[abridged] Among the different observational techniques used to select high-redshift galaxies, the hydrogen recombination line Lyman-alpha (Lya) is of particular interest as it gives access to the measurement of cosmological quantities such as the st
ar formation rate of distant galaxy populations. However, the interpretation of this line and the calibration of such observables is still subject to serious uncertainties. Therefore, it important to understand under what conditions the Lya line can be used as a reliable star formation diagnostic tool. We use a sample of 24 Lya emitters at z ~ 0.3 with an optical spectroscopic follow-up to calculate the Lya escape fraction and its dependency upon different physical properties. We also examine the reliability of Lya as a star formation rate indicator. We combine these observations with a compilation of Lya emitters selected at z = 0 - 0.3 to assemble a larger sample. The Lya escape fraction depends clearly on the dust extinction following the relation fesc(Lya) = C(Lya) x 10^(-0.4 E(B-V) k(Lya)), but with a shallower slope than previously reported, with k(Lya) ~ 6.67 and C(Lya) = 0.22. However, the correlation does not follow the expected curve for a simple dust attenuation. We explore the various mechanisms than lead to fesc(Lya) values above the continuum extinction curve, i.e. to an enhancement of the Lya output. We also observe that the strength of Lya and the escape fraction appear unrelated to the galaxy metallicity. Regarding the reliability of Lya as a star formation rate (SFR) indicator, we show that the deviation of SFR(Lya) from the true SFR (as traced by the UV continuum) is a function of the observed SFR(UV), which can be seen as the decrease of fesc(Lya) with increasing UV luminosity. Moreover, we observe a redshift-dependence of this relationship revealing the underlying evolution of fesc(Lya) with redshift.
We present reduced and calibrated high resolution Lyman-alpha (Lya) images for a sample of 6 local star forming galaxies. Targets were selected to represent a range in luminosity and metallicity and to include both known Lya emitters and non-emitters
. Far ultraviolet imaging was carried out with the Solar Blind Channel of the ACS on HST in the F122M (Lya on-line) and F140LP (continuum) filters. The resulting Lya images are the product of careful modeling of both the stellar and nebular continua, facilitated by supporting HST imaging in Ha and 5 continuum bands, combined with Starburst99 evolutionary synthesis models, and prescriptions for dust extinction on the continuum. In all, the resulting morphologies in Lya, Ha, and UV-continuum are qualitatively very different and we show that the bulk of Lya emerges in a diffuse component resulting from resonant scattering events. Lya escape fractions, computed from integrated Ha luminosities and recombination theory, are found never to exceed 14%. Internal dust extinction is estimated in each pixel and used to correct Lya fluxes. However, the extinction corrections are far too small (factors from 2.6 to infinity) to reconcile the global Lya luminosities with recombination theory. Surprisingly, when comparing the global equivalent widths of Lya and Ha, the two quantities appear anti-correlated, which may be due to the evolution of mechanical feedback. This calls for caution in the interpretation of Lya observations. The images presented have a physical resolution 3 orders of magnitude better than attainable at high-z from the ground with current instrumentation and our images may therefore serve as useful templates for comparing with observations and modeling of primeval galaxy formation. We therefore provide the reduced Lya, Ha, and continuum images to the community.
The Lyman-alpha (Lya) recombination line is a fundamental tool for galaxy evolution studies and modern observational cosmology. However, subsequent interpretations are still prone to a number of uncertainties. Besides numerical efforts, empirical dat
a are urgently needed for a better understanding of Lya escape process. We empirically estimate the Lyman-alpha escape fraction fesc(Lya) in a statistically significant sample of z ~ 0 - 0.3 galaxies in order to calibrate high-redshift Lyman-alpha observations. An optical spectroscopic follow-up of a sub-sample of 24 Lyman-alpha emitters (LAEs) detected by GALEX at z ~ 0.2-0.3, combined with a UV-optical sample of local starbursts, both with matched apertures, allow us to quantify the dust extinction through Balmer lines, and to estimate the Lyman-alpha escape fraction from the Halpha flux corrected for extinction in the framework of the recombination theory. The global escape fraction of Lyman-alpha radiation spans nearly the entire range of values, from 0.5 to 100 %, and fesc(Lya) clearly decreases with increasing nebular dust extinction E(B-V). Several objects show fesc(Lya) greater than fesc(continuum) which may be an observational evidence for clumpy ISM geometry or for an aspherical ISM. Selection biases and aperture size effects may still prevail between z ~ 0.2-0.3 LAEs and local starbursts, which may explain the difference observed for fesc(Lya).
Lyman-alpha is now widely used to investigate the galaxy formation and evolution in the high redshift universe. However, without a rigorous understanding of the processes which regulate the Lya escape fraction, physical interpretations of high-z obse
rvations remain questionable. We examine six nearby star-forming galaxies to disentangle the role of the dust from other parameters such as gas kinematics, geometry and ISM morphology in the obscuration of Ly-alpha. Thereby we aim to understand the Ly-a escape physics and infer the implications for high-redshift studies. We use HST/ACS to produce continuum-subtracted Lya maps, and ground-based observations (ESO/NTT and NOT) to map the Halpha emission and the extinction E(B-V) in the gas phase derived from the Balmer decrement Halpha/Hbeta. When large outflows are present, the Lya emission appears not to correlate with the dust content, confirming the role of the HI kinematics in the escape of Lya photons. In the case of a dense, static HI covering, we observe a damped absorption with a declining relationship between Lya and E(B-V). We found that the Lya escape fraction does not exceed 10% in all our galaxies and is mostly about 3% or below. Finally, because of the radiative transfer complexity of the Lya line, star formation rate based on Lya luminosity is underestimated with respect to that derived from UV luminosity. The failure of simple dust correction to recover the intrinsic Lya/Ha ratio or the total star formation rate should prompt us to be more cautious when interpreting high-z observations and related properties, such as SFRs based on Lya alone. To this end we propose a more realistic calibration for SFR(Lya) which accounts for dust attenuation and resonant scattering effects via the Lya escape fraction.
