Avishai Dekel The Hebrew University of Jerusalem

Jerusalem Winter School 2012/13

Feeding High-z Galaxies from the Cosmic Web

Jerusalem Winter School 2012/13 Lecture 1

z=0

z=2

z=8

Oxford Dictionary: simulatepretend to be, have or feel, hide under a false appearance, feign, put on, dissemble

Toy modelingCosmological simulations

Consider a spherical cow…

Toy modeling Cosmological simulations

Oxford Dictionary: simulatepretend to be have or feel, hide under a false appearance, feign, put on, dissemble

Consider a spherical cow…

1. Cold Streams from Cosmic Webco-planar, pancakes, angular momentum, mergers, outflows, detection in Lyα

2. Violent Disk Instability (VDI)in-situ clumps, evolution of instability

Key Concepts at high z

in-situ clumps, evolution of instability

3. Instability-Driven Inflow

spheroids, black Holes

Lecture 1: Outline

1. Cosmological inflow rate

2. Virial shock heating and cold flows

3. Cold streams from the cosmic web

4. Smooth flows and mergers4. Smooth flows and mergers

5. Stream penetration and fragmentation

6. Observing cold streams in Lyman alpha

7. Angular momentum: transport and exchange

8. Outflows

1. Cosmological Inflow rate

Neistein, van den Bosch, Dekel 2006Neistein & Dekel 2008Dekel et al. 2009; 2012

Genel et al. 2008Genel et al. 2008

From EPS and N-body simulations (5-10% accuracy)

Accretion Rate into a Halo at z>1Gyr5.17)3/2()/()1( 1

02/1

13/2

11 ≈Ω=≈+= −−− Htttza mEdS cosmology at z>1

3300

30u cmg105.2 −−− ×≈= ρρρ a

u20032 200)4( ρ∆== RMRGMV 1

33/1

121002/1

33/1

12200 )1()1( −+≈+≈ zMRzMVVirial relations

PS self-invariant time 3/2∝∝=δω

nkPnszMsMM ∝+≈≈≈+≈ − 6/)3(14.0Gyr03.0)1( 112 βµβ&

2/5→µ PS self-invariant time 3/2)(/ taDaDc ∝∝=δω2/5.const/ −∝∝→=∂∂ aMM ωω &&

Toy model 12/5 Gyr03.0)1( −≈+≈ szsMM&

79.0)2/3()( 1)(

00 ≈=≈ −− steMzM zz αα

17.05.2

314.1

121

baryon )1(80 fzMyrMM +≈ −⊗

&

2/5→µ

0110 ttMMM →∆+= ))(()()|( 0101 MgMMfMMP σ=MM ∝&

EPS evolution of main progenitor

AMR Cosmological Simulations

Cosmological box, RAMSES (Teyssier), resolution 1 kpc

Zoom-in individual galaxies, ART (Kravtsov, Klypin), RAMSES Ceverino, Tweed, Dekel, Primack:

- 50 pc res. (30 galaxies)- 25 pc res., lower SFR, stronger feedback

Isolated galaxies, resolution 1-10 pc, RAMSES, Bournaud et al.

