MNRAS 465, 3637–3655 (2017) doi:10.1093/mnras/stw2973 Advance Access publication 2016 November 17 The production and escape of Lyman-Continuum radiation from star-forming galaxies at z ∼ 2 and their redshift evolution 1‹ 1,2 3 4 3,5 Jorryt Matthee, David Sobral, Philip Best, Ali Ahmad Khostovan, Ivan´ Oteo, Downloaded from https://academic.oup.com/mnras/article-abstract/465/3/3637/2544379 by Royal Observatory Library user on 03 October 2018 Rychard Bouwens1 and Huub Rottgering¨ 1 1Leiden Observatory, Leiden University, PO Box 9513, NL-2300 RA Leiden, the Netherlands 2Department of Physics, Lancaster University, Lancaster, LA1 4YB, UK 3Institute for Astronomy, University of Edinburgh, Royal Observatory, Blackford Hill, Edinburgh EH9 3HJ, UK 4University of California, Riverside, 900 University Ave, Riverside, CA 92521, USA 5European Southern Observatory, Karl-Schwarzschild-Str. 2, D-85748 Garching, Germany Accepted 2016 November 15. Received 2016 November 15; in original form 2016 May 25 ABSTRACT We study the production rate of ionizing photons of a sample of 588 Hα emitters (HAEs) and 160 Lyman-α emitters (LAEs) at z = 2.2 in the COSMOS field in order to assess the implied emissivity from galaxies, based on their ultraviolet (UV) luminosity. By exploring the rest-frame Lyman Continuum (LyC) with GALEX/NUV data, we find fesc < 2.8 (6.4) per cent through median (mean) stacking. By combining the Hα luminosity density with intergalactic medium emissivity measurements from absorption studies, we find a globally averaged fesc . +14.5 = of 5 9−4.2 per cent at z 2.2 if we assume HAEs are the only source of ionizing photons. We find similarly low values of the global fesc at z ≈ 3–5, also ruling out a high fesc at z < 5. These low escape fractions allow us to measure ξ ion, the number of produced ionizing photons per unit UV luminosity, and investigate how this depends on galaxy properties. We 24.77 ± 0.04 −1 25.14 ± 0.09 −1 find a typical ξ ion ≈ 10 Hz erg for HAEs and ξ ion ≈ 10 Hz erg for LAEs. LAEs and low-mass HAEs at z = 2.2 show similar values of ξ ion as typically assumed in the reionization era, while the typical HAE is three times less ionizing. Due to an increasing ξ ion with increasing EW(Hα), ξ ion likely increases with redshift. This evolution alone is fully in line with the observed evolution of ξ ion between z ≈ 2 and 5, indicating a typical value of 25.4 −1 ξ ion ≈ 10 Hz erg in the reionization era. Key words: galaxies: evolution – galaxies: high-redshift – cosmology: observations – dark ages, reionization, first stars. Assessing whether galaxies have been the main provider of 1 INTRODUCTION ionizing photons at z 5 (alternatively to active galactic nuclei, One of the most important questions in galaxy formation is whether AGNs; e.g. Madau & Haardt 2015; Giallongo et al. 2015;Weigel galaxies alone have been able to provide the ionizing photons et al. 2015) crucially depends on (i) precise measurements of the which reionized the Universe. Optical depth measurements from number of galaxies at early cosmic times, (ii) the clumping factor the Planck satellite place the mean reionization redshift between of the intergalactic medium (IGM, e.g. Pawlik, Schaye & Dalla z ≈ 7.8 and 8.8 (Planck Collaboration XLVII et al. 2016). The Vecchia 2015), (iii) the amount of ionizing photons that is produced end point of reionization has been marked by the Gun–Peterson (Lyman-Continuum photons, LyC, λ<912 Å) and (iv) the fraction trough in high-redshift quasars at z ≈ 5–6, with a typical neutral of ionizing photons that escapes into the IGM. All these numbers fraction of ∼10−4 (e.g. Fan et al. 2006; McGreer, Mesinger & are currently uncertain, with the relative uncertainty greatly rising D’Odorico 2015). Moreover, recent observations indicate that there from (i) to (iv). are large opacity fluctuations among various sightlines, indicating Many studies so far have focused on counting the number of an inhomogeneous nature of reionization (Becker et al. 2015). galaxies as a function of their ultraviolet (UV) luminosity (luminos- ity functions) at z > 7 (e.g. McLure et al. 2013; Bowler et al. 2014; Atek et al. 2015; Bouwens et al. 2015a; Finkelstein et al. 2015; Ishigaki et al. 2015; McLeod et al. 2015; Castellano et al. 2016; E-mail: [email protected] Livermore, Finkelstein & Lotz 2016). These studies typically infer C 2016 The Authors Published by Oxford University Press on behalf of the Royal Astronomical Society 3638 J. Matthee et al. luminosity functions with steep faint-end slopes, and a steepening information on the production rate of LyC photons and where it is of the faint-end slope with increasing redshift (see for example, the possible to obtain