pith. machine review for the scientific record. sign in

arxiv: 2605.07928 · v1 · submitted 2026-05-08 · 🌌 astro-ph.IM · astro-ph.GA· astro-ph.SR· physics.comp-ph

Recognition: 1 theorem link

· Lean Theorem

Systematic Comparison between Constrained Transport and Mixed Divergence Cleaning Methods for Astrophysical Magnetohydrodynamic Simulations

Kazunari Iwasaki, Kengo Tomida, Kenji Eric Sadanari, Shinsuke Takasao

Pith reviewed 2026-05-11 03:10 UTC · model grok-4.3

classification 🌌 astro-ph.IM astro-ph.GAastro-ph.SRphysics.comp-ph
keywords MHD simulationsconstrained transportdivergence cleaningDedner's schemesolenoidal constraintastrophysical magnetohydrodynamicsnumerical methods
0
0 comments X

The pith

Constrained transport outperforms Dedner's divergence cleaning in MHD simulations when magnetic fields localize or timesteps change suddenly.

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

The paper systematically compares constrained transport using staggered grids against Dedner's mixed divergence cleaning in astrophysical MHD simulations. It demonstrates that the original Dedner's scheme produces substantial artifacts when magnetic fields are strongly localized or timesteps vary abruptly. Through idealized tests and practical applications, the authors show that some earlier findings, such as extremely rapid magnetic field growth in early-Universe star formation, may stem from these numerical issues. They suggest modifications to strengthen divergence cleaning, yet conclude that constrained transport remains more accurate and reliable across many regimes.

Core claim

Numerical experiments show that the original Dedner's scheme becomes inaccurate when magnetic fields are strongly localized or when the timestep suddenly changes, while the constrained transport scheme maintains superior accuracy and reliability in enforcing the solenoidal constraint.

What carries the argument

Constrained transport on staggered grids versus Dedner's mixed divergence cleaning that introduces an additional variable to damp divergence errors in the MHD equations.

If this is right

  • Previous claims of extremely rapid magnetic field growth in early-Universe star formation may require re-evaluation with constrained transport.
  • Simulation codes using Dedner's scheme should consider adopting constrained transport or the proposed modifications for greater robustness.
  • Divergence errors in localized-field regimes can contaminate physical interpretations unless the numerical method is chosen carefully.
  • Modified Dedner's schemes may reduce but not eliminate the accuracy gap with constrained transport in variable-timestep runs.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Wider adoption of constrained transport could revise models of magnetic field amplification in collapsing astrophysical objects.
  • Systematic tests in codes with adaptive timesteps would reveal whether the artifacts persist beyond the reported experiments.
  • Other divergence-cleaning variants may share similar vulnerabilities to sudden field localization.

Load-bearing premise

The chosen idealized tests and practical applications represent the typical regimes where these methods are used in astrophysical MHD research.

What would settle it

A controlled MHD test with strongly localized magnetic fields and abrupt timestep changes in which Dedner's original scheme matches the accuracy of constrained transport without producing artifacts.

Figures

Figures reproduced from arXiv: 2605.07928 by Kazunari Iwasaki, Kengo Tomida, Kenji Eric Sadanari, Shinsuke Takasao.

