Pith. sign in

REVIEW 3 major objections 5 minor 43 references

Light curves and spectra for stellar collisions between main-sequence stars in galactic nuclei

T0 review · 3 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read High-velocity collisions between main-sequence stars in galactic nuclei should produce bright, short-lived ultraviolet transients that sub-day-cadence surveys can detect.

desk verdict First systematic CMFGEN grid for main-sequence stellar collisions; the qualitative picture is convincing, but the quantitative fitting formulae inherit an unquantified viewing-angle bias from the 3D-to-1D spherical averaging. read the letter →

arxiv 2502.05265 v4 pith:2C6HZSTZ submitted 2025-02-07 astro-ph.HE astro-ph.GAastro-ph.SR

Taeho Ryu , Luc Dessart This is my paper · ORCID
classification astro-ph.HEastro-ph.GAastro-ph.SR
keywords stellarcollisionsgalacticnucleinucleartransientsultravioletflareslightcurvesspectraradiativetransferhydrodynamics
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that when two ordinary main-sequence stars collide at thousands of kilometers per second in a galactic nucleus, the wreckage should shine as a fast, ultraviolet-bright transient — a distinct class of nuclear flare alongside tidal disruption events and supernovae. Combining 3D hydrodynamics simulations of the collision with 1D radiative-transfer calculations of what an observer would see, the authors predict bolometric luminosities of roughly $10^{41.5}$ to $10^{42.5}$ erg s$^{-1}$ at half a day, declining steeply as $t^{-2}$ to $t^{-4}$ toward an optical plateau near $10^{39}$ to $10^{41.5}$ erg s$^{-1}$ (absolute $V$ magnitude $-10$ to $-15$) after a few days, with peaks that may reach $10^{42}$ to $10^{43}$ erg s$^{-1}$. They find that the day-one luminosity, the UV and $V$-band magnitudes, and the Doppler width of H$\alpha$ correlate with the collision's impact parameter and kinetic energy, and they reduce these correlations to fitting formulae. If correct, these events are observable nuclear transients for sub-day-cadence surveys, and spectroscopy can separate them from Type II supernovae, whose spectra they resemble.

What carries the argument

The central machinery is a two-stage simulation pipeline. First, a 3D moving-mesh hydrodynamics code (AREPO) follows the collision of relaxed, terminal-age stellar models for 0.5 d, until the shocked debris is expanding quasi-spherically and homologously. Second, a spherically averaged, mass-conserving profile of that debris seeds a 1D, non-LTE, time-dependent radiative-transfer code (CMFGEN) that computes the emergent light curves and spectra, tracking the H I and He I-II recombination fronts as the debris cools. The load-bearing step is the spherical average taken at 0.5 d: it conserves mass per radius bin but not kinetic or radiation energy, which is why the paper's predictive content is carried by four fitting formulae connecting ejecta mass, kinetic energy, radiation energy, and the day-one observables ($L_{\rm bol}$, $M_{\rm UVW1}$, $M_V$, $V_{\rm H\alpha}$) to the impact parameter $b$ and collision kinetic energy $E_{\rm kin}$.

What would settle it

The off-axis, low-velocity cases are the test bed: recompute a $b = 0.5$, $V_{\rm rel} = 2500$ km s$^{-1}$ collision with angle-resolved transport, and if the predicted light curve shifts by the factor of 5 to 10 implied by the energy mismatch, the fitting formulae need a viewing-angle term. Observationally, a UV survey that catches a nuclear transient within hours can check the predicted sequence directly — an early UV metal-line forest, a $t^{-2}$ to $t^{-4}$ decay, and an H$\alpha$ Doppler velocity consistent with Eq. (7) — and any of those signatures failing would count against the model.

Watch

Extended reading notes

Core claim

The paper establishes that debris from high-velocity collisions between terminal-age main-sequence stars radiates a predictable electromagnetic signal: a bright ultraviolet flash with bolometric luminosity peaking at $\gtrsim10^{43}$ erg s$^{-1}$, fading as $L \propto t^{-2}$ to $t^{-4}$ down to roughly $10^{41}$ to $10^{42}$ erg s$^{-1}$ at 0.5 d, and leveling onto a plateau of $10^{39}$ to $10^{41.5}$ erg s$^{-1}$ after a few days. The total radiated energy stays below $10^{49}$ erg, only $10^{-3}$ to $10^{-5}$ of the initial collision kinetic energy. Spectra evolve from UV-dominated, crowded with resonance lines of C, N, O, and Fe, to optical-dominated with broad H I Balmer lines plus Na I and Ca II features, qualitatively the same progression as in Type II supernovae. A grid of nearly forty collision setups shows that ejecta mass, kinetic and radiative energy, and day-one luminosity are governed mainly by impact parameter $b$ and initial kinetic energy $E_{\rm kin}$, with H$\alpha$ Doppler velocity tracking the relative collision velocity, and the authors condense these trends into fitting formulae.

