Star Formation in Damped Lyman alpha Systems

Star Formation in Damped Lyman alpha Star Formation in Damped Lyman alpha SystemsSystems

Art Wolfe

Collaborators:

J.X. Prochaska, J. C. Howk, E.Gawiser, and K. Nagamine

DLAS are

•Definition of DLA: N(HI) > 2*1020 cm-2

•Distinguishing characteristic of DLAs : Gas is Neutral

DLAS are

•Definition: N(HI) > 2*1020 cm-2

•Distinguishing characteristic of DLAs : Gas is Neutral

How are DLAs heated?

•DLAs Dominate the Neutral Gas Content of the Universe at z=[0,5]

•Gas Content of DLAs at z=[3,4] Accounts for current visible Mass

•DLAs Serve as Important Neutral Gas Reservoirs for Star Formation

Relevance of DLAs for Star FormationRelevance of DLAs for Star Formation

•DLAs Dominate the Neutral Gas Content of the Universe at z=[0,5]

•Gas Content of DLAs at z=[3,4] Accounts for current visible Mass

•DLAs Serve as Important Neutral Gas Reservoirs for Star Formation

Prochaska,Herbert-Fort,& Wolfe ‘05

•DLAs Dominate the Neutral Gas Content of the Universe at z=[0,5]

•Gas Content of DLAs at z=[3,4] Accounts for current visible Mass

•DLAs Serve as Important Neutral Gas Reservoirs for Star Formation

CurrentVisibleMatter

NeutralGas atHigh z

Evidence for Star Formation in DLAs?

• Direct Detection of Starlight

• Increase of Metallicity with time

• Evidence for Feedback between Stars and Absorbing Gas

SFRs Inferred from DLA EmissionSFRs Inferred from DLA Emission

DLA Redshift Method SFR (My-

1)Ref.

0458-02 2.040 Ly-alpha > 1.5 Moller etal 04

0953+47 3.407 Ly-alpha 0.8 to 7.0 Bunker etal 05

2206-19A 1.921 Cntuum. 25 to 50 Moller etal 02

1210+17 1.892 H-alpha < 5.0 Kularni etal 01

1244-34 1.859 H-alpha < 1.6 Kulkarni etal 00

Comparison between DLA and LBG SFRs

• LBG SFRs between 3 and

100 solar masses per year

• A few DLAs located at either end of LBG distribution

•What is SFR Distribution

For a fair sample of DLAs?

CII* Technique for Measuring SFRs in DLAs

OutlineOutline

• Heating and Cooling of DLAs

• Inferring SFRs per unit Area from CII* Absorption

• Global Constraints

SFRs per unit Comoving Volumne

Background Radiation

• Relationship Between DLAs and LBGs

FUV Photon

Ionizing PhotonGrain

Grain Photoeletric Heating of Neutral Gas in DLAS

H II Region

Electron

Thermal Balance in DLAsThermal Balance in DLAs

Obtaining Cooling Rates from CII* AbsorptionObtaining Cooling Rates from CII* Absorption

• [C II] 158 micron transition dominates cooling of neutral gas in Galaxy ISM

• Spontaneous emission rate per atom lc=n[CII] obtained from strength of 1335.7 absorption and Lyman alpha absorption

• Thermal equilibrium condition lc= pe gives heating rate per atom

2121)IH(*)IIC(

][ ~ Ahn NN

IIC ν 2121)IH(*)IIC(

][ ~ Ahn NN

IIC ν

HIRES Velocity Profiles of Metal-Rich DLAHIRES Velocity Profiles of Metal-Rich DLA

• Multi-component structure of absorbing gas

• Velocity Structure of CII* and Resonance lines are similar

• Strength of CII* Absorption gives heating rate of the neutral gas

[C II] 158 micron Emission rates vs N(H I)[C II] 158 micron Emission rates vs N(H I)

• Median lc=10-26.6 ergs s-1 H-1 for positive Detections

• Upper limits tend to have low N(H I)

• DLA lc values about 30 times lower than for Galaxy: explained by lower dust content and similar SFRs per unit area

[C II] 158 micron Emission rates vs N(H I)[C II] 158 micron Emission rates vs N(H I)

Critical Emission Rate

Are DLAs Heated by Background RadiationAlone?

