2023MNRAS.519.4608P

ALMA observations have revealed that many high-redshift galaxies are surrounded by extended (10-15 kpc) [C II]-emitting haloes that are not predicted by even the most advanced zoom-in simulations. Using a semi-analytical model, in a previous work we suggested that such haloes are produced by starburst-driven, catastrophically cooling outflows. Here, we further improve the model and compare its predictions with data from seven star-forming ($10\lesssim \rm SFR/{\rm M}_{\odot }\, yr^{-1}\lesssim 100$) galaxies at z = 4-6, observed in the ALPINE survey. We find that (a) detected [C II] haloes are a natural by-product of starburst-driven outflows; (b) the outflow mass loading factors are in the range 4 ≲ η ≲ 7, with higher η values for lower mass, lower star formation rate systems, and scale with stellar mass as $\eta \propto M_*^{-0.43}$, consistently with the momentum-driven hypothesis. Our model suggests that outflows are widespread phenomena in high-z galaxies. However, in low-mass systems the halo extended [C II] emission is likely too faint to be detected with the current levels of sensitivity.