Modeling Multiphase Galactic Outflows: A Multifluid Moving Mesh Approach

F. Bollati, R. Weinberger · 2025

Read Modeling Multiphase Galactic Outflows: A Multifluid Moving Mesh Approach by F. Bollati, R. Weinberger (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-16971.txt.

Introduction

Galaxy winds carry both hot and cold gas, yet cheap simulations blur the cold clouds away. Francesco Bollati and Rainer Weinberger add a sub-grid model that tracks the cold phase as its own fluid, in work from twenty twenty-five. Drag and mixing terms, drawn from single-cloud math, couple the two phases.

Run on a starburst setup like the galaxy M82, the model matched fine-grained runs on speeds, density, and outflow rates. Cold winds grew on their own from hot wind meeting cold clouds. Mixing strength steered the outcome hard: stronger mixing fed the cold phase from the hot, slowed the hot, and sped the cold. Cosmology runs gain a workable wind.

Catalog record

CREATORF. Bollati, R. Weinberger
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE116 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-16971

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Modeling Multiphase Galactic Outflows: A Multifluid Moving Mesh Approach F. Bollati, R. Weinberger Abstract Outflows are a key part of the galactic gas cycle and crucial in shaping the star formation activity in their host galaxy. Yet, in simulations of galaxy evolution, modeling these outflows in their multi-phase nature and over the relevant timescales is an unsolved problem. We present a subgrid model for simulating multiphase galactic outflows in efficient, comparatively low-resolution simulations, designed for application in future cosmological simulations. The cold phase ( T = 10 4 T=10^{4} K) is treated as pressureless, and its interaction with the hot phase is captured through source terms representing drag and mixing. These terms are obtained using analytic drag and mixing terms for single clouds and convolving them with a cloud mass distribution consistent with high-resolution simulations. Applied to a setup resembling the starburst galaxy M82, the model reproduces the velocity, density, and mass outflow rates of high-resolution simulations that resolve individual cold clouds. Cold outflows emerge naturally from interactions between the hot wind and cold interstellar clouds, with drag and mixing both contributing to the acceleration. Varying the mixing strength strongly affects outflow properties: stronger mixing enhances mass transfer from hot to cold gas, reduces the hot phase velocity, and accelerates the cold phase, while also influencing the origin and composition of the cold outflow. Weak mixing produces cold gas mostly from preexisting interstellar clouds, whereas stronger mixing leads to substantial cold gas formation from the hot phase. This framework enables efficient simulations of multiphase galactic outflows while retaining key multi-component features of the outflow dynamics. 1 Introduction Galactic winds play a fundamental role in galaxy evolution as they expel gas from the host galaxy, heat and enrich the circum-galatic medium with metals and regulate star formation in the host galaxy. However, despite their perceived importance, the many details related to galactic winds are still unclear: which physical processes drive galactic outflows? How much mass and energy is actually removed from galaxies? Which phase dominates the outflow? How universal are these properties as a function of galaxy mass, morphology and redshift? Answering these questions is key to a more complete understanding of galaxy formation (see Thompson & Heckman 2024 for a recent review). From an observational perspective, galactic outflows are detected both in the local universe ( Heckman et al. 1990 , e.g.,) and at high redshift ( Shapley et al. 2003 ; Genzel et al. 2014 ; Förster Schreiber et al. 2019 ; Guo et al. 2023 ; Belli et al. 2024 , e.g.,) , establishing them as a common feature of star-forming galaxies. A key finding is that these winds are multiphase in nature, spanning an enormous range of gas temperatures from \sim 10 K to \gtrsim 10^{7} K, and traced