We present an analysis of the quenching of star formation in galaxies, bulges, and disks throughout the bulk of cosmic history, from z = 2 − 0. We utilise observations from the Sloan Digital Sky Survey and the Mapping Nearby Galaxies at Apache Point Observatory survey at low redshifts. We complement these data with observations from the Cosmic Assembly Near-Infrared Deep Extragalactic Legacy Survey at high redshifts. Additionally, we compare the observations to detailed predictions from the LGalaxies semi-analytic model. To analyse the data, we developed a machine learning approach utilising a Random Forest classifier. We first demonstrate that this technique is extremely effective at extracting causal insight from highly complex and inter-correlated model data, before applying it to various observational surveys. Our primary observational results are as follows: at all redshifts studied in this work, we find bulge mass to be the most predictive parameter of quenching, out of the photometric parameter set (incorporating bulge mass, disk mass, total stellar mass, and B/T structure). Moreover, we also find bulge mass to be the most predictive parameter of quenching in both bulge and disk structures, treated separately. Hence, intrinsic galaxy quenching must be due to a stable mechanism operating over cosmic time, and the same quenching mechanism must be effective in both bulge and disk regions. Despite the success of bulge mass in predicting quenching, we find that central velocity dispersion is even more predictive (when available in spectroscopic data sets). In comparison to the LGalaxies model, we find that all of these observational results may be consistently explained through quenching via preventative ‘radio-mode’ active galactic nucleus feedback. Furthermore, many alternative quenching mechanisms (including virial shocks, supernova feedback, and morphological stabilisation) are found to be inconsistent with our observational results and those from the literature.

We study quenching in seven green valley galaxies on kpc scales by resolving their molecular gas content using 12CO(1–0) observations obtained with NOrthern Extended Millimeter Array and Atacama Large Millimeter Array, and their star formation rate using spatially resolved optical spectroscopy from the Mapping Nearby Galaxies at Apache Point Observatory survey. We perform radial stacking of both data sets to increase the sensitivity to molecular gas and star formation, thereby avoiding biases against strongly quenched regions. We find that both spatially resolved gas fraction (fgas) and star formation efficiency ($\rm {SFE}$) are responsible for quenching green valley galaxies at all radii: both quantities are suppressed with respect to typical star-forming regions. fgas and $\rm {SFE}$ have roughly equal influence in quenching the outer disc. We are, however, unable to identify the dominant mechanism in the strongly quenched central regions. We find that fgas is reduced by $\rm \sim\! 1~dex$ in the central regions, but the star formation rate is too low to be measured, leading to upper limits for the $\rm {SFE}$. Moving from the outer disc to central regions, the reduction in fgas is driven by an increasing $\rm \Sigma _{\star }$ profile rather than a decreasing $\rm \Sigma _{H_{2}}$ profile. The reduced fgas may therefore be caused by a decrease in the gas supply rather than molecular gas ejection mechanisms, such as winds driven by active galactic nuclei. We warn more generally that studies investigating fgas may be deceiving in inferring the cause of quenching, particularly in the central (bulge-dominated) regions of galaxies.

We develop a 2D inclined rotating disc model, which we apply to the stellar velocity maps of 1862 galaxies taken from the MaNGA survey (SDSS public Data Release 15). We use a random forest classifier to identify the kinematic parameters that are most connected to galaxy quenching. We find that kinematic parameters that relate predominantly to the disc (such as the mean rotational velocity) and parameters that characterise whether a galaxy is rotation- or dispersion-dominated (such as the ratio of rotational velocity to velocity dispersion) are not fundamentally linked to the quenching of star formation. Instead, we find overwhelmingly that it is the absolute level of velocity dispersion (a property that relates primarily to a galaxy’s bulge/spheroidal component) that is most important for separating star forming and quenched galaxies. Furthermore, a partial correlation analysis shows that many commonly discussed correlations between galaxy properties and quenching are spurious, and that the fundamental correlation is between quenching and velocity dispersion. In particular, we find that at fixed velocity dispersion, there is only a very weak dependence of quenching on the disc properties, whereby more discy galaxies are slightly more likely to be forming stars. By invoking the tight relationship between black hole mass and velocity dispersion, and noting that black hole mass traces the total energy released by AGN, we argue that these data support a scenario in which quenching occurs by preventive feedback from AGN. The kinematic measurements from this work are publicly available.

The Sunyaev-Zel'dovich (SZ) effect can potentially be used to investigate the heating of the circumgalactic medium and subsequent suppression of cold gas accretion onto the host galaxy caused by quasar feedback. We use a deep ALMA observation of HE0515-4414 in band 4, the most luminous quasar known at the peak of cosmic star formation (z=1.7), to search for the SZ signal tracing the heating of the galaxy's halo. ALMA's sensitivity to a broad range of spatial scales enables us to disentangle emitting compact sources from the negative, extended SZ signal. We obtain a marginal S-Z detection (∼3.3σ) on scales of about 300 kpc (30-40 arcsec), at the 0.2 mJy level, 0.5 mJy after applying a correction factor for primary beam attenuation and flux that is resolved out by the array. We show that our result is consistent with a simulated ALMA observation of a similar quasar in the fable cosmological simulations. We emphasise that detecting an SZ signal is more easily achieved in the visibility plane than in the (inferred) images. We also confirm a marginal detection (3.2σ) of a potential SZ dip on smaller scales (<100 kpc) already claimed by other authors, possibly highlighting the complex structure of the halo heating. Finally, we use SZ maps from the fable cosmological simulations, convolved with ALMA simulations, to illustrate that band 3 observations are much more effective in detecting the SZ signal with higher significance, and discuss the optimal observing strategy.

No abstract