Figure 1
Figure 1. Figure 1: Distributions of the magnetic field strength with magnetic field directions (orange color maps with cyan arrows) and the divergence cleaning variable (shown in the lower halves of panels (b), (c), (e) and (f)) in the KHI test with uniform magnetic fields at t = 3. Panels (a) and (d) correspond to CT, (b) and (e) to D1v, and (c) and (f) to Dhv. The top and bottom rows show the low- (h = 1/128) and high-reso… view at source ↗
Figure 2
Figure 2. Figure 2: Profiles of the magnetic field strength at x = 0 in the KHI test with uniform magnetic fields at t = 3. The top, middle, and bottom rows show CT, D1v and Dhv models, and the left and right columns show the low- and high-resolution models, respectively. The labels (a) – (f) correspond to those in [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Comparison between the D1c models with different ch. From left to right, ch = 1.0, 2.0, 4.0, 8.0. The top and bottom rows show the low- (h = 1/128) and high-resolution (h = 1/256) models, respectively. performs somewhat inconsistently, here we focus on the models with L = 1 (D1c). First, we find that the sys￾tem is numerically unstable for a small transport speed (ch ≲ 1.0)10. Beyond that, the solutions be… view at source ↗
Figure 4
Figure 4. Figure 4: Same as [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Same as [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Time evolution of the magnetic field strength with magnetic field directions (orange color maps with cyan arrows) and the divergence cleaning variable (bottom half) in the D1v (top) and Dhv (bottom) models before and after the sudden change in the timestep [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Time evolution of the maximum divergence clean￾ing variable. is not constant. However, in the conventional imple￾mentation suggested in the original paper (A. Dedner et al. 2002), the transport speed in eq.(19) changes ac￾cording to the timestep. In practical applications, the timestep can change as the system evolves. Usually, the timestep changes gradually as long as the system evolves continuously. Howe… view at source ↗
Figure 8
Figure 8. Figure 8: Same as [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Results of the simulations of the collapsing clouds in the early Universe. (a) Density cross section with MeshBlock boundaries of CT, (b)–(f) Magnetic field strength of CT, D1v, D1c with ch = 50, Dhv, Dhc with ch = 50, respectively. The maximum magnetic field strength is shown on each panel. The different sizes of the MeshBlocks in (a) correspond to different AMR refinement levels [PITH_FULL_IMAGE:figures… view at source ↗
Figure 10
Figure 10. Figure 10: Zoom-in views of the collapsing clouds in the early Universe. Panels (a) and (b) show density cross sections with gas velocity directions for CT and D1c with ch = 50, respectively, while panels (c) and (d) show magnetic field strength with magnetic field directions for the same models. dius, the density increases as ρ ∝ r −3 but the magnetic field scales as B ∝ r −2 . Therefore, a power-law scaling of B ∝… view at source ↗
Figure 11
Figure 11. Figure 11: Time evolution of the timesteps as a function of the time to te (or the lookback time). The lines are plotted with offsets. or less constant. In all the models, the timesteps oc￾casionally get short for an instance (seen as downward spikes). In the present simulations, these are caused by transient numerical errors. For example, numerical fluxes near a strong discontinuity can be inaccurate and produce a … view at source ↗
Figure 12
Figure 12. Figure 12: Results of the simulations of the collapsing clouds in the present-day environment. From left to right: CT, D1v, and Dhv. From top to bottom: the gas density, plasma beta with magnetic field directions, and radial velocity with flow directions. 3.5.2. Results [PITH_FULL_IMAGE:figures/full_fig_p018_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: From top left to bottom right: vy/cs, ψ, pmag for Dedner’s method and pmag for the CT method. The data are taken at t = 509 in the numerical unit [PITH_FULL_IMAGE:figures/full_fig_p020_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: Evolution of the energy fluxes measured at the height of y = 1.1. Top: the absolute values of the vertical component of the Poynting flux (blue) and the ψ-related flux (gold). Bottom: the ratio of the two. spatial reconstruction to better resolve turbulent flows. For the calculation with Dedner’s method, we use the D1v method with variable ch and L = 1. 3.6.2. Results [PITH_FULL_IMAGE:figures/full_fig_p0… view at source ↗
Figure 15
Figure 15. Figure 15: Temporal evolution of the magnetic energies. The top and bottom panels show the magnetic energies in￾tegrated in y ≥ 1.1 and y < 1.1, respectively. The blue and orange lines denote the results for Dedner’s method and CT method, respectively. The bottom panels of [PITH_FULL_IMAGE:figures/full_fig_p021_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: shows the result of the KHI test with localized magnetic fields presented in Section 3.2. The amplitude of the divergence error increases with time as expected, but the scheme does not crash (i.e., does not cause a catastrophic explosion or negative density/pressure) even with the substantial divergence error. The mag￾netic fields show scaly patterns which are clearly artifi￾cial. While the artificially p… view at source ↗
Figure 17
Figure 17. Figure 17: Distributions of the thermal energy error with magnetic field directions in the circularly polarized Alfv´en wave test at t = 5.0 [PITH_FULL_IMAGE:figures/full_fig_p027_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: The time evolution of the maximum (solid) and minimum (dashed) thermal energy errors in the circularly polarized Alfv´en wave test. and in this coordinate system, v and B are initialized as v1 = 0, v2 = v⊥ sin(2πx1), v3 = v⊥ cos(2πx1), B1 = B0, B2 = B⊥ sin(2πx1), B3 = B⊥ cos(2πx1), where v⊥ = 0.1 and B⊥ = 0.1 set the amplitude of the wave. We adopt B0 = 1, corresponding to the Alfv´en speed of va = 1 and … view at source ↗
read the original abstract

