left panel) and entropy
(K = kTe/n2/3
e ; right panel).
we ﬁnd that the center of the large scale emission distribution is offset by 38 kpc to the
north-northwest of the central emission peak and the brightest cluster galaxy (BCG).
In the residual image, obtained by dividing our Chandra image with the best ﬁt elliptical
double beta-model (see the right panel of Fig. 7.1), the tail-like excess emission
extending north-northwestward practically disappears, indicating that this apparent
excess emission is the result of the bright cluster core being displaced southward of the
center of the global emission distribution. The residual image reveals a surface brightness
excess at the projected distance of r ∼ 280 kpc to the south of the BCG, which
coincides with some bright galaxies identiﬁed in 2MASS images. Its southern edge has
a hint of a sharp bow-shock shaped morphology. The residual image also reveals a
surprising, sharp surface brightness discontinuity that is curved away from the core,
at r ∼ 120 kpc (3.55 arcmin) to the southeast of the BCG. Fitting the surface brightness
distribution of the discontinuity with a broken power-law model, we ﬁnd a steepening
from γ1 = 0.3 to γ2 = 2.6.
The cluster core harbors a series of nested cold fronts (previously discussed in Ascasibar
& Markevitch 2006; Million et al. 2010a) that surround the cool core (see Fig. 7.2).
The outermost cold front wraps around the core from the east to the west and extends
out to r ∼ 43 kpc in the south. The best ﬁt density discontinuity at its southern part is
nj = 1.7 and the 99 per cent upper limit on its width is 5.3 kpc. Southeast of the cluster
core, the surface brightness discontinuity appears blurred and the GGM ﬁltered image
highlights that the cold front is broken up. The GGM ﬁltering also highlights multiple
edges to the south of the core seen in the Chandra image.
The sharpest, most prominent cold front extends from the northeast to the southwest
of the BCG at r ∼ 4.5–10 kpc, and has a northward ‘tongue’-like extension (see
Fig. 7.3). Its best ﬁt density jump is nj = 2 and the front appears remarkably sharp. We
ﬁnd that the 99 per cent upper limit on its width is 1.5 kpc.
7.3 Results 129
-0.5 -0.4 -0.3 -0.2 -0.1 0.00049 0.1 0.2 0.3 0.4 0.5
Figure 7.9: Top: Underlying models of the “unperturbed” surface brightness distribution
of the Ophiuchus cluster. From left to right: spherically-symmetric β-model,
patched β-models with σ = 70 arcsec (removes large-scale asymmetry), σ = 30 arcsec,
and σ = 10 arcsec (see Zhuravleva et al. 2015, for details). Bottom: residual images of
the SB ﬂuctuations in the Ophiuchus cluster obtained from the initial image divided by
the underlying model on the corresponding upper panel. The smaller the σ, the smaller
the structures included to the model and the less structures remain in the residual im-
age.
The compact, bright innermost cluster core is displaced from the center of the BCG
by 2.2 kpc in projection (see also Hamer et al. 2012). The compact core has a surface
brightness discontinuity at its west side and an extension, a possible Kelvin-Helmholtz
roll, to the north. In contrast to the other nearby cooling cores, the core of the Ophiuchus
cluster does not show the presence of any obvious X-ray cavities produced by
the active galactic nucleus (AGN). However, the presence of very small cavities, with
radius of up to 1.5 kpc, to the southeast of the core cannot be ruled out.
7.3.2 X-ray spectroscopy
Our temperature map in Fig. 7.4 reveals a remarkably steep temperature gradient in
the core of the cluster, with the azimuthally averaged temperature increasing from
∼ 2 keV in the core of the cluster up to ∼ 8 keV at r ∼ 20 kpc. Outside r ∼ 30 kpc the
temperature distribution is remarkably uniform with kT ∼ 9.5–10.5 keV.
Zooming-in on the innermost r = 10 kpc core of the cluster, using lower signal-tonoise
(S/N=50) maps, the high temperature gradient in the core becomes even clearer
(see Fig. 7.5). Along the northwestern direction, the gradient of the projected temperature
reaches dkT/dr ∼ 1 keV/kpc. While we model the ICM using a singletemperature
plasma model, the plasma in the innermost cluster core is multiphase.
130 Chandra study of the Ophiuchus cluster
Figure 7.10: The amplitude of velocity ﬂuctuations in the 1.5–5 arcmin annulus in the
Ophiuchus cluster obtained by assuming that the amplitudes of the density and velocity
ﬂuctuations are proportional to each other on each scale (see the text and Zhuravleva
et al. 2014a; Gaspari et al. 2014, for detail). In order to determine the amplitude of
density ﬂuctuations, the large scale surface brightness gradient has been modeled with
a spherically-symmetric β-model (top hatched region) and with more ﬂexible models
shown in Fig. 7.9. While the top hatched region is clearly affected by the large-scale
asymmetry in the cluster, all of the models that account for the asymmetry indicate velocities
V1D,k 100 km s−1 on scales 100 kpc. The dashed line indicates the expected
shape of a Kolmogorov spectrum.
7.3 Results 131
The soft emission of a multiphase plasma biases the best ﬁt hydrogen column density,
NH, low. The central panel of Fig. 7.5 shows low best ﬁt NH in the central four spatial
bins, at radii r 3.5–4.5 kpc. Outside this innermost region, the best ﬁt line-of-sight
absorbing hydrogen column density distribution is relatively uniform and its value,
NH ∼ 4 × 1021 cm−2, is approximately two times higher than the column densities
measured by the Leiden-Argentine-Bonn survey of Galactic H I (Kalberla et al. 2005).
The difference is probably due to molecular gas and dust along our sight-line close
to the Galactic plane. Fixing the NH to 4 × 1021 cm−2 in our model (the best ﬁt value
determined in regions outside the cluster core), we ﬁt a two-temperature model to the
spectra extracted from the four central regions (regions where Fig. 7.5 shows a low NH).
With respect to a ﬁt with a single temperature plasma and free NH, for one additional
free parameter, the χ2 of the ﬁt improves by 210 for 4693 degrees of freedom. This
best ﬁt model indicated the presence of a 0.82 ± 0.02 keV and a 2.75 ± 0.09 keV plasma
phase with a metallicity of 2.14 ± 0.26 Solar. Assuming the 0.82 keV phase is radiatively
cooling, we also ﬁt a model of a collisionaly ionized plasma and a cooling ﬂow, cooling
from 0.82 keV, and ﬁnd a best ﬁt mass deposition rate of ˙M = 0.97 ± 0.12 M yr−1.
Similarly to other systems, such as M 87 (Werner et al. 2010), Perseus cluster (Sanders
& Fabian 2007), Centaurus cluster (Sanders et al. 2016), the coolest gas phases are associated
with Hα ﬁlaments in the core of the cluster (Hamer et al. 2012).
The metallicity map in the right panel of Fig. 7.4 reveals a strongly centrally peaked
iron abundance distribution. The apparent metallicity dip in the innermost cluster core
(see right panel of Fig. 7.5) disappears when ﬁt with a two-temperature model, which
indicates a relatively high central metallicity of 2.14 ± 0.26 Solar (see also the previous
paragraph). The metallicity distribution shows a relatively sharp discontinuity at
the southeastern edge of the outer cold front at r ∼ 43 kpc and a trail of excess iron
abundance extends to the north of the core.
The projected entropy distribution is asymmetric (see the left panel of Fig. 7.6), with
a clear discontinuity at the southern cold front and a low entropy ‘tail’ extending northward.
While, as indicated by our deprojection analysis (see the next paragraph), the
true pressure peaks at the BCG, the projected pressure in the bright cluster core is
biased low (see the right panel of Fig. 7.6). This bias is due to the underestimated
density, which is the result of the adopted constant line-of-sight bin size1. The detected
azimuthal variations are, however, robust. Similar to the entropy distribution,
the pressure distribution also shows a prominent excess to the north of the core.
Taking the X-ray surface brightness peak as the center of the system and assuming
spherical symmetry, we determine the radial proﬁles of the deprojected spectral
properties along ﬁve different azimuths, chosen based on the observed morphological
characteristics (the azimuth angles are measured counterclockwise from the west):
west (305–30 degress), north (30–110), east (110–205), southeast (205–260), and south
(260–305). Fig. 7.7 shows the deprojected electron density (ne) and temperature (kTe)
proﬁles and both the projected and deprojected iron abundances distributions (the de-
1When looking at a spherically symmetric cluster with a centrally peaked density distribution, the
effective line-of-sight length, from which the dominant fraction of photons is emitted, is smaller in the
core and increases with radius.
132 Chandra study of the Ophiuchus cluster
Figure 7.11: Left panel: The BCG harbors a relatively faint, point-like, unresolved radio
source with a 30 mJy ﬂux density at 1.4 GHz. The white ellipse in the bottom left corner
shows the beam size of 1.17 ×2.48 . Central panel: A peculiar, ﬁlamentary radio source,
located about 360 kpc to the north of the cluster core (see also Murgia et al. 2010). This
object with no optical or X-ray counterpart is a likely radio phoenix, a source revived
by adiabatic compression by gas motions in the ICM. Right panel: The spectral index
map of the likely radio phoenix shows a relatively steep spectrum across the source.
projection analysis introduces a substantial noise into the measured Fe abundance values).
Using these measurements, we determine the total pressure (P = nkT) and entropy
(K = kTe/n2/3
e ) proﬁles shown in Fig. 7.8. In the innermost r ∼2 kpc region, the
best ﬁt iron abundance peaks at ∼ 2.3 Solar and the entropy drops to ∼ 2.6 keV cm2 .
7.3.3 The power-spectra of surface brightness ﬂuctuations
Following the methods described in Churazov et al. (2012), Zhuravleva et al. (2015,
2016), and Ar´evalo et al. (2015), we perform an analysis of the power-spectra of the
surface brightness ﬂuctuations in the Ophiuchus cluster. Using theoretical arguments
supported by numerical simulations, Zhuravleva et al. (2014b) and Gaspari et al. (2014)
show that in galaxy clusters, where the gas motions are subsonic, the root mean squared
amplitudes of the density and line-of-sight-component velocity ﬂuctuations are proportional
to each other on each scale, δρk/ρ0 ≈ ηV1D,k/cs, where ρ0 is the unperturbed
gas density, cs is the sound speed and η ≈ 1 is a proportionality coefﬁcient. Cosmological
simulations show that η ≈ 1 with a scatter of 30 per cent (Zhuravleva et al.
2014b) and hydrodynamic simulations conﬁrm that the proportionality coefﬁcient is of
the order of unity as long as conduction is completely suppressed (Gaspari et al. 2014).
To avoid complications due to the steep temperature gradient in the core of the cluster,
we analyzed an image in the annular region of r = 1.5–5 arcmin around the surface
brightness peak, where the ICM appears approximately isothermal (see the left panel
of Fig. 7.4). The image was extracted in the 1.5–3.5 keV energy range, where the emissivity
is approximately independent of temperature and proportional to the density
squared. We ignored the soft band where we are affected by the high Galactic column
density. To remove the global surface brightness gradient, we divided the cluster
7.3 Results 133
Figure 7.12: The Ophiuchus cluster also harbors two bright narrow-angle tail radio
galaxies (see also Murgia et al. 2010). The radio emitting tails of these infalling galaxies
are over 60 kpc long.
image by a model of an “unperturbed” surface brightness distribution (see Fig. 7.9).
Our simplest assumption is a spherically symmetric β-model which, however, does
not account for the prominent global asymmetry. To remove the asymmetry, as shown
in Fig. 7.9, we experimented with β-models patched on several different scales (see
Zhuravleva et al. 2015, for details).
Assuming volume ﬁlling turbulence and a sound speed of 1500 km s−1, we calculate
the velocity ﬂuctuations from the density ﬂuctuations obtained by using a range of
models for the underlying global density distribution. We ﬁnd that after accounting for
the azimuthal variation in the underlying model, the derived volume ﬁlling velocities
are V1D,k 100 km s−1 on scales 100 kpc.
To determine the nature of the southern surface brightness excess (see Fig. 7.1), we
used a method introduced in Ar´evalo et al. (2015) and Zhuravleva et al. (2016), and
analyzed the statistical properties of the emissivity ﬂuctuations in two energy bands:
1.5–3.5 keV and 3.5–7.5 keV. For the gas temperatures in the Ophiuchus cluster, we
would expect that while ‘isobaric’ gas perturbations, associated with slow displacement
of gas, will result in a ratio of the power-spectra derived in the hard and soft
energy bands of R ∼ 0.9, ‘adiabatic’ perturbations, associated with weak shocks, will
result in R ∼ 1.1. Our measured ratios, after accounting for the large statistical uncertainties,
are R > 1 consistent with ‘adiabatic’ perturbations, suggesting the possible
presence of a weak shock or a separate sub-halo moving through the cluster. The statistics
associated with this feature is, however, low and a more direct conﬁrmation of the
nature of this feature will require deeper observations.
134 Chandra study of the Ophiuchus cluster
7.3.4 Radio properties
We detect a point-like radio source associated with the center of the BCG with a ﬂux
density of 30 mJy and spectral index α = −0.55. The source is both compact and
optically thin, typical of very low level accretion. As can be seen from the left panel of
Fig. 7.11, we ﬁnd no evidence of extended structure near the BCG above the rms level
of 10 µJy. The upper limit on the polarization of the source is 0.2 per cent.
Our radio data also show two head-tail radio galaxies (see Fig. 7.12) previously studied
by Murgia et al. (2010) and Govoni et al. (2010). However, the brightest (630 mJy)
and perhaps most interesting radio source in the Ophiuchus cluster is a peculiar, ﬁlamentary
object (see the central panel of Fig. 7.11), located about 360 kpc to the north
of the cluster core. It is about 200 arcsec long (120 kpc at the redshift of the cluster)
and has no optical or X-ray counterpart. The object has been previously identiﬁed in
GMRT data by Murgia et al. (2010) as a possible unusual radio relic. We ﬁnd a steep
spectral index of α ∼ −1.8 to −2.5 that is relatively uniform across the source (see the
right panel of Fig. 7.11). This index is steeper than the slope of α = −1.01 ± 0.03 measured
between 74 MHz and 1.4 GHz by Murgia et al. (2010), indicating the presence of
a break at ∼ 1 GHz. The source shows signs of polarization, with the brightest knot,
located near the middle of the source, being polarized to roughly 18 per cent.
The core of the Ophiuchus cluster also harbors a faint radio minihalo (see Govoni
et al. 2009; Murgia et al. 2009, and our discussion below), however, our high resolution
JVLA observation in the A conﬁguration is not sensitive to the faint extended emission.
7.4 Discussion
7.4.1 Dynamical activity and the truncated cooling core
The ∼ 38 kpc offset of the centroid of the large scale X-ray emission distribution from
the central BCG and the prominent southern surface brightness excess indicate that
the gravitational potential in the cluster core has been strongly perturbed. The maps
of projected thermodynamic properties support this picture. In a galaxy cluster in approximate
hydrostatic equilibrium, with subsonic gas motions, the pressure distribution
traces the gravitational potential of the system. However, contrary to most cooling
core clusters, the pressure distribution in the Ophiuchus cluster is highly azimuthally
asymmetric around the BCG and displays a prominent region of excess pressure in the
north. These results strongly indicate that the BCG, which presumably traces the dark
matter density peak, is offset from the cluster’s global center of mass. Such a perturbed
gravitational potential is most likely the result of a relatively recent encounter with an
in-falling sub-cluster.
The multiple concentric cold fronts in the core of the cluster are also most likely
due to gas sloshing in a changing gravitational potential (e.g. Ascasibar & Markevitch
2006; Markevitch & Vikhlinin 2007) which, following an encounter with a sub-cluster,
is expected to be evolving towards a global equilibrium. While the southern outer cold
front, at r ∼ 43 kpc from the BCG, indicates southward gas motions, the prominent
7.4 Discussion 135
eastern inner cold front, at r ∼ 5 kpc from the BCG, shows that the gas in the cluster
core is swinging northeastward, most likely following the innermost part of the dark
matter potential. As the dense low entropy ICM in the cluster core moves through the
ambient ICM, it encounters ram-pressure and, as proposed by Million et al. (2010a),
the outer layers of the cool core get stripped, resulting in the highly unusual steep
temperature and metallicity gradients (for comparison see the entropy proﬁles in Cavagnolo
et al. 2009; Panagoulia et al. 2014). The entropy and metallicity distributions
in Fig. 7.4–7.6 also show clear evidence for such ram-pressure stripping. On the large
scale maps, we see an excess of low-entropy, high-metallicity gas northwest of the core
of the cluster. On smaller scales, we see a similar excess along a perpendicular axis, to
the southwest. In the innermost cluster core (see Fig. 7.5), we see an excess of cooler
gas to the southeast. The observed disruption of the cool core in the Ophiuchus cluster
is particularly interesting in the light of simulations, which indicate that cool cores
are remarkably resilient and survive late major mergers (Burns et al. 2008; Poole et al.
2008).
The southern surface brightness excess, seen at the bottom of the residual image in
the right panel of Fig. 7.1, is an obvious candidate for a sub-cluster that could have
encountered the cluster core relatively recently. The appearance of its southern edge is
suggestive of a bow-shock like shape, which would indicate a merger taking place
close to the plane of the sky. Furthermore, as discussed in Section 7.3.3, the cross
spectrum analysis also suggests the possible presence of a weak shock. Assuming a
merger in the plane of the sky, occurring at the sound speed in the kT = 8.5 keV ICM
(cs = 1500 km s−1), we estimate that the closest passage to the cluster core would have
occurred less than ≈ 200 Myr ago. A recent study of spectroscopic redshifts by Durret
et al. (2015) indicates that the 1.1 × 1015 M Ophiuchus cluster is not undergoing a major
merger and the BCG is close to the center of the global gravitational potential, where
the BCG velocity is consistent with the mean cluster velocity ∆v = 47 ± 97 km s−1.
However, a merger with a M ∼ 1014 M system would still be consistent with the
optical data.
Although our data clearly show that the Ophiuchus cluster is dynamically active,
it is unlikely that the truncation of the cool core has been caused by a single, recent
merging event. Comparison to numerical simulations (e.g. Ascasibar & Markevitch
2006; ZuHone et al. 2011; Roediger et al. 2011) indicates that the observed complex,
developed sloshing patterns are likely to have been triggered by an encounter over 1
Gyr ago. The merger with the southern sub-cluster might therefore be the latest in a
chain of systems merging with the Ophiuchus cluster.
7.4.2 The radio phoenix
Another indication of dynamical activity in the Ophiuchus cluster are the two narrowangle
tail radio sources (see Fig. 7.12 and Murgia et al. 2010) which are infalling towards
the cluster center, and the peculiar, ﬁlamentary radio source detected at r ∼
360 kpc (10 arcmin) north of the BCG (see the central panel of Fig. 7.11). This reliclike
source has no optical/near-infrared or X-ray counterpart. The object lacks a clear
radio core, and its spectral index of α ∼ −2 is similar to that of radio relics (see also
136 Chandra study of the Ophiuchus cluster
Murgia et al. 2010). However, the relatively small size (about 120 × 30 kpc), unusual
morphology, high surface brightness, and the central location within the Ophiuchus
cluster make it unlikely that this source is a radio relic. It appears more similar to radio
phoenixes (Kempner et al. 2004; van Weeren et al. 2009; de Gasperin et al. 2015), sources
revived by adiabatic compression by a shock or gas motions propagating through the
ICM as shown by Ensslin & Gopal-Krishna (2001) and Ensslin & Br¨uggen (2002). The
vortex-like structure at its northern tip might be associated with a L ∼ 15 kpc turbulent
eddy in the ambient ICM. Contrary to the recently identiﬁed radio phoenix in
Abell 1033 (de Gasperin et al. 2015), the source in the Ophiuchus cluster shows signs of
polarization, which may be due to its larger distance from the cluster center resulting
in a smaller amount of intervening matter.
7.4.3 The cold fronts
The inner eastern cold front (Fig. 7.3) is remarkably sharp, with a width smaller than
1.5 kpc. This is the tightest measured upper limit on a physical width of a cluster cold
front. However, because the ICM density in the cluster core is high, the Coulomb mean
free path for particles diffusing from the cooler bright, dense side of the discontinuity
to the hotter outer ICM is about an order of magnitude smaller than this upper limit.
Diffusion has previously been shown to be suppressed at cold front interfaces with
upper limits in Abell 3667 and the Virgo cluster smaller or comparable to the Coulomb
mean free path (Vikhlinin et al. 2001b; Markevitch & Vikhlinin 2007; Werner et al. 2016).
The ‘tongue’-like extension to the north (Fig. 7.3) may be due to the onset of a
Rayleigh-Taylor instability, which develops when the displaced, dense gas subjected to
ram-pressure from the ambient ICM spreads sideways, eventually sprouting a tongue
of low entropy material which separates and starts sinking towards the minimum of
the gravitational potential. Note the remarkable morphological similarity of this feature
(although on a much smaller physical scale) to the results of the simulation in
Fig. 7 of Ascasibar & Markevitch (2006).
The outer southern cold front appears signiﬁcantly more disturbed, with multiple
edges and azimuthal variation. Fig. 7.2 shows that while the southern edge of the
front is relatively sharp (width smaller than 5.3 kpc), its southeastern side appears
smeared. Azimuthal variations and smearing has also been observed at the prominent
cold fronts in the Virgo and Centaurus clusters by Werner et al. (2016) and Sanders et al.
(2016). They propose that the smearing is most likely due to Kelvin-Helmholtz instabilities,
which have previously also been observed in NGC 7618 and UGC 12491 (Roediger
et al. 2012). Constrained hydrodynamic simulations reproducing the Virgo cluster
by Roediger et al. (2013) demonstrate that viscosities of about 10 per cent of the Spitzer
value or larger are very efﬁcient at preventing the development of Kelvin-Helmholtz
instabilities, indicating that the viscosity is suppressed in that system. Magnetohydrodynamic
simulations by ZuHone et al. (2011, 2013) show that magnetic ﬁelds provide
only partial protection against hydrodynamic instabilities. The apparent breakup
of the southern cold front is thus consistent with the previous results which suggest
that the viscosity of the ICM is suppressed with respect to the temperature dependent
Spitzer value.
7.4 Discussion 137
While the multiple surface brightness edges to the south of the cluster center, seen
in the right panel of Fig. 7.2, could be due to Kelvin-Helmholtz instabilities, as discussed
by Roediger et al. (2013), they could also be due to gas depletion by ampliﬁed
magnetic ﬁelds underneath the cold front, as discussed in Werner et al. (2016). Simulations
of ZuHone et al. (2011, 2015) show that gas sloshing may result in wide bands
of ampliﬁed magnetic ﬁelds extending relatively far below cold fronts. The increased
magnetic pressure in such layers may push gas out, decreasing its density, creating
surface brightness edges associated with the magnetized bands.
7.4.4 Turbulence and thermal conduction in the hot ICM
Even though our observations clearly show that the core of the Ophiuchus cluster is
strongly dynamically active, the amplitude of density ﬂuctuations in the r = 1.5–5 arcmin
(53–177 kpc) region is relatively low, indicating velocities smaller than ∼ 100
km s−1 on scales 100 kpc. Given that the cooling core is offset from the global center
of mass and the sub-cluster that initiated the unusually strong sloshing is likely to
have passed close to the cluster center, we would expect the gas motions associated
with the dynamical activity to be higher and to have cascaded down to a volume ﬁlling
turbulence in the investigated region. Moreover, such low ICM velocities would be
surprising in a disturbed cluster with a galaxy velocity dispersion of ∼ 954 ± 58 km s−1
(Durret et al. 2015). Viscosity is unlikely to suppress gas motions on the relatively large
scales that we are probing (see e.g. Gaspari et al. 2014), and the morphology of cluster
cold fronts indicates that the viscosity in the ICM is likely lower than the temperature
dependent Spitzer value by more than an order of magnitude (see Sect. 7.4.3 and
Roediger et al. 2013, 2015; Werner et al. 2016). Our relation between the power spectrum
of the density ﬂuctuations and the power-spectrum of velocity ﬂuctuations was
derived under the assumption that thermal conduction in the ICM is fully suppressed.
However, thermal conduction could damp the density ﬂuctuations by a factor ∼ 2–
3 on the observed scales, while leaving the velocity cascade unaltered (Gaspari et al.
2014). Such damping would then result in our underestimate of velocities.
The large temperature gradient, of the order of dkT/dr ∼ 1 keV/kpc, in the core of
the Ophiuchus cluster indicates that the thermal conduction is suppressed along the
radial direction. Such suppression might be the result of azimuthal magnetic ﬁelds
that will, in the absence of signiﬁcant turbulence, arise in cooling cores due to heatﬂux
driven buoyancy instability (HBI; Quataert 2008; Parrish & Quataert 2008; Parrish
et al. 2009, 2010). Gas sloshing can also result in azimuthally aligned magnetic ﬁelds
(ZuHone et al. 2011). Observations of cold fronts (Ettori & Fabian 2000; Xiang et al.
2007) and merging clusters of galaxies (Markevitch et al. 2003) also indicate that thermal
conduction in the ICM is signiﬁcantly suppressed (by a factor of order ∼ 102) with
respect to the Spitzer value. The alignment of magnetic ﬁeld lines alone, however,
would not explain the observed suppression of the heat ﬂux. Comparisons of numerical
simulations that include anisotropic conduction to observations indicate that
thermal conduction is also suppressed by over an order of magnitude along the magnetic
ﬁled lines (ZuHone et al. 2013; Werner et al. 2016). However, because thermal
conduction is a strong function of temperature (∝ T5/2), even a suppressed conduc-
138 Chandra study of the Ophiuchus cluster
tivity would be relatively efﬁcient in the ∼ 9 keV ICM outside of the cooling core of
the Ophiuchus cluster. Conduction could, in principle, contribute both to the observed
isothermality and to the damping of density ﬂuctuations. It would suppress the perturbations
preferentially on smaller scales, steepening the observed power-spectrum.
Although the observed velocity power-spectrum in Fig. 7.10 appears steeper than the
Kolmogorov spectrum, the large systematic uncertainties due the subtraction of the
underlying model do not allow us to make a ﬁrm conclusion about the slope.
A relatively low level of surface brightness ﬂuctuations is also measured in the
kT ∼ 3.5 keV AWM 7 cluster, where conduction is expected to be much less signifﬁcant
(Sanders & Fabian 2012). It is, however, possible that this system is dynamically
more relaxed and does therefore not harbor strong volume ﬁlling turbulence. Conduction
would probably also have a less signiﬁcant effect on the surface brightness
ﬂuctuations in the cool cores of the Perseus and Virgo clusters (see Zhuravleva et al.
2014a, 2015; Ar´evalo et al. 2015). Comparison of the power-spectra of surface brightness
ﬂuctuations and direct line broadening measurements with the Hitomi satellite
will soon provide important quantitative constraints on transport processes and the
microphysics of the ICM (see the discussion in Gaspari et al. 2014; ZuHone et al. 2015).
7.4.5 The concave surface brightness discontinuity: giant AGN outburst
or merger related gas dynamics?
Concave surface brightness discontinuities, similar to the one observed at r ∼ 120 kpc
southeast of the core of the Ophiuchus cluster, but on smaller physical scales, are also
observed in the Perseus cluster, Abell 1795, Abell 2390, and Centaurus cluster, where
they were interpreted as the inner walls of cavities resulting from AGN activity (Walker
et al. 2014; Sanders et al. 2016). Low frequency radio observations of the Ophiuchus
cluster performed by the GMRT also reveal steep spectrum radio emission beyond this
feature (see the radio source E in Murgia et al. 2010). The radius of curvature of the observed
concave feature is ∼ 180 kpc, therefore, if it is associated with the inner wall of
a spherical cavity, then the pV work required to displace the ICM is about 5 × 1061 ergs
(comparable with the most powerful AGN outburst known in MS 0735.6+7421; McNamara
et al. 2005; Vantyghem et al. 2014). However, a buoyantly rising AGN inﬂated
cavity would more likely have an ellipsoidal shape and a smaller size, requiring less
but still a substantial amount of AGN power. Such a powerful AGN outburst would
heat and displace a large amount of ICM in the cluster core and, in combination with
the ongoing merger, contribute siniﬁcantly to the disruption of the cooling core (see
e.g. Ehlert et al. 2011b).
However, the extremely steep metallicity and entropy gradients in the Ophiuchus
cluster would most likely get erased by such a powerful AGN outburst, arguing against
this scenario. It is more likely that the discontinuity is the result of merger related
gas dynamics. Similar concave discontinuities appear at certain stages of mergers in
simulations of infall of gas-rich subclusters into cooling core clusters (see Fig. 22 in
Ascasibar & Markevitch 2006).
