Reconstructing the Globular Cluster Initial Mass Function from Present-day Globular Cluster Systems

arXiv:2605.27509 · astro-ph.GA · Submitted 2026-05-26 · 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: "Reconstructing the Globular Cluster Initial Mass Function from Present-day Globular Cluster Systems".

Jocelyn: Here's the gist of the paper. The paper reconstructs the high-mass regime of the Globular Cluster Initial Mass Function (GCIMF) by inverting mass loss due to dynamical evolution across a wide range of host galaxy masses, which is crucial for understanding the initial conditions of globular cluster systems The Core Problem and Methodology The near-universal turnover mass of the present-day globular cluster mass function (GCMF), MTO ∼ 2 × 10 5 M⊙, is a well established observational feature across galaxies of different types and masses, providing an important empirical benchmark for understanding the globular cluster initial mass function (GCIMF) Competing explanations of this property invoke either dynamical evolution from an initial power-law distribution or an imprint of cluster formation physics We address this problem by reconstructing the high-mass regime of the GCIMF by inverting the mass loss due to dynamical evolution of present-day globular cluster systems across a wide range of host galaxy masses Our method is based on an environment-dependent mass-loss model calibrated by direct N-body simulations in time-dependent tidal fields, enabling a mapping between observed cluster masses and their progenitor values without assuming a priori a functional form for the GCIMF Data Sources and Sample Construction To obtain good statistics and cover a wide range of galaxy types and masses, the authors compile a large sample of GC systems from different public sources The resulting sample combines galaxies with individually measured GC masses, photometric GC catalogs from the ACS Virgo Cluster Survey, and large statistical catalogs for which only integrated properties of the GC system are available For galaxies where individual GC masses are available, such as the Milky Way (MW) and nearby systems, they apply the mass-loss inversion cluster by cluster They also construct synthetic present-day GC systems for the dwarf galaxies analyzed in Paper I Together, these datasets provide 451 present-day GC mass distributions across a wide range of galaxy masses and environments Dynamical Modeling and Inversion The long-term dynamical mass evolution of GCs is modeled using an environment dependent calibration introduced and tested in Paper I The forward evolution of a cluster over time t can be written as M(t) = M0

1 − A Mα 0 t: To reconstruct progenitor masses, they invert this relation for each observed present-day cluster mass Mobs by numerically solving Mobs = M0

1 − A Mα 0 t: This formulation assumes that the average mass-loss rate remains approximately constant over the lifetime of the cluster, such that cumulative mass loss scales linearly with time Key Findings on GCIMF Variation The reconstructed GCIMF is built in logarithmic mass bins, and they characterize the high-mass end of the distribution by fitting a power law of the form dN/d log M ∝ M−β The fit is performed in log-log space above the turnover region The resulting slopes, β, vary systematically across galaxies and show a positive correlation with host galaxy halo mass Specifically, more massive galaxies tend to exhibit steeper reconstructed slopes Implications for Cluster Formation Models The study suggests that the commonly assumed universal power-law GCIMF may not hold across all galaxy types and masses Instead, the initial cluster mass distribution and/or the efficiency of cluster survival may depend systematically on host galaxy properties The agreement between independent approaches supports a picture in which GCs form with a power-law-like mass distribution at early times, which is subsequently modified by environmental effects and dynamical mass loss This has important implications for models of GC formation in cosmological simulations and for the interpretation of highredshift proto-globular cluster candidates observed with the James Webb Space Telescope The method constrains robustly the upper portion of the GCIMF, while the low-mass regime remains uncertain due to the preferential disruption of low-mass clusters across cosmic time The authors present a novel approach to link observed present-day GCs with their progenitor populations, providing a bridge between local GC systems and their high-redshift origins This implies that the degree of modification depends on host galaxy mass Future work should extend this analysis to larger galaxy samples and incorporate models that account for the disruption of clusters no longer present today The reconstructed GCIMF should be interpreted as the progenitor mass distribution of surviving clusters rather than the full initial cluster population The high-mass regime provides a more reliable probe of the GCIMF shape The authors find that dwarf galaxies, where cluster disruption is inefficient, closely approximate the true initial distribution APPENDIX A This appendix presents the reconstructed GCIMFs for the remaining galaxies with individually measured GC masses discussed in Subsection 2.1 These figures complement the Milky Way example shown in Section 4 and illustrate the reconstruction procedure across different galaxy environments and mass scales, including M31 and nearby dwarf galaxy systems The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.50 ± 0.15 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 0.76 ± 0.11 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.76 ± 0.15 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.05 ± 0.08 in dN/d log M The scatter of the data around the linear relation, σ = 0.368, reflects the galaxy to galaxy variation in β The best fit parameters are β = (0.699 ± 0.038) log(Mh) − (6.698 ± 0.428) This confirms that the linear fit captures the average behavior and establishes that more massive galaxies tend to host reconstructed GCIMFs with steeper high-mass slopes The scatter of the data around the linear relation, σ = 0.368, reflects the galaxy to galaxy variation in β REFERENCES Adamo, A., Bradley, L.

Vera: D., Vanzella, E., et al.

Jocelyn: 2024, Nature, 632, 513 Andersson, E.

Vera: P., Mac Low, M.-M., Agertz, O., Renaud, F., & Li, H.

Title and authors: Vera: So we're looking at this paper titled "Reconstructing the Globular Cluster Initial Mass Function from Present-day Globular Cluster Systems," and the authors Elizabeth Moreno-Hilario, Hui Li, and Luis A. Martinez-Medina are tackling a big problem here.

