2022MNRAS.512.3030V

Dark matter haloes that reach the H I-cooling mass without prior star formation or external metal pollution represent potential sites for the formation of small - extremely faint - Population III galaxies at high redshifts. Gravitational lensing may in rare cases boost their fluxes to detectable levels, but to find even a small number of such objects in randomly selected regions of the sky requires very large areas to be surveyed. Because of this, a small, wide-field telescope can in principle offer better detection prospects than a large telescope with a smaller field of view. Here, we derive the minimum comoving number density required to allow gravitational lensing to lift such objects at redshift z = 5-16 above the detection thresholds of blind surveys carried out with the James Webb space telescope (JWST), the Roman space telescope (RST) and Euclid. We find that the prospects for photometric detections of Pop III galaxies are promising, and that they are better for RST than for JWST and Euclid. However, the Pop III galaxies favoured by current simulations have number densities too low to allow spectroscopic detections based on the strength of the He II1640 emission line in any of the considered surveys unless very high star formation efficiencies (ε >= 0.1) are evoked. We argue that targeting individual cluster lenses instead of the wide-field surveys considered in this paper results in better spectroscopic detection prospects, while for photometric detection, the wide-field surveys perform considerably better.