7.5 Conclusions 139
7.4.6 Suppressed AGN feedback
Because of the ram pressure experienced by the ICM (in contrast to the stars and dark
matter associated with the BCG, which are effectively collisionless), the gas motions in
the changing gravitational potential well can lead to the separation of the ICM density
peak from the BCG. Such a separation is clearly observed in the Ophiuchus cluster,
where the X-ray surface brightness peak is displaced from the center of the BCG by
2.2 kpc in projection and spatially coincides with Hα nebulae. This dense displaced
gas displays multi-temperature structure consistent with a radiative cooling rate of
˙M = 0.97 ± 0.12 M . Because this gas is displaced from the vicinity of the AGN, it can
continue cooling without triggering an AGN feedback response. Similar cooling, offset
from the AGN, has been seen in Sersic 159-03, Abell 3444 and Abell 1991, Abell 2146
(e.g. Werner et al. 2011; Hamer et al. 2012; McDonald et al. 2015; Russell et al. 2012; Canning
et al. 2012). The lack of observed X-ray cavities in the cluster core and the weak
point-like radio emission, lacking lobes or jets, indicate that the AGN may currently
be largely dormant, which is consistent with the cooling taking place offset from the
central supermassive black hole. The innermost surface brightness discontinuity, just
2 kpc west of the X-ray peak, separates the cooling multiphase gas from the ambient
ICM.
Mazzotta & Giacintucci (2008) propose that turbulence driven by gas sloshing reaccelerates
preexisting cosmic rays from the central AGN producing diffuse radio minihaloes.
The presence of the radio mini-halo in the Ophiuchus cluster, discovered by
Govoni et al. (2009) using VLA data at 1.4 GHz in the D conﬁguration, is consistent
with this picture. Murgia et al. (2009) concluded that the emissivity of the mini-halo
of the Ophiuchus cluster (along with the mini-haloes in Abell 1835 and Abell 2029) is
low, similar to the emissivities of radio haloes of merging clusters rather than to other
previously known mini-haloes. Within the framework of the turbulent re-acceleration
scenario, the relatively low surface brightness of the mini-halo in such a dynamically
active cluster core implies the lack of a substantial amount of preexisting cosmic rays,
consistent with a picture of a largely dormant AGN.
7.5 Conclusions
• The Ophiuchus cluster hosts a truncated cool core, with a temperature increasing
from kT ∼ 1 keV in the center to kT ∼ 9 keV at r ∼ 30 kpc. Beyond r ∼ 30 kpc
the ICM appears remarkably isothermal.
• The core is dynamically disturbed with multiple sloshing induced cold fronts,
with indications for both Rayleigh-Taylor and Kelvin-Helmholtz instabilities. The
sloshing is the result of the strongly perturbed gravitational potential in the cluster
core, with the BCG displaced southward from the global center of mass.
• The residual image reveals a likely sub-cluster south of the core at the projected
distance of r ∼ 280 kpc.
140 Chandra study of the Ophiuchus cluster
• The cluster harbors a peculiar, ﬁlamentary radio source with no optical or X-ray
counterpart, located about 360 kpc to the north of the cluster core. The object is a
likely radio phoenix, a source revived by adiabatic compression by gas motions
in the ICM.
• Even though the Ophiuchus cluster is strongly dynamically active, the amplitude
of density ﬂuctuations outside of the cooling core is low, indicating velocities
smaller than ∼ 100 km s−1. The density ﬂuctuations might be damped by
thermal conduction in the hot and remarkably isothermal ICM, resulting in our
underestimate of gas velocities.
• We ﬁnd a surprising, sharp surface brightness discontinuity that is curved away
from the core, at r ∼ 120 kpc to the southeast of the cluster center. We conclude
that this feature is most likely due to gas dynamics associated with a merger and
not a result of an extraordinary AGN outburst.
• The cooling core lacks any observable X-ray cavities and the AGN only displays
weak, point-like radio emission, lacking lobes or jets, indicating that currently it
may be largely dormant. The lack of strong AGN activity may be due to bulk
of the cooling taking place offset from the central supermassive black hole. The
observed unchecked cooling where a largely dormant AGN is offset from the
cooling ﬂow solidiﬁes the idea that AGN play a key role in maintaining the cooling/heating
balance in cluster cores.
Acknowledgments
Support for this work was provided by the National Aeronautics and Space Administration
through Chandra Award Number GO4-15126X issued by the Chandra X-ray
Observatory Center, which is operated by the Smithsonian Astrophysical Observatory
for and on behalf of the National Aeronautics Space Administration under contract
NAS8-03060. This work was supported in part by the US Department of Energy under
contract number DE-AC02-76SF00515. The authors thank Georgiana Ogrean and John
ZuHone for discussions, and Evan Million and Steven Ehlert for their help with the
observing proposal.
Chapter 8
The nature of ﬁlamentary cold gas in
the core of the Virgo Cluster
N. Werner1, J. B. R. Oonk2, R. E. A. Canning1, S. W. Allen1,3, A. Simionescu1, J. Kos2,4,
R. J. van Weeren5, A. C. Edge6, A. C. Fabian7, A. von der Linden1, P. E. J. Nulsen5, C. S. Reynolds8,
M. Ruszkowski9,10
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall, Stanford, CA 94305-4085, USA
2ASTRON, Netherlands Institute for Radio Astronomy, P.O. Box 2, 7990 AA Dwingeloo,
The Netherlands
3SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025,
USA
4Faculty of Mathematics and Physics, University of Ljubljana, Jadranska 19, 1000 Ljubljana,
Slovenia
5Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA
02138, USA
6Institute for Computational Cosmology, Department of Physics, Durham University,
Durham, DH1 3LE, UK
7Institute of Astronomy, Madingley Road, Cambridge CB3 0HA, UK
8Department of Astronomy and the Maryland Astronomy Center for Theory and Computation,
University of Maryland, College Park, MD 20742, USA
9Department of Astronomy, University of Michigan, 500 Church Street, Ann Arbor, MI
48109, USA
10Michigan Center for Theoretical Physics, 3444 Randall Lab, 450 Church St, Ann Arbor,
MI 48109, USA
Published in the Astrophysical Journal, volume 767, pages 153–162, 2013
142 Filamentary cold gas in M 87
Abstract
We present a multi-wavelength study of the emission-line nebulae located ∼38 arcsec
(3 kpc in projection) southeast of the nucleus of M 87, the central dominant galaxy
of the Virgo Cluster. We report the detection of far-infrared (FIR) [C II] line emission
at 158 µm from the nebulae using observations made with the Herschel Photodetector
Array Camera & Spectrometer (PACS). The infrared line emission is extended and
cospatial with optical Hα+ [N II], far-ultraviolet C IV lines, and soft X-ray emission.
The ﬁlamentary nebulae evidently contain multi-phase material spanning a temperature
range of at least 5 orders of magnitude, from ∼ 100 K to ∼ 107 K. This material
has most likely been uplifted by the Active Galactic Nucleus from the center of M 87.
The thermal pressure of the 104 K phase appears to be signiﬁcantly lower than that of
the surrounding hot intra-cluster medium (ICM) indicating the presence of additional
turbulent and magnetic pressure in the ﬁlaments. If the turbulence in the ﬁlaments
is subsonic then the magnetic ﬁeld strength required to balance the pressure of the
surrounding ICM is B ∼ 30 − 70 µG. The spectral properties of the soft X-ray emission
from the ﬁlaments indicate that it is due to thermal plasma with kT ∼ 0.5–1 keV,
which is cooling by mixing with the cold gas and/or radiatively. Charge exchange
can be ruled out as a signiﬁcant source of soft X-rays. Both cooling and mixing scenarios
predict gas with a range of temperatures. This is at ﬁrst glance inconsistent
with the apparent lack of X-ray emitting gas with kT < 0.5 keV. However, we show
that the missing very soft X-ray emission could be absorbed by the cold gas in the
ﬁlaments with an integrated hydrogen column density of NH ∼ 1.6 × 1021 cm−2, providing
a natural explanation for the apparent temperature ﬂoor to the X-ray emission
at kT ∼ 0.5 keV. The FIR through ultra-violet line emission is most likely primarily
powered by the ICM particles penetrating the cold gas following a shearing induced
mixing process. An additional source of energy may, in principle, be provided by X-ray
photoionization from cooling X-ray emitting plasma. The relatively small line ratio of
[O I]/[C II]< 7.2 indicates a large optical depth in the FIR lines. The large optical depth
in the FIR lines and the intrinsic absorption inferred from the X-ray and optical data
imply signiﬁcant reservoirs of cold atomic and molecular gas distributed in ﬁlaments
with small volume ﬁlling fraction, but large area covering factor.
8.1 Introduction
Massive giant elliptical galaxies in the centers of clusters with central cooling times
shorter than the Hubble time frequently display spectacular optical emission-line nebulae
(e.g. Johnstone et al. 1987; Heckman et al. 1989; Donahue et al. 1992; Crawford
et al. 1999; McDonald et al. 2010). Associated cold molecular gas has also been detected
in many of these systems, from near-infrared (NIR) K-band H2 line emission (e.g. Jaffe
& Bremer 1997; Falcke et al. 1998; Donahue et al. 2000; Hatch et al. 2005; Jaffe et al. 2005;
Johnstone et al. 2007; Oonk et al. 2010) and CO observations (e.g. Edge 2001; Edge &
Frayer 2003; Salom´e & Combes 2003; McDonald et al. 2012). While the NIR spectra
show H2 line ratios characteristic of collisionally excited 1000–2000 K molecular gas,
8.1 Introduction 143
Table 8.1: Summary of observations and of the properties of the FIR lines integrated
over the full 5×5 spaxel (47 ×47 ) ﬁeld of view of Herschel PACS.
Line Wavlegth Exposure Flux Inst. FWHM Obs. FWHM Shift
(µm) (s) (10−14 erg s−1 cm−2) (km s−1) (km s−1) (km s−1)
C II 157.7 3000 5.13 ± 0.5 240 361 ± 27 −62 ± 11
O I 63.2 3312 < 36.71 85 ... ...
O IB 145.5 4480 < 2.9a 255 ... ...
the CO emission traces the coldest (<50 K) molecular gas phases. Optical Hα+[N II]
line emission arises from a thin ionized skin of 104 K gas on underlying reservoirs of
cold neutral and molecular gas. CO observations indicate that the amounts of molecular
gas in cluster cores can be large (Edge 2001), e.g. the cold gas mass in the center
of the Perseus Cluster is approaching 1011 M (Salom´e et al. 2006, 2008, 2011). The
observed level of star formation in these systems is, however, typically relatively small
(e.g. O’Dea et al. 2008). Despite signiﬁcant efforts to understand the origin and the excitation
mechanism of these nebulae, their nature, energy source, and detailed physics
remain poorly understood.
The relatively nearby (d ∼16.7 Mpc, Blakeslee et al. 2009) giant elliptical galaxy M 87
(NGC 4486), at the centre of the Virgo Cluster, harbors a well-known extended optical
Hα+[N II] ﬁlament system (Ford & Butcher 1979; Sparks et al. 1993). The proximity of
this system allows us to study the physics of the emission-line nebulae in greater detail
than is possible elsewhere. The Hα+[N II] nebulae in M 87 were found to spatially coincide
with ﬁlamentary soft X-ray emission (Young et al. 2002; Sparks et al. 2004), which
can be modeled as ∼0.5 keV plasma in collisional ionization equilibrium (Werner et al.
2010). The emission-line nebulae also spatially coincide with C IV line emission at farultraviolet
wavelengths (FUV), which typically arises in gas at temperature ∼105 K
(Sparks et al. 2009, 2012). C IV line emission has so far not been detected in extended
ﬁlaments in any other central cluster galaxy. Werner et al. (2010) showed that all of
the bright Hα and UV ﬁlaments in M 87 are found in the downstream region of a <3
Myr old shock front revealed by X-ray observations with Chandra (Million et al. 2010d).
This argues that the generation of Hα, UV, and soft X-ray emission in M 87 is coupled
to shocks in the hot X-ray emitting plasma. Based on these observations Werner et al.
(2010) proposed that shocks induce shearing around the cooler, denser gas ﬁlaments,
which promotes mixing with the ambient hot intra-cluster medium (ICM) via instabilities.
By enhancing the rate at which hot thermal particles come into contact with the
colder gas phases, mixing can in principle supply the power and the ionizing particles
needed to explain the NIR to FUV line emission (Ferland et al. 2008, 2009; Fabian et al.
2011).
The launch of the Herschel space observatory, with its unprecedented sensitivity to
far-infrared (FIR) line emission (Pilbratt et al. 2010), has opened new opportunities
to study the coldest gas phases in galaxies. Herschel has enabled the ﬁrst detections
of atomic cooling lines of [C II], [O I], and [N II] in the X-ray bright cores of Abell 1068,
144 Filamentary cold gas in M 87
Figure 8.1: The FIR [C II]λ157 µm line obtained from the central spaxel (9.4 ×
9.4 arcsec2) of the Herschel PACS rebinned data cube. The mean line centroid is blueshifted
with respect to M 87 (at z = 0.004360) by v = −123 ± 5 km s−1. The blue line
indicates the detector noise.
Abell 2597 (Edge et al. 2010), and the Centaurus and Perseus clusters (Mittal et al. 2011,
2012). These lines are the dominant cooling lines for neutral interstellar gas and can be
used as diagnostics to infer temperatures, densities, and radiation ﬁelds (e.g. Kaufman
et al. 1999). Because carbon is the fourth most abundant element in the Universe and
has a low ionization potential, the 158 µm [C II] line is the most ubiquitous and beststudied
(e.g. Malhotra et al. 1997, 2001). It is a tracer of gas with a temperature of
∼ 100 K.
To study the cold gas phases associated with the ﬁlamentary line emission nebulae
in M 87 we observed the regions of brightest Hα+[N II] ﬁlaments, extending to the
southeast of the nucleus of the galaxy, with the Herschel Photodetector Array Camera &
Spectrometer (PACS) at the wavelengths of [C II]λ157µm, [O I]λ63µm and [O IB]λ145µm.
Here we report the results of these observations, as well as a reanalysis of deep Chandra
X-ray, Hubble Space Telescope (HST) optical and UV data, and long slit spectra obtained
with the William Herschel Telescope (WHT). Sect. 9.2 describes the observations, data
reduction, and analysis of the Herschel PACS, HST, Chandra, and WHT data. In Sect.
9.3, we summarize the results, and in Sect. 9.4 discuss the implications of our observations
for the nature and the energy sources of the ﬁlaments. Our main conclusions
are summarized in Sect. 9.5. We assume a distance to M 87 of 16.7 Mpc (Blakeslee
et al. 2009), which implies a linear scale of 81 pc arcsec−1. The redshift of M 87 is
z = 0.004360.
8.1 Introduction 145
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
[CII] flux
1 kpc
0.1 0.15 0.2 0.25 0.3
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
[CII] velocity
1 kpc
-150 -100 -50 0 50 100 150
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
[CII] velocity dispersion
1 kpc
20 40 60 80 100 120 140 160
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
Northern filament
Southern filament
Ha from HST WFPC2
1 kpc
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
FUV from HST ACS/SBC
1 kpc
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
0.7-0.9 keV X-ray image
1 kpc
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
0.5 keV plasma
1 kpc
10 20 30 40 50 60 70
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
Region
1 keV plasma
1 kpc
100 200 300
187.722 187.720 187.718 187.716 187.714 187.712 187.710
12.39212.39012.38812.38612.38412.382
Right ascension
Declination
2 keV plasma
1 kpc
300 400 500 600 700 800
Figure 8.2: FIR [C II] maps (top panels; we only show spatial bins where the signal-tonoise
ratio of the integrated [C II] ﬂux is greater than 2) and Hα+[N II], FUV, and X-ray
images of the ﬁlamentary multiphase gas southeast of the nucleus of M 87. All images
show the same 47 × 47 arcsec region of the sky. The projected distance of the bright
ﬁlaments from the nucleus is 38 arcsec (3 kpc). Top left panel: Map of the integrated
[C II] line ﬂux in units of 10−14 erg s−1 cm−2 per 6 ×6 spaxel obtained with Herschel
PACS. Top central panel: The velocity distribution of the [C II] emitting gas, in units of
km s−1, relative to the systemic velocity of M 87, v = 1307 km s−1. Top right panel: Map
of the velocity dispersion, σ, of the [C II] emitting gas. Middle row left panel: Hα+[N II]
image obtained with HST WFPC2. The contours of these ﬁlaments are over-plotted
on the Herschel PACS maps in the top row. On this HST WFPC2 image, contours of
6 cm radio emission from Hines et al. (1989) are over-plotted in red. Middle row central
panel: FUV image showing C IV line emission obtained with HST ACS/SBC. Middle row
right panel: Chandra soft X-ray image, extracted in the 0.7–0.9 keV band. Bottom panels:
Spatial distribution of the emission measure, Y = nHnedV, (in 1058 cm−3 arcsec−2) of
the 0.5 keV (bottom left), 1.0 keV (bottom central), 2.0 keV (bottom right) plasma detected
at 99.7 per cent conﬁdence.
146 Filamentary cold gas in M 87
8.2 Observations and data analysis
8.2.1 Far Infrared spectroscopy with Herschel PACS
We observed the FIR cooling lines of [C II]λ157µm, [O I]λ63µm, and [O IB]λ145µm with
the PACS integral-ﬁeld spectrometer (Poglitsch et al. 2010) on the Herschel space observatory
(Pilbratt et al. 2010). The observations, with a duration of 10880 seconds, were
performed on January 4th, 2012 (Obs. ID: 1342236278). Table 9.2 gives a summary
of the observations. Columns list the observed lines, their rest frame wavelengths,
observation durations, the observed line ﬂuxes and the 2σ upper limits, the spectral
resolution full-width-at-half-maximum (FWHM) of the instrument at the given wavelength,
the observed FWHM, and the observed line shift with respect to the systemic
velocity of M 87 (v = 1307 km s−1).
The observations were taken in line spectroscopy mode with chopping-nodding to
remove the telescope background, sky background and dark current. A chopper throw
of 6 was used. All three line observations were taken in pointed mode centered on (α,
δ)=(12:30:51.5, +12:23:07) (J2000).
The observations were reduced using the HIPE software version 8.2.0, using the
PACS ChopNodLineScan pipeline script for pointed observations. This script processes
the data from level 0 (raw channel data) to level 2 (ﬂux calibrated spectral cubes).
During the ﬁnal stage of reduction the data were spectrally and spatially rebinned
into a 5×5×λ cube. In the following we will refer to this cube as the re-binned cube.
Each spatial pixel, termed spaxel, in this cube has a size of 9.4 ×9.4 . The cube thus
provides us with a ﬁeld of view (FoV) of 47 ×47 .
For the wavelength re-gridding we set the parameters oversample and upsample equal
to 2 and 1 respectively. This means that one spectral bin corresponds to the native
resolution of the PACS instrument (see the PACS Data Reduction Guide2 for further
information). Larger values for both these two parameters were investigated and did
not change the results.
For the [C II] line we projected the re-binned cube onto the sky using the specProject
task in HIPE and the hrebin task in IDL. In the following we will refer to this cube as
the projected cube. Upon projecting the observed [C II] data from the telescope frame
to the sky we have chosen a resolution of 6 in order to Nyquist sample the beam,
the FWHM of which is 12 at the observed wavelength of the line. We only consider
spatial bins where the signal-to-noise ratio of the integrated [C II] ﬂux is greater than 2.
8.2.2 HST Hα+[N II] and FUV data
To study the detailed morphology of the ﬁlamentary emission-line nebulae, we have
also analyzed Hα+[N II] images taken with the Wide Field Planetary Camera 2 (WFPC2)
through the F658N ﬁlter for 2700 seconds (proposal ID: 5122). The central wavelength
of the ﬁlter is at 6591 ˚A and its bandwidth is 29 ˚A. It transmits the Hα line at λ = 6563 ˚A,
2http://herschel.esac.esa.int/twiki/pub/Public/PacsCalibrationWeb/
PDRG Spec May12.pdf
8.3 Results 147
with transmission T = 0.80, and two [N II] lines at λ = 6584 ˚A, with T = 0.20 and at
λ =6548 ˚A, with T = 0.76, at the systemic velocity of M 87 (v = 1307 km s−1). We
subtracted the emission of the underlying stellar population of the galaxy using a red
continuum image taken with WFPC2 through the ﬁlter F814W (proposal ID: 5941).
A FUV image was obtained using the Advanced Camera for Surveys Solar Blind
Channel (ACS/SBC) (proposal ID: 11861) through the ﬁlter F150LP (1630 seconds),
which covers the C IV line at λ = 1549 ˚A (see Sparks et al. 2009).
8.2.3 Chandra X-ray data
Extensive Chandra X-ray observations of M 87 were made between July 2002 and November
2005 using the Advanced CCD Imaging Spectrometer (ACIS). The total net exposure
time after cleaning is 574 ks. The data reduction is described in Million et al.
(2010d). We extracted six background subtracted, ﬂat-ﬁelded, narrow band images between
0.3 keV and 2.0 keV with a 0.492 × 0.492 arcsec2 pixel scale (raw detector pixels).
To measure the detailed properties of the soft X-ray emission, we also extracted spectra
from regions determined using the Contour Binning algorithm (Sanders 2006), as described
in Werner et al. (2010). Spectral modeling has been performed with the SPEX
package (Kaastra et al. 1996) in the 0.5–7.0 keV band.
8.2.4 Long slit optical spectra
Long slit spectroscopy was performed using the Intermediate dispersion Spectrograph
and Imaging System (ISIS) at the 4.2 m William Herschel Telescope on the island of La
Palma on February 10, 2011. The total on-source integration time was 3500 s, split into
7 exposures, for the blue and red arm simultaneously. The data were reduced using
standard IRAF (Image Reduction and Analysis Facility) procedures. The CCD images
were bias subtracted, ﬂat ﬁelded, wavelength calibrated, ﬂux calibrated, and merged.
Spectra were shifted into the heliocentric velocity frame. Cosmic rays were removed
with the CRNEBULA subroutine, designed for diffuse objects. Blue and red arm images
were treated separately during the whole analysis.
8.3 Results
8.3.1 Cold and warm gas phases
We detect the ﬁlamentary emission-line nebulae across all wavebands studied, from
the FIR to the soft X-rays. Using Herschel PACS, we detect [C II]λ157µm line emission
(see Fig. 8.1) and determine 2σ upper limits for the [O I]λ63µm and [O IB]λ145µm
cooling lines. The properties of the FIR lines, spatially integrated over the 5×5 spaxel
(47 ×47 ) Herschel PACS ﬁeld of view, are summarized in Table 9.2. The 2σ upper
limits on the ﬂuxes of the [O I]λ63µm and [O IB]λ145µm lines given in Table 9.2 were
determined assuming that their velocity widths are equal to the FWHM of the [C II]
line.
148 Filamentary cold gas in M 87
27.5 30 32.5 35 37.5 40 42.5
Composite
H II
LINER
Seyfert
Log10([OIII]5007/Hβ)
−0.75
−0.5
−0.25
0
0.25
0.5
0.75
Log10( [N II]6583 / Hα )
−0.75 −0.5 −0.25 0 0.25 0.5 0.75
100 cm-3
30 cm-3
500 cm-3
[SII]λ6716/[SII]λ6731
0.75
1
1.25
1.5
1.75
2
Projected distance from nucleus / arcsec
27.5 30 32.5 35 37.5 40 42.5
Figure 8.3: Left panel: Optical diagnostic diagram (a BPT diagram, Baldwin et al. 1981)
showing the [O III]/Hβ against the [N II]/Hα ﬂux observed in the southern ﬁlament
at different radii from the nucleus. The brightening shades of gray indicate increasing
projected distance from the nucleus in arcsec. While the side of the ﬁlament
closest to the nucleus has spectra similar to Seyfert like AGN, at larger radii the gas
has a LINER-like emission spectrum. The solid and dashed lines are from Kewley
et al. (2006) and Osterbrock & Ferland (2006), respectively. Right panel: The measured
[S II]λ6716/[S II]λ6731 line ratios, a good probe of gas density, as a function of radius.
The full, dashed, and dotted lines indicate ratios corresponding to electron densities of
ne = 30 cm−3, ne = 100 cm−3, and ne = 500 cm−3, respectively (Osterbrock & Ferland
2006).
8.3 Results 149
The [C II]λ157µm line emission is extended and spatially coincident with the Hα+[N II]
ﬁlaments (see the top left panel of Fig. 8.2). We see two prominent ﬁlamentary structures,
which we will call the ‘southern ﬁlament’ and the ‘northern ﬁlament’. The
brightest southern ﬁlamentary structure falls on the central spaxel of the PACS spectrometer.
The [C II] ﬂux detected from the central spaxel (9.4 × 9.4 arcsec2) in the rebinned
cube is (8.5 ± 0.7) × 10−15 erg s−1 cm−2. The top central panel of Fig. 8.2 shows
a map of the velocity distribution of the [C II] line emitting gas with respect to the velocity
of M 87. This image shows an interesting dichotomy, with the northern ﬁlaments
receding at about +140 km s−1 and the bright southern ﬁlament (most of the emission
of which falls on the central 9.4 × 9.4 arcsec2 spaxel of the detector) moving in the opposite
direction along our line of sight at −123 ± 5 km s−1, with respect to M 87. These
velocities show good agreement with the radial velocities of the corresponding optical
ﬁlaments measured using long-slit spectroscopy (Sparks et al. 1993). The top right
panel of Fig. 8.2 shows a map of the velocity dispersion of the [C II] line emitting material.
While the northern, receding ﬁlaments show a signiﬁcant velocity dispersion of
σ = FWHM/2.355 ∼ 124 km s−1 (FWHM=
√
3772 − 2402 = 291 km s−1, corrected for
the instrumental resolution FWHM of 240 km s−1), the velocity dispersion measured
in the southern ﬁlaments, σ ∼ 55 km s−1 (FWHM=
√
2732 − 2402 = 130 km s−1), is just
marginally above the instrumental resolution.
The middle row left panel of Fig. 8.2 shows the Hα+[N II] image obtained with HST
WFPC2. Contours of 6 cm radio emission from Hines et al. (1989) are over-plotted,
showing that in projection the side of the southern ﬁlament closest to the nucleus
overlaps with the radio lobes. The narrowest resolved ﬁlaments of Hα+[N II] emission
have a diameter of only 0.4 arcsec, which corresponds to 32 pc. The total count
rate measured in our narrow band image from the bright southern ﬁlamentary structure
within a 15.7 ×5.7 box-like region centered at (α, δ)=(12:30:51.48,+12:23:07.33) is
6.94 counts s−1. Assuming an [N II]λ6548/Hα ﬂux ratio of 0.81 and [N II]λ6584/Hα=2.45
(Ford & Butcher 1979) and taking into account the ﬁlter throughput at the wavelengths
of the lines (see Sect. 8.2.2), the Hα ﬂux of the southern ﬁlament is fHff = 9.0 ×
10−15 erg s−1 cm−2.
Sparks et al. (2009) reported that the Hα+[N II] emission-line nebulae are co-spatial
with FUV emission (see the middle row central panel of Fig. 8.2). Using subsequent
spectroscopic measurements with the HST Cosmic Origin Spectrograph, Sparks et al.
(2012) showed that the FUV emission in the northern ﬁlaments is due to C IV line emission.
We ﬁnd that the C IV ﬂux, measured within the same aperture as the Hα+[N II]
ﬂux above, is fCIV = 1.2 × 10−14 erg s−1 cm−2.
The optical spectral properties of the southern ﬁlament show an intriguing rapid
decrease in the [O III]/Hβ ratio with increasing distance from the nucleus, dropping
monotonically from ∼2.5 at a projected distance of 28 arcsec to ∼0.7 at 43 arcsec (see
Fig. 8.3). The diagram in the left panel of Fig. 8.3, showing the [O III]/Hβ over the
[N II]/Hα ratios (a BPT diagram, Baldwin et al. 1981), indicates that the side of the ﬁlament
closest to the nucleus, which also overlaps with the radio lobes (see middle row
left panel of Fig. 8.2), has spectra similar to Seyfert like active galactic nuclei (AGN).
The [O III]/Hβ line ratios in this part of the ﬁlament are larger than in the nucleus of
150 Filamentary cold gas in M 87
M 87, where we measure a ratio of ∼ 1.75. At larger radii, the gas has a LINER3-like
emission spectrum, which is more similar to the emission line spectra typically seen in
extended ﬁlaments around brightest cluster galaxies in cluster cores.
