The Hot Circumgalactic Medium
predicting X-ray signatures of gas motions and feedback around galaxies
Overview
Every galaxy is surrounded by a vast reservoir of diffuse gas, the circumgalactic medium (CGM), which extends out to its virial radius and beyond. In galaxies with mass comparable to the Milky Way and above, a substantial fraction of this gas is hot enough (106-107 K) to emit primarily in X-rays. This hot phase is shaped by the same processes that regulate galaxy evolution more broadly: cosmological accretion, galactic rotation, and outflows driven by stellar and active galactic nucleus (AGN) feedback. Because the hot CGM is faint and diffuse, and the Milky Way has its own hot CGM which is much brighter, it is generally not possible to observe it with current X-ray observatories such as Chandra and XMM-Newton, except in bright nearby systems and only in the inner regions. However, future missions with X-ray integral field unit (IFU) microcalorimeter instruments provide the spectral resolution to observe the emission lines from the hot CGM in external galaxies which are cosmologically redshifted away from the bright lines of the Milky Way (see Figure 1. Cosmological hydrodynamical simulations play a central role in predicting what these observations should show and in identifying which signatures can actually distinguish between competing models of galactic feedback.
Mapping the Line-of-Sight Velocity Field of the Hot CGM
Because the hot CGM is optically thin and diffuse, any given line of sight through it can pass through gas participating in several distinct kinds of motion at once: rotation co-planar with the galactic disk, inflows associated with cosmological accretion, and outflows driven by stellar and AGN feedback. Disentangling these components from real spectra requires first understanding, in a controlled simulated setting, how each contributes to the velocity field and how well microcalorimeter-class spectral resolution can separate them.
In (ZuHone et al., 2024), my collaborators and I used the TNG50 cosmological simulation to characterize the line-of-sight velocity structure of the hot, X-ray-emitting CGM around a sample of nearby, Milky Way-mass simulated disk galaxies, and generated synthetic X-ray observations to assess what future instruments could recover from it. We found that stellar- and AGN-driven outflows produce the fastest, most readily detectable motions (roughly 200-500 km/s), while rotational motion in the hot gas is slower (roughly 100-200 km/s) but still measurable, particularly in galaxies viewed close to edge-on. Slow inflows near the galactic plane (roughly 50-100 km/s), by contrast, are difficult to isolate in projection, since they tend to be blended with faster components along the same sight line, though multi-component spectral fitting may still be able to recover them. We also found that the velocity structure inferred from the data is sensitive to which emission line is used to trace it, since different ions and transitions preferentially trace gas at different temperatures, an important consideration for planning and interpreting future microcalorimeter observations of the CGM.
X-ray Signatures of Galactic Feedback Across Simulation Models
The kinematic study above used a single simulation and feedback model, raising the question of how much these predictions depend on the specific implementation of stellar and AGN feedback used in a given simulation, an important question since different cosmological codes make substantially different choices for the subgrid physics governing feedback.
In (Silich et al., 2025), my collaborators and I extended this approach across seven different cosmological hydrodynamical simulation suites, examining 28 Milky Way-mass disk galaxies at redshift zero and comparing their predicted X-ray surface brightness morphology, velocity structure, temperature distributions, and emission-line signatures. We found that the X-ray brightness morphology of the hot CGM varies significantly from one feedback model to another, with some simulations producing prominent outflow-enhanced regions and bubble-like structures while others show much more azimuthally uniform emission; simulations lacking cosmic ray physics, however, consistently produced radial surface brightness profiles well described by a single power law (roughly r-3 between 20 and 200 kpc), with feedback physics primarily adding scatter around this trend rather than changing its slope. Velocity maps revealed bulk rotation of the CGM together with high-velocity biconical outflows, most prominent in simulations with strong AGN feedback, which also tended to produce extended regions of enhanced temperature in their large-scale outflows; simulations including cosmic ray physics instead predicted systematically cooler CGM gas, since cosmic ray pressure support suppresses some of the compressive heating present in purely thermal-feedback models. Finally, we found that different emission lines trace distinct phases of the gas: lower-energy lines such as O VII preferentially trace the volume-filling, quiescent component of the CGM, while higher-energy lines such as Fe XVII selectively highlight the high-velocity, feedback-driven outflows, underscoring the value of combining multiple lines to build a complete picture of the CGM’s thermal and dynamical state with future X-ray missions.