Magnetohydrodynamic (MHD) simulations are indispensable research infrastructure in astrophysics today. In order to satisfy the solenoidal constraint of the MHD equations on discretized grids, modern simulation codes often employ either constrained transport (CT) with a staggered grid or divergence cleaning using an additional variable. We compare CT and Dedner's mixed divergence cleaning schemes systematically, and find that the divergence cleaning scheme can produce substantial artifacts in certain situations. Through numerical experiments including both idealized tests and practical applications, we show that the original implementation of Dedner's scheme becomes inaccurate when magnetic fields are strongly localized or when the timestep suddenly changes. We find that some previous results, such as the extremely rapid growth of magnetic fields during star formation in the early Universe, may be affected by the spurious behavior of the divergence cleaning scheme. We propose a few modifications to improve the robustness of the divergence cleaning method. Nevertheless, we find that the CT scheme is more accurate and reliable in many situations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript performs a systematic numerical comparison of constrained transport (CT) on staggered grids versus Dedner's mixed divergence cleaning for enforcing the solenoidal constraint in astrophysical MHD simulations. Through idealized test problems and practical applications, it identifies regimes where the original Dedner implementation produces substantial artifacts, specifically with strongly localized magnetic fields or abrupt timestep changes. The authors suggest that this behavior may have influenced certain prior results, including claims of extremely rapid magnetic field growth during early-Universe star formation, propose modifications to improve cleaning robustness, and conclude that CT is more accurate and reliable in many situations.

Significance. If the empirical findings hold, the work is significant for the astrophysical MHD community because it supplies concrete evidence of limitations in a widely adopted divergence-cleaning technique and offers practical guidance on method selection and fixes. The side-by-side testing across idealized and applied cases, together with the proposed modifications, directly aids reproducibility and reliability of future simulations.

major comments (2)
  1. Abstract and discussion: the assertion that 'some previous results, such as the extremely rapid growth of magnetic fields during star formation in the early Universe, may be affected by the spurious behavior of the divergence cleaning scheme' rests on extrapolation from the idealized and practical tests rather than a direct head-to-head re-simulation of the cited historical runs at equivalent resolution and initial conditions using CT within the same code base. This step is load-bearing for the strongest implication drawn from the comparison.
  2. Results section (practical applications): the manuscript does not report quantitative error norms, convergence rates, or direct side-by-side divergence-error time series for the practical astrophysical cases, making it difficult to assess the magnitude and statistical significance of the reported artifacts relative to CT.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the constructive and detailed review, as well as for recognizing the potential significance of our findings for the astrophysical MHD community. We address the two major comments point by point below, indicating the revisions we plan to make.

read point-by-point responses

  1. Referee: Abstract and discussion: the assertion that 'some previous results, such as the extremely rapid growth of magnetic fields during star formation in the early Universe, may be affected by the spurious behavior of the divergence cleaning scheme' rests on extrapolation from the idealized and practical tests rather than a direct head-to-head re-simulation of the cited historical runs at equivalent resolution and initial conditions using CT within the same code base. This step is load-bearing for the strongest implication drawn from the comparison.

