Modeling Photoionized Turbulent Material in the Circumgalactic Medium

Abstract

The circumgalactic medium (CGM) of nearby star-forming galaxies shows clear indications of O VI absorption accompanied by little to no detectable N V absorption. This unusual spectral signature, accompanied by highly non-uniform absorption from lower ionization state species, indicates that the CGM must be viewed as a dynamic, multiphase medium, such as occurs in the presence of turbulence. Motivated by previous isotropic turbulent simulations, we carry out chemodynamical simulations of stratified media in a Navarro-Frenk-White (NFW) gravitational potential with a total mass of 10 12 M and turbulence that decreases radially. The simulations assume a metallicity of 0.3 Z , a redshift zero metagalatic UV background, and they track ionizations, recombinations, and species-by-species radiative cooling using the MAIHEM package. We compare a suite of ionic column densities with the COS-Halos sample of low-redshift star-forming galaxies. Turbulence with an average onedimensional velocity dispersion ≈ 40 km s −1 , corresponding to an energy injection rate of ≈ 4 × 10 49 erg yr −1 , produces a CGM that matches many of the observed ionic column densities and ratios. In this simulation, the N N V /N O VI ratio is suppressed from its equilibrium value due to a combination of radiative cooling and cooling from turbulent mixing. This level of turbulence is consistent with expectations from observations of better constrained, highermass systems, and could be sustained by energy input from supernovae, gas inflows, and dynamical friction from dark matter subhalos. We also conduct a higher resolution ≈ 40 km s −1 run which yields smaller-scale structures, but remains in agreement with observations.

A Graphical Interpretation of Circumgalactic Precipitation

“…The critical velocity dispersion implied by the Voit (2018) model is broadly consistent with the findings of Gaspari et al (2013) for numerical simulations that drive turbulence, as well as with the observed velocity dispersions of galaxy-cluster cores that appear to be precipitating (Gaspari et al 2018). The turbulence cannot be much greater without either damping the perturbations through mixing with the ambient gas or overheating the ambient medium through turbulent dissipation (Gaspari et al 2013(Gaspari et al , 2017Banerjee & Sharma 2014;Buie et al 2018Buie et al , 2020. Around a galaxy like the Milky Way, the predicted one-dimensional velocity dispersion of hot gas in a precipitating CGM is therefore σ t ≈ 50-70 km s −1 .…”

Interpreting Observations of Absorption Lines in the Circumgalactic Medium with a Turbulent Medium

“…To this end, we can turn again to the example presented in Figure 1. Gray et al (2015); Gray & Scannapieco (2016); Buie et al (2018) showed that at low amounts of turbulence, σ 3D 10 km s −1 (top panel), MAIHEM mass fractions agree well with the pure photoionization case computed with CLOUDY (Ferland et al 2013).…”

“…A key result of this analysis is that turbulence as included in these models promotes the density and temperature gradients needed to sustain both low, intermediate, and high ions simultaneously (Buie et al 2018). We see, however, that the MAI-HEM model does not provide a perfect description of the data for intermediate and high ions.…”

“…In addition to the CGM being multiphase, Werk et al (2016) find the higher ionization state gas to have turbulent velocities of 50 -75 km s −1 contributing to the line widths of ions, which do not find a natural explanation in PIE models. To explore the dynamics of such turbulence in the CGM, Buie et al (2018) modeled isotropic turbulence using MAIHEM 1 (Models of Agitated and Illuminated Hindering and Emitting Media), in attempts to explain the puzzling presence of O VI but lack of N V absorption feature. This work showed that turbulence with σ 1D ≈ 60 km s −1 replicates many of the observed features within the CGM, such as clumping of low ionization-state ions and the existence of O VI at moderate ionization parameters, over-predicting however the amount of N V.…”

Non-isobaric Thermal Instability

“…Integral field unit (IFU) observations would be particularly useful, as the marked increases and decreases in brightness over the course of an oscillation would be correlated with the relative blueshifts and redshifts that track the compression and expansion velocities. No such correlations are expected for a turbulent multiphase environment, such as the ISM (and perhaps even the CGM; see Buie et al 2018). Computations of synthetic IFU observations for this signature APPENDIX A. LINEARIZED SOLUTIONS TO THE EQUATIONS OF NON-ADIABATIC HYDRODYNAMICS Here we linearize the equations of gas dynamics to derive the dispersion relation in equation (5), as well as the analytic solutions quoted in equation (11).…”