# Radio continuum size evolution of star-forming galaxies over 0.35 < z < 2.25

## TL;DR

This study investigates how the radio sizes of star-forming galaxies evolve from redshift 0.35 to 2.25, revealing size stability over cosmic time and differences between main-sequence and starburst galaxies, with implications for understanding galaxy growth.

## Contribution

It provides the first systematic analysis of radio continuum size evolution of SFGs over 0.35<z<2.25, highlighting size constancy and the role of starburst phases in galaxy evolution.

## Key findings

- Median radio size remains nearly constant over cosmic time.
- Main-sequence galaxies are generally more extended than starbursts.
- Size of stellar component is larger than radio emission, indicating concentrated star formation.

## Abstract

We present the first systematic study of the radio continuum size evolution of star-forming galaxies (SFGs) over the redshift range $0.35<z<2.25$. We use the VLA COSMOS 3GHz map (noise $\rm rms=2.3\,\mu Jy \,beam^{-1}$, $\theta_{\rm beam}=0.75\,\rm arcsec$) to construct a mass-complete sample of 3184 radio-selected SFGs that reside on and above the main-sequence (MS) of SFGs. We find no clear dependence between the radio size and stellar mass, $M_{\star}$, of SFGs with $10.5\lesssim\log(M_\star/\rm M_\odot)\lesssim11.5$. Our analysis suggests that MS galaxies are preferentially extended, while SFGs above the MS are always compact. The median effective radius of SFGs on (above) the MS of $R_{\rm eff}=1.5\pm0.2$ ($1.0\pm0.2$) kpc remains nearly constant with cosmic time; a parametrization of the form $R_{\rm eff}\propto(1+z)^\alpha$ yields a shallow slope of only $\alpha=-0.26\pm0.08\,(0.12\pm0.14)$ for SFGs on (above) the MS. The size of the stellar component of galaxies is larger than the extent of the radio continuum emission by a factor $\sim$2 (1.3) at $z=0.5\,(2)$, indicating star formation is enhanced at small radii. The galactic-averaged star formation rate surface density $(\Sigma_{\rm SFR})$ scales with the distance to the MS, except for a fraction of MS galaxies ($\lesssim10\%$) that harbor starburst-like $\Sigma_{\rm SFR}$. These "hidden" starbursts might have experienced a compaction phase due to disk instability and/or merger-driven burst of star formation, which may or may not significantly offset a galaxy from the MS. We thus propose to jointly use $\Sigma_{\rm SFR}$ and distance to the MS to better identify the galaxy population undergoing a starbursting phase.