iaptoulouse 3/19/09hakim atek lyman-alpha emitters in the local universe atelier muse – toulouse,...

IAPToulouse 3/19/09 Hakim ATEK

Lyman-alpha Emitters in the Local Universe

Atelier MUSE – Toulouse, 19 mars 2009

Hakim ATEKDaniel Kunth, Matthew Hayes, Claus Leitherer, J. Miguel Mas-Hesse, Göran Östlin, Artashes Petrosian, Daniel Schaerer, Anne Verhamme


The Lyman-alpha Puzzle

Early observations with IUE:

• Ly was not detected in many cases

• A damped absorption was observed instead

• When detected, Ly was well below the theoritical value predicted by recombination theory

Which factors are actually driving the visibility of Lyman-alpha emissionin star-forming galaxies ?

This question will remain unclear after the IUE era …


The Lyman-alpha Puzzle

Attempts to solve the puzzle in the early 90’s

Charlot & Fall (1993) Giavalisco et al. (1996)

Attenuation by dust : Only very early starbursts are detectable ?

-- Ly NOT correlated with dust-- Ly emission, even after extinction correction, is still below CASE B


New Insight with HST/GHRS

High resolution spectroscopy with HST/GHRS of local BCGs:

First results by Kunth et al. (1994) :

Very low metallicity and dust deficient galaxy IZw 18 :

• A prominent Ly emission is expected for a young unevolved starburst

IZw18[O/H] =-1.66WLy <- 30Å

Suprisingly, a strong damped absorption is observed


New Insight with HST/GHRS

Similar results by Thuan and Izotov (1997) : Low metallicity BCGs SBS 0335-052 & Tol 65

A prominent Ly emission was foundin a more metallic and dusty starburst Haro 2 Lequeux et al. 1995


GHRS Sample of BCGs: The Role of Kinematics

8 local BCGs: 4 Ly emitters, 4 absorbers

• In all Ly emitters the neutral, metallic absorption lines were blueshifted by 100-400 km/s.

• In the damped systems, the neutral absorptions were always at the systemic velocities

• All Ly emission lines showed a clear P Cyg profile, indicating the presence of an expanding shell of neutral gas.

The profiles could be well fitted assuming the measured expansion velocity

Kunth et al. 1998

IAP

Lyman Break Galaxies: have P-Cyg profiles

superwinds of neutral gas at z ~ 3 : outflows are ubiquitous

Only 25% of the LBG sample show Lya in emission !!

Shapley et al. 2003

Lyman Break Galaxies

Toulouse 3/19/09 Hakim ATEK


HST/STIS: Spatial Analysis

Long slit spectroscopy with STIS To map the kinematics of the neutral gas In 3 galaxies: Mas-Hesse et al. (2003) • Ly emission extends over more than 10” (~1 kpc)

• Large expanding shell in 2 cases with An ionised front in IRAS 08+65 associated With a secondary Ly emission

• No leaking photons from IZw 18 over the whole slit

The visibility of Lya might be driven mostly by the neutral gas distribution (porosity) and kinematics


HST/ACS Imaging of Local Starburst Galaxies

Need to map the diffuse Lya emission: complementary to UV-bright targetedspectroscopic studies

Resonant scattering Emission and absorption at small scale

To infer the relative importance of the factors regulating the detectability of Lya emission

Emission“Absorption”

Haro 2 (STIS) Haro 11 (ACS)

Kunth et al. 2003


HST/ACS Imaging : Pilot Study

6 selected local galaxies : - 0.009 < z < 0.029 - 40 orbits in total- 0.03” sampling span a range of : - Lya morphology and profiles - Luminosity and metallicity - dust

Preliminary results: - Complex emission + absorption - Lya emission with no continuum counterpart - Global damped absorption

Continuum subtraction issue

-20 7.9

-19

-15

-21

-21

-17

8.7

8.8 7.9

7.6

7.3

Kunth et al. 2003

HS

T /

AC

S /

F5

50M

MB Z


HST/ACS Imaging: SBC Observations

F122M

F140LP

Haro 11 SBS 0335 IRAS 08 Tol 65 NGC 6090 ESO 338

= -2

= +1

= 0


HST/ACS Imaging

Continuum subtraction technique


HST/ACS Imaging Results: Haro 11

FUV H Ly

• Net Ly emitter• Ly does NOT resemble FUV • Ly does NOT resemble H • 90% of flux in diffuse compnt.