HUJI: Ceverino, Goerdt, Mandelker, Tweed, Zolotov, …

UCSC: Fumagalli, Moody, Mozena, …

Collaborators: Bournaud, Teyssier, Krumholz, …

Toy Model vs SimulationsDekel, Zolotov, Tweed, Cacciato, Ceverino, Primack 2013

12/5 Gyr03.0)1( −=+= szsMM& 79.0)( )(0

0 ≈≈ −− αα zzeMzM

Steady State: SFR Governed by Accretion

*accgas MMM &&& −=

sfrgas* τMM =&Kennicutt SFR

Mass conservation

bfaf −=&If vary on a timescale < τsfr1

sfracc,−τM&

)1()1( /*

//acc

ττττ tacc

taccgas

tgas eMMeMMeMM −−− −==−= &&&&&

acc*gas 0 MMM &&& →→Steady state after τsfr

sfracc

acc 4/

τ≈≥ tdtMd

M&

&

sfr1sfr

1sfr 4/

τττ

≈≈−

−

tdtd Gyr75.0ff

-1ffsfr ≈≈ tετ

0023.014.005.03

1ff ≈××≈≈t

t

t

t

t

t

t

t vir

vir

disk

disk

ff

Timescale for change of parameters from z=4 to z<1

*acc accaccgasgas

Cosmological Steady State: SFR ~ accretion rate into the disk

sfraccgas,gas MMM &&& −=

gasMM =&

Mass conservation:

0gasaccgas,sfr ≈≈ MMM &&&

sfr

gassfr τ

M =&

But gas accumulates in disks at earlier times and in smaller masses

109

106

halob MfSFR &=

Bouche et al. 09

Star-formation history:

1011

2x1010Mmin=et al. 09

2. Virial shock heating and cold flows

Rees & Ostriker 1977; Silk 1977; Binney 1977

Birnboim & Dekel 2003; Dekel & Birnboim 2006Keres et al. 2005,2008; Ocvirk et al. 2008

Birnboim & Dekel 03, Keres et al 05, Dekel & Birnboim 06

11 −− < compresscool tt

- Critical halo mass ~~1012Mʘ

3. Virial Shock Heating

- Critical halo mass ~~10 Mʘ

- Cold streams penetrate the ….hot medium at z>2

Kravtsov

Old Standard Picture of Infall to a Disc

Perturbed expansion

Halo virialization

Gas infall, shock heating

Rees & Ostriker 77, Silk 77, White & Rees 78, …

Gas infall, shock heatingat the virial radius

Radiative cooling

Accretion to disc if tcool<tff

Stars & feedbackM<Mcool ~1012-13M⊙

Cooling rate

-22

-21

mic

[erg

cm

3 /s]

Z=0

Z=0.05Z=0.3

-24

-23

4 5 6 7 8

log

Λm

ic

log T [K]

Line emission Bremsstrahlung

particleper e gper molecules / ]s g [erg )( -1-1-2

22

χµρµ

χA

A NTN

q Λ=

T-1

Brems.

Rees & Ostriker 77; Silk 77; White & Rees 78; Blumenthal et al. 86

Cooling vs Free Fall

104 105 106 107

+3

+1

T

galaxies

tcool<tff

He

H2

Brems.

virial velocity

log gas density

H

10 30 100 300 1000

-1

-3

-5

CDM

galaxies

clusters

upper bound too big

Consider a spherical cow…

Gas through shock: heats to virial temperature

compression on a dynamical timescale versus radiative cooling timescale

11 −− < compresscool tt

Shock-stability analysis: post-shock pressure vs. gravitational collapse

3

4

5

21

V

Rt scompress ≈≡

ρρ&

Shock Stabilitypost-shock pressure vs. gravitational collapse

Birnboim & Dekel 03

with cooling rate q (internal energy e):

cool

compeff 21

5

3

5

)(ln

)(ln

t

t

e

q

d

Pd−=−=≡

ρργ

ργ

&

s

P

∂∂

=ρ

γln

lnadiabatic:

3/4>γstable:

3

4

5

21 shockcomp V

Rt ≈≡

ρρ&

prepost2

cool 416

3

),(ρρ

ρ≈≈

Λ∝≡ VT

ZT

T

q

et

cool213)(ln ted ρρ &

qVPe −−= &&

7

10>effγ

Stability criterion:

11 −− < compresscool tt

Growth of a Massive Galaxy T °K

shock-heated gas

1011Mʘ 1012Mʘ

“disc”

shock-heated gas

Spherical hydro simulation Birnboim & Dekel 03

A Less Massive Galaxy

shockedcold infall

T °K

1011Mʘ

“disc”

Spherical hydro simulation Birnboim & Dekel 03

z=4M=3x1011M

Tvir=1.2x106KRvir=34 kpc

Hydro Simulation: ~Massive M=3x1011M

Kravtsov et al.

virial shock

Kravtsov et al.