Jocelyn: Basically, they're trying to figure out what happened to globular clusters in the first place by looking at the ones we see today across all kinds of galaxies.

Subrahmanyan: It sounds like they are trying to bridge that gap between how we see the present-day population and what actually formed when these systems were first assembling.

Vera: That's right, and what they do is invert the mass loss due to dynamical evolution across a wide range of host galaxy masses. They use N-body simulations calibrated by direct N-body simulations in time-dependent tidal fields to map observed cluster masses back to their progenitor values without assuming a specific functional form for the initial mass function beforehand.

Jocelyn: So, they are using those simulated tidal fields to essentially rewind the clock on how much mass was stripped away over cosmic time, which is smart because it connects what we measure now to what it might have been then.

Subrahmanyan: And this moves us past just assuming a universal power-law distribution for the initial mass function, suggesting instead that the initial cluster mass distribution and or the efficiency of cluster survival may depend systematically on host galaxy properties.

Vera: They found that when they reconstructed the high-mass end of their GCIMF, they characterized it by fitting a power law of the form dN/d log M proportional to M to minus beta, and this slope beta varies systematically across galaxies and shows a positive correlation with host galaxy halo mass.

Jocelyn: That dependence on host galaxy mass is significant because it suggests that more massive galaxies tend to exhibit steeper reconstructed slopes for the initial mass function.

Subrahmanyan: If this holds up, it implies that the commonly assumed universal power-law GCIMF may not be consistent across all galaxy types and masses, meaning the initial cluster mass distribution might be modified by environmental effects and dynamical mass loss after formation.

Title and authors: Vera: The paper suggests a picture where GCs form with a power-law like mass distribution at early times, which is then subsequently modified by environmental effects and dynamical mass loss as they evolve. This has important implications for models of GC formation in cosmological simulations and for the interpretation of high-redshift proto-globular cluster candidates observed with the James Webb Space Telescope.

Jocelyn: They also noted that the low-mass regime remains uncertain because of the preferential disruption of low-mass clusters across cosmic time, which is a limitation they point out.

Subrahmanyan: And they do say that their method reconstructs only the progenitor mass distribution of surviving clusters, rather than the full initial cluster population, which means we should interpret this high-mass regime as a more reliable probe of the GCIMF shape.

Vera: To get to this reconstruction, they compiled a large sample of GC systems from different public sources combining galaxies with individually measured masses and photometric catalogs from things like the ACS Virgo Cluster Survey.

Jocelyn: They also constructed synthetic present-day GC systems for dwarf galaxies analyzed in a prior paper, which helps them cover a wide range of galaxy masses and environments, resulting in four hundred fifty-one present-day GC mass distributions across that spectrum.

Subrahmanyan: The methodology is based on an environment dependent mass-loss model calibrated by direct N-body simulations in time-dependent tidal fields, enabling a mapping between observed cluster masses and their progenitor values without assuming a priori a functional form for the GCIMF.

Vera: They used the forward evolution equation M(t) equals M0 one minus A M alpha zero t, and they invert this relation for each observed present-day cluster mass, Mobs, by numerically solving that equation to reconstruct progenitor masses.

Jocelyn: That formulation assumes that the average mass-loss rate stays approximately constant over the lifetime of the cluster, so cumulative mass loss scales linearly with time in their model.

Title and authors: Subrahmanyan: The key finding here is the systematic variation in slopes, beta, across different systems and galaxies, which they found to be captured by a linear relation where beta equals zero point six nine nine plus or minus zero three eight times the logarithm of the host galaxy mass minus six point six nine eight plus or minus zero four two eight.

Vera: That linear fit captures the average behavior, and it confirms that more massive galaxies tend to host reconstructed GCIMFs with steeper high-mass slopes, which is a direct result of their analysis.

Jocelyn: So, what this means for us is that the degree of modification depends on the host galaxy mass. Dwarf galaxies, where cluster disruption is inefficient, closely approximate the true initial distribution according to their findings.

Subrahmanyan: If you look at these results in the bigger cosmic picture, it supports a scenario where GCs form with a power-law mass distribution early on, and that initial distribution is then modified by environmental effects as they evolve.

Vera: The paper provides an empirical bridge between local GC systems and their high-redshift origins, showing that the degree of modification depends on host galaxy mass. Future work should extend this analysis to larger galaxy samples and incorporate models that account for the disruption of clusters no longer present today.

Jocelyn: It really gives us a better way to probe the upper portion of the GCIMF, even though they admit that the low-mass regime is still uncertain due to cluster disruption.

Subrahmanyan: To wrap up this paper, "Reconstructing the Globular Cluster Initial Mass Function from Present-day Globular Cluster Systems," we see that it provides a novel approach linking observed present-day GCs with their progenitor populations.

Vera: It implies that the degree of modification depends on host galaxy mass, and for those who are driving or cooking right now, this is how astronomers try to reverse-engineer the early universe from what's left today.

Jocelyn: We're ready for a new paper next time, maybe something about gravitational waves or something completely different.

Subrahmanyan: I think this work sets a strong constraint on the high-mass end of the GCIMF, and it’s definitely an important piece of data for modeling cluster formation in cosmological simulations.

The paper's summary: Vera: So, to summarize this paper, they're taking all those globular clusters we see right now and using math to figure out what their original masses were back when they first formed in those galaxies.