Load-bearing premise

That the 3D collision debris can be squeezed into a spherically symmetric 1D model at half a day without losing the light curve; the paper itself notes this fails for off-axis, slower collisions, where the 1D and 3D energy budgets differ by a factor of 5 to 10 and viewing angle starts to matter.

Editorial extensions

If this is right

  • The brightest phase is short, roughly 0.1 d, so sub-day-cadence UV surveys are required to catch these events near their $10^{42}$ to $10^{43}$ erg s$^{-1}$ peak; after about a week they fade onto a faint optical plateau.
  • A measured day-one luminosity combined with the H$\alpha$ Doppler velocity pins down the collision kinetic energy and impact parameter, while the individual stellar masses and the relative velocity remain degenerate without spectroscopy.
  • Giant-star collisions remain the brighter siblings: at equal kinetic energy they outshine main-sequence collisions by up to two orders of magnitude, so surveys should see the giant variety first.
  • Because the spectra resemble Type II supernovae, these flares could be misclassified without prompt UV spectroscopy, where the C, N, O, and Fe resonance-line forest is the distinguishing signature.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the bright phase lasts only about a day, surveys with nightly cadence would systematically catch only the plateau, implying that current rate limits on these events are set by cadence rather than sensitivity and that the true occurrence could be higher than archival searches suggest.
  • Each detected collision yields a direct measurement of the relative velocity between stars at a known distance from the black hole, so a catalog of these events could map the velocity dispersion profile of nuclear clusters the way masers map disks.
  • The models leave a He-rich remnant in most cases; if that remnant stays in the cluster and collides again, the nucleus could show repeated flares from the same site, a signature testable in long-baseline UV monitoring.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. This paper models collisions between terminal-age main-sequence stars in galactic nuclei, combining 3D AREPO hydrodynamics with 1D CMFGEN non-LTE radiative transfer. It presents bolometric, UVW1-band, and V-band light curves and optical-UV spectra for a grid of 35 collision configurations, identifies correlations between collision parameters (masses, relative velocity, impact parameter) and observables, and provides fitting formulae (Eqs. 1-7). The central claims are that the debris produce bright, fast-fading UV transients with bolometric luminosities reaching roughly 1e42-1e43 erg/s near 0.5 d, that the spectra morph from UV- to optical-dominated and resemble Type II supernovae, and that these events are observable as nuclear transients with sub-day cadence surveys.

Significance. If correct, the paper would establish main-sequence stellar collisions as a distinct class of nuclear transients and provide a systematic grid of predictions for ULTRASAT/UVEX. Strengths include the broad parameter coverage, the resolution tests and energy-conservation checks in the AREPO simulations, the explicit labeling of Eqs. (1)-(7) as least-squares fits to the simulation sample, and the public release of the radiative-transfer outputs. The main caveat is that the quantitative correlations and the individual spectra rely on a 1D spherical reduction of the 3D debris, and the paper itself documents cases where this reduction is not valid. The qualitative picture of bright, fast-fading UV transients is plausible, but the quantitative scaling relations and the b=0.5 light curves and spectra are not yet established at the same level.

