[CII] as a global cold gas tracer in novel hydrodynamical simulations

arXiv:2608.22080 · astro-ph.GA · Submitted 2026-08-22 · Read on arXiv

Listen

Radio episode about this paper

Transcript

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

Vera: I'm Vera.

Jocelyn: And I'm Jocelyn, here with Subrahmanyan.

Vera: We're looking at the paper: "

CII: as a global cold gas tracer in novel hydrodynamical simulations".

Jocelyn: Here's the gist of the paper.

C II: emission is a powerful tool for studying the gas content in galaxies, especially at early stages of evolution <ref:2608.22080#pg2> Given its low excitation potential, the

C II: 158 µm line is the main coolant of neutral gas in photodissociation regions and giant molecular clouds, hence a tracer of all phases of the interstellar medium <ref:2608.22080#pg4> However, there is no consensus on what is the nature of the gas traced by this emission <ref:2608.22080#pg2>.

Vera: Simulation Setup and Methodology The study utilized four pre-prepared simulations involving an isolated galaxy and three merger configurations: a co-rotating, counter-rotating, and perpendicular major mergers of two Milky Way-mass-size galaxies <ref:2608.22080#pg6> The analysis employed a postprocessing semi-analytical model to simulate the

C II: emission.

Jocelyn: The simulations were performed using pgadget3-k, which includes the krome chemistry package The subgrid physics treats the ISM as a multiphase medium allowing coexistence between hot-diffuse gas and cold-dense gas Stars form from cold-dense gas, adopting an initial mass function by Salpeter with a lower and upper cut-off of 0.1 M⊙, and 120 M⊙ respectively.

Vera: Results on

C II: Tracers The results indicate that around 50 per cent of the total

C II: emission is traced by molecular gas, while 30 to 40 per cent is traced by atomic and ionised hydrogen regardless of the initial configuration of the merger simulation.

Paper summary: Vera: So, they set up four different simulations. One isolated galaxy and three different merger setups: co-rotating, counter-rotating, and perpendicular major mergers of Milky Way mass galaxies >

Jocelyn: They used a postprocessing semi-analytical model to simulate the CII emission for these systems >

Subrahmanyan: What’s interesting is that they reproduced fundamental relations like the CII luminosity and the star formation rate relation, called the LCII-SFR relation >

Vera: And they found that in mergers, there’s an extra component in the CII emission that you can’t explain just by looking at star formation alone >

Jocelyn: They also found that during quiet or starburst phases, galaxies actually deviate from that LCII-SFR relation >

Conclusion: Vera: Looking at the title of this paper, "CII as a global cold gas tracer in novel hydrodynamical simulations," it sounds like they’re trying to define where this emission fits into our understanding of galactic evolution >

Jocelyn: It seems the main implication is that the nature of CII emission isn't fixed; it really depends on how much interaction and mergers a galaxy has gone through over time >

Subrahmanyan: The paper suggests that mergers and starbursts become more efficient drivers of this CII emission because they cause collisions with molecular hydrogen >

Vera: So, it’s not just one thing tracing gas; it's the evolutionary history of the galaxy itself that dictates what you see in the CII signal >

Instituto de Astrofísica, Pontificia Universidad Católica de Chile · Centro de Astro-Ingeniería, Pontificia Universidad Católica de Chile

astro-ph.GA

Submitted: 2026-08-22

Updated: 2026-11-05

Comments: 14 pages, 12 figures, published in Astronomy and Astrophysics

Journal ref: A&A, 715, A114 (2026)

DOI: 10.1051/0004-6361/202661156

License: http://creativecommons.org/licenses/by-nc-sa/4.0/

Importance score: 83/100

The gist: [C II] emission is a powerful tool for studying the gas content in galaxies, especially at early stages of evolution <ref:2608.22080#pg2> Given its low excitation potential…

Key concepts

[CII] Emission
[CII] emission at 158 $\mu$m is a key tracer for neutral gas, especially in cold regions like giant molecular clouds. It is used to study the total gas content across different phases of the interstellar medium, revealing where neutral hydrogen resides.
L[CII]-SFR Relation
This relation connects the [CII] luminosity (a measure of gas) to the star formation rate (a measure of star birth). The paper shows that mergers create an extra component in this relation, meaning [CII] emission is not solely driven by the current rate of star formation alone.
Molecular Gas Tracing
The study found that approximately 50% of total [CII] luminosity originates from molecular gas (H2). This confirms that [CII] is a robust tracer for the total molecular content in simulated galaxies, regardless of their evolutionary stage or merger configuration.
Merger Effects on Gas
Galaxy mergers and starbursts act as efficient drivers for [CII] emission because they cause collisions with molecular hydrogen. These interactions can increase the amount of [CII] seen globally and change the fraction of gas traced by warm atomic hydrogen.

Terminology

Summary

[C II] emission is a powerful tool for studying the gas content in galaxies, especially at early stages of evolution <ref:2608. p. 0

Where the paper says it

Received 28 May 2026 ABSTRACT Context. [C II] emission is a powerful tool for studying the gas content in galaxies, especially at early stages of evolution.

22080#pg2> Given its low excitation potential, the [C II] 158 µm line is the main coolant of neutral gas in photodissociation regions and giant molecular clouds, hence a tracer of all phases of the interstellar medium <ref:2608. p. 0
Where the paper says it

[C II] emission is a powerful tool for studying the gas content in galaxies, especially at early stages of evolution. Given its low excitation potential, the [C II] 158 µm line is the main coolant of neutral gas in photodissociation regions and giant molecular clouds, hence a tracer of all phases of the interstellar medium.

22080#pg4> However, there is no consensus on what is the nature of the gas traced by this emission <ref:2608. p. 0
Where the paper says it

Given its low excitation potential, the [C II] 158 µm line is the main coolant of neutral gas in photodissociation regions and giant molecular clouds, hence a tracer of all phases of the interstellar medium. However, there is no consensus on what is the nature of the gas traced by this emission.