Jocelyn: Exactly. The core idea is that you take the present-day cluster mass and work backward through the history of mass loss caused by the galaxy's gravitational pull, which is what shapes the initial mass function.

Subrahmanyan: It’s about inverting that process. They use an environment dependent model calibrated by N-body simulations to map those observed masses back to their progenitor values across many different host galaxies.

Vera: That’s the methodology we talked about before, but what this paper really delivers is how this reconstruction changes our view on the initial conditions of globular cluster systems.

Jocelyn: It moves us away from just guessing a universal power-law distribution and shows that the initial mass function isn't necessarily the same everywhere.

Subrahmanyan: They found that when you look at the high-mass end of this reconstructed mass function, the slope changes depending on how big and dense the host galaxy halo is.

Vera: That correlation they found, where more massive galaxies have steeper slopes in their initial distribution, that’s a big piece of data for us.

Jocelyn: So what does it mean for the world? It suggests that the way globular clusters form and survive depends systematically on the environment they live in.

Subrahmanyan: It supports a picture where GCs start out with a power-law mass distribution, but that initial shape gets modified by environmental effects and dynamical mass loss over time.

Vera: They’re using this to constrain the high-mass end of the GCIMF directly, which is something we can’t do easily otherwise.

Jocelyn: And they found that in dwarf galaxies, where clusters are less disrupted, their reconstructed initial distribution actually looks pretty close to what we might expect from a more universal scenario.

Subrahmanyan: That difference between massive and dwarf systems suggests that the degree of modification depends on the host galaxy mass itself.

Vera: So this gives us a way to link those local GC systems with their origins at high redshift, even if it's just for the high-mass part of the distribution.

Jocelyn: It’s a bridge, for sure. But they also admitted that figuring out the low-mass end is still tricky because low-mass clusters get destroyed so efficiently across cosmic time.

Subrahmanyan: That limitation is important; it tells us where our current understanding stops and where we need more simulations to go.

Vera: This work really constrains the upper portion of the GCIMF, which is a much more reliable probe for us right now.

Jocelyn: It provides a solid empirical link between what’s left today and what must have been there billions of years ago in those early galaxy systems.

Subrahmanyan: Next time, we should look at how these environmental effects scale, maybe using those proposed scaling relations they mentioned in their methods.

The paper's improvements: Vera: This paper lays out how they can actually make this reconstruction method more powerful for future work, and it’s interesting because they aren't just stopping at their first results.

Jocelyn: Right, so they suggest ways to refine the model itself to get cleaner data from the messy real sky.

Subrahmanyan: They are looking at how to predict those mass-loss rates more accurately, moving beyond just a simple calibration for a specific galaxy.

Vera: They propose using scaling relations like log A equals a one log Mh plus b one log Mh and alpha equals a two log Mh plus b two log Mh, all derived from those N-body simulations they used.

Jocelyn: That would let us apply this reconstruction method to any galaxy in the sample, not just the ones they measured individual cluster masses for.

Subrahmanyan: If that works, it means we could predict the expected GCIMF slope for a new galaxy just by knowing its host halo mass and environment.

Vera: That would be a massive improvement because it turns a specific observation into a general prediction about star formation history in galaxies.

Jocelyn: It’s about making the model more general and less dependent on the specific quirks of one host galaxy.

Subrahmanyan: This directly addresses the issue we talked about earlier: whether the initial mass function is truly universal or if it varies systematically with host properties.

Vera: It allows us to test that hypothesis by running a single reconstruction model across a vast range of environments.

Jocelyn: And they also talk about how their current method only reconstructs the surviving clusters, which is a limitation they acknowledge.

Subrahmanyan: They suggest future work needs to incorporate models that account for the disruption of clusters that aren't there today, which would be a big step forward.

Vera: So, they are aiming to build a more complete picture by trying to model the full evolution, not just the end state.

Jocelyn: It’s about improving our ability to connect those high-redshift progenitors with what we see in the present day, even if it means tackling some tough modeling challenges.

Subrahmanyan: This suggests that confirming or disproving a universal GCIMF will now require observational data from galaxies across much wider mass scales and environments.

Conclusion: Vera: So we’ve seen how this paper works to take today's globular clusters and reverse-engineer their initial masses using dynamical mass loss models across different host galaxies, and they’ve shown how this process changes depending on where you are in the galaxy.

Jocelyn: It really gives us a way to connect the present-day population of GCs with what formed them at high redshift, even though they admit that the low-mass end is still uncertain because of disruption.

Subrahmanyan: This implies that we need to be careful when we talk about a universal initial mass function and instead focus on how host galaxy properties modify the distribution over time.

Vera: They constrain the high-mass end of this GCIMF robustly, which is a crucial piece of data for testing formation models in cosmology.

Jocelyn: It’s an empirical bridge between local systems and those early universe candidates we see with telescopes like James Webb.

Subrahmanyan: If you want to confirm a universal power-law GCIMF, you’re going to need observations from both very massive and relatively low-mass galaxies.

Vera: Exactly, because these results show that dwarf galaxies are actually the best proxies for that initial distribution if disruption is inefficient there.

Jocelyn: So this paper provides a much more nuanced view of GC formation than just looking at one type of galaxy.

Subrahmanyan: It’s a solid step toward modeling the assembly history of galaxy halos, and I’ll keep an eye on how these scaling relations they propose actually behave in larger simulations.

