Cluster outskirts and the missing baryons

The XMM Cluster Outskirts Project
Dominique Eckert
Department of Astronomy, University of Geneva
S. Ettori, S. Molendi, E. Pointecouteau, C. Tchernin, F. Vazza, S. De Grandi, S.
Paltani, M. Roncarelli, G. Hurier, F. Gastaldello, M. Rossetti, ...
May 11, 2016
D. Eckert
X-COP
The outskirts of galaxy clusters
Why pushing toward the outskirts? (R > R500 )
Contain ∼ 90% of the
volume and ∼ 50% of the
mass!
Understand the build-up of
galaxy clusters
Search for the filaments of
the cosmic web
Estimate the global baryon
budget
D. Eckert
et white
al. 2006
Fig. 1 Simulated galaxyRoncarelli
cluster. The
circles ind
outwards, respectively (adapted from Roncarelli et al
the soft (0.5–2) keV band. The color scale spans 16 o
to highlight cluster outskirts. Right: Temperature ma
11 keV (red).
X-COP
Planck Sunyaev-Zeldovich measurements
Recently: Planck
measures the SZ effect
beyond the virial radius
Combined with X-ray
data, we can reconstruct:
kT =
PSZ
,
nX −ray
−5/3
K = PSZ nX −ray
Assuming hydrostatic
equilibrium we can also
reconstruct mass profiles:
Planck Collaboration V 2012
dP
GM(< r )
= −ρ
dr
r2
D. Eckert
X-COP
which are the largest ever flown on an X-ray satellite, for a combined effective area of 3,000 cm at 1 keV
and an effective area of 13 arcsec HEW. In AO-13 I was awarded a VLP (ID: 074441) for a total observing
The
Cluster
Outskirts
Large
Programme
timeXMM
of 1207 ks (335
hours) on this
major observatory.Very
This is the
largest program
awarded this year. This
VLP follows a pilot study based on two clusters (282 ks, ID: 069444 and 072524). In the pilot study (A2142
(X-COP)
and A780) we demonstrated that XMM-Newton is capable of detecting diffuse X-ray emission out to the
virial radius provided that the right observing strategy is used. In total, this project will benefit from a total
allotted time of nearly 1.5 Ms on XMM-Newton. This demonstrates that the science developed in X-COP was
XMM
AO-13
VLP,
1.5 Ms:
Construct
a sample of 13 clusters
highly
prioritized
by the
varioustotal
XMM-Newton
selection
panels.
In
X-COP
a detailed
X-ray mapping
of the entire
volume of and
13 clusters
in the redshift
attotal,
0.04
< zwill<provide
0.1 with
high-S/N
Planck
detection
XMM
range 0.04-0.1 at unprecedented depth. The list of clusters is provided in the Table below.
mapping of the entire azimuth
Cluster
Redshift
Mass [1014 M]
Planck S/N
A2319
0.0557
5.83
30.8
A3266**
0.0589
4.56
27.0
A2142*
0.090
8.15
21.3
A2255
0.0809
3.74
19.4
A2029
0.0766
7.27
19.3
A3158
0.059
3.65
17.2
A85
0.0555
5.32
16.9
A1795
0.0622
5.53
15.0
A644
0.0704
3.88
13.9
RXC J1825
0.065
2.62
13.4
A1644
0.0473
2.93
13.2
ZwCl 1215
0.0766
3.59
12.8
A780*
0.0538
1.89
-
Clusters identified by * were part of the pilot program. A similar program for A3266** is already publicly
D. Eckert
X-COP
available.
Mapping clusters out to Rvir with XMM and Planck
Abell 2142 (z = 0.09): pilot system
10-12
2
Energy flux [erg/(cm sarcmin )]
27.6
27.5
10-13
Declination
2
27.4
10-14
27.3
10-15
27.2
27.1
10-16
27.0
10-17
26.9
500
1000
1500
2000
2500
3000
R [kpc]
26.8
240.1 240.0 239.9 239.8 239.7 239.6 239.5 239.4 239.3 239.2 239.1
Right ascension
With the appropriate bkg modeling XMM can trace efficiently the
ICM out to Rvir
D. Eckert
X-COP
Calibration using deep fields
Mean radial profiles for 22 blank fields (total 1.3 Ms)
Eckert et al. subm.