The right hand panel of Fig. 8.3 shows the measured [S II]λ6716/[S II]λ6731 line
ratio, a good probe of gas density, as a function of radius. The full, dashed, and dotted
lines indicate ratios corresponding to electron densities of ne = 30 cm−3, ne =
100 cm−3, and ne = 500 cm−3, respectively (Osterbrock & Ferland 2006). The value
ne = 30 cm−3 is very close to the low density limit, below which the [S II] line ratios
do not provide useful constraints on gas density. Along most of the ﬁlament, the line
ratios are close to this limit indicating that the electron density of the 104 K phase is
ne 30 cm −3. At radii between 29–30.5 arcsec, the measured line ratios are outside
of the low density limit. The data point closest to the nucleus, at 28 arcsec, indicates a
possible increase in density, but with large error bars.
8.3.2 The soft X-ray emission
Young et al. (2002) and Sparks et al. (2004) found good spatial correlation between
Hα+[N II] line emission and soft X-ray emission. Here, we have analyzed almost four
times as much Chandra data, which provides excellent photon statistics in the regions
of interest. We extracted six narrow band images between 0.3 keV and 2.0 keV and
found that the ﬁlaments stand out clearly in the 0.7-0.9 keV band (see the middle row
right panel of Fig. 8.2). The spatial correlation between the 0.7–0.9 keV X-rays and
the Hα+[N II] line emission is remarkable: everywhere, where we see Hα+[N II] emitting
gas, we also see an excess of soft X-rays. On the other hand, in the 0.5–0.7 keV
and 0.9–1.2 keV bands the ﬁlaments do not stand out clearly against the surrounding
emission, indicating that most of their X-ray emission comes from the Fe XVII
and Fe XVIII lines in the 0.7–0.9 keV band. The excess X-ray ﬂux of the ﬁlaments in
this band is 1.1 × 10−14 ergs s−1 cm−2 and their average X-ray surface brightness is
1.2 × 10−16 ergs s−1 cm−2 arcsec−2.
Assuming that the soft X-ray emission is of a thermal origin, we follow the analysis
of Werner et al. (2010) and ﬁt to each spatial bin a model consisting of collisionally
ionized equilibrium plasmas at three ﬁxed temperatures (0.5 keV, 1.0 keV, and 2.0 keV),
with variable normalizations and common metallicity. The bottom panels of Fig. 8.2
show the spatial distributions of the emission measures of the individual temperature
components. The 0.5 keV component closely follows the distribution of the cold gas
phases. This component is required to ﬁt the Fe XVII and Fe XVIII lines in the 0.7–
0.9 keV band, which are only present in regions where Hα+[N II] emission has also
been detected. The spatial correlation between Hα+[N II] and the softest X-ray emitting
component is, however, not perfect: the ratio of Hα+[N II] ﬂux to the ∼0.5 keV emission
component is ∼ 5 times larger in the southern than in the northern ﬁlament. The
1 keV component is enhanced as well in the regions where cold gas is present, but
it is spatially more extended and is also seen in many regions where cold gas is not
detected (see Werner et al. 2010). The spatial distribution of the 2 keV plasma does not
3Low-ionization nuclear emission-line region (Heckman 1980)
8.4 Discussion 151
correlate with the Hα+[N II] emission.
Because the soft X-ray line emission observed in the ﬁlaments may be due to cooling
gas, we ﬁtted the spectra with a model consisting of a single-temperature plasma
in collisional ionization equilibrium plus an isobaric cooling ﬂow, which models cooling
between two temperatures at a certain metallicity and mass deposition rate. The
upper temperature and the metallicity of the cooling ﬂow component were tied to the
temperature and metal abundance of the single-temperature plasma. Both components
were absorbed by a Galactic absorption column density NH = 1.9 × 1020 cm−2
(Kalberla et al. 2005). Fixing the lower temperature cut-off of the cooling ﬂow model
to kTlow = 0.5 keV, we obtained a best ﬁt upper temperature of kTup = 1.90 ± 0.04 keV,
a mass deposition rate of ˙M = (2.46 ± 0.13) × 10−2 M yr−1, and a metallicity of
Z = 1.46 ± 0.09 Solar (relative to the Solar values by Grevesse & Sauval 1998). Even
though the spectra indicate that above kT ∼ 0.5 keV the gas may be cooling, the spectral
signatures of gas cooling below 0.5 keV - soft X-ray emission, including the O VII
line - are missing (see also the high resolution spectra in Werner et al. 2006, 2010).
In order to test whether the missing soft X-ray ﬂux might have been absorbed by
the cold gas in the ﬁlaments, we ﬁxed the mass deposition rate in the model to the
best ﬁt value obtained with a lower temperature cut-off set to 0.5 keV, extended this
cut-off down to a 1000 times lower temperature of 0.5 eV (the lowest value allowed
by the model). Furthermore, we assumed that the cooling X-ray plasma is intermixed
with the cold gas and the cooling happens in 13 different regions along our line-ofsight
separated from each other by intrinsic cold absorbers (the emission from the ﬁrst
cooling region is affected by one absorber, from the second by two absorbers, from the
third by three etc.). The different cooling regions are assumed to have the same cooling
rate and the different absorbers are assumed to have the same hydrogen column
density. Fitting this model to the data, we ﬁnd an integrated intrinsic hydrogen column
density of NH = (1.58 ± 0.21) × 1021 cm−2 and an absorbed bolometric ﬂux of
3.9 × 10−14 erg s−1 cm−2.
8.4 Discussion
8.4.1 Magnetized ﬁlaments of multi-phase material
In the ﬁlamentary emission-line nebulae to the southeast of the nucleus of M 87, we
detect gas spanning a temperature range of at least 5 orders of magnitude, from ∼
100 K to ∼ 107 K. The [C II], Hα+[N II], and C IV emitting gas phases are consistent
with being co-spatial, forming a multi-phase medium. Soft X-ray emission in the 0.7–
0.9 keV band is also always present in regions where Hα+ [N II] emission is seen. Its
spectral shape is consistent with line emission of ∼0.5 keV thermal plasma, previously
detected in the XMM-Newton high resolution reﬂection grating spectra integrated over
the central region of M 87 (Werner et al. 2006, 2010). We see no evidence of X-ray
emitting gas with temperature kT < 0.5 keV.
The [C II] 158 µm line is about 1500 times stronger than the CO (1→0) rotational
line in normal galaxies and Galactic molecular clouds, and 6300 times more intense in
152 Filamentary cold gas in M 87
starburst galaxies and Galactic star forming regions (Crawford et al. 1985; Stacey et al.
1991). Given a [C II] ﬂux of 8.5 × 10−15 erg s−1 cm−2 observed from the brightest region
of the southern ﬁlament, the expected CO (1→0) ﬂux is 1.3–5.3 × 10−18 erg s−1 cm−2,
consistent with the upper limit of 2 × 10−17 erg s−1 cm−2 for CO (1→0) at 2.6 mm observed
in the same region (Salom´e & Combes 2008). Using the standard CO luminosity
to H2 conversion factors the inferred range of CO luminosities corresponds to an H2
mass of 0.4–2 × 106 M . However, because the [C II]/CO line ratios in these ﬁlaments
may be different from those in star-forming galaxies and their heating mechanism may
be different from that in more normal molecular clouds, their true molecular mass may
be outside of this range.
Assuming that all the observed gas phases are in thermal pressure equilibrium with
the ambient ICM (assuming p = 0.11 − 0.22 keV cm−3, for a distance range of r =
3 − 6 kpc from the nucleus depending on the position of the ﬁlament along our line
of sight, Churazov et al. 2008), the ∼ 100 K [C II] emitting gas forms a network of
narrow ﬁlaments with densities of the order of (1.3 − 2.6) × 104 cm−3 (higher than the
critical density of 3 × 103 cm−3 above which the gas is in local thermal equilibrium with
the level populations determined by collisions) with volume ﬁlling fraction of fV ∼
(1 − 2) × 10−4 in the bright southern ﬁlament. Although the presence of O I 6300 ˚A
lines in the optical spectra (see also Ford & Butcher 1979) indicates the existence of large
partially ionized regions in the ﬁlaments, the presence of [S II] lines most probably
indicates a fully ionized ∼ 104 K phase surrounding the colder gas. The density of this
Hα+[N II] emitting gas, assuming thermal pressure equilibrium with the ambient ICM,
should be ∼ (1.3 − 2.6) × 102 cm−3.
The [S II]λ6716/[S II]λ6731 line ratios, however, indicate much lower particle densities
in the 104 K phase (n ∼ 2ne 60 cm−3) implying the presence of additional
signiﬁcant non-thermal pressure components in the ﬁlaments, such as turbulence and
magnetic ﬁelds. Assuming isotropic micro-turbulence with a characteristic velocity
vturb in a medium with a sound speed cs, the turbulent pressure support will be pturb =
1/2 γ ptherm M2, where γ is the adiabatic index assumed to be 5/3 and M = vturb/cs
is the Mach number. However, assuming characteristic turbulent velocities of the order
of the sound speed in the 104 K phase (cs = 15 km s−1), the sum of thermal and
turbulent pressure, pturb + ptherm = 0.09 keV cm−3 (1.4 × 10−10 dyne cm−2), is still
smaller than the surrounding ICM pressure. If the turbulence in the Hα emitting phase
is subsonic, then the additional pressure will be provided by magnetic ﬁelds. A magnetic
pressure of pmag = B2/8π ∼ (0.3 − 2.1) × 10−10 dyne cm−2, needed to keep the
ﬁlaments in pressure equilibrium with the surrounding ICM, requires magnetic ﬁelds
of B = 28 − 73 µG. This is similar to the values inferred using arguments based on
the integrity of Hα+[N II] ﬁlaments in the Perseus Cluster (Fabian et al. 2008). Radio
observations of the Faraday rotation measure in a number of cooling cores revealed
magnetic ﬁeld strengths of 10–25 µG (Taylor et al. 2001, 2007; Allen et al. 2001; Feretti
et al. 1999). Signiﬁcant imbalance in thermal pressure of the gaseous ﬁlaments in coolcore
clusters has previously also been inferred between the warm molecular hydrogen
gas seen in NIR (nT ∼ 108−9 cm −3 K) and the ionized gas phases seen in the optical
(nT ∼ 105 cm−3 K), emphasizing the need for dynamic models for the ﬁlaments (Jaffe
et al. 2001; Oonk et al. 2010).
8.4 Discussion 153
The smallest resolved diameter of the Hα+[N II] ﬁlaments is 32 pc, about half of
the 70 pc diameter observed in the more distant Perseus Cluster (Fabian et al. 2008).
However, even the narrowest resolved ﬁlaments are likely to consist of many smaller
strands. They are surrounded by hotter 105 K C IV emitting gas. The space between
these ﬁlaments is ﬁlled by X-ray emitting gas. If the soft X-ray line emission is due to
∼0.5 keV gas, then based on our three-temperature ﬁts (see Sect. 9.3) and assuming
thermal pressure equilibrium with the ambient ICM, its mass in the southern ﬁlament
is ∼ 105 M and its emitting volume is ∼0.016 kpc3.
8.4.2 The energy source of the ﬁlaments
Sparks et al. (2009) show that no main sequence stars of early spectral type are present
in the ﬁlaments and therefore neither the FUV nor the Hα+[N II] emission can be due to
photoionization by hot UV emitting stars. The [O III]/Hβ emission line ratios are high
on the side of the southern ﬁlament closest to the nucleus, which also overlaps with
the radio lobes (see middle row left panel of Fig. 8.2), and then decrease rapidly with
increasing distance from the nucleus. The fact that the line ratios in the ﬁlament reach
values which are higher than those measured in the nucleus argues against photoionization
by the AGN. It seems more likely that the increased [O III]/Hβ ratios are due
to internal shocks produced as the radio lobes push against the cold gas. These shocks
seem to affect only a part of the ﬁlament and the emitted Hα ﬂux does not decrease
with the decreasing [O III]/Hβ emission line ratios. Therefore, they can be ruled out as
energy source powering the emission of these nebulae.
Ferland et al. (2009) showed that the broad band (optical to mm) emission-line
spectra of nebulae around central galaxies of cooling core clusters most likely originate
in gas exposed to ionizing particles. Such ionizing particles can be either relativistic
cosmic rays or hot particles penetrating into the cold gas from the surrounding
ICM. Saturated conduction from the hot into the cold phase of the southern ﬁlament
would produce a heat ﬂux of 5φpcs ∼ 0.1 erg s−1 cm−2, where φ ∼ 1 accounts
for the uncertain physics, p = 0.22 keV cm −3 is the gas pressure and cs =
510 km s−1 is the sound speed in the hot 1 keV phase (Cowie & McKee 1977; Fabian
et al. 2011). Following the argument of Fabian et al. (2011), given a typical Hα surface
brightness of ∼ 2 × 10−16 erg s−1 cm−2 arcsec−2, corresponding to an emitted ﬂux of
1.1 × 10−4 erg s−1 cm−2 and assuming that the total broad band emitted ﬂux is 20 times
larger 2.2 × 10−3 erg s −1 cm−2, the required efﬁciency with which impinging thermal
particles penetrate and excite the cold gas is of the order of f = 0.022 (f = 0.044 for a
ﬁlament at r = 6 kpc from the nucleus), which is reasonable. However, these particles
must somehow overcome the obstacle presented by the magnetic ﬁelds believed to be
at the interface of the cold and hot gas phases.
Werner et al. (2010) show that all of the bright Hα+[N II] and UV ﬁlaments are seen
in the downstream region of a < 3 Myr old AGN induced M 1.2 shock front (Million
et al. 2010d) that seems to have just passed the ﬁlaments. Based on these observations
they propose that shocks induce shearing around the cold, ﬁlamentary, dense gas,
thereby promoting mixing of the cold gas with the ambient hot ICM via instabilities
(e.g. Friedman et al. 2012). Instabilities and turbulence may also be driven by shear-
154 Filamentary cold gas in M 87
ing motions between the hotter X-ray gas and the ﬁlaments, as they move through the
ambient ICM, both as they are being uplifted and as they fall back. This type of more
continuous shear induced mixing could perhaps be in the long term more important
for energizing the ﬁlaments than shocks. Fabian et al. (2011) propose that the sub-mm
through UV emission seen in NGC 1275 in the core of the Perseus Cluster is due to the
hot ICM efﬁciently penetrating the cold gas through reconnection diffusion (Lazarian
et al. 2010, 2011). The same process may also be operating in the ﬁlaments of M 87 with
the fast reconnection induced by shearing instabilities and turbulence. Shearing and
magnetic reconnection are, however, not essential to bring the cold gas into contact
with the hot ICM. The support from magnetic ﬁelds tied to the surrounding ionized
phase may become unstable if the layer of dense cold neutral gas is deep enough, and
therefore it may fall through into direct contact with the hot gas. The hot ICM particles
penetrating the cold gas then produce secondary electrons that excite the observed FIR
through UV emission as shown by Ferland et al. (2009). Mixing of the cold and hot gas
also naturally account for the presence of the ∼ 105 K intermediate-temperature gas
phase (geometric mean of the hot and cold gas temperatures) predicted by the mixing
layers calculated by Begelman & Fabian (1990), although mixing layers have most
probably a more complicated temperature structure (Esquivel et al. 2006). As the X-ray
emitting gas cools through mixing, the mass of the southern ﬁlament may grow by as
much as 4 × 10−2 M yr−1.
The particle heating model of Ferland et al. (2009), calculated considering optically
thin emission from a cell of gas, predicts [O I]/[C II] ∼ 21 and is inconsistent with
the 2σ upper limit of 7.2 observed in M 87. The ratios observed in the Perseus and
Centaurus clusters, and in Abell 1068 and 2597 are even lower. This discrepancy has
been discussed in a recent study of NGC 1275 by Mittal et al. (2012), who conclude
that the line ratio suggests that the lines are optically thick, implying a large reservoir
of cold atomic gas, which was not accounted for in previous inventories of the ﬁlament
mass. The same can be concluded for M 87, where the large optical depth of the FIR
lines (Canning et al. in prep.) and the intrinsic absorption inferred from the X-ray data
(see below) imply signiﬁcant reservoirs of cold atomic and molecular gas.
8.4.3 X-ray line emission from cold gas due to charge exchange and
inner shell ionization?
As ions from the hot ICM penetrate into the ﬁlaments, they pick up one or more electrons
from the cold gas. The impinging ions thus get into excited states and their
deexcitation may produce X-ray radiation. Astrophysical X-ray emission due to this
‘charge exchange’ process was ﬁrst discovered in the comet Hyakutake (Lisse et al.
1996; Cravens 1997) and since then it has been identiﬁed as the mechanism responsible
for X-rays from planets (e.g. Gladstone et al. 2002; Dennerl 2002), the geocorona
(Wargelin et al. 2004; Fujimoto et al. 2007), the heliosphere (Robertson et al. 2001), and
interstellar medium in particular at interfaces between hot gas and cool clouds (see the
review by Dennerl 2010, and references therein). Recently, charge exchange has also
been found to possibly contribute signiﬁcantly to the soft X-ray line emission from
8.4 Discussion 155
star-forming galaxies (Liu et al. 2011, 2012). Fabian et al. (2011) conclude that in the
Perseus Cluster charge exchange may account for a few per cent of the soft X-ray emission
from the ﬁlaments. They also speculate that if the ﬁlaments are composed of many
small strands, the same ion might pass from the hot into the cold and back to the hot
phase repeatedly, getting collisionally ionized while passing through the hot ICM, thus
enhancing the charge exchange emission to the observed level.
After hydrogen and helium, the most abundant ions penetrating the cold gas are
O8+, C6+, N7+, Ne10+, Mg12+, Mg11+ (with concentrations 8.4 × 10−4, 3.6 × 10−4, 1.1 ×
10−4, 1.1 × 10−4, 2.5 × 10−5, 1.07 × 10−5 relative to hydrogen by number). Charge
transfer onto these ions is expected to produce mainly hydrogen and helium like line
emission in the X-ray band. In the 1 keV plasma surrounding the ﬁlaments, 23% of
Fe is in the form of Fe20+, another 23% in Fe21+, 18% in Fe19+, and 17% in Fe22+,
with relative concentrations smaller than 1.1 × 10−5. The charge exchange line ﬂuxes
produced by Fe will therefore be signiﬁcantly lower, with most of the line ﬂux in Fe IXX
to Fe XXII lines, in the spectral band above 0.9 keV. Moreover, laboratory measurements
of X-ray spectral signatures of charge exchange in L- shell Fe ions by Beiersdorfer et al.
(2008) found that the n ≥ 4 → n = 2 transitions are strongly enhanced relative to
n = 3 → n = 2, shifting the line emission to higher energies. Because charge transfer
processes and the corresponding X-ray emission are expected to occur inside of the
dense cold clouds, absorption may signiﬁcantly decrease the observed ﬂuxes of the
lowest energy X-ray lines such as C VI, N VII, O VII, and O VIII. The Ne IX, Ne X, Mg XI,
and Mg XII lines in the 0.9–1.9 keV band are, however, not expected to be affected by
absorption strongly and their enhancement might indicate charge exchange. Given
that none of these lines are enhanced in the ﬁlaments, and no strong charge exchange
emission is expected in the 0.7–0.9 keV band, we conclude that the observed soft X-ray
emission is not due to charge transfer onto the impinging ions.
The ions and electrons penetrating from the surrounding hot ICM into the cold ﬁlaments
will also ionize the inner electron shells of the atoms and molecules in the cold
clouds. This inner shell ionization may in principle also contribute to the X-ray emission
of the ﬁlaments, but given the energetics, it is not expected to be signiﬁcantly
stronger than the X-ray emission due to charge exchange. We would primarily expect
to see X-ray lines from few times ionized C, N, O, Ne, and Mg, with the Ne and Mg
lines in the band above 0.6 keV. Detailed predictions of this process are beyond the
scope of this paper.
Given that the spectral shape of the ﬁlaments is consistent with line emission of
∼0.5 keV thermal plasma previously resolved within the central region of M 87 with
XMM-Newton RGS (Werner et al. 2006, 2010) and that charge exchange as a signiﬁcant
source of X-ray emission from the ﬁlaments can be ruled out, we conclude that the
observed soft X-ray emission is most likely due to thermal plasma. We note, however,
that although charge exchange is not responsible for the observed excess X-ray emission
in the 0.7 − 0.9 keV band, the level which is required from the particle heating
model is observationally acceptable.
156 Filamentary cold gas in M 87
8.4.4 AGN uplift induced cooling instabilities or mixing?
The observed velocity distribution of the [C II] (see top central panel of Fig. 8.2) and
optical emission lines (Sparks et al. 1993) indicates that the northern ﬁlaments are
currently being uplifted by the counter-jet (which is oriented at < 19◦ from our lineof-sight,
Biretta et al. 1999) moving away from M 87 with a line-of-sight velocity of
vLOS = +140 km s−1. The southern ﬁlaments, in contrast, are falling back towards M 87
with vLOS = −123 km s−1. Assuming uplift at the Keplerian velocity of ∼ 400 km s−1,
the southern ﬁlament would take 8 × 106 yr to get to its current position. Given that
this ﬁlament is most likely already falling back, its age is 107 yr. Uplift of low entropy,
metal enriched X-ray emitting gas by buoyantly rising relativistic plasma from
the AGN jets of M 87 has been previously inferred based on radio and X-ray data (Churazov
et al. 2001; Forman et al. 2005; Simionescu et al. 2008; Werner et al. 2010). Optical
emission line nebulae also appear to have been drawn out by rising radio bubbles in
other nearby systems (e.g. the Perseus and Centaurus clusters, Hatch et al. 2006; Canning
et al. 2011).
The initially uplifted material might have already been a multi-phase mixture of low
entropy X-ray emitting plasma and dusty cold gas, as is observed in the core of M 87
(Sparks et al. 1993). As the relatively dense, low entropy X-ray emitting plasma was removed
from the direct vicinity of the AGN jets it might have cooled radiatively, further
contributing to the cold gas mass of the uplifted material (see Revaz et al. 2008). The
multi-phase X-ray emitting gas, with the 0.5 keV phase co-spatial with the Hα+[N II]
ﬁlaments and surrounded by a spatially more extended 1 keV plasma, indicates that
such cooling may be taking place. Cooling instabilities are predicted to take place in
cooling core clusters within radii where the ratio of the cooling time to free fall time
is tcool/tff 10 (Sharma et al. 2012a; McCourt et al. 2012; Gaspari et al. 2012, this
condition is fulﬁlled at the radii where the M 87 ﬁlaments are seen). AGN induced uplift
of multiphase gas may facilitate the onset of such local cooling instabilities. Some
Hα+[N II] and soft X-ray ﬁlaments lie at the edges of cavities ﬁlled by radio emitting
plasma, indicating that they might be sheets of gas seen in projection that have been
swept up and compressed by the expanding and rising cavities. Even though such
compression would ﬁrst heat the gas, eventually it will accelerate its cooling. This
picture of radiatively cooling X-ray gas is, however, complicated by the observed temperature
ﬂoor at 0.5 keV. Several explanations were proposed to address the missing
soft X-ray emission, including mixing (Fabian et al. 2002), anomalous O/Fe abundance
ratios, and absorption (Sanders & Fabian 2011).
The mass deposition rate, inferred from the temperature distribution of the X-ray
gas in the southern ﬁlament above kT = 0.5 keV, assuming an isobaric cooling ﬂow
model, is ˙M = (2.46 ± 0.13) × 10−2 M yr−1. This cooling rate could only be reconciled
with the apparent lack of kT < 0.5 keV plasma if the cold gas in the ﬁlaments absorbs
the soft X-rays, modifying the observed spectral shape in a way that results in an apparent
temperature cut-off. The X-ray spectral lines providing the most diagnostic power
at these temperatures are the Fe XVII and the O VII lines. The ionic fraction of Fe XVII is
above 0.3 across a relatively large temperature range of 0.2–0.6 keV, but the O VII line
becomes observable only at kT < 0.4 keV. Absorption is expected to strongly suppress
8.4 Discussion 157
the O VII line emitted by the coldest gas phases. The Fe XVII lines will be affected by absorption
to a much lesser extent. Therefore when modeling a spectrum that is affected
by intrinsic absorption, we may conclude that the cooling stopped at the temperature
where we expect to start to see the O VII line4. In Sect. 9.3, we show that the required
integrated hydrogen column density, distributed in multiple layers (the cooling takes
place between the threads) along our line-of-sight is NH = (1.58 ± 0.21) × 1021 cm−2.
If the hot ICM indeed cools radiatively between the cold ﬁlaments, then the large area
covering fraction of the many small strands may provide a sufﬁcient absorption column
to the cooling X-ray emitting plasma to explain the apparent temperature ﬂoor at
∼0.5 keV.
Soft X-ray photons from the cooling X-ray plasma absorbed by the cold gas would
produce energetic photoelectrons which would contribute to the heating and ionization
of the cold gas (in a process similar to the secondary electron production by cosmic
rays or impinging hot ICM particles, Ferland et al. 2009). Photoionization of emission
line nebulae by X-rays has been considered before (e.g. Heckman et al. 1989; Donahue
& Voit 1991; Crawford & Fabian 1992; Jaffe & Bremer 1997) and has been found insufﬁcient
to ionize and power the observed line emission. However, when considering
photoionization by primarily soft X-rays, from cooling plasma which surrounds the
cold ﬁlamentary gas with a small volume ﬁlling fraction, the efﬁciency of this process
increases signiﬁcantly (Oonk 2011). According to our spectral modeling, where we
assumed that all the soft X-ray emission is due to radiative isobaric cooling, intrinsic
absorption reduces the observed ﬂux by ∼ 3.9 × 10−14 erg s−1 cm−2. If the soft X-ray
emitting plasma ﬁlls the space between the cold ﬁlaments and the cold gas absorbs
approximately the same amount of energy in all directions, then the total X-ray energy
absorbed by the cold gas is ∼ 1.3 × 1039 erg s−1. With a total emitted energy from UV
to sub-mm about 20 times larger than the observed Hα luminosity of 3.0 × 1038 erg s−1,
X-ray photoionization may, in principle, contribute a signiﬁcant fraction of the energy
needed to power the ﬁlaments.
The cooling rates above were deduced under the assumption that all the soft X-ray
line emitting gas originates from radiative cooling. The kT ∼ 0.5 keV gas could, in
principle, also be produced by mixing of the hot ICM with the cold ﬁlaments. More
developed mixing might be responsible for the higher ratio of 0.5 keV thermal to
Hα+[N II] line emission in the southern ﬁlament (which is most likely falling back
towards M 87), with respect to the northern ﬁlament (which is likely younger, currently
being uplifted). Such mixing is, however, not expected to result in a gas with
kT ∼ 0.5 keV - it would result in a mixing layer with a wide range of temperatures
(Esquivel et al. 2006). Therefore, both in the cooling and in the mixing scenario for
the origin of the soft X-ray emission from the ﬁlaments, the explanation of the 0.5 keV
cutoff requires intrinsic absorption by cold gas.
The internal reddening in the optical band calculated from the observed line ratio
of Hα/Hβ = 4 in the ﬁlament (Ford & Butcher 1979) is, E(B − V) = 0.28. Using the
4We note that only a small fraction of the observed X-ray emission originates in the ﬁlaments and
is affected by intrinsic absorption. The dominant emission component from the ambient ICM remains
unaffected by cold gas in the ﬁlaments, and therefore ﬁtting the spectra with a Galactic absorption as a
free parameter will not result in a signiﬁcantly higher best ﬁt NH.
158 Filamentary cold gas in M 87
Galactic transformation between reddening and column density of neutral hydrogen,
NH/E(B − V) = 5.8 × 1021 cm−2 mag−1 (Bohlin et al. 1978), we infer an intrinsic neutral
hydrogen column density of NH = 1.6 × 1021 cm−2 - sufﬁcient to absorb the soft
X-ray emission from gas with kT < 0.5 keV. We note that substantial reddening has
been observed in a number of Hα nebulae in brightest cluster galaxies of cooling core
clusters (e.g. Allen et al. 1995; Crawford et al. 1999).
A similar temperature ﬂoor at ∼0.5 keV is observed in the Perseus Cluster, Centaurus
Cluster, 2A 0335+096, Abell 262, Abell 3581, HCG 62, and Abell 2052 (Sanders &
Fabian 2007; Sanders et al. 2008, 2009, 2010a; de Plaa et al. 2010) and in the cooling uplifted
multi-phase gas in the core of the galaxy cluster S´ersic 159-03 (Werner et al. 2011).
The coolest ∼0.5 keV gas phases are co-spatial with ﬁlamentary optical emission-line
nebulae in all of these systems. In the Perseus and Centaurus clusters, copious amounts
of FIR line emission has also been discovered, indicating the presence of large reservoirs
of cold atomic gas (Mittal et al. 2011, 2012). Following the arguments outlined
above, the low temperature X-ray cutoffs observed in those systems may therefore
also, at least partly, result from intrinsic absorption.
