A "universal" supernova efficiency to drive the HI turbulence in nearby star-forming galaxies

arXiv:2610.10698 · astro-ph.GA · Submitted 2026-10-07 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: Today's paper: "A "universal" supernova efficiency to drive the HI turbulence in nearby star-forming galaxies".

Jocelyn: The gist The observed HI kinetic energy in dwarf galaxies requires only ≲ 3% of the SN energy,

Vera: First, who's behind it and why it matters.

Paper summary: Vera: Building on what we discussed, this paper, "A "universal" supernova efficiency to drive the HI turbulence in nearby star-forming galaxies," focuses on dwarf irregular galaxies specifically, which are those challenging environments with low masses and low star formation rates. The authors investigate the origin of neutral atomic gas turbulence here.

Jocelyn: They look at the observed HI velocity dispersion in these dwarfs, finding it’s typically around six to twelve km/s, which often seems too high for just thermal motions alone. This excess is what they attribute to turbulence between fluid elements.

Subrahmanyan: The paper builds on prior work from spirals showing that disc flaring prolongs the time it takes for energy to dissipate, which makes supernova feedback a more viable way to sustain that turbulence in those lower star formation rate environments.

Vera: They measure the SN efficiency, which is defined as the fraction of supernova energy required specifically to sustain HI turbulence in these dwarf galaxies. They test if this efficiency is universal across a sample representative of the star-forming main sequence.

Jocelyn: From their observed HI distributions and kinematics, they derive radial profiles for the kinetic energy per unit area and model them as a sum of thermal energy and turbulent energy from supernovae, constrained by the observed surface density of star formation.

Subrahmanyan: They consider pure cold neutral medium phases and warm neutral medium phases as limiting cases, along with a mixed two-phase scenario to constrain the supernova efficiency and the warm neutral medium fraction using hierarchical Bayesian inference.

Conclusion: Vera: So to wrap up this paper, "A "universal" supernova efficiency to drive the HI turbulence in nearby star-forming galaxies," the core finding is that they found a single, universal supernova efficiency value for driving HI turbulence across many different dwarf galaxies.

Jocelyn: They found that when you look at the observed kinetic energy in these dwarf galaxies, it only requires about zero point zero two or two percent of the total supernova energy to keep things moving turbulently.

Subrahmanyan: That means this efficiency is not wildly dependent on whether the galaxy is a dwarf irregular or a spiral, which is pretty big for building models of galaxy evolution.

Vera: They show that this two percent efficiency holds up across their sample of star-forming main sequence galaxies, matching what they see in both dwarfs and spirals.

Jocelyn: It’s within about a factor two of the uncertainty they have, and it lines up well with what models predict for how supernova remnants evolve.

Subrahmanyan: This gives us a concrete benchmark for how much energy SNe transfer into the cold interstellar medium in those sub-grid models.

Vera: The paper suggests that supernovae alone are enough to drive this turbulence without needing extra mechanisms like gravitational instabilities to explain the motions in dwarf galaxies.

Jocelyn: They did mention that their method is a bit different from other ways people study this, and they haven't fully settled if SNe are truly sufficient on their own.

Subrahmanyan: That’s a fair point because this work provides a solid starting point for setting up the fraction of supernova energy that gets transferred into the cold ISM in those sub-grid models.

Lauri Sassali, Cecilia Bacchini, Aku Venhola, Giuliano Iorio, Antonino Marasco

Space Physics and Astronomy Research Unit, University of Oulu · DARK, Niels Bohr Institute, University of Copenhagen · Institut de Ciències del Cosmos (ICCUB), Universitat de Barcelona · INAF – Padova Astronomical Observatory

astro-ph.GA

Submitted: 2026-10-07

Updated: 2026-10-07

Comments: Accepted for publication in A&A on 24/09/2026. 20 pages, 10 figures, 3 tables

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 72/100

The gist: The gist The observed HI kinetic energy in dwarf galaxies requires only ≲ 3% of the SN energy, and that a "universal" efficiency of ≈ 2% explains the HI turbulence across our main-sequence sample.