    Authors: We agree that a direct head-to-head re-simulation of the exact cited runs would constitute stronger evidence. Our current claim is based on the fact that the artifact-triggering conditions we identified (strongly localized magnetic fields and abrupt timestep changes) are present in the setups of the referenced early-Universe star-formation simulations that employed Dedner's cleaning. In the revised manuscript we will moderate the language in the abstract and discussion to describe this as a plausible concern that may warrant re-examination, rather than a definitive statement that prior results are affected. We will also add a short paragraph explaining the mapping between our test conditions and those in the cited literature. revision: partial

  2. Referee: Results section (practical applications): the manuscript does not report quantitative error norms, convergence rates, or direct side-by-side divergence-error time series for the practical astrophysical cases, making it difficult to assess the magnitude and statistical significance of the reported artifacts relative to CT.

    Authors: We accept this criticism. The revised manuscript will include quantitative L2 and L∞ norms of the divergence error for both CT and Dedner runs in the practical-application sections, together with time-series plots of the divergence error and, where the problem setup permits, a brief convergence study. These additions will allow readers to evaluate the size and significance of the observed differences. revision: yes

standing simulated objections not resolved
  • Performing a direct re-simulation of the specific historical runs cited in the abstract and discussion at equivalent resolution and initial conditions using CT within the original code bases; such an exercise lies outside the scope of the present study and would require substantial additional code development and computational resources.

Circularity Check

0 steps flagged

No circularity: empirical side-by-side numerical comparison

full rationale

The paper performs direct numerical experiments comparing established CT and Dedner divergence-cleaning schemes on idealized tests plus practical applications. All claims (artifact identification, proposed modifications, and the inference that prior rapid B-field growth results may be affected) rest on those simulations rather than any derivation, fitted parameter, or self-referential prediction. No load-bearing self-citation, uniqueness theorem, or ansatz is invoked; the work is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper relies on the standard requirement that MHD simulations preserve a divergence-free magnetic field; no free parameters are fitted, no new physical entities are postulated, and no ad-hoc axioms beyond the domain assumption of the solenoidal constraint are introduced.

axioms (1)
  • domain assumption The MHD equations require that the magnetic field remains divergence-free (div B = 0) on discretized grids.
    This is the fundamental numerical constraint stated in the abstract as the motivation for both methods.

pith-pipeline@v0.9.0 · 5492 in / 1307 out tokens · 57462 ms · 2026-05-11T03:10:39.193866+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

59 extracted references · 59 canonical work pages

  1. [1]

    2021, PrincetonUniversity/athena: Athena++ v21.0, 21.0 Zenodo, doi: 10.5281/zenodo.4455880

    Athena++ development team. 2021, PrincetonUniversity/athena: Athena++ v21.0, 21.0 Zenodo, doi: 10.5281/zenodo.4455880

    work page doi:10.5281/zenodo.4455880 2021

  2. [2]

    Balsara, D. S. 2004, ApJS, 151, 149, doi: 10.1086/381377

    work page doi:10.1086/381377 2004

  3. [3]

    S., & Kim, J

    Balsara, D. S., & Kim, J. 2004, ApJ, 602, 1079, doi: 10.1086/381051

    work page doi:10.1086/381051 2004

  4. [4]

    Bonnor, W. B. 1956, MNRAS, 116, 351

    work page 1956

  5. [5]

    U., & Barnes, D

    Brackbill, J. U., & Barnes, D. C. 1980, Journal of Computational Physics, 35, 426, doi: 10.1016/0021-9991(80)90079-0

    work page doi:10.1016/0021-9991(80)90079-0 1980

  6. [6]

