Efficient and Accurate Force Replay in Cosmological-baryonic Simulations

Abstract

We construct time-evolving gravitational potential models for a Milky Way–mass galaxy from the FIRE-2 suite of cosmological-baryonic simulations using basis function expansions. These models capture the angular variation with spherical harmonics for the halo and azimuthal harmonics for the disk, and the radial or meridional plane variation with splines. We fit low-order expansions (four angular/harmonic terms) to the galaxy's potential for each snapshot, spaced roughly 25 Myr apart, over the last 4 Gyr of its evolution, then extract the forces at discrete times and interpolate them between adjacent snapshots for forward orbit integration. Our method reconstructs the forces felt by simulation particles with high fidelity, with 95% of both stars and dark matter, outside of self-gravitating subhalos, exhibiting errors ≤4% in both the disk and the halo. Imposing symmetry on the model systematically increases these errors, particularly for disk particles, which show greater sensitivity to imposed symmetries. The majority of orbits recovered using the models exhibit positional errors ≤10% for 2–3 orbital periods, with higher errors for orbits that spend more time near the galactic center. Approximate integrals of motion are retrieved with high accuracy even with a larger potential sampling interval of 200 Myr. After 4 Gyr of integration, 43% and 70% of orbits have total energy and angular momentum errors within 10%, respectively. Consequently, there is higher reliability in orbital shape parameters such as pericenters and apocenters, with errors ∼10% even after multiple orbital periods. These techniques have diverse applications, including studying satellite disruption in cosmological contexts.

Copyright and License

Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

Acknowledgement

A.A. and R.E.S. acknowledge support from the Research Corporation through the Scialog Fellows program on Time Domain Astronomy, from NSF grants AST-2007232 and AST-2307787, and from NASA grant 19-ATP19-0068. R.E.S. is supported in part by a Sloan Fellowship. A.W. received support from the NSF via CAREER award AST-2045928 and grant AST-2107772, NASA ATP grant No. 80NSSC20K0513, and HST grant No. GO-16273 from STScI. S.L. acknowledges support from NSF grant AST-2109234 and HST grant No. AR-16624 from STScI. E.C.C. acknowledges support for this work provided by NASA through the NASA Hubble Fellowship Program grant No. HST-HF2-51502 awarded by the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., for NASA, under contract NAS5-26555. N.S. is supported by an NSF Astronomy and Astrophysics Postdoctoral Fellowship under award AST-2303841.

This research is part of the Frontera computing project at the Texas Advanced Computing Center (TACC). Frontera is made possible by National Science Foundation award OAC-1818253. Simulations in this project were run using Early Science Allocation 1923870, and analyzed using computing resources supported by the Scientific Computing Core at the Flatiron Institute. This work used additional computational resources of the University of Texas at Austin and TACC, the NASA Advanced Supercomputing (NAS) Division and the NASA Center for Climate Simulation (NCCS), and the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant No. OCI-1053575.

Funding

A.A. and R.E.S. acknowledge support from the Research Corporation through the Scialog Fellows program on Time Domain Astronomy, from NSF grants AST-2007232 and AST-2307787, and from NASA grant 19-ATP19-0068. R.E.S. is supported in part by a Sloan Fellowship. A.W. received support from the NSF via CAREER award AST-2045928 and grant AST-2107772, NASA ATP grant No. 80NSSC20K0513, and HST grant No. GO-16273 from STScI. S.L. acknowledges support from NSF grant AST-2109234 and HST grant No. AR-16624 from STScI. E.C.C. acknowledges support for this work provided by NASA through the NASA Hubble Fellowship Program grant No. HST-HF2-51502 awarded by the Space Telescope Science Institute, which is operated by the Association of Universities for Research in Astronomy, Inc., for NASA, under contract NAS5-26555. N.S. is supported by an NSF Astronomy and Astrophysics Postdoctoral Fellowship under award AST-2303841.

Data Availability

FIRE-2 simulations are publicly available (A. Wetzel et al. 2023) at http://flathub.flatironinstitute.org/fire. Additional FIRE simulation data are available at https://fire.northwestern.edu/data. A public version of the Gizmo code is available at http://www.tapir.caltech.edu/~phopkins/Site/GIZMO.html.

Software References

IPython (F. Perez & B. E. Granger 2007), Matplotlib (J. D. Hunter 2007), Numpy (C. R. Harris et al. 2020), Gizmo Analysis (A. Wetzel & S. Garrison-Kimmel 2020a), Agama (E. Vasiliev 2019b), Rockstar (P. S. Behroozi et al. 2012), Halo Analysis (A. Wetzel & S. Garrison-Kimmel 2020b), CMasher (E. van der Velden 2020).

Files

Arora_2024_ApJ_977_23.pdf

Files (5.2 MB)

Name	Size	Download all
Arora_2024_ApJ_977_23.pdf md5:49b07029afcd94573eec7563201ab5cc	5.2 MB	Preview Download

	All versions	This version
Views	0	0
Downloads	0	0
Data volume	0 Bytes	0 Bytes