Less Massive M=1.8x1010M

coldinfall

virial radius

z=9M=1.8x1010

Tvir=3.5x105

Rvir=7 kpc

coldinfall

Shock-Heating Scale

Mvir [Mʘ]

Birnboim & Dekel 03 Dekel & Birnboim 06

stable shock

6x1011 Mʘ

Keres et al 05

unstable shock

typical halos

The Critical Mass: Cosmological Simulations

virial radius

SPH Keres et al 2005, AMR Kravtsov et al

M<<1012Mʘ cold flows M>1012Mʘ virial shock heating

radius

Fraction of Cold Gas in Halos: Cosmological simulations (Kravtsov)

cold

hot

Birnboim, Dekel, Neistein 2007

d(Entropy)/dt

Virial shock: rapid expansion from the inner halo to Rvir

Necessary condition M>Mcrit Trigger: minor merger

Two Galaxy Types: Bi-modality

Bi-modality in color, SFR, bulge/disk

E/SE/S00/Sa/Sa00..6565<z<<z<00..7575

Disks and IrregularsDisks and Irregulars

Bell

M*crit~3x1010Mʘ

Color-Magnitude bimodality & B/D depend on environment ~ halo mass

environment density: low high very high

SDSS: Hogg et al. 03

spheroids

disks

Mhalo>6x1011

“cluster”Mhalo<6x1011

“field”

Galaxy Bi-modality about Mcrit

While halos grow by smooth accretion and mergers

Dekel & Birnboim 06

M<Mcrit: The Blue Sequencecold gas supply → disk growth & star formation

Stellar feedback regulates SFR → long duration

mergers & disk instability → bulges

M>Mcrit: The Red Sequenceshock-heated gas + AGN fdbk → no new gas supply

gas exhausted → SFR shuts off

passive stellar evolution → red & dead

further growth by gas-poor mergers

In a standard Semi Analytic Model (GalICS)

not red enough

excess of big blueCattaneo, Dekel, Devriendt, Guiderdoni, Blaizot 06z=0

data ---sam ---

color

no red sequence at z~1

too few galaxies at z~3

colo

r u

-r

magnitude Mr

star formation at low z

With Shutdown Above 1012 Mʘco

lor

u-r

magnitude Mr

Standardco

lor

u-r

magnitude Mr

With Shutdown Above 1012 Mʘco

lor

u-r

magnitude Mr

From Blue to Red Sequence by Shutdown Dekel & Birnboim 06

Mvir [Mʘ]

redshift z

all cold

hot

1014

1013

1012

1011

1010

109

0 1 2 3 4 5

in hot

cold

Mshock

3. Cold Streams from the Cosmic Web

Dekel et al. 2009Danovich, Dekel, Hahn, Teyssier 2012Danovich, Dekel, Hahn, Teyssier 2012

Pichon et al. 2012Kim et al. 2012

At High z, in Massive Halos: Cold Streams in Hot Halos

Totally hot at z<1

in M>Mshock

Cold streams

shock

Cold streams at z>2

no shock

coolingDekel & Birnboim 2006

Kravtsov et al

cold flows

disk

Mass Distribution of Halo Gas

density

Temperature

adiabatic infall

shock-heated

flows

Mvir [Mʘ]

all hot

1014

1013

1012

cold filaments

in hot medium

MshockMshock>>M*

Mshock~M*

Cold Streams in Big Galaxies at High z

M*

10

1011

1010

109

0 1 2 3 4 5

redshift z

all cold

Mshock

Dekel & Birnboim 06

high-sigma halos: fed by relatively thin, dense filaments → cold narrow streams

the millenium cosmological simulation

typical halos: reside in relatively thick filaments, fed ~spherically → no cold streams

Ms~M*

At high z, M halos are high-σ peaks:

Origin of dense filaments in hot halos (M≥Mshock) at high z

At low z, Mshock halos are typical: they reside in thicker filaments of comparable density

Ms>>M*

Large-scale filaments grow self-similarly with M*(t) and always have typical width ~R* ∝M*

1/3

At high z, Mshock halos are high-σ peaks: they are fed by a few thinner filaments

of higher density

Narrow dense gas streams at high z versus spherical infall at low z

high z low z

Ocvirk, Pichon, Teyssier 08

M=1012Mʘ ~~MPSM=1012Mʘ >>MPS

Mstream

Mshock

DB06cold filaments

in hot medium

Critical Mass and Cold Streams in Hot Halos: Cosmological Simulations

Ocvirk, Pichon, Teyssier 08

Mshock

Cold Sterams at ~1 kpc Resolution

- >90% of influx in cold streams

- Penetration: V~~Vvir dM/dt(r)~~const

- Hot accretion negligible

– recycled outflows

100 kps

Dekel et al 09 Nature

.

Cold streams through hot halos

Cold streams through hot halos

outflows

Flux per solid angle

Dekel et al 09

Streams riding DM filaments of Cosmic Web. High-z massive galaxies form at the nodes.

Cosmic-web Streams feed galaxies:mergers and a smoother component

AMR RAMSES Teyssier, Dekel

box 300 kpc

res 30 pc

z = 5.0 to 2.5

Tweed, Dekel, TeyssierRAMSES Res. 50 pc

Streams Feeding a Hi-z Galaxy

Streams Feeding a Hi-z Galaxy

Ceverino, Dekel, Primack ART res. 35-70 pc

Cold Streams Penetrate through Hot Halos

Mv > 1012 M

100 kpc Agertz et al 09

Co-planar Streams and Pancakesinflux M

yr-1rad-2

Danovich, Dekel, Teyssier, Hahn

200

50

1-2Rvir

Streams in a Pancake

influx Myr-1rad-2

Streams in a Pancake

Flows into pancakes, and along pancakes to filaments

The stream plane extends from r<0.4Rv to r>5Rv

Influx at Rvir: 70% in streams, 20% in pancakes

MW4 z=2.3

MW4 z=7

influx Myr-1rad-2

Influx in the streams: 55% in 1 stream, 90% in 3 streams

Distribution of Influx in Streams and Pancakes

Influx:70% in streams 20% in pancakes

streams

pancakes

>50% in 1 stream >90% in 3 streams

streams

Origin of Structure near a Node of the Cosmic Web

A dominant filament in a dominant sheet within a smoothing volume

3 major filaments in a plane (Arnold et al. 1982)

A honeycomb:maximum volume of voids minimum length of filaments

1 4 5

Pancakes of low Entropy

MW1

Entropy Influx

Hahn

MW5

MW4

4. Smooth Flows and Mergers

Dekel et al. 2009

Dekel, Teyssier et al 09; MareNostrum RAMSES sims 1kpc res

Stream Clumpiness - Mergers

Mass input to galaxies (all along streams)

- Major mergers >1:3 <10%

- Major+minor mergers >1:10 ~33%

- Miniminors and smooth flows ~67%M=1012Mʘ z=2.5

Distribution of gas inflow rateCosmological hydro simulations (MareNostrum, Dekel et al. 09)

mergers >1:10

)|( MMP &

smooth flows

Galaxy density at a given gas inflow rate Dekel et al. 2009, Mare Nostrum Simulation