22080#pg2>.

Simulation Setup and Methodology

The study utilized four pre-prepared simulations involving an isolated galaxy and three merger configurations: a co-rotating, counter-rotating, and perpendicular major mergers of two Milky Way-mass-size galaxies <ref:2608.22080#pg6> The analysis employed a postprocessing semi-analytical model to simulate the [C II] emission. p. 0

Where the paper says it

We used four pre-prepared simulations performed with a version of p-gadget-3, which includes the krome chemistry package. We implemented a postprocessing semi-analytical model to simulate the [C II] emission.

The simulations were performed using pgadget3-k, which includes the krome chemistry package The subgrid physics treats the ISM as a multiphase medium allowing coexistence between hot-diffuse gas and cold-dense gas Stars form from cold-dense gas, adopting an initial mass function by Salpeter with a lower and upper cut-off of 0. p. 2
Where the paper says it

The ISM is treated as a multiphase medium that allows coexistence and material exchange between the hot-diffuse gas phase and the cold-dense gas phase (Scannapieco et al. Stars are assumed to form from the cold-dense gas, for which we adopt an initial mass function (IMF) by Salpeter (1955), with a lower and upper cut-off of 0.

1 M⊙, and 120 M⊙ respectively. p. 2
Where the paper says it

Stars are assumed to form from the cold-dense gas, for which we adopt an initial mass function (IMF) by Salpeter (1955), with a lower and upper cut-off of 0. 1 M⊙ , and 120 M⊙ , respectively.

Results on [C II] Tracers

The results indicate that around 50 per cent of the total [C II] emission is traced by molecular gas, while 30 to 40 per cent is traced by atomic and ionised hydrogen regardless of the initial configuration of the merger simulation. The [C II] luminosity-star formation rate relation, L[CII]-SFR, shows that mergers exhibit an extra component for the [C II] emission not traced by star formation alone. p. 0

Where the paper says it

We analysed an isolated Milky Way-mass-size galaxy, and three merger configurations: a co-rotating, counter-rotating, and perpendicular major mergers of two Milky Way-mass-size galaxies. We reproduce fundamental relations such as the [C II] luminosity and the star formation rate, L[CII]-SFR relation, in which mergers show an extra component for the [C II] emission not traced by the star formation alone.

During quiescent or starburst stages, galaxies deviate from the L[CII]-SFR relation. p. 0
Where the paper says it

We reproduce fundamental relations such as the [C II] luminosity and the star formation rate, L[CII]-SFR relation, in which mergers show an extra component for the [C II] emission not traced by the star formation alone. During quiescent or starburst stages, galaxies deviate from the L[CII]-SFR relation.

Evolution of Gas Phases During Interactions

The analysis of the [C II] emission fraction reveals how different gas phases contribute to the total luminosity over time. For merger simulations, the L[CII] on a global scale traces mostly H2 (∼ 50 per cent), while H I and H II account for 30-40 per cent of the total luminosity. p. 0

Where the paper says it

We find that in our simulations [C II] is a robust tracer of molecular gas, as well as atomic gas in both cold and warm phases on a global scale. Our results indicate that around 50 per cent of the total [C II] emission is traced by molecular gas, while 30 to 40 per cent is traced by atomic and ionised hydrogen regardless of the initial configuration of the merger simulation.

Interactions can decrease the fraction of the emission traced by the diffuse warm H I gas by ∼ 20-30 per cent. p. 9
Where the paper says it

3 (Harrington et al. Globally, the interaction and coalescence of the simulated galaxies produces a strong decrease in the L[CII] fraction traced by H I, by reducing it by ∼ 20-30 per cent.

Globally, simulated mergers show that H2 in diffuse state is a more predominant contributor to the total L[CII] compared to the Iso galaxy. p. 4
Where the paper says it

Hence, a comparison between simulated and observations should be done with caution. As the system evolves, the isolated galaxy (upper left panel) moves onto the observed relations, globally in good agreement with the trend reported by Herrera-Camus et al.

p. 4
Where the paper says it

Hence, a comparison between simulated and observations should be done with caution. As the system evolves, the isolated galaxy (upper left panel) moves onto the observed relations, globally in good agreement with the trend reported by Herrera-Camus et al.

Global Relations and Conclusions

The simulations successfully reproduce the observed L[CII]-SFR relation, with the best agreement with the relation of Schaerer et al. (2020) for high-redshift galaxies The [C II] emission is a robust tracer of molecular gas in simulated galaxies, even in different stages of their evolution. p. 6

Where the paper says it

In all the tested scenarios, the L[CII] correlates strongly with the total molecular mass. Therefore, L[CII] proves to be a robust tracer of the total molecular content in our simulated galaxies, even in different stages of their evolution.

The findings suggest that the nature of the [C II] emission depends strongly on the merger history of galaxies, in which mergers and starbursts can act as more efficient drivers of [C II] emission through collision with molecular hydrogen. p. 0
Where the paper says it

Conclusions. Our findings suggest that the nature of the [C II] emission depends strongly on the merger history of galaxies, in which mergers and starbursts can act as more efficient drivers of [C II] emission through collision with molecular hydrogen.

Future Directions

It remains to be explored the [C II] behaviour on resolved scales. A following paper will visit this idea by doing a spatially resolved study and looking for extended [C II] emission in galaxy mergers. p. 1

Where the paper says it

This extended emission is produced by cold gas in the tidal tails between the main galaxy and the satellite. Given that galaxy mergers can perturb the kinematics and modify the chemical signatures in interacting galaxies, studying how interactions influence [C II] emission is of particular interest to interpret observations.