Quiescent soft protons must be taken into account; with new
calibration we reach a precision of 5% on background subtraction
D. Eckert
X-COP
Mapping clusters out to Rvir with XMM and Planck
Y parameter
Abell 2142 (z = 0.09): MILCA component separation (Hurier et al.
2013)
10
−5
10
−6
χ
10−7
2.5
2
1.5
1
0.5
0
− 0.5
−1
− 1.5
−2
3
102
102
10
R [kpc]
10
Planck nicely detects several individual clusters out to Rvir
D. Eckert
X-COP
3
The “clumping bias”
Properties of
The accretion flow on galaxy
clusters is clumpy and
asymmetric
relaxed
post merger
relaxed
post merger
Vazza, DE et al. 2013
1.00e-18
D. Eckert
1.48e-15
7.41e-15
Figure 2. Top panels: X-ray flux in the [0.5-2] keV (in [erg/(s · cm2 )]) of
X-COPE3B-merging). Bottom panels: X-ray flux of clumps identified by our proced
The “clumping bias”
Properties of
The accretion flow on galaxy
clusters is clumpy and
asymmetric
X-ray signal biased towards
high-density regions:
C2 =
relaxed
post merger
relaxed
post merger
hρ 2 i
>1
hρi2
The gas density measured from X-ray
observations is biased high in the presence of inhomogeneities
Vazza, DE et al. 2013
1.00e-18
D. Eckert
1.48e-15
7.41e-15
Figure 2. Top panels: X-ray flux in the [0.5-2] keV (in [erg/(s · cm2 )]) of
X-COPE3B-merging). Bottom panels: X-ray flux of clumps identified by our proced
A2142 clumping factor
By comparing mean and median SB we can recover the gas
clumping factor
1.6
1.5
3
ne [1/cm ]
1.4
10− 3
C
1.3
1.2
1.1
10− 4
1
0.9
R [kpc]
10
3
R [kpc]
10
3
Tchernin, DE et al. subm.
We observe significant clumping beyond R500
D. Eckert
X-COP
A2142 entropy profile
K/K500
Graph
1
10
3
R [kpc]
Tchernin, DE et al. subm.
The entropy flattens beyond R500 when clumps are not excised...
D. Eckert
X-COP
K/K500
A2142 entropy profile
1
10
3
R [kpc]
Tchernin, DE et al. subm.
The entropy flattens beyond R500 when clumps are not excised...
but not when clumping is taken into account!
D. Eckert
X-COP
A2142 mass profile
Mtot/M
1015
1014
10
3
R [kpc]
Tchernin, DE et al. in prep.
Hydrostatic profiles consistent with weak lensing, galaxy dynamics;
fgas converges to the cosmic value
D. Eckert
X-COP
Accreting substructures in A2142
27.6
27.5
Declination
27.4
27.3
27.2
27.1
27.0
26.9
26.8
240.1 240.0 239.9 239.8 239.7 239.6 239.5 239.4 239.3 239.2 239.1
Right ascension
Eckert et al. 2014
D. Eckert
X-COP
Accreting substructures in A2142
Eckert et al. 2014
XMM imaging in the soft band is very efficient at detecting
accreting substructures
D. Eckert
X-COP
Ram-pressure stripping and thermal conduction
This by far the largest stripped
structure seen so far, projected
length 800 kpc
Infall velocity v ∼ 1, 200 km s−1
⇒ the feature has been surviving
in the cluster environment for at
least 600 Myr
Thermal conduction in the ICM is
inhibited by a factor & 400
o
th
sm
b
c
th
fip
ra
f
la
(tf
tu
th
pn
fu
d
in
m
eI
bA
eR
cto
is
T
sm
le
th
k
2s
fu
g1
N
(
bT
dn
n
a
3s
s
W
o
tw
pth
s
IC
b
is
c
th
to
fi
ta
pr
Fig. 3. Mid-plane cuts of δρ/ρ for the models with M ∼ 0.25. From
−2
& Churazov
top to bottom: f = 0, 10−3 , 10Gaspari
, 10−1 (the latter
very similar to 2013
f =1
run). The color coding is blue → white → red: -40% → 0% → 40%.
(
page 8 of 17
D. Eckert
X-COP
tu
The same at high resolution...