    L., Norman, M

    Bryan, G. L., Norman, M. L., O’Shea, B. W., et al. 2014, ApJS, 211, 19, doi: 10.1088/0067-0049/211/2/19

    work page doi:10.1088/0067-0049/211/2/19 2014

  7. [7]

    Jones, T. W. 2009, ApJS, 182, 519, doi: 10.1088/0067-0049/182/2/519

    work page doi:10.1088/0067-0049/182/2/519 2009

  8. [8]

    2002, Journal of Computational Physics, 175, 645, doi: 10.1006/jcph.2001.6961

    Dedner, A., Kemm, F., Kr¨ oner, D., et al. 2002, Journal of Computational Physics, 175, 645, doi: 10.1006/jcph.2001.6961

    work page doi:10.1006/jcph.2001.6961 2002

  9. [9]

    1955, Zeitschrift fur Astrophysik, 36, 222

    Ebert, R. 1955, Zeitschrift fur Astrophysik, 36, 222

    work page 1955

  10. [10]

    R., & Hawley, J

    Evans, C. R., & Hawley, J. F. 1988, ApJ, 332, 659, doi: 10.1086/166684

    work page doi:10.1086/166684 1988

  11. [11]

    2006, A&A, 457, 371, doi: 10.1051/0004-6361:20065371

    Fromang, S., Hennebelle, P., & Teyssier, R. 2006, A&A, 457, 371, doi: 10.1051/0004-6361:20065371

    work page doi:10.1051/0004-6361:20065371 2006

  12. [12]

    , keywords =

    Fryxell, B., Olson, K., Ricker, P., et al. 2000, ApJS, 131, 273, doi: 10.1086/317361

    work page doi:10.1086/317361 2000

  13. [13]

    A., & Stone, J

    Gardiner, T. A., & Stone, J. M. 2005, Journal of Computational Physics, 205, 509, doi: 10.1016/j.jcp.2004.11.016

    work page doi:10.1016/j.jcp.2004.11.016 2005

  14. [14]

    F., & Stone, J

    Hawley, J. F., & Stone, J. M. 1995, Computer Physics Communications, 89, 127, doi: 10.1016/0010-4655(95)00190-Q

    work page doi:10.1016/0010-4655(95)00190-q 1995

  15. [15]

    Hirano, S., & Machida, M. N. 2022, ApJL, 935, L16, doi: 10.3847/2041-8213/ac85e0

    work page doi:10.3847/2041-8213/ac85e0 2022

  16. [16]

    doi:10.1093/mnras/stv2180 , eprint =

    Hopkins, P. F., & Raives, M. J. 2016, MNRAS, 455, 51, doi: 10.1093/mnras/stv2180

    work page doi:10.1093/mnras/stv2180 2016

  17. [17]

    2013, in Astronomical Society of the Pacific Conference Series, Vol

    Iwasaki, K., & Inutsuka, S.-I. 2013, in Astronomical Society of the Pacific Conference Series, Vol. 474, Numerical Modeling of Space Plasma Flows (ASTRONUM2012), ed. N. V. Pogorelov, E. Audit, & G. P. Zank, 239

    work page 2013

  18. [18]

    2023, A&A, 673, A66, doi: 10.1051/0004-6361/202245359

    Keppens, R., Popescu Braileanu, B., Zhou, Y., et al. 2023, A&A, 673, A66, doi: 10.1051/0004-6361/202245359

    work page doi:10.1051/0004-6361/202245359 2023

  19. [19]

    H., Li, Z.-Y., Chen, C.-Y., Tomida, K., & Zhao, B

    Lam, K. H., Li, Z.-Y., Chen, C.-Y., Tomida, K., & Zhao, B. 2019, MNRAS, 489, 5326, doi: 10.1093/mnras/stz2436

    work page doi:10.1093/mnras/stz2436 2019

  20. [20]