SFR ~(1/2) inflow rate

&& ≈

Star-Forming Gal’s

Sub-Millimeter Gal’s

SFR~

*2MM && ≈

Mergers have a Limited Contribution to SFR

At a given dM/dt, 3/4 galaxies are fed by smooth flows

Bright SMGs are mergers & smooth flows

SFGs are miniminor mergers, i.e. smooth flows

In-situ vs Ex-situ Star FormationTweed, Zolotov, Dekel, Ceverino et al. 2012

ex-situ

in disk

in bulge

From z=4 to z=1In-situ SFR 77%-57%Ex-situ mergers 23%-43%

Mass Added in Major vs Minor Mergers

minorM

ddM )( ω

Neistein & Dekel 2008

)(tDcδω =

15.2 Gyr)1(04.0 −+−≈ zω&

ωω&

&

M

ddM

M

M )(=

major

5. Stream penetration and fragmentation

Dekel et al. 2009Dekel, Zolotov, Tweed, Ceverino, Primack 2013Dekel, Zolotov, Tweed, Ceverino, Primack 2013

Baryon Penetration to the Disk: ~50%

)(8.00

0zzeMM −−=

2/51 )1(Gyr03.0 zMM += −&

Rvir

Toy models versus simulations (ART 50 pc):Dekel, Zolotov, Tweed, Cacciato, Ceverino, Primack

EPS approximation:

Rvir

0.1 Rvir

A proxy for SFR

Baryon Penetration to the Disk: ~50%

)(8.0 0zzeMM −−=

2/51 )1(Gyr03.0 zMM += −&

Rvir

Toy models versus simulations (ART 50 pc):Dekel, Zolotov, Tweed, Cacciato, Ceverino, Primack

EPS approximation:

)(8.00

0zzeMM −−=Rvir

0.1 Rvir

A proxy for SFR

disk

streams

Ceverino, Dekel, Bournaud, Primack ART 35-70pc resolution

Streams Break Up - the “Messy” Region

interaction region

Higher resolution reveals smaller clumps:- Merging galaxies with DM halos- Baryonic clumps – hydro and thermal instabilities

Supersonic cold stream in a hot medium (2D)

Burkert, Ntormousi, Forbes, Dekel, Birnboim…

Supersonic cold stream in a hot medium (2D)

Burkert, Ntormousi, Forbes, Dekel, Birnboim…

6. Observing Cold Streams in Lyman alpha

Emission: Goerdt et al. 2010, Kasen et al. 2011Absorption: Fumagalli et al. 2011, Goerdt et al. 2011

ART code (Klypin, Kravtsov)Simulations: Ceverino, Dekel, Bournaud 2010

Lyman-alpha Emission (LA Blobs)

Extended source of cold H is provided by the inflowing

Radiative transport of UV & Lyα, fluorescence from stars, dustGoerdt et al. 10; Kasen, Ceverino, Fumagalli, Dekel, Prochaska, Primack

z=4.5

Mv~1012M

L~1043 erg s-1

d~100 kpc

100

kpc

provided by the inflowing (clumpy) streams

Energy is provided by: 1. inflow down the gravitational …..potential gradient 2. fluorescence by stars

Yet to be incorporated: AGN, enhanced outflows

z=4.5

Mv~1012M

L~1043 erg s-1

d~100 kpc

100

kpc

Gravity Powers Lyman-alpha Emission

∂∂

r

φMf=(r)E ccheat& 25.3

482.1

12143 )1(102.1 zMfsergE cheat +×≈ −

Half the luminosity

In AMR simulations: Vinflow~Vvir ~const. → potential gain dissipates to Lα

Half the luminosity outside 0.3Rv

LABs from galaxies at z=2-4 are inevitable Have cold streams been detected ?

Gravitational heating is generic (e.g. clusters)

Lyman-alpha from Cold streams

Goerdt, Dekel, Sternberg, Ceverino, Teyssier, Primack 09

100 kpc

Surface brightness L ~~1043-44 erg s-1

T=(1-5)x104 K n=0.01-0.1 cm-3 NHI~~1020 cm-2 pressure equilib.