major comments (3)
  1. [§3.1, Table A.1, §4.3] The 1D spherical-averaging step is not validated for a substantial part of the radiative-transfer sample, and Table A.1 shows that the problem is not confined to the excluded Vrel=2500 km/s, b=0.5 models. For example, ms10/ms9/vrel5/b0p5 has Erad,ej(1D)/Erad,ej(3D) of about 4.2 and ms10/ms9/vrel10/b0p5 has a ratio of about 4.6, while the included model ms10/ms1/vrel2p5/b0p5 has a ratio of about 3.5; several other b=0.5 models show ratios of 2-3. Because a mass-conserving spherical average does not preserve the density, temperature, and velocity structure that governs photon diffusion, the resulting Lbol and spectra for these models effectively correspond to different initial conditions. These models nevertheless enter the correlations behind Eqs. (4)-(7), so the quoted 1sigma uncertainties of 36% (2-3%, 10%) for Lbol, magnitudes, and VHalpha describe scatter around a fit, not the systematic error of the 1D mapping. The qualitative conclusion that off-axis collisions produce fainter transients may survive, but the quantitative b-dependence and the spectra for b=0.5 are not established. I ask the authors to either restrict the radiative-transfer sample and the fitting formulae to configurations where the 1D reduction is quantitatively valid, or to demonstrate robustness with angle-resolved or 3D radiative-transfer tests for representative b=0.5 cases.
  2. [§3.2, Table 1] The model ms3/ms1/vrel2p5/b0p5 has Lbol,1d = 4.97e39 erg/s, a factor of about 7 fainter than the next-faintest entry in Table 1. This is precisely one of the cases where the 1D initialization is most questionable (b=0.5, Vrel=2500 km/s), yet it is included in the correlations used to infer collision parameters. A single outlier produced by an invalid 1D mapping can bias the fitted slope in Eq. (4). The paper should show the fits with and without such marginal models, or replace them with a treatment that does not assume spherical symmetry.
  3. [§3.2, §5] The abstract and the summary present 'peaking at ≳10^43 erg/s' as a simulation result, but the earliest CMFGEN output is at 0.5 d and the peak is an extrapolation based on the Lbol ∝ t^-2 to t^-4 fading measured at 0.5-1 d. This extrapolation is reasonable, but it should be labeled as such, and the peak duration of order 0.1 d is inferred rather than simulated. Please add an explicit caveat or accompany the extrapolation with an estimate based on the stored radiation energy and diffusion time from the 3D AREPO runs.
minor comments (5)
  1. [§1, §2.2] There are typographical errors: 'magenta-shaed' should be 'magenta-shaded' in §1, and 'di ffuse' should be 'diffuse' in §2.2.
  2. [Table 1] The model name 'ms10/ms1/relv2p5/b0p1' in Table 1 is inconsistent with the naming convention used elsewhere; it should be 'ms10/ms1/vrel2p5/b0p1'.
  3. [Fig. 1] The units on the vertical axis appear corrupted as 'M⊙ pc□3'; this should presumably be M⊙ pc^-3.
  4. [Abstract] The abstract states that the total radiated energy is less than 1e49 erg and corresponds to 1e-3 to 1e-5 of the initial kinetic energy, but this quantity is not defined or derived in the body of the paper; please add a definition or remove the statement.
  5. [§4.3] The discussion of the weak b=0.25 correlation attributes the scatter to the small sample size (six models); given that small size, the b-dependence in Eqs. (4)-(6) would benefit from a bootstrap or leave-one-out estimate of the fit uncertainty.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reduction: light curves and spectra are forward-modeled, and the fitting formulae are explicitly descriptive; the 1D/3D non-conservation is a validity limitation, not circularity.

full rationale

The paper's chain is: 3D AREPO hydrodynamics, spherical-averaged 1D profiles at 0.5 d, CMFGEN NLTE time-dependent radiative transfer, then light curves, spectra, and correlations. No equation in this chain reduces by construction to another equation's output. The fitting formulae (Eqs. 1-7) are least-squares fits to the paper's own simulation outputs and are explicitly labeled as fits; footnote 6 states 'All fitting coefficients presented in the paper were obtained using the curve_fit function from the SciPy package,' and Section 4 calls them 'fitting formulae that describe the correlations.' They are therefore descriptive summaries, not independent predictions, and presenting them as usable for parameter inference is an application, not a circular derivation. The self-citations to AREPO (Springel 2010), CMFGEN (Hillier & Dessart 2012), Ryu et al. (2024a), and Dessart et al. (2024) are code and methodology citations with independent provenance; no uniqueness theorem or ansatz is imported by self-citation, and no central claim is forced by an unverified prior result. The paper itself flags its main validity limitation in Section 3.1: the spherical average conserves mass per bin but 'the kinetic energy and radiation energies ... were not strictly conserved,' and for b = 0.5 with Vrel = 2500 km/s the 1D-3D difference is 'a factor of 5-10, indicating our assumption of spherical symmetry for expanding ejecta is not valid in such cases. This also implies that viewing angle effects are important for these collisions.' The conclusion repeats that 'observable properties would depend on viewing angle' near the supermassive black hole. These passages are important correctness and robustness caveats for the quantitative correlations and spectra of off-axis, low-velocity cases, but they are not circularity: the radiative-transfer calculation is a nontrivial forward model of the supplied profiles, and the light curves are checked against SN 1987A and Type II SNe behavior. The fitting formulae are not elevated to predictions, and the derivation does not presuppose its conclusions. Hence no circular step is present.