Elizabeth Moreno-Hilario, Hui Li, Luis A. Martinez-Medina

Department of Astronomy, Tsinghua University · Instituto de Astronomía, Universidad Nacional Autónoma de Mîxico

astro-ph.GA

Submitted: 2026-05-26

Updated: 2026-10-28

Comments: Accepted for publication in ApJ. 16 pages, 10 figures, 3 tables

DOI: 10.3847/1538-4357/aeb2e9

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

Importance score: 87/100

The gist: The paper reconstructs the high-mass regime of the Globular Cluster Initial Mass Function (GCIMF) by inverting mass loss due to dynamical evolution across a wide range of host galaxy masses…

Key concepts

Globular Cluster Initial Mass Function (GCIMF)
This is a theoretical distribution describing how many globular clusters of different masses were initially formed in a galaxy. The paper uses observations to try and figure out what this original distribution looked like before dynamical processes changed the cluster masses we see today.
Mass Loss Inversion
The method involves taking the mass of a cluster as it exists today and mathematically 'undoing' the expected mass loss that occurred during its evolution. This allows researchers to estimate the larger, original mass that the cluster must have started with.
Host Galaxy Halo Mass Dependence
The research found that how steeply the high-mass end of the GCIMF slopes changes depends on how massive and dense a galaxy's dark matter halo is. More massive galaxies show steeper reconstructed slopes, indicating different initial formation conditions for their clusters.

Terminology

Summary

The paper reconstructs the high-mass regime of the Globular Cluster Initial Mass Function (GCIMF) by inverting mass loss due to dynamical evolution across a wide range of host galaxy masses, which is crucial for understanding the initial conditions of globular cluster systems

The Core Problem and Methodology

The near-universal turnover mass of the present-day globular cluster mass function (GCMF), MTO ∼ 2 × 10 5 M⊙, is a well established observational feature across galaxies of different types and masses, providing an important empirical benchmark for understanding the globular cluster initial mass function (GCIMF) Competing explanations of this property invoke either dynamical evolution from an initial power-law distribution or an imprint of cluster formation physics We address this problem by reconstructing the high-mass regime of the GCIMF by inverting the mass loss due to dynamical evolution of present-day globular cluster systems across a wide range of host galaxy masses Our method is based on an environment-dependent mass-loss model calibrated by direct N-body simulations in time-dependent tidal fields, enabling a mapping between observed cluster masses and their progenitor values without assuming a priori a functional form for the GCIMF

Data Sources and Sample Construction

To obtain good statistics and cover a wide range of galaxy types and masses, the authors compile a large sample of GC systems from different public sources The resulting sample combines galaxies with individually measured GC masses, photometric GC catalogs from the ACS Virgo Cluster Survey, and large statistical catalogs for which only integrated properties of the GC system are available For galaxies where individual GC masses are available, such as the Milky Way (MW) and nearby systems, they apply the mass-loss inversion cluster by cluster They also construct synthetic present-day GC systems for the dwarf galaxies analyzed in Paper I Together, these datasets provide 451 present-day GC mass distributions across a wide range of galaxy masses and environments

Dynamical Modeling and Inversion

The long-term dynamical mass evolution of GCs is modeled using an environment dependent calibration introduced and tested in Paper I The forward evolution of a cluster over time t can be written as M(t) = M0 [1 − A Mα 0 t] To reconstruct progenitor masses, they invert this relation for each observed present-day cluster mass Mobs by numerically solving Mobs = M0 [1 − A Mα 0 t] This formulation assumes that the average mass-loss rate remains approximately constant over the lifetime of the cluster, such that cumulative mass loss scales linearly with time

Key Findings on GCIMF Variation

The reconstructed GCIMF is built in logarithmic mass bins, and they characterize the high-mass end of the distribution by fitting a power law of the form dN/d log M ∝ M−β The fit is performed in log-log space above the turnover region The resulting slopes, β, vary systematically across galaxies and show a positive correlation with host galaxy halo mass Specifically, more massive galaxies tend to exhibit steeper reconstructed slopes

Implications for Cluster Formation Models

The study suggests that the commonly assumed universal power-law GCIMF may not hold across all galaxy types and masses Instead, the initial cluster mass distribution and/or the efficiency of cluster survival may depend systematically on host galaxy properties The agreement between independent approaches supports a picture in which GCs form with a power-law-like mass distribution at early times, which is subsequently modified by environmental effects and dynamical mass loss This has important implications for models of GC formation in cosmological simulations and for the interpretation of highredshift proto-globular cluster candidates observed with the James Webb Space Telescope The method constrains robustly the upper portion of the GCIMF, while the low-mass regime remains uncertain due to the preferential disruption of low-mass clusters across cosmic time

Conclusion

The authors present a novel approach to link observed present-day GCs with their progenitor populations, providing a bridge between local GC systems and their high-redshift origins This implies that the degree of modification depends on host galaxy mass Future work should extend this analysis to larger galaxy samples and incorporate models that account for the disruption of clusters no longer present today The reconstructed GCIMF should be interpreted as the progenitor mass distribution of surviving clusters rather than the full initial cluster population The high-mass regime provides a more reliable probe of the GCIMF shape The authors find that dwarf galaxies, where cluster disruption is inefficient, closely approximate the true initial distribution

APPENDIX A

This appendix presents the reconstructed GCIMFs for the remaining galaxies with individually measured GC masses discussed in Subsection 2.1 These figures complement the Milky Way example shown in Section 4 and illustrate the reconstruction procedure across different galaxy environments and mass scales, including M31 and nearby dwarf galaxy systems The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. p. 7

Where the paper says it

Reconstructed GCIMF (12 Gyr ago) for the Milky Way (yellow histogram) compared to the present-day GCMF (teal histogram) of 167 GCs. The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.

p. 7
Where the paper says it

Reconstructed GCIMF (12 Gyr ago) for the Milky Way (yellow histogram) compared to the present-day GCMF (teal histogram) of 167 GCs. The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.

