On the quenching of star formation in observed and simulated central galaxies: evidence for the role of integrated AGN feedback

ABSTRACT In this paper, we investigate how massive central galaxies cease their star formation by comparing theoretical predictions from cosmological simulations: EAGLE, Illustris, and IllustrisTNG with observations of the local Universe Sloan Digital Sky Survey (SDSS). Our machine learning (ML) classification reveals supermassive black hole mass (MBH) as most predictive parameter in determining whether a galaxy is forming or quenched at redshift z = 0 all three simulations. This predicted consequence active galactic nucleus (AGN) quenching reflected observations, where it true for range indirect estimates MBH via proxies well its dynamical measurements. partial correlation analysis shows that other parameters lose strong association quiescence, once correlations are accounted for. simulations, demonstrate integrated power output AGN, rather than instantaneous activity, which causes to quench. Finally, analyse change molecular gas content star-forming passive populations. We find both fractions (fgas) efficiencies (SFEs) decrease upon transition quiescence but SFE more fgas ML passive/star-forming classification. These trends SDSS closely recovered direct contrast made Illustris. conclude viable AGN feedback prescription can be achieved combination preventative turbulence injection together quench galaxies.