Magnetic fields in and beyondclusters of galaxies

Klaus Dolag

Perseus

Hydra

Coma

A3627

Virgo

Pavo

Centaurus

Donnert, Dolag et al. 2009

17/06/2009 – p. 1

IntroductionCMB

Cosmic structure today

TemperatureDensity

275 Mpc

(t = 0.38 Myr)

(t = 13.7 Gyr)

Borgani, Murante, Springel, Diaferio, Dolag et al. 2004

The cosmic web today (z = 0) is mainly accessible throughsimulations (warm, thin). Model predictions for~B are importantfor propagation of ultra high energetic cosmic rays (UHECRs).

17/06/2009 – p. 2

Introduction

Temperature

275 Mpc

40 Mpc

Density

"Zoomed" Simulation of a galaxy clustert = 0.38 Myr

t = 13.7 Gyr

Dolag, Borgani, Murante & Springel 2008

Clusters form at the nodes of the cosmic web and can be usedas a tool to understand the physical state of diffuse baryons.

17/06/2009 – p. 2

IntroductionObservations (⇒), Simulations (⇐) and the role of~B:

DSSVLA

Galaxies (optical, radio):⇒ Interaction with the ICM

⇐ Galaxies in dense environment(stripping , distribution of metals)

⇐ Magnetic fieldseeding(outflows)

Snowden, ROSAT

ICM (X-ray, thermal bremsstrahlung):⇒ Dynamical state of ICM

⇐ Non thermalpressuresupport

⇐ Turbulence, viscosity, shocks

Deiss, Effelsberg

ICM (radio, synchrotron radiation):⇒ Distribution of ~B, CRs (diffuse + RM)

⇐ Evolution andbuildup of ~B

⇐ Accelerationandpropagation of CRs17/06/2009 – p. 2

Faraday Rotation (RM)

9h15m41.0s9h15m41.5s9h15m42.0s9h15m42.5s9h15m43.0s

-11:53:00

-11:52:50

-11:52:40

RA

DEC

-2000 0 2000 4000 6000 (RAD/M/M)

50 100 150 200

200

400

600

-100 -50 0 50 (counts)

3C449 Hydra

Feretti et al. 1999

Taylor & Perley 1993

150

kpc

20 kpc

High quality Rotation Measure maps across the lobes of thecentral radio source in 3C449 (left) and Hydra (right).

• Cool core versus cluster wide turbulence / fields structure?• Origin of cosmological magnetic fields ?• Extension, structure and evolution of magnetic fields ?

17/06/2009 – p. 3

Faraday Rotation (RM)B(r) = B0

(

1 + (r/rc)2)−1.5µ

, |Bk|2 ∝ k−n

2 MpcA119

Murgia, Govoni, Feretti, Giovannini, Dallacasa, Fanti, Taylor & Dolag 2004

⇒ B0 = 5µG n = 2 µ = 0.517/06/2009 – p. 3

Faraday Rotation (RM)Coma

3 Mpc

Bonafede, Feretti, Govoni, Murgia, Giovannini, Dolag & Taylor 2009

n

1.0

1.5

2.0

2.5

3.0

3.5

10 2 3 4 5

100 200 300 400 500maxλ

6

17/06/2009 – p. 3

Radio Halos (synchrotron)

Feretti 1999

Radio Halo

Radio Relic

1 Mpc

Dolag & Ensslin 2000

Cluster widediffuse synchrotron emissionconnected tomerger events,periferal emission directly connected toshocks.

• Radio halo: Turbulence, shocks, secondary ?• Relics: Primary from shocks or compressed radio plasma ?

17/06/2009 – p. 4

Simulation Network

−4.5 −4.0 −3.5 −3.0 −2.5 −2.0 −1.5

Virgo

Hydra

Perseus

Coma

Centaurus

2log( )ρ /cm ][g

−11.0−13.0−15.0−17.0−19.0−21.0−23.0

λ γlog( )γ 2[ /cm /s/arcmin ]2

Virgo

Hydra

Perseus

Coma

Centaurus

−11.0−13.0−15.0−17.0−19.0−21.0−23.02log(Lx) [erg/cm /s/arcmin ]2

−11.0 1.0−2.0−5.0−8.0−17.0 −15.0

log(P )ν [mJy/arcmin ]2

1.0−1.0−3.0−5.0−7.0−9.0−11

log(|B|) µ[ G]

Perseus

Hydra

Coma

A3627

Virgo

Pavo

Centaurus

(M)δMechanism

???Efficiency

CR−p CR−e

−3.0−4.0−5.0 −2.0 −1.0 0.0 1.0

log( )T [keV]

−9.5 −8.5 −7.5 −6.5 −5.5 −4.5log(Y)

−10.5

Temperature

Density

Structure Formation

ICs (Cosmology)

Turbulence

−ray SBγ Radio SB

Cosmic Rays

AGN ?