a complete selection of star-forming galaxies. recent review from Finkelstein 2015), leading to a high number of In this paper, we use a large sample of Hα emitters (HAEs) and faint galaxies. Assuming ‘standard’ values for the other parameters Lyα emitters (LAEs) at z = 2.2 to constrain fesc and measure ξ ion such as the escape fraction, simplistic models indicate that galax- and how this may depend on galaxy properties. Our measurements ies may indeed have provided the ionizing photons to reionize the of ξ ion rely on the assumption that fesc is negligible (<10 per cent), Universe (e.g. Madau, Haardt & Rees 1999;Robertsonetal.2015), which we validate by constraining fesc with archival GALEX NUV and that the ionizing background at z ∼ 5 is consistent with the imaging and by comparing the estimated emissivity of HAEs with Downloaded from https://academic.oup.com/mnras/article-abstract/465/3/3637/2544379 by Royal Observatory Library user on 03 October 2018 derived emissivity from galaxies (Choudhury et al. 2015; Bouwens IGM emissivity measurements from quasar absorption lines (e.g. et al. 2015b). However, without validation of input assumptions Becker & Bolton 2013). Combined with rest-frame UV photometry, regarding the production and escape of ionizing photons (for exam- accurate measurements of ξ ion are possible on a source by source ple, these simplistic models assume that the escape fraction does not basis for HAEs, allowing us to explore correlations with galaxy depend on UV luminosity), the usability of these models remains properties. Since only a handful of LAEs are detected in Hα (see to be evaluated. Matthee et al. 2016), we measure the median ξ ion from stacks of The most commonly adopted escape fraction of ionizing photons, LAEs from Sobral et al. (2016a). fesc, is 10–20 per cent, independent of mass or luminosity (e.g. Mitra, We describe the galaxy sample and definitions of galaxy proper- Choudhury & Ferrara 2015;Robertsonetal.2015). However, hy- ties in Section 2. Section 3 presents the GALEX imaging. We present drodynamical simulations indicate that fesc is likely very anisotropic upper limits on fesc in Section 4. We indirectly estimate fesc from the and time dependent (Cen & Kimm 2015;Maetal.2015). An es- Hα luminosity function and the IGM emissivity in Section 5 and cape fraction which depends on galaxy properties (for example, a measure the ionizing properties of galaxies and its redshift evolution higher fesc for lower mass galaxies, e.g. Paardekooper, Khochfar in Section 6. Section 7 discusses the implications for reionization. & Dalla Vecchia 2015) would influence the way reionization hap- Finally, our results are summarized in Section 8. We adopt a CDM −1 −1 pened (e.g. Sharma et al. 2016). Most importantly, it is impossible cosmology with H0 = 70 km s Mpc , M = 0.3 and = 0.7. to measure fesc directly at high redshift (z > 6) because of the high Magnitudes are in the AB system. At z = 2.2, 1 arcsec corresponds opacity of the IGM for ionizing photons (e.g. Inoue et al. 2014). to a physical scale of 8.2 kpc. Furthermore, to estimate fesc it is required that the intrinsic amount of ionizing photons is measured accurately, which requires accurate understanding of the stellar populations, star formation rate (SFR) 2 GALAXY SAMPLE and dust attenuation (i.e. De Barros et al. 2016). We use a sample of Hα selected star-forming galaxies from Nevertheless, several attempts have been made to measure fesc, the High-z Emission Line Survey (HiZELS; Geach et al. 2008; both in the local Universe (e.g. Leitherer et al. 1995; Deharveng Sobral et al. 2009)atz = 2.2 in the COSMOS field. These galaxies et al. 2001; Leitet et al. 2013; Alexandroff et al. 2015)andat were selected using narrow-band (NB) imaging in the K band with intermediate redshift, z ∼ 3, where it is possible to observe red- the United Kingdom InfraRed Telescope. HAEs were identified shifted LyC radiation with optical CCDs (e.g. Inoue, Iwata & among the line emitters using BzK and BRU colours and photomet- Deharveng 2006; Boutsia et al. 2011; Vanzella et al. 2012; Bergvall ric redshifts, as described in Sobral et al. (2013), and thus have a et al. 2013;Mostardietal.2015). However, the number of reliable photometric redshift of z = 2.22 ± 0.02 where the error is due to direct detections is limited to a handful, both in the local Uni- the width of the NB filter. In total, there are 588 HAEs at z = 2.2 in verse and at intermediate redshift (e.g. Borthakur et al. 2014;Izotov COSMOS.1 et al. 2016a,b; De Barros et al. 2016; Leitherer et al. 2016), and HAEs are selected to have EW0,Hα+[NII] > 25 Å. Since the COS- strong limits of fesc 5–10 per cent exist for the majority (e.g.