D. Eckert
X-COP
Hydra A: A bent ram-pressure stripped tail
D. Eckert et al.: A textbook example of ram-pressure stripping in the Hydra A/A780 cluster
Another galaxy group 1.1 Mpc South of the core of Hydra A
is roughly a factor of 6 larger than
e currently associated with either
suming that all the excess emission
A SE region, is associated with gas
ups potential well we come to a conr the total gas mass of the group of
still low by a factor of 2.5 with reWhile gas mass estimates for groups
than those for clusters, this results
our group may have already lost a
as prior to the stripping event we
. Irrespective of what the starting
might have been, the current data is
orm us that a significant fraction of
ed from the group. Assuming that
rest the tip be made of gas that is
und to the group, we estimate the
De Grandi, DE et al. 2016
een 1 and 5×1011 M⊙ , i.e. half and Fig. 16. From Roediger et al. (2015a). Snapshot of a highas mass, depending on whether the resolution simulation of the hot atmosphere of an elliptical
Ideal was
laboratories
ram-pressure
stripping
and
ICM
physics
galaxy
falling into a galaxy
cluster.
The
bottom
of the poa A SE region
donated byto
thestudy
tential well is marked with the black cross. The arrows show
f the group falling onto A2142 the velocity field of the fluid. Several features can be seen:
have performed an estimate of the the cold front at the peak with the KH rolls on the side, the
D. Eckert
X-COP
Abell 2744 (z = 0.306): the Pandora cluster
Abell 2744 is one of the HST “Frontier Fields” clusters
Jauzac et al. 2015
Jauzac et al. 2015: We detected ∼ 50 lensed galaxies in this
cluster, corresponding mass model known at 1% precision
D. Eckert
X-COP
XMM-Newton observation of Abell 2744
We discovered 5 regions of extended X-ray emission radially
connected to the cluster
-30.10
9.99e-14
4.38e-14
-30.20
N
1.98e-14
-30.40
NW
9.74e-15
-30.50
5.50e-15
SE
SW
3.71e-15
-30.60
S
2.95e-15
-30.70
Declination
-30.30
E
2.63e-15
3.90
3.80
3.70
3.60
3.50
3.40
3.30
Right ascension
Eckert et al. 2015
D. Eckert
X-COP
Hot gas filaments in Abell 2744
Significant extended emission detected in the direction of the
filaments out to ∼ 4 Mpc
Cluster
SX [counts s-1 arcmin-2]
10-2
South
North-West
East
-3
10
10-4
10-5
10-6
2
4
6
8
10
12
14
Radius [arcmin]
Eckert et al. 2015
D. Eckert
X-COP
Hot gas filaments in Abell 2744
-30.40
-30.50
-30.60
-30.70
Declination
-30.30
-30.20
-30.10
The filamentary structures correspond with overdensities of cluster
galaxies (spectroscopically confirmed)...
3.90
3.80
3.70
3.60
3.50
3.40
3.30
Right ascension
Eckert et al. 2015
D. Eckert
X-COP
Hot gas filaments in Abell 2744
-30.35
-30.40
-30.45
-30.50
-30.55
-30.60
Declination
-30.30
-30.25
-30.20
-30.15
The filamentary structures correspond with overdensities of cluster
galaxies ... and DM (CFHT weak lensing)!
3.80
3.70
3.60
3.50
3.40
3.30
Right ascension
Eckert et al. 2015
D. Eckert
X-COP
Is it the WHIM?
480
DAVEŠ ET AL.
contribution t
s~1 sr~1 keV
WFN00). Dire
background f
yields similar c
principle, plac
warm-hot tem
These limit
using similar m
papers argue
mation of viria
greatly exceed
niÐcant nong
baryon, is requ
virialized obje
In this section
WHIM gas, a
ferent physica
models assum
Davé et al. 2001
substantially
FIG. 6.È Contours in temperature and density for simulation D1 at
one of densit
7
z
\
0,
enclosing
10%,
50%,
and
90%
of
the
baryons
in
the
range
shown.
We are observing
gas with overdensity ∼ 200 and T ∼ 10 K: calculations o
Density and temperature are correlated in the WHIM regime. Thick line
scaling of o/o6 \for
T /104.7
the warm-hot
temperature
consistentindicates
with apredictions
thein high-T
part
of therange.