Fardal et al 01; Furlanetto et al 05; Dijkstra & Loeb 09

Cold streams as Lyman-alpha BlobsGoerdt, Dekel, Sternberg, Ceverino, Teyssier, Primack 09

100 kpc

Matsuda et al 06-09

Lyman-alpha Luminosity Function

Matsuda et al 09

Isophotal area and kinematics also consistent with data

Detection of Inflow in Lya Emission

Most are red-shifted -> outflows

5% are blue-shifted -> inflow

Kulas, Shapley et al. 2011 Systemic z from Hα

Kasen et al. 12

Cold Streams as Absorption Systems

SLLS 19.0-20.3cm-2

DLA >20.3cm-2

LLS 17.2-19.0cm-2

Fumagalli, Prochaska, Kasen, Dekel, Ceverino, Primack 2011; Goerdt et al. 2012

MFP 15.5-17.2cm-2

19.0-20.3cm-2

Fumagalli, Prochaska, Kasen, Dekel, Ceverino, Primack 11

500

average line profile

9%

HI Absorption Systems

SLLS ionized-neutral

DLA neutral, thick

LLS ionized, HI thick

MFP thin-thick

SLLSLLS

DLA

MFP

Stacked absorption line profile is weak because of low sky coverage

Inflow signal consistent with observatios (Steidel et al. 10)

Inflow undetectable in metals because of low Z and coverage

Lya Image – radiative transfer

Kasen et al 11: including Lya multiple scattering, UV bkgd, Fluorescence from stars

7. Angular Momentum: Transport and Exchange

Danovich et al. 2012Pichon et al. 2012; Kim et al. 2012Pichon et al. 2012; Kim et al. 2012

In-streaming Extended Rotating Disk

- AM by transverse motion of streams – impact parameter

- Streams transport AM into the inner halo

- One stream is dominant

- Higher J/M at later times → inside-out disk buildup

100 kpc Agertz et al 09

Disk AM Buildup by Streams

disk

streams

AM Exchange in the Inner Halo

Is AM amplitude conserved to within a factor of 2?

AMdiskAM(r)Danovich, Dekel, Hahn, Teyssier 2012

Pichon et al. 2012; Kim et al. 2012

interaction region

disk

Ceverino, Dekel, Bournaud 2010 ART 50 pc resolution

Torques & AM exchange in the inner halo ~0.3Rv

AM is not conserved all the way to the disk!

Angular Momentum on Halo Scale

Only little alignment between stream plane and AM at RvAMStrPln

Most of the AM in one stream

Disk is not aligned with AM at r>0.3Rvir

AMdiskAMRv

AMdiskAM(r) AMRvAM(r)

8. Outflows

- What drives the massive outflows in massive galaxies?

– How do the outflows affect the inflows? …Need to maintain Inflow + Reservoir = SFR + Outflow

Theory Challenge: Inflow and Outflow

Tweed, Dekel, Teyssier

RAMSES 70-pc resolution

Inflows & Outflows

Outflows find their way out through the dilute medium no noticeable effect on the dense cold rapid inflows

Gas density Temperature Metallicity

no noticeable effect on the dense cold rapid inflows

dilute hot high Z

House, Tweed, Ceverino et al.

outflow η~0.7

Inflows and Outflows

in-streaming ~1.7SFR

Conclusions

- Cosmological inflow rate Mdot/M ~ 0.03 Gyr-1 (1+z)2.5 M α exp(-0.8z)

- Steady state: SFR ~ accretion rate, Mgas~const., as long as τsfr ~ ε-1tff

- Shock heating scale Mv~1012M

- cold flows / hot medium - bimodality

- High-z massive galaxies fed by cosmic-web streams, smooth & mergers

Certain robust predictions for high-z galaxies in λCDM cosmology

- ~3 co-planar streams (70%) in a pancake (20%)

- ~50% penetration into the galaxy; instabilities, messy region near disk

- Streams (mostly one) transport AM. AM exchange in the inner halo.

- Cold streams are observable in Lya emission (LAB) and absorption (LLS)

- Outflows are in harmony with the in-streams (tentative)

50 Mpc

100 kpc

100 kpc

Cosmic Web, Cold Streams, Clumpy Disks & Spheroids

stars

5 kpc 5 kpc