The authors acknowledge support from the European Research Executive Agency HORIZON-MSCA-2021-SE-01 Research and Innovation programme under the Marie Skłodowska-Curie grant agreement number 101086388 (LACEGAL). p. 8
Where the paper says it

(2021) considers a dependence on metallicity so that at a given MHI, higher metallicity implies larger L[CII], we coloured by the median gas metallicity of each bin, with metallicity defined as the ratio of metals in the gas phase to the total gas mass Zgas = MZ/Mgas. The simulated L[CII] does have a dependence on gas metallicity in some cases, where for a given mass of H I, higher metallicity gas has higher [C II] luminosities in agreement with Heintz et al.

p. 8
Where the paper says it

(2021) considers a dependence on metallicity so that at a given MHI, higher metallicity implies larger L[CII], we coloured by the median gas metallicity of each bin, with metallicity defined as the ratio of metals in the gas phase to the total gas mass Zgas = MZ/Mgas. The simulated L[CII] does have a dependence on gas metallicity in some cases, where for a given mass of H I, higher metallicity gas has higher [C II] luminosities in agreement with Heintz et al.

p. 8
Where the paper says it

(2021) considers a dependence on metallicity so that at a given MHI, higher metallicity implies larger L[CII], we coloured by the median gas metallicity of each bin, with metallicity defined as the ratio of metals in the gas phase to the total gas mass Zgas = MZ/Mgas. The simulated L[CII] does have a dependence on gas metallicity in some cases, where for a given mass of H I, higher metallicity gas has higher [C II] luminosities in agreement with Heintz et al.

The authors thank the EvolGal4D team and the Highredshift Galaxies team for the interesting discussions throughout this work. p. 5
Where the paper says it

This points to a decoupling between the SFR and the L[CII], in which additional physical processes might also be contributing to the [C II] emission on a global scale in our simulations. This is particularly interesting as the L[CII]-SFR relation is usually estimated using star-forming galaxies.

p. 5
Where the paper says it

This points to a decoupling between the SFR and the L[CII], in which additional physical processes might also be contributing to the [C II] emission on a global scale in our simulations. This is particularly interesting as the L[CII]-SFR relation is usually estimated using star-forming galaxies.

p. 5
Where the paper says it

This points to a decoupling between the SFR and the L[CII], in which additional physical processes might also be contributing to the [C II] emission on a global scale in our simulations. This is particularly interesting as the L[CII]-SFR relation is usually estimated using star-forming galaxies.

The authors gratefully thank Manuel Aravena for the insightful discussions that helped define the scope of this work. p. 2
Where the paper says it

We aim to model [C II] emission self-consistently to investigate its evolution during galaxy mergers, to constrain the physical origin of the emission, and to provide predictions for future observations. It should be noted that providing predictions of how our simulated [C II] emission would be observed by ALMA is beyond the scope of this work and should include effects such as CMB supression, beam dilution, and sensitivity limitations.

p. 2
Where the paper says it

We aim to model [C II] emission self-consistently to investigate its evolution during galaxy mergers, to constrain the physical origin of the emission, and to provide predictions for future observations. It should be noted that providing predictions of how our simulated [C II] emission would be observed by ALMA is beyond the scope of this work and should include effects such as CMB supression, beam dilution, and sensitivity limitations.

p. 2
Where the paper says it

We aim to model [C II] emission self-consistently to investigate its evolution during galaxy mergers, to constrain the physical origin of the emission, and to provide predictions for future observations. It should be noted that providing predictions of how our simulated [C II] emission would be observed by ALMA is beyond the scope of this work and should include effects such as CMB supression, beam dilution, and sensitivity limitations.

PBT acknowledges partial funding by FondecytANID 1240465/2024. p. 1
Where the paper says it

(ORCHIDS, Aravena et al. 2024) using JWST/NIRSpec, which aims to understand the nature of the extended [C II] emission in star-forming galaxies at z ∼ 5 − 6.

p. 1
Where the paper says it

(ORCHIDS, Aravena et al. 2024) using JWST/NIRSpec, which aims to understand the nature of the extended [C II] emission in star-forming galaxies at z ∼ 5 − 6.

The authors acknowledge support from FONDECYT grant No. 1252054. PBT and JGL acknowledge support from ANID BASAL project FB210003. ES acknowledges funding by Fondecyt-ANID Postdoctoral 2024 Project N°3240644. p. 1
Where the paper says it

(ORCHIDS, Aravena et al. 2024) using JWST/NIRSpec, which aims to understand the nature of the extended [C II] emission in star-forming galaxies at z ∼ 5 − 6.

p. 1
Where the paper says it

(ORCHIDS, Aravena et al. 2024) using JWST/NIRSpec, which aims to understand the nature of the extended [C II] emission in star-forming galaxies at z ∼ 5 − 6.

p. 1
Where the paper says it

(ORCHIDS, Aravena et al. 2024) using JWST/NIRSpec, which aims to understand the nature of the extended [C II] emission in star-forming galaxies at z ∼ 5 − 6.

The authors acknowledge support from the European Research Executive Agency HORIZON-MSCA-2021-SE-01 Research and Innovation programme under the Marie Skłodowska-Curie grant agreement number 101086388 (LACEGAL). The authors thank the EvolGal4D team and the Highredshift Galaxies team for the interesting discussions throughout this work. The authors gratefully thank Manuel Aravena for the insightful discussions that helped define the scope of this work. The authors acknowledge support from FONDECYT grant No. 1252054. PBT and JGL acknowledge support from ANID BASAL project FB210003. ES acknowledges funding by Fondecyt-ANID Postdoctoral 2024 Project N°3240644. The authors acknowledge support from the European Research Executive Agency HORIZON-MSCA-2021-SE-01 Research and Innovation programme under the Marie Skłodowska-Curie grant agreement number 101086388 (LACEGAL). The authors thank the EvolGal4D team and the Highredshift Galaxies team for the interesting discussions throughout this work. The authors gratefully thank Manuel Aravena for the insightful discussions that helped define the scope of this work. PBT acknowledges partial funding by FondecytANID 1240465/2024. The authors acknowledge support from ANID BASAL project FB210003. ES acknowledges funding by Fondecyt-ANID Postdoctoral 2024 Project N°3240644.