Thermal Equilibria: lThermal Equilibria: lcc versus density versus density

DLAs with Detected N(CII*)

lc versus ndiagrams

Thermal Equilibria with local FUV Radiation Thermal Equilibria with local FUV Radiation IncludedIncluded

Thermal Equilibria with local FUV Radiation Thermal Equilibria with local FUV Radiation IncludedIncluded

Two-Phase Models of DLAs with Positive DetectionsTwo-Phase Models of DLAs with Positive Detections

WNM

WNM

CNM

•“CNM Model”

•“WNM Model”

DLAs with Upper LimitsOn N(CII*):

lc versus n diagrams

WNM Phase Model for DLAs with Upper LimitsWNM Phase Model for DLAs with Upper Limits

WNM

Multi-phase Models and SFRsMulti-phase Models and SFRs

DLAs with lc > 10-27.1

1. CII* Forms in CNM Phase: moderate SFR/Area

2. CII* Forms in WNM Phase: high SFR/Area

DLAs with lc < 10-27.1

1. CII* Forms in WNM Phase: Background Heating Alone

2. CII* Forms in WNM Phase: moderate SFR/Area

Multi-phase Models and SFRsMulti-phase Models and SFRs

DLAs with lc > 10-27.1

1. CII* Forms in CNM Phase: moderate SFR/ H I Area

2. CII* Forms in WNM Phase: high SFR/ H I Area

DLAs with lc < 10-27.1

1. CII* Forms in WNM Phase: Background Heating Alone

2. CII* Forms in WNM Phase: moderate SFR/ H I Area

SFR or Luminosity perSFR or Luminosity per unit Comoving volumeunit Comoving volume

Observed

De-reddened

Giavalisco etal ‘04

Global ConstraintsGlobal Constraints

Consequences of LBG ConstraintsConsequences of LBG Constraints

• Most DLA models predict νDLA >> ν

LBG: high JνCII*

• This rules out models with inefficient heating

-All models where CII* absorption occurs in WNM -Models where CII* absorption occurs in CNM gas heated by FUV radiation incident on large grains

• Even with efficient heating, νDLA =ν

LBG

• Strong overlap between DLAs and LBGs

1. DLAs with lc >10-27.1 ergs s-1 H-1

DLAs

LBGs in DLAs with LBGs in DLAs with llcc > > 1010-27.1-27.1 ergs s ergs s-1 -1 HH-1-1

LBG

Dust

DLA

DLA Gas May Replenish LBG Star DLA Gas May Replenish LBG Star Formation ActivityFormation Activity

• LBG Star Formation Rate Requires “Fuel”

• DLA Gas would sustain SFRs for ~ 2 Gyr.

• Replenishment from IGM may be required

Supporting Evidence for this ScenarioSupporting Evidence for this Scenario

1. Detection of DLA absorption in an LBG

2. Evidence for DLA-LBG cross correlation

3. Evidence for Grain photoelelctric heating

4. Independent Evidence for CNM Gas

An LBG Galaxy Associated with a DLA (Moller etal ‘02)An LBG Galaxy Associated with a DLA (Moller etal ‘02)

•SFR=25 to 50 Myr-1

An LBG Galaxy Associated with a DLA (Moller etal ‘02)An LBG Galaxy Associated with a DLA (Moller etal ‘02)

8.4 kpc

LyEmission [O III] Emission

Preliminary DLA-LBG Cross-Correlation FunctionPreliminary DLA-LBG Cross-Correlation Function(Cooke etal 2005)(Cooke etal 2005)

LBG-DLALBG-DLA:

r=4.25,=1.11

LBG-LBG:LBG-LBG:

r=3.96,=0.15

Mpc

Mpc

Nature of DLAs with lNature of DLAs with lcc < 10 < 10-27.1 -27.1 ergs s ergs s-1 -1 HH- 1- 1

Low CII* Absorption

ImplicationsImplications• Local Source of Heat Input Required for the 40% of DLAs with lc > 10-

27.1 ergs s-1 H-1

• These DLAs likely heated by attenuated FUV radiation emitted by embedded LBG.

• In these DLAs, gas producing CII* absorption is CNM.

• Background Radiation heats the 60% of DLAs with lc < 10-27.1 ergs s-1 H-1. Gas is WNM.

• LBGs may be in subset of DLAs in which starburst activity occurs. DLA gas may fuel star formation

DLA Age-Metallicity RelationshipDLA Age-Metallicity Relationship

• Sub-solar metals at all z

• Statistically Significant evidence for increase of metals with time

• Most DLAs detected at epochs prior to formation of Milky Way Disk

• Mixed Evidence for star formation

Incidence of DLAs per unit Absorption Incidence of DLAs per unit Absorption DistanceDistance

Equivalence Between Bulge & Uniform Disk ScenariosEquivalence Between Bulge & Uniform Disk Scenarios

•Disk ScenarioSource

Field

•Bulge Scenario

Source

Field

•Mean Intensities: JνB=Jν

D if LνH I the same•Comoving Luminosity Densities, ν

B=νD

Bolometric Backgrounds at z=0 due to Bolometric Backgrounds at z=0 due to Sources at z > zSources at z > zminmin

Multi-phase Diagram for Typical DLAMulti-phase Diagram for Typical DLA

Evidence Against WNM gas in a DLAEvidence Against WNM gas in a DLA

• SiII* Absorption sensitive to warm gas

• Absence of SiII* Absorption implies T < 800 K for CII* Gas

Evidence for Grain Photoelectric HeatingEvidence for Grain Photoelectric Heating

• Statistically significant correlation between lc and dust-to-gas ratio

• Solid curves are lines of constant Jν

• Upper limits are at lowLow dust-to-gas ratios