through both emission and absorption features. The cold molecular phase ( T\sim 10 K) is revealed by tracers such as CO ( Bolatto et al. 2013 ; Leroy et al. 2015 ; Veilleux et al. 2020 ; Fluetsch et al. 2021 , e.g.,) and PAH emission ( Fisher et al. 2025 , e.g.,) . The cool neutral phase ( T\sim 10^{2} K) is detected via H i ( Martini et al. 2018 ; Watts et al. 2024 , e.g.,) and low-ionization lines such as Na D, O i , and Mg i ( Rupke et al. 2005 ; Davies et al. 2024 ) . A warm ionized phase ( T\sim 10^{4-5} K) is traced by H \alpha ( McKeith et al. 1995 ; Westmoquette et al. 2009 , e.g.,) and metal lines including Mg ii , O ii , O iii ,and Si iv ( Rubin et al. 2011 ; Chisholm et al. 2017 ; Concas et al. 2019 ; Ha et al. 2025 , e.g.,) . The warm–hot phase ( T\sim 10^{5-6} K) is detected in higher ionization tracers such as O vi ( Ashley et al. 2020 ; Ha et al. 2025 , e.g.,) , while the hottest gas ( T\gtrsim 10^{6} K) is observed in diffuse X-ray emission ( Strickland & Heckman 2009 ; Lopez et al. 2020 , e.g.,) . Nearby archetypal starbursts such as M82 and NGC 253 show evidence for all of these phases coexisting in the same outflow ( Bolatto et al. 2013 ; Lopez et al. 2020 , e.g.,) , while at higher redshift typically only a subset of phases are detected, often limited to ionized or neutral gas ( Davies et al. 2024 ; Watts et al. 2024 ; Belli et al. 2024 , e.g.,) . Despite these advances, significant uncertainties remain: the relative mass-loading factors of different phases, their energy and momentum budgets, and the physical mechanisms that generate and maintain multiphase structure are still poorly constrained observationally. Addressing these open questions increasingly relies on numerical simulations, which can disentangle the role of different processes in shaping the multiphase outflow. From a theoretical perspective, simulating multiphase outflows resolving even the T\sim 10^{4} K phase has been proven challenging. Cloud crushing simulations ( Klein et al. 1994 ; Xu & Stone 1995 , e.g.,) provide a controlled setup to study at high-resolution the interaction between a single cold ( T\sim 10^{4} K) cloud and the hot wind in which it is embedded. The general picture emerging from cloud–wind interaction simulations is that the relative motion between clouds and the hot flow excites hydrodynamical instabilities—primarily Kelvin–Helmholtz and Rayleigh–Taylor instabilities—at the phase interface. These instabilities generate turbulence and create a mixing layer between the two phases, characterized by intermediate temperatures ( Gronke & Oh 2018 ; Fielding et al. 2020 ) . Within this layer, two competing processes operate: turbulent mixing, which strips material from and can ultimately destroy the clouds, and radiative cooling, which can condense hot material onto the cloud surface and promote cloud mass growth ( Gronke & Oh 2018 ; Gronke & Oh 2020 ; Fielding et al. 2020 ; Tan et al. 2021 ; Kanjilal et al. 2021 ; Abruzzo et al. 2022 , e.g.,) . Both processes are accompanied by associated momentum and energy exchange, while in parallel the clouds are accelerated by ram pressure drag. The outcome of cloud survival or destruction is therefore set by the relative timescales of these competing effects ( Fielding & Bryan 2022 ) . A large body of work has explored how additional physical processes alter the cloud–wind interaction, including conduction ( Brüggen & Scannapieco 2016 , e.g.,) , magnetic fields ( McCourt et al. 2015 ; Cottle et al. 2020 ; Sparre et al. 2020 , e.g.,) , cosmic rays ( Wiener et al. 2019 ; Brüggen & Scannapieco 2016 , e.g.,) , viscosity ( Marin-Gilabert et al. 2025 , e.g.,) , the pressure gradients of expanding winds ( Dutta et al. 2025 , e.g.,) , and turbulence in the hot medium ( Das et al. 2024 ; Marin-Gilabert et al. 2025 , e.g.,) . More recent studies have begun to examine populations of clouds embedded in hot flows or turbulent environments. These works suggest that the collective interaction is not a simple superposition of individual cloud–wind encounters, but instead gives rise to more complex emergent behaviors ( Villares et al. 2024 ; Tan & Fielding 2024

Read the complete text