Appendix Details

In Appendix A, the formation of a [C II] bridge for the Counter simulation during the 1st apocentre is analyzed as an example. The [C II] emission accounts for almost ∼ 10 per cent of the total L[CII] within this region at the first apocentre, pointing to a decoupling between the emission of [C II] and the presence of young and hot recently formed stars. p. 1

Where the paper says it

3 eV), it is the strongest emission line from cool gas with temperatures ≲ 104 K residing in galaxies. It accounts for up to 1 per cent of the total infrared luminosity (Stacey et al.

Model Validation

The model validation involved testing how well fundamental relations are reproduced using the Iso simulation as a control case. The isolated galaxy shows a clear correlation between L[CII] and SFR, with a Spearman correlation coefficient of 0. p. 4

Where the paper says it

(2015) and Schaerer et al. p. 1

Where the paper says it

2015) and for a diversity of redshift (De Looze et al. Schaerer et al.

The Iso case shows a clear correlation between L[CII] and SFR, with a Spearman correlation coefficient of 0.

98 (p < 0.001). The merger simulations follow the expected L[CII]-SFR relation, with the best agreement with the relation of Schaerer et al. p. 5
Where the paper says it

7 dex for the Iso, Corot, Counter, and 90- deg simulations, respectively. As we previously mentioned, our tested scenarios reproduce the observed L[CII]-SFR relation, with the best agreement with the relation of Schaerer et al.

(2020) for high-redshift galaxies. p. 5
Where the paper says it

As we previously mentioned, our tested scenarios reproduce the observed L[CII]-SFR relation, with the best agreement with the relation of Schaerer et al. (2020) for high-redshift galaxies.

The simulated galaxies deviate from the L[CII]-SFR relation when they are either starburst (τdep ∼ 0. p. 6
Where the paper says it

Considering a normal star-forming galaxy with a τdep= 1 Gyr, we show in blue colours the starburst phases and in brown the quiescent phases of the evolving galaxies in each simulation. The merger cases cross the observed relations when the galaxies are normal star-forming galaxies, and deviate from the relation when the galaxies are either in starburst phase with τdep∼ 0.

1 Gyr) or quiescent (τdep ∼ 10 Gyr), suggesting that the L[CII]-SFR might not hold in extreme phases of the evolution. p. 6
Where the paper says it

1 Gyr, or in quiescent phase with τdep∼ 10 Gyr (Kennicutt 1998). This might indicate that the L[CII]- SFR relation might not hold as well in galaxies in extreme phases of their evolution.

The simulations follow the expected relation and overall trend, with a median deviation for SFR of log(SFR/M⊙yr−1) ∼1, and with respect to the relation from Schaerer et al. p. 5
Where the paper says it

4 we discuss the nature of this remaining emitting gas. Our simulations follow the expected relation and overall trend, with a median deviation for SFR of log(SFR/M⊙yr−1 ) ∼1, and with respect to the relation from Schaerer et al.

(2020), of-0. p. 0
Where the paper says it

This is known as ‘CO-dark’ gas (Madden et al. 2020; Tacconi et al.

p. 0
Where the paper says it

This is known as ‘CO-dark’ gas (Madden et al. 2020; Tacconi et al.

7 dex, -0.9 dex, -0.9 dex, and-0. p. 5
Where the paper says it

(2020), of -0. 9 dex, and -0.

p. 5
Where the paper says it

(2020), of -0. 9 dex, and -0.

7 dex for the Iso, Corot, Counter, and 90-deg simulations respectively. p. 5
Where the paper says it

9 dex, and -0. 7 dex for the Iso, Corot, Counter, and 90- deg simulations, respectively.

p. 5
Where the paper says it

9 dex, and -0. 7 dex for the Iso, Corot, Counter, and 90- deg simulations, respectively.

The merger simulations cross the observed relations when the galaxies are normal star-forming galaxies. p. 6
Where the paper says it

Considering a normal star-forming galaxy with a τdep= 1 Gyr, we show in blue colours the starburst phases and in brown the quiescent phases of the evolving galaxies in each simulation. The merger cases cross the observed relations when the galaxies are normal star-forming galaxies, and deviate from the relation when the galaxies are either in starburst phase with τdep∼ 0.

Gas Phase Tracers

The total L[CII] correlates strongly with the total molecular mass, proving [C II] to be a robust tracer of the total molecular content in simulated galaxies. The simulations follow the observed relations with H2, showing mean deviations of 0. p. 6

Where the paper says it

2024), are also displayed. Our simulations follow the observed relations with H2, showing mean deviations of 0.

6 dex, 0.6 dex, 0.5 dex, and 0. p. 5
Where the paper says it

(2020), of -0. 9 dex, and -0.

3 dex at a H2 mass of log(MH2/M⊙) ∼9 with respect to the relation by Zanella et al. p. 6
Where the paper says it

5 dex, and 0. 3 dex at a H2 mass of log(MH2 /M⊙) ∼ 9 with respect to the relation by Zanella et al.

(2018) for the Iso, Corot, Counter, and 90-deg simulations respectively. p. 6
Where the paper says it

3 dex at a H2 mass of log(MH2 /M⊙) ∼ 9 with respect to the relation by Zanella et al. (2018) for the Iso, Corot, Counter, and 90-deg simulations, respectively.

The simulations follow the expected relations and overall trend, with a median deviation for SFR of log(SFR/M⊙yr−1) ∼1, and with respect to the relation from Schaerer et al. p. 5
Where the paper says it

4 we discuss the nature of this remaining emitting gas. Our simulations follow the expected relation and overall trend, with a median deviation for SFR of log(SFR/M⊙yr−1 ) ∼1, and with respect to the relation from Schaerer et al.

(2020), of-0.7 dex, -0.9 dex, -0.9 dex, and-0.7 dex for the Iso, Corot, Counter, and 90-deg simulations respectively. The relation of [C II] and H I with the gas metallicity exhibits a more complex behaviour. p. 8
Where the paper says it

Mergers are chaotic events that impact the chemical evolutionary history of the interacting galaxies, in particular the gas phase (Jara-Ferreira in prep, Perez et al. Therefore, as expected, the relation of [C II] and H I with the gas metallicity exhibits a more complex behaviour.