8.5 Conclusions
We performed a multi-wavelength study of the emission-line nebulae located southeast
of the nucleus of M 87. Our main conclusions may be summarized as follows:
• We detect FIR [C II] line emission at 158 µm with Herschel PACS. The line emission
is extended and co-spatial with optical Hα+[N II], FUV C IV lines, and soft X-ray
emission.
• The ﬁlamentary nebulae contain multi-phase material spanning a temperature
range of at least 5 orders of magnitude, from ∼ 100 K to ∼ 107 K. This material
has most likely been uplifted by the AGN from the center of M 87.
• The thermal pressure of the 104 K phase appears to be signiﬁcantly lower than
that of the surrounding hot ICM indicating the presence of additional turbulent
and magnetic pressure in the ﬁlaments. If the turbulence in the ﬁlaments is subsonic
then the magnetic ﬁeld strength required to balance the pressure of the
surrounding ICM is B ∼ 30 − 70 µG.
• The spectral properties of the soft X-ray emission from the ﬁlaments indicate that
it is due to thermal plasma with kT ∼ 0.5–1 keV, which is cooling by mixing with
the cold gas and/or radiatively. Charge exchange can be ruled out as a signiﬁcant
source of soft X-rays.
• Both cooling and mixing scenarios predict gas with a range of temperatures. This
is at ﬁrst glance inconsistent with the apparent lack of X-ray emitting gas with
kT < 0.5. However, we show that the missing very soft X-ray emission could
be absorbed by the cold gas in the ﬁlaments with an integrated hydrogen column
density of NH ∼ 1.6 × 1021 cm−2, providing a natural explanation for the
8.5 Conclusions 159
apparent temperature ﬂoor to the X-ray emission at kT ∼ 0.5 keV. The internal
reddening observed in the optical band indicates a similar level of intrinsic ab-
sorption.
• The FIR through UV line emission is most likely primarily powered by the ICM
particles penetrating the cold gas following a shearing induced mixing process.
An additional source of energy may, in principle, be provided by X-ray photoionization
from cooling X-ray emitting plasma.
• The relatively small line ratio of [O I]/[C II]< 7.2 indicates a large optical depth in
the FIR lines. The large optical depth at FIR and the intrinsic absorption inferred
from the X-ray and optical data all imply signiﬁcant reservoirs of cold atomic
and molecular gas distributed in ﬁlaments with small volume ﬁlling fraction, but
large area covering factor.
Acknowledgments
This work is based in part on observations made with Herschel, a European Space
Agency Cornerstone Mission with signiﬁcant participation by NASA. Support for this
work was provided by NASA through award number 1428053 issued by JPL/Caltech.
Support for this work was provided by NASA through Einstein Postdoctoral grant
numbers PF9-00070 (AS) and PF2-130104 (RvW) awarded by the Chandra X-ray Center,
which is operated by the Smithsonian Astrophysical Observatory for NASA under
contract NAS8-03060. SWA acknowledges support from the U.S. Department of Energy
under contract number DE-AC02-76SF00515. MR acknowledges NSF grant AST
1008454.
160 Filamentary cold gas in M 87
Chapter 9
The origin of cold gas in giant elliptical
galaxies and its role in fueling
radio-mode AGN feedback
N. Werner1,2, J. B. R. Oonk3, M. Sun4, P. E. J. Nulsen5, S. W. Allen1,2,6, R. E. A. Canning1,2,
A. Simionescu7, A. Hoffer8, T. Connor8, M. Donahue8, A. C. Edge9, A. C. Fabian10,
A. von der Linden1,2,11, C. S. Reynolds12, M. Ruszkowski13,14
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall, Stanford, CA 94305-4085, USA
2Department of Physics, Stanford University, 382 Via Pueblo Mall, Stanford, CA 94305-
4060, USA
3ASTRON, Netherlands Institute for Radio Astronomy, P.O. Box 2, 7990 AA Dwingeloo,
The Netherlands
4Department of Physics, University of Alabama in Huntsville, Huntsville, AL 35899,
USA
5Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA
02138, USA
6SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025,
USA
7Institute of Space and Astronautical Science (ISAS), JAXA, 3-1-1 Yoshinodai, Chuo-ku,
Sagamihara, Kanagawa, 252-5210 Japan
8Physics & Astronomy Department, Michigan State University, East Lansing, MI 48824-
2320, USA
9Institute for Computational Cosmology, Department of Physics, Durham University,
Durham, DH1 3LE, UK
10Institute of Astronomy, Madingley Road, Cambridge CB3 0HA, UK
11Dark Cosmology Centre, Niels Bohr Institute, University of Copenhagen, Juliane
Maries Vej 30, DK-2100 Copenhagen, Denmark
12Department of Astronomy and the Maryland Astronomy Center for Theory and
Computation, University of Maryland, College Park, MD 20742, USA
13Department of Astronomy, University of Michigan, 500 Church Street, Ann Arbor,
162 Cold gas in giant ellipticals
MI 48109, USA
14Michigan Center for Theoretical Physics, 3444 Randall Lab, 450 Church St, Ann Arbor,
MI 48109, USA
Published in the Monthly Notices of the Royal Astronomical Society, volume 439,
pages 2291–2306, 2014
Abstract
The nature and origin of the cold interstellar medium (ISM) in early type galaxies are
still a matter of debate, and understanding the role of this component in galaxy evolution
and in fueling the central supermassive black holes requires more observational
constraints. Here, we present a multi-wavelength study of the ISM in eight nearby,
X-ray and optically bright, giant elliptical galaxies, all central dominant members of
relatively low mass groups. Using far-infrared spectral imaging with the Herschel Photodetector
Array Camera & Spectrometer (PACS), we map the emission of cold gas
in the cooling lines of [C II]λ157µm, [O I]λ63µm, and [O Ib]λ145µm. Additionally, we
present Hα+[N II] imaging of warm ionized gas with the Southern Astrophysical Research
(SOAR) telescope, and a study of the thermodynamic structure of the hot X-ray
emitting plasma with Chandra. All systems with extended Hα emission in our sample
(6/8 galaxies) display signiﬁcant [C II] line emission indicating the presence of reservoirs
of cold gas. This emission is co-spatial with the optical Hα+[N II] emitting nebulae
and the lowest entropy soft X-ray emitting plasma. The entropy proﬁles of the hot
galactic atmospheres show a clear dichotomy, with the systems displaying extended
emission line nebulae having lower entropies beyond r 1 kpc than the cold-gas-poor
systems. We show that while the hot atmospheres of the cold-gas-poor galaxies are
thermally stable outside of their innermost cores, the atmospheres of the cold-gas-rich
systems are prone to cooling instabilities. This provides considerable weight to the
argument that cold gas in giant ellipticals is produced chieﬂy by cooling from the hot
phase. We show that cooling instabilities may develop more easily in rotating systems
and discuss an alternative condition for thermal instability for this case. The hot
atmospheres of cold-gas-rich galaxies display disturbed morphologies indicating that
the accretion of clumpy multiphase gas in these systems may result in variable power
output of the AGN jets, potentially triggering sporadic, larger outbursts. In the two
cold-gas-poor, X-ray morphologically relaxed galaxies of our sample, NGC 1399 and
NGC 4472, powerful AGN outbursts may have destroyed or removed most of the cold
gas from the cores, allowing the jets to propagate and deposit most of their energy further
out, increasing the entropy of the hot galactic atmospheres and leaving their cores
relatively undisturbed.
9.1 Introduction 163
Table 9.1: Summary of the sample of galaxies. The distances in column 2 are from
†Tonry et al. (2001) and ‡Blakeslee et al. (2009). The K-band luminosities in column
5 were determined using the 2MASS survey (Jarrett et al. 2003). The Hα+[N II] luminosities
in column 6 are from Macchetto et al. (1996). The luminosities marked with ‘ ’
have been revised in Sect. 9.3 and Table 9.4. The 1.4 GHz radio luminosities in column
7 are from Condon et al. (1998) and Condon et al. (2002). Bolometric X-ray luminosities
in column 8 are from O’Sullivan et al. (2001). The jet powers for NGC 4636, NGC 4472,
and NGC 5846 are from Allen et al. (2006), for NGC 5044 from David et al. (2009), for
NGC 1399 from Shurkin et al. (2008), and for NGC 5813 from Randall et al. (2011b). For
the jet-powers in NGC 5044 and NGC 5813 no errorbars were published (Randall et al.
2011b; David et al. 2009). Star-formation rates were estimated using a combination of
FUV data from Galex and MIR data from Wise, employing the technique developed by
Hao et al. (2011).
Galaxy d Scale z LK LHff+[N II] Lradio LX Pjet SFR
(Mpc) (arcsec kpc−1) (×1040 erg s−1) (×1039 erg s−1) (×1038 erg s−1) (×1041 erg s−1) (×1041 erg s−1) M yr−1
NGC 1399 20.9‡ 9.9 0.004753 2.69 9.3 1.52 5.68 21.9 ± 7 0.171
NGC 4472 16.7‡ 12.4 0.003326 3.97 5.8 1.20 2.96 80.7 ± 23.5 0.099
NGC 4636 14.7† 14.0 0.003129 1.20 5.5 0.28 3.32 3.0 ± 0.8 0.027
NGC 5044 31.2† 6.6 0.009280 1.65 45.6 0.59 5.87 6.0 0.073
NGC 5813 32.2† 6.4 0.006578 2.30 14.0 0.26 10.63 1.2 0.063
NGC 5846 27.1† 7.6 0.005717 2.11 24.6 0.26 6.25 7.4 ± 2.6 0.074
NGC 6868 26.8† 5.6 0.009520 1.74 23.8 1.00 1.01 – 0.078
NGC 7049 29.9† 7.1 0.007622 2.32 31.8 0.84 1.23 – 0.117
9.1 Introduction
Recent observations and simulations strongly suggest that the growth and evolution of
giant early type galaxies is closely tied to that of their central supermassive black holes
through a well regulated feedback cycle (e.g. Silk & Rees 1998; Magorrian et al. 1998;
Croton et al. 2006; Sijacki et al. 2007). Many fundamental aspects of this feedback process,
such as the nature of the material feeding the black holes in active galactic nuclei
(AGN), jet formation, and the heating, cooling and detailed physics of the interstellar
medium (ISM) are not well understood. Among the best laboratories to test theories
for the formation and growth of massive galaxies are nearby giant ellipticals, groups
and clusters of galaxies: systems where hot X-ray emitting plasma may be cooling and
accreting onto the galaxies and interacting with their AGN.
In massive galaxy clusters, with central cooling times shorter than the Hubble time,
the central brightest cluster galaxies (BCG) are often surrounded by spectacular, ﬁlamentary
optical Hα+[N II] emission-line nebulae extending up to 70 kpc from the core
of the BCG (e.g. Cowie et al. 1983; Johnstone et al. 1987; Heckman et al. 1989; Donahue
et al. 1992; Crawford et al. 1999; McDonald et al. 2010). These nebulae of ionized gas
seem to be co-spatial with warm (1000-2000 K) molecular hydrogen seen in the nearinfrared
(e.g. Jaffe & Bremer 1997; Falcke et al. 1998; Donahue et al. 2000; Edge et al.
2002; Hatch et al. 2005; Jaffe et al. 2005; Johnstone et al. 2007; Oonk et al. 2010; Lim et al.
2012) and with large quantities (108 − 1011.5 M ) of cold (< 50 K) molecular gas traced
by CO (e.g. Edge 2001; Edge & Frayer 2003; Salom´e & Combes 2003; McDonald et al.
164 Cold gas in giant ellipticals
Table 9.2: Summary of observations.
Galaxy Herschel Herschel Chandra Detector Chandra SOAR/SOI SOAR/SOI SOAR/SOI
Obs. ID Exp. (s) Obs. ID Exp. (ks) Obs. date ﬁlter Exp (s)
NGC1399 1342239492 8161 319 ACIS-S 49.9 2012 Oct. 9 CTIO 656375-4/6916-78 3 × 600 / 3 × 600
4172 ACIS-I 36.7 2013 Aug. 4 Goodman 1.68 long slit, KOSI600 grating 3 × 480
9530 ACIS-S 59.3 2013 Sep. 8 Goodman 3.0 long slit, KOSI600 grating 4 × 480
NGC4472 1342234992 8161 321 ACIS-S 19.1
NGC4636 1342236884 5442 323 ACIS-S 35.8 2009 June 1 CTIO 6600/75 3 × 780
CTIO 6120/140 3 × 500
NGC5044 1342238376 24723 9399 ACIS-S 53.4 2010 April 10 CTIO 6649/76 3 × 1200
3926 ACIS-I 54.5 CTIO 6520/76 3 × 720
4415 ACIS-I 73.3
NGC5813 1342238158 10880 5907 ACIS-S 47.4 2008 July 6 CTIO 6600/75 3 × 900
9517 ACIS-S 98.8 CTIO 6120/140 3 × 720
NGC5846 1342238157 5442 788 ACIS-S 17.3 2009 June 1 CTIO 6600/75 3 × 900
7923 ACIS-I 75.9 CTIO 6120/140 3 × 540
NGC6868 1342215929 5442 3191 ACIS-I 18.6 2012 Oct. 9 CTIO 6600/75 2 × 900, 1 × 600
11753 ACIS-I 56.8 2012 Oct. 9 CTIO 6563/75 2 × 900, 1 × 600
NGC7049 1342216658 5442 5895 ACIS-I 2.2 2012 Oct. 9 CTIO 6600/75 2 × 900, 1 × 600
2012 Oct. 9 CTIO 6563/75 1 × 900, 2 × 600
2012). Recently, observations with the Herschel Space Observatory (Pilbratt et al. 2010)
also revealed the presence of far-infrared (FIR) cooling lines of [C II], [O I], and [N II]
in the X-ray bright cores of Abell 1068, Abell 2597 (Edge et al. 2010), and the Centaurus,
Perseus, and Virgo clusters (Mittal et al. 2011, 2012; Werner et al. 2013). In nearby
systems, such as the Centaurus, Perseus, and Virgo clusters, where the spatial distribution
of the [C II] line could be mapped, it was found to be extended and co-spatial
with optical Hα+[N II], far-ultraviolet C IV (in M 87, Sparks et al. 2012), and soft X-ray
emission. The ﬁlamentary nebulae in the centers of massive cool-core galaxy clusters
thus contain multi-phase material spanning a temperature range of over 5 orders of
magnitude, from < 50 K to ∼ 107 K. This gas often appears to be interacting with
the jets and the buoyant relativistic plasma from the central AGN (Fabian et al. 2003b;
Hatch et al. 2006; Canning et al. 2013). Despite the large quantities of cold gas, many
extended nebulae show no evidence for recent star-formation.
The majority of AGN in these systems are in a quiescent, so called ‘radio’-mode,
where they are accreting at a modest rate. The accretion ﬂows in this mode, though optically
faint, often drive powerful relativistic jets, extending to large distances, which
can have a profound impact on their surroundings. Using Chandra X-ray observations
of nine nearby, X-ray luminous elliptical galaxies, Allen et al. (2006) found a tight correlation
between the Bondi accretion rates (see Bondi 1952) from the hot X-ray emitting
ISM and the power in the relativistic jets (though see Russell et al. 2013a). The jet powers
were determined from the work required to inﬂate bubbles of relativistic plasma
associated with the most recent cavities in the hot X-ray emitting atmospheres of the
galaxies (Churazov et al. 2002). The relationship between jet power and accretion rate
of the hot ISM, however, raises the question: what is the role of the cold gas in the
AGN feedback cycle?
Here we present a multi-wavelength study of 8 nearby, X-ray bright, giant elliptical
galaxies, all central dominant members of relatively low mass groups. We use FIR data
obtained with the Herschel Photodetector Array Camera & Spectrometer (PACS) in the
lines of [C II]λ157 µm, [O I]λ63µm, and [O Ib]λ145µm that are excellent probes of cold
∼ 100 K gas; optical data from the Southern Astrophysical Research (SOAR) telescope
in the lines of Hα+[N II] probing warm, ionized ∼10,000 K gas; and X-ray data from
9.2 Observations and data analysis 165
the Chandra X-ray Observatory, tracing the hot 5-20 million K plasma permeating these
systems.
9.1.1 The galaxy sample
Our target list is drawn from the parent sample of Dunn et al. (2010), who identiﬁed the
optically and X-ray brightest giant elliptical/S0 galaxies within a distance d ≤ 100 Mpc
and with declination Dec. ≥ −45. Motivated by the goal of studying the properties
of cold gas in these types of systems in detail, we selected only those galaxies with
distance d < 35 Mpc and with relatively bright Hα+[N II] emission as reported by
Macchetto et al. (1996). All of the systems in our sample are the central dominant
galaxies of their respective groups. The properties of the galaxies are summarized
in Table 9.1. The table lists the distances of the galaxies (Tonry et al. 2001; Blakeslee
et al. 2009), their K-band near-infrared luminosities determined using the magnitudes
measured by the 2MASS survey (Jarrett et al. 2003), Hα+ [N II] luminosities (Macchetto
et al. 1996), radio luminosities at 1.4 GHz (Condon et al. 1998, 2002), bolometric X-ray
luminosities (O’Sullivan et al. 2001), jet-powers (Allen et al. 2006; David et al. 2009;
Shurkin et al. 2008; Randall et al. 2011b), and star-formation rates (SFR) estimated using
a combination of far-ultraviolet (FUV) data from Galex and mid- infrared (MIR) data
from Wise, employing the technique developed by Hao et al. (2011).
The near-infrared K-band luminosities, and by implication the stellar masses, span a
range of approximately a factor of three, with NGC 4636 being the least and NGC 4472
the most luminous. The Hα+[N II] luminosities span a range of a factor of eight, with
NGC 5044 being the most luminous, and with NGC 4636 and NGC 4472 on the low luminosity
end. NGC 5044 boasts a particularly spectacular network of radial ﬁlaments,
extending out to a radius of at least r ∼10 kpc (Gastaldello et al. 2009; David et al.
2011). The SFRs are small, typically of the order of 0.1 M yr−1, and they do not correlate
with the Hα luminosities. All eight galaxies have central, active radio jets, also
spanning a range of radio luminosities and jet-powers. The X-ray luminosities of the
galaxies span a range of an order of magnitude.
9.2 Observations and data analysis
9.2.1 Far-infrared spectroscopy with Herschel PACS
We observed the FIR cooling lines of [C II]λ157µm, [O I]λ63µm, and [O Ib]λ145µm in
our sample of 8 galaxies with the PACS integral-ﬁeld spectrometer (Poglitsch et al.
2010) on the Herschel Space Observatory. Table 9.2 gives a summary of the observations.
The observations were taken in line spectroscopy mode with chopping-nodding to
remove the telescope background, sky background and dark current. A chopper throw
of 6 arcmin was used. For 7 systems, with a relatively compact Hα emission region,
the observations were taken in pointed mode targeting the centers of the galaxies. For
NGC 5044, we used raster mapping in 3 × 3 steps of 23.5 arcsec to match the extent
of the Hα nebula. The observations were reduced using the HIPE software version
166 Cold gas in giant ellipticals
8.2.0, using the PACS ChopNodLineScan pipeline script. This script processes the
data from level 0 (raw channel data) to level 2 (ﬂux calibrated spectral cubes).
During the ﬁnal stage of the reduction the data were spectrally and spatially rebinned
into 5 × 5 × λ cubes. In the following we will refer to these cubes as the rebinned
cubes. Each spatial pixel, termed spaxel, in these cubes has a size of 9.4 ×
9.4 arcsec2. The cubes thus provide us with a ﬁeld of view (FoV) of 47 × 47 arcsec2.
For the wavelength regridding the parameters oversample and upsample were set
to 2 and 1, respectively. This means that one spectral bin corresponds to the native
resolution of the PACS instrument.
The integrated [C II] line ﬂuxes were in all cases, except for NGC 5044, obtained
by spatially integrating the 5 × 5 spaxels from the rebinned cubes. For NGC 5044 the
integrated line ﬂux is obtained after projecting the individual rebinned cubes onto the
sky. No point-spread function (PSF) correction is applied as the [C II] emission, in all
cases where it is detected, is found to be extended.
To visualize the extent of the [C II] emission in our galaxies, we have created sky
maps of the [C II] emission by using the specProject task in HIPE and the hrebin
task in IDL. In the following we will refer to these data cubes as the projected cubes.
A pixel size of 6 arcsec was chosen in order to Nyquist sample the beam, the fullwidth-at-half-maximum
(FWHM) of which is 12 arcsec at the observed wavelength
of the [C II] line. We only consider spatial bins where the signal-to-noise ratio of the
integrated line ﬂux is greater than 2. We have compared the integrated line ﬂuxes
obtained from the level 2 rebinned cubes with those from the projected cubes and ﬁnd
that they are consistent. Velocity and velocity width maps were constructed by ﬁtting
a single Gaussian to the projected data.
For the [O I] and [O Ib] lines, we report line ﬂuxes obtained from either the central
spaxel or the central 3 × 3 spaxels. The PSF corrected ﬂuxes should be viewed as lower
limits to the true [O I] and [O Ib] line emission in these galaxies. For NGC 5044 the
integrated [O I] and [O Ib] ﬂuxes are obtained after projecting the individual rebinned
cubes onto the sky.
9.2.2 Hα+[N II] imaging and spectroscopy
We performed narrow-band imaging observations for all systems in our sample, except
for NGC 4472, using the SOAR Optical Imager (SOI) on the 4.1 m SOAR telescope.
See Table 9.2 for details on the observations. For each galaxy, we obtained
two narrow-band images, one centered on the Hα emission and the other in an adjacent
emission-line free band. We reduced the images using standard procedures in the
IRAF MSCRED package. The pixels were binned by a factor of two, for a scale of 0.154
arcsec per pixel. The typical seeing was ∼1 . Spectrophotometric standard stars were
observed for each exposure. More detail on the SOI data reduction can be found in
Sun et al. (2007). For continuum subtraction, we follow the isophote ﬁtting method
described in Goudfrooij et al. (1994). We assumed line ratios of [N II]λ6583/Hα = 1.5
and [N II]λ6548/[N II]λ6583 = 1/3, which are consistent with the typical values in
Goudfrooij et al. (1994).
9.2 Observations and data analysis 167
Table 9.3: Summary of observations with Herschel PACS and the properties of the FIR
lines.
galaxy Line Observation duration Line Flux in central spaxel Observed FWHM Line shift Integrated Line Flux
(s) (10−14 erg s−1 cm−2) (km s−1) (km s−1) (10−14 erg s−1 cm−2)
NGC 1399 C IIλ157.7µm 2250 < 0.2 300f - O
Iλ63.2µm 2484 < 2.8 300f - O
Ibλ145.5 3360 < 0.2 300f - NGC
4472 C IIλ157.7µm 2250 0.2 ± 0.04 255 ± 36 99 ± 15 1.04 ± 0.16
O Iλ63.2µm 2484 < 2.5 255f - O
Ibλ145.5µm 3360 < 0.2 255f - NGC
4636 C IIλ157.7µm 1500 2.6 ± 0.06 361 ± 6 22 ± 3 10.52 ± 0.37
O Iλ63.2µm 1656 1.3 ± 0.2 (1.9) 233 ± 29 75 ± 12 O
Ib λ145.5µm 2240 < 0.5 233f - NGC
5044 C IIλ157.7µm 6750 - - - 31.73 ± 4.76
O Iλ63.2µm 7452 - - - 7.1 ± 1.8
O Ibλ145.5µm 10080 - - - 1.0 ± 0.5
NGC 5813 C IIλ157.7µm 3000 1.4 ± 0.04 419 ± 10 96 ± 4 8.55 ± 0.26
O Iλ63.2µm 3312 1.0 ± 0.2 (1.5) 273 ± 36 30 ± 15 O
Ibλ145.5µm 4480 0.2 ± 0.04 (0.4) 608 ± 88 100 ± 37 1.4 ± 0.2
NGC 5846 C IIλ157.7µm 1500 2.2 ± 0.06 477 ± 10 −25 ± 4 13.61 ± 0.36
O Iλ63.2µm 1656 < 2.5 477f - O
Ibλ145.5µm 2240 < 0.3 477f - NGC
6868 C IIλ157.7µ 1500 5.7 ± 0.08 510 ± 6 125 ± 3 21.61 ± 0.44
O Iλ63.2µm 1656 3.0 ± 0.3 506 ± 31 137 ± 13 5.9 ± 0.8
O Ibλ145.5µm 2240 0.5 ± 0.06 579 ± 50 125 ± 21 2.0 ± 0.3
NGC 7049 C IIλ157.7µm 1500 2.5 ± 0.06 395 ± 7 78 ± 3 22.49 ± 0.45
O Iλ63.2µm 1656 < 3.3 395f - O
Ibλ145.5µm 2240 < 0.2 395f - Notes:
In brackets, we indicate the PSF corrected ﬂuxes determined assuming the emission
comes from a point-source, correcting the measured spaxel ﬂux upward to account
for the beam size. The PSF corrected ﬂuxes can therefore be considered a lower limit
to the ﬂux integrated over the whole PACS area.
Table 9.4: Hα+[N II] ﬂuxes from the SOAR telescope.
Galaxy fHff+[NII]
(10−13 erg s−1 cm−2)
NGC 1399 < 0.34
NGC 4636 2.7 ± 0.4
NGC 5044 7.6 ± 0.9
NGC 5813 2.2 ± 0.3
Table 9.5: Deprojected thermodynamic properties at r ∼ 0.5 kpc.
Galaxy ne kT Pe K tcool
(cm−3) (keV) (keV cm−3) (keV cm2) (107 yr)
NGC 1399 0.167 ± 0.005 0.950 ± 0.009 0.158 ± 0.005 3.14 ± 0.07 4.6
NGC 4472 0.154 ± 0.008 0.860 ± 0.012 0.132 ± 0.007 2.99 ± 0.12 4.3
NGC 4636 0.072 ± 0.001 0.535 ± 0.020 0.039 ± 0.001 3.11 ± 0.07 5.2
NGC 5044 0.068 ± 0.005 0.582 ± 0.037 0.039 ± 0.004 3.51 ± 0.27 5.9
NGC 5813 0.074 ± 0.003 0.616 ± 0.041 0.046 ± 0.005 3.50 ± 0.31 5.7
NGC 5846 0.076 ± 0.006 0.677 ± 0.033 0.052 ± 0.005 3.77 ± 0.26 6.2
NGC 6868 0.072 ± 0.002 0.768 ± 0.036 0.055 ± 0.003 4.45 ± 0.22 7.8
168 Cold gas in giant ellipticals
0.5 kpc
Ha+[NII]
NGC 4636 2 kpc
Ha+[NII]
NGC 5044 2 kpc
Ha+[NII]
NGC 5813
1 kpc
Ha+[NII]
NGC 5846 2 kpc
Ha+[NII]
NGC 6868 2 kpc
Ha+[NII]
NGC 7049
Figure 9.1: Hα+[N II] images of the galaxies with extended emission line nebulae in
our sample obtained with the 4.1 m SOAR telescope.
On August 4 and September 8 2013, we also took spectra of NGC 1399 using the
Goodman spectrograph on the SOAR telescope. We used a 600 lines per mm grating
for the wavelength range of 4350–6950 ˚A. The ﬁrst run was taken with a 1.68 slit,
along the NS direction. The second run was taken with a 3.0 slit, at a position angle
of 302 degrees (almost along the major axis of the galaxy). We took three 480 second
exposures on the ﬁrst night and four 480 second exposures on the second night. Before
and after each exposure we took a quartz ﬂat and a comparison spectrum of an
Fe lamp. Bias correction, ﬂat-ﬁelding, and wavelength calibration were all performed
using standard IRAF procedures. For ﬂux calibration, we took spectra of the spectrophotometric
standard stars LTT7379 and LTT1020.
9.2.3 Chandra X-ray data
The archival Chandra data were reprocessed using the CIAO (version 4.3) software
package. We cleaned the data to remove periods of anomalously high background.
Table 9.2 summarizes the net exposure times after cleaning. Our data reduction procedures
are described in detail in Million et al. (2010c,d).
To produce 2D maps of projected thermodynamic quantities, we identiﬁed spatial
regions using the contour binning algorithm (Sanders 2006), which groups neighbouring
pixels of similar surface brightness until a desired signal-to-noise ratio threshold
is met. We adopted a signal-to-noise ratio of 18–25 (∼320–630 counts per region) in
the 0.6–2.0 keV band, which gives us small enough regions to resolve substructure, yet
provides enough counts to achieve better than 5 per cent precision on the temperature.
