Stellar-gas kinematic misalignments in EAGLE: lifetimes and longevity of misaligned galaxies

TL;DR

The paper investigates how stellar-gas kinematic misalignments in galaxies with $M_{*}>10^{9.5}\,M_\odot$ persist and relax from formation to relaxation across $0<z<1$ using the EAGLE simulation. By extracting misalignments over complete formation-to-relaxation windows and computing 3D misalignment angles, dynamical and torquing times, and relaxation paths, the authors quantify relaxation times and their dependence on morphology, gas content, inflow, environment, and mergers. They find that the median relaxation time is about $0.5$ Gyr, with a substantial fraction ($\sim20\%$) lasting longer than $1$ Gyr; long-lived misalignments are associated with higher stellar masses, lower star-forming gas fractions, elevated gas inflow, and central positions in dense environments. Mergers contribute modestly to misalignment formation ($\sim10-21\%$ depending on redshift), suggesting diverse formation pathways, including halo cooling, are important for sustaining misalignments. Overall, the results indicate that unstable misalignments are not predominantly merger-driven and that long relaxation times are not common, providing constraints on ISM replenishment mechanisms in ETGs and LTGs alike.

Abstract

The dominant processes by which galaxies replenish their cold gas reservoirs remain disputed, especially in massive galaxies. Stellar-gas kinematic misalignments offer an opportunity to study these replenishment processes. However, observed distributions of these misalignments conflict with current models of gas replenishment in early-type galaxies (ETGs), with longer relaxation timescales suggested as a possible solution. We use the EAGLE simulation to explore the relaxation of unstable misaligned gas in galaxies with masses of $M_{*}\geqslant \mathrm{10^{9.5}}$ M$_\odot$ between $0<z<1$. We extract misalignments from formation to relaxation providing a sample of $\sim3200$ relaxations. We find relaxation timescales tend to be short-duration, with median lifetimes of $\sim0.5$ Gyr, though with a notable population of unstable misalignments lasting $\gtrsim1$ Gyr. Relaxation time distributions show a log-linear relationship, with $\approx20$ per cent of unstable misalignments persisting for $\gtrsim3$ torquing times. Long-lived unstable misalignments are predominantly found in galaxies with higher stellar masses, lower star-forming gas fractions, higher ongoing gas inflow, and which reside in the centres of dense environments. Mergers only cause $\approx10$ per cent of unstable misalignments among galaxies at $z<0.35$, and $\approx21$ per cent at $0.35<z<1.0$ in EAGLE. We conclude that, at least in EAGLE, unstable kinematic misalignments are not predominantly driven by gas-rich minor mergers at any redshift probed. Additionally, processes that significantly extend relaxation times are not dominant in the galaxy population. Instead, we see a diverse formation pathway for misalignments such as through hot halo cooling.

Figures (15)

• Figure 1: Galaxy stellar mass function at $z=0.1$ of the total galaxy population (solid purple line) in the Ref-L100N1504 simulation in comparison with our sub-sample of galaxies (dashed red line) selected using the criteria listed in Section \ref{['sample:']}. There is no mass bias in our sub-sample compared to the general galaxy population.
• Figure 2: Projected 2D (cyan/orange) and 3D (blue/red) stellar-gas$_{\rm{SF}}$ misalignment angle distributions for LTGs (top) and ETGs (bottom) for our sample of misalignments at $z=0.1$. The vertical dashed black line denotes the boundary between aligned ($<30^{\circ}$) and misaligned galaxy populations ($>30^{\circ}$). Observational results from Bryant2019 are included in grey and should be compared against $\psi_{\rm{2D}}$ distributions. Errors are given as Poisson uncertainties. The simulated and observed galaxies exhibit fairly similar distributions.
• Figure 3: Stellar-gas$_{\rm{SF}}$ misalignment angle fractions for LTGs (solid lines) and ETGs (dash-dotted lines) between $0<z<1$. Percentages are given with respect to the total galaxies of a given morphology at a given redshift. We classify galaxies as aligned with $\psi_{\rm{3D}}<30^{\circ}$ (blue), misaligned galaxies with $\psi_{\rm{3D}}>30^{\circ}$ (red), and counter-rotating galaxies with $\psi_{\rm{3D}}>150^{\circ}$ (purple). Errors are given as Poisson uncertainties (shaded regions). We see a slight upward trend in the misalignment fraction for both ETGs and LTGs across this redshift range, coinciding with an increase in counter-rotating fraction for both galaxy populations.
• Figure 4: Evolution of two galaxies that experience long relaxation periods. Left: Galaxy A displaying an unstable misalignment matching our criteria with $t_{\rm{relax}}\approx3.78$ Gyr $\approx15.0\bar{t}_{\rm{dyn}}\approx3.5\bar{t}_{\rm{torque}}$. Right: Galaxy B shows an initial unstable misalignment and relaxation into the stable counter-rotating regime with $t_{\rm{relax}}\approx0.35$ Gyr $\approx2.0\bar{t}_{\rm{dyn}}\approx1.1\bar{t}_{\rm{torque}}$, before the cumulative effects of star formation induce a second unstable misalignment driven by a stellar angular momentum flip back into a co-rotating state with $t_{\rm{relax}}\approx2.55$ Gyr $\approx12.0\bar{t}_{\rm{dyn}}\approx7.5\bar{t}_{\rm{torque}}$. In order, the rows show: (1) the 3D misalignment angle between stars and gas$_{\rm{SF}}$ (black) within $r_{50}$ ($1\sigma$ error bars shown in grey) and the 3D misalignment angle between the stars/gas$_{\rm{SF}}$ at $r_{50}$ and the DM at 30 pkpc, with the horizontal dark and light grey shaded regions representing the stable regime ($<20^{\circ}$ and $>160^{\circ}$) and aligned/counter-rotating regime ($<30^{\circ}$ and $>150^{\circ}$), respectively
• Figure 5: Illustration of the evolution of two galaxies shown in Figure \ref{['plot: evolutions']} in five key stages. The net angular momentum vectors of the stellar (yellow) and gas$_{\rm{SF}}$ (blue) are represented by arrows. Top: evolution of Galaxy A beginning with (I) an initial flyby of a gas-rich satellite that causing a stellar-DM misalignment; (II) the formation of an unstable misalignment from the accretion of gas stripped from the satellite during a second fly-by; (III) merging of the satellite galaxy with the central galaxy and the resulting morphological transformation of the galaxy to a more spheroidal stellar mass distribution, while maintaining the unstable misalignment through continued accretion of stripped gas; (IV) a decrease in the inflow rate allows the gas disc to begin precessing and relaxing into the galactic plane; (V) the gas disc becomes aligned with the stellar angular momentum. Bottom: evolution of Galaxy B beginning with (I) a gas-rich merger that disrupts the existing gas disc; (II) the quick formation of a large misaligned gas disc that quickly settles to become counter-rotating; (III) a new generation of counter-rotating stars begin to form from the counter-rotating gas which begins to build a new counter-rotating stellar disc; (IV) the stellar angular momentum eventually flips, manifesting as a 'pseudo-relaxation'; (V) the counter-rotating stars dominate the net angular momentum and the galaxy appears as aligned.