The best-fitting relation for the galaxy sample is log(L[CII]/L⊙) = (3. p. 8
Where the paper says it

Therefore, as expected, the relation of [C II] and H I with the gas metallicity exhibits a more complex behaviour. We estimate the best-fitting relation for the galaxy sample, across their evolution and including all the simulations as log(L[CII]/L⊙) = (3.

78 ± 0.11) + (0.32 ± 0.01) × log(MHI/M⊙)+(−0. p. 7
Where the paper says it

09)×log(Z/Z⊙) + (1. 12) +log(MHI/M⊙) by Heintz et al.

18±0.04)×log(Z/Z⊙). p. 1
Where the paper says it

Schaerer et al. (2020) obtained a relation of log(L[CII]/L⊙) = 6.

The simulated galaxies move across the observed regions. p. 8
Where the paper says it

04)×log(Z/Z⊙). In all merger experiments, the simulated galaxies move across the observed regions.

The simulations allow us to investigate the distribution of L[CII] across different tracers as done above. p. 9
Where the paper says it

In a followup paper, we explore the spatial extent of this stripped gas and compare how the different configurations of mergers affect this outcome. Simulations allow us to investigate the distribution of L[CII] across different tracers as done above, an analysis that cannot be performed directly with observational data.

Gas Phase Contributions

For merger simulations, [C II] emission on a global scale traces mostly H2 (∼ 50 per cent), while H I and H II account for 30-40 per cent of the total luminosity. The simulations show that interactions can decrease the fraction of the emission traced by the diffuse warm H I gas by ∼ 20-30 per cent. p. 0

Where the paper says it

We find that in our simulations [C II] is a robust tracer of molecular gas, as well as atomic gas in both cold and warm phases on a global scale. Our results indicate that around 50 per cent of the total [C II] emission is traced by molecular gas, while 30 to 40 per cent is traced by atomic and ionised hydrogen regardless of the initial configuration of the merger simulation.

Globally, simulated mergers show that H2 in diffuse state is a more predominant contributor to the total L[CII] compared to the Iso galaxy. The peak in L[CII] around the first pericentre corresponds to gas being ionised by the interaction and decoupled from SF. p. 4
Where the paper says it

During the first percentre, for both Corot and Counter mergers, a peak in L[CII] is accompanied by a decrease in SFR. This points to emitting gas that is not directly linked to SF, but rather to gas probably shocked during the interaction.

This stripped gas continues to emit long after the coalescence of the galaxy pair. p. 9
Where the paper says it

This gas continues to emit [C II] forming a bridge during the 1st apocentre, and remains constantly emitting after the coalescence (see appendix A). In a followup paper, we explore the spatial extent of this stripped gas and compare how the different configurations of mergers affect this outcome.

Final Summary

The authors conclude that [C II] emission involves a complex interplay of different physical properties of the gas phase. The study provides new insights into how [C II] behaves as a cold gas tracer on a global scale during interactions and mergers of controlled simulated systems. p. 2

Where the paper says it

: [C II] as a global cold gas tracer in novel hydrodynamical simulations ting [C II] around merging systems at high redshift. Some numerical works have studied how dwarf galaxy mergers affect the SFR of the interacting galaxies (Bisbas et al.

The authors suggest that interactions/mergers are efficient mechanisms behind the ionisation of the emitting gas in their simulations. p. 6
Where the paper says it

short depletion times), it is not in conditions of forming stars. This indicates a more complex evolution during mergers, which affects the thermodynamical properties of the gas in conditions to emit in [C II] , and suggests that interactions/mergers are efficient mechanisms behind the ionisation of the emitting gas in our simulations.

The authors note that it remains to be explored the [C II] behaviour on resolved scales.

Appendix B Constants

Table B.1 presents the constants used to estimate the L[CII] and the respective reference following Casavecchia et al. p. 4

Where the paper says it

: [C II] as a global cold gas tracer in novel hydrodynamical simulations total C+ (nC+ ) Λ[CII] = A21∆E21· nC+ (nH2 γ H2 12 + nHIγ HI 12 + neγ e 12) nH2 (γ H2 12 + γ H2 21 ) + nHI(γ HI 12 + γ HI 21 ) + ne(γ e 12 + γ e 21) + A21 (2) where γ H2 12 , γ HI 12 , and γ e 12 are the excitation collisional rates, and γ H2 21 , γ HI 21 , and γ e 21 are the de-excitation collisional rates for H2, H I, and e respectively. We display the excitation collisional rates and constants used in Table B.

(2025). The constants include γe 21 2. p. 4
Where the paper says it

: [C II] as a global cold gas tracer in novel hydrodynamical simulations total C+ (nC+ ) Λ[CII] = A21∆E21· nC+ (nH2 γ H2 12 + nHIγ HI 12 + neγ e 12) nH2 (γ H2 12 + γ H2 21 ) + nHI(γ HI 12 + γ HI 21 ) + ne(γ e 12 + γ e 21) + A21 (2) where γ H2 12 , γ HI 12 , and γ e 12 are the excitation collisional rates, and γ H2 21 , γ HI 21 , and γ e 21 are the de-excitation collisional rates for H2, H I, and e respectively. We display the excitation collisional rates and constants used in Table B.

8 × 10−7T −0.5 100 cm3 s−1 (3), γHI 21 8 × 10−10T 0. p. 2
Where the paper says it

First, we follow the evolution of [C II] and C+self-consistently along and after the coalescence. Second, our chemical model is able to account for cold gas at temperatures lower than 100 K, thus we also can separate the hydrogen gas in each phase, which allow us to effectively follow each component self-consistenlty over time.