    Larson, R. B. 1969, MNRAS, 145, 271, doi: 10.1093/mnras/145.3.271

    work page doi:10.1093/mnras/145.3.271 1969

  21. [21]

    A., Schleicher, D

    Latif, M. A., Schleicher, D. R. G., & Khochfar, S. 2023, ApJ, 945, 137, doi: 10.3847/1538-4357/acbcc2

    work page doi:10.3847/1538-4357/acbcc2 2023

  22. [22]

    A., Schleicher, D

    Latif, M. A., Schleicher, D. R. G., & Schmidt, W. 2014, MNRAS, 440, 1551, doi: 10.1093/mnras/stu357

    work page doi:10.1093/mnras/stu357 2014

  23. [23]

    N., Hirano, S., & Basu, S

    Machida, M. N., Hirano, S., & Basu, S. 2025, ApJ, 988, 6, doi: 10.3847/1538-4357/addc56

    work page doi:10.3847/1538-4357/addc56 2025

  24. [24]

    2016, A&A, 587, A32, doi: 10.1051/0004-6361/201526371

    Masson, J., Chabrier, G., Hennebelle, P., Vaytet, N., & Commer¸ con, B. 2016, A&A, 587, A32, doi: 10.1051/0004-6361/201526371

    work page doi:10.1051/0004-6361/201526371 2016

  25. [25]

    2019, PASJ, 71, 83, doi: 10.1093/pasj/psz064

    Matsumoto, Y., Asahina, Y., Kudoh, Y., et al. 2019, PASJ, 71, 83, doi: 10.1093/pasj/psz064

    work page doi:10.1093/pasj/psz064 2019

  26. [26]

    2024, A&A, 686, A253, doi: 10.1051/0004-6361/202348405

    Mauxion, J., Lesur, G., & Maret, S. 2024, A&A, 686, A253, doi: 10.1051/0004-6361/202348405

    work page doi:10.1051/0004-6361/202348405 2024

  27. [27]

    C., Zier, O., Naab, T., et al

    Mayer, A. C., Zier, O., Naab, T., et al. 2025, MNRAS, 537, 379, doi: 10.1093/mnras/staf027

    work page doi:10.1093/mnras/staf027 2025

  28. [28]

    2010, Journal of Computational Physics, 229, 2117, doi: 10.1016/j.jcp.2009.11.026

    Mignone, A., & Tzeferacos, P. 2010, Journal of Computational Physics, 229, 2117, doi: 10.1016/j.jcp.2009.11.026

    work page doi:10.1016/j.jcp.2009.11.026 2010

  29. [29]

    2012, ApJS, 198, 7, doi: 10.1088/0067-0049/198/1/7

    Mignone, A., Zanni, C., Tzeferacos, P., et al. 2012, ApJS, 198, 7, doi: 10.1088/0067-0049/198/1/7

    work page doi:10.1088/0067-0049/198/1/7 2012

  30. [30]

    2005, Journal of Computational Physics, 208, 315, doi: 10.1016/j.jcp.2005.02.017

    Miyoshi, T., & Kusano, K. 2005, Journal of Computational Physics, 208, 315, doi: 10.1016/j.jcp.2005.02.017

    work page doi:10.1016/j.jcp.2005.02.017 2005

  31. [31]

    2016, MNRAS, 463, 477, doi: 10.1093/mnras/stw2004

    Mocz, P., Pakmor, R., Springel, V., et al. 2016, MNRAS, 463, 477, doi: 10.1093/mnras/stw2004

    work page doi:10.1093/mnras/stw2004 2016

  32. [32]

    2014, MNRAS, 442, 43, doi: 10.1093/mnras/stu865

    Mocz, P., Vogelsberger, M., & Hernquist, L. 2014, MNRAS, 442, 43, doi: 10.1093/mnras/stu865

    work page doi:10.1093/mnras/stu865 2014

  33. [33]