50 ± 0.15 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 0. p. 7
Where the paper says it

The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. 18 in dN/d log M.

p. 7
Where the paper says it

The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. 18 in dN/d log M.

76 ± 0.11 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. p. 7
Where the paper says it

The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. 18 in dN/d log M.

p. 7
Where the paper says it

The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. 18 in dN/d log M.

76 ± 0.15 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. p. 7
Where the paper says it

The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. 18 in dN/d log M.

p. 7
Where the paper says it

The dashed line shows the best-fit high-mass power-law slope, yielding β = 1. 18 in dN/d log M.

05 ± 0.08 in dN/d log M The scatter of the data around the linear relation, σ = 0. p. 8
Where the paper says it

038) log(Mh) − (6. p. 8

Where the paper says it

The best fit parameters are β = (0. 038) log(Mh)−(6.

p. 8
Where the paper says it

The best fit parameters are β = (0. 038) log(Mh)−(6.

428) (12) The scatter of the data around the linear relation, σ = 0.

p. 8
Where the paper says it

038) log(Mh) − (6. 428) (12) The scatter of the data around the linear relation, σ = 0.

368, reflects the galaxy to galaxy variation in β The best fit parameters are β = (0. p. 8
Where the paper says it

368, reflects the galaxy to galaxy variation in β, showing that galaxies of similar halo mass can host GC systems with slightly different high-mass slopes. This confirms that the linear fit captures the average behavior and establishes that more massive galaxies tend to host reconstructed GCIMFs with steeper high-mass slopes.

p. 8
Where the paper says it

368, reflects the galaxy to galaxy variation in β, showing that galaxies of similar halo mass can host GC systems with slightly different high-mass slopes. This confirms that the linear fit captures the average behavior and establishes that more massive galaxies tend to host reconstructed GCIMFs with steeper high-mass slopes.

699 ± 0.038) log(Mh) − (6.698 ± 0.428) This confirms that the linear fit captures the average behavior and establishes that more massive galaxies tend to host reconstructed GCIMFs with steeper high-mass slopes The scatter of the data around the linear relation, σ = 0. p. 8
Where the paper says it

368, reflects the galaxy to galaxy variation in β, showing that galaxies of similar halo mass can host GC systems with slightly different high-mass slopes. This confirms that the linear fit captures the average behavior and establishes that more massive galaxies tend to host reconstructed GCIMFs with steeper high-mass slopes.

368, reflects the galaxy to galaxy variation in β

REFERENCES

Adamo, A., Bradley, L. D., Vanzella, E. p. 0

Where the paper says it

Mowla et al. Vanzella et al.

, et al. 2024, Nature, 632, 513 Andersson, E. p. 0
Where the paper says it

Garcia et al. Andersson et al.

P., Mac Low, M. p. 1
Where the paper says it

Because YMCs represent the closest observable analogues to young GCs, it is often assumed that GCs formed with a similar initial mass distribution. If this is the case, the present-day peaked GCMF must arise primarily from dynamical evolution processes that preferentially destroy low-mass clusters over cosmic time.

-M., Agertz, O., Renaud, F., & Li, H. 2024, A&A, 681, A28 Baumgardt, H. p. 2
Where the paper says it

Starting with the MW, we adopt dynamical GC masses from the Galactic Globular Cluster Database (H. Baumgardt et al.

, & Hilker, M. p. 3
Where the paper says it

Baumgardt & M. Hilker 2018).

2018, MNRAS, 478, 1520 Bell, E. p. 3
Where the paper says it

Baumgardt & M. Hilker 2018).

F., McIntosh, D. H., Katz, N., & Weinberg, M. D. 2003, ApJS, 149, 289 Brodie, J. p. 0
Where the paper says it

Because most GCs have survived for more than 10 Gyr, their present-day properties encode information about both their formation conditions and their subsequent dynamical evolution within their host galaxies (J. Brodie & J.

P., & Strader, J. p. 0
Where the paper says it

Brodie & J. Strader 2006).

2006, ARA&A, 44, 193 Caldwell, N. p. 0
Where the paper says it

Brodie & J. Strader 2006).

, & Romanowsky, A. p. 3
Where the paper says it

Caldwell & A. Romanowsky (2016), the sample includes 436 GCs.

J. 2016, ApJ, 824, 42 Chen, Y. p. 3
Where the paper says it

Caldwell & A. Romanowsky (2016), the sample includes 436 GCs.

, & Gnedin, O. p. 1
Where the paper says it

Numerical studies further showed that cluster disruption rates depend sensitively on environmental properties such as the strength of the galactic tidal field and the orbital histories of clusters (O. Gnedin & J.

p. 1
Where the paper says it

Numerical studies further showed that cluster disruption rates depend sensitively on environmental properties such as the strength of the galactic tidal field and the orbital histories of clusters (O. Gnedin & J.

p. 1
Where the paper says it

Numerical studies further showed that cluster disruption rates depend sensitively on environmental properties such as the strength of the galactic tidal field and the orbital histories of clusters (O. Gnedin & J.