Point sources were identiﬁed using the CIAO task WAVDETECT and excluded from
all regions used for spectral analysis. Separate photon-weighted response matrices and
9.3 Results 169
effective area ﬁles were constructed for each spectral region.
We modelled the spectra extracted from each spatial region with the SPEX package
(Kaastra et al. 1996). The spectrum for each region was ﬁtted with a model consisting
of an absorbed single-phase plasma in collisional ionization equilibrium, with the
temperature and spectral normalization (emission measure) as free parameters. The
line-of-sight absorption column densities, NH, were ﬁxed to the values determined by
the Leiden/Argentine/Bonn radio survey of H I (Kalberla et al. 2005).
From the best ﬁt emission measures, Y = nHnedV, where nH = ne/1.2 is the
hydrogen number density and V is the emitting plasma volume, we determined the
projected electron densities, ne, for each region assuming a ﬁxed column depth of
l = 20 kpc. Using these electron densities and the best ﬁt plasma temperatures (kT), we
determined the pressures (Pe = nekT) and entropies (K = kT/n2/3
e ). We note that the
choice of the assumed column depth for these projected maps is arbitrary and it does
not affect the magnitude of the observed azimuthal variations in the derived thermodynamic
properties.
Because the statistical quality of the data for the X-ray faintest galaxies in our sample,
NGC 6868 and NGC 7049, does not allow us to produce 2D maps of thermodynamic
properties, we only present X-ray images for these systems. Backgroundsubtracted
images were created in six narrow energy bands, spanning 0.5–2.0 keV.
These were ﬂat ﬁelded with respect to the median energy for each image and then
co-added to create the broad-band X-ray images.
To determine the deprojected thermodynamic properties of the hot gas, we extracted,
for each galaxy (except NGC 7049), a set of azimuthally averaged spectra from concentric
annuli. We modeled these spectra simultaneously in the 0.6–2.0 keV band, with the
XSPEC (version 12.5 Arnaud 1996) spectral ﬁtting package, using the projct model (see
Werner et al. 2012, for details). We have determined the azimuthally-averaged deprojected
radial proﬁles of electron density and temperature, from which we derived the
deprojected radial proﬁles for entropy, electron pressure, and cooling time. We deﬁne
the cooling time as the gas enthalpy divided by the energy radiated per unit volume
of the plasma:
tcool =
5
2(ne + ni)kT
neniΛ(T)
, (9.1)
where the ion number density ni = 0.92ne, and Λ(T) is the cooling function for Solar
metallicity tabulated by Schure et al. (2009). We caution that, because the deprojected
thermodynamic properties are determined assuming spherical symmetry, for the 5/7
analyzed galaxies with disturbed morphologies, the derived values have signiﬁcant
systematic uncertainties.
9.3 Results
Using Herschel PACS, we detect FIR [C II]λ157µm line emission in 7/8 systems (see Table
9.3). The detections of [O I]λ63.2µm and [O Ib]λ145.5µm lines are relatively weak,
and are only seen in 4/8 and 3/8 systems, respectively. No cold neutral gas is detected
170 Cold gas in giant ellipticals
in NGC 1399 and the [C II] detection in NGC 4472 is relatively weak. Table 9.3 lists
the observed lines for all the target galaxies, their rest frame wavelengths and observation
durations. For the central spaxel, we present the observed line ﬂuxes and the
2σ upper limits where appropriate, the observed FWHM, and the observed line shifts
with respect to the systemic velocity determined based on optical data. In the brackets,
we give the PSF corrected ﬂuxes determined assuming that the emission comes from
a point-source, correcting the measured spaxel ﬂux upward to account for the beam
size. The PSF corrected ﬂuxes can therefore be considered a lower limit to the ﬂux
integrated over the whole PACS area. For sources that allow us to conﬁrm that the
emission is extended we report non-PSF corrected ﬂuxes. The last column gives the
line ﬂuxes spatially integrated over the 5 × 5 spaxel (47 × 47 ) Herschel PACS ﬁeld of
view.
The [O I]/[C II] line ratios determined from the central spaxel in systems with signiﬁcant
[O I] detections are in the range of 0.4–0.8. The 2σ upper limits on the ﬂuxes of
the [O I]λ63µm and [O Ib]λ145µm lines given in Table 9.3 were determined assuming
that their velocity widths are equal to the FWHM of the [C II] line.
As indicated in Section 9.1.1 and in Table 9.1, according to Macchetto et al. (1996)
all our targets are Hα luminous. Fig. 9.1 shows narrow band Hα+[N II] images obtained
with the SOAR telescope of our six targets with extended line-emitting nebulae.
The nebulae in NGC 6868 and NGC 7049 show regular extended disk-like morphologies
reaching radii r ∼ 2 − 3 kpc (see also Macchetto et al. 1996). On the other
hand, NGC 5044, NGC 5813, NGC 5846, and NGC 4636 show ﬁlamentary emission.
NGC 5044 has the largest optical emission line luminosity, with a rich, dense network
of ﬁlaments extending from the core out to a radius of ∼ 10 kpc (see also David et al.
2011; Gastaldello et al. 2009). The nebulae in NGC 5813 show a radial bipolar distribution
in the direction of buoyant bubbles rising in the hot X-ray emitting atmosphere of
the galaxy (see also Randall et al. 2011b). The bright Hα+[N II] emission in NGC 4636
and NGC 5846 appears relatively compact, concentrated in the innermost r 1.5 kpc
of the galaxies, but the morphology of the nebulae is clearly ﬁlamentary exhibiting
signs of interaction with the central AGN.
Our new SOAR data allow robust constraints on the Hα+[N II] ﬂux in four galaxies,
NGC 5044, NGC 4636, NGC 5813, and NGC 1399 (see Table 9.4). The NGC 5846
data were not taken in photometric conditions, and for NGC 6868 and NGC 7049, the
ﬁlters available at the time of the observations made the continuum subtraction less
reliable than what is achieved for other galaxies. Our Hα+[N II] ﬂux for NGC 4636 is
consistent with the values published by Macchetto et al. (1996) and Buson et al. (1993).
For NGC 5813, our result is consistent with Goudfrooij et al. (1994) but is almost a
factor of two higher than the ﬂux measured by Macchetto et al. (1996). For NGC 5044,
our ﬂux is signiﬁcantly higher than the values measured by Goudfrooij et al. (1994) and
Macchetto et al. (1996), but is consistent with the result of Rickes et al. (2004). The large
spatial extent of the nebula surrounding NGC 5044 combined with the small ﬁeld-ofview
of the instrument (1.4 × 2.2 ) may have prevented Macchetto et al. (1996) from a
robust determination of the local background.
Contrary to the results of Macchetto et al. (1996), we detected no Hα+[N II] emission
in NGC 1399. Goudfrooij et al. (1994) derived an Hα+[N II] ﬂux of 4.2 ± 0.2 × 10−14 erg
9.3 Results 171
Figure 9.2: The Hα+[N II] luminosity (bottom panel) from SOAR for NGC 1399,
NGC 5044, NGC 5813, and NGC 4636 (see Table 9.4) and from Macchetto et al. (1996)
for NGC 4472, NGC 5846, NGC 6868, and NGC 7049 (see Table 9.1), 70/24 µm infrared
continuum ratio (middle panel, Temi et al. 2007a), and K-band luminosity (top panel,
based on the 2MASS survey, Jarrett et al. 2003) plotted against the spatially integrated
[C II] luminosity measured by Herschel PACS. The grey doted lines on the bottom panel
indicate constant [C II] over Hα+[N II] luminosity ratios of 0.4 and 0.8. These ratios
are remarkably similar for 6/8 galaxies in our sample. The outliers, NGC 4472 and
NGC 1399, have little or no cold gas. They are indicated in red.
172 Cold gas in giant ellipticals
Figure 9.3: Jet powers (Allen et al. 2006; Shurkin et al. 2008; David et al. 2009; Randall
et al. 2011b) plotted against the spatially integrated [C II] luminosity measured by Herschel
PACS. Galaxies with little or no cold gas are indicated in red. For the jet-powers
in NGC 5044 and NGC 5813 no errorbars were published (Randall et al. 2011b; David
et al. 2009).
9.3 Results 173
52.0 51.0 12:42:50.0 49.0 48.0
50.040.030.020.010.02:41:00.040:50.0
Right ascension
Declination
NGC 4636
[CII] flux
0.5 kpc
0.5 1 1.5 2 2.5
52.0 51.0 12:42:50.0 49.0 48.0
50.040.030.020.010.02:41:00.040:50.0
Right ascensionDeclination
NGC 4636
[CII] velocity
0.5 kpc
-140 -120 -100 -80 -60 -40 -20 0 20 40 60
52.0 51.0 12:42:50.0 49.0 48.0
50.040.030.020.010.02:41:00.040:50.0
Right ascension
Declination
NGC 4636
[CII] velocity dispersion
0.5 kpc
100 150 200 250 300
28.0 26.0 24.0 22.0 13:15:20.0
22:30.0-16:23:00.030.024:00.0
Right ascension
Declination
NGC 5044
[CII] flux
2 kpc
0.1 0.2 0.3 0.4 0.5 0.6
28.0 26.0 24.0 22.0 13:15:20.0
22:30.0-16:23:00.030.024:00.0
Right ascension
Declination
NGC 5044
[CII] velocity
2 kpc
-120 -100 -80 -60 -40 -20 0 20 40 60
28.0 26.0 24.0 22.0 13:15:20.0
22:30.0-16:23:00.030.024:00.0
Right ascension
Declination
NGC 5044
[CII] velocity dispersion
2 kpc
80 100 120 140 160 180 200 220
13.0 12.0 11.0 15:01:10.0 09.0
30.020.010.01:42:00.050.041:40.0
Right ascension
Declination
1 kpcNGC 5813
[CII] flux
0.1 0.2 0.3 0.4 0.5
13.0 12.0 11.0 15:01:10.0 09.0
30.020.010.01:42:00.050.041:40.0
Right ascension
Declination
1 kpcNGC 5813
[CII] velocity
-40 -20 0 20 40 60 80 100 120 140 160
13.0 12.0 11.0 15:01:10.0 09.0
30.020.010.01:42:00.050.041:40.0
Right ascension
Declination
NGC 5813
[CII] velocity dispersion
1 kpc
80 100 120 140 160 180 200 220 240 260 280
31.0 15:06:30.0 29.0 28.0 27.0
50.040.030.020.010.01:36:00.035:50.0
Right ascension
Declination
2 kpcNGC 5846
[CII] flux
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
31.0 15:06:30.0 29.0 28.0 27.0
50.040.030.020.010.01:36:00.035:50.0
Right ascension
Declination
2 kpcNGC 5846
[CII] velocity
-100 -50 0 50 100 150
31.0 15:06:30.0 29.0 28.0 27.0
50.040.030.020.010.01:36:00.035:50.0
Right ascension
Declination
2 kpcNGC 5846
[CII] velocity dispersion
100 150 200 250 300
Figure 9.4: Left panels: Maps of the integrated [C II] line ﬂux in units of
10−14 erg s−1 cm−2 per 6 × 6 spaxel obtained with Herschel PACS for NGC 4636,
NGC 5044, NGC 5813, and NGC 5846. Central panels: The velocity distribution of
the [C II] emitting gas, in units of km s−1, relative to the systemic velocity of the host
galaxy. Right panels: Map of the velocity dispersion, σ, of the [C II] emitting gas. Contours
of the Hα+[N II] emission are overlaid.
174 Cold gas in giant ellipticals
57.0 56.0 55.0 20:09:54.0 53.0 52.0 51.0
22:20.030.040.050.0-48:23:00.010.020.0
Right ascension
Declination
2 kpcNGC 6868
[CII] flux
0.2 0.4 0.6 0.8 1 1.2 1.4 1.61.8 2
57.0 56.0 55.0 20:09:54.0 53.0 52.0 51.0
22:20.030.040.050.0-48:23:00.010.020.0
Right ascension
Declination
2 kpcNGC 6868
[CII] velocity
-100 -50 0 50 100 150 200 250
57.0 56.0 55.0 20:09:54.0 53.0 52.0 51.0
22:20.030.040.050.0-48:23:00.010.020.0
Right ascension
Declination
2 kpcNGC 6868
[CII] velocity dispersion
100 120 140 160 180 200 220 240 260
03.0 02.0 01.0 21:19:00.059.0 58.0 18:57.0
33:10.020.030.040.050.0-48:34:00.010.0
Right ascension
Declination
2 kpcNGC 7049
[CII] flux
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
03.0 02.0 01.0 21:19:00.059.0 58.0 18:57.0
33:10.020.030.040.050.0-48:34:00.010.0
Right ascension
Declination
2 kpcNGC 7049
[CII] velocity
-200 -100 0 100 200 300 400
03.0 02.0 01.0 21:19:00.059.0 58.0 18:57.0
33:10.020.030.040.050.0-48:34:00.010.0
Right ascension
Declination
2 kpcNGC 7049
[CII] velocity dispersion
120 140 160 180 200 220 240 260 280
Figure 9.5: Same as Fig. 9.4, but for NGC 6868 and NGC 7049.
s−1 cm−2, much smaller than the value measured by Macchetto et al. (1996). Our SOI
Hα+[N II] imaging analysis of NGC 1399, with either adjacent narrow-band or R band
image for the continuum subtraction, results in negative net ﬂux around the center of
the galaxy, indicative of Hα absorption. Therefore, we took additional long-slit spectra
with the Goodman spectrograph on SOAR. Our Goodman spectra give a 5 σ limit on
the EW of the emission line of 0.1 ˚A. Assuming that all the Hα+[N II] emission is from
the central 12 radius, as shown by the Hα+[N II] image from Macchetto et al. (1996),
this EW upper limit translates to an Hα+[N II] ﬂux limit of 3.4 × 10−14 erg s−1 cm−2.
The lack of a spectroscopic detection of optical emission lines in NGC 1399 is inconsistent
with previous reports of detections through narrow-band ﬁlters. We suspect that
the previously reported detections of morphologically smooth line emission, where the
emitting region has the same morphology as the old stellar populations, may be due
to color gradients in the stellar population of the galaxy. Currently, we do not have
SOAR data for NGC 4472.
In Fig. 9.2, we plot the Hα+[N II] luminosities (bottom panel), 70/24 µm infrared
continuum ratios (middle panel, Temi et al. 2007a), and the K-band luminosity (top
panel, based on the 2MASS survey, Jarrett et al. 2003) against the spatially integrated
[C II] luminosity measured by Herschel PACS. The ratio of the [C II] over Hα+[N II]
luminosity is remarkably similar ∼ 0.4–0.8 for 6/8 galaxies in our sample. For the
same 6/8 systems, the [C II]λ157µm line emission is spatially extended and appears to
follow the distribution of the Hα+[N II] ﬁlaments (see the left panels of Fig. 9.4 and 9.5).
In the extended network of ﬁlaments of NGC 5044 at r 1 kpc, the [C II]/(Hα+[N II])
ratios also remain similar as a function of position. In the two FIR faint systems (shown
9.3 Results 175
0.0005 0.001 0.0015 0.002 0.0025 0.003
NGC 4636
Pressure
0.5 kpc
10 15 20 25 30 35 40 45 50 55 60
NGC 4636
Entropy
0.5 kpc
0.004 0.006 0.008 0.01 0.012 0.014
NGC 5044
Pressure
2 kpc
10 15 20 25 30 35 40 45 50
NGC 5044
Entropy
2 kpc
0.004 0.006 0.008 0.01 0.012 0.014
NGC 5813
Pressure
1 kpc
10 15 20 25 30 35 40
NGC 5813
Entropy
1 kpc
0.008 0.012 0.016 0.02 0.024
1 kpcNGC 5846
Entropy
4 6 8 10 12 14 16 18 20 22 24
1 kpcNGC 5846
Entropy
Figure 9.6: 2D map of the hot gas pressure (in units of keV cm−3 × l
20kpc
−1/2
) and
entropy (in units of keV cm2 × l
20kpc
1/3
) with the Hα+[N II] contours overlaid. The
maps were obtained by ﬁtting each region independently with a single temperature
thermal model, yielding 1σ fractional uncertainties of ∼3–5 per cent in temperature,
entropy, and pressure. Point sources were excluded from the spectral analysis and
therefore appear as white circles on the maps.
0.02 0.04 0.06 0.08 0.1
1 kpc
Pressure
NGC 1399
10 20 30 40 50 60
1 kpc
Entropy
NGC 1399
0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045
1 kpc
Pressure
NGC 4472
5 10 15 20 25 30 35
1 kpc
Entropy
NGC 4472
Figure 9.7: 2D maps of the hot gas pressure and entropy for the two galaxies with little
or no [C II] emitting cold gas detected.
176 Cold gas in giant ellipticals
by the red data points) the ionized Hα emission is either concentrated in the nucleus
of the galaxy (NGC 4472, Macchetto et al. 1996) or is not present (NGC 1399).
The [C II] luminosities, which are likely excellent tracers of the cold molecular gas
mass (Crawford et al. 1985), do not correlate with the K-band luminosities (see Table
9.1), which trace the stellar masses of the galaxies. There is also no correlation
between the [C II] luminosities of systems with extended emission line nebulae and
the 70/24 µm infrared continuum ratios. In the FIR faint systems indicated by the red
data points these ratios are smaller (see the central panel of Fig. 9.2). The 70/24 µm
infrared continuum ratio is a good tracer of interstellar dust (Temi et al. 2007a).
Fig. 9.3 shows that the [C II] luminosities in systems with spatially extended emission
do not correlate with the jet powers determined from the work required to inﬂate
bubbles of relativistic plasma associated with the cavities in the hot X-ray emitting atmospheres
of the galaxies listed in Table 9.1. In our relatively small sample, the galaxies
with little or no cold gas (indicated in red) have higher jet powers than the systems
with signiﬁcant [C II] detections.
The central panels of Fig. 9.4 and 9.5 show the velocity distributions of the [C II]
line emitting gas with respect to the velocities of the host galaxies. In NGC 6868 and
NGC 7049 the velocity distribution of the [C II] emitting gas indicates that the cold gas
forms a rotating disk. This is consistent with the overall morphology of the optical line
emission nebulae. The velocity distribution of the [C II] emitting gas in the other four
systems with extended emission does not show any coherent structure.
The projected velocity dispersions measured from the [C II] lines along our line of
sight are generally in the range between 100 km s−1 and 200 km s−1, with the highest
measured values reaching ∼ 280 km s−1.
Fig. 9.6 shows maps of the pressure and entropy in the hot X-ray emitting intragroup/interstellar
medium for the galaxies where the cold gas has a ﬁlamentary morphology.
In all of these systems, the ﬁlaments are co-spatial with the lowest entropy
hot X-ray emitting gas. They also appear to anti-correlate and interact with the AGN
inﬂated bubbles of relativistic plasma, which provide signiﬁcant non-thermal pressure
in the X-ray atmospheres of the galaxies and therefore appear as regions of low
projected X-ray pressure on our maps. This is most clearly seen for the southwestern
ﬁlament in NGC 5813, which appears to be pushed aside by the large, older, outer bubble.
The inner portion of the ﬁlament also appears on the side and partly on the top
of the small, younger, inner southwestern bubble that is currently being inﬂated in the
core (for detailed discussion on the bubbles see Randall et al. 2011b). Similar interaction,
with the ﬁlaments being pushed around by the bubbles, is seen in NGC 5846 and
NGC 4636, both of which show ‘cavities’ in the Hα+[N II] images that coincide with
pressure decrements in their X-ray spectral maps. NGC 5044 shows a strongly disturbed
morphology due to both AGN activity and the intra-group medium sloshing
in the gravitational potential of the system (Gastaldello et al. 2009; David et al. 2009,
2011; O’Sullivan et al. 2013).
Fig. 9.7 shows the pressure and entropy maps for NGC 4472 and NGC 1399 - systems
with little or no cold and warm gas. Compared to the systems with ﬁlamentary
Hα nebulae, their thermodynamic maps have relatively regular morphologies. These
systems are among the nearby giant ellipticals with the most relaxed X-ray morpholo-
9.4 Discussion 177
gies at small radii (Werner et al. 2012).
NGC 6868 and NGC 7049 do not have X-ray data with the quality that would allow
us to extract detailed maps of thermodynamic properties and in Fig. 9.8 we only
show their X-ray images. The images show extended X-ray haloes, co-spatial with the
Hα+[N II] emission, surrounding the galaxies. The X-ray image of NGC 6868 shows a
relatively complex disturbed morphology (see also Machacek et al. 2010). The X-ray
data of NGC 7049 are very shallow and only allow us to detect the extended X-ray
emission. Four galaxies - NGC 5044, NGC 6868, NGC 7049, and NGC 5813 - have an
X-ray point source associated with the central AGN.
The deprojected central densities and pressures measured at r = 0.5 kpc in the morphologically
relaxed NGC 1399 and NGC 4472 appear signiﬁcantly higher than in the
other, more disturbed, [C II] bright systems (see Table 9.5). All [C II] bright galaxies
appear to have similar central densities and pressures. The entropies at r = 0.5 kpc
span a small range of 3.0–4.5 keV cm2 and the central cooling times are 4–8 × 107 yr.
At radii r 1 kpc the entropies of the galaxies containing cold gas (except NGC 6868)
are systematically lower than those of NGC 1399 and NGC 4472 (see Fig. 9.9). The
measured relatively low central densities and pressures of the morphologically disturbed
systems may partly be due to the additional non-thermal pressure support in
the central regions of the these systems. However, we also suspect a bias due to departures
from spherical symmetry. Due to the non-uniform surface brightness within
the circular annuli used in the deprojection analysis, densities in the outer shells may
be somewhat overestimated, leading to underestimated central densities. For the morphologically
disturbed systems, we repeated the deprojection analysis using wedges
and found, that despite the systematic uncertainties, at radii r 1 kpc the difference
in the shapes of the proﬁles shown in Fig. 9.9 remains robust.
9.4 Discussion
9.4.1 Properties of the cold ISM
The observed high [C II] luminosities indicate the presence of large reservoirs of cold
gas. Cold gas is evidently prevalent in massive giant elliptical galaxies with spatially
extended Hα+[N II] nebulae, even though their stellar populations are old and appear
red and dead (Annibali et al. 2007). NGC 5044 has previously also been detected in the
CO(2–1) line with the 30 m IRAM telescope with a total ﬂux of 6.6 × 10−17 erg s−1 cm−2
(Jeremy Lim & Francoise Combes, private communication). Assuming a CO(2–1) to
CO(1–0) ratio of 0.8, we obtain a [C II]λ158µm over CO(1–0) ﬂux ratio of 3817, which is
within the range of the ratios (1500–6300) observed in normal and star forming galaxies
and in Galactic molecular clouds (Crawford et al. 1985; Stacey et al. 1991).
Assuming the cold ∼100 K [C II] emitting gas is in thermal pressure equilibrium
with the surrounding hot X-ray emitting plasma, it will presumably form ﬁlaments
with densities ∼ 104 cm−3 and small volume ﬁlling fractions. As discussed in Fabian
et al. (2003b, 2008, 2011) and Werner et al. (2013), the emission line ﬁlaments are likely
to consist of many strands that have small volume ﬁlling fractions but large area cov-
178 Cold gas in giant ellipticals
2 kpcNGC 6868
0.5-2 keV image
0.5 1 1.5 2 2.5 33.5
0.5-2 keV image
NGC 7049 2 kpc
0.02 0.06 0.1 0.14
Figure 9.8: 0.5–2 keV Chandra images of the X-ray faintest galaxies in our sample,
NGC 6868 and NGC 7049, with the Hα+[N II] contours overlaid. The statistical quality
of the X-ray data does not allow us to produce 2D maps of thermodynamic properties
for these two systems.
ering factors. The soft X-ray emission that is always associated with ﬁlaments in cool
core clusters (Sanders & Fabian 2007; Sanders et al. 2008, 2009, 2010a; de Plaa et al.
2010; Werner et al. 2011, 2010, 2013) indicates the presence of cooling hot plasma distributed
between the threads. The hot plasma may be cooling both radiatively and by
mixing with the cold gas in the ﬁlaments. Part of the soft X-ray emission from this
cooling plasma gets absorbed by the cold gas in the strands (Werner et al. 2013).
The velocities inferred from the [C II] line emission are consistent with those measured
from the Hα+[N II] lines by Caon et al. (2000), indicating that the different observed
gas phases form multiphase ﬁlaments where all the phases move together. The
velocity structure measured in NGC 6868 and NGC 7049 indicates that the cold gas in
these systems is distributed in large extended rotating disks. Given the expected densities
of the [C II] emitting gas, its volume ﬁlling fraction must be low and the disks are
also likely to be formed from many small clouds, sheets, and ﬁlaments.
The observed velocity dispersions of the [C II] emitting gas are similar to the range
of values measured from CO line widths and using optical and near-infrared integral
ﬁeld spectroscopy of 2000–10,000 K gas in the cooling cores of clusters (e.g. Edge 2001;
Ogrean et al. 2010; Oonk et al. 2010; Farage et al. 2012; Canning et al. 2013). Observations
indicate that, due to their small volume ﬁlling fraction and large area covering
factors, ﬁlaments may be blown about by the ambient hot gas, potentially serving as
good tracers of its motions (Fabian et al. 2003b). The measured velocity dispersions
of the [C II] emitting gas, which reﬂect the range of velocities of the many small gas
clouds/ﬁlaments moving through the galaxy, may therefore be indicative of the motions
in the hot ISM. They are in general consistent with the limits and measurements
9.4 Discussion 179
0.1 1
110100
kT/n2/3
(keVcm2
)
Radius (kpc)
no extended nebulae
with extended nebulae
Figure 9.9: Entropy proﬁles for a sample of morphologically relaxed giant ellipticals
(NGC 4472, NGC 1399, NGC 4649, NGC 1407, NGC 4261; Werner et al. 2012), with no
extended optical emission line nebulae (in black), and for ﬁve galaxies from our sample
with [C II] emission indicating the presence of reservoirs of cold gas (NGC 5044,
NGC 5813, NGC 5846, NGC 4636, NGC 6868 - in red). The entropy in galaxies containing
cold gas is systematically lower at r 1 kpc. The only exception is NGC 6868 the
entropy proﬁle of which follows that of the cold-gas-poor galaxies.
180 Cold gas in giant ellipticals
obtained through X-ray line broadening and resonant line scattering observations with
the Reﬂection Grating Spectrometers on XMM-Newton (Werner et al. 2009; de Plaa et al.
2012; Sanders & Fabian 2013).
The mid-infrared Spitzer spectra of all of our [C II] bright galaxies overlapping with
the sample of Panuzzo et al. (2011) - NGC 4636, NGC 5044, NGC 5813, NGC 5846,
NGC 6868 - show the presence of dust, warm H2 molecular gas, and for NGC 5044,
NGC 6868, and NGC 4636 also PAH emission. Temi et al. (2007a) show that, for
NGC 4472 and NGC 1399, galaxies with little or no [C II] detected, the infrared spectral
energy distributions (SEDs) can be fully explained by circumstellar dust in the context
of a steady state model of dust production in normal stellar mass loss followed
by sputtering by the hot X-ray emitting gas. However, for all of the galaxies with extended
[C II] emission, the infrared SEDs observed with IRAS and Spitzer indicate the
presence of true interstellar dust, well in excess of the predictions of the steady state
model (Knapp et al. 1989; Temi et al. 2007a,b). Although there are early reports of detections
of H I emission in NGC 5846 and NGC 4636 (Bottinelli & Gouguenheim 1977,
1979; Knapp et al. 1978), they were not conﬁrmed by later re-observations (Krishna
Kumar & Thonnard 1983; Lake & Schommer 1984; Knapp et al. 1985).
Werner et al. (2013) showed that the [S II]λ6716/6731 line ratios of the ﬁlaments
in M 87 indicate very low densities in the Hα+[N II] emitting 10,000 K phase. They
concluded that, assuming subsonic turbulence in the ﬁlaments, the presence of significant
magnetic pressure is required to keep this warm ionized gas in pressure equilibrium
with the surrounding intra-cluster medium, indicating that the ﬁlaments are
supported by magnetic ﬁelds of B = 28 − 73µG. The assumption of subsonic turbulence
is motivated by the lack of [O III] line emission, which would be produced by
strong shocks. For 5/8 galaxies overlapping with the sample of Annibali et al. (2010),
the [S II]λ6717/6731 line ratios point to very low densities, ne < 26 cm−3, in the ionized
gas, indicating that it is supported by magnetic ﬁelds. Such low densities appear
to be common in the ionized, extended emission line nebulae in the cores of galaxy
clusters. Signiﬁcant magnetic ﬁelds threading the emission line nebulae were also inferred
using arguments based on the integrity of the ﬁlaments in the Perseus Cluster
(Fabian et al. 2008) and based on radio observations of the Faraday rotation measure
in cooling core clusters (Taylor et al. 2001, 2007; Allen et al. 2001; Feretti et al. 1999).