Observables

X−ray SB

thermal SZ

Dolag, Hansen, Roncarelli & Moscardini 2005 Dolag, Grasso, Springel & Tkachev 2004/2005Donnert, Dolag, Cassano & Brunetti 2009

Seed Magnetic Field

Magnetic Field Evolution

Shock Statistics

UHECR−

Feedback ?Star Formation ? Dissipation ?

Numerics ?Resolution ?

Magnetic Field

Magnetic Pressure

Compression

Coupling to Star Formation

Viscosity ?Numerics ?Sub−Grid Model ?

Description ?Diffusion ?

Detection ?Numerics ?

Deflection17/06/2009 – p. 5

MHD with SPH

Why SPH ?

• Perfect self gravity⇒ excellent to capture structure

formation

• Lagrangian⇒ excellent advection

• Highly adaptive⇒ large dynamical range

• Low dissipation⇒ no numerical mixing

• Efficient⇒ large simulations

• Simple⇒ allows to include various physical processes

• Ideal MHD:+ Induction equation + Lorenz force⇒ Many details to obtain stable and reliable

implementation

17/06/2009 – p. 6

MHD with SPHCode verification in 1D shock tube tests (Ryu & Johns 1995)(Dolag et al. 1999, 2001, 2005) ...Code verification in various 2D tests

Athena

Art. Dissipation B smooth

SPH−MHD Athena

Art. Dissipation

SPH−MHD

B smooth

Magnetic field in the vortex (Orszag & Tang 1979) and the rotor (Balsare

& Spicer 1999) test problem. Results for Anthena (top left) andSPH-MHD runs (Dolag & Stasyszyn 2008).

17/06/2009 – p. 6

Cosmological MHD simulations

Origin• Primordial• Battery• Dynamo (Turbulence)• Stars• Supernova• Galactic Winds• AGN, Jets• Shocks Rees 1994

+ further amplification bystructure formation

- dissipation ?

17/06/2009 – p. 7

Cosmological MHD simulations5 Mpc

Dolag et al. 1999/2002

First cluster MHD simulations (Dolag et al. 1999/2002)• Simulations reproduce the radial shape of the RM signal

⇒ Magnetic power spectrum of clustersn ≈ 2.3 − 3.1• Magnetic field configuration driven by cluster dynamics

⇒ Initial magnetic fieldstructure not important• Initial fields of≈ (0.2 − 1) × 10−11 G are sufficient

⇒ in range formany modelsfor magnetic seed fields17/06/2009 – p. 7

Cosmological MHD simulations

(counts)

RA

DEC

(RAD/M/M)

684 kpc

68.4 Mpc

6.84 Mpc

684 Mpc

RM

SimulationObservation

Feretti et al. 1999

3C449

“Zoomed” cluster simulations (Dolag & Stasyszyn 2008)17/06/2009 – p. 7

Cosmological MHD simulations

⇒ Several times confirmed since early work⇒ Generic feature from structure formation

17/06/2009 – p. 7

Cosmological MHD simulations

But: Central part steepens strongly with resolution.⇒ Resolutionis important

17/06/2009 – p. 7

Cosmological MHD simulations

Attention : Also depends strongly on dissipation⇒ Numerical dissipationis important

17/06/2009 – p. 7

Cosmological MHD simulations

(counts) (counts) (counts) (counts)

RA

DEC

(RAD/M/M)

−3 3 −30 30 −70 70 −600 600

3000x220x130x10x3C449

Feretti 1999 Dolag 2005 Dolag 2006 Dolag 2009 Dolag 2009

(counts)70−70

Observed and simulated RM maps up to highest resolutionsimulation: MHD, 20 Million particles withinRvir,mDM = 107M⊙/h, ǫGrav = 1kpc/h(work in progress).

17/06/2009 – p. 7

Cosmological MHD simulations

3C440 (Feretti 1999)220x (Dolag 2009)

3000x (Dolag 2009)

130x (Dolag 2006)10x (Dolag 2005)

Structure function derived from observed and simulated RMmaps up to highest resolution simulation: Indication for need ofmagnetic dissiation (work in progress).