Key concepts

HI Turbulence: This refers to the chaotic, non-uniform motions within the neutral atomic gas (HI) in galaxies. It's measured by velocity dispersion and is thought to be caused by energy injection from supernovae, as it exceeds what thermal motions alone can explain.
SN Efficiency ($\eta$): This is a key parameter representing the fraction of supernova energy that successfully transfers into driving turbulence within the cold gas. The study tests if this efficiency stays consistent across different galaxies.
Two-Phase Scenario: The model considers the neutral gas in two states: cold (dense, star-forming) and warm (diffuse). The paper uses a mixed two-phase scenario to better reflect the actual gas distribution, finding a universal efficiency of about 1.2% for this case.

Terminology

Velocity Dispersion — 17× in this paper · explained in 2 episodes
Stellar Feedback — 14× in this paper · explained in 2 episodes
Metallicity — 5× in this paper · explained in 3 episodes
Degeneracy — 4× in this paper · explained in 4 episodes
Abundance — 3× in this paper · explained in 1 episode
Bayesian Inference — 3× in this paper · explained in 5 episodes
Core — 3× in this paper · explained in 2 episodes
Consistency — 2× in this paper · explained in 3 episodes
Trace — 2× in this paper · explained in 2 episodes
Adam — 1× in this paper · explained in 2 episodes
Bayesian Framework — 1× in this paper · explained in 3 episodes
Benchmark — 1× in this paper · explained in 2 episodes
Initial Mass Function (IMF) — 1× in this paper · explained in 2 episodes

Summary

The gist The observed HI kinetic energy in dwarf galaxies requires only ≲ 3% of the SN energy, and that a universal efficiency of ≈ 2% explains the HI turbulence across our main-sequence sample.

Introduction and Motivation

The origin of cold gas turbulence in galaxies has long been debated, with supernovae (SNe) being considered insufficient to sustain it against dissipation, especially in low star formation rate environments like dwarf irregular galaxies<ref:2610.10698#pg6>. The observed HI velocity dispersion in star-forming dwarf galaxies is typically 6 − 12 km/s, which often exceeds what can be explained by a purely thermal origin<ref:2610.10698#pg4>. This excess is ascribed to turbulence, indicating chaotic motions between fluid elements<ref:2610.10698#pg5>. Numerical simulations suggest turbulence should dissipate on timescales of order 10 Myr, necessitating a continuous energy source to sustain it<ref:2610.10698#pg6>.

Methodology and Modeling

The study models the observed kinetic energy as the sum of turbulent and thermal components, defined by Emod(R) = Eturb(R) + Eth(R)<ref:2610.10698#pg4>. The timescale for turbulence dissipation is defined as τd ≡ Eturb / E˙diss = LD / vturb<ref:2610.10698#pg4>. The turbulent energy injected by SNe is calculated using Eturb,SNe = ηΣSFR fccESN / (2hvturb)<ref:2610.10698#pg5>. The thermal energy per unit area is modeled as Eth = ECNM + EWNM, involving the WNM-to-total HI fraction fw<ref:2610.10698#pg4>. A hierarchical Bayesian inference framework is employed to constrain the SN efficiency η and the WNM fraction fw<ref:2610.10698#pg5>.

Results for Characteristic Efficiencies

The analysis yields characteristic SN efficiencies for different gas phases in individual dwarf galaxies<ref:2610.10698#pg4>. For a purely cold medium (fw = 0), the median efficiency is ⟨ηCNM⟩ ≈ 0.031, with values generally below 0.1<ref:2610.10698#pg5>. In the pure warm medium case (fw = 1), the characteristic efficiencies are much lower, with a median of ⟨ηWNM⟩ ≈ 0.001<ref:2610.10698#pg5>. The most realistic two-phase scenario yields a median universal efficiency of ⟨η2ph⟩ ≈ 0.012<ref:2610.10698#pg5>.