    , keywords =

    Pakmor, R., Bauer, A., & Springel, V. 2011, MNRAS, 418, 1392, doi: 10.1111/j.1365-2966.2011.19591.x

    work page doi:10.1111/j.1365-2966.2011.19591.x 2011

  34. [34]

    Powell, K. G. 1994, An approximate Riemann solver for magnetohydrodynamics (that works more than one dimension), Tech. rep

    work page 1994

  35. [35]

    G., Roe, P

    Powell, K. G., Roe, P. L., Linde, T. J., Gombosi, T. I., & De Zeeuw, D. L. 1999, Journal of Computational Physics, 154, 284, doi: 10.1006/jcph.1999.6299

    work page doi:10.1006/jcph.1999.6299 1999

  36. [36]

    doi:10.1111/j.1365-2966.2005.09360.x , eprint =

    Price, D. J., & Monaghan, J. J. 2005, MNRAS, 364, 384, doi: 10.1111/j.1365-2966.2005.09576.x

    work page doi:10.1111/j.1365-2966.2005.09576.x 2005

  37. [37]

    J., Wurster, J., Tricco, T

    Price, D. J., Wurster, J., Tricco, T. S., et al. 2018, PASA, 35, e031, doi: 10.1017/pasa.2018.25

    work page doi:10.1017/pasa.2018.25 2018

  38. [38]

    E., Omukai, K., Sugimura, K., Matsumoto, T., & Tomida, K

    Sadanari, K. E., Omukai, K., Sugimura, K., Matsumoto, T., & Tomida, K. 2021, MNRAS, 505, 4197, doi: 10.1093/mnras/stab1330 29

    work page doi:10.1093/mnras/stab1330 2021

  39. [39]

    E., Omukai, K., Sugimura, K., Matsumoto, T., & Tomida, K

    Sadanari, K. E., Omukai, K., Sugimura, K., Matsumoto, T., & Tomida, K. 2023, MNRAS, 519, 3076, doi: 10.1093/mnras/stac3724

    work page doi:10.1093/mnras/stac3724 2023

  40. [40]

    E., Omukai, K., Sugimura, K., Matsumoto, T., & Tomida, K

    Sadanari, K. E., Omukai, K., Sugimura, K., Matsumoto, T., & Tomida, K. 2024, PASJ, 76, 823, doi: 10.1093/pasj/psae051

    work page doi:10.1093/pasj/psae051 2024

  41. [41]

    C., Stehle, R., & Papaloizou, J

    Spruit, H. C., Stehle, R., & Papaloizou, J. C. B. 1995, MNRAS, 275, 1223

    work page 1995

  42. [42]

    P., & Price, D

    Steinwandel, U. P., & Price, D. J. 2025, arXiv e-prints, arXiv:2511.19615, doi: 10.48550/arXiv.2511.19615

    work page doi:10.48550/arxiv.2511.19615 2025

  43. [43]

    M., & Gardiner, T

    Stone, J. M., & Gardiner, T. 2009, NewA, 14, 139, doi: 10.1016/j.newast.2008.06.003

    work page doi:10.1016/j.newast.2008.06.003 2009

  44. [44]

    The Astrophysical Journal Supplement Series , author =

    Stone, J. M., Tomida, K., White, C. J., & Felker, K. G. 2020, ApJS, 249, 4, doi: 10.3847/1538-4365/ab929b

    work page doi:10.3847/1538-4365/ab929b 2020

  45. [45]

    2019, Computer Physics Communications, 245, 106866, doi: https://doi.org/10.1016/j.cpc.2019.106866

    Teunissen, J., & Keppens, R. 2019, Computer Physics Communications, 245, 106866, doi: https://doi.org/10.1016/j.cpc.2019.106866

    work page doi:10.1016/j.cpc.2019.106866 2019

  46. [46]

    Tomida, K., Okuzumi, S., & Machida, M. N. 2015, ApJ, 801, 117, doi: 10.1088/0004-637X/801/2/117

    work page doi:10.1088/0004-637x/801/2/117 2015

  47. [47]