07 100 cm3 s−1 (3), and γH2 21 3. p. 2
Where the paper says it

First, we follow the evolution of [C II] and C+self-consistently along and after the coalescence. Second, our chemical model is able to account for cold gas at temperatures lower than 100 K, thus we also can separate the hydrogen gas in each phase, which allow us to effectively follow each component self-consistenlty over time.

8 × 10−10T 0.14 100 cm3 s−1 (2). p. 2
Where the paper says it

First, we follow the evolution of [C II] and C+self-consistently along and after the coalescence. Second, our chemical model is able to account for cold gas at temperatures lower than 100 K, thus we also can separate the hydrogen gas in each phase, which allow us to effectively follow each component self-consistenlty over time.

The constants also include A21 2. p. 4
Where the paper says it

: [C II] as a global cold gas tracer in novel hydrodynamical simulations total C+ (nC+ ) Λ[CII] = A21∆E21· nC+ (nH2 γ H2 12 + nHIγ HI 12 + neγ e 12) nH2 (γ H2 12 + γ H2 21 ) + nHI(γ HI 12 + γ HI 21 ) + ne(γ e 12 + γ e 21) + A21 (2) where γ H2 12 , γ HI 12 , and γ e 12 are the excitation collisional rates, and γ H2 21 , γ HI 21 , and γ e 21 are the de-excitation collisional rates for H2, H I, and e respectively. We display the excitation collisional rates and constants used in Table B.

4 × 10−6 s−1 (1) and ∆E21 1. p. 3
Where the paper says it

2012; Casavecchia et al. 2025) we can express Λ[CII] ≡ nC+,2A21∆E21 (1) where nC+,2 is the number density of the C+ ions in the excited upper state, A21 = 2.

259 × 10−14erg (4). p. 3
Where the paper says it

4 × 10−6 s −1 (Draine 2011) is the Einstein coefficient for spontaneous emission, and ∆E21 = 1. 259×10−14 erg (Santoro & Shull 2006) is the energy separation between levels.

The excitation temperature is Texc = E21/k = 91. p. 4
Where the paper says it

We display the excitation collisional rates and constants used in Table B. 1, while we obtain the de-excitation collisional rates from the relation given by the detailed balance γ k 12 = (g2/g1)γ k 21 exp (−E21/KT), k = H2, HI, e (3) with g1, g2 are the statistical weights, which for the J1 = 1/2 −→ J2 = 3/2 jump corresponds to g1 = 2 and g2 = 4, and the excitation temperature to Texc = E21/k = 91.

2 K (3).

Appendix A Bridge Formation

Appendix A analyzes the formation of a [C II] bridge for the Counter simulation during the 1st apocentre as an example. The projection maps show that only 0. p. 4

Where the paper says it

As an example, in Fig. 2, face-on projection maps of the ionised carbon mass (upper rows) and of the [C II] luminosity, L[CII] (bottom rows) at different stages of the evolution of the Iso (upper panel) and Corot (bottom panel) simulations are displayed.

83 per cent of the total instantaneous SFR is located in the bridge, while [C II] emission accounts for almost ∼ 10 per cent. p. 0
Where the paper says it

We find that in our simulations [C II] is a robust tracer of molecular gas, as well as atomic gas in both cold and warm phases on a global scale. Our results indicate that around 50 per cent of the total [C II] emission is traced by molecular gas, while 30 to 40 per cent is traced by atomic and ionised hydrogen regardless of the initial configuration of the merger simulation.

The predominance of [C II] over the star formation rate in the gas bridge is prevalent during the interaction, pointing to a decoupling between the emission of [C II] and the presence of young and hot recently formed stars. p. 1
Where the paper says it

Additionally, [C II] has also been proposed as a tracer of the star formation rate (SFR) of galaxies. This is because newly formed O and B stars produce far-ultraviolet photons that heat the gas on dust grains and polycyclic aromatic hydrocarbons via photoelectric effect (Helou et al.

Figure A.1 Details

Fig. A.1 shows projection maps of instantaneous SFR (upper panel) and L[CII] (bottom panel) for the Counter merger at the moment of 1st apocentre. p. 4

Where the paper says it

As an example, in Fig. 2, face-on projection maps of the ionised carbon mass (upper rows) and of the [C II] luminosity, L[CII] (bottom rows) at different stages of the evolution of the Iso (upper panel) and Corot (bottom panel) simulations are displayed.

White rectangle denotes a 20 kpc×25 kpc region enclosing the gas bridge between the galaxy pair. p. 4
Where the paper says it

As seen, the L[CII] traces the spiral structure and arms in the Iso experiment, pointing to a connection between star formation and L[CII], given that these are zones of active star formation. In the mergers, the L[CII] mainly traces the central parts of the galaxy, where star formation is occurring, as well as the stripped gas from the central parts of the galaxies, which forms a bridge between the galaxy pair.

Figure A.1 Inset Labels

Inset labels show the percentage of the total instantaneous SFR (upper panel) and of total L[CII] (bottom panel) within this region at the first apocentre. The inset labels show that only 0. p. 7

Where the paper says it

After the galaxies merge, the entire system is depicted as a whole. Each coloured circle represents the galaxy at a given merger time (inset labels).

83 per cent of the total instantaneous SFR is located in the bridge, and [C II] emission accounts for almost ∼ 10 per cent. p. 1
Where the paper says it

3 eV), it is the strongest emission line from cool gas with temperatures ≲ 104 K residing in galaxies. It accounts for up to 1 per cent of the total infrared luminosity (Stacey et al.

Figure A.1 L[CII] Projection

Fig. A.1 shows projection maps of ionised carbon mass (upper row) and of L[CII] (bottom rows) at different stages of the evolution of the Iso (upper panel) and Corot (bottom panel) simulations. p. 4

Where the paper says it

As an example, in Fig. 2, face-on projection maps of the ionised carbon mass (upper rows) and of the [C II] luminosity, L[CII] (bottom rows) at different stages of the evolution of the Iso (upper panel) and Corot (bottom panel) simulations are displayed.