This magnetic support may be slowing or preventing the gravitational collapse of any
molecular gas clouds traced by the [C II] line emission that exceed the Jeans mass, preventing
them from forming stars (e.g. see the discussion for NGC 1275 in Ho et al.
2009).
9.4.2 Heating and ionization of the cold ISM
In the six systems with signiﬁcant cold gas mass reservoirs, the [C II] line emission
appears to be co-spatial with the Hα+[N II] emission, and with the lowest entropy Xray
emitting plasma. In the same galaxies, the ratios of [C II]/(Hα+[N II]) emission
appear to be similar (0.4–0.8), indicating that in these systems the [C II] and Hα+[N II]
emission are powered by the same energy source.
9.4 Discussion 181
Photoionization by young hot stars or by the central AGN is negligible as a source
of excitation for the nebulae in these galaxies. Ferland et al. (2009) showed that the
broad band emission-line spectra of the ﬁlamentary nebulae around central galaxies of
cooling core clusters most likely originate in gas exposed to ionizing particles, either
relativistic cosmic rays or hot X-ray emitting plasma penetrating into the cold gas. If
the magnetized, ionized Hα emitting phase forms a thin skin on the underlaying cold
neutral gas, then the hot plasma particles must somehow overcome the obstacle presented
by the magnetic ﬁelds. Based on the observations of the emission line ﬁlaments
in M 87, Werner et al. (2010) and Werner et al. (2013) propose that shocks propagating
within the hot X-ray emitting plasma as well as the movement of the ﬁlaments
through this ambient hot gas (both as they are being uplifted and as they fall back)
induce shearing around the ﬁlaments, thereby promoting mixing of the cold gas with
the ambient hot medium via instabilities (e.g. Friedman et al. 2012). Fabian et al. (2011)
propose that the ionizing hot plasma in the core of the Perseus Cluster penetrates the
cold ﬁlaments through magnetic reconnection diffusion (Lazarian et al. 2010, 2011),
which may be induced by shearing instabilities and turbulence. The penetration of the
cold gas by the hot plasma particles implies that the X-ray emitting gas cools through
mixing and thus the ﬁlaments of cold gas grow continuously in mass. If the mixing of
the cold and hot gas phases turns more violent and the energy input rate due to penetrating
hot plasma becomes larger than the cooling rate, this process may also lead to
the destruction of the ﬁlaments.
On the other hand, according to the scenario proposed by Churazov et al. (2013),
the ﬁlaments may be powered by the reconnection of the stretched magnetic ﬁelds in
the wakes of AGN inﬂated buoyantly rising bubbles, and do not necessarily grow in
mass. The fact that the ratios of the [C II]/(Hα+[N II]) emission appear similar in both
ﬁlamentary nebulae and rotating disks indicates that, despite the different morphologies,
the gas is heated and ionized by the same energy source in both types of systems.
But because the cold gas in the disky nebulae is not distributed in the wakes of AGN
inﬂated bubbles, the scenario proposed by Churazov et al. (2013) could not explain
their energy source. However, in the reconnection model, the powering of the cold gas
could also be driven by the orbiting motion of the magnetized blobs/ﬁlaments through
the hot plasma. This may lead to the sliding of the cold gas along the opposite-polarity
magnetic ﬁelds and to the subsequent reconnection.
Despite its relatively low X-ray luminosity, the central pressure of the hot X-ray
emitting plasma in the disky NGC 6868 is similar to the central pressures in the ﬁlamentary
systems (see Table 9.5) indicating that in the absence of magnetic ﬁelds the
heat ﬂux from the hot into the cold phase would also be similar in both types of systems.
It therefore appears that if the cold and hot phase come into direct contact - either
due to turbulence, shearing motions, or the magnetic ﬁelds tied to the ionized phase
becoming unstable - hot gas particles penetrating into the cold gas provide a viable
mechanism for powering the ﬁlaments in all of the systems in our sample. The penetrating
particles will heat and increase the degree of ionization of the molecular gas
clouds present within the ﬁlaments, increasing their Jeans mass and slowing down the
collapse of the gas clouds in the presence of magnetic ﬁelds.
This model, calculated by Ferland et al. (2009) considering emission from an opti-
182 Cold gas in giant ellipticals
cally thin cell of gas, however, predicts [O I]/[C II]∼ 21, signiﬁcantly higher than our
observed range of 0.5–0.7. The ratios previously observed in the Perseus and Centaurus
clusters, and in Abell 1068 and Abell 2597 are also low, in the range of 0.3–1 (Edge
et al. 2010; Mittal et al. 2011, 2012). Mittal et al. (2012) conclude that these line ratios
suggest that the lines are optically thick, implying a large reservoir of cold gas, which
was not accounted for in previous inventories of the ﬁlament masses. Our measured
ratios are consistent with the values determined previously in normal and starburst
galaxies (Malhotra et al. 2001).
9.4.3 Origin of the cold ISM
A hotly debated question in the literature is whether the cold gas seen in the central
giant elliptical galaxies in clusters and groups, and in early type galaxies with X-ray
haloes, originated from the radiative cooling of the hot plasma, from stellar mass loss,
or whether it was accreted in mergers with gas rich galaxies.
The difference between the entropy proﬁles of the systems with cool gas and those
without it in Fig. 9.9 supports the argument that the cold gas originates by cooling from
the hot phase. In more massive clusters, Hα ﬁlaments are always co-spatial with soft
X-ray emitting ∼0.5 keV gas (Sanders & Fabian 2007; Sanders et al. 2008, 2009, 2010a;
de Plaa et al. 2010; Werner et al. 2011, 2010, 2013), indicating that the cold gas must be
related to the hot phase. Furthermore, signiﬁcant star formation and line emitting gas
around the central galaxy are only found in systems where the central cooling time is
short, or, equivalently, the central entropy is low (Rafferty et al. 2008; Cavagnolo et al.
2008).
The parameter
ΠF =
κT
nenHΛ(T)r2
, (9.2)
where κ is the thermal conductivity, T is the temperature, Λ is the cooling function, and
ne and nH are the electron and hydrogen number densities, respectively, is a measure
of the ratio of the conductive heating rate to the radiative cooling rate on scales comparable
to r, i.e. the “Field criterion” for thermal instability (Field 1965). Fig. 9.10 shows
the Field criterion, ΠF, as a function of the radius, where the systems with substantial
quantities of cold gas are plotted in red. There is a clear dichotomy with the coldgas-rich
systems remaining unstable out to relatively large radii. This result strongly
indicates that the cold gas is produced chieﬂy by thermally unstable cooling from the
hot phase. A similar result was found for a sample of central galaxies in more massive
clusters by Voit et al. (2008), who found minimum values of ΠF 5 in all systems with
star formation. Voit et al. (2008) argued that the effective conductivity is suppressed
by a factor of about 5 by magnetic ﬁelds, in which case all of their systems with young
stars are thermally unstable by the Field criterion.
The Field criterion alone is insufﬁcient to ensure that the gas is thermally unstable.
In a spherical atmosphere, the gas is supported by buoyancy, so that an overdense
blob tends to fall inwards on the free-fall timescale towards its equilibrium position
where the surrounding gas has the same speciﬁc entropy and requires the same speciﬁc
9.4 Discussion 183
0.1 1
0.1110100
Fieldstabilityparameter
Radius (kpc)
no extended nebulae
with extended nebulae
Figure 9.10: The Field criterion, ΠF = κT/(nenHΛ(T)r2), as a function of radius for the
galaxies that display extended emission line nebulae and [C II] line emission (shown
in red) and for a sample of arguably cold-gas-poor systems (shown in black). The clear
dichotomy, with the cold gas-rich systems being thermally unstable out to relatively
large radii, indicates that the cold gas originates from the cooling hot phase. The galaxy
NGC 6868, which shows evidence for a rotating disk of gas, is shown with a dashed
line.
184 Cold gas in giant ellipticals
heating rate. This suppresses the growth of thermal instability unless the cooling time
is comparable to the free-fall time or shorter (Cowie et al. 1980; Balbus & Soker 1989).
Using numerical simulations, Sharma et al. (2012b) found that thermal instability is
only signiﬁcant when the cooling time is less than ∼ 10 free fall times (see also McCourt
et al. 2012; Kunz et al. 2012; Gaspari et al. 2012, 2013). However, this condition can
be circumvented if an overdense gas blob is prevented from falling to its equilibrium
position by the magnetic ﬁelds which pin the cloud to the ambient ICM (Nulsen 1986),
or by rotational support in addition to buoyancy.
If a cooling cloud has sufﬁcient angular momentum that is retained while it cools,
it can cool unstably onto a non-radial orbit. This provides an alternative condition for
thermal instability, that the viscous diffusion length in a cooling time,
√
νtc, where ν
is the kinematic viscosity and tc is the cooling time, must be smaller than the size of a
cloud. Analogous to the Field criterion, we deﬁne the stability parameter
Πv =
νtc
r2
, (9.3)
which needs to be smaller than about unity for thermal instability to develop on scales
comparable to r. Viscous stresses in a plasma (Braginskii 1965) are much less sensitive
to the structure of the magnetic ﬁeld than conduction (e.g., Nulsen & McNamara
2013) and, although the stresses have a different tensor character in magnetized and
unmagnetized plasmas, the viscosity itself is the same. Because the underlying process
is diffusion controlled by Coulomb collisions in both cases (electrons for ΠF and
ions for Πv), the ratio of these two parameters is almost constant at Πv/ΠF 0.0253.
Numerical simulations are needed to determine a more accurate stability threshold,
but we expect that the viscous diffusion length needs to be somewhat smaller than r
for instability by this mechanism. The condition ΠF < 5 translates to Πv 0.13, or
the diffusion length in one cooling time being 0.36r. Although it is unclear which
stability threshold takes precedence, we conclude that it is sufﬁcient to test only one of
these parameters, which we take to be ΠF, for stability.
Our Herschel data reveal that in two galaxies, NGC 6868 and NGC 7049, the cold gas
forms rotating disks with radii r ∼ 3 kpc. The disks of cold gas in these systems rotate
with orbital velocities of up to 250 km s−1. If this gas cooled from the hot phase, its
rotation implies that the hot atmospheres of these systems have signiﬁcant net angular
momentum and the hot gas cools onto non- radial orbits. In the case of NGC 6868, this
angular momentum may have been provided by gas sloshing indicated by the Chandra
data (Machacek et al. 2010). Because the atmosphere is aspherical, heating originating
from the center of the galaxy cannot balance the cooling rate locally throughout the atmosphere
and so should not be expected to stop hot gas from cooling onto an extended
disk. Because of this, and because rotational support prevents infall, instabilities may
develop more easily and the gas may become thermally unstable at higher ΠF than in
non-rotating systems. This is consistent with our results presented in Fig. 9.10, where
ΠF for NGC 6868, shown with a dashed line, is higher than ΠF for the other systems
containing non-rotating cold gas. The Field criterion still needs to be met, but the behavior
of the stability parameter in NGC 6868 also indicates that the viscous stability
criterion takes precedence.
9.4 Discussion 185
However, in the context of the radiative cooling model, the presence of dust and
PAHs within the cold gas clouds (Panuzzo et al. 2011) is intriguing. It suggests that
at least some of the cold material originated from stellar mass loss (see also Voit &
Donahue 2011; Donahue et al. 2011, for galaxy clusters). Our understanding of dust
formation is, however, limited and it cannot be ruled out that dust may somehow form
within the cold gas clouds (Fabian et al. 1994). Recent Herschel observations suggest
that in the most massive galaxies, the dust mass is unconnected to the stellar populations,
leading to the suggestion that the dust and the cold gas in these systems have
been acquired externally (Smith et al. 2012). Rawle et al. (2012) show that the dustto-stellar
mass ratios in BCGs without line emitting nebulae are much smaller than in
systems with cold gas. In the absence of cold gas clouds, the dust produced by stellar
mass loss will get destroyed by sputtering and the tenuous gas, lost by stars, may get
more easily mixed with and assimilated into the hot ISM. Therefore, if a galaxy loses
all of its cold gas content (e.g. due to AGN activity), then stellar mass loss will likely
become inefﬁcient in rebuilding its cold gas reservoir. A galaxy that loses its cold gas
content will therefore remain free of gas and dust until it again accretes cold material
from the cooling hot phase or in a wet merger. This may have happened to NGC 1399
and NGC 4472. On the other hand, if a galaxy already contains clouds of cold gas (it
either never lost its cold ISM or it accreted some cold material externally) then these
clouds and ﬁlaments will help preserving the products of stellar mass loss (both the
gas and the dust) by assimilating them as they plow through the galaxy.
In the systems with ﬁlamentary cold gas, the AGN jets and the buoyant bubbles
of relativistic plasma seem to interact with the ﬁlaments, dragging them out from the
center of the galaxy. As these ﬁlaments move through the hot ISM they may get penetrated
by hot particles, which deposit energy into the cold gas heating and ionizing
it. This interaction can both destroy the cold gas and contribute to the growth of its
mass, depending on the energy input rate from the hot into the cold phase. In the context
of this model, the total gas/dust mass will not correlate with the stellar mass of
the galaxy, but it will depend on the balance between heating and cooling and on the
interaction of the cold nebulae with the AGN (see also Mathews et al. 2013).
9.4.4 Fueling the AGN
The jet-powers, determined from the work required to inﬂate bubbles of relativistic
plasma associated with the cavities in the hot X-ray emitting atmospheres, do not increase
with the amount of cold gas in these galaxies. On the contrary, the two galaxies
that lack any signiﬁcant reservoirs of cold gas have the largest jet-powers in our sample.
These galaxies, NGC 1399 and NGC 4472, also have the largest hot ISM densities
in their cores. Accretion of hot gas in galaxies with high core densities can naturally
result in higher jet powers, as has been demonstrated by the correlation between the
Bondi accretion rates of hot gas and the observed jet powers in such ellipticals (Allen
et al. 2006). The pressure and entropy distributions of the hot ISM in the cores of these
two galaxies look remarkably regular, with the X-ray cavities/radio-lobes appearing
at larger radii of ∼4 kpc and ∼8 kpc for NGC 4472 and NGC 1399, respectively.
Other nearby giant elliptical galaxies that appear morphologically relaxed in their
186 Cold gas in giant ellipticals
cores, such as NGC 4649, NGC 1407, NGC 4261, or NGC 1404, also do not display
extended optical emission line nebulae (Macchetto et al. 1996; Baldi et al. 2009; Tremblay
et al. 2009; Moustakas et al. 2010), indicating that they too are relatively depleted
of cold gas. Interestingly, some of these morphologically relaxed cold-gas-poor systems,
such as NGC 4261, have very large jet powers, with jets puncturing through
the X-ray emitting galactic atmosphere to form giant lobes in the surrounding intragroup
medium (O’Sullivan et al. 2011). For others, such as NGC 1404, NGC 4649, and
NGC 1407, however, the current jet powers are more modest (Shurkin et al. 2008; Giacintucci
et al. 2012).
Galaxies containing large reservoirs of cold gas, on the other hand, in general display
strongly disturbed X-ray morphologies, possibly relating to earlier AGN outbursts
impacting the cores of these systems. In principle, the presence of cold gas may
affect both accretion and jet heating in these systems. Accretion of clumpy multiphase
gas (hot plasma mixed with cooler, ﬁlamentary gas) may result in variable accretion
rates and power output of the AGN jets, potentially triggering sporadic, larger outbursts
than would be typical from the steady accretion of hot gas alone. Because of the
relatively small volume ﬁlling fraction of this cooler gas, these outbursts may be relatively
short (arguments for sporadic cold/multi-phase accretion in radio mode AGN
in galaxy clusters have also been presented by Rafferty et al. 2006; Russell et al. 2013a).
As demonstrated by Morganti et al. (2013), radio-mechanical feedback does not only
operate on the hot tenuous atmospheres of galaxies but also on the dense cold gas,
which can be accelerated by jets to high speeds. Recent observations using ALMA also
show that radio mode AGN activity can drive massive outﬂows of molecular gas from
BCGs (Russell et al. 2013b; McNamara et al. 2013). In systems with large cold gas reservoirs,
jets may be more easily slowed by coupling to this material, and may therefore
deposit most of their energy at smaller radii. This will result in more disturbed central
thermodynamic distributions and, in particular may decrease the central hot ISM
density, resulting in lower hot gas accretion rates between the relatively brief periods
of multiphase accretion. Any reduction in the accretion rates from the hot gas would
result in smaller jet powers, consistent with the observations presented here.
Numerical simulations (Wagner & Bicknell 2011; Wagner et al. 2012) have shown
that radio jets can be efﬁcient in clearing the central regions of galaxies of their cold gas,
provided that the cold gas clouds are porous, clumpy, and have small volume ﬁlling
fractions, with individual clouds being relatively small, which again are characteristics
consistent with the observations discussed here. Powerful jets may thus eventually
destroy or remove most of the cold gas in the cores of the galaxies, allowing the jets to
propagate further out and deposit their energy at larger radii, increasing the entropy
of the hot galactic atmospheres. This in turn will allow the density of the hot gas in
the core of the galaxy to increase, thus increasing the jet power. The central regions
of galaxies cleared of cold gas may acquire relaxed morphologies within the central
r ∼ 5 kpc over ∼ 5 × 107 yr or a few sound crossing times. The similarity of the
thermodynamic proﬁles of elliptical galaxies with highly relaxed X-ray morphologies
(see black data points in Fig. 9.9 and Werner et al. 2012) indicates that, after elliptical
galaxies are cleared of cold gas, jet heating and cooling of their hot atmospheres comes
to an equilibrium in an approximately universal manner. The galaxies may maintain
9.5 Conclusions 187
these equilibria until they are disturbed by a merger event.
9.5 Conclusions
Using FIR, optical, and X-ray data, we study the ISM in eight nearby, X-ray and optically
bright, giant elliptical galaxies, all central dominant members of relatively low
mass groups. We ﬁnd that:
• All systems with extended Hα emission in our sample (6/8 galaxies) display signiﬁcant
[C II] line emission that traces ∼100 K gas.
• This emission is co-spatial with the optical Hα+[N II] emitting nebulae and the
lowest entropy soft X-ray emitting plasma.
• These systems have similar [C II]/(Hα+[N II]) ratios of 0.4–0.8, indicating that the
[C II] and Hα+[N II] emission are powered by the same energy source. The likely
dominant source of energy is the hot X-ray emitting plasma penetrating into the
cold gas.
• The entropy proﬁles of the hot galactic atmospheres show a clear dichotomy, with
the systems displaying extended emission line nebulae having lower entropies
beyond r 1 kpc than the cold-gas-poor systems. We show that while the hot
atmospheres of the cold-gas-poor galaxies are thermally stable outside of their innermost
cores, the atmospheres of the cold-gas-rich systems are prone to cooling
instabilities. This provides considerable weight to the argument that cold gas in
giant ellipticals is produced chieﬂy by cooling from the hot phase. We show that
cooling instabilities may develop more easily in rotating systems and discuss an
alternative condition for thermal instability for this case.
• The hot atmospheres of cold-gas-rich galaxies display disturbed morphologies
indicating that the accretion of clumpy multiphase gas in these systems may result
in variable power output of the AGN jets, potentially triggering sporadic,
larger outbursts. The jets may be more easily slowed by coupling to the dense
cold material and may therefore deposit most of their energy at smaller radii,
resulting in the observed disturbed central thermodynamic distributions.
• In the two cold-gas-poor, X-ray morphologically relaxed galaxies of our sample,
NGC 1399 and NGC 4472, powerful AGN outbursts may have destroyed or removed
most of the cold gas from the cores, allowing the jets to propagate and
deposit most of their energy further out, increasing the entropy of the hot galactic
atmospheres and leaving their cores relatively undisturbed. These observations
indicate that radio-mechanical AGN feedback is likely to play a crucial role in
clearing giant elliptical galaxies of their cold gas, keeping them ‘red and dead’.
188 Cold gas in giant ellipticals
Acknowledgments
This work is based in part on observations made with Herschel, a European Space
Agency Cornerstone Mission with signiﬁcant participation by NASA. Support for this
work was provided by NASA through award number 1428053 issued by JPL/Caltech.
This work is based in part on observations obtained at the Southern Astrophysical
Research (SOAR) telescope, which is a joint project of the Minist´erio da Ciˆencia, Tecnologia,
e Inovac¸˜ao (MCTI) da Rep´ublica Federativa do Brasil, the U.S. National Optical
Astronomy Observatory (NOAO), the University of North Carolina at Chapel Hill
(UNC), and Michigan State University (MSU). SWA acknowledges support from the
U.S. Department of Energy under contract number DE-AC02-76SF00515. MR acknowledges
the NSF grant AST 1008454 and NASA ATP grant (12-ATP12-0017).
Chapter 10
A uniform metal distribution in the
intergalactic medium of the Perseus
cluster of galaxies
N. Werner1,2, O. Urban1,2,3, A. Simionescu1,2,4, S. W. Allen1,2,3
1Kavli Institute for Particle Astrophysics and Cosmology, Stanford University, 452 Lomita
Mall, Stanford, CA 94305-4085, USA
2Department of Physics, Stanford University, 382 Via Pueblo Mall, Stanford, CA 94305-
4060, USA
3SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, CA 94025,
USA
4Institute of Space and Astronautical Science (ISAS), JAXA, 3-1-1 Yoshinodai, Chuo-ku,
Sagamihara, Kanagawa 252-5210, Japan
Published in Nature, volume 7473, pages 656–658, 2013
Most of the metals (elements heavier than helium) ever produced by stars in
the member galaxies of galaxy clusters currently reside within the hot, X-ray emitting
intra-cluster gas. Observations of X-ray line emission from this intergalactic
medium have suggested a relatively small cluster-to-cluster scatter outside of the
cluster centers (Matsushita 2011; Leccardi & Molendi 2008) and enrichment with
iron out to large radii (Fujita et al. 2008b; Simionescu et al. 2011; Urban et al. 2011)
leading to the idea that the metal enrichment occurred early in the history of the
Universe (Fujita et al. 2008b). Models with early enrichment predict a uniform
metal distribution at large radii in clusters, while late-time enrichment, favored by
some previous studies (Balestra et al. 2007; Maughan et al. 2008), is expected to introduce
signiﬁcant spatial variations of the metallicity. To discriminate clearly between
these competing models, it is essential to test for potential inhomogeneities by measuring
the abundances out to large radii along multiple directions in clusters, which
has not hitherto been done. Here we report a remarkably uniform measured iron
abundance, as a function of radius and azimuth, that is statistically consistent with
190 Metal enrichment of the Perseus cluster
a constant value of ZFe = 0.306 ± 0.012 Solar out to the edge of the nearby Perseus
Cluster. This homogeneous distribution requires that most of the metal enrichment
of the intergalactic medium occurred before the cluster formed, likely over 10 billion
years ago, during the period of maximal star formation and black hole activity.
Between 2009 and 2011, we obtained a total of 84 observations of the Perseus Cluster
with the Suzaku X-ray satellite, as a Key Project for that mission. The pointings covered
eight azimuthal directions from the cluster center out to an offset angle of 2 degrees,
with a total exposure time of over 1 million seconds. We have analyzed the data from
all three functioning X-ray Imaging Spectrometers, extracting spectra from annuli centered
on the cluster center. We modeled the spectra from each of the 76 independent
regions as a single-temperature thermal plasma in collisional ionization equilibrium,
with the temperature, iron abundance, and spectrum normalization included as free
parameters (Urban et al. 2014).
The measured radial and azimuthal variation of the iron abundance of the hot intracluster
medium (ICM) out to the edge of the Perseus Cluster along the eight different
directions is presented in Fig. 10.1. We deﬁne this ‘edge as r200, the radius within which
the mean enclosed mass density of the cluster is 200 times the critical density of the
Universe at the cluster redshift (for the Perseus Cluster r200 = 1.8 Mpc, corresponding
to 82 arcmin4). We have tested the statistical uniformity of the measured iron abundance
at radii r > 400 kpc (i.e. beyond the central metallicity peak associated with
the brightest cluster galaxy Simionescu et al. 2011; De Grandi et al. 2004) by modeling
the data from all radii and azimuths with a constant abundance. The measured chisquare
value of 65.8 for 75 degrees of freedom is consistent with the null hypothesis of
a constant iron abundance with a value of ZFe = 0.306 ± 0.012 Solar (68% conﬁdence
limit; we adopt a Solar iron abundance of 3.24 × 10−5 relative to hydrogen by number
Feldman 1992). This abundance is also consistent with the mean value measured
at intermediate radii (∼ 0.2–0.5r200) for a sample of 48 clusters previously observed
with XMM-Newton (Leccardi & Molendi 2008). The iron abundance values measured
at large radii in the less massive Virgo Cluster are lower, although these are likely biased
by the multi-temperature structure of the gas in that system (Urban et al. 2011).
The azimuthally averaged iron abundance proﬁle for the Perseus Cluster is shown in
Fig. 10.2.
Enrichment scenarios in which the ICM is predominantly enriched with metals after
the formation of clusters predict a non-uniform metal distribution, which is inconsistent
with the observations presented here. Ram-pressure stripping (Gunn & Gott 1972)
would result in a negative gradient in the metal abundance proﬁles out to large radii, as
well as signiﬁcant azimuthal variations, with higher metal abundances expected along
directions connecting to surrounding large scale structure ﬁlaments (Domainko et al.
2006). Furthermore, metals ejected by galaxies at later times should roughly follow
the distribution of galaxies, producing an approximately constant metal mass to light
ratio, which is also inconsistent with observations (Matsushita et al. 2013). Although
large scale sloshing motions (Simionescu et al. 2012) may be able to mix metals to some
degree, the strong gradients in the entropy distributions of cluster atmospheres (Voit
191
et al. 2005), including Perseus (Simionescu et al. 2011; Urban et al. 2014), make the
ICM convectively stable, prohibiting the efﬁcient mixing of metals that are initially
non-uniformly distributed across large radial ranges.
The observed uniform iron abundance distribution at large radii in the Perseus Cluster
requires early enrichment of the intergalactic gas in the proto-cluster environment.
This enrichment was most likely driven primarily by galactic winds (De Young 1978) energetic
outﬂows of metal enriched gas - which are expected to be strongest at epochs
around the peak of star formation and active galactic nuclei (AGN) activity (redshifts
z ∼ 2–3, or lookback times of 10–12 billion years Madau et al. 1996; Brandt & Hasinger
2005). The combined energy of supernova explosions (De Young 1978) and AGN (Fabjan
et al. 2010) must have been strong enough to expel most of the metals from the
galaxies at early times, and enrich and mix the intergalactic gas. This gas was later accreted
by clusters and virialized (increasing its entropy through shock heating) to form
the present ICM.
The total iron mass in the ICM of the Perseus Cluster, calculated from the measured
iron abundance and ICM density proﬁles (Urban et al. 2014), is about 50 billion Solar
masses, with about 60% residing beyond 0.5r200. The dominant fraction of this iron was
likely supplied by type Ia supernovae (SNIa), which are thought to have produced 60-
90% of iron in the Perseus Cluster (Matsushita et al. 2013) depending on the assumed
supernova yields. Based on our measurements, we estimate that at least 40 billion
SNIa contributed to the chemical enrichment of the proto-cluster environment that
later formed the Perseus Cluster. For such a large iron mass to be expelled by the
winds from the galaxies at early times, a signiﬁcant fraction of these SNIa must have
exploded shortly after the epoch of peak star-formation. This is consistent with recent
ﬁndings based on SNIa delay time distributions (Mannucci et al. 2006; Maoz et al.
2012), which imply that a large fraction of SNIa explode less than ∼ 5 × 108 years after
the formation of the progenitor binary system (prompt SNIa). SNIa with longer delay
times will continue contributing to the enrichment of the ICM after clusters virialize,
and are expected to be partly responsible for the metallicity peaks surrounding the
brightest cluster galaxies in the centers of many clusters (B¨ohringer et al. 2004).