Assumptions & free parameters 7 free parameters · 6 assumptions · 0 invented entities

The central claims rest on standard stellar structure inputs, the assumed adiabatic hydrodynamics, and the spherical mapping into CMFGEN. The fitting formulae add seven sets of coefficients fitted to the simulation grid. These are the main purchased inputs; no entities are invented.

free parameters (7)
  • Eq. 1 normalization and exponent = -0.63, 1.3
    Fitted to Mej/(M1+M2) versus b/0.1 across the simulated grid; central to the ejecta-mass correlation.
  • Eq. 2 normalization and exponent = -0.5, 1.4
    Fitted to Ekin,ej/Ekin versus b/0.1; describes the kinetic-energy correlation.
  • Eq. 3 normalization and exponents = 6e46 erg, 1.3, -0.6, 1.1
    Fitted to Erad,ej versus Ekin and b; central to the radiation-energy correlation.
  • Eq. 4 normalization and exponents = 8e40 erg, -1 on b, linear Ekin
    Fitted to Lbol,1d versus Ekin and b; the main luminosity forecasting formula.
  • Eq. 5 magnitude coefficients = -2, -14.5, -0.1
    Fitted to MUVW1,1d versus Ekin and b.
  • Eq. 6 magnitude coefficients = -1.6, -12.5, -0.07
    Fitted to MV,1d versus Ekin and b.
  • Eq. 7 velocity coefficients = 5500 km/s, -0.25, 1.3
    Fitted to VHalpha,1d versus Vrel and b; used to break the mass-velocity degeneracy.
assumptions (6)
  • domain assumption The debris expansion is quasi-spherical and homologous by 0.5 d, allowing 1D RT.
    Invoked in Section 3.1 when constructing spherical averages for CMFGEN; the paper notes it breaks down for b=0.5 and Vrel=2500 km/s.
  • domain assumption Hydrodynamics can be treated as adiabatic until 0.5 d.
    Section 2.2 states the hydrodynamics simulations do not account for radiative energy loss; the authors argue excess heat radiates away in the first RT timestep.
  • domain assumption Terminal-age main-sequence stellar models at solar metallicity represent the relevant collision population.
    Section 2.1 evolves models to central H mass fraction 0.01 with Z=0.02; earlier evolutionary stages would be more easily destroyed.
  • domain assumption Accretion power from the supermassive black hole is negligible at times shorter than a week.
    Section 3.1 states accretion would dominate at late times; the early optically-thick phase is assumed unaffected.
  • standard math Standard ideal gas plus radiation pressure equation of state with LTE.
    Section 2.1 gives P = aT^4/3 + rho kT/(mu mp), used for all hydrodynamic runs while debris is optically thick.
  • domain assumption All isotopes are stable, with no radioactive decay heating.
    Section 3.1 states there is no contribution from radioactive decay and no non-thermal effects from such decays.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Light curves and spectra for stellar collisions between main-sequence stars in galactic nuclei." pith.science (2026). https://pith.science/paper/2C6HZSTZ

@misc{pith2026250205265,
  author       = {Pith},
  title        = {Pith review of: Light curves and spectra for stellar collisions between main-sequence stars in galactic nuclei},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2C6HZSTZ}},
  note         = {Machine review of arXiv:2502.05265}
}
abstract