(Hayes et al. 2007a; Östlin et al. 2007)


HST/ACS Imaging Results: ESO 338-04

FUV H Ly

• Net Ly emitter• Ly largely symmetric around one knot• Ly uncorrelated with H


HST/ACS Imaging Results: SBS 0335-052

FUV H Ly

• Net Ly absorber• Ring-like Ly leakage • Ha follows FUV tightly• Lya almost exact mirror


HST/ACS Imaging Results: IRAS 08+65

FUV H Ly

• Strong nuclear emission • Diffuse component


HST/ACS Imaging Results: NGC 6090

FUV H Ly

• Strong + diffuse Lya from NE core• Weak diffuse surrounding SW• Brightest Ha region not the brightest in Lya


The Role of Dust in Lya Obscuration: Haro 11

• Diffuse emission component independent of the dust• Emission from knot C with E(B-V) ~ 0.4• Absorption from knot A with E(B-V) ~ 0.2• EW(Ly) vs EW(H) • Ly/H above the theoretical value (8.7 case B extinction corrected) enhanced Ly/H ratio

Neufeld, 1991


The Role of Dust in Lya Obscuration: IRAS 08+65

• Spread Ly emission over the extinction range• No correlation EW(Ly) EW(H) • Ly/H independent of the extinction

Role of the gas kinematics


The Role of Dust in Lya Obscuration: SBS 0335

• Absoprtion correlates with extinctionRole of the static column density

• EW(Ly) declines with the dust content


HST/ACS Imaging Results

Escape fraction decreases with dust No clear correlation between EW(Ly) and dust

Different evolutionnary effects :

mixed effects: EW(lya) shouldbe not correlated with age


HST/ACS Imaging Results

Star Formation Rate calibration :

What fraction of Ly photons actually escape from the starburst ? simple dust correction of Ly luminosity fails to recover the intrinsic SFR !

Atek et al. 2008

IAP

Lya from emission to absorption

- Profile well fitted with a shell model using observational constraints on Hi column density and dust

- Considering observer’s sight-line and N(H)=4x10^21 cm-2 damped absorption is reproduced with NO DUST E(B-V) = 0

- Lya absorbers : with static and high N(Hi) only a small amount of dust is needed to suppress Lya emission

- Lya emitters : in presence of outflows high N(Hi) and high dust content is needed.

A young, metal- and dust-poor galaxy shows a strong damped absorption !!

(atek et al. 2009a)

IAP

Lya from emission to absorption

Radiation transport of Lya explains the observed profile variations across the galaxy

IAP

LAEs number density

MUSE Prospects

Need to observe galaxies at higher redshift Is the evolution of fesc ratio always related to the dust content ?

Lya variations at a small scale Emission/absorption usually not local Variety of profiles (The case of P Cygni)

High spatial resolution needed (AO, space facilities)

IFU mode capabilities with high sampling

Investigate spectroscopically the ISM kinematics

complementary to GALEX sample at z ~ 0.3 (atek et al. 2009b)

Verhamme et al 2008


IAP

LAEs number density

Le Delliou et al. (2005)

Discrepancy between models and observationsnow less severe:

Past:- Completeness limit (instr. Sensitivity)- Radiation transfer effects

Now:

New instrumental capabilities and detection techniques and new models seem to converge with actualobservations

BUT models assumptions still too simple e.g :using a constant escape fraction …

LAEs number density


Nagamine et al. (2008)


Summary

Reliability of Lya emission as a COSMOLOGICAL TOOL ? Lya escape is multi-parameter process :

Intrinsic EW(Lya) (age, mass) Line broadening Static HI coverage: Lya greatly attenuated by dust Out flowing HI gas: effect of dust is reduced Multi-phase medium: Lya escape if favored

Low escape fractions (< 15%) -- dust corrections fail

New routes to SFR(Lya) calibration !!

Demonstrates the need for a detailed, statistically significant investigation !!

Instrinsic Emission

Radiation Transfer

iaptoulouse 3/19/09hakim atek lyman-alpha emitters in the local universe atelier muse – toulouse,...

Documents