A ‘phase transition’ in the physics of accretion onto galaxies
In the concordance model of cosmology, galaxies form in the center of dark matter halos, where the latter develop from primordial fluctuations. Already in ’78 White & Rees argued that the physics of gas in dark matter halos, and how it accretes onto the central galaxy, crucially depend on whether the gas cooling time is longer or shorter than its dynamical time. Despite that four decades have since passed, and that the idealized theory of this transition has been significantly developed (Birnboim & Dekel ’03, Dekel & Birnboim ’06, Fielding et al. ’17), this expected ‘phase transition’ in the nature of accretion has neither been detected in observations nor accurately identified in cosmological simulations. In a series of four papers (Stern et al. 2018, Stern et al. 2019, and two in prep.) I and collaborators revisit this predicted transition using (1) analytic calculations, (2) idealized hydro-simulations, (3) the FIRE cosmological simulations, and (4) UV absorption observations.
(1) Cooling flow solutions to the 1D steady-state flow equations (adapted from Stern et al. 2019 and Stern et al., in prep.). Top panels show gas temperature while bottom panels show the inflow Mach number of the inflow. If the mass inflow rate or gas density is low (left panels) then the gas is hot (T~T_vir) and pressure-supported (Mach<1) down to the galaxy scale (<~10kpc). If the mass inflow rate or gas density are relatively high (right panels), at the galaxy scale the flow is cool and in free-fall.
(2) Idealized 3D hydrodynamic simulations of gas in 1012 Msun halos, for different density normalizations. The gas is multi-phased and free-falling in the high density sim on the left, while it is single-phased and pressure-supported in the low density sim on the right (Stern et al. 2019, simulations developed by D. Fielding).
(3) FIRE cosmological simulations: (TBD)
(4) UV absorption observations: using HST-COS, Tumlinson et al. (2011, panel on the right) discovered that OVI absorption columns through halos of blue star-forming galaxies significantly differ from OVI columns through halos of red quiescent galaxies.
In Stern et al. (2018) we demonstrated that the OVI observed around blue galaxies is consistent with a volume-filling free-falling gas phase, corresponding to the free-falling halo gas seen in the theoretical figures above. The O VI dichotomy around blue/red galaxy can hence be explained if the volume-filling gas phase in the halos of blue galaxies is free-falling, while it is pressure-supported in red galaxies. The panel on the bottom-right shows that the relation between OVI column and OVI width is consistent with velocity shear across the absorber induced by tidal forces, if the OVI volume density is ~10-9 – 10-8.5 cm-3. The panel on the bottom-left demonstrates that the observed column ratios of HI/OVI and upper limits on NV/OVI are consistent with cool gas with a hydrogen volume density of ~10-4.5 cm-3, corresponding to a OVI volume density of 10-8.5 cm-3 for solar metallicity (blue line) or 10-9 cm-3 for third-solar metallicity (green line). That, is the absorption line ratios and velocity widths are consistent with free-falling gas with the same parameters.
Multi-Density Photoionization Modeling
In an earlier paper (Stern et al. 2016) , we derived the distribution of absorber sizes in the halos of low-redshift, Milky-Way-sized galaxies. The result is shown in this video and is based on a photoionization modeling method for CGM absorption lines surveys, applied to the COS-Halos survey. The new method relaxes the assumption used by almost all previous studies that the absorbing gas has a single density and physical scale, instead allowing the gas to span a range of densities and scales. Our analysis suggests a total absorbing gas mass which is <20% of gas mass estimates based on standard modeling methods. The new method can be applied to additional surveys of CGM absorption lines.