Have we detected one of the sources responsible for an early reionization of the Universe?

In a recent paper Pello et al. have reported a candidate z = 10 galaxy, A1835-1916, which was found in a near-infrared survey of the central regions of the gravitational lensing cluster A1835. If this detection is confirmed and the detection rate turns out to be typical, then the volume-averaged ultraviolet emissivity must be rising rapidly with increasing redshift. For a magnification resulting from gravitational lensing by a factor M greater than or similar to 25 estimated by Pello et al., the inferred star formation rate at z = 10 would be about one order of magnitude higher than estimates of the star formation rate density at z = 6. Objects at z = 10 would contribute substantially to the total source counts at 1.6 mum and the estimated space density of sources may exceed the space density of dark matter haloes in a LambdaCDM model. We therefore argue that if A1835-1916 is indeed at z = 10 then either the magnification factor may have been overestimated or the galaxy has a top-heavy initial mass function. Sources with the UV flux and space density of A1835-1916 produce similar to33(fesc)(M/25) hydrogen ionizing photons per hydrogen atom per Hubble time, where f (esc) is the escape fraction of ionizing photons. This rate should be sufficient to reionize most of the diffuse hydrogen in the Universe at redshift 10. We further use a correlation between the equivalent width and the redshift of the Lyalpha emission line with respect to the systemic redshift observed in Lyman-break galaxies to obtain constraints on the ionization state of the surrounding intergalactic medium (IGM) from the Gunn-Peterson absorption. These constraints also argue in favour of the surrounding IGM being fully ionized. Pello et al. may thus have detected a population of sources, which is responsible for the large electron scattering optical depth indicated by the cross-power spectrum of the temperature and polarization fluctuations of the cosmic microwave background as measured by WMAP.