A unique prediction of the early enrichment scenario is that essentially all galaxy
clusters with masses comparable to the Perseus Cluster should have homogeneous
iron abundance distributions of about one-third Solar at large radii. Also, contrary
to some initial ﬁndings (Balestra et al. 2007; Maughan et al. 2008), there should be no
substantial redshift evolution in the ICM metallicity outside of the central regions of
clusters, out to z ∼ 2. The presence and strength of such evolution are a matter of ongoing
debate (Leccardi & Molendi 2008; Baldi et al. 2012; Andreon 2012). The observed
large iron abundance of the high-entropy gas in the outskirts of the Perseus Cluster is
also consistent with the idea that the highest-energy cosmic rays (above a few 1018 eV)
may be primarily iron nuclei (Abraham et al. 2010) accelerated by cluster formation
shocks (Norman et al. 1995). Additionally, because much of the metal rich ICM seen
at large radii has been accreted from the surrounding large scale structure ﬁlaments,
our model predicts that the tenuous warm-hot intergalactic medium (WHIM) permeating
the cosmic web, in which up to half of the baryons in the Universe currently
reside (Cen & Ostriker 1999; Dav´e et al. 2001), is likely to be substantially enriched in
192 Metal enrichment of the Perseus cluster
Figure 10.1: Iron abundance proﬁles expressed in Solar units (Feldman 1992) measured
along eight different directions in the Perseus Cluster. At the temperature of the
Perseus Cluster, the iron abundance measurement is driven by the Fe-K lines. Compared
to the Fe-L lines, these lines are signiﬁcantly less prone to systematic biases associated
with the presence of multi-temperature gas. The spectral extraction regions
span a relatively small range in radius and azimuth, where the plasma can be well approximated
as single-temperature. There is no evidence for a bias in the temperature
measurements due to gas clumping (Urban et al. 2014). We have veriﬁed that spectral
ﬁts with the Fe-L line complex (0.8–1.5 keV energy range) excluded give consistent
results to those in the full 0.7–7 keV band. Collisional ionization and electron-ion equilibrium
are likely to hold out to r200 in the Perseus Cluster, where the equilibration
timescales are only ∼ 3.5 × 108 yr and ∼ 7.3 × 108 yr, respectively.Within a radius
r ∼ 400 kpc (∼ 0.2r200; not shown here) metal enrichment is strongly inﬂuenced by the
brightest cluster galaxy and the metallicity is centrally peaked (Simionescu et al. 2011;
De Grandi et al. 2004). Along the eastern direction, large scale sloshing motions are
present and are uplifting the ICM from smaller radii out to r ∼ 650 kpc (Simionescu
et al. 2012). Thus we also do not present results for r < 650 kpc in this direction.
Beyond these radii, the iron abundance measurements are consistent with a radially
and azimuthally constant value of ZFe = 0.306 ± 0.012 Solar, indicated by the dotted
lines. Expressed in units of an older, and historically more commonly used Solar iron
abundance value (Anders & Grevesse 1989), our best ﬁt constant iron abundance is
ZFe = 0.212 ± 0.008 Solar. The plotted error bars are 68% conﬁdence intervals.
193
Figure 10.2: Azimuthally averaged iron abundance proﬁle for the Perseus Cluster. The
proﬁle was determined by simultaneous ﬁtting of the spectra at a given radius for all
eight azimuthal directions. While the iron abundance was assumed to be the same in
all directions in the ﬁts, the temperatures and spectral normalizations were allowed to
vary independently along the different azimuths. The dotted line shows the best ﬁt
constant iron abundance value of ZFe = 0.306 Solar. The plotted error bars are 68%
conﬁdence intervals. Our iron abundance values are lower than the value measured
in the 650–1100 kpc range with XMM-Newton (Matsushita et al. 2013); however, the
XMM-Newton measurement suffers from signiﬁcant systematic uncertainties due to
the high and variable particle background of that instrument compared to Suzaku.
194 Metal enrichment of the Perseus cluster
metals. The same is true for the hot circumgalactic gas accreted by galaxies from the
intergalactic-medium.
Acknowledgements
The authors are grateful for discussions with other members of the Perseus Cluster
Suzaku Key project collaboration, as well as with Yu Lu, Paul Simeon, and Roger
Blandford. This work was supported by the Suzaku grants NNX09AV64G, NNX10AR48G,
NASA ADAP grant NNX12AE05G, and by the U.S. Department of Energy under contract
number DE-AC02-76SF00515.
Chapter 11
Summary and outlook
In the enclosed papers, we advanced our understanding of the radio-mode AGN feedback,
and of the microphysics and the chemical enrichment of the ICM. We provided
some of the clearest views to date of AGN feedback in action, stripping central cluster
galaxies of their lowest entropy gas (Chapter 2 and 3; Werner et al. 2010, 2011); developed
the use of resonance scattering of X-ray spectral lines to measure AGN induced
turbulence in the hot ISM of giant elliptical galaxies (Chapter 5; Werner et al. 2009);
placed constraints on the viscosity, conductivity and the structure of intra-cluster magnetic
ﬁelds (Chapter 6 and 7; Werner et al. 2015, 2016); discovered large reservoirs of
cold gas in nearby giant elliptical galaxies (previously thought to be ‘red and dead’)
and showed that the cold gas observed in giant ellipticals is produced chieﬂy by thermally
unstable cooling from the hot phase (Chapter 9 and 8; Werner et al. 2014, 2013a);
and by pioneering the application of X-ray measurements to study the outskirts of
galaxy clusters, we discovered that the hot intergalactic medium in cluster outskirts
is clumpy and must have experienced the bulk of its metal enrichment over 10 billion
years ago (these studies were published in articles in Science and Nature by Simionescu
et al. 2011 and Werner et al. 2013b; see Chapter 10).
Our knowledge will be signiﬁcantly advanced in the coming years due to the improved
observing capabilities of the Hitomi (ASTRO-H) satellite, launched on 2016
February 17. Next to the standard X-ray observables such as the density, temperature,
and metallicity, the revolutionary capabilities of Hitomi, for the ﬁrst time, enable
direct measurements of the ICM dynamics. The X-ray calorimeters on Hitomi provide
a factor of 30 improvement in spectral resolution compared to the current state-of-theart
CCD-type detectors. This enables detailed measurements of line centroid shifts and
velocity broadening of X-ray emission lines, allowing us to tightly constrain both coherent
motions and small scale turbulence in the ICM (see Fig. 11.1). Hitomi will thus
perform the ﬁrst direct kinematic measurements of the interaction of the ICM with the
jets and the buoyant relativistic plasma.
I led the development of the observing strategy for some of the highest priority
galaxy and cluster observations for the Hitomi satellite and I am leading the analysis of
some of the ﬁrst data obtained by this mission. Breakthroughs in our understanding
of the intimate connection between the IGM, star formation and the growth of supermassive
black holes are virtually guaranteed. I plan to use deep observations of the
196 Summary and outlook
Figure 11.1: These simulated Hitomi (ASTRO-H) spectra for a 100 ks observation of
the core of the Perseus Cluster illustrate the revolutionary diagnostic capabilities of
the new calorimeter-type Soft X-ray Spectrometer (SXS). The 5 eV spectral resolution
of SXS provides a factor of 30 improvement compared to the current state of-the-art
CCD-type detectors, allowing to separate previously unresolved spectral lines. The
spectrum in the full energy band (left panel) illustrates the wealth of lines and chemical
elements that we will detect. The zoom-in to the energy band of the He-like Fe-K line
complex (right panel) shows spectra without velocity broadening and with a broadening
of σv = 330 km s−1, illustrating the power of high resolution X-ray spectroscopy in
measuring the ICM velocities.
197
Figure 11.2: A composite optical (HST), infrared (Spitzer), and X-ray (Chandra) image of
the starburst galaxy M 82. The X-ray emission is shown in blue. The collective energy
of exploding supernovae and stellar winds blows the metal rich gas out of the galaxy,
enriching and heating the surrounding intergalactic medium. Hitomi observations of
this galaxy and of NGC 253 will allow us to measure the velocity, thermal state, and
the metallicity of the out-ﬂowing hot phase and estimate its momentum.
198 Summary and outlook
nearest, brightest galaxy clusters to measure and compare the turbulent velocities on
different spatial scales. Mapping the turbulent motions of the ICM in nearby galaxy
clusters will provide critical information about the driving and dissipation scales of
turbulence, which on small scales is sensitive to viscosity. While for a low-viscosity
ICM we expect to see relatively strong small scale turbulence, a highly viscous ICM
will be less turbulent, with gas motions, especially on small scales, quickly dissipating
into heat. The high-quality Hitomi spectra will also allow us to resolve spectral lines
from different ionization levels and thus measure the detailed temperature structure of
the gas, placing important constraints on radiative cooling, multi-phasedness, and on
the dissipation and distribution of heat in the ICM. In the same time, the spectacular,
sub-arcsecond imaging capabilities of Chandra will remain unsurpassed for the next
20 years, and ultra-deep observations with this instrument will continue to provide
opportunities for breakthroughs in our understanding of the microphysics of the hot
plasma (e.g. see Zhuravleva et al. 2014a; Werner et al. 2016).
Most of the metals in the IGM/ICM are believed to have escaped the galaxies due
to supernova- and AGN-driven outﬂows (see the outﬂows from M 82 in Fig. 11.2) and
only a signiﬁcantly smaller fraction of metals escapes due to ram-pressure stripping
and galaxy-galaxy interactions. The physics and the energetics of these outﬂows, and
the mixing of the metals with the ambient IGM, are poorly understood. Hitomi will
allow us, for the ﬁrst time, to measure the velocity, thermal state, and the metallicity
of the outﬂowing hot phase and estimate its momentum. These results will provide
crucial ingredients for galaxy formation models.
Our experience with Hitomi, will be invaluable in the development of the approved,
future, L-class (budget of 1 Billion EUR) X-ray mission Athena (see Fig. 11.3). I also
plan to help in the development of a future satellite aimed at detecting and mapping
the tenuous warm-hot intergalactic medium permeating the cosmic web (DIOS being
proposed to JAXA). Together with new instruments operating in other wavelengths
(e.g. the James Webb Space Telescope, 30-meter class optical and near-infrared telescopes),
the progress in technology in the next decade will allow the extension of this
research to the faint unvirialized intergalactic gas and the high-redshift Universe.
199
Figure 11.3: Athena, the future, large (budget of over 1 Billion EUR) X-ray mission
approved by the European Space Agency. Hitomi (ASTRO-H) can be considered a
pathﬁnder for Athena. We will capitalize on our experience with Hitomi in the development
of this ambitious European mission.
200 Summary and outlook
References
Abraham, J., Abreu, P., Aglietta, M., et al. 2010, Physical Review Letters, 104, 091101
Allen, S. W., Dunn, R. J. H., Fabian, A. C., Taylor, G. B., & Reynolds, C. S. 2006, MNRAS,
372, 21
Allen, S. W., Evrard, A. E., & Mantz, A. B. 2011, ARA&A, 49, 409
Allen, S. W., Fabian, A. C., Edge, A. C., Bohringer, H., & White, D. A. 1995, MNRAS,
275, 741
Allen, S. W., Rapetti, D. A., Schmidt, R. W., et al. 2008, MNRAS, 383, 879
Allen, S. W., Taylor, G. B., Nulsen, P. E. J., et al. 2001, MNRAS, 324, 842
Andernach, H., Baker, J. R., von Kap-Herr, A., & Wielebinski, R. 1979, A&A, 74, 93
Anders, E. & Grevesse, N. 1989, Geochim. Cosmochim. Acta, 53, 197
Andreon, S. 2012, A&A, 546, A6
Annibali, F., Bressan, A., Rampazzo, R., Zeilinger, W. W., & Danese, L. 2007, A&A, 463,
455
Annibali, F., Bressan, A., Rampazzo, R., et al. 2010, A&A, 519, A40
Ar´evalo, P., Churazov, E., Zhuravleva, I., Forman, W. R., & Jones, C. 2015, ArXiv e-
prints
Arnaud, K. A. 1996, in Astronomical Society of the Paciﬁc Conference Series, Vol. 101,
Astronomical Data Analysis Software and Systems V, ed. G. H. Jacoby & J. Barnes,
17
Ascasibar, Y. & Markevitch, M. 2006, ApJ, 650, 102
Bˆırzan, L., Rafferty, D. A., McNamara, B. R., Wise, M. W., & Nulsen, P. E. J. 2004, ApJ,
607, 800
Baars, J. W. M., Genzel, R., Pauliny-Toth, I. I. K., & Witzel, A. 1977, A&A, 61, 99
Balbus, S. A. & Soker, N. 1989, ApJ, 341, 611
Baldi, A., Ettori, S., Molendi, S., et al. 2012, A&A, 537, A142
Baldi, A., Forman, W., Jones, C., et al. 2009, ApJ, 707, 1034
Baldwin, J. A., Phillips, M. M., & Terlevich, R. 1981, PASP, 93, 5
Balestra, I., Tozzi, P., Ettori, S., et al. 2007, A&A, 462, 429
Balucinska-Church, M. & McCammon, D. 1992, ApJ, 400, 699
Begelman, M. C. & Fabian, A. C. 1990, MNRAS, 244, 26P
Beiersdorfer, P., Schweikhard, L., Liebisch, P., & Brown, G. V. 2008, ApJ, 672, 726
Belsole, E., Sauvageot, J. L., B¨ohringer, H., et al. 2001, A&A, 365, L188
Bertin, E. 2006, in Astronomical Society of the Paciﬁc Conference Series, Vol. 351, Astronomical
Data Analysis Software and Systems XV, ed. C. Gabriel, C. Arviset, D. Ponz,
& S. Enrique, 112–+
202 References
Bertin, E., Mellier, Y., Radovich, M., et al. 2002, in Astronomical Society of the Paciﬁc
Conference Series, Vol. 281, Astronomical Data Analysis Software and Systems XI,
ed. D. A. Bohlender, D. Durand, & T. H. Handley, 228–+
Best, P. N., Kauffmann, G., Heckman, T. M., et al. 2005, MNRAS, 362, 25
Biller, B. A., Jones, C., Forman, W. R., Kraft, R., & Ensslin, T. 2004, ApJ, 613, 238
Biretta, J. A., Sparks, W. B., & Macchetto, F. 1999, ApJ, 520, 621
Bˆırzan, L., McNamara, B. R., Nulsen, P. E. J., Carilli, C. L., & Wise, M. W. 2008, ApJ,
686, 859
Blakeslee, J. P., Jord´an, A., Mei, S., et al. 2009, ApJ, 694, 556
Bland-Hawthorn, J. & Maloney, P. R. 1999, ApJ, 510, L33
Bock, D., Large, M. I., & Sadler, E. M. 1999, AJ, 117, 1578
Bohlin, R. C., Savage, B. D., & Drake, J. F. 1978, ApJ, 224, 132
B¨ohringer, H., Matsushita, K., Churazov, E., Finoguenov, A., & Ikebe, Y. 2004, A&A,
416, L21
B¨ohringer, H., Matsushita, K., Churazov, E., Ikebe, Y., & Chen, Y. 2002, A&A, 382, 804
B¨ohringer, H., Nulsen, P. E. J., Braun, R., & Fabian, A. C. 1995, MNRAS, 274, L67
Bonamente, M., Lieu, R., Mittaz, J. P. D., Kaastra, J. S., & Nevalainen, J. 2005, ApJ, 629,
192
Bondi, H. 1952, MNRAS, 112, 195
Bottinelli, L. & Gouguenheim, L. 1977, A&A, 60, L23
Bottinelli, L. & Gouguenheim, L. 1979, A&A, 74, 172
Braginskii, S. I. 1965, Reviews of Plasma Physics, 1, 205
Brandt, W. N. & Hasinger, G. 2005, ARA&A, 43, 827
Bregman, J. N. & Parriott, J. R. 2009, ApJ, 699, 923
Br¨uggen, M. 2003, ApJ, 592, 839
Br¨uggen, M. & Kaiser, C. R. 2002, Nature, 418, 301
Buote, D. A. 2000a, ApJ, 539, 172
Buote, D. A. 2000b, MNRAS, 311, 176
Burns, J. O., Hallman, E. J., Gantner, B., Motl, P. M., & Norman, M. L. 2008, ApJ, 675,
1125
Buson, L. M., Sadler, E. M., Zeilinger, W. W., et al. 1993, A&A, 280, 409
Canning, R. E. A., Fabian, A. C., Johnstone, R. M., et al. 2011, MNRAS, 417, 3080
Canning, R. E. A., Russell, H. R., Hatch, N. A., et al. 2012, MNRAS, 420, 2956
Canning, R. E. A., Sun, M., Sanders, J. S., et al. 2013, ArXiv e-prints
Caon, N., Macchetto, D., & Pastoriza, M. 2000, ApJS, 127, 39
Cappellaro, E., Evans, R., & Turatto, M. 1999, A&A, 351, 459
Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345, 245
Cavagnolo, K. W., Donahue, M., Voit, G. M., & Sun, M. 2008, ApJ, 683, L107
Cavagnolo, K. W., Donahue, M., Voit, G. M., & Sun, M. 2009, ApJS, 182, 12
Cen, R. & Ostriker, J. P. 1999, ApJ, 514, 1
Chandran, B. D. G., Cowley, S. C., Ivanushkina, M., & Sydora, R. 1999, ApJ, 525, 638
Chandrasekhar, S. 1961, Hydrodynamic and hydromagnetic stability (Oxford: Clarendon,
1961)
Churazov, E., Br¨uggen, M., Kaiser, C. R., B¨ohringer, H., & Forman, W. 2001, ApJ, 554,
261
203
Churazov, E., Forman, W., Jones, C., & B¨ohringer, H. 2000, A&A, 356, 788
Churazov, E., Forman, W., Jones, C., & B¨ohringer, H. 2003, ApJ, 590, 225
Churazov, E., Forman, W., Jones, C., Sunyaev, R., & B¨ohringer, H. 2004, MNRAS, 347,
29
Churazov, E., Forman, W., Vikhlinin, A., et al. 2008, MNRAS, 388, 1062
Churazov, E. & Inogamov, N. 2004, MNRAS, 350, L52
Churazov, E., Ruszkowski, M., & Schekochihin, A. 2013, ArXiv e-prints
Churazov, E., Sunyaev, R., Forman, W., & B¨ohringer, H. 2002, MNRAS, 332, 729
Churazov, E., Tremaine, S., Forman, W., et al. 2010, MNRAS, 404, 1165
Churazov, E., Vikhlinin, A., Zhuravleva, I., et al. 2012, MNRAS, 421, 1123
Combes, F., Young, L. M., & Bureau, M. 2007, MNRAS, 377, 1795
Condon, J. J., Cotton, W. D., & Broderick, J. J. 2002, AJ, 124, 675
Condon, J. J., Cotton, W. D., Greisen, E. W., et al. 1998, AJ, 115, 1693
Cowie, L. L., Fabian, A. C., & Nulsen, P. E. J. 1980, MNRAS, 191, 399
Cowie, L. L., Hu, E. M., Jenkins, E. B., & York, D. G. 1983, ApJ, 272, 29
Cowie, L. L. & McKee, C. F. 1977, ApJ, 211, 135
Cravens, T. E. 1997, Geophys. Res. Lett., 24, 105
Crawford, C. S., Allen, S. W., Ebeling, H., Edge, A. C., & Fabian, A. C. 1999, MNRAS,
306, 857
Crawford, C. S. & Fabian, A. C. 1992, MNRAS, 259, 265
Crawford, C. S., Sanders, J. S., & Fabian, A. C. 2005, MNRAS, 361, 17
Crawford, M. K., Genzel, R., Townes, C. H., & Watson, D. M. 1985, ApJ, 291, 755
Croton, D. J., Springel, V., White, S. D. M., et al. 2006, MNRAS, 365, 11
Das, P., Gerhard, O., Churazov, E., & Zhuravleva, I. 2010, MNRAS, 409, 1362
Dav´e, R., Cen, R., Ostriker, J. P., et al. 2001, ApJ, 552, 473
David, L. P., Jones, C., Forman, W., et al. 2009, ApJ, 705, 624
David, L. P., O’Sullivan, E., Jones, C., et al. 2011, ApJ, 728, 162
de Gasperin, F., Ogrean, G. A., van Weeren, R. J., et al. 2015, MNRAS, 448, 2197
De Grandi, S., Ettori, S., Longhetti, M., & Molendi, S. 2004, A&A, 419, 7
De Lucia, G. & Blaizot, J. 2007, MNRAS, 375, 2
de Plaa, J., Werner, N., Bykov, A. M., et al. 2006, A&A, 452, 397
de Plaa, J., Werner, N., Simionescu, A., et al. 2010, A&A, 523, A81
de Plaa, J., Zhuravleva, I., Werner, N., et al. 2012, A&A, 539, A34
De Young, D. S. 1978, ApJ, 223, 47
De Young, D. S. 2003, MNRAS, 343, 719
den Herder, J. W., Brinkman, A. C., Kahn, S. M., et al. 2001, A&A, 365, L7
Dennerl, K. 2002, A&A, 394, 1119
Dennerl, K. 2010, Space Sci. Rev., 157, 57
Dickey, J. M. & Lockman, F. J. 1990, ARA&A, 28, 215
Domainko, W., Mair, M., Kapferer, W., et al. 2006, A&A, 452, 795
Donahue, M., de Messi`eres, G. E., O’Connell, R. W., et al. 2011, ApJ, 732, 40
Donahue, M., Mack, J., Voit, G. M., et al. 2000, ApJ, 545, 670
Donahue, M., Stocke, J. T., & Gioia, I. M. 1992, ApJ, 385, 49
Donahue, M. & Voit, G. M. 1991, ApJ, 381, 361
Doron, R. & Behar, E. 2002, ApJ, 574, 518
204 References
Dunn, R. J. H., Allen, S. W., Taylor, G. B., et al. 2010, MNRAS
Dunn, R. J. H. & Fabian, A. C. 2006, MNRAS, 373, 959
Dunn, R. J. H. & Fabian, A. C. 2008, MNRAS, 385, 757
Dunn, R. J. H., Fabian, A. C., & Taylor, G. B. 2005, MNRAS, 364, 1343
Durret, F., Wakamatsu, K., Nagayama, T., Adami, C., & Biviano, A. 2015, A&A, 583,
A124
Eckert, D., Molendi, S., & Paltani, S. 2011, A&A, 526, A79
Edge, A. C. 2001, MNRAS, 328, 762
Edge, A. C. & Frayer, D. T. 2003, ApJ, 594, L13
Edge, A. C., Oonk, J. B. R., Mittal, R., et al. 2010, A&A, 518, L46
Edge, A. C., Stewart, G. C., Fabian, A. C., & Arnaud, K. A. 1990, MNRAS, 245, 559
Edge, A. C., Wilman, R. J., Johnstone, R. M., et al. 2002, MNRAS, 337, 49
Edwards, L. O. V., Robert, C., Moll´a, M., & McGee, S. L. 2009, MNRAS, 396, 1953
Ehlert, S., Allen, S. W., von der Linden, A., et al. 2011a, MNRAS, 411, 1641
Ehlert, S., Allen, S. W., von der Linden, A., et al. 2011b, MNRAS, 411, 1641
Ensslin, T. A. & Br¨uggen, M. 2002, MNRAS, 331, 1011
Ensslin, T. A. & Gopal-Krishna. 2001, A&A, 366, 26
Erben, T., Schirmer, M., Dietrich, J. P., et al. 2005, Astronomische Nachrichten, 326, 432
Esquivel, A., Benjamin, R. A., Lazarian, A., Cho, J., & Leitner, S. N. 2006, ApJ, 648, 1043
Ettori, S. & Fabian, A. C. 2000, MNRAS, 317, L57
Fabian, A. C., Allen, S. W., Crawford, C. S., et al. 2002, MNRAS, 332, L50
Fabian, A. C., Johnstone, R. M., & Daines, S. J. 1994, MNRAS, 271, 737
Fabian, A. C., Johnstone, R. M., Sanders, J. S., et al. 2008, Nature, 454, 968
Fabian, A. C., Mushotzky, R. F., Nulsen, P. E. J., & Peterson, J. R. 2001, MNRAS, 321,
L20
Fabian, A. C., Sanders, J. S., Allen, S. W., et al. 2003a, MNRAS, 344, L43
Fabian, A. C., Sanders, J. S., Crawford, C. S., et al. 2003b, MNRAS, 344, L48
Fabian, A. C., Sanders, J. S., Taylor, G. B., et al. 2006, MNRAS, 366, 417
Fabian, A. C., Sanders, J. S., Williams, R. J. R., et al. 2011, MNRAS, 417, 172
Fabjan, D., Borgani, S., Tornatore, L., et al. 2010, MNRAS, 401, 1670
Falcke, H., Rieke, M. J., Rieke, G. H., Simpson, C., & Wilson, A. S. 1998, ApJ, 494, L155
Farage, C. L., McGregor, P. J., & Dopita, M. A. 2012, ApJ, 747, 28
Feigelson, E. D., Wood, P. A. D., Schreier, E. J., Harris, D. E., & Reid, M. J. 1987, ApJ,
312, 101
Feldman, U. 1992, Phys. Scr, 46, 202
Feretti, L., Dallacasa, D., Govoni, F., et al. 1999, A&A, 344, 472
Ferland, G. J., Fabian, A. C., Hatch, N. A., et al. 2008, MNRAS, 386, L72
Ferland, G. J., Fabian, A. C., Hatch, N. A., et al. 2009, MNRAS, 392, 1475
Ferrarese, L., Ford, H. C., & Jaffe, W. 1996, ApJ, 470, 444