High-velocity stellar collisions in galactic nuclei produce ejecta that generate potentially observable electromagnetic radiation, making them promising nuclear transients. However, the photometric and spectroscopic properties of these collisions, which would more frequently involve main-sequence stars, remain largely unexplored. Here, using 3D hydrodynamics and 1D radiative-transfer simulations, we investigate the properties and observables of the debris produced in high-velocity collisions between terminal-age main-sequence stars, covering a wide range of collision configurations. The ejecta produce bright ultraviolet (UV) flares with bolometric luminosities typically peaking at $\gtrsim10^{43}$ erg s$^{-1}$, declining steeply as $t^{-2}-t^{-4}$ to reach $\gtrsim10^{41}-10^{42}$ erg s$^{-1}$ at 0.5 d and leveling off on a plateau at $10^{39}-10^{41.5}$ erg s$^{-1}$ ($M_V$ between $-$10 to $-$15 mag) after a few days. Their spectra evolve considerably during the first few days, morphing from UV- to optical-dominated. The UV range shows numerous resonance transitions from metals like C, N, and O, whereas the optical primarily shows H I Balmer lines. These properties are qualitatively similar to those observed, as well as obtained in models of Type II supernovae. Observables from these events exhibit clear correlations with collision configurations, including impact parameter, relative velocity, and stellar masses. We provide fitting formulae to describe these correlations. Detecting these flares requires sub-day cadence surveys such as ULTRASAT, combined with spectroscopic observations to disentangle degeneracies and infer collision characteristics.

Figures

Figures reproduced from arXiv: 2502.05265 by the authors.

Figure 1
Figure 1. Ranges of the relative velocity between stars and stel [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Internal structure of stellar models. Density (top) and H mass fraction (bottom) profiles of our main-sequence stellar models [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Gas density distribution at successive moments in main-sequence star collisions with [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Evolution of the ejecta kinetic and radiative energies in a sample of stellar-collision simulations. We show radiation energy [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 7
Figure 7. Figure 7: Spectral evolution from the UV to the optical and from [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]
Figure 6
Figure 6. Figure 6: Photometric properties for main-sequence star collisions [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: Spectral comparisons of models grouped by the same values of [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]
Figure 9
Figure 9. Figure 9: Ejecta mass Mej (top) and its fraction of the total mass Mej/(M1 + M2) (bottom), at 0.5 d after the collision as a func￾tion of initial kinetic energy Ekin,coll. We use different markers to distinguish collisions of different masses, while different col￾ors represent d…
Figure 10
Figure 10. Figure 10: Same as Fig [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]
Figure 11
Figure 11. Figure 11: Bolometric luminosity Lbol,1d (top-left) and UVW1-band magnitudes MUVW1,1d (top-right), V-band magnitudes MV,1d (bottom-left) as a function of initial kinetic energy Ekin,coll and the Doppler velocity VHα,1d (bottom-right) at maximum absorp￾tion measured in the Hα lin…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 24 canonical work pages

  1. [1]

    Colliding red giants in galactic nuclei: Shocks, jets, impact on the ISM, X- and gamma-rays, neutrinos, fusion ignition and afterglow

    Amaro Seoane , P. 2023 a , arXiv e-prints, arXiv:2307.10330

    work page Pith review arXiv 2023

  2. [2]

    2023 b , , 947, 8

    Amaro Seoane , P. 2023 b , , 947, 8

    work page 2023

  3. [3]

    Bahcall , J. N. & Wolf , R. A. 1976, , 209, 214

    1976

  4. [4]

    2013, , 434, L26

    Balberg , S., Sari , R., & Loeb , A. 2013, , 434, L26

    2013

  5. [5]

    & Yassur , G

    Balberg , S. & Yassur , G. 2023, , 952, 149

    work page 2023

  6. [6]

    & Hilker , M

    Baumgardt , H. & Hilker , M. 2018, , 478, 1520

    2018

  7. [7]

    2024, , 974, L22

    Brutman , Y., Steinberg , E., & Balberg , S. 2024, , 974, L22

    work page 2024

  8. [8]

    Burbidge , E. M. & Sandage , A. 1958, , 128, 174

    work page 1958

Show all 43 references
  1. [9]

    2016, , 823, 102

    Choi , J., Dotter , A., Conroy , C., et al. 2016, , 823, 102

    2016

  2. [10]

    Cohen , J. G. 1983, , 270, L41

    1983

  3. [11]

    & Hillier , D

    Dessart , L. & Hillier , D. J. 2011, , 415, 3497

    2011

  4. [12]

    & Hillier , D

    Dessart , L. & Hillier , D. J. 2019, , 622, A70

    2019

  5. [13]

    J., Waldman , R., & Livne , E

    Dessart , L., Hillier , D. J., Waldman , R., & Livne , E. 2013, , 433, 1745

    2013

  6. [14]