Y. 2023, MNRAS, 522, 5638 Claeyssens, A. p. 8
Where the paper says it

In a recent work, A. Claeyssens et al.

, Adamo, A. p. 0
Where the paper says it

Vanzella et al. Adamo et al.

, Kokorev, V., et al. 2026, arXiv e-prints de Boer, T. p. 3
Where the paper says it

For the Fornax dwarf spheroidal galaxy, GC masses are taken from T. de Boer & M.

J. L., & Fraser, M. p. 3
Where the paper says it

de Boer & M. Fraser (2016).

2016, A&A, 590, A35 Dornan, V. p. 3
Where the paper says it

Caldwell & A. Romanowsky (2016), the sample includes 436 GCs.

, & Harris, W. p. 1
Where the paper says it

Jord´an et al. Harris et al.

E. 2026, ApJ, 998, 41 Fall, S. p. 1
Where the paper says it

Young massive cluster (YMC) populations observed in nearby starforming galaxies typically follow a power-law mass function with slope dN/dM ∝ M−2 (e. Fall 1999; N.

M., & Zhang, Q. p. 1
Where the paper says it

McLaughlin 1994; S. Zhang 2001).

1999, ApJL, 527, L81 Fardal, M. p. 0
Where the paper says it

Martinez-Medina 2 1Department of Astronomy, Tsinghua University, 30 Shuangqing Road, Beijing 100084, China 2 Instituto de Astronom´ıa, Universidad Nacional Aut´onoma de M´exico, A. 04510, CDMX, M´exico Submitted to ApJL ABSTRACT The near-universal turnover mass of the present-day globular cluster mass function (GCMF), MTO ∼ 2 × 105 M⊙, is a well established observational feature across galaxies of different types and masses, providing an important empirical benchmark for understanding the globular cluster initial mass function (GCIMF).

A., Weinberg, M. D., Babul, A., et al. 2013, MNRAS, 434, 2779 Forbes, D. p. 0
Where the paper says it

A large fraction of present-day GCs are metal-poor and have ages of ≳ 12 Gyr, indicating that they formed at high redshift, likely around the epoch of reionization (D. Forbes et al.

A., Read, J. I., Gieles, M. p. 1
Where the paper says it

Gnedin 2023; M. Gieles & O.

, & Collins, M. L. M. 2018, MNRAS, 481, 5592 Garcia, F. p. 0
Where the paper says it

Rieder et al. Garcia et al.

A. B., Ricotti, M., Sugimura, K., & Park, J. 2023, MNRAS, 522, 2495 Gieles, M. p. 1
Where the paper says it

Gnedin 2023; M. Gieles & O.

, & Gnedin, O. Y. 2023, MNRAS, 522, 5340 Gnedin, O. p. 1
Where the paper says it

Pfeffer et al. Gnedin 2023; M.

Y., & Ostriker, J. p. 1
Where the paper says it

Gnedin & J. Ostriker 1997).

P. 1997, ApJ, 474, 223 Harris, W. p. 1
Where the paper says it

Gnedin & J. Ostriker 1997).

E., Harris, G. p. 1
Where the paper says it

Jord´an et al. Harris et al.

p. 1
Where the paper says it

Jord´an et al. Harris et al.

L., & Hudson, M. J. 2015, ApJ, 806, 36 Harris, W. p. 3
Where the paper says it

Harris et al. (2015), which gives Mh = 2.

E., Harris, G. L. H., & Alessi, M. 2013, ApJ, 772, 82 Hunter, J. p. 3
Where the paper says it

Fardal et al. (2013), who derive a dynamical mass model based on kinematic and structural constraints of the Andromeda system.

D. 2007, Computing in Science and Engineering, 9, 90 Jord´an, A. p. 4
Where the paper says it

Jord´an et al. (2007), who found that the turnover is slightly smaller in dwarfs compared to more massive galaxies.

, McLaughlin, D. p. 1
Where the paper says it

Early analytical studies demonstrated that dynamical processes such as two-body relaxation, tidal evaporation, and gravitational shocks in a galactic potential can transform an initially featureless or power-law cluster mass distribution into a peaked luminosity function over a Hubble time (D. McLaughlin 1994; S.

E., Cˆot´e, P., et al. 2007, ApJS, 171, 101 Jord´an, A. p. 4
Where the paper says it

Jord´an et al. (2007), who found that the turnover is slightly smaller in dwarfs compared to more massive galaxies.

, Peng, E. W., Blakeslee, J. P., et al. 2009, ApJS, 180, 54 Karim, N. p. 3
Where the paper says it

The halo mass is adopted from the recent dynamical analysis of N. Karim et al.

, Collins, M. L. M., Forbes, D. p. 0
Where the paper says it

A large fraction of present-day GCs are metal-poor and have ages of ≳ 12 Gyr, indicating that they formed at high redshift, likely around the epoch of reionization (D. Forbes et al.

A., & Read, J. I. 2024, MNRAS, 530, 4936 Kimm, T. p. 0
Where the paper says it

At the same time, increasingly high-resolution hydrodynamical simulations are beginning to resolve the formation of bound star clusters in cosmological environments (e. Kimm et al.