Ferrari, C., Govoni, F., Schindler, S., Bykov, A. M., & Rephaeli, Y. 2008, SSR, 134, 93
Field, G. B. 1965, ApJ, 142, 531
Ford, H. C. & Butcher, H. 1979, ApJS, 41, 147
Forman, W., Jones, C., Churazov, E., et al. 2007, ApJ, 665, 1057
Forman, W., Nulsen, P., Heinz, S., et al. 2005, ApJ, 635, 894
Foster, A. R., Ji, L., Smith, R. K., & Brickhouse, N. S. 2012, ApJ, 756, 128
205
Friedman, S. H., Heinz, S., & Churazov, E. 2012, ApJ, 746, 112
Fujimoto, R., Mitsuda, K., Mccammon, D., et al. 2007, PASJ, 59, 133
Fujita, Y., Hayashida, K., Nagai, M., et al. 2008a, PASJ, 60, 1133
Fujita, Y., Tawa, N., Hayashida, K., et al. 2008b, PASJ, 60, S343
Gaspari, M., Churazov, E., Nagai, D., Lau, E. T., & Zhuravleva, I. 2014, A&A, 569, A67
Gaspari, M., Ruszkowski, M., & Oh, S. P. 2013, MNRAS, 432, 3401
Gaspari, M., Ruszkowski, M., & Sharma, P. 2012, ApJ, 746, 94
Gastaldello, F., Buote, D. A., Temi, P., et al. 2009, ApJ, 693, 43
Gastaldello, F. & Molendi, S. 2004, ApJ, 600, 670
Gavazzi, G., Boselli, A., V´ılchez, J. M., Iglesias-Paramo, J., & Bonfanti, C. 2000, A&A,
361, 1
Ghizzardi, S., De Grandi, S., & Molendi, S. 2014, A&A, 570, A117
Ghizzardi, S., Rossetti, M., & Molendi, S. 2010, A&A, 516, A32
Giacintucci, S., O’Sullivan, E., Clarke, T. E., et al. 2012, ApJ, 755, 172
Gilfanov, M. R., Sunyaev, R. A., & Churazov, E. M. 1987, Soviet Astronomy Letters, 13,
3
Gitti, M., Brighenti, F., & McNamara, B. R. 2012, Advances in Astronomy, 2012, 950641
Gladstone, G. R., Waite, J. H., Grodent, D., et al. 2002, Nature, 415, 1000
Gonzal´ez-Riestra, R. 2004, http://xmm.vilspa.esa.es/docs/documents/CAL-TN-
0058-1-0.ps.gz
Goudfrooij, P., Hansen, L., Jorgensen, H. E., & Norgaard-Nielsen, H. U. 1994, A&AS,
105, 341
Govoni, F., Dolag, K., Murgia, M., et al. 2010, A&A, 522, A105
Govoni, F., Murgia, M., Markevitch, M., et al. 2009, A&A, 499, 371
Greisen, E. W. 2003, Information Handling in Astronomy - Historical Vistas, 285, 109
Grevesse, N. & Sauval, A. J. 1998, Space Science Reviews, 85, 161
Gunn, J. E. & Gott, III, J. R. 1972, ApJ, 176, 1
Hamer, S. L., Edge, A. C., Swinbank, A. M., et al. 2012, MNRAS, 421, 3409
Hao, C.-N., Kennicutt, R. C., Johnson, B. D., et al. 2011, ApJ, 741, 124
Hasegawa, T., Wakamatsu, K.-i., Malkan, M., et al. 2000, MNRAS, 316, 326
Hatch, N. A., Crawford, C. S., Fabian, A. C., & Johnstone, R. M. 2005, MNRAS, 358, 765
Hatch, N. A., Crawford, C. S., Johnstone, R. M., & Fabian, A. C. 2006, MNRAS, 367, 433
Heckman, T. M. 1980, A&A, 87, 152
Heckman, T. M., Baum, S. A., van Breugel, W. J. M., & McCarthy, P. 1989, ApJ, 338, 48
Heinz, S. & Churazov, E. 2005, ApJ, 634, L141
Henry, J. P., Evrard, A. E., Hoekstra, H., Babul, A., & Mahdavi, A. 2008, ArXiv e-prints
Hicks, A. K. & Mushotzky, R. 2005, ApJ, 635, L9
Hines, D. C., Eilek, J. A., & Owen, F. N. 1989, ApJ, 347, 713
Ho, I.-T., Lim, J., & Dinh-V-Trung. 2009, ApJ, 698, 1191
Humphrey, P. J., Buote, D. A., Brighenti, F., Gebhardt, K., & Mathews, W. G. 2008, ApJ,
683, 161
Humphrey, P. J., Buote, D. A., Brighenti, F., Gebhardt, K., & Mathews, W. G. 2009, ApJ,
703, 1257
Humphrey, P. J., Buote, D. A., Gastaldello, F., et al. 2006, ApJ, 646, 899
Ichinohe, Y., Werner, N., Simionescu, A., et al. 2015, MNRAS, 448, 2971
206 References
Inogamov, N. A. & Sunyaev, R. A. 2003, Astronomy Letters, 29, 791
Irwin, J. A., Athey, A. E., & Bregman, J. N. 2003, ApJ, 587, 356
Jaffe, W. & Bremer, M. N. 1997, MNRAS, 284, L1
Jaffe, W., Bremer, M. N., & Baker, K. 2005, MNRAS, 360, 748
Jaffe, W., Bremer, M. N., & van der Werf, P. P. 2001, MNRAS, 324, 443
Jaffe, W., Ford, H. C., Ferrarese, L., van den Bosch, F., & O’Connell, R. W. 1993, Nature,
364, 213
Jaffe, W. & McNamara, B. R. 1994, ApJ, 434, 110
Jarrett, T. H., Chester, T., Cutri, R., Schneider, S. E., & Huchra, J. P. 2003, AJ, 125, 525
Johnstone, R. M., Fabian, A. C., & Nulsen, P. E. J. 1987, MNRAS, 224, 75
Johnstone, R. M., Hatch, N. A., Ferland, G. J., et al. 2007, MNRAS, 382, 1246
Jones, C., Forman, W., Vikhlinin, A., et al. 2002, ApJ, 567, L115
Kaastra, J. S., Ferrigno, C., Tamura, T., et al. 2001, A&A, 365, L99
Kaastra, J. S., Mewe, R., & Nieuwenhuijzen, H. 1996, in UV and X-ray Spectroscopy of
Astrophysical and Laboratory Plasmas p.411, K. Yamashita and T. Watanabe. Tokyo
: Universal Academy Press
Kaastra, J. S., Tamura, T., Peterson, J. R., et al. 2004, A&A, 413, 415
Kaiser, C. R. 2003, MNRAS, 343, 1319
Kalberla, P. M. W., Burton, W. B., Hartmann, D., et al. 2005, A&A, 440, 775
Kaufman, M. J., Wolﬁre, M. G., Hollenbach, D. J., & Luhman, M. L. 1999, ApJ, 527, 795
Kempner, J. C., Blanton, E. L., Clarke, T. E., et al. 2004, in The Riddle of Cooling Flows
in Galaxies and Clusters of galaxies, ed. T. Reiprich, J. Kempner, & N. Soker
Kennicutt, Jr., R. C. 1998, ApJ, 498, 541
Keshet, U. 2012, ApJ, 753, 120
Keshet, U., Markevitch, M., Birnboim, Y., & Loeb, A. 2010, ApJ, 719, L74
Kewley, L. J., Groves, B., Kauffmann, G., & Heckman, T. 2006, MNRAS, 372, 961
Kim, D.-W. & Fabbiano, G. 2004, ApJ, 611, 846
Knapp, G. R., Faber, S. M., & Gallagher, J. S. 1978, AJ, 83, 11
Knapp, G. R., Guhathakurta, P., Kim, D.-W., & Jura, M. A. 1989, ApJS, 70, 329
Knapp, G. R., Turner, E. L., & Cunniffe, P. E. 1985, AJ, 90, 454
Krishna Kumar, C. & Thonnard, N. 1983, AJ, 88, 260
Kunz, M. W., Bogdanovi´c, T., Reynolds, C. S., & Stone, J. M. 2012, ApJ, 754, 122
Kunz, M. W., Schekochihin, A. A., & Stone, J. M. 2014, Physical Review Letters, 112,
205003
Lake, G. & Schommer, R. A. 1984, ApJ, 280, 107
Large, M. I., Mills, B. Y., Little, A. G., Crawford, D. F., & Sutton, J. M. 1981, MNRAS,
194, 693
Lazarian, A., Kowal, G., Vishniac, E., & de Gouveia Dal Pino, E. 2011, Planet. Space Sci.,
59, 537
Lazarian, A., Santos-Lima, R., & de Gouveia Dal Pino, E. 2010, in Astronomical Society
of the Paciﬁc Conference Series, Vol. 429, Numerical Modeling of Space Plasma
Flows, Astronum-2009, ed. N. V. Pogorelov, E. Audit, & G. P. Zank, 113
Leccardi, A. & Molendi, S. 2008, A&A, 487, 461
Lim, J., Ohyama, Y., Chi-Hung, Y., Dinh-V-Trung, & Shiang-Yu, W. 2012, ApJ, 744, 112
Lisse, C. M., Dennerl, K., Englhauser, J., et al. 1996, Science, 274, 205
207
Liu, J., Mao, S., & Wang, Q. D. 2011, MNRAS, 415, L64
Liu, J., Wang, Q. D., & Mao, S. 2012, MNRAS, 420, 3389
Liuzzo, E., Taylor, G. B., Giovannini, G., & Giroletti, M. 2009, A&A, 501, 933
Lodders, K. 2003, ApJ, 591, 1220
Macchetto, F., Pastoriza, M., Caon, N., et al. 1996, A&AS, 120, 463
Machacek, M. E., O’Sullivan, E., Randall, S. W., Jones, C., & Forman, W. R. 2010, ApJ,
711, 1316
Madau, P., Ferguson, H. C., Dickinson, M. E., et al. 1996, MNRAS, 283, 1388
Magorrian, J., Tremaine, S., Richstone, D., et al. 1998, AJ, 115, 2285
Maia, M. A. G., da Costa, L. N., Willmer, C., Pellegrini, P. S., & Rite, C. 1987, AJ, 93, 546
Malhotra, S., Helou, G., Stacey, G., et al. 1997, ApJ, 491, L27
Malhotra, S., Kaufman, M. J., Hollenbach, D., et al. 2001, ApJ, 561, 766
Malyshkin, L. 2001, ApJ, 554, 561
Mannucci, F., Della Valle, M., & Panagia, N. 2006, MNRAS, 370, 773
Mantz, A., Allen, S. W., Ebeling, H., & Rapetti, D. 2008, MNRAS, 387, 1179
Maoz, D., Mannucci, F., & Brandt, T. D. 2012, MNRAS, 426, 3282
Markevitch, M., Mazzotta, P., Vikhlinin, A., et al. 2003, ApJ, 586, L19
Markevitch, M., Ponman, T. J., Nulsen, P. E. J., et al. 2000, ApJ, 541, 542
Markevitch, M. & Vikhlinin, A. 2007, Phys. Rep., 443, 1
Markevitch, M., Vikhlinin, A., & Mazzotta, P. 2001, ApJ, 562, L153
Mathews, W. G. 1990, ApJ, 354, 468
Mathews, W. G. & Brighenti, F. 2003, ARA&A, 41, 191
Mathews, W. G., Temi, P., Brighenti, F., & Amblard, A. 2013, ApJ, 768, 28
Matsushita, K. 2011, A&A, 527, A134
Matsushita, K., Belsole, E., Finoguenov, A., & B¨ohringer, H. 2002, A&A, 386, 77
Matsushita, K., Sakuma, E., Sasaki, T., Sato, K., & Simionescu, A. 2013, ApJ, 764, 147
Matsuzawa, H., Matsuoka, M., Ikebe, Y., Mihara, T., & Yamashita, K. 1996, PASJ, 48,
565
Mauch, T., Murphy, T., Buttery, H. J., et al. 2003, MNRAS, 342, 1117
Maughan, B. J., Jones, C., Forman, W., & Van Speybroeck, L. 2008, ApJS, 174, 117
Mazzotta, P. & Giacintucci, S. 2008, ApJ, 675, L9
McCourt, M., Sharma, P., Quataert, E., & Parrish, I. J. 2012, MNRAS, 419, 3319
McDonald, M. & Veilleux, S. 2009, ApJ, 703, L172
McDonald, M., Veilleux, S., Rupke, D. S. N., & Mushotzky, R. 2010, ApJ, 721, 1262
McDonald, M., Wei, L. H., & Veilleux, S. 2012, ApJ, 755, L24
McDonald, M., Werner, N., Oonk, J. B. R., & Veilleux, S. 2015, ApJ, 804, 16
McMullin, J. P., Waters, B., Schiebel, D., Young, W., & Golap, K. 2007, in Astronomical
Society of the Paciﬁc Conference Series, Vol. 376, Astronomical Data Analysis
Software and Systems XVI, ed. R. A. Shaw, F. Hill, & D. J. Bell, 127
McNamara, B. R. & Nulsen, P. E. J. 2007, ARA&A, 45, 117
McNamara, B. R., Nulsen, P. E. J., Wise, M. W., et al. 2005, Nature, 433, 45
McNamara, B. R., Russell, H. R., Nulsen, P. E. J., et al. 2013, ArXiv e-prints
Merloni, A., Heinz, S., & di Matteo, T. 2003, MNRAS, 345, 1057
Million, E. T. & Allen, S. W. 2009, MNRAS, 399, 1307
Million, E. T., Allen, S. W., Werner, N., & Taylor, G. B. 2010a, MNRAS, 405, 1624
208 References
Million, E. T., Allen, S. W., Werner, N., & Taylor, G. B. 2010b, MNRAS, 405, 1624
Million, E. T., Allen, S. W., Werner, N., & Taylor, G. B. 2010c, MNRAS, 405, 1624
Million, E. T., Werner, N., Simionescu, A., et al. 2010d, MNRAS, 407, 2046
Mittal, R., O’Dea, C. P., Ferland, G., et al. 2011, MNRAS, 418, 2386
Mittal, R., Oonk, J. B. R., Ferland, G. J., et al. 2012, ArXiv e-prints
Molendi, S. 2002, ApJ, 580, 815
Morganti, R., Fogasy, J., Paragi, Z., Oosterloo, T., & Orienti, M. 2013, ArXiv e-prints
Moustakas, J., Kennicutt, Jr., R. C., Tremonti, C. A., et al. 2010, ApJS, 190, 233
Murgia, M., Eckert, D., Govoni, F., et al. 2010, A&A, 514, A76
Murgia, M., Govoni, F., Markevitch, M., et al. 2009, A&A, 499, 679
Nagai, D., Vikhlinin, A., & Kravtsov, A. V. 2007, ApJ, 655, 98
Narayan, R. & Medvedev, M. V. 2001, ApJ, 562, L129
Norman, C. A., Melrose, D. B., & Achterberg, A. 1995, ApJ, 454, 60
Nulsen, P. E. J. 1986, MNRAS, 221, 377
Nulsen, P. E. J. & McNamara, B. R. 2013, Astronomische Nachrichten, 334, 386
Nulsen, P. E. J., McNamara, B. R., Wise, M. W., & David, L. P. 2005, ApJ, 628, 629
O’Dea, C. P., Baum, S. A., Privon, G., et al. 2008, ApJ, 681, 1035
Ogrean, G., van Weeren, R., Jones, C., et al. 2015, arXiv: 1505.05560
Ogrean, G. A., Hatch, N. A., Simionescu, A., et al. 2010, MNRAS, 406, 354
Oonk, J. B. R. 2011, PhD thesis, Ph. D. thesis, University of Leiden (2011). Advisor:
W. Jaffe. ISBN: 9789461910318. pp. 226
Oonk, J. B. R., Jaffe, W., Bremer, M. N., & van Weeren, R. J. 2010, MNRAS, 405, 898
Osterbrock, D. E. & Ferland, G. J. 2006, Astrophysics of gaseous nebulae and active
galactic nuclei
O’Sullivan, E., David, L. P., & Vrtilek, J. M. 2013, ArXiv e-prints
O’Sullivan, E., Forbes, D. A., & Ponman, T. J. 2001, MNRAS, 328, 461
O’Sullivan, E., Ponman, T. J., & Collins, R. S. 2003, MNRAS, 340, 1375
O’Sullivan, E., Worrall, D. M., Birkinshaw, M., et al. 2011, MNRAS, 1180
Owen, F. N., Eilek, J. A., & Kassim, N. E. 2000, ApJ, 543, 611
Owen, F. N., Eilek, J. A., & Keel, W. C. 1990, ApJ, 362, 449
Owers, M. S., Nulsen, P. E. J., Couch, W. J., & Markevitch, M. 2009, ApJ, 704, 1349
Panagoulia, E. K., Fabian, A. C., & Sanders, J. S. 2014, MNRAS, 438, 2341
Panuzzo, P., Rampazzo, R., Bressan, A., et al. 2011, A&A, 528, A10
Parriott, J. R. & Bregman, J. N. 2008, ApJ, 681, 1215
Parrish, I. J. & Quataert, E. 2008, ApJ, 677, L9
Parrish, I. J., Quataert, E., & Sharma, P. 2009, ApJ, 703, 96
Parrish, I. J., Quataert, E., & Sharma, P. 2010, ApJ, 712, L194
Peterson, J. R. & Fabian, A. C. 2006, Phys. Rep., 427, 1
Pilbratt, G. L., Riedinger, J. R., Passvogel, T., et al. 2010, A&A, 518, L1
Poglitsch, A., Waelkens, C., Geis, N., et al. 2010, A&A, 518, L2
Pollack, L. K., Taylor, G. B., & Allen, S. W. 2005, MNRAS, 359, 1229
Poole, G. B., Babul, A., McCarthy, I. G., Sanderson, A. J. R., & Fardal, M. A. 2008,
MNRAS, 391, 1163
Pratt, G. W., Arnaud, M., Piffaretti, R., et al. 2010, A&A, 511, A85+
Pratt, G. W., Arnaud, M., & Pointecouteau, E. 2006, A&A, 446, 429
209
Quataert, E. 2008, ApJ, 673, 758
Quillen, A. C., Zufelt, N., Park, J., et al. 2008, ApJS, 176, 39
Rafferty, D. A., McNamara, B. R., & Nulsen, P. E. J. 2008, ApJ, 687, 899
Rafferty, D. A., McNamara, B. R., Nulsen, P. E. J., & Wise, M. W. 2006, ApJ, 652, 216
Randall, S. W., Forman, W. R., Giacintucci, S., et al. 2011a, ApJ, 726, 86
Randall, S. W., Forman, W. R., Giacintucci, S., et al. 2011b, ApJ, 726, 86
Rawle, T. D., Edge, A. C., Egami, E., et al. 2012, ApJ, 747, 29
Rebusco, P., Churazov, E., Sunyaev, R., B¨ohringer, H., & Forman, W. 2008, MNRAS,
384, 1511
Reiprich, T. H. & B¨ohringer, H. 2002, ApJ, 567, 716
Reiss, I. & Keshet, U. 2014, Physical Review Letters, 113, 071302
Revaz, Y., Combes, F., & Salom´e, P. 2008, A&A, 477, L33
Rickes, M. G., Pastoriza, M. G., & Bonatto, C. 2004, A&A, 419, 449
Robertson, I. P., Cravens, T. E., Snowden, S., & Linde, T. 2001, Space Sci. Rev., 97, 401
Roediger, E., Br¨uggen, M., Simionescu, A., et al. 2011, MNRAS, 413, 2057
Roediger, E., Kraft, R. P., Forman, W. R., Nulsen, P. E. J., & Churazov, E. 2013, ApJ, 764,
60
Roediger, E., Kraft, R. P., Machacek, M. E., et al. 2012, ApJ, 754, 147
Roediger, E., Kraft, R. P., Nulsen, P. E. J., et al. 2015, ApJ, 806, 104
Roediger, E. & Zuhone, J. A. 2012, MNRAS, 419, 1338
Rossetti, M., Eckert, D., De Grandi, S., et al. 2013, A&A, 556, A44
Russell, H. R., McNamara, B. R., Edge, A. C., et al. 2013a, MNRAS, 432, 530
Russell, H. R., McNamara, B. R., Edge, A. C., et al. 2013b, ArXiv e-prints
Russell, H. R., McNamara, B. R., Sanders, J. S., et al. 2012, MNRAS, 423, 236
Ruszkowski, M., Br¨uggen, M., & Begelman, M. C. 2004a, ApJ, 611, 158
Ruszkowski, M., Br¨uggen, M., & Begelman, M. C. 2004b, ApJ, 615, 675
Sakelliou, I., Peterson, J. R., Tamura, T., et al. 2002, A&A, 391, 903
Salom´e, P. & Combes, F. 2003, A&A, 412, 657
Salom´e, P. & Combes, F. 2008, A&A, 489, 101
Salom´e, P., Combes, F., Edge, A. C., et al. 2006, A&A, 454, 437
Salom´e, P., Combes, F., Revaz, Y., et al. 2011, A&A, 531, A85
Salom´e, P., Revaz, Y., Combes, F., et al. 2008, A&A, 483, 793
Sanders, J. S. 2006, MNRAS, 371, 829
Sanders, J. S. & Fabian, A. C. 2002, MNRAS, 331, 273
Sanders, J. S. & Fabian, A. C. 2006, MNRAS, 371, 1483
Sanders, J. S. & Fabian, A. C. 2007, MNRAS, 381, 1381
Sanders, J. S. & Fabian, A. C. 2008, MNRAS, 390, L93
Sanders, J. S. & Fabian, A. C. 2011, MNRAS, 412, L35
Sanders, J. S. & Fabian, A. C. 2012, MNRAS, 421, 726
Sanders, J. S. & Fabian, A. C. 2013, MNRAS, 429, 2727
Sanders, J. S., Fabian, A. C., Allen, S. W., et al. 2008, MNRAS, 385, 1186
Sanders, J. S., Fabian, A. C., Allen, S. W., & Schmidt, R. W. 2004, MNRAS, 349, 952
Sanders, J. S., Fabian, A. C., Frank, K. A., Peterson, J. R., & Russell, H. R. 2010a, MNRAS,
402, 127
Sanders, J. S., Fabian, A. C., Smith, R. K., & Peterson, J. R. 2010b, MNRAS, 402, L11
210 References
Sanders, J. S., Fabian, A. C., & Taylor, G. B. 2009, MNRAS, 396, 1449
Sanders, J. S., Fabian, A. C., Taylor, G. B., et al. 2016, [arXiv: 1601.01489]
Scannapieco, E. & Br¨uggen, M. 2008, ApJ, 686, 927
Schekochihin, A. A., Cowley, S. C., Kulsrud, R. M., Hammett, G. W., & Sharma, P. 2005,
ApJ, 629, 139
Schekochihin, A. A., Cowley, S. C., Kulsrud, R. M., Rosin, M. S., & Heinemann, T. 2008,
Physical Review Letters, 100, 081301
Schuecker, P., Finoguenov, A., Miniati, F., B¨ohringer, H., & Briel, U. G. 2004, A&A, 426,
387
Schure, K. M., Kosenko, D., Kaastra, J. S., Keppens, R., & Vink, J. 2009, A&A, 508, 751
Serra, P. & Oosterloo, T. A. 2010, MNRAS, 401, L29
Sharma, P., McCourt, M., Quataert, E., & Parrish, I. J. 2012a, MNRAS, 420, 3174
Sharma, P., McCourt, M., Quataert, E., & Parrish, I. J. 2012b, MNRAS, 420, 3174
Sharma, P., Parrish, I. J., & Quataert, E. 2010, ApJ, 720, 652
Shepherd, M. C., Pearson, T. J., & Taylor, G. B. 1995, in Bulletin of the American Astronomical
Society, Vol. 27, Bulletin of the American Astronomical Society, ed. B. J. Butler
& D. O. Muhleman, 903–+
Shurkin, K., Dunn, R. J. H., Gentile, G., Taylor, G. B., & Allen, S. W. 2008, MNRAS, 383,
923
Sijacki, D., Springel, V., Di Matteo, T., & Hernquist, L. 2007, MNRAS, 380, 877
Silk, J. & Rees, M. J. 1998, A&A, 331, L1
Simionescu, A., Allen, S. W., Mantz, A., et al. 2011, Science, 331, 1576
Simionescu, A., B¨ohringer, H., Br¨uggen, M., & Finoguenov, A. 2007, A&A, 465, 749
Simionescu, A., Roediger, E., Nulsen, P. E. J., et al. 2009a, A&A, 495, 721
Simionescu, A., Werner, N., B¨ohringer, H., et al. 2009b, A&A, 493, 409
Simionescu, A., Werner, N., Finoguenov, A., B¨ohringer, H., & Br¨uggen, M. 2008, A&A,
482, 97
Simionescu, A., Werner, N., Forman, W. R., et al. 2010, MNRAS, 405, 91
Simionescu, A., Werner, N., Urban, O., et al. 2013, ApJ, 775, 4
Simionescu, A., Werner, N., Urban, O., et al. 2012, ApJ, 757, 182
Simionescu, A., Werner, N., Urban, O., et al. 2015, ApJ, 811, L25
Sirothia, S. K. 2009, MNRAS, 398, 853
Smith, M. W. L., Gomez, H. L., Eales, S. A., et al. 2012, ApJ, 748, 123
Smith, R. K., Brickhouse, N. S., Liedahl, D. A., & Raymond, J. C. 2001, ApJ, 556, L91
Soker, N., Blanton, E. L., & Sarazin, C. L. 2004, A&A, 422, 445
Sparks, W. B., Donahue, M., Jord´an, A., Ferrarese, L., & Cˆot´e, P. 2004, ApJ, 607, 294
Sparks, W. B., Ford, H. C., & Kinney, A. L. 1993, ApJ, 413, 531
Sparks, W. B., Pringle, J. E., Carswell, R. F., et al. 2012, ApJ, 750, L5
Sparks, W. B., Pringle, J. E., Donahue, M., et al. 2009, ApJ, 704, L20
Spitzer, L. 1962, Physics of Fully Ionized Gases (New York: Interscience (2nd edition))
Stacey, G. J., Geis, N., Genzel, R., et al. 1991, ApJ, 373, 423
Stone, J. M., Gardiner, T. A., Teuben, P., Hawley, J. F., & Simon, J. B. 2008, ApJS, 178,
137
Sun, M. 2009, ApJ, 704, 1586
Sun, M., Donahue, M., & Voit, G. M. 2007, ApJ, 671, 190
211
Sun, M., Voit, G. M., Donahue, M., et al. 2009, ApJ, 693, 1142
Sunyaev, R. A., Norman, M. L., & Bryan, G. L. 2003, Astronomy Letters, 29, 783
Sutherland, R. S. & Dopita, M. A. 1993, ApJS, 88, 253
Tamura, T., Bleeker, J. A. M., Kaastra, J. S., Ferrigno, C., & Molendi, S. 2001, A&A, 379,
107
Tamura, T., Kaastra, J. S., Makishima, K., & Takahashi, I. 2003, A&A, 399, 497
Taylor, G. B. 1996, ApJ, 470, 394
Taylor, G. B., Barton, E. J., & Ge, J. 1994, AJ, 107, 1942
Taylor, G. B., Fabian, A. C., Gentile, G., et al. 2007, MNRAS, 382, 67
Taylor, G. B., Govoni, F., Allen, S. W., & Fabian, A. C. 2001, MNRAS, 326, 2
Temi, P., Brighenti, F., & Mathews, W. G. 2007a, ApJ, 660, 1215
Temi, P., Brighenti, F., & Mathews, W. G. 2007b, ApJ, 666, 222
Tittley, E. R. & Henriksen, M. 2005, ApJ, 618, 227
Tonry, J. L., Dressler, A., Blakeslee, J. P., et al. 2001, ApJ, 546, 681
Tremblay, G. R., Chiaberge, M., Sparks, W. B., et al. 2009, ApJS, 183, 278
Turk, M. J., Smith, B. D., Oishi, J. S., et al. 2011, ApJS, 192, 9
Urban, O., Simionescu, A., Werner, N., et al. 2014, MNRAS, 437, 3939
Urban, O., Werner, N., Simionescu, A., Allen, S. W., & B¨ohringer, H. 2011, MNRAS,
414, 2101
van Weeren, R. J., R¨ottgering, H. J. A., Br¨uggen, M., & Cohen, A. 2009, A&A, 508, 75
Vantyghem, A. N., McNamara, B. R., Russell, H. R., et al. 2014, MNRAS, 442, 3192
Vikhlinin, A., Kravtsov, A. V., Burenin, R. A., et al. 2008, ArXiv e-prints
Vikhlinin, A., Markevitch, M., & Murray, S. S. 2001a, ApJ, 551, 160
Vikhlinin, A., Markevitch, M., & Murray, S. S. 2001b, ApJ, 549, L47
Vikhlinin, A., Markevitch, M., Murray, S. S., et al. 2005, ApJ, 628, 655
Vikhlinin, A. A. & Markevitch, M. L. 2002, Astronomy Letters, 28, 495
Voit, G. M., Cavagnolo, K. W., Donahue, M., et al. 2008, ApJ, 681, L5
Voit, G. M. & Donahue, M. 2011, ApJ, 738, L24
Voit, G. M., Kay, S. T., & Bryan, G. L. 2005, MNRAS, 364, 909
Wagner, A. Y. & Bicknell, G. V. 2011, ApJ, 728, 29
Wagner, A. Y., Bicknell, G. V., & Umemura, M. 2012, ApJ, 757, 136
Wakamatsu, K., Malkan, M. A., Nishida, M. T., et al. 2005, in Astronomical Society of
the Paciﬁc Conference Series, Vol. 329, Nearby Large-Scale Structures and the Zone
of Avoidance, ed. A. P. Fairall & P. A. Woudt, 189
Walker, S. A., Fabian, A. C., & Kosec, P. 2014, MNRAS, 445, 3444
Wargelin, B. J., Markevitch, M., Juda, M., et al. 2004, ApJ, 607, 596
Werner, N., Allen, S. W., & Simionescu, A. 2012, MNRAS, 425, 2731
Werner, N., B¨ohringer, H., Kaastra, J. S., et al. 2006, A&A, 459, 353
Werner, N., Durret, F., Ohashi, T., Schindler, S., & Wiersma, R. P. C. 2008,
Space Sci. Rev., 134, 337
Werner, N., Kaastra, J. S., Takei, Y., et al. 2007, A&A, 468, 849
Werner, N., Oonk, J. B. R., Canning, R. E. A., et al. 2013, ApJ, 767, 153
Werner, N., Simionescu, A., Million, E. T., et al. 2010, MNRAS, 407, 2063
Werner, N., Sun, M., Bagchi, J., et al. 2011, MNRAS, 415, 3369
Werner, N., Zhuravleva, I., Churazov, E., et al. 2009, MNRAS, 398, 23
212 References
Werner, N., ZuHone, J. A., Zhuravleva, I., et al. 2016, MNRAS, 455, 846
Xiang, F., Churazov, E., Dolag, K., Springel, V., & Vikhlinin, A. 2007, MNRAS, 379, 1325
Xu, H., Kahn, S. M., Peterson, J. R., et al. 2002, ApJ, 579, 600
Young, A. J., Wilson, A. S., & Mundell, C. G. 2002, ApJ, 579, 560
Zhuravleva, I., Churazov, E., Arevalo, P., et al. 2015, arXiv: 1501.07271
Zhuravleva, I., Churazov, E., Arevalo, P., et al. 2016, ArXiv e-prints
Zhuravleva, I., Churazov, E., Schekochihin, A. A., et al. 2014a, Nature, 515, 85
Zhuravleva, I., Churazov, E. M., Schekochihin, A. A., et al. 2014b, ApJ, 788, L13
ZuHone, J. A., Kunz, M. W., Markevitch, M., Stone, J. M., & Bifﬁ, V. 2015, ApJ, 798, 90
ZuHone, J. A., Markevitch, M., & Johnson, R. E. 2010, ApJ, 717, 908
ZuHone, J. A., Markevitch, M., & Lee, D. 2011, ApJ, 743, 16
ZuHone, J. A., Markevitch, M., Ruszkowski, M., & Lee, D. 2013, ApJ, 762, 69