    Dessart , L., Ryu , T., Amaro Seoane , P., & Taylor , A. M. 2024, , 682, A58

    2024

  7. [15]

    & Benz , W

    Freitag , M. & Benz , W. 2005, , 358, 1133

    2005

  8. [16]

    2010, Reviews of Modern Physics, 82, 3121

    Genzel , R., Eisenhauer , F., & Gillessen , S. 2010, Reviews of Modern Physics, 82, 3121

    2010

  9. [17]

    2021, , 59, 21

    Gezari , S. 2021, , 59, 21

    2021

  10. [18]

    2009, , 692, 1075

    Gillessen , S., Eisenhauer , F., Trippe , S., et al. 2009, , 692, 1075

    2009

  11. [19]

    Harris , W. E. 2010, arXiv e-prints, arXiv:1012.3224

    2010 arXiv

  12. [20]

    Hillier , D. J. & Dessart , L. 2012, , 424, 252

    2012

  13. [21]

    Hu , B. X. & Loeb , A. 2024, , 689, A23

    2024

  14. [22]

    S., Bauer , E

    Jermyn , A. S., Bauer , E. B., Schwab , J., et al. 2023, , 265, 15

    2023

  15. [23]

    R., Harrison , F

    Kulkarni , S. R., Harrison , F. A., Grefenstette , B. W., et al. 2021, arXiv:2111.15608

    2021 arXiv

  16. [24]

    Ledoux, W. P. 1947, ApJ, 105, 305

    1947

  17. [25]

    T., R \"o pke , F

    Ohlmann , S. T., R \"o pke , F. K., Pakmor , R., & Springel , V. 2017, , 599, A5

    2017

  18. [26]

    2016, , 455, 1134

    Pakmor , R., Springel , V., Bauer , A., et al. 2016, , 455, 1134

    2016

  19. [27]

    2011, , 192, 3

    Paxton , B., Bildsten , L., Dotter , A., et al. 2011, , 192, 3

    2011

  20. [28]

    2013, , 208, 4

    Paxton , B., Cantiello , M., Arras , P., et al. 2013, , 208, 4

    2013

  21. [29]

    & Meylan , G

    Pryor , C. & Meylan , G. 1993, in Astronomical Society of the Pacific Conference Series, Vol. 50, Structure and Dynamics of Globular Clusters, ed. S. G. Djorgovski & G. Meylan , 357

    1993

  22. [30]

    C., Naoz , S., Gautam , A

    Rose , S. C., Naoz , S., Gautam , A. K., et al. 2020, , 904, 113

    2020

  23. [31]

    C., Naoz , S., Sari , R., & Linial , I

    Rose , S. C., Naoz , S., Sari , R., & Linial , I. 2023, , 955, 30

    2023

  24. [32]

    M., & Ohlmann , S

    Ryu , T., Amaro Seoane , P., Taylor , A. M., & Ohlmann , S. T. 2024 a , , 528, 6193

    2024

  25. [33]

    2024 b , arXiv e-prints, arXiv:2410.00148

    Ryu , T., Sills , A., Pakmor , R., de Mink , S., & Mathieu , R. 2024 b , arXiv e-prints, arXiv:2410.00148

    2024 arXiv

  26. [34]

    Sandage , A. R. 1953, , 58, 61

    1953

  27. [35]

    2024, , 964, 74

    Shvartzvald , Y., Waxman , E., Gal-Yam , A., et al. 2024, , 964, 74

    2024

  28. [36]

    D., & Demarque , P

    Sills , A., Bailyn , C. D., & Demarque , P. 1995, , 455, L163

    1995

  29. [37]

    2010, , 401, 791

    Springel , V. 2010, , 401, 791

    2010

  30. [38]

    E., et al

    Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261

    2020

  31. [39]

    & Ryu , T

    Wang , C. & Ryu , T. 2024, arXiv e-prints, arXiv:2410.10314

    2024 arXiv

  32. [40]

    2020, , 248, 32

    Weinberger , R., Springel , V., & Pakmor , R. 2020, , 248, 32

    2020

  33. [41]

    & Ryu , T

    Wevers , T. & Ryu , T. 2023, arXiv e-prints, arXiv:2310.16879

    2023 arXiv

  34. [42]

    , " * write output.state after.block = add.period write newline

    ENTRY address archiveprefix author booktitle chapter edition editor howpublished institution eprint journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sent...

  35. [43]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.