, Cen, R., Rosdahl, J., & Yi, S. K. 2016, ApJ, 823, 52 Kruijssen, J. p. 1
Where the paper says it

More recent theoretical models have emphasized the strong dependence of cluster formation and survival on the evolving galactic environment, including the density of the interstellar medium and the dynamical history of the host galaxy (J. Kruijssen 2015).

M. D. 2015, MNRAS, 454, 1658 Li, H. p. 1
Where the paper says it

More recent theoretical models have emphasized the strong dependence of cluster formation and survival on the evolving galactic environment, including the density of the interstellar medium and the dynamical history of the host galaxy (J. Kruijssen 2015).

, & Gnedin, O. Y. 2019, MNRAS, 486, 4030 McCrady, N. p. 1
Where the paper says it

Fall 1999; N. McCrady & J.

, & Graham, J. p. 1
Where the paper says it

McCrady & J. Graham 2007).

R. 2007, ApJ, 663, 844 McLaughlin, D. p. 1
Where the paper says it

Early analytical studies demonstrated that dynamical processes such as two-body relaxation, tidal evaporation, and gravitational shocks in a galactic potential can transform an initially featureless or power-law cluster mass distribution into a peaked luminosity function over a Hubble time (D. McLaughlin 1994; S.

E., & Pudritz, R. p. 1
Where the paper says it

McLaughlin & R. Pudritz (1996) investigated the formation of GC systems within protogalactic environments and suggested that the observed mass distribution could arise from a combination of formation physics and subsequent dynamical disruption.

E. 1996, ApJ, 457, 578 Moreno-Hilario, E. p. 0
Where the paper says it

Draft version May 28, 2026 Typeset using LATEX twocolumn style in AASTeX7. 1 Reconstructing the Globular Cluster Initial Mass Function from Present-Day Globular Cluster Systems Elizabeth Moreno-Hilario , 1 Hui Li , 1 and Luis A.

, Martinez-Medina, L. p. 0
Where the paper says it

1 Reconstructing the Globular Cluster Initial Mass Function from Present-Day Globular Cluster Systems Elizabeth Moreno-Hilario , 1 Hui Li , 1 and Luis A. Martinez-Medina 2 1Department of Astronomy, Tsinghua University, 30 Shuangqing Road, Beijing 100084, China 2 Instituto de Astronom´ıa, Universidad Nacional Aut´onoma de M´exico, A.

A., Li, H., Souza, S. O., & Perez-Villegas, A. 2024, MNRAS, 527, 2765 Mowla, L. p. 0
Where the paper says it

However, recent observations with the James Webb Space Telescope have begun to reveal candidate proto–globular clusters at redshifts up to z ∼ 10 in strongly lensed systems (e. Mowla et al.

, Iyer, K. G., Desprez, G., et al. 2022, ApJL, 937, L35 Pace, A. p. 0
Where the paper says it

Martinez-Medina 2 1Department of Astronomy, Tsinghua University, 30 Shuangqing Road, Beijing 100084, China 2 Instituto de Astronom´ıa, Universidad Nacional Aut´onoma de M´exico, A. 04510, CDMX, M´exico Submitted to ApJL ABSTRACT The near-universal turnover mass of the present-day globular cluster mass function (GCMF), MTO ∼ 2 × 105 M⊙, is a well established observational feature across galaxies of different types and masses, providing an important empirical benchmark for understanding the globular cluster initial mass function (GCIMF).

B. 2024, arXiv e-prints Parmentier, G. p. 1
Where the paper says it

Similarly, G. Parmentier & G.

, & Gilmore, G. p. 1
Where the paper says it

Parmentier & G. Gilmore (2007) showed that rapid gas expulsion from embedded protoclusters within a powerlaw molecular cloud mass spectrum can produce a bellshaped cluster mass function at early times, even before long-term dynamical evolution becomes dominant.

2007, MNRAS, 377, 352 Peng, E. p. 3
Where the paper says it

(2009) and E. Peng et al.

W., Jord´an, A. p. 1
Where the paper says it

acteristic peaked or log-normal shape with a turnover mass near MTO ∼ 2 × 105 M⊙ (A. Jord´an et al.

, Cˆot´e, P., et al. 2008, ApJ, 681, 197 Pfeffer, J. p. 3
Where the paper says it

Peng et al. (2008), selecting objects with pGC > 0.

, Kruijssen, J. p. 1
Where the paper says it

More recent theoretical models have emphasized the strong dependence of cluster formation and survival on the evolving galactic environment, including the density of the interstellar medium and the dynamical history of the host galaxy (J. Kruijssen 2015).

M. D., Crain, R. A., & Bastian, N. 2018, MNRAS, 475, 4309 Posti, L. p. 3
Where the paper says it

Baumgardt & M. Hilker 2018).

, & Helmi, A. p. 3
Where the paper says it

3 × 1012 M⊙ (L. Helmi 2019).

2019, A&A, 621, A56 Rejkuba, M. p. 1
Where the paper says it

The apparent universality of this feature has made the GCLF a useful distance indicator (M. Rejkuba 2012), but it also raises a fundamental challenge for models of GC formation and evolution.

2012, Ap&SS, 341, 195 Rieder, S. p. 0
Where the paper says it

Kimm et al. Rieder et al.

, Dobbs, C., Bending, T., Liow, K. Y., & Wurster, J. 2022, MNRAS, 509, 6155 Rodriguez C. p. 1
Where the paper says it

Gnedin 2019; C. Rodriguez et al.