Universal Efficiency Findings

The hypothesis of a single universal SN efficiency is tested across the sample, and the results are consistent with this hypothesis<ref:2610.10698#pg5>. When considering both dwarfs and spirals together, a single universal SN efficiency of ηgb,2ph ≈ 0.02 (≈ 2%) can simultaneously reproduce the observed HI kinetic energy profiles across the star-forming main sequence<ref:2610.10698#pg5>. This value is within a factor 2 against uncertainties and is remarkably similar to that predicted by models of SN remnant evolution<ref:2610.10698#pg5>.

Implications and Discussion

The findings suggest that the SN efficiency remains similar across galaxies with different properties, offering valuable constraint for feedback recipes in galaxy evolution models<ref:2610.10698#pg5>. The volumetric star formation law is consistent with being driven by SN feedback only, where the gas velocity dispersion regulates the disc thickness through pressure-gravity balance<ref:2610.10698#pg4>. Furthermore, the required SN efficiency of ≈ 2% is within physically motivated limits of ≈ 10% predicted by SN remnant models<ref:2610.10698#pg5>. The study provides a benchmark for numerical and semi-analytical models of galaxy formation and evolution<ref:2610.10698#pg5>. The results indicate that SNe alone are sufficient to drive HI turbulence in dwarf galaxies, with no mandatory contribution from other sources of turbulence<ref:2610.10698#pg5>. This study offers a useful benchmark for numerical and semi-analytical models of galaxy formation and evolution<ref:2610.10698#pg5>. The universal SN efficiency obtained here could serve as an empirically-motivated value for setting up the fraction of SN energy that is transferred to the cold ISM in these sub-grid models<ref:2610.10698#pg5>. The future direction includes testing whether this SN efficiency is truly universal across cosmic time<ref:2610.10698#pg5>. The Square Kilometre Array may allow to push the studies on the origin of turbulence to even more extreme environments<ref:2610.10698#pg5>.

--- Page 20 ---

The results are statistically more robust, yet in agreement, with the median values of the characteristic SN efficiencies obtained by B20a<ref:2610.10698#pg5>. The universal SN efficiency obtained here could serve as an empirically-motivated value for setting up the fraction of SN energy that is transferred to the cold ISM in these sub-grid models, reducing the degrees of freedom<ref:2610.10698#pg5>. The most drastic change occurs in DDO 168, which clearly illustrates the degeneracy problem<ref:2610.10698#pg5>. In DDO 168, the SN efficiency found in this case is ηgb,2ph ≈ 0.013, much lower than the value η2ph ≈ 0.071 found in Sect. 4.1.3<ref:2610.10698#pg5>. This is because the WNM fraction increases from fw ≈ 0.180 to fw ≈ 0.907 to compensate for the different contribution of SN energy in the two cases<ref:2610.10698#pg5>. The resulting posteriors for both ηgb,2ph and fw are very well constrained in all three scenarios<ref:2610.10698#pg5>. In contrast to the characteristic SN efficiency, the resulting posteriors for both ηgb,2ph and fw are very well constrained in all three scenarios<ref:2610.10698#pg5>. The most drastic change occurs in DDO 168, which clearly illustrates the degeneracy problem<ref:2610.10698#pg5>. In DDO 168, the SN efficiency found in this case is ηgb,2ph ≈ 0.013, much lower than the value η2ph ≈ 0.071 found in Sect. 4.1.3<ref:2610.10698#pg5>. This is because the WNM fraction increases from fw ≈ 0.180 to fw ≈ 0.907 to compensate for the different contribution of SN energy in the two cases<ref:2610.10698#pg5>. The resulting posteriors for both ηgb,2ph and fw are very well constrained in all three scenarios<ref:2610.10698#pg5>. In contrast to the characteristic SN efficiency, the resulting posteriors for both ηgb,2ph and fw are very well constrained in all three scenarios<ref:2610.10698#pg5>. The most drastic change occurs in DDO 168, which clearly illustrates the degeneracy problem<ref:2610.10698#pg5>. In DDO 168, the SN efficiency found in this case is ηgb,2ph ≈ 0.013, much lower than the value η2ph ≈ 0.071 found in Sect. 4.1.3<ref:2610.10698#pg5>. This is because the WNM fraction increases from fw ≈ 0.180 to fw ≈ 0.907 to compensate for the different contribution of SN energy in the two cases<ref:2610.10698#pg5>. The resulting posteriors for both ηgb,2ph and fw are very well constrained in all three scenarios<ref:2610.10698#pg5>. In contrast to the characteristic SN efficiency, the resulting posteriors for both ηgb,2ph and fw are very well constrained in all three scenarios<ref:2610.10698#pg5>. The most drastic change occurs in DDO 168, which clearly illustrates the degeneracy problem<ref:2610.10698#pg5>. In DDO 168, the SN efficiency found in this case is ηgb,2ph ≈ 0.013, much lower than the value η2ph ≈ 0.071 found in Sect. 4.1.3<ref:2610.10698#pg5>.