17/06/2009 – p. 7

Origin of Magnetic Fields

Seeding from galactic outflows (Donnert, Dolag et al. 2008)17/06/2009 – p. 8

Origin of Magnetic Fields

Different wind parameters (Donnert, Dolag et al. 2008)17/06/2009 – p. 8

Propagation of UHECRs

Trajectories of Cosmic Rays diffusing through the cluster core(Rordorf, Grasso & Dolag 2004)

17/06/2009 – p. 9

Propagation of UHECRs

Perseus

Coma

Virgo

HydraCentaurus

Pavo

A3627

0.0 1.0 2.0 3.0 4.0 5.0 [Degrees]

Full sky deflection signal for4 × 1019eV Cosmic Rays withoutlosses, using a sphere of 110 Mpc radius andB0 = 10−5µG(Dolag, Grasso, Springel & Tkachev 2004)

17/06/2009 – p. 9

Propagation of UHECRs

Perseus

Coma

Virgo

HydraCentaurus

Pavo

A3627

0.0 0.2 0.4 0.6 0.8 1.0 [Degrees]

Full sky deflection signal for4 × 1019eV Cosmic Rays withoutlosses, using a sphere of 110 Mpc radiusB0 = 0.2 × 10−5µG(Dolag, Grasso, Springel & Tkachev 2005)

17/06/2009 – p. 9

Propagation of UHECRs

Perseus

Coma

Virgo

HydraCentaurus

Pavo

A3627

0.0 0.1 0.2 0.3 0.4 0.5 [Degrees]

Full sky deflection signal for4 × 1019eV Cosmic Rays withoutlosses, using a sphere of 110 Mpc radius from galactic outflows(Just for you !)

17/06/2009 – p. 9

Propagation of UHECRs

Sky coverage of deflection signal for4 × 1019eV Cosmic Rayswithout losses, using a sphere of 110 Mpc radius for all models(Also just for you !)

17/06/2009 – p. 9

Propagation of UHECRs

Sky maps of UHECRs emitted uniformly from M87 with 1000(upper right), 100, 10 and 1 EeV (lower left)(Dolag, Kachelriess, Semikoz 2008), see talk by Dimitri Semikoz.

17/06/2009 – p. 9

Investigating magnetic fieldsRotation Measure:

RM ∝

∫

ne B‖ dl

⇒ additional proportionality to density

⇒ not very good for low density regions

Also background/foreground subtraction not trivial for next

generation of instruments like LOFAR, SKA or EVLA !

Other methods:

• Propagation of UHECRs (this talk)

• Blazar halos(Dolag, Kachelriess, Ostapchenko & Tomas 2009)

(see talk by Dimitri Semikoz)

• Dynamics in structure formation !

17/06/2009 – p. 10

Investigating magnetic fieldsGalaxy formation studies with RAMSES:

Teyssier 2009

17/06/2009 – p. 10

Investigating magnetic fieldsCosmological simulation with Gadget:

Bini = 10−5µG, work in progress,(Stasyszyn & Dolag 2009)

17/06/2009 – p. 10

Conclusions

• Predicted magnetic field structure reflects formationof galaxy clusters

• Galactic outflows are valid sources for cluster fields• Predicted (complex) field structure in galaxy clusters

is compatible with RM measures• Filaments might host remaining signatures of magnetic

field origin• UHECR propagation very promising

⇒ Pointing back to sources might be possible⇒ Complications from propagation in nearby clusters⇒ non defection of UHECRs measure small~B-fields

• Blazar halos allow to probe even lower~B-fields⇒ see talk by Dimitri Semikoz

17/06/2009 – p. 11

S.O.S

S.O.S

B

S.O.S

S.O.S

Stars

Radio Emission

hard/soft X−RayRadio Ghosts

Sharp Edges (cold fronts !)

Cooling

Some Baryonic Matter

Lots of Dark Matter

Turbulence

Thermal Conduction

Lots of Dark Energy ?

EUV excess ?

Do

we u

nd

ersta

nd

ou

r "

wo

rld

" !?

Galaxy clusters as physics laboratory:

Thermal Emission

WHIM

17/06/2009 – p. 12

Comparison Project

17/06/2009 – p. 13

Comparison Project

z=0.25

GadgetENZOz=1.0 z=0.5

z=0.25 z=0.11

z=1.0

z=0.11

z=0.5

17/06/2009 – p. 14

Comparison Projectz=1z=1

Gadgetz=0.5

z=0.11z=0.25 z=0.25

z=0.5

z=0.11

ENZO

17/06/2009 – p. 14

Comparison Project

17/06/2009 – p. 14

Turbulence in ClustersOld viscosity scheme New viscosity scheme

• Instabilities less damped (e.g. Kelvin-Helmholtz).

⇒ Inset of turbulence

⇒ Enlarged energy-fraction in gas velocity

Dolag, Vazza, Brunetti, Tormen & Springel 2004

17/06/2009 – p. 15

Turbulence in Clusters

0 0.0001 0.0002 0.0003 0.0004 dT/T

200 400 600 800 1000

200

400

600

800

1000

g8.gas.a.z

0 0.0001 0.0002 0.0003 dT/T

200 400 600 800 1000

200

400

600

800

1000

g8.gas_nv.a.z

Standard (left) and reduced (right) Viscosity, Compton-y map

17/06/2009 – p. 16