L., Hafen Z., Grudi´c M. Y., et al. 2023, MNRAS, 521, 124 Tudorica, A. p. 3
Where the paper says it

For the dwarf galaxy IKN, GC masses are taken from A. Tudorica et al.

, Georgiev, I. Y., & Chies-Santos, A. L. 2015, A&A, 581, A84 van der Walt, S. p. 1
Where the paper says it

More recent theoretical models have emphasized the strong dependence of cluster formation and survival on the evolving galactic environment, including the density of the interstellar medium and the dynamical history of the host galaxy (J. Kruijssen 2015).

, Colbert, S. C., & Varoquaux, G. 2011, Computing in Science and Engineering, 13, 22 Vanzella E. p. 0
Where the paper says it

Mowla et al. Vanzella et al.

, Claeyssens A. p. 8
Where the paper says it

In a recent work, A. Claeyssens et al.

, Welch B., et al. 2023, ApJ, 945, 53 Vanzella E. p. 0
Where the paper says it

Mowla et al. Vanzella et al.

, Messa, M., Adamo A., et al. 2026, A&A, 705, A171 Vesperini E. p. 0
Where the paper says it

One of the most remarkable observational properties of GC systems is the near-universal shape of their mass and luminosity distributions (E. Vesperini 2001).

2001, MNRAS, 322, 247 Virtanen P. p. 3
Where the paper says it

(2003): log(M/Lz) = 0. 322 (g − z) − 0.

, Gommers R., Oliphant T. E., et al. 2020, Nature Methods, 17, 261 Wes McKinney. p. 1
Where the paper says it

However, such simulations often adopt simplified setups in which clusters evolve on circular or eccentric orbits within fixed galactic potentials. While these controlled experiments provide valuable insight into cluster disruption processes, their idealized nature limits the accuracy with which these prescriptions can be extrapolated to the complex, time-varying environments of cosmological galaxy formation.

2010, in Proceedings of the 9th Python in Science Conference Zhang Q. p. 1
Where the paper says it

McLaughlin 1994; S. Zhang 2001).

, & Fall, S. p. 1
Where the paper says it

Young massive cluster (YMC) populations observed in nearby starforming galaxies typically follow a power-law mass function with slope dN/dM ∝ M−2 (e. Fall 1999; N.

M. 1999, ApJL, 527, L81

This appendix presents the reconstructed GCIMFs for the remaining galaxies with individually measured GC masses discussed in Subsection 2.1 These figures complement the Milky Way example shown in Section 4 and illustrate the reconstruction procedure across different galaxy environments and mass scales, including M31 and nearby dwarf galaxy systems The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.50 ± 0.15 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 0.76 ± 0.11 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.76 ± 0.15 in dN/d log M The dashed line shows the best-fit high-mass power-law slope, yielding β = 1.05 ± 0.08 in dN/d log M The scatter of the data around the linear relation, σ = 0.368, reflects the galaxy to galaxy variation in β The best fit parameters are β = (0.699 ± 0.038) log(Mh) − (6.698 ± 0.

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. Improved AI system can perform empirical reconstruction of globular cluster initial mass functions (GCIMFs) by inverting present-day cluster masses using an environment dependent mass-loss model calibrated on N-body simulations.

  2. The improved system can systematically map the high-mass slope of the GCIMF as a function of host galaxy halo mass, providing empirical constraints like the derived linear relation beta = (0.699 ± 0.038) log(Mh) − (6.698 ± 0.428).

  3. The system can distinguish between scenarios where the GCIMF is universal versus varying with host galaxy properties by comparing reconstructed GCIMFs across systems spanning dwarf to Milky Way-like hosts, testing whether the commonly assumed universal power-law GCIMF is consistent with the observed properties of GC systems.

  4. The AI can constrain the high-mass end of the GCIMF by recovering only progenitor masses of surviving clusters, providing a direct empirical link between present-day globular cluster systems and their high-redshift progenitors.

  5. The system can predict environmentally dependent mass-loss rates for any galaxy in the sample by using proposed scaling relations such as log A = a1 log Mh + b1 and alpha = a2 log Mh + b2, derived from N-body simulations.

Abstract

The near-universal turnover mass of the present-day globular cluster mass function (GCMF), M TO about 2 times 10 5 M, is a well established observational feature across galaxies of different types and masses, providing an important empirical benchmark for understanding the globular cluster initial mass function (GCIMF). Competing explanations of this property invoke either dynamical evolution from an initial power-law distribution or an imprint of cluster formation physics. We address this problem by reconstructing the high-mass regime of the GCIMF by inverting the mass loss due to dynamical evolution of present-day globular cluster systems across a wide range of host galaxy masses. Our method is based on an environment-dependent mass-loss model calibrated by direct N-body simulations in time-dependent tidal fields, enabling a mapping between observed cluster masses and their progenitor values without assuming a priori a functional form for the GCIMF. We apply our method to galaxies spanning halo masses of about10 9 - 10 12 M, combining systems with individually measured globular cluster masses as well as large statistical samples constructed from observed global properties. The recovered GCIMFs are systematically shifted towards higher masses and exhibit a power-law behavior at the high-mass end. The inferred slopes vary across galaxies and show a strong correlation with host halo mass, with more massive galaxies exhibiting steeper high-mass slopes. Our results suggest that the slope of the GCIMF depends on the galactic properties and provides a direct empirical link between present-day globular cluster systems and their high-redshift progenitors.

Sources

Related papers