    Tomida, K., & Stone, J. M. 2023, ApJS, 266, 7, doi: 10.3847/1538-4365/acc2c0

    work page doi:10.3847/1538-4365/acc2c0 2023

  48. [48]

    2013, ApJ, 763, 6, doi: 10.1088/0004-637X/763/1/6

    Tomida, K., Tomisaka, K., Matsumoto, T., et al. 2013, ApJ, 763, 6, doi: 10.1088/0004-637X/763/1/6

    work page doi:10.1088/0004-637x/763/1/6 2013

  49. [49]

    S., & Price, D

    Tricco, T. S., & Price, D. J. 2012, Journal of Computational Physics, 231, 7214, doi: 10.1016/j.jcp.2012.06.039

    work page doi:10.1016/j.jcp.2012.06.039 2012

  50. [50]

    S., Price, D

    Tricco, T. S., Price, D. J., & Bate, M. R. 2016, Journal of Computational Physics, 322, 326, doi: 10.1016/j.jcp.2016.06.053

    work page doi:10.1016/j.jcp.2016.06.053 2016

  51. [51]

    N., & Inutsuka, S

    Tsukamoto, Y., Iwasaki, K., Okuzumi, S., Machida, M. N., & Inutsuka, S. 2015, MNRAS, 452, 278, doi: 10.1093/mnras/stv1290

    work page doi:10.1093/mnras/stv1290 2015

  52. [52]

    H., Tomida, K., & Hsu, C.-Y

    Tu, Y., Li, Z.-Y., Lam, K. H., Tomida, K., & Hsu, C.-Y. 2024, MNRAS, 527, 10131, doi: 10.1093/mnras/stad3843 T´ oth, G. 2000, Journal of Computational Physics, 161, 605, doi: https://doi.org/10.1006/jcph.2000.6519

    work page doi:10.1093/mnras/stad3843 2024

  53. [53]

    2018, A&A, 615, A5, doi: 10.1051/0004-6361/201732075

    Chabrier, G. 2018, A&A, 615, A5, doi: 10.1051/0004-6361/201732075

    work page doi:10.1051/0004-6361/201732075 2018

  54. [54]

    2013, ESAIM: Proc., 43, 180, doi: 10.1051/proc/201343012

    Vides, J., Audit, E., Guillard, H., & Nkonga, B. 2013, ESAIM: Proc., 43, 180, doi: 10.1051/proc/201343012

    work page doi:10.1051/proc/201343012 2013

  55. [55]

    2009, ApJ, 696, 96, doi: 10.1088/0004-637X/696/1/96

    Wang, P., & Abel, T. 2009, ApJ, 696, 96, doi: 10.1088/0004-637X/696/1/96

    work page doi:10.1088/0004-637x/696/1/96 2009

  56. [56]

    Weinberger, V

    Weinberger, R., Springel, V., & Pakmor, R. 2020, ApJS, 248, 32, doi: 10.3847/1538-4365/ab908c

    work page doi:10.3847/1538-4365/ab908c 2020

  57. [57]

    J., & Bate, M

    Wurster, J., Price, D. J., & Bate, M. R. 2016, MNRAS, 457, 1037, doi: 10.1093/mnras/stw013

    work page doi:10.1093/mnras/stw013 2016

  58. [58]

    Xu, W., & Kunz, M. W. 2021, MNRAS, 502, 4911, doi: 10.1093/mnras/stab314

    work page doi:10.1093/mnras/stab314 2021

  59. [59]

    2016, Frontiers in Astronomy and Space Sciences, Volume 3 - 2016, doi: 10.3389/fspas.2016.00006

    Zhang, M., & Feng, X. 2016, Frontiers in Astronomy and Space Sciences, Volume 3 - 2016, doi: 10.3389/fspas.2016.00006

    work page doi:10.3389/fspas.2016.00006 2016