Erratum: The spectral evolution of the first galaxies. III. Simulated james webb space telescope spectra of reionization-epoch galaxies with lyman-continuum leakage (Astrophysical Journal (2017) 836 (78) DOI: 10.3847/1538-4357/836/1/78)

Due to a mistake, the BPASS v2.0 binary-star model (Stanway et al. 2016) used in Figure 6(b) of the published article featured a different initial mass function (IMF) than the Geneva single-star model used for comparison. While the Geneva model used the Kroupa universal IMF (Kroupa 2001) with upper slope α = 2.3 (dN dM ∝ M-α) at 0.5 ≤ M(M⊙) ≤ 100, the BPASS model used a slope of α = 2.0 throughout the same mass interval. Hence, in its original form, that figure actually illustrated the combined effect of assuming binary stars and adopting a more extreme IMF. This resulted in a greater number of massive stars and a boosted ionizing flux for the BPASS model. In the revised Figure 6(b), we show the corrected comparison with identical IMF slopes (α = 2.3). In this case, the shift between the two sets of simulated galaxies in the EW(Hβ)-β diagram is more modest than in the published article. The galaxies simulated using the BPASS binary-star model are shifted by an average Δ(EW(Hβ)) ≈ 0.2 compared to galaxies simulated using the Geneva single-star model. The difference in UV slope β between the two sets of simulated galaxies is also smaller in the corrected version (Δ(β) < 0.1 on average). The risk of misclassifying high-fesc objects due using a stellar evolutionary model not representative of z > 7 galaxies is therefore somewhat smaller than indicated in the published article. (Figure Presented).