Gas in and around galaxies in the Simba simulations
Appleby, Sarah Ceridwen
Galaxy evolution is an interplay of physical processes across a wide range of cosmological size scales, from star formation and black hole growth on sub-pc scales, to interactions with their environments on Mpc scales. The region surrounding individual galaxies within their host halos is known as the Circumgalactic Medium (CGM), and connects galaxies to the wider Intergalactic Medium (IGM). The CGM acts as both a reservoir of inflowing gas and a repository for outflowing gas as it is transported outwards via galactic feedback. Quasar absorption line surveys have revealed the CGM to be a complex, multiphase environment in which ions tracing cool and hot gas are present within the same absorber systems. Cosmological simulations are a powerful tool for building a theoretical understanding of these interconnected processes and resulting observations. They provide a 3D view of galaxies and their host halos, in which we intrinsically know the physical conditions of the constituent dark matter and baryons. As such, simulations are useful both for testing galaxy evolution models against the real universe, and for interpreting complex observations in a physical context. In this thesis I present detailed comparisons of simulations to observations of star formation and quenching in galaxies, and of absorbers in galactic halos in an effort to constrain physical models. I use Simba, a suite of state of the art cosmological simulations with realistic sub-grid physical models based on high resolution zoom simulations, including novel treatments of dust evolution and of black hole growth and feedback. I investigate the nature of gas in galaxies and their halos, and show comparisons with observations. In addition, I isolate the impacts of galactic feedback and ionising radiation on the results by varying elements of the input physical models. In Chapter 3, I present radial profiles of star forming and green valley galaxies in Simba at z=0 and compare with observations of MaNGA spatially resolved spectroscopy. My analysis shows strong central depressions in star formation rate (SFR), specific SFR (sSFR), and gas fraction in Simba's green valley galaxies and massive star-forming systems, qualitatively as] observed, owing to Active Galactic Nuclei (AGN) X-ray feedback, which pushes central gas radially outwards. These effects are less pronounced at higher redshift, owing to less powerful AGN feedback. I also present a comparison of galaxy half light radii to observations at various redshifts, finding that predictions for star-forming galaxy sizes are accurate but that quenched galaxies are impacted by numerical heating. In Chapter 4, I characterise the low redshift CGM in Simba and examine the impact of differing feedback prescriptions. Low mass galaxies live in multiphase, diverse halos, whereas high mass galaxies live in halos dominated by hot gas. Halo baryon fractions are generally < 50% of the cosmic fraction due to stellar feedback at low masses, and 'jet-mode' AGN feedback at high masses. I present a comparison of absorption statistics to the COS-Halos and COS-Dwarfs surveys: Simba reproduces HI absorption well around star forming galaxies and broadly reproduces absorption of selected metal lines, including the observed dichotomy in OVI absorption between star forming and quenched galaxies. These predictions of metal line absorption are sensitive to the choice of photo-ionising background. Varying the galactic feedback prescription shows that stellar feedback enriches the CGM, while AGN feedback sets the gas temperature phase. In Chapter 5 I extend this analysis with a theoretical perspective of the CGM in Simba using a sample of CGM absorbers from a representative sample of Simba galaxies, selected across a range of stellar masses and sSFR. Absorbers are more abundant around low mass, star forming galaxies; the CGM of green valley galaxies is more similar to that of quenched galaxies. The absorber overdensity depends on the sSFR of the galaxy, while the absorber temperature is set by galaxy stellar mass. Absorption from low ions arises from cold, dense CGM gas, whilst absorption from higher ions traces hotter, more diffuse gas. I also examine the proportion of absorption arising from collisionally ionised gas, and from gas that is associated with satellite galaxies. Finally, in Chapter 6 I develop a tree-based machine learning mapping between the observable properties and the underlying physical conditions of the absorbers in the CGM absorber sample from Chapter 5. Such a mapping has the potential to be used to interpret CGM observations within a full cosmological context, assuming the Simba galaxy evolution model.