Convergence tests of SIDM N-body simulations
Data files
Sep 26, 2024 version files 229.19 GB
-
C10T225_Nbody.zip
51.89 GB
-
C10T225_reduced.zip
85.08 KB
-
C10T225LOWRES_Nbody.zip
7.75 GB
-
C10T225LOWRES_reduced.zip
90.32 KB
-
C10T9_Nbody.zip
66.30 GB
-
C10T9_reduced.zip
105.81 KB
-
C10T9LOWRES_Nbody.zip
9.61 GB
-
C10T9LOWRES_reduced.zip
110.82 KB
-
C50T225_Nbody.zip
26.12 GB
-
C50T225_reduced.zip
76.43 KB
-
C50T225LOWRES_Nbody.zip
6.36 GB
-
C50T225LOWRES_reduced.zip
71.19 KB
-
C50T9_Nbody.zip
53.36 GB
-
C50T9_reduced.zip
94.64 KB
-
C50T9LOWRES_Nbody.zip
7.81 GB
-
C50T9LOWRES_reduced.zip
90.14 KB
-
README.md
4.86 KB
Abstract
Self-interacting dark matter (SIDM) theory predicts that dark matter halos experience core collapse, a process where the halo's inner region rapidly increases in density and decreases in size. The N-body simulations used to study this process can suffer from numerical errors when simulation parameters are selected incorrectly. Optimal choices for simulation parameters are well-studied for cold dark matter (CDM) but are not deeply understood when self-interactions are included. In order to perform reliable N-body simulations and model core-collapse accurately we must understand the potential numerical errors, how to diagnose them, and what parameter selections must be made to reduce them. We use the Arepo N-body code to perform convergence tests of core-collapsing SIDM halos across a range of halo concentrations and SIDM cross-sections and quantify potential numerical issues related to mass resolution, timestep size, and gravitational softening length. Our tests discover that halos with fewer than 105 simulation particles, a resolution typically not met by subhalos in N-body simulations, suffer from significant discreteness noise that leads to variation and extreme outliers in the collapse rate. At our lowest resolution of N=104 particles, this collapse time variation can reach as high as 20%. At this low resolution, we also find a bias in collapse times and a small number of extreme outliers. Additionally, we find that simulations that run far beyond the age of the Universe, which have been used to calibrate SIDM gravothermal fluid models in previous work, have a sensitivity to the timestep size that is not present in shorter simulations or simulations using only CDM. Our work shows that choices of simulation parameters that yield converged results for some halo masses and SIDM models do not necessarily yield convergence for others.
README: Convergence tests of SIDM N-body simulations
The data here are the product of N-body simulations of self-interacting dark matter (SIDM). The simulations were run with the N-body code Arepo (https://doi.org/10.1111/j.1365-2966.2009.15715.x), including SIDM modifications (https://doi.org/10.1111/j.1365-2966.2012.21182.x).
Each compressed folder in this dataset corresponds to a set of convergence test simulations for a particular dark matter halo and SIDM cross-section. The folder names follow the same convention shown in Table 1 of our publication - where the number following C refers to the initial halo concentration, and the number following T refers to the estimated collapse time in Gyr. For example, C10T225 refers to the simulation set with a concentration of 10 and an approximate collapse time of 225 Gyr. A folder name that is followed by LOWRES contains data for the low-resolution N-body simulations. The resolutions in the standard folders are 10^4, 10^5, and 10^6 particles, while the "low resolution" simulations are 10^4.25, 10^4.5, and 10^4.75.
For each simulation halo, there are two types of compressed folders:
These folders contain the central density data from analyzing N-body simulations. Each CSV file is labeled CTxx with some number xx and corresponds to a particular set of numerical parameters. These parameters are the softening length, timestep size (quantified through the "eta" parameter discussed in our manuscript), and number of particles. The first column of each CSV file is time (Gyr), and the second column is the halo central density (dimensionless, normalized by the NFW scale density rho_s). Each folder has a file called param_key.txt, which matches each CTxx to the numerical parameters used in the corresponding N-body simulation. The columns of param_key.txt are, in order: simulation label, number of particles, softening length (kpc), and eta.
These folders contain particle data for the N-body simulations. Each subfolder is labeled CTxx as in the reduced data and can be matched with numerical parameters with the same param_key.txt. Each CTxx subfolder contains simulation snapshots saved in the format snapshot_xxx.hdf5, where xxx is the snapshot number.
Each snapshot stores the following particle data:
Data Path | Description | Units |
---|---|---|
['PartType1/Coordinates'] | 3D particle positions | kpc |
['PartType1/ParticleIDs'] | Particle identification numbers for tracking between snapshots | |
['PartType1/Potential'] | Particle gravitational potential | (km/s)^2 |
['PartType1/SIDM_NumTotalScatter'] | Number of SIDM scattering events the particle has experienced before the snapshot | |
['PartType1/Velocities'] | 3D particle velocities | km/s |
and the following general data:
Data Path | Description | Units |
---|---|---|
['Header'].attrs['Time'] | Snapshot time | Arepo time units (multiply by 0.978462 to change to Gyr) |
['Header'].attrs['MassTable'][1] | Particle mass | 10^10 solar masses |
Information on the hdf5 file format: https://www.hdfgroup.org/solutions/hdf5/
Information on reading hdf5 files in Python: https://www.h5py.org/
Information on Arepo hdf5 files: https://arepo-code.org/wp-content/userguide/snapshotformat.html
Each CTxx subfolder also contains two other files: energy.txt and parameters-usedvalues. parameters-usedvalues include the parameters used in running the corresponding simulation. A description of the parameter names can be found at https://arepo-code.org/wp-content/userguide/parameterfile.html. energy.txt contains information on the system energy as a function of time. A more detailed description of this file can be found at https://arepo-code.org/wp-content/userguide/diagnosticfiles.html.
Methods
This dataset contains snapshots of simulations run in the N-body code Arepo, with the addition of a scattering interaction between dark matter particles. The snapshots are organized by simulation run and are saved as hdf5 files output by Arepo. The snapshots have not been processed, aside from the removal of some data (such as local particle density) to save space. The results of our publication can be replicated using the remaining particle data: mass, positions, velocities, potentials, number of scattering events, and ID numbers.
In addition to the N-body output, we also share reduced data files. These files contain halo central densities versus time and were computed from the N-body data using the method described in section II.D of our publication.