Hot accretion onto spiral galaxies: the origin of extended and warped HI discs

TL;DR

This work investigates the origin of extended and warped HI discs in local spiral galaxies by simulating hot, rotating circumgalactic media (CGM) with misaligned rotation axes relative to pre-existing galactic discs. Using idealized 3D hydrodynamic simulations with the GIZMO code, the authors show that hot CGM inflows continuously condense into cool HI discs at radii near the circularization radius, forming long-lived warps whose planes align with the CGM rotation axis. The results explain why HI discs extend beyond stellar discs, predict warps trace the angular momentum of the hot CGM, and imply that warp observations can constrain CGM accretion rates and metallicity. This hot-accretion scenario also reconciles the so-called HI desert by predicting that the cool gas observed in warps originates from hot CGM rather than the inner CGM, offering a new observational handle on disc galaxy evolution.

Abstract

Gas accretion, hot ($\sim 10^6\,{\rm K}$) atmospheres, and a tilt between the rotation axes of the disc and the atmosphere are all robust predictions of standard cosmology for massive star-forming galaxies at low redshift. Using idealized hydrodynamic simulations, we demonstrate that the central regions of hot galaxy atmospheres continuously condense into cool ($\sim10^4\,{\rm K}$) discs, while being replenished by an inflow from larger scales. The size and orientation of the condensed disc are determined by the angular momentum of the atmosphere, so it is large and often tilted with respect to the pre-existing galaxy disc. Continuous smooth accretion from hot atmospheres can thus both provide the necessary fuel for star formation and explain the observed ubiquity of extended and warped HI discs around local spirals. In this hot accretion scenario, cool gas observations cannot be used to trace the source of the HI, warps out to halo radii, consistent with recent indications of a lack of $21\,{\rm cm}$ emission from the halos of nearby galaxies (the `HI desert'). Observations of HI warps formed via hot accretion can be used to constrain the angular momentum, accretion rate, and gas metallicity of hot galaxy atmospheres, important parameters for disc galaxy evolution that are hard to determine by other means.

Figures (12)

• Figure 1: Illustration of the initial conditions in the simulations: a galactic disc surrounded by a hot rotating CGM with a tilted axis. The primed and non-primed coordinate systems respectively describe the CGM and disc orientations, with $z$ and $z^\prime$ along the rotation axes. The $y=y'$ axis is oriented along the line of nodes, and the tilt angle is marked by $\theta_{\rm tilt}$.
• Figure 2: An H i warp condensing out of a tilted hot CGM. Panels show edge-on projections of the $\theta_{\rm tilt} =30^\circ$ simulation at $t = 400$ Myr, after the hot inflow reached a steady state. Gas temperature is shown on the left, H i column density in the centre, and velocity along the projected axis on the right. White streamlines in the left panel trace the velocity field in the projected plane. The pre-existing disc is apparent in the $z=0$ plane with a maximum extent of $4R_{\rm d}=10\,{\rm kpc}$. A cool, integral-sign H i warp is apparent, aligned with the $30^\circ$-tilted midplane of the hot CGM. The streamlines demonstrate that the hot CGM is inflowing towards the disc and is the source of the warp. Cooling of the hot CGM from $\approx10^6\,{\rm K}$ to $\approx10^4\,{\rm K}$ at the warp boundary increases $N_{\text{H{\sc i}}}$ by orders of magnitude.
• Figure 3: Similar to Fig. \ref{['fig:xz_projection']}, for a face-on projection. White streamlines on the left demonstrate how the hot CGM spirals inward toward the disc, cooling abruptly from $\approx10^6\,{\rm K}$ to $\approx10^4\,{\rm K}$ at the warp boundary. The right panel shows that at $|x|\gtrsim 10\,{\rm kpc}$ the warp is inflowing towards the disc midplane.
• Figure 4: Formation of H i warps from rotating hot CGM with different tilt angles. Panels show edge-on projections of H i column density in the $t= 400\,{\rm Myr}$ snapshots of the four simulations, with red streamlines tracing the velocity field in the projected plane. Tilt angles are noted on top and are evident from the symmetry axis of the streamlines. The hot inflow cools in the midplane of the tilted CGM in all four cases, creating a H i warp when $\theta_{\rm tilt}\neq0^\circ$. Observed angles of H i warps thus trace the rotation axis of the hot CGM.
• Figure 5: Angular momentum profiles of cool and hot gas in CGM inflows. Panels show the magnitude (top) and angle to the disc axis (bottom) of the mean specific angular momentum vector at each spherical radius, for both cool gas ($T<10^{5}\,{\rm K}$, solid) and hot gas ($T> 10^{5}\,{\rm K}$, dashed). Colors denote the four simulations, which differ in CGM tilt angle, as evident in the hot gas profiles in the bottom panel at large radii. The $10\,{\rm kpc}$ size of the pre-existing disc is marked. The H i warp is evident as an increase in cool gas angle from $\approx 0^\circ$ at $r\lesssim10\,{\rm kpc}$ to $\approx\theta_{\rm tilt}$ at $r\approx 20\,{\rm kpc}$. The cool warp rotates somewhat slower than circular orbits (gray line in top panel), while hot gas rotates slower than cool gas at all radii.