9/19/2014 Claus Leitherer: Lyman Continuum Leakage 1 Lyman Continuum Leakage in the Local Universe Claus Sanch Tim Janice Sally Roderik Leitherer Borthakur.

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 1

Lyman Continuum Leakage in the Local Universe

Claus Sanch Tim Janice Sally Roderik Leitherer Borthakur Heckman Lee Oey Overzier STScI JHU JHU STScI UMich ON

9/19/2014 2Claus Leitherer:

Lyman Continuum Leakage

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 3

o Can normal (non-AGN) galaxies account for the ionizing photons required for re-ionization?

o Since H I is very effective in absorbing Lyman continuum photons, only a small fraction of the available photons will escape from the galaxy.

o What is fesc?

o Cosmological models suggest fesc ≈ 20% is needed for reionization.

Science Question

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 4

o Local galaxies are not necessarily representative of the galaxy population at high redshift.

o However, they are excellent training sets for studying which factors determine fesc.

Galaxy Sample

Redshift Range

Spatial Information

Solid Angle

Ancillary Data

Observational Challenge

Sample Definition

IGM Properties

~0.0

~1

>3

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 5

Sample 1: starburst galaxies with outflows at z ≈ 0

Lyman Continuum Leakage Candidates

Sample 2: Lyman-break analogs with weak Hα emission at z ≈ 0.2

Zackrisson et al. (2013)

Lyman Continuum Escape in Local Starburst Galaxies with Outflows

o Pilot study as a proof of concept; therefore the sample is driven by instrument requirements

o Preexisting UV spectra with F(1100) > 11014

o v > 10,000 km s1 > 30 Å

o Presence of O stars from UV spectra

9/19/2014 6Claus Leitherer:

Lyman Continuum Leakage

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 7

o The three galaxies were observed before with FUSE (Pellerin & Robert 2007; Grimes et al. 2009; Leitet et al. 2013)

o Several studies specifically targeted the Lyman continuum

Galaxy Sample

Galaxy 

Type 

E(BV)MW

 (mag)vLSR

(km s1)D

(Mpc)MB

(mag)E(BV)int

(mag) 12+log O/H SFRUV

(M⊙yr1)

Tol 0440-381 HII 0.014 12,249 167 20.2 0.17 8.2 4

Tol 1247-232 HII 0.077 14,390 207 21.0 0.09 8.1 11

Mrk 54 BCG 0.013 13,450 191 22.2 0.14 8.6 12

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 8

Grimes et al. (2009): Tol 0440-381

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 9

o ISM lines in FUSE spectra are blueshiftedo Large-scale outflows driven by SNe and stellar windso Correlates with SFR/Mo Inhomogeneous ISM creates favorable conditions for

Lyman continuum escape

9/19/2014 10Claus Leitherer:

Lyman Continuum Leakage

9/19/2014 11Claus Leitherer:

Lyman Continuum Leakage

9/19/2014 12Claus Leitherer:

Lyman Continuum Leakage

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 13

o Spectral region around the Lyman break

o Corrected for intrinsic reddening

o Restframeo ~35 Å of Lyman

continuumo Geocoronal

Lyman lines denoted

o Dashed: modelo T=20 Myr, const.

SF, Kroupa IMF, Geneva models with rotation

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 14

Lyman continuum escape fractions

Escape Fraction Tol 0440-381 Tol 1247-232 Mrk 54

fesc(gas only) 0.265±0.235 0.075±0.088 0.055±0.100fesc(gas+dust) 0.015±0.014 0.017±0.019 0.005±0.009

o Total escape fractions are of order 1% or lesso Consistent with optical depth and covering

factor suggested by CII 1036

Local Lyman-break Analogs as Candidates for Leakage

o Heckman et al. (2005): GALEX discovery of a sample of UV-bright galaxies with properties similar to those of LBG’s Lyman-break analogs (LBA’s)

o L ≈ 1011 L⊙

o SFR ≈ 101 – 102 M⊙yr1

o z ≈ 0.2

9/19/2014 15Claus Leitherer:

Lyman Continuum Leakage

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 16

Overzier et al. (2009): HST UV+ optical imaging actively star-forming galaxies

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 17

o Heckman et al. (2011): HST COS spectroscopy of a subsample with compact cores

Properties of J0921+4509

o Hα derived SF rate significantly lower than UV+IR derived rate density bounded or inhomogeneous?

o Lyα profile shows blueshifted emission; models suggest low due to small covering factor

o IS absorption lines are optically thick, yet have residual intensity picket-fence model with covering factor of 60%

9/19/2014 18Claus Leitherer:

Lyman Continuum Leakage

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 19

o COS spectrum of J0921+4509 (Borthakur et al. 2014)o Young starburst dominated by OB stars; consistent with

profile fits to CIV 1550, NV 1240, and OVI 1035o Predict the expected Lyman-break from models

absolute fesc

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 20

• Borthakur et al. (2014):

• Comparison of the observed break to models

• Dashed: 912+/912-

• Observed: 0.9 dex

• Model: 0.5 dex

• fesc = 0.37± 0.05

• Accounting for dust: fesc ≈ 1%

Take-away Pointso COS is well suited for Lyman continuum

observations in the local universe.o Local star-forming galaxies are excellent

training sets studying Lyman leakage.o The observed local star-forming galaxies

have dust corrected escape fractions of 1% or less.

o In the absence of dust absorption, the galaxies would have fesc compatible with the value required for reionization

9/19/2014 21Claus Leitherer:

Lyman Continuum Leakage

Back-up Slides

9/19/2014 22Claus Leitherer:

Lyman Continuum Leakage

Reliability of the Models

o Model predicts an intrinsic Lyman break of ~0.4 dex for a standard population

o Kroupa IMF, constant star formation, T=20 Myr, Geneva evolution models with rotation, spherically extended, expanding, blanketed non-LTE atmospheres

o How do model assumptions and uncertainties affect the prediction?

9/19/2014 23Claus Leitherer:

Lyman Continuum Leakage

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 24

Varying the duration of the SF episode has little effect for T>5 Myr as massive stars have reached equilibrium

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 25

Varying the IMF exponent between 1.6 and 3 changes the flux level by a factor of ~100 but has a small effect on the break itself

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 26

Extended model atmospheres accounting for winds decrease the break by up to 0.15 dex

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 27

Evolution models with rotation have hotter, more luminous stars and increase the Lyman continuum flux by ~0.2 dex

9/19/2014Claus Leitherer:

Lyman Continuum Leakage 28

Chemical composition has a surprisingly small influence due to the compensating effects of atmospheres and evolution models