The interaction between AGN and the ICM

1
The interaction between AGN and the ICM

• Marcus Brüggen
• Mateusz Ruszkowski, Sebastian Heinz, Mitch
  Begelman, Elke Rödiger, Matthias Hoeft, Paola
  Rebusco, Hans Böhringer, Eugene Churazov, Aurora
  Simionescu
• Garching, 8 August 2006

2
Birzan et al. 2004
3
Why bother?
4
AGN feedback may be responsible for exponential
cut-off of LF
With AGN
Without AGN
Croton et al 2004
5
Two Questions

• What heats cooling flows?
• What enriches the ICM with metals?

6
Simulations of jets/bubbles in clusters
Bubbles in equilibrium
conduction
Jets
Cluster model in hydrostatic equilibrium
viscosity
2D/3D
radiative cooling
HD/MHD
Grid/SPH
7
Brüggen Kaiser, Nature 418, 301 (2002)
8
dens
T
X-ray
Reynolds, Heinz, Begelman (2001)
9
Temperature
Entropy
Bolometric emissivity
Dalla Vecchia et al., MNRAS 355, 995 (2004)
10
No bubbles
increasing bubble energy
Temperature profile
Entropy (index) profile
Dalla Vecchia et al., MNRAS 355, 995 (2004)
11
Dalla Vecchia et al., MNRAS 355, 995 (2004)
12
Left 0.3-1.5 keV unsharp-masked X-ray image
(width 1 0.98 arcsec, width 2 9.8 arcsec).
Regions brighter than average are light, those
fainter are dark. Right Overlay of contours from
H-alpha image.
Fabian et al. 2003
13
Microphysics of the ICM
viscosity
conductivity
But
conductivity dominates dissipation by up to a
factor of 40
magnetic fields affect conductivity more strongly
14
The effect of viscosity on bubble stability
Reynolds et al., MNRAS 357, 242 (2005)
15
inner 254 kpc of simulation shown
Vernaleo Reynolds (2006)
T
Low-density channel prevents isotropic heating
P
S
16
This is what a real cluster looks like...
17
FLASH simulations...
Shortly Relativistic accretion onto NS
Flame-vortex interactions
Compressed turbulence
Type Ia Supernova
Orzag/Tang MHD vortex
Wave breaking on white dwarfs
Intracluster interactions
Nova outbursts on white dwarfs
Rayleigh-Taylor instability
Laser-driven shock instabilities
Helium burning on neutron stars
Cellular detonation
Magnetic Rayleigh-Taylor
Richtmyer-Meshkov instability
18
Award on SGI Columbia
SGI Altix 10240 Procs, 20 TB shared memory
19
3D Simulations of dissipation in clusters
3D FLASH AMR simulations 7 levels of
refinement 20483 effective zones Initial model
S2 from Springel et al. 2001 (6 keV, 7 1014
Msol) Box size 2 /h Mpc, gas 714000
collisionless particles. recurrent source in the
centre Hot gas is injected into spherical
region of 13 kpc diameter (average L
8.3x1044erg/s, P 30 Myrs) Modify momentum
energy equation to account for viscosity
(Spitzer) Thermal conduction
20
Brüggen, Ruszkowski Hallman (2005)
21
cluster extracted from a full COSMOLOGICAL
SIMULATION
2.5Mpc
400kpc
Brüggen, Ruszkowski Hallman (2005)
22
Buoyantly rising bubbles temperature slice

Brüggen, Ruszkowski Hallman (2005)
23
Periodic behavior reflects periodicity of the
source
Different annuli (10 kpc)
Weakly nonlinear
Ruszkowski, Bruggen Begelman (2004b)
24
SYNTHETIC Chandra OBSERVATIONS

• background subtracted
• vignetting-corrected
• 200-ks Chandra ACIS-I
• unsharp masked photon image 0.3-1.5 keV

25
Which one is the synthetic observation of a
simulated cluster?
26
Fe line signature of bubbles
27
Hoeft Brüggen (2005)
28
(No Transcript)
29
Cluster metallicities
ICM metallicity is about 1/3 solar
Cool core clusters
Non-cool core clusters
Uniform metal distribution Early enrichment by
Type II SNe
Peaked abundance profile centred on BCG Type
Ia Metal distribution is broader than BCG light
distribution
30
XMM
Chandra
Expected profile from SNIa in 8 Gyrs
Metal profile 0.3 solar
Rebusco et al., MNRAS 359, 1041 (2005)
31
(No Transcript)
32
Influence of bubbles on metal distribution in ICM

• How far out can bubbles carry metals?
• How efficient is this process?
• On which timescales does this work?
• How does it depend on cluster bubble
  parameters?
• Does this process cause characteristic features,
  such as
• anisotropies?

33
Roediger, Brüggen, Rebusco, Böhringer Churazov
(2006)
34
Roediger, Brüggen, Rebusco, Böhringer Churazov
(2006)
35
Roediger, Brüggen, Rebusco, Böhringer Churazov
(2006)
36
Extensive study of the effect of bubble-induced
motions on metallicity profiles in clusters using
3D hydrodynamic simulations by
Roediger, Brüggen, Rebusco, Böhringer Churazov,
MNRAS submitted (2006)

• explored dependence of metal profiles on
• bubble size
• bubble position
• cluster parameters
• recurrence times

• Some results
• inferred diffusion coefficients for Perseus at
  1029 cm2s-1 at radius of 10 kpc
• this is in good agreement with diffusion
  coefficients inferred from observations
• in hydrostatic cluster model resulting metal
  distribution very elongated along
• direction of the bubbles

37
The answer is blowing in the wind....
For more on this see talk by Sebastian Heinz...
Heinz, Brüggen, Young, Leveque (2006),
astro-ph/0606664
38
For more info and better movies see talk by
Sebastian Heinz
Heinz, Brüggen, Young, Leveque (2006),
astro-ph/0606664
39
Open Questions
How does feedback operate? How are the bubbles
inflated? Where and how is energy
dissipated? How long do bubbles live? Are they
stabilised? B-fields? What is the equation of
state of the bubbles? How important are
microphysical processes such as viscosity and
conduction? Is multiphase physics
important? What are the relative contributions
to the kinetic energy input of the central AGN
and other processes such as mergers etc.?
40
3C 338 and Abell 2199 Johnstone et al. 2002
Chandra image 1.7 GHz radio
Multiple fossil bubbles not aligned with the
current jet axis

41
How does the AGN know, how much heating is
required?

• feedback accretion onto central AGN triggers
  outburst
• luminosity of AGN proportional to accretion rate

42
Quasi-hydrostatic 1D models

• solve hydrostatic equation in fixed cluster
  potential (NFW)
• entropy evolves subject to cooling, heating
  conduction
• explicit integration in time to compute
  evolution of cluster
• b.c. M(0) 0 and P(r200) const.

43
varying efficiency