Semi-empirical Predictions for Ultra-deep Radio Counts of Star-forming Galaxies with the SKAO

Abstract

Star-forming galaxies (SFGs) dominate the faint radio sky at flux densities below 0.1 mJy. Identifying these systems through a multiwavelength approach is essential to tracing the cosmic history of star formation. Upcoming surveys with the Square Kilometre Array Observatory (SKAO) in its AA4 configuration for the Mid array will probe these faint populations, offering unprecedented insights into the star formation activity of galaxies across cosmic time. Semi-empirical models, built on minimal assumptions and empirical galaxy relations, provide an efficient framework to study galaxy evolution using recent radio and optical/near-infrared (NIR) data. We developed SEMPER (Semi-EMPirical model for Extragalactic Radio emission) to predict the radio luminosity functions and number counts of SFGs. SEMPER combines redshift-dependent stellar mass functions from deep NIR surveys with empirical relations such as the galaxy main sequence and the IR/radio correlation, to characterise the radio properties of massive, high-redshift galaxies. The model shows excellent agreement with recent deep radio observations and naturally predicts a substantial population of massive, dust-obscured galaxies already in place at early epochs. In this chapter, we extend the SEMPER framework to SKA surveys by including an evolving starburst fraction and computing differential number counts at 1.4 GHz for both lensed and unlensed SFGs. Furthermore, we predict the cosmic star formation rate density (SFRD) traced by radio-emitting galaxies up to $z\approx10$. Our results show that SKA surveys will probe the faintest flux-density regimes, dominated by galaxies powered by star formation, and that <20 hours of SKA-Mid Band 2 observations will recover at least $\approx$20% of the total SFRD predicted by SEMPER, including contributions from optically/NIR-dark systems up to $z\approx 6$.