The projections along the z-axis have been done considering a pixel size of 0. p. 4
Where the paper says it

2, face-on projection maps of the ionised carbon mass (upper rows) and of the [C II] luminosity, L[CII] (bottom rows) at different stages of the evolution of the Iso (upper panel) and Corot (bottom panel) simulations are displayed. These projections along the z-axis have been done considering a pixel size of 0.

5 kpc to be consistent with the spatial resolution of the simulations. p. 4
Where the paper says it

These projections along the z-axis have been done considering a pixel size of 0. 5 kpc to be consistent with the spatial resolution of the simulations.

As seen, L[CII] traces the spiral structure and arms in the Iso experiment, pointing to a connection between star formation and L[CII], given that these are zones of active star formation. p. 4
Where the paper says it

5 kpc to be consistent with the spatial resolution of the simulations. As seen, the L[CII] traces the spiral structure and arms in the Iso experiment, pointing to a connection between star formation and L[CII], given that these are zones of active star formation.

In the mergers, the L[CII] mainly traces the central parts of the galaxy, where star formation is occurring, as well as the stripped gas from the central parts of the galaxies, which forms a bridge between the galaxy pair. p. 4
Where the paper says it

As seen, the L[CII] traces the spiral structure and arms in the Iso experiment, pointing to a connection between star formation and L[CII], given that these are zones of active star formation. In the mergers, the L[CII] mainly traces the central parts of the galaxy, where star formation is occurring, as well as the stripped gas from the central parts of the galaxies, which forms a bridge between the galaxy pair.

Figure 3 Evolution

Fig. 3 shows L[CII] (black lines) and SFR (grey lines) evolution in time for our tested scenarios: Iso, Corot, Counter, and 90-deg mergers. p. 6

Where the paper says it

L[CII] (black lines) and SFR (grey lines) evolution in time. Each row shows the evolution of these quantities for our tested scenarios: Iso, Corot, Counter, and 90-deg mergers.

For the mergers, key moments are shown as coloured lines as shown in the bottom panel. p. 6
Where the paper says it

Each row shows the evolution of these quantities for our tested scenarios: Iso, Corot, Counter, and 90-deg mergers. For the mergers, key moments are shown in coloured lines as shown in the bottom panel.

Figure 4 L[CII]-SFR Relation

Fig. 4 shows the L[CII]-SFR relation for each simulation, colour-coded by time. p. 4

Where the paper says it

Thus, in Fig. 4 we show the L[CII]-SFR relation for each simulation, colour-coded by time.

The isolated galaxy is in good agreement with the observed relations by HerreraCamus et al. p. 4
Where the paper says it

Hence, a comparison between simulated and observations should be done with caution. As the system evolves, the isolated galaxy (upper left panel) moves onto the observed relations, globally in good agreement with the trend reported by Herrera-Camus et al.

(2015) and Schaerer et al. (2020). After the coalescence, the resulting systems are located within the observed range, settling into a constant L[CII] (log (L[CII]/L⊙) ∼ 6. p. 4
Where the paper says it

After this, the galaxies get quenched and their SF ceases, which translates into lower a L[CII], showing a loop. For the Corot and Counter cases, the merged galaxy settles into a constant L[CII], that is, for a diversity of SFR (log (SFR/M⊙) = [−2, 0]), constant L[CII] is observed, which shows as a cluster of circles around Article number, page 5 of 15

5-7), that is, for a diversity of SFRs, a determined L[CII] is observed. p. 4
Where the paper says it

After this, the galaxies get quenched and their SF ceases, which translates into lower a L[CII], showing a loop. For the Corot and Counter cases, the merged galaxy settles into a constant L[CII], that is, for a diversity of SFR (log (SFR/M⊙) = [−2, 0]), constant L[CII] is observed, which shows as a cluster of circles around Article number, page 5 of 15

Figure 5 Oxygen Abundance Relation

Fig. 5 shows the L[CII]-SFR relation for the tested scenarios coloured-coded by the oxygen abundance of the galaxy’s gas phase. p. 7

Where the paper says it

6 12 + log(O/H) Fig. L[CII]-SFR relation for the tested scenarios coloured-coded by the oxygen abundance of the galaxy’s gas phase.

The plateau formed around log (L[CII]/L⊙) ∼ 6. p. 6
Where the paper says it

4) given that, as the galaxies evolve, they form stars that will enrich the gas phase with metals. The plateau formed around log (L[CII]/L⊙) ∼ 6.

5-7 in the Corot and Counter mergers does not show a particular dependence on the level of abundance of oxygen, but rather has a constant value. p. 6
Where the paper says it

The plateau formed around log (L[CII]/L⊙) ∼ 6. 5-7 in the Corot and Counter mergers, does not show a particular dependence on the level of abundance of oxygen, but rather has a constant value.

For the Iso case, higher oxygen abundances lie at lower L[CII] and SFR values. p. 6
Where the paper says it

5-7 in the Corot and Counter mergers, does not show a particular dependence on the level of abundance of oxygen, but rather has a constant value. For the Iso case, we see that higher oxygen abundances lie at lower L[CII] and SFR values.

The 90-deg case shows a weak dependence on the level of oxygen enrichment, in which at a given SFR, higher L[CII] translates into higher oxygen abundances. p. 6
Where the paper says it

For the Iso case, we see that higher oxygen abundances lie at lower L[CII] and SFR values. The 90- deg case shows a weak dependence on the level of oxygen enrichment, in which at a given SFR, higher L[CII] translates into higher oxygen abundances.

Figure 6 Depletion Time Relation

Fig. 6 studies the L[CII]-SFR relation for the tested scenarios coloured-coded by the galaxy’s depletion time. p. 7

Where the paper says it

0 lo g( d e p G yrs 1 ) Fig. L[CII]-SFR relation for the tested scenarios coloured-coded by the galaxy’s depletion time.