WHIM. that all soft Xb
our simulatio
D. Eckert
X-COP
Toward the next decade: The sub-halo mass function
Key question: How active are structure formation processes in the
current Universe?
-53.00
31.00
-7.00
-53.20
30.80
-7.20
-7.40
-7.60
Declination
30.60
Declination
Declination
-53.40
-53.60
30.40
-53.80
30.20
-7.80
-54.00
30.00
-8.00
125.00
A644
124.80
124.60
124.40
124.20
Right ascension
124.00
123.80
A3158
-54.20
RXC J1825
29.80
56.50
56.00
55.50
55.00
Right ascension
277.00
276.80
276.60
276.40
XMM is very efficient at detecting low-mass structures around
nearby clusters
D. Eckert
X-COP
276.20
Right ascension
276.00
275.80
Toward the next decade: The sub-halo mass function
Key question: How active are structure formation processes in the
current Universe?
-53.00
31.00
-7.00
-53.20
30.80
-7.20
-7.40
-7.60
Declination
30.60
Declination
Declination
-53.40
-53.60
30.40
-53.80
30.20
-7.80
-54.00
30.00
-8.00
125.00
A644
124.80
124.60
124.40
124.20
Right ascension
124.00
123.80
A3158
-54.20
RXC J1825
29.80
56.50
56.00
55.50
55.00
Right ascension
277.00
276.80
276.60
276.40
XMM is very efficient at detecting low-mass structures around
nearby clusters
Possible strategy: Expand the X-COP program to a
sufficient number of local clusters to measure the
sub-halo mass function
D. Eckert
X-COP
276.20
Right ascension
276.00
275.80
Toward the next decade: The high-T phase of the WHIM
Key question: Can we detect the hottest phase of the WHIM?
The FOV and sensitivity of XMM are best matched for filaments
around intermediate-redshift clusters (here MACS 0717)
D. Eckert
X-COP
Toward the next decade: The high-T phase of the WHIM
Key question: Can we detect the hottest phase of the WHIM?
The FOV and sensitivity of XMM are best matched for filaments
around intermediate-redshift clusters (here MACS 0717)
Possible strategy: Target the most massive clusters in
the redshift range 0.3 − 0.5 with deep XMM
observations
D. Eckert
X-COP
Summary
Combining X-ray and SZ data in cluster outskirts highlights
the importance of clumping
XMM is efficient at finding accreting subhalos around the virial
radius
Expect many more results on cluster outskirts with the recent
X-COP very large programme
We discovered 3 filaments radially connected to A2744
Properties consistent with WHIM, gas fraction 5-10%
The Next Decade: Constrain the sub-halo mass function and
search for filaments around intermediate-redshift clusters
D. Eckert
X-COP
Backup Slides
Gas clumping factor
Azimuthal median is robust against inhomogeneities
10
Azimuthal mean
Azimuthal median
1
SX
10-1
10
-2
10
-3
10
-4
10-1
1
R/R200
Eckert et al. 2015
D. Eckert
X-COP
Gas clumping factor
ROSAT/PSPC ENZO NR GADGET NR GADGET CSF+AGN
R 200
2
Data
V13
1.8
R13 NR
R13 CSF+AGN
R13 Residual
C
1.6
1.4
1.2
1
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
R/R500
Eckert et al. 2015
Hydrodynamical simulations predict too many substructures in
the outskirts
Including AGN + SN feedback improves the match
D. Eckert
X-COP
Galaxies and DM overdensity in the filaments
10
Lensing signal
Cluster
South
North-West
Ngal [arcmin-2]
1
10-1
2
4
6
8
10
10-3
Cluster
South
North-West
East
70
East
Surface Mass Density [h M kpc-2]
Galaxy density
12
10-4
1
10
Radius [arcmin]
Radius [arcmin]
Eckert et al. 2015
Excess galaxy and DM density is observed in the regions
encompassing the filaments. The gas fraction in the filaments is
5-10%
D. Eckert
X-COP
Nature of the filaments
0.01
10−3
normalized counts s−1 keV−1
Spectral analysis reveals thermal gas
with T ∼ 1 keV
East
0.5
1
2
Energy (keV)
5
Eckert et al. 2015
We are observing diffuse hot gas originating from the LSS and
heated up by the gravitational pull of A2744
D. Eckert
X-COP