MISTRAL: a model for AGN winds from radiatively efficient accretion in cosmological simulations

TL;DR

Farcy et al. introduce MISTRAL, a subgrid wind model for AGN feedback from radiatively efficient accretion, implemented in Arepo and calibrated within IllustrisTNG physics. They present two implementations: mistral-continuous (continuous radial momentum deposition) and mistral-stochastic (stochastic bipolar momentum deposition). In idealized Milky Way–mass and cosmological $z=2$ zoom simulations, mistral-stochastic generates large-scale bipolar winds that efficiently regulate star formation and BH growth, reproduce the empirical stellar-to-halo mass relation, and yield BH–stellar mass trends without SMBH-mass–dependent tuning, while mistral-continuous tends to produce weaker regulation and fountain-like gas cycling. Across halo masses from $M_{200}\sim 10^{12}$ to $3 imes10^{13} m M_\odot$, mistral-stochastic provides self-consistent feedback that suppresses inflows and enhances outflows, with implications for interpreting JWST-era observations of high-redshift galaxies and quasars ($M_{200} o 10^{12}-3 imes10^{13} m M_\odot$).

Abstract

Feedback from active galactic nuclei (AGN) is crucial for regulating galaxy evolution. Motivated by observations of broad absorption line winds from rapidly accreting supermassive black holes (SMBHs), we introduce the Mistral AGN feedback model, implemented in the Arepo code. Mistral comes in two versions: continuous radial (Mistral-continuous) and stochastic bipolar momentum deposition (Mistral-stochastic). Using the framework of the IllustrisTNG simulations, we explore the effect of Mistral on BH and galaxy properties, through an idealized Milky Way-mass galaxy and cosmological zoom simulations run down to $z=2$. Unlike standard thermal AGN feedback prescriptions, Mistral generates galaxy-scale winds that mimic outflows driven by BH accretion. Mistral-continuous produces short-lived galactic fountains, and is inefficient at regulating the growth of massive galaxies at $z=2$. In contrast, Mistral-stochastic efficiently suppresses star formation in massive galaxies, reproduces the empirical stellar-to-halo mass relation, and yields a consistent trend of BH-stellar mass evolution. By supporting large-scale outflows while simultaneously preventing gas inflows, Mistral-stochastic additionally regulates the cold and hot gas fractions at both galaxy and halo scales. Mistral-stochastic therefore works self-consistently across the halo mass range explored $\left(10^{12}-3\times10^{13}\,\rm M_\odot\right)$, without adopting a SMBH-mass dependent AGN feedback scheme such as the one used in IllustrisTNG. Our model is a promising tool for predicting the impact of AGN winds on galaxy evolution, and interpreting the growing population of high-redshift galaxies and quasars observed by JWST. This work is part of the "Learning the Universe" collaboration, which aims to infer the physical processes governing the evolution of the Universe.

Figures (19)

• Figure 1: Fraction of the inflowing gas mass rate $\dot{M}_{\rm inf}$ that contributes to the BH mass accretion rate $\dot{M}_{\rm BH}$ (black lines, left y-axis) and to the BH energy rate $\dot{E}_{\rm BH}$ (blue lines, right y-axis) as a function of the AGN wind feedback coupling efficiency $\epsilon_{\rm w}$. Solid, dashed and dotted lines respectively correspond to an average wind velocity of $10^5$, $10^4$ and $10^3\,\rm km\,s^{-1}$.
• Figure 2: Clockwise: schematic illustration of the isotropic thermal and random kinetic models used in TNG, mistral-stochastic and mistral-continuous. For each model, we represent the BH sink particle as a yellow star, surrounded by gas cells depicted by blue circles. The red circle shows the extent of the spherical smoothing volume with radius $h_{\rm BH}$. With the isotropic thermal model, each gas cell receives a mass-weighted fraction of the BH energy in a thermal form (Eq. \ref{['eq:ethcell']}). With the random kinetic model, gas cells receive a momentum kick in the same random direction, provided that enough BH energy has accumulated (Eq. \ref{['eq:ekincell']}). In the TNG simulations, this model acts when the Eddington ratio is below a certain threshold $\chi$ (Eq. \ref{['eq:chi']}), and the isotropic thermal model operates otherwise. With mistral-continuous, cells are kicked radially away from the BH, with a velocity weighted by the alignment of the cell to the angular momentum of the gas surrounding the BH (Eq. \ref{['eq:vkick']}). With mistral-stochastic, cells are kicked at a velocity $v_{\rm w}$ with a probability $\mathbb{P}$ (Eq. \ref{['eq:effproba-outf']}), in a direction parallel or anti-parallel to the gas angular momentum.
• Figure 3: Edge-on 80 kpc wide slices of the gas density (top rows) and temperature (bottom rows). From left to right and top to bottom, we show the no bh, fiducial tng, random kinetic, isotropic thermal, mistral-continuous and mistral-stochastic simulations. A 10 kpc width scale bar is plotted in the lower left corner, and the time at which these maps are plotted is indicated in the lower right corner. In the density maps, grey arrows show the gas velocity field overlaid.
• Figure 4: Probability density function of the gas radial velocity with data stacked from 500 to 600 Myr. We restrict the material used in the analysis to gas within half the galaxy virial radius and outside the galaxy disc. We show simulations without BH physics ( no bh) in grey dotted, and with the fiducial tng, random kinetic, isotropic thermal, mistral-continuous and mistral-stochastic feedback models in dashed black, solid orange, green, blue and purple, respectively.
• Figure 5: Phase diagrams of gas temperature against the gas radial velocity, color-coded by the mean hydrogen number density. From top to bottom and left to right, we show the no bh, isotropic thermalfiducial tngmistral-continuous, random kinetic and mistral-stochastic simulations, with data stacked from 500 to 600 Myr (21 snapshots). Black lines show the $1\sigma$ contours for $n_{\rm H}=10^{-2},10^{-1}$ and $1\,\rm cm^{-3}$.