The merger cases cross the observed relations when the galaxies are normal star-forming galaxies, and deviate from the relation when they are either in starburst phase with τdep∼ 0. p. 6
Where the paper says it

Considering a normal star-forming galaxy with a τdep= 1 Gyr, we show in blue colours the starburst phases and in brown the quiescent phases of the evolving galaxies in each simulation. The merger cases cross the observed relations when the galaxies are normal star-forming galaxies, and deviate from the relation when the galaxies are either in starburst phase with τdep∼ 0.

1 Gyr, or in quiescent phase with τdep∼ 10 Gyr. p. 6
Where the paper says it

Considering a normal star-forming galaxy with a τdep= 1 Gyr, we show in blue colours the starburst phases and in brown the quiescent phases of the evolving galaxies in each simulation. The merger cases cross the observed relations when the galaxies are normal star-forming galaxies, and deviate from the relation when the galaxies are either in starburst phase with τdep∼ 0.

Figure 7 Specific SFR Relation

Fig. 7 shows the L[CII]-SFR relation for the tested scenarios coloured-coded by the galaxy’s specific star formation rate. p. 8

Where the paper says it

5 log(sSFR/yr) Fig. L[CII]-SFR relation for the tested scenarios coloured-coded by the galaxy’s specific star formation rate.

Systems that deviate from the relation are either very efficient or very inefficient in transforming gas into stars, implying that even if there is gas available (i. p. 6
Where the paper says it

7 we show the L[CII]-SFR plane coloured by the specific star formation rate, sSFR = SFR/M∗. Systems that deviate from the relation are either very efficient or very inefficient in transforming gas into stars, implying that even if there is gas available (i.

e. short depletion times), it is not in conditions of forming stars. p. 6
Where the paper says it

Systems that deviate from the relation are either very efficient or very inefficient in transforming gas into stars, implying that even if there is gas available (i. short depletion times), it is not in conditions of forming stars.

Figure 8 Molecular Gas Mass Relation

Fig. 8 shows the total L[CII] as a function of the system’s molecular gas mass H2. p. 8

Where the paper says it

Total L[CII] as a function of the system’s molecular gas mass H2. Each column shows the global trend for the tested scenarios as the median value in mass bins of 0.

In all tested scenarios, L[CII] correlates strongly with the total molecular mass. p. 6
Where the paper says it

(2020) is estimated for the Iso, Corot, Counter, and 90-deg simulations, respectively. In all the tested scenarios, the L[CII] correlates strongly with the total molecular mass.

The best fitting relation for the whole sample is log(L[CII]/L⊙) = (1. p. 8
Where the paper says it

Therefore, as expected, the relation of [C II] and H I with the gas metallicity exhibits a more complex behaviour. We estimate the best-fitting relation for the galaxy sample, across their evolution and including all the simulations as log(L[CII]/L⊙) = (3.

11 ± 0.16) + (0.63 ± 0.02) × log(MH2/M⊙), in very good agreement with the reported observational slopes. p. 6
Where the paper says it

We estimate the bestfitting relation for the whole sample of galaxies, including all the simulations and all snapshots, as log(L[CII]/L⊙) = (1. 02) × log(MH2 /M⊙), in very good agreement with the reported observational slopes.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improve hydrodynamical simulation models to include self-consistent tracing of different gas phases based on temperature, as this allows for a self-consistent tracking of each component over time. This enables the system to differentiate between H2, H I, and H II contributions to [C II] emission more accurately than current methods which might oversimplify the gas phase.

  2. Enhance the simulation subgrid physics by improving the treatment of cold dense gas and molecular clouds to better resolve their substructure or account for possibly missed high-density regions due to numerical resolution through a more refined clumping factor Cp approach. This will improve accuracy in tracing emission from these specific phases, which are crucial during mergers.

  3. Develop postprocessing models that can provide predictions for future observations by incorporating effects such as CMB suppression, beam dilution, and sensitivity limitations. This will allow researchers to translate simulation results into more realistic observational constraints when comparing against ALMA data.

  4. Implement a method to robustly distinguish between emission sources during merger events by analyzing the evolution of the L[CII] fraction traced by H2, H I, and H II across different temporal scales. This helps in understanding whether the peak in L[CII] around the first pericentre corresponds to star formation or gas being decoupled from SF.

  5. Integrate a metallicity dependence into the [C II]-SFR relation modeling to better capture how more enriched gas tends to have higher luminosities, due to the larger availability of carbon in the medium. This will improve predictions for galaxies at different stages of evolution based on their chemical history.

Abstract

[CII] emission is a powerful tool for studying the gas content in galaxies, especially at early stages of evolution. Given its low excitation potential, the [CII] 158 μm line is the main coolant of neutral gas in photodissociation regions and giant molecular clouds, hence a tracer of all phases of the interstellar medium. However, there is no consensus on what is the nature of the gas traced by this emission. We aim to provide insights into the physics of the [CII] emission and predictions for future observations. We used four pre-prepared simulations performed with a version of p-gadget-3, which includes the krome chemistry package. We implemented a postprocessing semi-analytical model to simulate the [CII] emission. We analysed an isolated Milky Way-mass-size galaxy, and three merger configurations: a co-rotating, counter-rotating, and perpendicular major mergers of two Milky Way-mass-size galaxies. We reproduce fundamental relations such as the [CII] luminosity and the star formation rate, L[CII]-SFR relation, in which mergers show an extra component for the [CII] emission not traced by the star formation alone. During quiescent or starburst stages, galaxies deviate from the L[CII]-SFR relation. We find that [CII] is a robust tracer of molecular gas, as well as atomic gas in both cold and warm phases on a global scale. Our results indicate that around 50 per cent of the total [CII] emission is traced by molecular gas, while 30 to 40 per cent is traced by atomic and ionised hydrogen regardless of the configuration. Our findings suggest that the nature of the [CII] emission depends strongly on the evolutionary history of galaxies, in which mergers and starbursts act as more efficient drivers of [CII] emission through collision with molecular hydrogen.

Sources

Related papers