Improvements for AI systems

Improved feedback recipes for galaxy evolution models can incorporate a universal SN efficiency of approximately 2% to drive HI turbulence across the star-forming main sequence, as this value is remarkably similar to that predicted by models of SN remnant evolution (e.g. Cimatti et al. 2019, Sect. 8.7).
AI systems can be trained to determine the optimal fraction of WNM gas based on observed HI kinetic energy profiles, as the study found that if both the SN efficiency and the WNM fraction are allowed to vary across galaxies, there is no obvious preference for HI being dominated by the WNM.
Numerical simulations of galactic feedback can be constrained by using a universal SN efficiency to reduce degrees of freedom in sub-grid prescriptions, as this value could serve as an empirically-motivated value for setting up the fraction of SN energy that is transferred to the cold ISM in these sub-grid models, reducing the degrees of freedom.
AI can be used to assess the robustness of turbulence drivers by comparing observed HI velocity dispersions against models constrained by a two-phase medium scenario, which was found to reproduce the observed energies very well in most galaxies, better than in the extreme cases discussed in Sects. 4.1.1 and 4.1.2.
AI systems can be refined to handle observational uncertainties by employing hierarchical Bayesian inference with non-centered parametrization, which has been shown to improve sampling efficiency when the true quantities are weakly informed by the data and to alleviate pathological posterior geometries, often described as 'funnel'-shaped structures.

Abstract

The origin of cold gas turbulence in galaxies has long been debated. Supernovae (SNe), the dominant form of stellar feedback, were thought insufficient to sustain it against dissipation, especially in galactic environments with low star formation rates (SFRs). We investigate the origin of the neutral atomic gas (HI) turbulence in dwarf irregular galaxies, a challenging regime of low masses and SFRs. We build on a recent study of spirals showing that disc flaring prolongs the dissipation timescale, reducing the energy needed to sustain turbulence and making SN feedback viable. We measure the SN efficiency, that is the fraction of SN energy required to sustain HI turbulence, in 14 dwarfs, and test whether a "universal" efficiency holds across a sample representative of the star-forming main sequence. From the observed HI distributions and kinematics, we derive the radial profiles of the HI kinetic energy per unit area and model them as the sum of thermal energy and turbulent energy from SNe, constrained by the observed SFR surface density. We consider pure cold and warm neutral medium (WNM) phases as limiting cases, as well as a mixed two-phase scenario, obtaining the SN efficiency and WNM fraction via hierarchical Bayesian inference. We find that the observed HI kinetic energy in dwarfs requires only 3% of the SN energy, and that a "universal" efficiency of about 2% explains the HI turbulence across our main-sequence sample. Our findings suggest that the SN efficiency remains similar across galaxies with different properties, offering valuable constraint for feedback recipes in galaxy evolution models.